Research and the Quality of Science Education

RESEARCH AND THE QUALITY OF SCIENCE EDUCATION

Research and the Quality of Science Education Edited by

KERST BOERSMA .

Utrecht University, The Netherlands

MARTIN GOEDHART University of Groningen, The Netherlands

ONNO DE JONG Utrecht University, The Netherlands and

HARRIE EIJKELHOF Utrecht University, The Netherlands

A C.I.P. Catalogue record for this book is available from the Library of Congress.

ISBN-10 ISBN-10 ISBN-13 ISBN-13

1-4020-3672-8 (HB) 1-4020-3673-6 (e-book) 978-1-4020-3672-9 (HB) 978-1-4020-3673-6 (e-book)

Published by Springer, P.O. Box 17, 3300 AA Dordrecht, The Netherlands. www.springeronline.com

Printed on acid-free paper

All Rights Reserved © 2005 Springer No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Printed in the Netherlands.

TABLE OF CONTENTS ix

PREFACE PART 1: THE QUALITY OF SCIENCE EDUCATION Wolf-Michael Roth, From normal to revolutionary science education

3

Piet Lijnse, Reflections on a problem posing approach

15

Svein Lie, How can large international comparative studies contribute to the quality of science education?

27

Wilmad Kuiper, Kerst Boersma, Jan van den Akker, Towards a more curricular focus In international comparative studies on mathematics and science education 41

PART 2: SCIENCE CURRICULUM INNOVATION Jon Ogborn, 40 Years of curriculum development

57

Hanna Westbroek, Kees Klaassen, Astrid Bulte, Albert Pilot, Characteristics of meaningful chemistry education 67 Mary Ratcliffe, Richard Harris, Jenny McWhirter, Cross-curricular collaboration in teaching social aspects of genetics 77 Russel Tytler, School innovation in science: change, culture, complexity Maria Andrée, Ways of using “everyday life” in the science classroom

89 107

PART 3: SCIENCE TEACHER EDUCATION Dimitris Psillos, Anna Spyrtou, Petros Kariotoglou, Science teacher education: issues and proposals 119 Paul Denley, Keith Bischop, Outcomes of professional development in primary science: developing a conceptual framework 129 Rachel Mamlok-Naaman, Oshrit Navon, Miriam Carmeli, Avi Hofstein, Chemistry teachers research their own work: two case studies 141 Tina Jarvis, Anthony Pell, The relationships between primary Teachers’ attitudes and cognition during a two year science in-service programme 157 v

vi

TABLE OF CONTENTS

Machiel Stolk, Astrid Bulte, Onno de Jong, Albert Pilot, Teaching concepts in contexts: designing a chemistry teacher course in a curriculum innovation

169

Virginie Albe, Laurence Simonneaux, Epistemological thought and role-playing: impact on pre-service teachers' opinions on mobile phone risks 181

PART 4: TEACHING-LEARNING SEQUENCES IN SCIENCE EDUCATION Martine Méheut, Teaching–learning sequences tools for learning and/or research 195 John Leach, Jaume Ametller, Andy Hind, Jenny Lewis, Philip Scott, Designing and evaluating short science teaching sequences: improving student learning 209 Björn Andersson, Frank Bach, Mats Hagman, Clas Olander, Anita Wallin Discussing a research programme for the improvement of science teaching

221

Zahava Scherz, Ornit Spektor-Levy, Bat Sheva Eylon, “Scientific communication” : an instructional program for high-order learning skills and its impact on students’ performance 231

PART 5: TEACHING THE NATURE OF SCIENCE Stein Dankert Kolstø, Idar Mestad, Learning about the nature of scientific knowledge: the imitating-science project

247

Saouma Boujaoude, Suha Sowwan, Fouad. Abd-El-Khalick, The effect of using drama in science teaching on students' conceptions of the nature of science 259 Sverre Pettersen, The relevance of teaching about the "Nature of Science" to students of the health sciences

269

Jim Ryder, Andy Hind, John Leach, Teaching about the epistemology of science in school science classrooms: case studies of teachers' experiences 283

PART 6: MODELS, MODELLING AND ANALOGIES IN SCIENCE EDUCATION Wolter Kaper, Martin Goedhart, A three-phase design for productive use of analogy in the teaching of entropy 297 Barbara Crawford, Michael Cullin, Dynamic assessments of preservice teachers’ knowledge of models and modelling 309

TABLE OF CONTENTS

vii

Rosária Justi, John K. Gilbert, Investigating teachers’ ideas about models and modelling – some issues of authenticity

325

Silke Mikelskis-Seifert, Antje Leisner, Investigation of effects and stability in teaching model competence

337

Allan Harrison, Onno de Jong, Using multiple analogies: case study of a chemistry teacher’s preparations, presentations and reflections 353

PART 7: DISCOURSE AND ARGUMENTATION IN SCIENCE EDUCATION Jonathan Osborne, The role of argument in science education

367

Sibel Erduran, Jonathan Osborne, Shirley Simon, The role of argumentation in developing scientific literacy 381 Phil Scott, Eduardo Mortimer, Meaning making in high school science classrooms: a framework for analysing meaning making interactions 395 Marida Ergazaki, Vassiliki Zogza, From a causal question to stating and testing hypotheses: exploring the discursive activity of biology students 407 María Pilar Jiménez-Aleixandre, Cristina Pereiro-Muñoz, Argument construction and change while working on a real environment problem 419

PART 8: TEACHING AND LEARNING SCIENTIFIC CONCEPTS Richard Gunstone, Brian McKittrick, Pamela Mulhall, Textbooks and their authors: another perspective on the difficulties of teaching and learning electricity 435 Kees Klaassen, The concept of force as a constitutive element of understanding the world 447 Jocelyn Locaylocay, Ed van den Berg, Marcelita Magno, Changes in college students’ conceptions of chemical equilibrium

459

Susann Hartmann, Hans Niedderer, Parallel conceptions in the demain of force and motion 471 Gultekin Cakmakci, Jim Donnelly, John Leach, A cross-sectional study of the understanding of the relationships between concentration and reaction rate among Turkish secondary and undergraduate students 483

NAME INDEX

499

PREFACE In August 2003 over 360 researchers met in The Netherlands to exchange experiences and discuss results in the field of science education research. The Conference was organized by the European Science Education Research Association (ESERA), the fourth since the foundation of the Association in 1995. The participants came from 39 countries, mainly from Europe, but also from other continents. Almost all European scholars with a long record of eminent work were present, but also many young researchers who were in the stage of preparing their PhD theses. The abstracts of the more than 300 papers were published in the Book of Abstracts; synopses were published on a CD-ROM. The general theme of the Conference was Research and the Quality of Science Education. This theme was chosen with the importance of science education at all levels of schooling in mind, formal and informal, from primary to higher education. The significance of science education is not only felt by teachers and school administrators, but also by many others: researchers, industrialists, politicians, and parents. Over the last decade science education has been a topic of public debate, related to the results of international comparisons (such as the TIMMS and the PISA studies), the fall of interest in science studies in higher education, and the shortage of teachers. At the same time educational research showed that learning results were often not as good as expected and that the motivation of pupils for science education was less than adequate. Also opinions on effective learning changed from a classical teaching methodology and content to approaches which put more emphasis on concept development, collaborative work, connections with the world outside the classroom (such as modern developments in science and technology), argumentation, modelling, the nature of science, and the use of computer technology. Many innovations have been initiated and practised by science educators, teacher trainers, national curriculum institutes, and professional scientific bodies. In such a dynamic educational setting, research plays an important role: it provides theoretical guidelines, it brings together knowledge and experiences from many countries, and it poses critical questions before, during, and after innovations. In this way it could (and in our opinion should) play a major role in monitoring and promoting the quality of science education. This book is not intended to be proceedings of the conference. The CD-ROM with three-page synopses fulfils this role. Our aim for this book is to publish just a selection of those papers which in our opinion are outstanding, representative of the progress in a variety of fields, and worthwhile enough to be made accessible to a larger audience. We selected around 40 of the 309 presented papers and invited the authors to rewrite their papers according to our format. Each of these rewritten papers was independently reviewed by two experts, and based on their comments, the editorial board returned all submitted papers with guidelines for improvement. Finally, 38 of the papers were approved for publication. ix

P REFACE

x

In order to facilitate reading, the papers were ordered according to the research fields they represent: • The quality of science education • Science curriculum innovation • Science teacher education • Teaching-learning sequences • The nature of science • Models and analogies • Discourse and argumentation • Teaching and learning of concepts. In most cases the position of each paper was clear, in some cases, if various themes were covered, we had to make a choice. For instance, many of the modelling papers dealt with teachers’ professional development. Finally, we would like to thank all those who contributed to the publication of this book: our colleague-organizers of the Conference, the authors, the reviewers, the secretaries of the Centre for Science and Mathematics Education, and the language editor.

The editors: Kerst Boersma Martin Goedhart Onno de Jong Harrie Eijkelhof

PART 1 The Quality of science education

FROM NORMAL TO REVOLUTIONARY SCIENCE EDUCATION

WOLFF-MICHAEL ROTH University of Victoria, Canada

ABSTRACT This paper has the explicit aim to raise questions about ourselves, in fact, to question the very ways in which we science educators do business and understand ourselves. Would it come as a surprise if some readers were upset with me for raising such questions?1 Negative responses to the issues I articulate in this paper are at the very heart of what my chapter is about. How does a community of practice renew itself when at the very moment that those of its members who propose change are often silenced by journal and book reviewers who see their power, which they have gained in the existing community, threatened by new or different ideas? And how can we begin talking about such issues without upsetting those who have different stakes and views? But then, we also need to ask, how can the science education community renew itself if there are gatekeepers who uphold the old order? That is, how can the science education community (of practice) change itself from doing normal science to doing revolutionary science?

1. INTRODUCTION Over the past decade since leaving fulltime classroom teaching, I developed interests and conducted research that took me beyond my root discipline, science education including social studies of science, anthropology of the workplace, and linguistics (pragmatics). Working and publishing in these fields, I encountered theoretical frameworks, ways of relating to the research participants, and forms of scholarship that differ from our discipline. Upon coming back from time to time to my root discipline, I come to see it differently, see it struggling with issues that elsewhere have been settled. With more than a little concern, I frequently see my own discipline plodding along instead of engaging in efforts that change the world. In this chapter I hold up a mirror, thereby allowing the science education community (including myself) to look at itself. The need for change in science education practices has emerged for me particularly while researching controversy and environmentalism in one community (e.g., Roth & Lee, 2002), on the one hand, and while researching in urban schools where approximately 90% of the students are from home conditions of relative poverty (e.g., Roth et al., 2004), on the other. In the first instance, I came to realize that it is not necessary for every citizen to know how to balance a chemical equation, 3 K. Boersma et al. (eds.), Research and the Quality of Science Education, 3—14. © 2005 Springer. Printed in the Netherlands.

4

FROM NORMAL TO REVOLUTIONARY SCIENCE EDUCATION

recite the Krebs cycle, or use Newton’s third law to explain some phenomenon; rather, what we need are structures that allow citizens to solve problems and controversy in a collective manner. More important than everyone knowing scientific facts and concepts is that everyone, whatever his or her predilections, penchants, and beliefs, can participate in collective decision-making. In the second instance, I realized that science education contributes to reproducing an unjust, iniquitous, and inequitable society (Hein, 2004). More science education is continuously producing scientists who build weapons of mass destruction and work for ruthless multi-national companies that exploit a planet, which, as a proverb among the First Nations people on the Canadian Northwest Coast goes, we did not inherit from our ancestors but are borrowing from our children. What we therefore need is a discipline that goes beyond interpreting science teachers and students in various ways; the point of the existence of science education has to be the production of a better world. When existing paradigms cease to function adequately—for example, in the exploration of an aspect of nature—substantial change (revolutions) is in order (Kuhn, 1970). Because of the nature of science education as an applied discipline, substantial change may occur at three levels. First, I think that there is a need to revisit the theoretical frameworks we use to understand the world. Second, there is a need to revisit the way in and for which we prepare future science teachers. Third, there is a need to theorize the second issue in ways that lead to change so that it contributes to the production of a more reflexive and equitable society. In the remainder of this chapter, I present a framework that allows us not only to understand teaching and learning, but also to reflect upon our own actions and how these co-produce some of the phenomena we report in our journals. This framework has allowed us (my colleagues, students, and me) to bring about changes in the way we teach science teachers, the way science teachers teach in one school, and in the way students participate and take charge of their own learning. Most importantly, as I articulate below, this approach has led us to an active participation of university supervisors, teachers in training, science methods instructors, school administrators, and researchers in the teaching of students. This, readers will readily recognize, constitutes a substantial (revolutionary) departure from current practices in our discipline. I begin by briefly articulating the framework that allows us to theorize not only the phenomena of interest, but, much as quantum theory has done for physicists, also allows us to theorize how any observer participant mediates the production of data. I then use this framework to look at a range of activities in science education practice and research to show how they constitute a radical departure from what science educators have done in the past. 2. AGENCY AND STRUCTURE In many disciplines, researchers recognize the productive nature of human agency: not only do humans react to sociomaterial (including their own bodily) conditions,

W.M. ROTH

5

but they also produce and reproduce these conditions. Thus, neither the environment nor bio/psychological factors determine human actions. Such an approach necessitates a theoretical frame in which human subjects and their environment are related dialectically; that is, they mutually constitute one another rather than being dualistically opposed, as is currently the case in other psychological, sociological, and even discursive approaches.2 Though differently articulated and named, the dialectic of agency and structure is fundamental to approaches in many disciplines, including cultural sociology (e.g., Sewell, 1992) and cultural-historical activity theory (Leont’ev, 1978). Agency requires structure—without the human body, articulated as it is with all of its components, we cannot think of someone who acts; structure requires agency—any cognition requires the active engagement of an organism in a structured world (e.g., von Uexküll, 1973/1928). Structures come in two kinds: within the agent, there are schemas; in the environment, there are sociomaterial resources. (The predicated sociomaterial is used to approach social and material phenomena symmetrically.) The two are again dialectically related, for the schemas allow us to recognize environmental structures for what they are; but the structures in the environment have led to the formation of the schemas in the first place. This may sound like a chicken-and-egg situation, which would be difficult to explain in traditional logic. But such systems are as easy to explain in dialectical logic, or even in chaos- and catastrophe-theoretic approaches, where new, multi-state variants emerge as complex systems move through bifurcation points (e.g., Roth & Duit, 2003). Cultural-historical activity theorists provided a useful heuristic for identifying structure in an activity system (Engeström, 1987). This heuristic includes material structures in the form of tools and objects, on the one hand, and social structures in the form of communities with their rules and divisions of labor, on the other hand (Figure 1). Thus, scientists who take the genes of corn plants as their object of inquiry may produce not only genetically modified corn but also research articles. In this productive activity, they draw on a variety of means which, in fact, mediate the engagement with the object. The outcomes of the activity are intended for a particular community that consumes the product, and they therefore mediate the productive process. Interactions with the community and interactions with the object are mediated by rules, such as codes of ethics or appropriate scientific procedure. Within the research group, a division of labor mediates the different forms of engagement with the research object (e.g., as head of lab, lab technician, postdoctoral fellow); within the community, division of labor mediates, for example, the production of tools or the role of the individual subject in the community (i.e., someone who does genetic engineering of corn [DISTRIBUTION in Figure 1]). It is important to note that the subjects not only produce outcomes that are consumed within the community, but also they produce and reproduce themselves as members of the community.

6

FROM NORMAL TO REVOLUTIONARY SCIENCE EDUCATION

Figure 1. The heuristic for articulating the indivisible unit of an activity system: no part can be understood independently of the others and the relations that they mediate. For example, subjects do not directly relate to the object, but the subject-object relation is mediated by, for example, the tools or the division of labor.

Sociocultural and cultural-historical systems are far from equilibrium (e.g., Prigogine & Stengers, 1979). In dialectical approaches to social theories, disequilibrium is theorized in terms of contradictions which constitute the drivers of change and development (Il’enkov, 1977). Like physical systems that operate in a state of disequilibrium, human activity systems are unpredictable because of bifurcations along their historical trajectories. Because any change ripples through and affects the entire system, understanding what happens within an activity system at any point in time requires a study of its dynamics and history. Further, it does not suffice to study static structures (e.g., material or schemata) to understand actions of a system; understanding requires the study of the actions as a function of the entire system. The interesting aspect about dialectical approaches is that they recognize contradictions in their own theorizing as necessary drivers for theoretical development. They are not master theories or grand narratives, as one reviewer suggested, but tools for raising doubt, thus enabling one to become self-reflexive and self-aware without falling into the trap of solipsism. The dynamics in activity systems are described at three levels. Activities, such as researching or schooling, are characterized by collective motives; goals motivate the individual actions that concretely realize the activity when properly sequenced. Activities and actions are dialectically related, because actions constitute an activity, but activities guide the nature and sequence of actions; the relation between activity (motive) and action (goal) is called sense. Although directed towards conscious goals, actions are realized in practice by unconscious operations. The relation between action and operations is again dialectical, because actions (goals) provide a referent for the nature and sequencing of operations, but the operations constitute actions; the relation is called reference. The two relations, sense and reference, also stand in a dialectical relation called meaning. That is, one can speak of meaning only

W.M. ROTH

7

when there are ongoing processes; meaning is neither an attribute of texts and images, nor is it behind or underneath them. The dialectical nature of the activityaction and action-operation relations has substantive consequences. Any action will be associated with a different sense if it is employed in a different activity system; any operation will be associated with a different referent if the goal has changed.

3. CREATING UNDERSTANDING IN SCIENCE EDUCATION The framework outlined so far has considerable consequences for the way we see and go about research in science education. I articulate and discuss two examples: interviewing as a way of getting at beliefs and conceptions, and researching classrooms. A science education researcher sits together with a science teacher for the purpose of conducting an interview about teaching and teacher beliefs. The outcome of such an interview is usually a text, often produced from a mechanical recording (video, audio) and more seldom from handwritten notes made during the interview. The interview text is normally taken as a data source for getting at teachers’ beliefs or ideas about teaching, which are taken to have a direct bearing on what a teacher does in the classroom. A framing within a dialectical approach shows us that the interview text is inherently the product of a system and therefore cannot be attributed to the interviewed teacher alone. Here, the activity is “interviewing for research purposes”; the motive is understanding or theorizing teaching. Any action, such as an uttered sentence (a speech act [Hanks, 1996]), is related to the motive of the activity; the sense associated with the action depends on the activity-action relation. Thus, anything a teacher says is uttered in relation to the motive, interviewing about beliefs and teaching, not with respect to the praxis of teaching. Whatever the outcome, it is not for teaching and its community, but for science educators and their community (Figure 2). The means drawn on in the production of the interview (e.g., belief discourse, talk about science pedagogy) are very different than the tools drawn on in teaching (e.g., enacted science pedagogy, physics discourse [about atoms]). Because actions enter a referential relation to operations, the latter will likely be different too, including, for example, the gestures and body positions, the stance, the signs of confidence produced.3 That is, the interview text cannot be ascribed to the interviewee; it bears all the marks of the entire system which it therefore reflects.

8

FROM NORMAL TO REVOLUTIONARY SCIENCE EDUCATION

Figure 2. In an interview about teaching and teaching beliefs, the researcher and teacher co-produce the interview text, which therefore cannot be understood as something that signifies the teacher alone. The text reflects the system as a whole, including teacher, researcher, community, tools, and the object (of inquiry).

Sociologists know very well that interviews are co-produced even when they are conducted under the most rigorous conditions in the production of preformatted questionnaire answers (e.g., Suchman & Jordan, 1990), and the relationship of anything said in an interview to what the practice is about has to be established empirically and cannot be taken for granted. Even practitioners have no better insight into their practice than theorists (e.g., Bourdieu, 1990). It is not surprising then that interviewees (e.g., scientists, engineers, teachers) often contradict themselves within minutes during the same interview when they describe or explain something from their practice. It would be much more challenging and much more scientific if science educators were explaining strong coherence (consistency of actions across contexts) than mocking themselves, as they have in the past, about any observed weak coherence (between beliefs and actions). Strong (thick) coherence is the exception; thin coherence, the rule (Sewell, 1999). In much of science education research, the independence of observer and the observed phenomenon is taken for granted. Researchers think of themselves as being able to act like flies on the wall or that they can be objective recording instruments. Thus, they observe classrooms or simply take recordings others have made (e.g., in the many analyses of TIMMS videos) and write research reports, destined for the community of science educators. They are then astonished that whatever they do and think has little or no bearing on science classrooms—during ESERA 2003, I overheard several different conversations to this effect. Having been a teacher for many years and having continued to teach with them over the past decade, I am not surprised by teachers’ distrust of researchers. Many teachers do not like to have researchers in their classrooms; some feel threatened, while others hate the disruptions that this might cause. If science educators truly want to contribute to classroom teaching, they have to change substantially (radically) the way they do business. Taking the theoretical lens that I propose here, the situation between teachers and researchers should not be astonishing. Science education researchers who study teaching and learning take science classrooms as their objects of inquiry (Figure 3); they record the events or take field notes, and subsequently, after having searched

W.M. ROTH

9

Figure 3. The researcher makes a classroom his/her object of inquiry, thereby objectifying the classroom, its events, participants, and objects. The products of research are consumed in a community with which the stakeholders in the school (teachers, students, parents, administrators) do not identify.

for patterns and themes, write articles about them that are shared with and read in the science education community. The literary genres used are those accepted by and consumed in the research community. In taking classrooms as their objects of inquiry, researchers objectify the classroom, the people that inhabit and populate it, and the events that occur in it. Teachers are not true participants; they are therefore not part of the subject in the research triangle (Figure 3). Rather, teachers find themselves in a subject-object relationship; typically, they might indicate that they do not want to be “lab rats,” research objects in another field. The outcomes of the research activity systems are intended for other science educators. It therefore comes as no surprise that teachers and students think that these researcher-oriented texts have little to say to them and that they are not useful tools for teaching and learning. Again, I would expect this because cultures are characterized by thin rather than thick coherence. 4. THE POINT IS TO CHANGE SCIENCE EDUCATION In this chapter, there is insufficient room to account for the trajectory that has led my colleagues and me to a radically different research and teaching practice—such accounts have been provided elsewhere (e.g., Roth, 2000; Roth & Tobin, 2002). It was an arduous and sometimes painful journey, contradictions moving us continuously ahead as we pitted our existing theories against our practice of teaching in some of the most difficult, urban schools of the United States. Here I simply sketch how research and teaching are intertwined in one U.S. urban high school that serves more than 2000 students, predominantly from (extreme) poverty. Our work is based on two practices that stand in a dialectic relationship: co-teaching | cogenerative dialoguing (Roth, 2002).

10

FROM NORMAL TO REVOLUTIONARY SCIENCE EDUCATION

Understanding activity, as evident from the exposition above, requires participation in the productive process—the kind of and reasons for choices made in some fields are only apparent to the practitioner oriented to the object (Bourdieu, 1990). Thus, co-teaching involves all non-student adults present in the science classroom and sometimes even some students (Figure 4, shaded system). Universitybased researchers and methods professors, and school-based supervisors and administrators are no longer allowed simply to watch, as if they were flies on the wall, and then sit in judgment over (write reports about) classroom processes, teachers, and students. This, most readers will recognize, is a radical (revolutionary) departure from current practice where university-based researchers and teacher educators pronounce judgments from on high. In our situation, however, everyone present contributes as a co-teacher to the teaching and attempts to address any problematic issue as it arises; therefore, the teacher collective does its best to support student learning. Teaching is a collective responsibility. Directly after class or after school—the frequency depending on the particular situation and the sociomaterial constraints—all participating teachers and students (or student representatives) meet to analyze what has happened. The motive of this form of activity is to generate theory and plans of actions to improve classroom events. As its name indicates, co-generative dialoguing is designed to have all participating stakeholders contribute to the generation of theory and action plans. For such plans to be useful, all stakeholders need to have the sense that they are in control of the object, the production means, and therefore the outcomes. As Figure 4 shows, the participants in co-generative dialoguing are the same as those in coteaching; however, the division of labor has changed. While in co-teaching there is a division between students and teachers, co-generative dialoguing provides all participants with equal opportunities and power for generating understanding, and plans of actions—in this way, some students in fact become teachers and some teachers (including university professors) become learners (e.g., Roth, Tobin, Zimmermann, Bryant & Davis, 2002). The outcomes of these co-generative dialogues, understanding and action plans are intended for classroom use by the same participants, and therefore have a much higher likelihood to lead to (lasting) change than in traditional science education. Rather than telling administrators, teachers, and students how to improve their practices, we engaged with them in trying to understand the events and contributing our little bit to help improve the situation. Interestingly enough, this model has been shown to be an extremely effective environment for learning to teach. Not only do new teachers (in training) learn to teach as they co-teach with more experienced others, but veteran teachers also improve their practices while co-teaching with less experienced teachers and even novices. It is immediately evident that in this generative model, science teachers in training and university-based supervisors and science methods teachers become resources for science teaching and learning in elementary and high schools. In the process they not only reproduce themselves, they also become better science teachers.

W.M. ROTH

11

Figure 4. The primary purposes of co-teaching | co-generative dialoguing are learning and teaching and the improvement of the learning environment. The outcomes of the collective, sense-making processes are fed back into the classroom.

It is evident that this practice constitutes a substantial, even radical or revolutionary departure from the predominant practice in science education in which professors and instructors generally do not contribute to science teaching in schools and often are far removed from what is workable in science classrooms. The talk on both sides of the school-university divide concerning ivory towers and real classrooms, or the theory-practice gap, provides ample evidence of a chasm. In our research approach, we see all university-based individuals and school administrators as important resources for the events in classrooms, resources that are currently under-utilized; and we see this form of practice as an important way not only to describe and understand but also to bring about real and lasting change in science education. Just imagine all professors, instructors, and science teacher aspirants contributing to teaching and learning in schools rather than talking about it in university lecture halls! This does not preclude our continued contribution to a literature for other university-based science educators, with its own genres; but the data have been coproduced while contributing to the activity that we all claim our own actions to be about: teaching and learning science in the classroom. 5. CODA In this chapter, I described a theoretical approach that is very different from those currently used in science education; in effect it questions the products of much of existing research, which does not account for the fact of its own contribution to the research outcomes. I also described a very different conception of science teaching and learning which requires researchers, science teacher educators, and science

12

FROM NORMAL TO REVOLUTIONARY SCIENCE EDUCATION

teachers in training to contribute actively to science teaching. Instituting such activity more broadly would constitute nothing less than a revolution. The theoretical framework allows us to understand in new ways many problems science educators face. For example, the “cook book” labs high school students conduct do not work because students follow steps, that is, implement operations without knowing or understanding the goals of the actions thus constituted. Furthermore, without an understanding of how these actions relate to the motive of activity (schooling), and therefore without the experience of sense, there is little chance that these laboratory tasks lead to anything that resembles science learning on a broader scale. Here, the analysis begins with the identification of a contradiction that, once recognized, can be used to drive change and development. The theoretical and practical approaches offered here not only are consistent but also lead to substantial change. Shifting from current ways of doing and thinking about science education research and practice to those proposed here requires nothing short of a radical change, a revolution. However, prerequisite to any revolution is a sense of malfunction and crisis, a sense “that existing institutions have ceased adequately to meet the problems posed by an environment that they have in part created” (Kuhn, 1970, p. 92). This growing sense is often restricted to a segment of the community. I intend this chapter as a mirror for other science educators and myself and point out some contradictions in our discipline. Consistent with my framework, I do not despair because I see contradictions and inconsistencies as opportunities for change, development, and growth. Change does not come easily, for “like the choice between competing political institutions, that between competing paradigms proves to be a choice between incompatible modes of community life” (p. 94). In the spirit of the power of collective action, I suggest that we engage together to bring about what now may look like revolutionary changes in the way in which we go about our daily business as science educators. The most important issue is this: the point of science education is change to make this a better, more just and equitable world. ACKNOWLEDGMENTS I am grateful to Kenneth Tobin, the members of the working group on culturalhistorical activity theory at the University of Victoria (Diego Ardenghi, Leanna Boyer, Damien Givry, Marines Goulart, JaeYoung Han, Michael Hoffmann, SungWon Hwang, Yew Jin Lee, and Lilian Pozzer-Ardenghi), and two reviewers for their helpful comments in revising an earlier version of this paper, in particular, in assisting me to find the right tone. ENDNOTES 1. Because the reviewers of this chapter made comments such as “Rather high-flown about revolutionary changes which would be necessary. Some relativisation and

W.M. ROTH

13

modesty would be better, but this is a question of personal taste”, or “The theoretical frame does not offer anything new and is not necessary for what is propagated”, I expect other readers to react in a similar way. 2. One reviewer suggested that there is nothing new to such a perspective. But in fact, a dialectical perspective on social-psychological phenomena is radically different from all other approaches (Engeström, 1987) that dichotomize individual and society or culture, intra- and inter-psychological phenomena, the subject and its object of action, and so forth. 3. Again, one reviewer criticized me on this point, suggesting that science educators have been aware of this. He (or she) wrote, “depending on the kinds of information sought, an interview may yield valid data—and triangulation with other data sources can increase validity further”. I am writing not about lack of validity but about the collective nature of interview texts, reflecting both interviewer and interviewee. Furthermore, triangulation does not make sense if particulars of the (changing) situations change the outcome, that is, the interview text. REFERENCES Bourdieu, P. (1990). The logic of practice. Cambridge, UK: Polity Press. Engeström, Y. (1987). Learning by expanding: An activity-theoretical approach to developmental research. Helsinki: Orienta-Konsultit. Hanks, W. F. (1996). Language and communicative practices. Boulder: Westview Press. Hein, G. H. (2004, January/February). Museum-school bridges: A legacy of progressive education. ASTC (Association for Science-Technology Centers) Dimensions, pp. 6–7. Il’enkov, E. (1977). Dialectical logic: Essays in its history and theory (Transl. by H. Campbell Creighton). Moscow: Progress. Kuhn, T. S. (1970). The structure of scientific revolutions (2nd ed.). Chicago: The University of Chicago Press. Leont’ev, A. N. (1978). Activity, consciousness and personality. Englewood Cliffs, NJ: Prentice Hall. Prigogine, I., & Stengers, I. (1979). La nouvelle alliance: Métamorphose de la science. Paris: Gallimard. Roth, W-M. (2000). Learning environments research, lifeworld analysis, and solidarity in practice. Learning Environments Research, 2, 225–247. Roth, W-M. (2002). Being and becoming in the classroom. Westport, CT: Ablex. Roth, W-M., & Duit, R. (2003). Emergence, flexibility, and stabilization of language in a physics classroom. Journal for Research in Science Teaching, 40, 869–897. Roth, W-M., & Lee, S. (2002). Scientific literacy as collective praxis. Public Understanding of Science, 11, 33–56. Roth, W-M., & Tobin, K. G. (2002). At the elbow of another: Learning to teach by coteaching. New York: Peter Lang. Roth, W-M., Tobin, K., Elmesky, R., Carambo, C., McKnight, Y., & Beers, J. (2004). Re/making identities in the praxis of urban schooling: A cultural historical perspective. Mind, Culture, & Activity, 11,

14

FROM NORMAL TO REVOLUTIONARY SCIENCE EDUCATION

Roth, W-M., Tobin, K., Zimmermann, A., Bryant, N., & Davis, C. (2002). Lessons on/from the dihybrid cross: An activity theoretical study of learning in coteaching. Journal of Research in Science Teaching, 39, 253–282. Sewell, W. H. (1992) A theory of structure: duality, agency and transformation, American Journal of Sociology, 98, 1–29. Sewell, W. H. (1999). The concept(s) of culture. In V. E. Bonnell & L. Hunt (Eds.), Beyond the cultural turn: New directions in the study of society and culture (pp. 35–61). Berkeley: University of California Press. Suchman, L. A., & Jordan, B. (1990). Interactional troubles in face-to-face survey interviews. Journal of the American Statistical Association, 85, 232-244. Uexküll, J. von (1973). Theoretische Biologie [Theoretical biology]. Frankfurt: Suhrkamp. (First published in 1928)

REFLECTIONS ON A PROBLEM POSING APPROACH

PIET LIJNSE Utrecht University, The Netherlands

ABSTRACT This paper describes some general aspects of the problem posing approach, as developed at the CSMEU. It describes why this approach has been developed; what didactical problem it tries to focus on; from what perspective this is done; to what didactical structures such an approach may lead, and what its application may involve for a teacher. The arguments are endorsed by examples taken from recent PhD work, but placed within a wider perspective.

1. INTRODUCTION In the recent past, much work has been done on the cognitive aspects of science learning, e.g., by developing and studying exemplary teaching sequences (Méheut & Psillos, 2004). However, Leach and Scott (2002) argue that in the latter work not enough attention has been given to the role of the teacher. Others focus on the role of motivation for science learning, while Osborne (this volume) emphasizes the importance of adequate scientific argumentation. This paper deals with a line of work at the CSMEU in which all these aspects more or less come together, i.e., the development of what we call a problem posing approach to science education. It addresses some small steps forward in our didactical insight, as this is the most that can be expected from science education research. The origin of this approach lies in our work on curriculum development, i.e. the former PLON-project (Lijnse et al., 1990). This project had a major influence on contextualising Dutch physics education, though its cognitive learning effects were not as positive as expected. In retrospect, we may say that we overestimated the positive influence of contexts on conceptual learning, particularly as far as the experienced functionality of the concepts to be learned is concerned. A main problem was that, though we did our utmost to make the contexts used relevant for our students, due to our mostly top down didactics, from their point of view students often got the idea that they had now to describe more or less familiar life-world contexts in a – for them – strange way of physics. Since then, we have been looking for ways to improve the quality of our didactical approach. We have done this by means of developmental research (Lijnse, 1995, 2003) which nowadays has probably become better known as design research (Cobb et al., 2003), i.e., 15 K. Boersma et al. (eds.), Research and the Quality of Science Education, 15—26. © 2005 Springer. Printed in the Netherlands.

16

REFLECTIONS ON A PROBLEM POSING APPROACH

developing, testing, and reflecting on actual teaching/learning processes in order to come to new didactical insights and theory. In fact, this connects in some way to much other research that has been done on developing research-inspired improved ways of teaching 2. WHAT IS THE PROBLEM? In the final decades of the last century, extensive reports were published on all kinds of conceptual problems that students appeared to have with the learning of science. In relation to these conceptual difficulties, other problems were also reported that have more to do with the way students perceive the detailed teaching/learning process. It appeared that during the process of teaching and learning, very often students do not see the point of what they are actually doing. This was not only the case in our context-related teaching, but it also applies, e.g., to the relation between theory and experiment as reported by Joling et al. (1988) who concluded, in an evaluation report about an innovative teaching method in a chemistry classroom, that students “carry out assignments without knowing what function they have. The relation between observations and conclusions becomes blurred due to a lack of purposiveness in the experiment”. However, the problem is much more general. To give another example, Gunstone (1992) reported as follows: “In the following typical example, the student (P) has been asked by the interviewer (O) about the purpose of the activity they have just completed. P: He talked about it……..That’s about all….. O: What have you decided it [the activity] is all about? P: I dunno, I never really thought about it …. just doing it – doing what it says its 8.5 …. just got to do different numbers and the next one we have to do is this [points in text to 8.6].”

In addition, Gunstone (1992) writes: “This problem of students not knowing the purpose(s) of what they are doing, even when they have been told, is perfectly familiar to any of us who have spent time teaching. The real issue is why the problem is so common and why it is very hard to avoid.” Now, in our approach we do not focus on explaining this problem, but on trying to find ways to avoid it. The commonality that Gunstone mentions, reflects an often occurring mismatch between the ways teachers and students perceive the teaching/learning process. In the teaching situation referred to by Gunstone, the teacher probably had a coherent conceptual pathway in mind, and thus also perceived his/her teaching activities as coherently aiming at a certain purpose, but from the point of view of the students this coherence broke down to separate learning activities that had to be worked through according to their number. Some of them may have been understandable, but others too difficult, thus blocking an experience of coherence and purpose. In our experience, this is not really amazing as, in spite of their perception, teachers often teach separate activities according to their number, i.e., they teach subsequent activities without relating them to one another. In such cases it is clear that students may wonder what they are supposed to

P. LIJNSE

17

do, as they do not (and cannot) experience and perceive the activities in the intended way. One could symbolise this as follows (Figure 1). Teacher’s perception of the teaching process

Teaching/learning activities

Students’ perception of the learning process Figure 1. The teaching process consists of subsequent activities that are not explicitly related to one another. Nevertheless, the teacher perceives his/her teaching as coherent and aiming at a certain purpose, while the students perceive the activities as largely non-related, more or less in a non-intended way, and with an unclear purpose. 3. OTHERS’ SOLUTIONS In the literature, we have seen many efforts to remedy conceptual shortcomings, most of them from a more or less constructivist perspective (Scott et al., 1992). A first step that constructivists often advocate involves that we should start conceptually from where students are and stimulate students’ ‘deeper’ thinking during the respective teaching/learning activities, e.g., by asking ‘deeper’ qualitative, conceptual questions. As a consequence, it may become clearer to students, what they are supposed to learn from a particular activity with the result that there is less conceptual confusion. Teachers’ perception of the teaching/learning process

Students’ perception of the teaching/learning process Figure 2. The curved arrows indicate activity-related deeper-thinking questions, which are to be used and monitored by the teacher (indicated by the vertical arrows), resulting in improved conceptual understanding and sometimes in some implicitly perceived backward coherence (as indicated by the two horizontal arrows).

18

REFLECTIONS ON A PROBLEM POSING APPROACH

However, such measures do not necessarily result in a sense of purpose, even though, implicitly, they may foster the perceived coherence between activities. The teacher must, of course, make sure that these deeper questions lead to the required understanding of the respective tasks. However, in examples from the literature that I have come across, it still seems that a teacher is not supposed to pay much explicit attention to activities' mutual coherence or to a sense of purpose (Figure 2). Others have, in addition, advocated the importance of paying explicit attention to more general aspects of meta-cognition. Students should learn to learn actively and cooperatively and to show good ‘learning behaviours’, i.e. to take responsibility for their own learning processes. An example of this is in the Australian PEEL project in which students were taught to ask reflective ‘self-questions’ like: How does today’s lesson connect with yesterday’s lesson?, Are there any new ideas today?, Am I clear about what I have to do? At the same time, teachers developed teaching strategies to foster such ‘quality learning’ (Figure 3). However, such procedures and strategies that aim at making students more aware of the quality of their learning, appeared not at all easy for teachers.

Figure 3. Now also explicit teaching activities (longer vertical arrows) are being used to stimulate the experience of a backward coherence between activities for students (horizontal arrows). 4. OUR APPROACH In light of our indicated problem, we think the approaches just described to be insufficient since they may lead to a backwardly experienced coherence but not to a forward-looking sense of purpose. Therefore, we have adopted the additional view that on content related grounds during students' learning process, it should, as much as possible, be clear to them why they are doing something and where it should be leading them. More precisely, students should at any time during their learning process be able to recognize the content-related point of what they are doing. We

P. LIJNSE

19

think that if this is the case, the process of teaching and learning will probably be more meaningful to them, and it then becomes more probable that they will construct or accommodate the required new knowledge on grounds that they themselves understand. An approach to science education that explicitly aims at this is called problem posing by us. In our problem posing approach, guided by the teacher, we want the students as much as possible to frame themselves or at least value, the problems that they will work on, in contrast to just solving a problem as put to them by the teacher (Taconis et al., 2001). The emphasis of a problem posing approach is thus on bringing students to such a position that they themselves come to experience a content-related sense of purpose and come to see the point of extending their existing conceptual knowledge and experiences in a certain direction, i.e. in the direction of the concepts to be taught. Thus formulated, this starting point seems rather trivial and hardly new at all, and indeed it is. Since in itself such a starting point doesn’t give any further detailed, didactical guidance, the real non-trivial didactical challenge lies in the quality with which it can be put into practice. Further, the challenge to such an approach is that it does not only ask for a considerable change in didactical contract, as compared to that which teachers and students are mostly accustomed; it also requires teaching activities which are as much as possible structured and formulated bottom-up, i.e., from the point of view of understanding, coherence, and purpose for students. In fact, one could say that in our approach we want students, guided by the teacher, to walk as much as possible on, what is for them, an explicit, rational, and meaningful pathway of questions and answers that eventually leads them to the concepts to be taught. Put in this way, it will be clear that our approach involves in principle nothing new, even though it appears to be rather difficult to put into practice. In fact, one could say that this approach includes naturally, a content-related ‘good argumentation’ viewpoint (Osborne, this volume). For fostering meaning-generated learning, we should make a distinction between seeing the point of something and liking that point; or, in other words, between having a motive for doing something and being motivated to do it. Much work has been done on the role of motivation in education, for example, Boekaerts (2002) writes: “By organizing learning situations in such a way that students are always encouraged to begin the learning process by generating learning goals from their own goal hierarchy, teachers allow their students to experience situational meaningfulness”, because “students who engage in meaning-generated learning, experience positive effect”. Therefore, Boekaerts pleads for more attention to socioemotional goals, as: “personal goals give meaning and organization, or in other words purpose, to a student’s adaptation processes in the classroom”. Examples of such personal goals, as given by Boekaerts, are: “be successful”, “be respected”, “make many friends”, and so on. Without arguing about the value of this position, we may expect that the learning of scientific content matter will not easily be perceived by students as personal goals of such a kind. Therefore, we do not try to relate to such general goals, but rather we focus on finding a way to engage students in meaning-generated learning by making them have content related motives for

20

REFLECTIONS ON A PROBLEM POSING APPROACH

learning some topic and its concepts, which should enable them to experience the teaching process as coherent, useful, and possibly also more interesting. Apart from a very careful design and outline of the detailed teaching activities from a student’s perspective, we have tried to achieve this ideal in the following way. First, we develop with students a means that allows them to look forward and that at the same time may serve as a means to monitor ‘how far we already have come’. This is done by starting with a global orientation in which a global motive is developed for the topic under concern that should enable students to have the required ‘sense of purpose’. From this global motive, a storyline is developed, e.g., by splitting the global motive up into several local motives that are developed bottom-up at appropriate places during teaching, e.g., by encouraging students to ask, value, or reflect on questions that have been worked on in previous activities, or which will be worked on in future activities. In fact, we have now developed several teaching/learning sequences from this perspective, which has led to the emergence of a certain pattern that, in our experience, also has prescriptive value as a heuristic for the design of new teaching sequences, including appropriate teacher preparation (Lijnse, 2000; Lijnse and Klaassen, 2004). To make myself more clear, let me show you an example of such a storyline as developed by Kortland (2001). He tackled the didactical problem of how to teach the ‘general skill’ of decision making, being formulated as being able to present an argued point of view, in relation to teaching about the environmental waste issue. Kortland developed a problem posing teaching sequence for 14 year old lower ability students which can be summarised in a didactical structure (Figure 4). This 10 lesson sequence focussed on the question of how to deal best with household package waste, from an environmental point of view. After an orientation on personal decision making about household waste, at the level of using both life-world knowledge and intuitive decision making, students come to the recognition that they first need to know more about household package waste. In this phase, students’ pre-knowledge is activated, structured, and productively used for formulating a knowledge need. Then, after having acquired and applied this knowledge in situations that ask for decision making and about which they have to present their point of view, they come to realise that it is not obvious at all what it means to present a ‘well argued’ opinion. As a consequence, in this phase a need emerges for some ‘norms’. Thus, in reflection, a (still contextualised) number of heuristic rules are made explicit and used, that help students to structure and check their reasoning. The resulting pattern (mentioned above) is illustrated in the general structure of Figure 4. In designing a teaching sequence, one should clearly establish and distinguish its main independent objectives on which one wants to work. Then for those main objectives, teaching/learning pathways are designed that start from where students are and lead in a bottom-up way to the intended end points. In the design, a central problem posing feature is the idea to intertwine these pathways in a ‘natural way’ for students via motives that are to be developed during the teaching process.

P. LIJNSE

21

Another aspect of this general pattern is that in the didactical structure the following phases can be distinguished, which relate to particular didactical functions that have to be fulfilled in such a way that they assure the necessary coherence and purpose in the activities of students. • Phase 1: Orienting and evoking a global interest in and motive for a study of the topic at hand. • Phase 2: Narrowing down this global motive to a content-specific knowledgeneed. • Phase 3: Extending students’ existing knowledge in view of the global motive and the more specifically formulated knowledge-need. • Phase 4: Applying this knowledge in situations for which the knowledge was meant. • Phase 5: Creating, in view of the global motive, a need for a reflection on the skill involved. • Phase 6: Developing a (possibly still contextualised) meta-cognitive tool for an improved performance of this skill. We remark that phases 2 and 5, consisting of creating relevant needs, represent one of the main points of a problem posing approach. Such phases are not present in teaching cycles as published in the literature (Abraham, 1998). Those cycles almost exclusively deal with cognitive learning, even though it is also often written that one should not forget about the importance of motives. In our approach, however, in some sense both cognitive learning and motives are taken together and integrated from the start. 5. OUR EXPERIENCES The teacher’s role Let me now focus on the didactical role of the teacher in this approach, as it has turned out that this role is not at all easy to fulfil. In fact, the teacher has two main content-related roles. In the first place, the didactical task at the conceptual level involves a change with respect to ‘traditional teaching’. The teaching has to be bottom-up, i.e., students have to have more opportunities either to be guided to ask their own questions or to value those brought forward by the teacher, and to follow their questions up by investigating and discussing their ideas, though within the intended sense of direction and purpose. A main didactical problem is, thus, how to set students initially on the right track. As regards this bottom-up character, a trial school teacher noted: "…I more often try to get into the skin of the pupils……. It has already yielded fruit (still to be seen whether it is ripe) in my daily teaching practice. Holding back, listening to pupils, adjusting a little later. A changed attitude with regard to pupils’ making notes of observations. Less direct explaining.” But also: “In fairness I have to tell that teaching in this way, with ‘holding back’ and ‘listening’, does require quite an effort. After these lessons I generally was more tired than after lessons taught in my old way. The question then presents itself whether that additional effort balances the achieved result. I do give this question a cautious ‘yes’ though.

22

REFLECTIONS ON A PROBLEM POSING APPROACH

ISSUE KNOWLEDGE

MOTIVE

a global orientation on environmental issues

DECISION-MAKING SKILL that asks for decisions to made

should result in a feeling that one could contribute to ‘a better environment’, if one knew more about the topic starting by focusing on general

on which is reflected in

knowledge about the (exemplary)

terms of known environmental criteria for decision making

packaging waste issue

resulting in a recognition that more specific, criteria-related issue knowledge is required operationalized in questions that ask for answers by means of investigations that result in the necessary knowledge

to be applied in appropriate decisionmaking situations resulting in a recognition that the presentation of an argued point of view asks for a reflection in terms of developing and making explicit a decision-making procedure (content and presentation standards)

leading to the expectation that such a procedure could also be useful in other environmental decision making provided that adequate issue knowl-edge can be obtained

Figure 4. A didactical structure for a problem posing approach about decision making on the waste issue. One could say that this comment largely resembles what is known about teachers’ experiences with ‘constructivist teaching approaches’. It has to do with

P. LIJNSE

23

giving students more construction space and thus more responsibility for their conceptual learning, which shifts the role of the teacher towards more guidance and procedural control. However, in our work this procedural control gets an extra dimension, as the teacher also has to make sure that students connect their learning experiences with the to-be-developed local and global motives; or, in other words, making sure that students experience the teaching/learning process as coherent and maintaining their sense of purpose and direction. The teacher, therefore, also has to monitor and guide the teaching/learning process at what could be called a meta-didactical level (Figure 5).

Global motive

Local motives

Figure 5. This figure shows that during the teaching process, the teacher has to develop global and local motives and has to monitor the teaching/learning process regularly in view of these motives in order to establish for students the intended coherence and sense of purpose. In practice, it appeared that this can best be done while rounding off previous and starting a new series of activities. So, the teacher must regularly focus on questions like: ‘how far have we come in answering our main questions’, ‘what problems did we already solve and which ones remain that we still have to work on’, ‘what new questions result from the foregoing’, etc. It is our experience that precisely this meta-didactical activity appears to be rather unusual and difficult for teachers. We call it a meta-level activity as it involves a reflection on the outcomes of the didactical process so far, resulting, if necessary, in restoring the results of poor previous didactical activities. Though such a reflection at the conceptual level is not uncommon for teachers, it is the relationship to motives that is rather new and difficult for them. In our experiments, the fact that trial school teachers were not always sufficiently able to deal with this meta-level appeared to have direct negative consequences for the experienced coherence and problem posing character of the teaching/learning process involved.

24

REFLECTIONS ON A PROBLEM POSING APPROACH

Low-level One of the reasons for this difficulty for teachers lies in the fact that our approach often involved cooperative reflections which we had mostly organised in plenary classroom discussions. In fact, this turned out to be a weak point of more general importance. Reflective classroom discussions are not only difficult to handle for most teachers, but also often for students, particularly if the preceding learning process has resulted in too much cognitive diversity. So, we are now trying to find ways to decrease the emphasis on classroom discussions by replacing them with other, easier methods to stimulate reflection among students. This is part of an effort to develop approaches that remain to an essential extent problem posing, but are easier to apply for teachers and thus easier to implement. Successful? In general, we found the development of teaching sequences according to the gradually emerging concept of a problem posing approach a rewarding research experience, but unfortunately also a rather difficult one, even for experienced researchers and teachers. This appears to be caused by a difficulty in acting and thinking really from a student’s perspective. We are always inclined to overestimate the clearness of what we say and write (including this paper). So, you might ask, has this effort been worthwhile? In fact, we have not yet done any comparative research, as we thought it to be more important to see to what extent we could succeed in reaching our aims by first designing and developing an ‘as-intended-going’ teaching/learning process. Therefore, we followed the actual teaching/learning process in our experiments to a rather detailed level, to see to what extent this process developed indeed as predicted, and thus, also to what extent it could be considered as sufficiently problem posing. In general, we succeeded to a reasonable degree in making students see the point of what they were doing, though further improvement is certainly required, particularly as many details, such as the specific order and wording of the tasks, and the extent to which the teacher was able and willing to follow the intended teaching process appeared to be rather critical. 6. ANYTHING NEW? But of course, you may be more critical and ask: does your approach actually involve anything new? As already indicated, that depends very much on the level of didactical detail with which you want to look at it. Some time ago, Robin Millar wrote me in an e-mail: ”Personally, as far as I understand it, I am not persuaded that the problem-posing approach is significantly different from most good practice in science education. I see little difference between trying to engineer matters so that pupils appear 'spontaneously' to pose a question, and the teacher proposing the question, provided he/she makes an effort to ensure that the pupils really understand the question before he/she starts telling them the answer (or trying to lead them towards the answer). Both are ways of achieving the same end: ensuring that the pupils really understand the question the teaching is about.”

P. LIJNSE

25

Though we hope that it has become clear that our approach does indeed involve a lot more than only making sure that students do really understand the teacher’s question, in principle we agree to a large extent with Millar’s point of ‘little difference’. However, we want to stress that, in saying so, he is referring to ‘good practice’. And how common is that good practice? We would conjecture, on the basis of our experience, that such ‘good practice’ is not at all common practice. In making this ‘good practice’ systematically the heart of our approach we want to support teachers in providing them with means that enable them to come closer to the intended good practice, and thereby, making them aware of the need for and helping them to improve the didactical quality of their work. As a trial school teacher who taught one of our teaching sequences wrote: “Does all this mean that in your physics classes you will tomorrow be able to begin with ‘holding back, postponing questions, letting pupils themselves think of experiments, carry those out and note down conclusions’? The answer to this question is a very distinct: no! Material that makes possible such an approach for other topics simply isn’t available. But if such material is to come, and it isn’t up to me to take care of that, I will surely use it.” REFERENCES Abraham, M.R. (1998). The learning cycle approach as a strategy for instruction in science. In B.J. Fraser & K.G. Tobin (Eds.) International Handbook of Science Education (pp.513524). Dordrecht: Kluwer. Boekaerts, M. (2002). Bringing about change in the classroom: strengths and weaknesses of the self-regulated learning approach. Learning and Instruction, 12, 589-604. Cobb, P., Confrey, J., diSessa, A., Lehrer, R. & Schauble, L.(2003). Design Experiments in Educational Research. Educational Researcher, 32(1), 9-13. Duit, R. & Treagust, D.F. (1998). Learning in science – From behaviourism towards social constructivism and beyond. In B.J.Fraser & G. Tobin (Eds.) International Handbook on Science Education (pp.3-25). Dordrecht: Kluwer. Gunstone, R. (1992). Constructivism and metacognition: theoretical issues and classroom studies. In R. Duit, F. Goldberg & H. Niedderer (Eds.), Research in Physics Learning: Theoretical Issues and Empirical Studies (pp.129-140). Kiel: IPN. Joling, E., van Lierop, A., van Soest, W., Kaper, W., ten Voorde, H.H., de Vos, W., Mellink, E., Snel, B. & Timmer, J. (1988). Chemie mavo: onderzoek naar het functioneren van een leergang scheikunde. Amsterdam: SCO. Klaassen, C.W.J.M. (1995). A problem posing approach to teaching the topic of radioactivity. Utrecht: CD-β Press. Kortland, J. (2001). A Problem Posing Approach to Teaching Decision Making about the Waste Issue. Utrecht: CD-ß Press. Leach, J. & Scott, P. (2002). Designing and Evaluating Science Teaching Sequences: An Approach Drawing upon the Concept of Learning Demand and a Social Constructivist Perspective on Learning. Studies in Science Education, 38, 115-142. Lijnse, P.L., Kortland, J., Eijkelhof, H.M.C., van Genderen, D. & Hooymayers, H.P. (1990). A thematic physics curriculum: a balance between contradictory curriculum forces. Science Education, 74, 95-103.

26

REFLECTIONS ON A PROBLEM POSING APPROACH

Lijnse, P.L. (1995). ‘Developmental research’ as a way to an empirically based ‘didactical structure’ of science. Science Education, 79, 189-99. Lijnse, P.L. (2000). Didactics of Science: the forgotten dimension in science education research? In R. Millar, J. Leach & J. Osborne (Eds.), Improving science education – The contribution of research (pp.308-326). Buckingham: Open University Press. Lijnse, P.L. (2003). Developmental Research: its aims, methods and outcomes. In D. Krnel (Ed.), Proceedings of the 6th ESERA Phd-Summerschool. University of Ljubljana. Lijnse, P.L. & Klaassen, C.W.J.M. (2004). Didactical structures as an outcome of research on teaching-learning sequences? International Journal of Science Education (in press). Méheut, M. & Psillos, D. (2004). Teaching-Learning Sequences. International Journal of Science Education (in press). Scott, P.H., Asoko, H.M. & Driver, R.H. (1992). Teaching for conceptual change: a review of strategies. In R. Duit, F.Goldberg & H. Niedderer (Eds.) Research in Physics Learning: Theoretical Issues and Empirical Studies (pp.310-329). Kiel: IPN. Taconis, R., Ferguson-Hessler, M.G.M. & Broekkamp, H. (2001) Teaching science problem solving: an overview of experimental work. Journal of Research in Science Teaching, 38, 442-468. Vollebregt, M.J. (1998). A Problem Posing Approach to Teaching an Initial Particle Model. Utrecht: CD-β Press.

HOW CAN LARGE INTERNATIONAL COMPARATIVE STUDIES CONTRIBUTE TO THE QUALITY OF SCIENCE EDUCATION?

SVEIN LIE University of Oslo, Norway

ABSTRACT In this paper the two international comparative studies IEA TIMSS and OECD PISA have been discussed by comparing their similarities and differences. A number of examples have been presented to demonstrate how findings in various areas are relevant to help improve science education. Focus are on students’ conceptual understanding, gender and school differences, relations to home background factors, and on what characteristics of instruction that seem to be related to high achievement. Furthermore, the assessment frameworks for the two studies are argued to be of influential importance in its own terms, but that any influence on national aims and curricula should be carefully considered only in a national context.

1. TWO LARGE-SCALE COMPARATIVE STUDIES This contribution focuses on the large comparative studies, i.e. the IEA TIMSS (Trends In Mathematics and Science Study) and the OECD PISA (Programme for International Student Assessment), and discusses some of their features that can contribute to increased quality of science education around the world. More and more countries are taking part in these international studies, as many as around 50 in the 2003 versions of these two studies. One of the main goals for these studies is to present reliable comparisons between countries in the simple form of league tables. Such information can make good headings in the media, but do not in itself represent any educational improvement. However, even such simple ranking measures can inform on national strengths and weaknesses, thus help defining problematic areas that could be the focus for increased attention or further investigations. And more important, in the process of planning structural, pedagogical or content reforms in science education there is rich information in the data from these studies to learn from other countries. The aim of the present paper is to point to some important areas in this respect. There have been many critics of international assessment studies from researchers in science education in the last years (e.g. Orpwood, 2000; Jenkins, 2000). However, such sound critiques mainly concern the restricted emphasis that is put on issues central for science education. They do not challenge the technical 27 K. Boersma et al. (eds.), Research and the Quality of Science Education, 27—40. © 2005 Springer. Printed in the Netherlands.

28

HOW CAN LARGE INTERNATIONAL COMPARATIVES STUDIES CONTRIBUTE

quality or the validity of the insight that can be accomplished provided the data are carefully interpreted in national contexts. One of the aims of the present paper is to argue for more use of the rich databases for the two studies. Data for earlier studies have been released for public use, and databases for TIMSS 2003 and PISA 2003 will be publicly available shortly. By referring to and giving some examples of various types of results, I have a hope that more researchers in science education will put interest and effort into exploring the rich source of data from their own research perspectives. There are lots of relevant data from all around the world waiting to be further explored: on responses to individual achievement items, on student attitudes towards science and their motivation and further education plans, on teaching styles and learning strategies, on pedagogical climate in schools and classrooms, on teachers’ education and pedagogical beliefs, and much more. The two studies at hand have much in common, but they also differ in certain important respects. Both studies have a science component and assess scientific knowledge and skills according to what may be called a two-dimensional content-by-behaviour grid. That is, one dimension is specifying the content domains and subdomains, while the other dimension specifies the types of competencies that are being measured for each domain. Furthermore, the two studies have similar main research questions: Countries and national groups of students are compared by their mean and spread in proficiency along the cognitive scales, and these scale scores are further related to data from student and school questionnaires. Thus a focus is on the relationships between cognitive outcomes and various affective, home background and contextual student, classroom and school variables. The studies are repeated every third (PISA) or fourth (TIMSS) year, and since many elements in the questionnaires and assessment instruments are kept constant, important trend indicators are being measured. The most profound difference between the two studies concerns how the scientific competencies are defined. In principle, TIMSS is measuring the “achieved” curriculum, that is, how well students have learned the core material common to the curriculum in the majority of participating countries. The framework (Mullis et al, 2001 for TIMSS 2003) describes this “material” and also makes clear how it is based on information on what is expected to be covered by instruction during the years up to and including the actual grade of assessment. PISA defines a very different point of departure in its framework (OECD 2000) by its focus on measuring students’ “preparedness for life”, or how well students are prepared to meet challenges of future knowledge societies. Obviously, it is no simple task to make predictions concerning what will be important in the future, and further, the response to this question also depends on what is meant by “important” and in what perspective importance is judged. Whereas the present national curricula are regarded as irrelevant as descriptors of such criteria, the actual principles for competencies to be measured are based on “what is regarded as important in a perspective of life-long learning” (ibid), according to a consensus among the OECD countries at the political level. TIMSS can be said to be “curriculum driven”, in the sense that the assessment framework is based on curricular considerations. PISA, on the other hand, may be characterized as “utility -” or “relevance driven” due to the focus on

S. LIE

29

what 15-year-olds will need in their future lives as citizens and what they can do with what they have learned. PISA’s definition of Scientific Literacy is the following: “Scientific literacy is the capacity to use scientific knowledge, to identify questions and to draw evidence-based conclusions in order to understand and help make decisions about the natural world and changes made to it through human activity”. (OECD 2000, p.76) This definition implies that PISA not only assesses students’ knowledge, but also examines their ability to reflect on their knowledge and experiences, and to apply that knowledge and experience to real world issues. The term “literacy” is used to sum up this broader concept of knowledge and skills. Another difference between the two studies should also be mentioned here. While in TIMSS intact classes are drawn from two particular populations or grades, in most countries grade 4 and 8, in the sampled schools (in TIMSS 1995 even two adjacent grades for each of the two populations), in PISA 15-year old students are sampled individually in each sampled school regardless of class and grade level. This difference also explains why teacher questionnaires are included in TIMSS, but regarded as not so relevant and therefore not included in PISA. 2. WHICH BENEFITS FOR PARTICIPATING COUNTRIES? An obvious question we have to face is what is meant by quality in a situation where each country has its own goals for science education in general and for a specific grade at school? While national assessment studies usually compare student outcomes with national cognitive and affective goals, international studies compare countries according to some defined criteria that depend on consensus between the participating countries. In particular, any considerations on the validity of the reported proficiency measures will refer to a framework document that clearly explains what is measured within the actual cognitive domains. And likewise, there must be a rationale and a clear description of any reported construct in the affective domain. From one country’s perspective the cognitive and affective results will provide important information only when interpreted in a context of national curriculum goals. If national goals are different, how can comparative assessments be meaningful? Are we comparing the incomparable? There are many possible answers to this question. One is that national goals for science in schools at lower secondary level are not very different around the world, at least not for certain groups of countries, like the OECD members. There are basic elements of science content and scientific ways of thinking that are taught in most countries, even if the emphasis put at each element at a particular grade level do vary a lot (Schmidt et al, 1997). Another way of answering starts from the problematic task of comparing outcome of national assessments with national goals, since such goals usually are not stated in a way that can be easily operationalized. Curricular guidelines often give directions on which “stars” one should be aiming at, but more seldom one can find clear descriptions of how “far” a certain percentage of students are supposed to reach on

30

HOW CAN LARGE INTERNATIONAL COMPARATIVES STUDIES CONTRIBUTE

their way: Are they supposed to leave the Solar system or to struggle in the nearest bush? In this situation international comparisons, in particular with “comparable” countries, can help provide a sort of an implicit standard of what can reasonably be expected at the actual grade level. Thus, comparative studies can help measuring “fulfilment” of national goals by providing an international (albeit tacit) interpretation of what is meant by “fulfilment”. For each participating country the comparative studies provide patterns of relative strengths and weaknesses. They offer opportunities to compare competencies and curricular emphases between countries in a way that allows each country to look at its own situation from an international perspective. By comparison one may learn from others where there are possibilities for improvement and how these may be brought about. The point is not trying to make blind copies of what is done in “successful” countries. Rather, by comparing with others one can get ideas to consider implemented in a national context, taking the local cultural, historical and pedagogical traditions into account. 3. UNITS, ITEMS AND COMPETENCIES IN PISA The items in PISA are organized in units, mostly based on authentic texts, for example from a newspaper or magazine article and so on. The units include both multiple-choice and constructed response item formats. Two items within one unit in PISA 2000, Semmelweis’ diary, will serve as examples (see Appendix). The students are first given a text from Semmelweis’ diary in which he relates how frightened he is because so many mothers die of puerperal fever. The figure shows the data Semmelweis collected about the number of deaths in the two wards at the hospital. In the first question (Question 1) the students are asked to give a reason, based on the data Semmelweis collected, why puerperal fever is unlikely to be caused by earthquakes. This question assesses the intellectual process of critically evaluating scientific data, for instance by comparing data for the two wards. In fact, an answer referring to what we know today about bacteria etc, albeit in itself a “correct” answer would not be a relevant response to the question. The last question (Question 4) of the unit requires a different type of competency. Here, relevant content knowledge and conceptual understanding are instrumental in responding correctly (alternative B). These two items also demonstrate two different item formats, open constructed (Q1) and multiple choice (Q4) formats. According to the definitions in PISA 2000 (OECD 2000, p.77) five types of competencies were defined: • Recognizing scientifically investigable questions • Identifying evidence needed in a scientific investigation • Drawing or evaluating conclusions • Communicating valid conclusions • Demonstrating understanding of scientific concepts For PISA 2003 and 2006 slightly different versions are being used. But the duality between what may be called “process skills” vs. “content knowledge” (Kjærnsli& Molander, 2003) is kept as a basic feature. This dichotomy contrasts the

S. LIE

31

first four of the five competencies above (process skills) on one hand with the fifth competency (content knowledge/conceptual understanding) on the other. By investigating the achievement data along this dichotomy, one can get valuable insight into characteristic differences between countries concerning their emphasis on and achievement within the “process” vs. the “content” aspect of science. Compared to TIMSS, PISA has a much stronger emphasis on the process aspect, a feature that can partly explain the remarkable high scores for the English-speaking countries in PISA. (OECD , 2001). In PISA 2000 there were too few items to make individual scale scores according to the dichotomy in question. Nevertheless, Kjærnsli & Molander (ibid) have categorized all the science items into two groups, depending on in which of the two aspects they belong, and compared percentages of correct responses for each item by country and by gender. They have demonstrated that students in all the English-speaking countries, Australia, Canada, United States, United Kingdom and Ireland, performed relatively better on items focusing on science process skills. On the other hand, “East European countries” (or rather, former European communist countries) achieve relatively better within the domain of conceptual understanding. The division into these two groups of countries; the English-speaking countries and the “East European” countries, seems to throw light on important cultural traditions. We know that English-speaking countries put more emphasis on the process aspect in their education system, and we also often see that Eastern European countries place more emphasis on factual knowledge and conceptual understanding. As exemplified above, by secondary in-depth studies of item-by-country interactions one may penetrate further into countries’ patterns of relative strengths and weaknesses. There are different traditions concerning curricular emphases in science around the world. Cluster analysis of percentage correct responses have proved a potent means of revealing similarities and differences between the countries, thus helping countries to understand its own tradition and how this contrasts other existing cultural contexts for science education (Zabulionis, 2001; Grønmo et al., 2004). 4. ITEMS, CONTENT DOMAINS AND STUDENT UNDERSTANDING IN TIMSS The main purpose of the achievement items in comparative studies is to provide the basis for valid and reliable scale scores within certain specified cognitive domains. However, item specific data can also provide valuable insight into students’ conceptual understanding and the nature and even origin of their misconceptions. In TIMSS responses to all open constructed (also called “free response”) items are coded according to a two-digit system, the first digit giving the quality or score points, whereas the second digits specifies the type of correct of wrong type of response (Lie, Taylor & Harmon, 1996). Thus important information on student responses is taken care of in the database and can be analyzed in secondary studies. A few diagnostic analyses have been presented (e.g. Angell, Kjærnsli & Lie, 2000; Kjærnsli, Angell & Lie, 2002), but there are opportunities for many more to come.

32

HOW CAN LARGE INTERNATIONAL COMPARATIVES STUDIES CONTRIBUTE

The database with data from countries all over the world, provide rich opportunities for comparing aspects of students’ thinking and (mis-)understanding in a variety of different educational contexts and traditions. In Norway we have published a diagnostic resource book for teachers that contain all released TIMSS 1995 science items with Norwegian and international response distribution and diagnostic comments and hints to teachers about the nature of misunderstandings and how teachers can address them in the classroom (Kjærnsli et al, 1999). Let us take a look at two TIMSS items for population 2 (12-13 year olds). The first (O12) is a multiple-choice item asking for simple factual knowledge: Air is made up of many gases. Which gas is found in the greatest amount? A. Nitrogen B. Oxygen C. Carbon dioxide D. Hydrogen Fact is that nitrogen is by far the most abundant gas, close to 80%, while oxygen covers most of what is left. However, the international response distributions among 13-year olds for the four alternatives are extremely different: 27%, 53%, 14% and 5% respectively and with only 2% missing responses. In most countries around the world, oxygen is the obvious and most popular response. And even carbon dioxide is picked by a substantial number of students. As Jenkins has put it in a comment on the results for this item: “Given that the composition of the air is a feature of school science education common to all countries in the TIMSS study, the responses to this test item can only be regarded as severely disappointing” (Jenkins, 2000). Here we can see a demonstration of the common conception that oxygen and air is more or less the same. Nitrogen, on the other hand, emerges as a rather obscure gas, about which not much is heard, neither at school nor in the media. Information like this can and should be brought forward to science teachers, who may take this into account when planning lessons on topics like the atmosphere, oxygen intake of humans or photosynthesis. Another item (Q18) asks a very fundamental question about the conservation of mass in physical processes. The required understanding is that when melting, ice is converted into an amount of water of exactly the same mass. A glass of water with ice cubes in it has a mass of 300 grams. What will the mass be immediately after the ice has melted? Explain your answer. Internationally, only 30% of the students got it right; stating that the mass will stay the same, with an adequate explanation. In addition, about 10% gave a correct response of 300 grams, but with no or inadequate explanation. Students that gave an incorrect response, were distributing themselves fairly equally between more than 300 g (17%) and less than 3 00g (12%). The fundamental aspect of mass being an invariant property during melting and freezing processes (as well as in and evaporation and condensation) is a very fundamental aspect of matter. One might suspect that this item would be much simpler for countries with a cold climate. However, this is not the case. In Norway, as in the Nordic countries in general, the response distribution is quite close to the international mean. During the marking process in Norway we looked into the explanations given to the incorrect responses (Kjærnsli, Angell & Lie 2002). Some interesting sources for the misconceptions

S. LIE

33

were evident. Most typically, a common way of reasoning starts from the fact that “ice is lighter than water, since it floats.” As is often the case, this daily life expression is in conflict with correct scientific terminology, which states that ice has lower density but is not “lighter”. Since the process of melting transforms ice into water (which is regarded as “heavier”), the conclusion for these students is that we will end up with more mass. Explanations of the correct responses could be very different and still receive full credit. The short and easy direct reference to the general principle of constant mass was not given by many. It may well be that some students know this principle, but still do not regard a statement of this “law” as appropriate. After all, referring to a general principle, albeit perfectly correct, really does not explain anything (“Why is it so?”), beyond restating the correct response (“that”). As soon as they try to give an explanation in terms of volume and density, or even microscopic particles, they tend to get confused (ibid). The two items discussed here, and also the two PISA items discussed above, are just examples of the rich diagnostic information from students worldwide that is available for secondary studies. TIMSS and PISA both offer rich opportunities for in-depth studies that can give a better “understanding of students’ understanding”, their conceptual frameworks as well as the nature and abundance of common misconceptions. 5. GENDER AND SCHOOL DIFFERENCES International studies are very well suited to unveil characteristic similarities and differences between countries concerning patterns of gender differences. If we measure gender differences as standardized differences (difference divided by standard deviation of the distribution) we can compare the magnitude across scales and studies, even achievement with affective measures. Since the first science study in the 1970s, the gender gap in favour of boys has decreased steadily via the second to the third (TIMSS) study. From a difference of more than 30 % of a standard deviation, the difference was around 15% in TIMSS 1995. In PISA the gender difference were not even statistical significant in most countries. As explained above, there are important differences between the PISA and TIMSS tests, so these outcomes cannot be directly compared, however. Another interesting aspect of gender differences is the interplay between cognitive and affective factors. In Norway grades 1-7 belong to primary schools, where no grades are given and the pressure to achieve is rather low. At grade 8 students enter secondary school and meet more formal science and more focus on achievement and grades. In TIMSS 1995 the achievement differences came out the same at both grade levels, 13% of a standard deviation. However, for a construct measuring positive attitude towards science, the gap increased dramatically from 10% to 32% of a standard deviation from grade 7 to grade 8. This finding sends out an important message from a perspective of gender balance in selection of science in upper secondary school: Even if gender differences in achievement are not high, the affective gender gap is dramatically widening and this is alarming, since students’ selections of subjects are very influenced by affective components.

34

HOW CAN LARGE INTERNATIONAL COMPARATIVES STUDIES CONTRIBUTE

What about differences between schools? A common measure of the magnitude of this difference is the percentage of the over-all variance of student achievement that can be ascribed to the differences between schools. In the Nordic countries as well as Japan and Korea, this percentage is around 10%, a low value indicating that most of the student variance can be found inside schools (and even classes). On the other hand, in countries with tracking or streaming systems at this level (e.g. German-speaking countries, Netherlands and USA), the between-school part of the variance can be well above 50%, indicating a more even distribution within schools, but with much higher diversity between schools. The same type of results emerged from PISA 2000. And furthermore, there is a tendency that in countries with higher diversity between schools there are stronger relationship between achievement and home background factors. This finding may be regarded as an argument for not separating students too early in the schooling system if one wants to counteract the tendency for schooling to reproduce socio-economic differences in the society. 6. THE IMPORTANCE OF HOME BACKGROUND In both TIMSS and PISA the relationship between home background factors and the various outcome measures constitute important parts of the research questions. PISA puts particularly strong emphasis on measuring rich and reliable home background variables since relations between achievement and home background are important indicators for all countries (OECD, 2001). An example will easily demonstrate the strength of this relation. Figure 1 displays how the over-all TIMSS science achievement in one country depends on the number of books at home. Obviously, this variable is an indirect indicator of the socio-economic status of the family, while the books themselves do not have any important direct influence on student achievement. 7. FOCUS ON SCIENCE TEACHING To follow up on what is just stated above, what are some findings concerning “good” science teaching? There are lots of reasons why one cannot expect very clear messages. One is discussed above, and furthermore, since students have had a long educational career, one cannot expect characteristics of instruction during the present school year to have strong influence on achievement outcomes. Nevertheless, it is possible to sum up some of the tendencies that have been revealed(e.g. Beaton et al, 1996; Martin et al, 2000; OECD 2001; Pelgrum & Plomp, 2002).

S. LIE

35

56 54

52

Science score

50

48 46

44

42 N=

248

454

0-10

1314

910

26-100 11-25

1149

Above 200 101-200

Nr of books in your home Figure 1: Science achievement (with 95% confidence intervals) as a function of numer of books at home. TIMSS 1995, Norwegian data. Science scale: Mean = 50, Standard deviation = 10 Figure 1 clearly demonstrates the strong relationship between achievement and home background that can be found in all countries, albeit to a various extent. This finding provides one part of the explanation why it is difficult to find pronounced relations between instruction variables and outcome in these studies, and also in other studies. Clearly, if we want to explore what seems to be “good practice”, the dominant dependence on student home background factors is a disturbing element. However, there are various ways the home background factors can be “corrected” for (e.g. Martin et al, 2000; Välijärvi & Malin, 2003, Turmo & Lie, 2004). 8. FOCUS ON SCIENCE TEACHING To follow up on what is just stated above, what are some findings concerning good” science teaching? There are lots of reasons why one cannot expect very clear messages. One is discussed above, and furthermore, since students have had a long educational career, one cannot expect characteristics of instruction during the

36

HOW CAN LARGE INTERNATIONAL COMPARATIVES STUDIES CONTRIBUTE

present school year to have strong influence on achievement outcomes. Nevertheless, it is possible to sum up some of the tendencies that have been revealed (e.g. Beaton et al, 1996; Martin et al, 2000; OECD 2001; Pelgrum & Plomp, 2002). These findings are, however, open for interpretations, and to the extent that the following points have been generalized beyond the stated findings themselves, I want to emphasize that they are interpretations: • It is not possible to describe any single “Best practice” for science teaching around the world. • Focus on learning goals is crucial regardless of the instruction method. This seems to be less easily obtained in some student-centred settings, like project work and internet surfing. • Time on task is crucial, in both mental and physical sense, and at school as well as at home. • Student-centred and teacher-lead instruction can both lead to high and low achievement, but there are tendencies that the latter teaching style is related to higher science achievement. • Computers provide in themselves by far no shortcut to scientific competencies. The potential for learning is obvious, but computers also do provide excellent opportunities to learn nothing. Internet surfing seem to correlate negatively with achievement. • Some countries (e.g. Japan and Korea) seem to a remarkable extent to be able to combine teacher-lead instruction with mentally involving all students in social learning situations. • There is no evidence that strong emphasis on student or teacher experiments is linked to higher achievement on TIMSS and PISA written tests, rather there are tendencies towards the opposite. Again, given that focus on learning goals are crucial, student activity provides in itself no simple way towards understanding. • There are mixed findings concerning what is sometimes called “constructivist teaching”, a rather misleading term. Tendencies are that one aspect of constructivism is linked positively with achievement, namely the learning strategy of linking new concepts to the already known, and the emphasis that is put on this aspect by teachers. The other side, the focus on “student-centred” approaches in the meaning of students’ independent learning work gets little support from the data. Instead, the crucial role of teacher guidance concerning learning activities and summing-up discussions clearly speak out from the results. 9. CONCLUDING REMARKS In this paper I have focused on some features of the two international studies, TIMSS and PISA. The two studies have been compared and typical differences and similarities have been discussed. The two studies are different, and in some respect they complement each other. That is, one cannot simply regard one as “better” than

S. LIE

37

the other. In future versions of the two studies, they may well be even more different, in order that more countries will find it useful to participate in both. I have given just a few examples of how findings from these studies may contribute to increased quality of science education. However, maybe the most important features of these comparative studies are provided by the assessment frameworks themselves. They do establish international frameworks to which countries can relate their national aims and curricula. They also provide a common base and language for discussing similarities and differences in school science worldwide. It is important to emphasize, however, that they do not establish any international standard or curriculum with which national curricula should match. Any influence from these studies should obviously be interpreted and framed in a national context. REFERENCES Angell, C., Kjærnsli, M. & Lie, S. (2000). Exploring Student Responses on FreeResponse Science Items in TIMSS. In: Shorrocks-Taylor, D. & Jenkins, E.W. (Eds.):Learning from Others. International Comparisons in Education. Dordrecht/Boston/London: Kluwer Academic Publishers, p. 159-187. Beaton, A. E., Martin, M. O., Mullis, I. V. S., Gonzales, E. J., Smith, T. A. & Kelly, D. L. (1996). Science Achievement in the Middle School Years. IEA’s Third International Mathematics and Science Study. Boston: TIMSS International Study Center, Lynch School of Education, Boston College. Grønmo, L.S., Kjærnsli, M. & Lie, S. (2004): Looking for Cultural and Geographical Factors in Patterns of Responses to TIMSS Items. Paper to the 1st IEA International research Conference, Lefkosia, Cyprus, May, 2004 Kjærnsli, M., Lie, S., Stokke, K.H. & Turmo, A. (1999): Hva i all verden kan elevene i naturfag? Oppgaver med resultater og kommentarer. (What in the world can the students in science? Items with results and comments. In Norwegian). ILS. University of Oslo. Kjærnsli, M., Angell, C. & Lie, S. (2002). Exploring Population 2 students' ideas about science. In: Robitaille, D.F. & Beaton, A.E. (eds.): Secondary Analysis of the TIMSS Data. Dordrecht/ Boston/ London. Kluwer Academic Publishers, p. 127-144. Kjærnsli, M. & Molander, B. O. (2003): Scientific literacy: Content Knowledge and Process Skills. In Lie, S., Linnakylä & Roe, A. (eds.): Northern Lights on PISA. Unity and Diversity in the Nordic countries. Oslo: ILS, University of Oslo. Jenkins, E.W.(2000): Making Use of International Comparisons of Student Achievement in Science and Mathematics. In: Shorrocks-Taylor, D. & Jenkins, E.W. (Eds.):Learning from Others. International Comparisons in Education. Dordrecht/Boston/London: Kluwer Academic Publishers, p. 137-157.

38

HOW CAN LARGE INTERNATIONAL COMPARATIVES STUDIES CONTRIBUTE

Lie, S., Taylor, A., & Harmon, M. (1996). Scoring techniques and criteria. In M. O. Martin & D. Kelly (Eds.), Third International Mathematics and Science Study technical report. Volume 1: Design and development. Boston.: International Study Center, Lynch School of Education, Boston College. Martin, M. O., Mullis, I. V. S., Gregory, K. D., Hoyle, C. & Shen, S. (2000): Effective Schools in Science and Mathematics. IEA’s Third International Mathematics and Science Study. Boston: TIMSS International Study Center, Lynch School of Education, Boston College. Mullis, I.V.S., Martin, M.O., Smith, T.A., Garden, R.A., Gregory, K.D., Gonzales, E.J., Chrostowski, S.J. & O’Connor, K.M. (2001). TIMSS Assessment and Specifications 2003. Boston: International Study Center, Lynch School of Education, Boston College. Orpwood, G. (2000). Diversity of Purpose in International Assessments: Issues arising from the TIMSS Tests of Mathematics and Science. In: ShorrocksTaylor, D. & Jenkins, E.W. (Eds.):Learning from Others. International Comparisons in Education. Dordrecht/Boston/London: Kluwer Academic Publishers, p. 49-62. OECD (2000). Measuring Student Knowledge and Skills. The PISA 2000 Assessment of Reading, Mathematical and Scientific Literacy. Paris: OECD Publications OECD (2001) Knowledge and Skills for Life. First results from PISA 2000. Paris: OECD Publications. Pelgrum, W.J. & Plomp, T. (2002): Indicators of ICT in Mathematics: Status and Covariation with Achievement measures. In: Robitaille, D.F. & Beaton, A.E. (eds.): Secondary Analysis of the TIMSS Data. Dordrecht/ Boston/ London. Kluwer Academic Publishers, p. 317-330. Schmidt, W.H., Raizen, S., Britton, E., Bianchi, L.J., & Wolfe, R.G. (1997). Many visions, many aims, Volume 2. A cross-national investigation of curricular intentions in school science. Dordrecht/Boston/London: Kluwer Academic Publishers. Turmo, A. & Lie, S. (2004) Hva kjennetegner norske skoler som skårer høyt i PISA2000? (What characterize Norwegian schools with high scores in PISA 2000? In Norwegian). Acta Didactica nr 1/2004. Oslo: ILS, University of Oslo. Välijärvi, J. & Malin, A. (2003): The two-level effect of socio-economic background. In Lie, S., Linnakylä, P. & Roe, A. (eds.): Northern Lights on PISA. Unity and Diversity in the Nordic Countries in PISA 2000, p. 123132. Oslo: ILS, University of Oslo. Zabulionis, A. (2001). Similarity of mathematics and science achievement of various nations. Educational Policy Analysis Archives 9 (33).

S. LIE

39

Appendix SEMMELWEIS’ DIARY TEXT 1 ‘July 1846. Next week I will take up a position as “Herr Doktor” at the First Ward of the maternity clinic of the Vienna General Hospital. I was frightened when I heard about the percentage of patients who die in this clinic. This month not less than 36 of the 208 mothers died there, all from puerperal fever. Giving birth to a child is as dangerous as first-degree pneumonia.’

These lines from the diary of Ignaz Semmelweis (1818-1865) illustrate the devastating effects of puerperal fever, a contagious disease that killed many women after childbirth. Semmelweis collected data about the number of deaths from puerperal fever in both the First and the Second Wards (see diagram).

Number of Deaths per 100 deliveries from puerperal fever Number of Deaths

F irst W d

Sec ond W d 1841

1842

1843

1844

1845

1846

Diagram

Physicians, among them Semmelweis, were completely in the dark about the cause of puerperal fever. Semmelweis’ diary again: ‘December 1846. Why do so many women die from this fever after giving birth without any problems? For centuries science has told us that it is an invisible epidemic that kills mothers. Causes may be changes in the air or some extraterrestrial influence or a movement of the earth itself, an earthquake. Nowadays not many people would consider extraterrestrial influence or an earthquake as possible causes of fever. We now know it has to do with hygienic conditions. But in the time Semmelweis lived, many people, even scientists, did! However, Semmelweis knew that it was unlikely that fever could be caused by extraterrestrial influence or an earthquake. He pointed at the data he collected (see diagram) and used this to try to persuade his colleagues.

40

HOW CAN LARGE INTERNATIONAL COMPARATIVES STUDIES CONTRIBUTE

Question 1: SEMMELWEIS’ DIARY Suppose you were Semmelweis. Give a reason (based on the data Semmelweis collected) why puerperal fever is unlikely to be caused by earthquakes.

Question 4: SEMMELWEIS’ DIARY Many diseases may be cured by using antibiotics. However, the success of some antibiotics against puerperal fever has diminished in recent years. What is the reason for this? A B C D

Once produced, antibiotics gradually lose their activity. Bacteria become resistant to antibiotics. These antibiotics only help against puerperal fever, but not against other diseases. The need for these antibiotics has been reduced because public health conditions have improved considerably in recent years

TOWARDS A MORE CURRICULAR FOCUS IN INTERNATIONAL COMPARATIVE STUDIES ON MATHEMATICS AND SCIENCE EDUCATION

WILMAD KUIPER¹, KERST BOERSMA², JAN VAN DEN AKKER³ ¹University of Twente, ²University of Utrecht, The Netherlands

ABSTRACT From international comparative studies (TIMSS, PISA) it appears that students in lower secondary education in the Netherlands perform relatively well in mathematics and science compared to their peers from other participating countries. Policy-makers, especially, are eager to bring these positive outcomes into the limelight. However, one may wonder whether, in case of the Netherlands, there is good reason for such zeal. An evaluation study, conducted by the Netherlands Inspectorate of Education, shows that lower secondary schools do not meet the quality required in implementing a curriculum reform that started in 1993, entitled ‘basic secondary education’. So, in spite of all rhetoric on the positive outcomes of TIMSS and PISA in the Netherlands, when putting the relatively good student performance in the context of the implementation of this ambitious curriculum reform, many people become puzzled. Research findings on the quality of mathematics and science education seem to be in conflict with the results of TIMMS and PISA. This conclusion and also the observation that international comparative assessment studies have serious difficulty in meeting the goal of providing proper interpretations of student achievement, especially from a curriculum perspective, give reason to attempt to disentangle the conflicting images.

1. INTRODUCTION AND PROBLEM STATEMENT The outcomes of international comparative studies like TIMSS and PISA get widespread attention in media and policy circles. Depending on the nature of the results, they tend to provoke a wide array of, often rhetorical, reactions. For example, the relatively poor performances of American 13 year old students in mathematics and science in TIMSS-1995 and TIMSS-Repeat 1999 gave cause to a still continuing flow of discussions, arguments, and reflections on origins of this problem (“the mathematics as well as the science curriculum is a mile wide and an inch deep”) and on possible solutions to it (‘rigorous’ and ‘demanding’ new standards). The weak performances of students in lower secondary education in Germany in TIMSS-1995, in TIMSS-Repeat 1999, and especially in PISA-2001 (with 31 participating countries Germany appeared 20th in the ranking for mathematics and science and 21st in the ranking for reading comprehension) caused a public debate that was dominated by great displeasure and concern about the quality of education in Germany. Also in the Netherlands the reactions poured into 41 K. Boersma et al. (eds.), Research and the Quality of Science Education, 41—54. © 2005 Springer. Printed in the Netherlands.

42

TOWARDS A MORE CURRICULAR FOCUS

the air, although they were quite different in nature due to the fact that – as it appears from TIMSS-1995, TIMSS-Repeat 1999, as well as PISA-2001 – students in lower secondary education perform relatively well in mathematics and science compared to their peers from other participating countries (Bos, Kuiper & Plomp, 1999; Bos & Vos, 2000; Kuiper, Bos & Plomp, 1999; Kuiper, Bos & Plomp, 1997; Wijnstra, 2001). Policy-makers are especially eager to bring these positive outcomes into the limelight. However, one may wonder whether, in case of the Netherlands, there is good reason for such zeal, and especially for the sense of self-satisfaction that some quotes and comments seem to convey. This is not because of the low response rates in TIMSS-1995 and PISA-2001 that may have biased the good results, but rather because of the outcomes of an evaluation study conducted almost at the same time by the Netherlands Inspectorate of Education (Inspectie van het Onderwijs, 1999ae). This evaluation study shows that secondary schools do not achieve the quality required in implementing a curriculum reform that began in 1993, entitled ‘basic secondary education’. An even less favorable picture emerges when the performances of Dutch students in TIMSS are contrasted with the demanding instructional and learning goals as defined at system level in terms of attainment targets, instead of with the international mean achievement score (which is common practice in international comparative studies). Others (for example, Boersma, 2000ab) criticize the new curriculum for mathematics, physics/chemistry, and biology, as it is overloaded and fragmented, lacks coherence and longitudinal alignment, is implemented without sufficient relevance for students, and is dominated by rather traditional modes of assessment. So, in spite of all rhetoric on the positive outcomes of TIMSS (and also PISA) in the Netherlands, when putting the relatively good student performance in the context of the ‘challenging’ implementation of the curriculum reform in lower secondary education, many people become puzzled. Research findings on the quality of mathematics and science education seem to be in conflict with the results of TIMMS and PISA. This conclusion and also the observation that international comparative assessment studies have serious difficulty in meeting the goal of providing proper interpretations of variations in student achievement in view of policy implications (Bos, 2002; Kellaghan, 1996), in general and especially from a curriculum perspective, led us to attempt to disentangle the conflicting images. The outcomes of this attempt are described in this chapter. We start with a more in-depth analysis of conflicting images in the Netherlands as appearing from main findings from TIMSS and the Evaluation Study by the Inspectorate of Education (ESIE). This analysis is meant to clarify the debate and to articulate a curricular focus in international comparative studies like TIMSS. There is a clear need for doing the latter, as it has also been cogently substantiated by Westbury (1992) in his analysis of differences in achievement – found in SIMS – between American and Japanese secondary school students. A conceptual focus that emphasizes the “fundamental salience of curriculum” (Westbury, 1992, p.23) offers a chance for a sharper understanding of (factors influential to) mathematics and science achievement,

W. KUIPER, K. BOERSMA, J. VAN DEN AKKER

43

which, in turn, is a prerequisite for more focused policy recommendations aiming at the enhancement of mathematics and science education. 2. CONCEPTUAL FRAMEWORK As a stepping-stone in our analysis, we start with the curriculum typology known from TIMSS and other IEA studies (Robitaille et al., 1993; Schmidt et al., 2001): the intended, the implemented, and the attained curriculum. In our definition a curriculum is ‘a plan for learning’ that, depending on its nature and scope, may pertain to several components (van den Akker, 2003): rationale; aims, goals, and objectives; contents; teacher’s role; student activities; materials and resources for teaching and learning; time allocation; location; and assessment modes and criteria. The intended curriculum refers to all those provisions aimed at being offered to students, including all those concepts, processes, and attitudes students are expected to study and learn. These may find expressions in formal documents (such as official attainment targets) and textbooks. The implemented curriculum is the curriculum as interpreted by teachers and made available to students (curriculum-in-action). The attained curriculum refers to that portion of the curriculum actually attained by students. This includes achievement measures as well as students’ attitudes, perspectives, and values. These three curriculum representations closely cohere. Also, there is never a linear, top-down transformation from curriculum intentions via implementation in teaching and learning settings to students’ outcomes. It is a complicated process in which much elaboration and adaptation may be needed and may occur. Also a lot of ‘noise’ may arise. Original intentions can be blurred, distorted, or even devastated. Also other, often more powerful variables than only the intended curriculum may have an effect on the implemented and the attained curriculum (Figure 1; based on van den Akker, 1998). Some of the variables in this curriculum transformation or ‘curriculum dilution’ process may also be non-curricular in nature, like sociocultural context (home, media, peers) and student characteristics (aptitude, motivation, gender). Nevertheless, in our analysis of the Netherlands case we focus on the curriculum levels depicted at the horizontal axis in the middle of the Figure: intended – implemented – attained. Findings about (school and socio-cultural) context, student characteristics, and teacher characteristics will be left aside for the greater part, as it is not our ambition to look for an explanation of the findings of each of two studies. Instead, we try to disentangle conflicting images by comparing main findings at the three curriculum levels within and across the two studies

44

TOWARDS A MORE CURRICULAR FOCUS

Intended curriculum

School context

Sociocultural context

Implemented curriculum

Attained curriculum

Teacher Characteristics

Student characteristics

Figure 1. Curriculum typology Nevertheless, in our analysis of the Netherlands case we focus on the curriculum levels depicted at the horizontal axis in the middle of the Figure: intended – implemented – attained. Findings about (school and socio-cultural) context, student characteristics, and teacher characteristics will be left aside for the greater part, as it is not our ambition to look for an explanation of the findings of each of two studies. Instead, we try to disentangle conflicting images by comparing main findings at the three curriculum levels within and across the two studies. 3. NETHERLANDS CASE: TIMSS AND ESIE This section encompasses an analysis of the main findings from TIMSS and ESIE, preceded by some context information about the implementation of basic secondary education and about the goals, design, and instrumentation of ESIE and TIMSS. Basic secondary education The formal implementation of basic secondary education started in 1993. It aims at raising the standard of lower secondary education and at ‘modernizing’ the curriculum while maintaining the existing structure of four student ability tracks. It entails a core curriculum of 15 subjects (including mathematics, physics/chemistry, biology) covering the first three years of secondary education. For each subject, attainment targets have been set which indicate the expected level of achievement in terms of knowledge, understanding, and skills. The modernization not only refers to an increase of the number of subjects up to 15, but also to an intended change of both the subjects’ contents (more application-oriented) and pedagogy (more activitybased and student-centered) at classroom level. As time has passed, complaints

W. KUIPER, K. BOERSMA, J. VAN DEN AKKER

45

began to pour in about the reform (see earlier). As an attempt to arrive at a solution, the government has prepared a proposal for a sweeping revision of basic secondary education from 2005 onwards. It is in this turbulent reform context that Secondary 1 and 2 students perform(ed) relatively well on TIMSS tests. ESIE ESIE was carried out by government order during the school year 1997-1998. It was a large-scale evaluation (120 schools) of the implementation of basic secondary education four to five years after its formal start in August 1993 (Inspectie van het Onderwijs, 1999a-e). Investigated was the extent to which the attainment targets as well as general skills (e.g., conducting a simple inquiry) were part of the intended curriculum at school level. For these purposes textbooks in use, additional curriculum materials, and schools’ work plans were analyzed, teacher questionnaires and interviews were administered, and lesson observations took place (Peters-Sips et al., 2000). As regards the implemented curriculum, lesson observations were conducted in order to obtain a picture of the quality of the teaching and learning at classroom level (van den Bergh, Zwarts & Peters-Sips, 2000). At the attained level secondary analyses took place, of student performances on drafts of so-called ‘basic secondary education tests’ constructed and administered by CITO in the spring of 1997 and 1998. TIMSS The TIMSS-1995 Population 2 study in the Netherlands, with the data collection in spring 1995, entailed the following components (Kuiper, Bos & Plomp, 1997): • Attained curriculum: the administration of a written mathematics and science test in Secondary 1 and 2 (95 schools) plus a performance assessment in Secondary 2 (49 schools). • Implemented curriculum: the administration of a student questionnaire (attitudes) and a teacher questionnaire (teaching practices; ‘opportunity-tolearn’ judgments, i.e. judgments on whether the content tested via a selection of items had been taught before test administration). • Intended curriculum: an expert appraisal on the appropriateness of the items from the written test for the attainment targets for mathematics, physics/chemistry, and biology. As part of TIMSS-1995 the Netherlands also contributed to an extensive crossnational analysis at the level of the intended curriculum, encompassing curriculum guides and textbooks in most common use (cf. Schmidt, McKnight et al., 1997; Schmidt, Raizen et al., 1997). TIMSS-1999 (data collection, spring 1999) consisted of the same components as TIMSS-1995 (Bos & Vos, 2000), except for: (i) the written test that was administered in Secondary 2 only (126 schools), (ii) teachers who made a ‘opportunity-to-learn’ judgment to all (and not only a selection of) written test items, and (iii) the cross-national curriculum analysis. In addition, only in the Netherlands,

46

TOWARDS A MORE CURRICULAR FOCUS

the performance assessment was repeated in the spring of 2000 (27 schools; Vos & Kuiper, 2004). Differences in research object A complicating factor in comparing main findings from TIMSS and ESIE is that there is a major difference between the two as regards research object. TIMSS primarily focuses on the measurement of students’ performances in mathematics and science: the attained curriculum. Data at the level of the implemented curriculum (e.g. teachers’ judgments about opportunity to learn) or at the level of the intended curriculum (match between test items and attainment targets) are meant as context information for interpreting students’ performances. ESIE has the implemented curriculum as its primary focus, with data about students’ outcomes as measured by some national tests (attained) as well as data about the match between the implemented curriculum and the attainment targets (intended) as secondary sources. Due to these differences, a direct comparison between the two studies is hard to make. Nevertheless, there are findings from both studies that seem to indicate that in this regard, the two studies are consistent with each other. As a consequence, both studies seem to provide us with sufficient input for the disentangling attempt envisaged. Main findings: Attained curriculum As far as the written test is concerned, Dutch Secondary 2 students performed relatively well in the TIMSS comparison in both 1995 and 1999. About two-thirds of the students achieved above the international mean for both mathematics and science (41 countries in 1995, 37 in 1999). However, Dutch students did not score outstandingly well in the TIMSS 1995 performance assessment, although curriculum experts judged the practical test as matching well with the attainment targets. The students’ overall mean achievement (average 61% correct), was near the international average (average 59% correct). Five years later the overall mean achievement (average 64% correct) had improved slightly but significantly, due to a better performance on the science tasks only. In ESIE, student performances on the ‘basic secondary education tests’ have been compared with the standard of achieving ‘above the level’, ‘on the level’, or ‘below the level’. For each student ability track and for each content area, these standards had been set by teachers and subject-matter experts. The picture for mathematics and for the science subjects didn’t appear to be univocal (Inspectie van het Onderwijs, 1999c, d, and e). The results roughly showed that for biology and physics/chemistry, students from the two higher ability tracks (havo, vwo) generally did not meet the standards of performing ‘at or above their level’, contrary to students from the two lower ability tracks (vbo, mavo) who generally performed ‘at or above their level’. For mathematics the results were the other way around. From other analyses at the aggregate level of the exact sciences (mathematics, physics/chemistry, biology as well as technology), it appeared that 67% of the students from the lowest ability track (vbo) performed ‘at or above their level’

W. KUIPER, K. BOERSMA, J. VAN DEN AKKER

47

(Inspectie van het Onderwijs, 1999a). The same was true for 67% of the students from the second highest ability track (havo) and for 53% of the students from the highest ability track (vwo). This means that 33% (vbo), 33% (havo) and 47% (vwo) of the students performed below the standard that had been set by teachers and subject matter experts for their ability level (for readers unfamiliar with the Dutch school system, a valuable source of information is http://www.minocw.nl). Reflective comments In TIMSS student achievement has been measured via international tests. Students’ performances on these tests have been expressed in a country’s mean score for both mathematics and science. This mean score determined a country’s position in the international ranking, and also its position relative to the international mean. The criterion for a country’s performance is referenced to the international mean, which in turn depends on the number of participating countries, as well as their performance levels. The substantial number of developing countries participating in TIMSS makes the performance of Dutch students appear relatively good, but one may wonder what is usefulness of such a comparison. In ESIE quite a different approach has been applied. Teachers and subject matter experts formulated standards for each student ability track. Next, it was determined to which extent students were able to meet those standards. In addition, those absolute standards were based on the attainment targets, which are a much more ambitious criterion than an international mean on a test that covers “an internationally consensual body of content defining mathematics (and science)” (Westbury, 1992, p. 19). Another difference with TIMSS is that the (secondary analyses of) student performances reported in ESIE were based on (drafts of) national tests. The administration of these laborious and time-consuming tests took some doing. As a consequence, the results may be disputed. So, in the two studies, student outcomes have been measured using different tests. Also there are large differences in standards that have been used as a reference. Differences in standards make it possible to judge student performance in TIMSS much more positively than those in ESIE. However, due to differences in test instruments used and standards set, making comparisons between student achievements in both studies does not make sense. Main findings: Implemented curriculum In TIMSS-1999 it appeared that 82% of the mathematics teachers and 64% of the science teachers determined each of the items from the written test appropriate to the implemented curriculum (Vos & Bos, 2000). For the performance assessment, 58% of the mathematics teachers and 46% of the science teachers came to such a judgment (Vos & Kuiper, 2004). So, the appropriateness of the written test, the written science test, and the performance test (both mathematics and science) to the implemented curriculum was good, less good, respectively moderate. This conclusion implies that part of the content tested had not been taught before test administration. The finding that students, in spite of this, performed relatively well

48

TOWARDS A MORE CURRICULAR FOCUS

is striking and suggests an influential role of extracurricular factors like school and socio-cultural context on student outcomes (Figure 1). Unfortunately, there are no findings, either from TIMSS or from ESIE, that might offer further clues in this respect. In order to picture the implemented curriculum in ESIE a distinction was made on five quality standards for teaching and learning at classroom level (van den Bergh et al., 2000): positive class climate, class management and teaching and learning approach, pedagogical content approach, promotion of active learning, and considering individual differences. Indicators were set for each of these standards. Based on observations, the various standards were rated in terms of: predominantly weak, weakness dominates strength, strength dominates weakness, and predominantly strong. Strengths of the implemented mathematics, physics/chemistry, and biology curricula appeared to be class climate as well as, but to a less extent, class management and instructional approach. The majority (varying from 61% to 83%) of the lessons observed were rated as ‘at least sufficient’ with regard to these two standards. The pedagogical content approach was rated as ‘not sufficient’ in half (physics/chemistry and biology) or one-third (mathematics) of the lessons. Promotion of active learning was rated as ‘not sufficient’ in about half of the mathematics, physics/chemistry, and biology lessons. In two-thirds or more of the lessons individual differences were not sufficiently considered. These findings brought the Inspectorate to the overall conclusion that a teaching approach that promotes active learning – one of the key-characteristics of the intended curriculum reform in basic secondary education – is still lacking. The Inspectorate’s conclusion, also reached by Kuiper (1993) in an earlier study on science teaching practices, seems to be to some extent in line with the TIMSS performance assessment findings. The performance assessments in 1995 and 2000 showed that not only student achievement, but also the appropriateness of the performance test to the implemented curriculum, turned out to be less suitable than hoped (Vos & Kuiper, 2004). The Inspectorate’s conclusion can be regarded as a support for the latter, as it is plausible that a predominantly instructivist and teachercentered approach has a detrimental effect on mastering practical skills. Reflective comments In order to get an understanding of the appropriateness of the international tests to the implemented curriculum, in this component of TIMSS the implemented curriculum has been conceived as ‘Did the students have the opportunity to learn the content tested?’. Those content coverage findings are crucial context information for interpreting student achievement, but of course the implemented curriculum represents more than only opportunities to learn content tested. In ESIE a much broader definition of the implemented curriculum was used, much more resembling the curriculum-in-action definition given in the beginning of this chapter. To some extent this broader definition can also be recognized as a guide to the TIMSS teacher questionnaires, but the latter instruments were designed and administered merely to obtain “some information about the implemented curriculum” (Beaton, Martin &

W. KUIPER, K. BOERSMA, J. VAN DEN AKKER

49

Mullis, 1997, p.53). Getting a thorough understanding of instructional practices using only written questionnaires is a utopian situation indeed. For that purpose, a larger array of research methods and instruments is needed, similar to what was done in ESIE. So, although there are differences in research focus between the two studies, in conceiving the implemented curriculum as well as in research methods and instruments, the findings that are available from TIMSS and ESIE as regards the implemented curriculum give support to the conclusion that the two studies are rather consistent. Main findings: Intended curriculum Mathematics education experts determined that, on average, 69% (1995) and 72% (1999) of the mathematics items from the TIMSS written test were appropriate for the attainment targets for mathematics at Secondary 2 level. Science education experts came to a comparable judgment for the science items (on average, 70% in 1995 and 69% in 1999). The appropriateness of the performance test was of about the same order: 9 out of 12 mathematics and science practical tasks were rated as matching the core objectives. As part of TIMSS-1999 (written test), the item ratings by mathematics and science education experts (intended curriculum) were compared with the ‘opportunity to learn’ item ratings by teachers (implemented curriculum). For that purpose the mathematics and science items were split up in a set ‘appropriate’ for the intended curriculum (111 mathematics items, 72%; 69 science items, 69%) and a set ‘not appropriate’ (44 mathematics items, 28%; 44 science items, 31%). Next, these four categories were cross-indexed with teachers’ ratings. Mathematics teachers generally appeared more positive in their ratings than mathematics experts. For science, however, the teachers’ ratings were generally consistent with the ratings made by the experts. ESIE showed that there was only a partial match between the intended curricula at school level – as appearing from analyses of textbooks, additional curriculum materials, schools’ work plans, teacher questionnaire and interview data, and lesson observations – and the attainment targets for mathematics, physics/chemistry, and biology. The match for the 15 subjects altogether (including mathematics, physics/chemistry, and biology) varies from, on the average, 40% for the lowest ability track to 59% for the highest track. The match for biology (32% - 46%) and physics/chemistry (33% - 54%) is less than the overall average in each student ability track; for mathematics (55% - 68%) it is the other way around (Peters-Sips et al., 2000). Another finding was that the general skill ‘conducting a simple inquiry’ is sufficiently part of the intended curriculum at school level for both biology (65% of the lessons) and physics/chemistry (62%), but not for mathematics (21%). Another general skill, ‘collaborating with peers’, is only sufficiently part of the intended school curriculum for physics/chemistry (62%); this general skill is only visible in 40% of the biology lessons and 46% of the mathematics lessons. A third component investigated in ESIE was the match between, on the one hand, the core objectives

50

TOWARDS A MORE CURRICULAR FOCUS

and the goals pertaining to general skills, and, on the other hand, textbooks in most common use. It appeared that textbooks have a substantial match with the attainment targets, but not with goals pertaining to general skills. ESIE also showed that: (i) textbooks generally reflect the attainment targets, but aims pertaining to general skills goals are much less visible; (ii) although teachers heavily rely on textbooks in their teaching practice, the implemented curriculum is only a slight representation of how attainment targets and general skills goals have been represented in textbooks. Reflective comments The TIMSS finding, that in 1995 and in 1999 Dutch students performed relatively well on the written mathematics and science test, which consisted of about 30% of items not appropriate for the attainment targets, indicates that the relatively good scores for these students are based on an item set that partly consisted of ‘policy irrelevant’ items. In addition, that about 30% of the items were not covered by the attainment targets, shows that more content has been tested than is covered by the attainment targets. Unfortunately, it has not been analyzed what that ‘more content’ refers to. However, another observation form ESIE, that teachers still teach ‘old’ content, nourishes the assumption that ‘more content’ refers to the old curriculum (i.e. that which preceded the basic secondary education era). So, the new still seems to be blended with the old. This seems to happen not only at the level of the intended curriculum but also at the implemented curriculum level. The latter can be inferred from the ESIE finding that the implemented curriculum matches poorly with textbooks which, in their turn, match well with the attainment targets and general skills goals. Next to this, it is not unlikely that variables in the socio-cultural context, as well as student characteristics, are influential. The TIMSS finding, however, that about 70% of the mathematics and science items from the written test match with the attainment targets, raises the question whether all attainment targets have been represented in the written test and, if not, which attainment targets have been omitted. Also no data are available on this topic. However, that the TIMSS written test does not fully cover the attainment targets goes without saying. 4. DISCUSSION In the foregoing we have tried to analyze conflicting images of the quality of lower secondary mathematics and science education in the Netherlands as appearing from two large-scale studies, TIMSS and ESIE. From this analysis a number of conclusions can be drawn, and some additional reflective comments made. It was emphasized already that the two studies differ in regard to the research object. In TIMSS the attained curriculum is the primary focus; in ESIE the emphasis is on the implemented curriculum. Our analysis has made clear that this difference in research object results not only in differences in outcomes of the two studies but also in difficulties in explaining those differences. However, research methods and instruments also differ. As a consequence, it is impossible to make comparisons

W. KUIPER, K. BOERSMA, J. VAN DEN AKKER

51

between a number of findings, a problem that becomes clearly obvious at the level of the attained curriculum. A third difference between the two studies deals, as it seems, with conceptualizing the curriculum. Due to its main interest in student achievement, in TIMSS the primary focus is on test-curriculum matching issues; in ESIE the researchers seem to rely on a broader curriculum concept. Notwithstanding these important conceptual and methodological differences, there seem to be findings at the level of the implemented and the intended curriculum, that give support to the conclusion that between the two studies there is also some ground in common. The area of the common ground, though, is small. It is so small that no proper explanation can be given for the expected occurrence of the process of ‘curricular dilution’ in ESIE findings vis-à-vis the unexpected non-occurrence (or better: the reverse) of this phenomenon in the TIMSS written test findings. Curriculum dilution appears from the ESIE findings in terms of the following areas: (i) unsatisfactory student achievement on basic secondary education tests (attained); (ii) in the context of teaching approaches in which the promotion of active learning is still lacking (implemented); (iii) against the background of attainment targets (intended) that only partially match with the intended curriculum at school level. In these findings a dilution process is visible that is consistent with the curriculum transformation process depicted in Figure 1. However, the main TIMSS findings seem in contrast, as, again roughly speaking, Dutch students perform relatively well on the written test (attained) – the appropriateness of the written test to the implemented curriculum was good for mathematics and less good for science (implemented), while almost one third of the items were rated not appropriate to the attainment targets (intended). A good explanation for the non-occurrence of the dilution process in the TIMSS findings cannot be given. When commenting and reflecting on the results from the two studies pertaining to the intended and implemented curriculum, the TIMSS findings (with support inferred from ESIE) seem to point more in the direction of the occurrence of a process of ‘curricular blending’. A further reflection on the curriculum typology, taken as the stepping-stone in our analysis, in relation to one of the main results of TIMSS brings us to a further comment. In TIMSS a partial match was found between the international test and the attainment targets for mathematics, physics/chemistry, and biology. This finding might suggest that a full match is something for which one should strive. However, as experiences with a national option mathematics test (administered in 1995 in addition to the written test; Kuiper, Bos & Plomp, 2000) have taught us, a proper match is not a guarantee for proper student achievement. Even more important, trying to realize a full coverage of the attainment targets in tests seems to be a kind of a top-down approach that doesn’t make sense in the Netherlands. It seems to be more fruitful to take the intended curriculum as a guide. A partial match between the test and the intended curriculum is not a problem. Via the attainment targets, an intended curriculum for mathematics, physics/chemistry, and biology has been framed that can and should be perceived as an area within which schools can make and account for their own choices. Such an approach fits the national government’s

52

TOWARDS A MORE CURRICULAR FOCUS

new education policy to give schools more autonomy and responsibility in making their own curricular choices. Enlarging schools’ autonomy may at the same time be a lever in creating more dynamics in the transformation process as depicted in Figure 1. Quite typical for the Netherlands is that, so far, the arrows between the three curricular appearances predominantly point to the right (from intended to attained) instead of also to the left (from attained to intended). In conclusion to our analysis we make two final comments. First, the linking of student performance on international tests, which are administered as part of studies like TIMSS (and PISA), also to national standards (ESIE) is to be preferred above making comparisons with only an international mean as reference. Such an approach puts achievement results in a nationally relevant perspective, which in turn is a prerequisite for inferring meaningful policy implications aiming at the improvement of the quality of mathematics and science education. Second, in international comparative studies there is also needed a broader conception of curriculum than mainly ‘content (to be) taught and learned’ and ‘goals and objectives (to be) achieved’. In presenting our conceptual framework, we have emphasized that content and goals/objectives are only two of nine components to which a curriculum can pertain. Data on the match between ‘content tested’ (attained), ‘content taught’ (implemented), and ‘content to be taught’ (intended) are very relevant, but a broadening of curricular focus to teaching practices, especially, will provide vital clues for interpreting student performance (see focus of Inspectorate Study). The implemented curriculum (still reflecting traditional features as ESIE clearly shows) is the link between the intended and the attained. In TIMSS, however, this curricular appearance currently it is too much like a black box to provide a worthwhile frame for interpreting (discrepancies between) the attained, implemented, and intended curriculum. In more practical terms this means that, if possible, there should be an attempt to connect future studies like TIMSS and ESIE to each other in such a way that the strengths of both are exploited. We have indicated in this paper the most salient issues to consider for this. In exploiting the strengths of several studies, it is probable that less energy will be needed to disentangle conflicting images. REFERENCES Beaton, A.E., Martin, M.O. & Mullis, I.V.S. (1997). Providing data for educational policy in an international context: The Third International Mathematics and Science Study (TIMSS). European Journal of Psychological Assessment, 13 (1), 49-85. Boersma, K.Th. (2000a). Het leerplan van de basisvorming als problem. In M. Peters-Sips, J. van der Linden & A. Wald (Eds.), Verder werken aan de basis. Basisvorming bundelt krachten (pp. 43-56). Utrecht: Inspectie van het Onderwijs. Boersma, K.Th. (2000b). Oorzaken en aanpak van overladenheid van het operationele curriculum van de basisvorming. Tijdschrift voor Onderwijsresearch, 25 (1/2), 110-117.

W. KUIPER, K. BOERSMA, J. VAN DEN AKKER

53

Bos, K.Tj. (2002). Benefits and limitations of large-scale international comparative assessment studies: The case of IEA’s TIMSS study (doctoral dissertation). Enschede: University of Twente. Bos, K.Tj. & Vos, F.P. (2000). Nederland in TIMSS-1999. Exacte vakken in leerjaar 2 van het voortgezet onderwijs. Enschede: Universiteit Twente. Bos, K.Tj., Kuiper, W. & Plomp, Tj. (1999). Student performance and curricular appropriateness in the Netherlands. Studies in Educational Evaluation, 25, 269-276. Inspectie van het Onderwijs (1999a). Werk aan de basis. Evaluatie van de basisvorming na vijf jaar. Algemeen rapport. Utrecht: Inspectie van het Onderwijs. Inspectie van het Onderwijs (1999b). Bijlagen bij ‘ Evaluatierapport onderwijsleerproces basisvorming’. Utrecht: Inspectie van het onderwijs. Inspectie van het Onderwijs (1999c). 3: Biologie in de basisvorming. Evaluatie van de eerste vijf jaar. Utrecht: Inspectie van het Onderwijs. Inspectie van het Onderwijs (1999d). 15: Natuur- en scheikunde in de basisvorming. Evaluatie van de eerste vijf jaar. Utrecht: Inspectie van het Onderwijs. Inspectie van het Onderwijs (1999e). 19: Wiskunde in de basisvorming. Evaluatie van de eerste vijf jaar. Utrecht: Inspectie van het Onderwijs. Kellaghan, T. (1996). IEA studies and educational policy. Assessment in Education,3 (2), 143-160. Kuiper, W. (1993). Curriculum reform and teaching practice (doctoral dissertation). Enschede: University of Twente. Kuiper, W., Bos, K.Tj. & Plomp, Tj. (1997). Wiskunde en de natuurwetenschappelijke vakken in leerjaar 1 en 2 van het voortgezet onderwijs. Nederlands aandeel in TIMSS populatie 2. Enschede: Universiteit Twente. Kuiper, W., Bos, K.Tj. & Plomp, Tj. (1999). Mathematics achievement in the Netherlands and appropriateness of the TIMSS mathematics test. Educational Research and Evaluation, 5 (2), 85-104. Kuiper, W., Bos, K.Tj. & Plomp, Tj. (2000). The TIMSS national option test mathematics. Studies in Educational Evaluation, 26. Peters-Sips, M., Zwarts, M., Van den Berg, H. & Schuurmans, L. (2000). Kwaliteit van het vakspecifieke aanbod. Tijdschrift voor Onderwijsresearch, 25 (1/2), 40-52. Robitaille, D.F., Schmidt, W.H., Raizen, S., McKnight, C. Britton, E. & Nicol, C. (1993). Curriculum frameworks for mathematics and science. TIMSS Monograph No. 1. Vancouver: Pacific Educational Press. Schmidt, W.H., McKnight, C.C., Houang, R.T., Wang, H., Wiley, D.E., Cogan, L.S. & Wolfe, R.G. (2001). Why schools matter. A cross-national comparison of curriculum and learning. San Francisco, CA: Jossey-Bass. Schmidt, W.H., McKnight, C.C., Valverde, G.A., Houang, R.T. & Wiley, D.E. (1997). Many visions, many aims (Volume 1). A cross-national investigation of curricular intentions in school mathematics. Dordrecht: Kluwer. Schmidt, W.H., Raizen, S.A., Britton, E.D., Bianchi, L.J. & Wolfe, R.G. (1997). Many visions, many aims (Volume 2). A cross-national investigation of curricular intentions in school science. Dordrecht: Kluwer. Van den Akker, J.J.H. (1998). De uitbeelding van het curriculum (orational address). Enschede: University of Twente. Van den Akker, J. (2003). Curriculum perspectives. An introduction. In J. van den Akker, W. Kuiper & U. Hameyer (Eds.), Curriculum landscapes and trends (pp. 1-14). Dordrecht: Kluwer.

54

TOWARDS A MORE CURRICULAR FOCUS

Van den Berg, H., Zwarts, M. & Peters-Sips, M. (2000). Kwaliteit van het onderwijsleerproces. Tijdschrift voor Onderwijsresearch, 25 (1/2), 20-39. Vos, P. & Kuiper, W. (2004, in press). Trends (1999-2003) in the TIMSS mathematics performance assessment in the Netherlands. Educational Research and Evaluation. Westbury, I. (1992). Comparing American and Japanese achievement: Is the United States really a low achiever? Educational Researcher, 21 (6), 18-24. Wijnstra, J.M. (2001). Bruikbare kennis en vaardigheden voor jonge mensen. Nederlandse uitkomsten van het OESO Programme for International Student Assessment op het gebied van begrijpend en studerend lezen, wiskunde en de natuurwetenschappelijke vakken in het jaar 2000. Arnhem: Citogroep.

PART 2 Science curriculum innovation

40 YEARS OF CURRICULUM DEVELOPMENT

JON OGBORN University of London, UK

ABSTRACT I discuss a number of features of world-wide science curriculum development, including the extent to which each development is local and specific, the relationship to issues and ideologies current at the time, the question of 'top-down' versus 'bottom-up' development, the role of didactic inventions and creativity, the relationship of development to research, and the question of ownership.

1. PERSONAL INTRODUCTION Thirty-five years ago I was asked to lead – with Paul Black – a national curriculum development project in the UK. That was Nuffield Advanced Physics (Ogborn, 1971). I thought of it as a unique experience. Then, thirty years later, I was asked to do the same again, for the Institute of Physics project Advancing Physics (Ogborn & Whitehouse, 2000; Ogborn & Whitehouse, 2001). Rarely is anyone invited to make the same mistakes twice over. At the risk of over-personalising what I have to say, it is from this standpoint that I have chosen to look back over forty years of curriculum development in the sciences. Before I get started, I should commend to you another account, by Myron Atkin and Paul Black, of their experiences in curriculum change, in their book Inside Science Education Reform (Atkin & Black, 2003). 2. CURRICULUM DEVELOPMENT WORLD-WIDE Since the late 1950s and early 1960s there has been a huge amount of science curriculum development work, varying widely in scale and in motivation. Some sought to refresh science teaching; some to make it more efficient. Some have been adopted (or imposed) nationally; others have made their way as a free choice on the part of schools. I have to be careful. I’m much better acquainted with efforts in the UK and in the English language, than with others. I have a natural bias towards work in physics education. Moreover, I have learned over the years that you cannot understand curriculum change in any country, without a clear understanding of the culture and the specific historical circumstances. This I signally lack in many cases. 57 K. Boersma et al. (eds.), Research and the Quality of Science Education, 57—65. © 2005 Springer. Printed in the Netherlands.

58

40 YEARS OF CURRICULUM DEVELOPMENT

I offer my remarks, not as a carefully worked-out theoretical scheme, nor as a well-researched narrative, but rather as a patchwork of thoughts that occur to me as I look back and reflect. 3. LOCAL SPECIFICITY The Devil, it is said, is in the details. This seems to be very true of curriculum development. The early large-scale developments in the USA, notably the Physical Sciences Study Committee (PSSC, 1960), Harvard Project Physics (Rutherford, 1970), Chem Study (Campbell, 1962) , the Chemical Bond Approach (CBA, 1962), and the Biological Sciences Curriculum Study (BSCS, 1959), all hoped to have an influence well beyond the confines of the USA. So they did, but more often through the fact of their existence than through direct adoption in other countries. The first reason is simple: they were all finely tuned to the needs of the American educational system. PSSC made good sense for a system in which high school students began their first substantial study of physics at age 16. But in the UK, where physics was taught from age 11, it made little sense. As a result, the sponsors of PSSC complained about it not being “translated into English”. A second reason has to do with ownership and creativity. The main reaction of teachers and educators in European countries to these US projects was to want to try to do it for themselves. Local pride, and local awareness of essential subtleties, played an important role. As a result, over the 1960s and 1970s, a variety of projects burgeoned throughout Europe: for example PLON in the Netherlands (PLON, 1985) and “Ask Nature” in Denmark (Thomsen, 1978), besides the dozen or more projects sponsored by the Nuffield Foundation in the UK. Each was very specific to its time and place. New teaching programmes have to be a very good fit to local circumstances, taking account of different structures of schooling, of different times available for teaching, of the varying prior knowledge of students, of the expectations and preparation of teachers, of official rules and regulations. But could we not all agree about the essential structure of physics, chemistry, or biology, and about good ways to approach the central concepts, and then tune these in detail to local circumstances? It turns out not to be so. Just as good architectural solutions often arise from turning disadvantages of an awkward site to positive advantage, so good educational solutions often capitalise on local problems and constraints, turning what looks like a difficulty into an opportunity. An example might be the emphasis in the UK projects on first hand laboratory work for students. UK science teachers found their school laboratories full of old pre-war apparatus. Students disliked the excessive amount of theory, with concepts not much linked to experimentation. The solution was to develop new equipment and to promote the notion of exploratory play with apparatus. This kept pupils and teachers happy, and was in tune with the general empiricism of Anglo-Saxon culture. The developers were surprised to find that teachers in France, Italy, Spain, or Portugal were unimpressed, giving rigorous theory a much higher valuation than did the empiricist English.

J. OGBORN

59

Perhaps the general message is that we are all rather blind to the specificities of our local circumstances. They are “just how things are”, and we are surprised when we find that they are very different for others. 4. ISSUES, IDEOLOGIES AND SLOGANS Development projects naturally address current educational issues. In the UK, the main educational issue actually changed while the first Nuffield projects were being developed. It changed from being how to develop a lively up-to-date science for selective secondary schools, to being how to develop a convincing science program for the all-ability comprehensive schools just then being introduced. The UK had, up until the 1970s, a divided system of secondary education. About 25% of the school population was selected for the academic “grammar” schools. The rest went mainly to “secondary modern” schools, whose curriculum was at best loosely specified. The Nuffield Foundation’s projects initially focused on the science curriculum for the selective schools. This was certainly in need of repair – dull and routine, with its structure largely inherited from the great 19th century textbooks. The new slogan was “Science for All”. But, like most slogans, it did not mean what it said. It meant, science to appeal to all the 25% selected for grammar schools, not just to future specialist scientists. It did not remotely mean science for students of all abilities. However, during this period the movement to replace the divided system by a comprehensive schooling system, actually “for all”, gathered strength. Thus in the UK, the issue became how to develop science courses genuinely designed for the whole school population. This became something of a national obsession, not shared by other countries. One slogan devised for this was “Relevance”. Complex issues need complex solutions, but they generally get simple slogans to encapsulate and make memorable these solutions: “Relevance”, “Ask Nature”, “Science for All”, “Hands On”, “Science Workshop”, “Learning by Doing”. Mao Zedong had a genius for inventing them, in a very different context. Be wary of these slogans. They are needed, even essential, to help people remember the point and perhaps to focus energy and enthusiasm. But they rarely speak plainly. I remember being asked near the start of my second development project Advancing Physics, what its slogan would be. I was at first embarrassed to find that I had no good answer. Maybe “Variety”, I said – if you want to appeal to more people you have to offer more ways of being attractive. The answer suggests its own limits. It cannot be right to focus a whole course on being attractive, at any cost. So there must be a basic truthfulness to the nature of the subject – in this case physics. But now this isn’t a slogan, but the statement of a complex problem. I can’t say that I’m sorry, even if it makes it hard to tell people what is the “essential new idea” behind Advancing Physics. In fact, I’m suspicious of any educational development that passionately believes in its own slogans. I don’t much believe in one-shot solutions – ‘magic bullets’.

60

40 YEARS OF CURRICULUM DEVELOPMENT 5. ‘TOP-DOWN’ VERSUS ‘BOTTOM-UP’

I vividly recall my introduction to the question of whether curriculum change should proceed from the top down – from experts to teachers – or should be bottom up, collecting ideas and good practice from teachers themselves. I was sitting in a grassroofed hut in the Kruger National Park in South Africa, alongside Professor Dieudonné, one of the famous Bourbaki mathematicians. We were in South Africa to talk about changes in the science and mathematics curriculum: I to talk about Nuffield Advanced Physics and Dieudonné to talk about the changes in the mathematics curriculum in France. He learned that I was a secondary school teacher. Graciously but sceptically, he asked me where the key ideas for Nuffield Advanced Physics came from. Who guided our work from above? Proudly I answered, “From us – from the team, all of us teachers”. “No”, he replied, “You misunderstand. Who really supplies the main ideas, the fundamental basis of the course?”. I gave the same answer. “Impossible”, he said, “New ideas come from the University – par definition.” Yet in fact, from his point of view, he was right. In mathematics, the desire for change stemmed from deep changes in mathematics itself. The Bourbaki mathematicians and others had sought to place mathematics on an entirely new rigorous foundation. So mathematicians found school mathematics almost unrecognisable as mathematics. They wanted a fresh start, beginning for example with the logic of sets. France was not alone in this movement for “the new mathematics”. The idea was sweeping the world. It has to be said that the introduction of ‘modern mathematics’ was not a complete success. Parents were disturbed to find their young children coming home from primary school talking of things the parents had never heard of: sets, unions, disjunctions. Many older teachers felt that alien ideas were being imposed on them; that their hard-won teaching skills were suddenly valueless. Nevertheless, the changes that had taken place in mathematics were real and were valuable. Gradually, as new teachers replaced older ones, some at least of the new thinking became naturalised in schools. A recent change in the sciences is the growing importance of digital imaging, together with new ways of imaging structures down to the molecular scale. Besides its many applications, digital imaging and communication is a whole new subject matter for which teaching methods need to be created. We attempted some of this in Advancing Physics, to the point of starting the physics course with an ultrasound image of a baby in the womb. Sooner or later, changes in scientific subjects start to affect the school science curriculum. In the case of biology, it has been sooner rather than later: DNA is, at 50 years of age, already firmly part of school biology. In physics, change is patchy, often later rather than sooner. Some glamorous parts of astronomy are present, if only as an option; so are simplified accounts of the quark structure of nucleons and mesons. But, with rare exceptions, the revolution introduced by quantum field theory remains unremarked; so indeed in large measure do Maxwell’s equations, and relativity, ancient though both are.

J. OGBORN

61

Thus, some curriculum change in science and mathematics is necessarily ‘topdown’. To return briefly to the hut in the Kruger National Park, it was disingenuous of me to tell Professor Dieudonné that Nuffield Advanced Physics simply worked ‘bottom up’, from teachers’ own ideas. Certainly we avidly collected ideas from the best teachers we could find. Fundamentally, though, this project was also ‘top down’, in the sense that the course was designed and built by a small central team, and then disseminated through a process of trials, and supported by a large scale training programme over several years. Not all necessary changes in the curriculum derive from changes in the subject matter. Often, the problems lie elsewhere, in changes in the nature of schooling and of society. In some such cases, the natural way to work is ‘bottom-up’, from teachers’ own expertise and ideas. An example, again from the UK, is the Secondary Science Review (West, 1983). Directed by Dick West, this did not attempt to create central teaching materials to solve its problem. Its problem was whether there could exist a viable science course that might meet the needs of all secondary pupils. The Review set about collecting and describing examples of good practice, and making them more widely known. It was driven by its own ideology, that of valuing the expertise of the practitioner. And indeed, it did succeed in building many groups of increasingly self-confident teachers who were made to feel that their efforts were valued and valuable. I cannot say that a large body of high quality teaching material emerged in this way. Indeed, the project published some rather unremarkable stuff. But that was not really the issue. The issue was political: to persuade parents, head teachers, other teachers, and local and central government officials that solutions could be found and that teachers could be trusted to find them. In this, the Review succeeded. I am sure that there are, and will be, other examples where ‘bottom-up’ is best. An instance may be the use of computers in science teaching, particularly computerbased laboratory work. This does involve changes of a fundamental kind, but changes essentially of classroom practice. Having teachers invent ways of exploiting these devices, and making their ideas widely known, may well be the best way forward. Let us not forget, also, the large amount of ‘invisible’ curriculum development that goes on through the pages of teachers’ journals, and at meetings for teachers, where good ideas are presented and exchanged. Indeed, I would think that any country should give high priority to stimulating such an infrastructure, to support and develop a sense of professional community amongst science teachers. The Internet offers scope for doing more in this direction. Involving teachers directly in curriculum development is widely seen as the right way forward. Myron Atkin and Paul Black (1996) report how, in the majority of the international sample of development projects that they surveyed for the OECD, considerable responsibility was devolved to teachers for deciding content and approaches. At the same time, I think that there remains a role for strong leadership and vision. Teachers will identify with a new course, not only because it came from

62

40 YEARS OF CURRICULUM DEVELOPMENT

other teachers, but also because it offers something strong and inspiring with which to identify. It really helps to think that you are part of something important. 6. INVENTIONS Every way to teach a given idea or skill was once invented by someone, and passed on to others. In science their traces are often to be seen in the science teaching apparatus stored in the laboratory cupboards. Often new technologies suggest new ways to teach. In chemistry, one such was the introduction of glassware for small-scale preparation of compounds. In biology, schools had to equip themselves for doing microbiology. The digital revolution swept away most of the analogue electrical meters in schools. And now, computerbased instrumentation has the power to change the way we teach much of experimental science. I am not sure how the mobile telephone and digital camera will change the way we teach about electromagnetic waves and digital communication, but I’m sure they will. My own personal interest, however, has been more in the invention of new ways to construct and present theoretical arguments. There are many important parts of science that languish untaught in schools because the theory is simply too difficult. One example is thermodynamics. The purely macroscopic theory is highly abstract and inaccessible. The statistical microscopic theory is more easily interpreted, but seems to require difficult statistical arguments. Thirty years ago, in Nuffield Advanced Physics, we found a way through these difficulties, using random simulations. I may as well tell you the origin of this line of thinking, to illustrate the chancy nature of didactic invention. In an early conversation about Nuffield Advanced Physics, Paul Black and I agreed that thermodynamics was probably too difficult for us. So I ‘wasted’ time dreaming of possible answers. Books like Henry Bent’s The Second Law showed that thermodynamics need not be dull and unintelligible. A chemist told me of his way of introducing the Boltzmann distribution, which struck me as incorrect. He agreed, but said that he didn’t know how to do better. Then, one sunny afternoon in Worcester, I had the idea of moving plastic chips representing quanta of energy around on a grid whose sites represented oscillators in a crystal. Astonishingly, the Boltzmann distribution seemed to appear. Paul Black recruited a mathematician to prove that the idea was right, and everything fell into place. Today, these ideas are alive and well in chemistry courses in the UK. A second long-standing obsession of mine has been inventing ways of exploiting the computational approach to solving differential equations to simplify the teaching of mechanics and other topics. This obsession also started in a very unlikely way. About January 1966, I was worrying about how to teach the wave mechanical account of the hydrogen atom. It occurred to me that one could solve the timeindependent radial Schrodinger equation very simply, step by step. This could be done graphically, without any heavy arithmetic or algebra. I was overjoyed to see the form of the radial wave-function for the ground state emerging on my graph paper.

J. OGBORN

63

Then I started worrying about how to reach that point with a class of students. I saw that the same graphical methods could be used for the first order equation for exponential decay, and for the second order equations for uniformly accelerated motion and harmonic oscillations. This was obvious to me because I had been lucky enough to be taught at Cambridge by the great Douglas Hartree, whose lectures inspired us with the notion that very simple step-by-step arithmetic methods could both solve difficult problems and illuminate their inner structure. The result was that Nuffield Advanced Physics used computational methods to understand simple differential equations, at a time when the only way for a school to use a computer was to send a deck of Hollerith cards to the computer centre of a university or a commercial company. The more general question now is whether computational modelling can radically simplify and illuminate the reasoning needed in mechanics and other problems. It seems clear to me that it can. I have no insight at all into why this didactic invention has proved so difficult for most teachers to accept. Anyway, for me this process of la transposition didactique is as fascinating and intellectually demanding a process as anything I know. But it is also wayward and subject to chance, as is anything creative. 7. RESEARCH At this meeting, you will be expecting me to tell you how crucial research in science education has been for curriculum development, and how important it is that research underpins future development. Could there be a hint of your self-interest here? The fact is that research has been important, but only in a limited number of cases. In France, the curriculum in optics was reformed on the basis of very good research by Laurence Viennot and her colleagues into problems of understanding light. Paul Black and Wynne Harlen devised a primary science programme based on their research project SPACE (Black & Harlen, 1990). In the USA Lillian MacDermott, Joe Redish, and Barbara White are amongst those who have built teaching materials around research results (see for example Redish, 2003). Important though these efforts are, I remain a shade sceptical. Research can often point the way to the existence of a problem. It less often points directly to the solution. An example is that we can now be quite sure, from a massive body of research, that students find Newton’s laws unbelievable, and create for themselves ideas about forces needed to keep objects in uniform motion. I have my ideas about where the deep difficulty lies; so no doubt do you. But none of us seem to be able to break through. This means being modest about what research can contribute to curriculum development, and admitting that there are cases where insight, intuition, experience of teaching, and deep knowledge of the subject are at least equally valuable sources of ideas about how to teach.

64

40 YEARS OF CURRICULUM DEVELOPMENT

Finally, I must mention the crucial area of research in assessment and evaluation. I shall say no more about it here, though, for lack of space, and because I don't feel that I have anything interesting to say about it, despite its importance. 8. OWNERSHIP Stimulated by the experience of first leading the Advancing Physics team and then of standing back, letting go and watching the teachers of the course take over, I recently wrote about the question of who owns a curriculum development (Ogborn 2002). I conclude that it is the teachers who teach and transform it, not those who originally develop it. The reason is the absolute inevitability of teachers transforming any set of didactic intentions and suggestion, in the act of turning them into real everyday teaching. There simply is no such thing as “doing exactly what the developer intended”. There was a time, after the first wave of curriculum development, when developers saw (often with horror) what was happening to their materials, and started to speak about creating “teacher-proof” materials. A vain hope, it has turned out, and for good reason. “Putting across the ideas of a project” is essentially a matter of communication. The 'obvious' or 'common sense' view of communication is that ideas should be transmitted clearly. This view of communication underpins much of commercial and political life, particularly the notion of 'Accountability'. But for me, communication is, always and everywhere, transformative. My role, as curriculum developer, becomes like that of a person who says, “Be reasonable, see it my way”. It is reasonable for me to tell people, as clearly and forcefully as I can, what would be my way. But it is equally reasonable for them not to agree. Indeed, this (“my way”) is not even possible: there is certain to be some transformation, however small, in the act of communication. One of the strongest conclusions to come out of decades of studies is that innovations succeed when teachers feel a sense of ownership of them. The seemingly simple question, “If this is the best, why should everybody not do it?” has to be given a subtle answer, namely that “the best” is an elusive thing, not always the same for everybody. A teacher willingly and enthusiastically teaching an “inferior” course, will do a better job than if obliged to teach a “better” one. 9. COMETH THE MOMENT The possibility of curriculum development depends on being lucky in catching the right moment. There are times when teachers are ready for change. There are times when the political will is there. There are times when the resources can be found. I was very lucky to be around at two times when curriculum development was possible, even welcomed. The first occasion was one when the example of what could be done was shown by the USA and combined with a new post-war sense of the desirability of change, encouraged the Nuffield Foundation to put substantial resources into science education. The second occasion arose because there was to be

J. OGBORN

65

a generally welcomed and overdue broadening of the curriculum. This combined with the fact that the professional association of physicists, the Institute of Physics, was worried about the decline in numbers of students taking physics, and had (briefly) some money to spare. So Advancing Physics happened. Thus, my final message to those who would like to be involved in curriculum development is: ‘be lucky’. Do your best to live in interesting times. REFERENCES Atkin, J. M. & Black, P.J. (2003). Inside Science Education Reform. Buckingham: Open University Press and New York: Teachers College Press. BSCS (1959). BSCS Biology NY: Kendall Hunt Black, P.J. & Atkin, J. M. (Eds.) (1996). Changing the subject. London: Routledge. Black, P.J. & Harlen, W. (1990). ‘Primary Science in the National Curriculum: the SPACE Approach’. Links, (15) 3, 17-20 Campbell, J. A. (Ed.) (1962). Chem Study. Berkeley: Lawrence Hall of Science. CBA (1962). Chemical Bond Approach. NY: Chemical Education Publishing Co. Ogborn, J. (Ed.) (1971). Nuffield Advanced Physics. Harmondsworth: Penguin. Ogborn, J. & Whitehouse, M. (Eds.) (2000). Advancing Physics AS. Bristol: Institute of Physics Publishing. Ogborn, J. & Whitehouse, M. (Eds.) (2001). Advancing Physics A2. Bristol: Institute of Physics Publishing. Ogborn, J. (2002). ‘Ownership and Transformation: Teachers using curriculum innovations’. Physics Education, 37 (2), 142-146. PLON (1985). Curriculum materials. Utrecht: University of Utrecht. PSSC (1960). Physics. NY: D C Heath. Redish, E. F. (2003). Teaching Physics with the Physics Suite. NY: John Wiley. Rutherford, F. J. (1970). Project Physics. NY: Holt. Thomsen, P. (Ed.) (1978). Ask Nature. Copenhagen: Royal Danish School of Educational Studies. West, R. (1983). Science Education 11-16: Proposals for Action and Consultation. London: Secondary Science Curriculum Review.

CHARACTERISTICS OF MEANINGFUL CHEMISTRY EDUCATION HANNA WESTBROEK, KEES KLAASSEN, ASTRID BULTE, ALBERT PILOT Utrecht University, The Netherlands ABSTRACT In this paper we elaborate on three potential strategies to promote meaningful chemistry education: using relevant contexts, offering content on a need-to-know basis, and making students feel that their input matters. We illustrate that it is educationally worthwhile to incorporate these characteristics, through our work on a particular chemistry module. Such emphasis leads to concrete, empirically based designs of modules and to heuristic guidelines for educational design decisions. It also productively informs further theorizing, such as an improved conceptualisation of the relations between the three characteristics. We therefore suggest that the type of investigation discussed in this paper, and the scenario-based design method which goes along with it, deserves a more prominent place in science education research.

1. INTRODUCTION From several analyses of science education, three main problematic features of student learning emerge: ‘rhetoric of conclusions’ (Schwab, 1962; De Vos et al., 2002), ‘incoherencies’ (Roberts, 1982; De Vos & Pilot, 2001), and ‘lack of student input’ (Lemke, 1990). It is argued that these problematic features, which play a role at the levels of the entire curriculum, of one module, and one lesson, contribute to an experienced alienation and lack of sense of direction for students. Several attempts have been made to pay attention to these problematic features by designing more ‘meaningful’ chemistry education, ranging from more or less isolated projects, designed as small-scale experiments or welcome additions to the existing curriculum, to projects that aim to reform the curriculum. Some well-known projects are: PLON, Salters’, ChemCom, Chemie im Kontext, The Wide River Project, and Chemistry in Products. Even if not explicitly labelled as such, in all of these projects three intertwined ‘characteristics of meaningful education’ generally can be seen to play a role: (1) context, (2) need to know, and (3) attention to student input. We adopt these characteristics, at least tentatively, as offering potential solutions for the problematic features indicated. With respect to the context characteristic, it can be argued that a well defined, and, for students, recognisable context will motivate them and provide the concepts involved with a distinct function and therefore meaning. When the emphasis has shifted from ‘getting an overview of the conceptual products of chemistry’ to the ‘functionality of concepts in relation to a certain relevant, recognisable context’, the ‘rhetoric of conclusions’ and ‘incoherencies’ features can be avoided. Instead, a consistent development of 67 K. Boersma et al. (eds.), Research and the Quality of Science Education, 67—76. © 2005 Springer. Printed in the Netherlands.

68

CHARACTERISTICS OF MEANINGFUL CHEMISTRY EDUCATION

concepts might be achieved. It can also be argued that addressing students’ questions on a need to know basis will provide for an increasing involvement of students in the teaching-learning process, because they will see the point of what they learn every step of the way. This need to know characteristic can avoid ‘incoherencies’ and students asking ‘why are we learning this?’. Together with the context characteristic, the development of a consistent emphasis might be enabled. The attention to student input characteristic is closely related to the need-to-know characteristic. If one proposes to incorporate a need-to-know approach in the design of a teaching-learning process, attention to input from students seems inevitable if the latter are to experience the functionality of ‘what comes next’. Obviously, this characteristic addresses the problematic ‘lack of student input’ feature. It is one thing to argue for these three characteristics as providing general design directions for meaningful science education. It is quite another to work out or incorporate these characteristics in concrete designs of science modules so that the characteristics contribute in an empirically justified and theoretically explainable way to solving the identified problems. In this paper we first of all illustrate, with our work on a particular chemistry module, that this emphasis on characteristics is educationally very worthwhile. It leads to non-trivial findings and challenges, and may also productively inform further theorizing. We close with a plea that this often neglected and underestimated area deserves a more prominent place in educational research. 2. GIVING CONTENT TO THE THREE CHARACTERISTICS OF MEANINGFUL EDUCATION Research strategy Taking as the object of our research ‘to adequately incorporate the three characteristics of meaningful learning in concrete designs of science modules’ has had several implications. Firstly, it has determined the questions we aim to answer. For instance, do students really experience what they are doing now, as enabling them to get a better understanding of this context (need to know), is the context really motivating enough to make them want to reach that understanding (context), and will they really feel that their input matters in reaching that understanding (attention to student input)? If the answers to these questions is yes, do they do this for the reasons we intended; and if not, can we understand why and improve matters? Secondly, if we want to answer such questions we need a specific, suitable research strategy: scenario-based developmental research (Lijnse, 1995). In what we call a ‘scenario’, we give an argued expectation about what will happen with respect to every step of the designed teaching-learning process. These expectations are described down to the level of expected answers, emerging discussions, questions raised with students, and so on. This includes an explanation of why this will contribute to the intended aims. In the case at hand, it concerns especially the steps in which we have incorporated one or more of the ‘characteristics of meaningful education’.

H.B. WESTBROEK, C.W.J.M. KLAASSEN, A.M.W. BULTE, A. PILOT

69

The scenario also serves two research purposes. First, in the design phase it forces us actually to think through in detail, and as well as we can, what students will make of the designed activities. Secondly, the scenario forms the framework for the interpretation of the implemented teaching-learning process. Discrepancies between the intended and implemented teaching-learning process give us indications of where further research is needed, which can lead to adjustments in the design, the scenario, and/or the assumptions underlying the design. Such adjustments should be tested in a next cycle. This paper concerns our research efforts with respect to just one chemistry module, which is about ‘water quality’. Our aim is to give an exemplar of our research strategy and of the kind of findings and challenges which have arisen. Within the scope of this paper, we cannot give more than impressions and cannot do justice to the detailed design and evaluation processes on which they are based. We first give an overview of the research design. We then present some design features of the first version of our module. We subsequently present some findings with respect to the three characteristics, and how we think this has led to some theoretical progress. Finally, we give some preliminary findings with the third version of our module. Research design The first version of the teaching sequence was tried at two different schools. School A is a protestant school, situated in a small village near Utrecht. The general teaching approach is traditional, with largely teacher centred, whole class instruction. Two teachers, A1 and his class of 19 students, and A2 and his class of 23 students, were involved. The classes consisted of 14-15 year old higher ability students, and the trials took place in April-May, 2001. Shortly afterwards (MayJune, 2001) the first version was tried at the other school. School B is situated in Amsterdam and bases its pedagogy on the Montessori vision: students work at their own pace, individually or in small groups, and the teacher guides these processes, giving students a lot of individual attention. In this trial, one teacher, B1, and his class of seven 15-16 year old higher ability students were involved. The second version of the teaching sequence was tried five times. The trials took place at the same two schools A and B, with the same three teachers, A1, A2 and B1, and with students of the same (i.e. higher) ability level. The teaching sequence was first put into practice at school B, in March-April of 2002. Three classes of respectively 25, 25 and 28 students were involved. Shortly afterwards, in May-June of 2002, the teaching sequence was put into practice in two classes at school A: a class of teacher A1, with 27 students, and a class of teacher A2, involving 28 students. Preparation of teachers Looking back, the teachers were prepared for their roles in a rather superficial way. In fact, the role of the teacher was not explicitly worked out until the third version of the design. In two successive meetings the actual learning materials (a syllabus), the

70

CHARACTERISTICS OF MEANINGFUL CHEMISTRY EDUCATION

teacher guide, and the intended outcomes were discussed with the teachers. The guide provided an overview of the general ideas behind the design, the learning goals, a possible planning of each lesson, a time indication, and a proposal for an instructional format (class discussion, group work, individual work, etc.) for each learning activity. Before every lesson, the teacher’s planning was discussed (together with the designer/ researcher) and it was decided in broad outlines how he (the teacher) could ‘set the stage’ for the learning activities (which we evaluated afterwards). Data All lessons were observed and audio-taped, and all group discussions among students were audio-taped. Students were asked questions during activities about what they were doing and thinking. All students’ work sheets were collected; all students did a final test and were asked to fill out an evaluative questionnaire. A selection of students was interviewed at the end of the teaching sequence. All meetings with the teachers, including those just before and after a lesson, were audio-taped. Data analysis As we will illustrate below, it soon became clear in the trials with the first version that the didactical structure and outline of the teaching learning activities would need a thorough revision. A detailed analysis of the data was therefore pointless. Clues for revision were based on comparing more general observations of the implemented teaching learning process (also using information from the evaluative interviews with teachers and students) with more general expectations of the sequence of activities. The implemented teaching learning process of the second version, however, was compared with the detailed expectations and justifications of each learning activity (the scenario). Class observations provided ‘first impressions’ but the primary information sources were the audio-taped class discussions and discussions between the students. Together with students’ work sheets, the implemented teaching learning process was reconstructed. Interviews mainly helped to clarify some of the students’ utterances and to verify interpretations. Design features of the first version of the module The module Water Quality is intended for 14-15 year old pre-A level students. It takes eight 50-minute lessons and by the end should lead students to the formulation of a criteria-based judgement about the quality of water samples. Its first version was designed around a framework inspired by Rivet et al.. (2000), and consisted of a driving question and sub-questions. The driving question, which was meant to contribute to the context characteristic, was: Is the water in our neighbourhood clean enough? From this driving question we carefully derived the sub-questions, by putting ourselves in the knowledge-position of students and thinking about what they still needed to know before they would be able to answer the driving question (need to know). This led to sub-questions like: What is the water used for? What are relevant parameters and norms? How can we test these parameters? How accurate are the tests? Answers to the sub-questions were supposed to lead gradually to an

H.B. WESTBROEK, C.W.J.M. KLAASSEN, A.M.W. BULTE, A. PILOT

71

answer to the driving question. We tried to realise the attention to student input characteristic by letting each group work on ‘its own water’ and by letting them present their own findings to their classmates in a peer-discussion. These presentations, moreover, were to contribute to the insight that there is ‘a common procedure’ for testing the quality of waters having different functions. Findings and challenges with respect to ‘context’ Students were very interested in the project and in the driving question. In general, they seemed to work enthusiastically and purposefully. We concluded that there were no real challenges with respect to establishing a relevant context for students. In the second version of our module we therefore also used a driving question to realize the context characteristic, though we did rethink the idea of sub-questions for reasons related to the need-to-know characteristic. Findings and challenges with respect to ‘need to know’ The first version of our module embodied the need-to-know characteristic because students were expected to see the sub-questions as a functional step to answering the driving question. In this sense we expected our approach to lead to a meaningful integration of content and context. It turned out, however, that students had not actually experienced the relevance of quite a lot of the sub-questions and therefore of the content involved. For example, they did not integrate the concepts they addressed in the section called ‘Principles and accuracy of test method’ in the context of answering the driving question. They just did the section and never thought about it again when doing their own tests on their water samples. What did we learn from these experiences? The sub-questions, each relating to some content matter students would need to know before they could work on their projects, had been carefully thought out and were designed to use and built on prior experiences and knowledge of students. The problem was that whereas we did know, for example, that in order to arrive at a reliable judgement of water quality, students would have to get a deeper insight into the accuracy and reliability of the test methods, students did not know this at the time the issues of accuracy and reliability were raised in the module. In this sense it can be said that we had incorrectly assumed that they would almost automatically share our motive for introducing the issues of accuracy and reliability in the light of the driving question.* Some of our findings also contained clues as to how students might have experienced the usefulness of what they were to learn. One group of students had a test result only just within the norm and had judged their swimming water as ‘good’. When they presented and discussed their findings with the class, an argument arose concerning whether one could really conclude this. At that point the sub-question, *

Here we do not just wish to illustrate the (not very surprising) point that the perceived logic (by students) of a teaching sequence can differ from that intended by the developers. If we had not made our detailed expectations explicit, and had not observed the whole teachinglearning process in detail, the findings reported probably would have stayed unnoticed, given that on the whole students were highly motivated and actively involved.

72

CHARACTERISTICS OF MEANINGFUL CHEMISTRY EDUCATION

‘How accurate are the tests?’ did become relevant for them in the light of their attempts to answer the driving question. At that point it would have been worthwhile to learn about the accuracy and reliability of the test methods. We concluded that learning activities should be designed in such a way that they give rise to a ‘knowledge need’ that students can appreciate as such, and on content-related grounds at the time it is raised and not later (as with the example of accuracy just mentioned), or not at all (as with many of the other sub-questions). The specific ‘knowledge need’ raised in one learning activity is to be elaborated upon in the next, which in turn is to induce a further ‘knowledge need’ and so on, till eventually the (top-down formulated) learning goals are reached. In this respect, Klaassen (1995) refers to the successive development of ‘content related motives’ in students. Perhaps it is important to emphasize that the aim is not merely to generate some motivation or arouse some curiosity. Most teachers will feel that this is part of their job, and cherish successful moments. Our aim, however, runs deeper than this. It is to design activities thoroughly in such a way that students both develop and follow a content-related sense of purpose, and to research systematically whether the result is as expected. We hope we have clarified that this poses a non-trivial challenge. We have tried to meet it in the second version of our module. Findings and challenges with respect to ‘attention to student input’ In the first version of our Water Quality module, we tried to realise the attention to students input characteristic by letting students test ‘their’ own water and present their own results. It did contribute to students feeling that their input mattered. However, when the second version of our module, in which we had paid special attention to the need-to-know characteristic, was put into practice, an additional important aspect of the attention to students input characteristic became apparent. This was the need for a much more detailed approach on the level of interactions between teachers and students. It turned out that the students seemed to appreciate the logic of the unfolding learning activities more when the teachers in their interaction with students paid serious attention to students’ input on the level of each learning activity. The problem was that two of the three participating teachers tended to ignore this input. They had not been sufficiently prepared for the fact that the input of students on this level of detail was an important characteristic of the designed teaching-learning process; nor were they sufficiently prepared for how, in the concrete details, such a process should be guided. The teachers relied on their own teaching style which was dominated by what Lemke (1990) has called a ‘triadic dialogue structure’. These teachers also frequently used the content as a means to control the class. As a result the motivation to get a deeper insight in a certain content shifted from ‘wanting to know in the light of the driving question’ (content related motives) to ‘important for the test’. The students of these teachers quite often failed to appreciate the logic of the learning activities in contrast to the students of the third teacher involved, who appreciated his students’ input much more. As part of the design of the third version of our module we therefore tried to find solutions to the following non-trivial problems:

H.B. WESTBROEK, C.W.J.M. KLAASSEN, A.M.W. BULTE, A. PILOT – –

73

Which interaction structures between teachers and students promote (in each specific learning activity) students’ perceptions that their input matters? How should such interaction structures in the teaching-learning process be implemented?

Reconceptualisation of the relations between the three characteristics After the second cycle, we began to rethink the relations between the three characteristics, in particular the connection between context and need to know. When viewed in isolation, the way we had embodied the context characteristic seemed to be proper, in the sense that students were interested in the driving question and were generally motivated by it. When viewed in relation to the need-to-know characteristic, however, we began to doubt this conclusion. Although the idea of creating content-related motives through learning activities had helped in the sense that sub-questions in the second version were increasingly raised by students, we still felt that the idea of a driving question was not a sufficiently strong guideline; that is, that students eventually had to be able to answer the driving question was not in itself sufficient to help us determine the nature and order of motives to be created. We concluded that we had to find another way of giving content to the context characteristic than through the use of a driving question. A new way, moreover, that would fit nicely with the ‘content related motives’ way of giving content to the need-to-know characteristic. Partly inspired by the interpretation that Van Oers (1998) and Van Aalsvoort (2000) have given to the concept of context, we have now come to think of it as a ‘communal enterprise or practice’. The idea is that by gradually involving students in an ‘instructional form’ of an existing chemical practice, the need to know characteristic could be realised in the following sense. In order to be able to perform actions effectively in that emerging instructional form of a chemical practice, students will gradually have to extend their knowledge and skills in the direction of the learning goals we have set. That is, the knowledge and skills to be learned would then be functional for participating in the ‘instructional practice’ (cf. Bulte et al., 2002). We will now elaborate this idea by tentatively formulating some conditions under which developing an instructional form on an existing practice seems appropriate; appropriate that is, both in the sense of motivating students and providing them with a general sense of purpose (as a wellchosen driving question managed to do) and in the sense of fitting well with the other characteristics (as the driving question failed to do). An obvious first condition then is that students appreciate the purpose or interest that is served by the existing practice. This condition seems to be met by the practice of ‘testing and judging water quality’, given that the aim of this practice concerns such human needs as clear drinking or swimming water. An instructional form of this practice can begin to emerge by giving students real water samples derived from real situations in which judging water quality plays a central role. The broad purpose and sense of direction within this instructional practice is then established: to gain knowledge about what is involved in deciding if water quality is good enough for a certain goal by simulating in some relevant respects the original practice. If students

74

CHARACTERISTICS OF MEANINGFUL CHEMISTRY EDUCATION

can identify with the roles they are given in this practice, this will stimulate their involvement. Moreover, to the extent that playing their role will be necessary for them to find out (what is involved in deciding) if the quality of the water samples is good enough, this will contribute to the realization of the attention to student input characteristic. A second condition has to do with linking the context to the need to know characteristic. It involves the typical heuristic or procedure that is used in the existing practice in order to attain the purpose it serves. In the case of judging water quality, for instance, this procedure involves steps like: determining the water quality demands based on the water’s use, testing the water, comparing the test results with the appropriate norms and parameters, and so on. The second condition is that this typical procedure should already be familiar to students, at least in a rudimentary form. Working through this procedure should indicate what the next logical step in the process will be. In this sense the procedure can provide the strong guidance that the driving question could not. It continually drives students to take the next step in the procedure or to find out something that will enable them to take this next step. We think that in the case of testing and judging water quality, the condition just mentioned is satisfied. Students do have at least a rough sense of the steps involved in the procedure to arrive at such a decision, even though they may lack the specific chemical knowledge to take those steps. For example, they are very much aware that water can contain all kinds of contaminants, and that it is logical to test whether the water is contaminated. What they do not know, at least in any detail, is what the contaminants might be and how to test for them. At this point, students can derive the relevant information from the existing practice. The design of the third version of our module was guided by the ideas just mentioned. The challenge lay in exploiting students’ intuitive appreciation of the logic of the successive procedural steps (which may have a different sequence than in the actual chemical practice), in a subtle interplay between the existing chemical practice and its evolving instructional form, so that students successively stumble on a gap in their knowledge that they know needs to be filled in order to take the next step in the procedure and finally to complete it. Preliminary findings with the third version of the module The third version has recently been put to the test with two new teachers, using interaction structures to promote a feeling among students that their input matters. Data have not been analysed in detail yet, but our first impressions are that students were motivated to know more about what is involved in the practice of testing and judging water quality. They mostly appeared to experience the logic and usefulness of the successive learning activities, which now induce content-related motives by using students' intuitive notions about ‘the next step of the procedure’. We observed students acting as we expected them to in the scenario; that is, the designed teaching-learning activities in general gave rise as intended to student questions and discussions. Also, the teachers were prepared much better for their role. Of course, all of this needs extensive backing. As before, a detailed analysis of the data may

H.B. WESTBROEK, C.W.J.M. KLAASSEN, A.M.W. BULTE, A. PILOT

75

force us to withdraw some of the first impressions given here; it will certainly give a much richer picture. 3. DISCUSSION We think our findings have broader implications, even though the previous discussion concerned only one chemistry module which was tested in only a few classes with only a few teachers. Like us in testing the first version of our module, many researchers report success in establishing relevant context by means of interesting and challenging themes, story lines or driving questions (Campbell et al., 1994; Schwartz, 1999; Rivet et al., 2000). Few of those researchers, however, have at the same time studied the realization of the need-to-know characteristic in any detail. To the extent that they have, the findings are in line with our experiences described above, namely ‘that students did not see a connection between the content presented in the class and the context’ (Rivet et al., 2000). Similar findings are reported about some PLON units (e.g., Eijkelhof, 1990). We therefore hypothesise that our re-conceptualisation of the characteristics of context, need to know, and attention to student input, by means of which we think to have tightened the connections between them, may also be useful for others. The concept of an ‘instructional form of a chemical practice’ may serve as a kind of framework for the design of modules, as in fact it does in our group (cf. Bulte et al., 2003). This is not, of course, in the sense of providing an algorithm, but as providing heuristic guidelines for arriving at didactical decisions. It is an evolving theoretical framework which both directs and is informed by detailed empirical classroom research concerning the question, whether a design lives up to its intentions. We especially want to stress this latter point. Even though within the scope of this paper we have not been able to present detailed analyses, our findings are backed by, and could not have been reached without, a detailed investigation of the relationship between the intended and implemented teaching-learning process. This type of investigation is rarely done. Mostly design plays a subservient role, and the teaching-learning process as such is not investigated, but rather some general issue like ‘motivation’ or ‘quality of argumentation’. Research that addresses in some detail the teacher-student interaction generally only does so in the context of traditional science education. It is for this reason that Lijnse (2001) has called the type of investigation discussed in this paper, the scenario-based design method that goes along with it, and the kind of theoretical framework it delivers, a ‘forgotten dimension in science education’. REFERENCES Bulte, A.M.W., Klaassen, C.W.J.M., Westbroek, H.B., Stolk, M, Prins, G.J., Genseberger, R. , De Jong, O. & Pilot, A. (2002). Modules for a new Chemistry Curriculum: Research on a Meaningful relation between Contexts and Concepts. Paper presented at the Second international IPN-YSEG Symposium: Context-based Curricula, October 10-13, 2002, Kiel, Germany.

76

CHARACTERISTICS OF MEANINGFUL CHEMISTRY EDUCATION

Bulte, A.M.W., Klaassen, C.W.J.M. & Pilot, A. (2003). Involving students in a communication practice about healthy and poisonous effects of substances Paper presented at the fourth International Conference of the European Science Education Research Association, August 19-23, Noordwijkerhout, The Netherlands. Campbell, B., Lazonby, J., Millar, R, Nicolson, P., Ramsden, J. & Waddington, D. (1994). Science: The Salters’ Approach – A Case Study of the Process of Large Scale Development. Science Education, 78, 415-447. De Vos, W., Bulte, A.M.W. & Pilot, A. (2002). Chemistry curricula for general education: analysis and elements of a design. In: J.K. Gilbert, O. De Jong, R. Justi, D.F. Treagust & J.H. Van Driel (Eds.), Chemical education: towards research-based practice (pp. 101124). Dordrecht: Kluwer Academic Publishers. De Vos, W. & Pilot, A. (2001). Acids and bases in layers. Journal of Chemical Education, 78, 494-499. Eijkelhof, H.M.C. (1990). Radiation and risk in physics education. Utrecht: CD-ß Press. Klaassen, C.W.J.M. (1995). A Problem-posing approach to teaching the topic of radioactivity. Utrecht: CD-ß Press. (http://www.library.uu.nl/digiarchief/dip/diss/01873016/inhoud.htm) Lemke, J.L. (1990). Talking Science. Norwood: Ablex. Lijnse, P.L. (1995). ‘Developmental Research’ as a way to an empirically based didactical structure of science Science Education, 79, 189-199. Lijnse, P.L. (2001). Didactics of Science: the forgotten Dimension in Science Education? In: R. Millar, J. Leach, J. Osborne (Eds.),Improving Science Education: the Contribution of Research (pp. 308-326). Buckingham: Open University Press. Rivet, A., Singer, J., Schneider, R., Kraijick, J. & Marx, R. (2000). The Evolution of Water: Designing and Developing Effective Curricula. Paper presented at the annual conference of the National Association for Research in Science Teaching. New Orleans, USA. Roberts, D.A. (1982). Developing the concept of ‘Curriculum Emphases’ in science education. Science education, 66, 243-260. Schwab, J.J. (1962). The teaching of science as inquiry. In: J.J. Schwab & P.F. Brandwein (Eds.), The teaching of science (pp. 3-103). Cambridge: Harvard University Press. Schwartz, A.T. (1999). Creating a context for Chemistry. Science and Education, 8, 605-618. Van Aalsvoort, J.M. (2000). Chemistry in products. Utrecht: CD-ß Press. Van Oers, B. (1998). From context to contextualizing. Learning and Instruction, 8, 473-488.

CROSS-CURRICULAR COLLABORATION IN TEACHING SOCIAL ASPECTS OF GENETICS

MARY RATCLIFFE, RICHARD HARRIS, JENNY MCWHIRTER University of Southampton, UK

ABSTRACT Science teachers can lack pedagogic skill and confidence in handling multi-faceted socio-scientific issues. This project explored the development, implementation, and evaluation of a ‘cross-curricular’ day as a suitable vehicle in eight different schools for both engaging 14-16 year old pupils in active consideration of social aspects of genetics and enabling science and humanities teachers to collaborate in planning and delivery. The cross-curricular research team planned a programme of activities, involving volunteer teams of teachers in development. Pupils in participating schools generally found the day stimulating, increasing their understanding of genetics and appreciation of social aspects. However, implementation showed that some teachers missed important learning opportunities as a result of lack of critical scaffolding of pupils’ discussions and limited expertise in ethical analysis. Cross-curricular collaboration was successful in presenting pupils with a holistic experience but had limitations in developing teachers’ expertise. Continuing professional development for both science and humanities teachers is needed to address socioscientific issues effectively.

1. BACKGROUND The research reported here was commissioned and funded by the Wellcome Trust – an independent biomedical research charity which aims to improve human and animal health. A previous Wellcome Trust project found that science teachers addressed social aspects of biomedical science infrequently and with lack of confidence. Although humanities teachers showed greater willingness to engage pupils in such discussion, few teachers of any discipline addressed ethical aspects of scientific advancements (Levinson & Turner, 2001). Levinson and Turner’s study recommended that a ‘collapsed’ or cross-curricular day in which science and humanities teachers collaborate in design and delivery might be an effective way of engaging pupils with socio-scientific issues. This suggestion for a ‘collapsed day’, including an integrated model of teaching and equal participation by all teachers, arose from a synthesis of teachers’ views through interviews but with little empirical basis for its construction and effectiveness. The aim of our project was thus to explore the feasibility and effectiveness of cross-curricular collaboration through the development, implementation, and evaluation of a programme for a ‘collapsed day’ on a biomedical issue, examining the barriers, opportunities, and outcomes at each stage. 77 K. Boersma et al. (eds.), Research and the Quality of Science Education,77—88. © 2005 Springer. Printed in the Netherlands.

78

COLLABORATION IN TEACHING SOCIAL ASPECTS OF GENETICS

The difficulties of addressing social and ethical implications of advances in scientific research are related to at least three factors: the nature of the socioscientific issues themselves; the pedagogical strategies adopted by teachers of different disciplines; the compartmentalised nature of the secondary curriculum. Aspects of genetics were chosen as a focus in this project because these emerged as most popular with teachers in an initial survey. Advances in gene therapy and genetic engineering raise issues of both private and public morality (Warnock, 2001). Of importance in considering, for example, the implications of genetic testing for individuals and society may be understanding of: the underpinning of genetics and the nature of science; the nature of decision-making processes, probability, and cost-benefit analysis; the nature of media-reporting; the social context of the issue; personal and societal value judgements and ethical reasoning. The multi-faceted nature of socio-scientific issues suggests that for individuals to develop an informed view on any issue they should have a good understanding of all the aspects (Ratcliffe & Grace, 2003). Each aspect can be explored individually within different subject areas, but this approach runs the risk that full consideration of the issue does not occur. A ‘collapsed day’ implies a holistic approach in which the different facets of the issue are brought together. Thus, one aspect explored in this project was the extent to which the different facets were addressed and supported. Socio-scientific issues raise pedagogical challenges for teachers in considering educational purpose and appropriate teaching strategies. There may be a hierarchy of purposes for considering an issue holistically: from sharing individual perspectives on the issue; reaching an understanding of the variety of available subjective responses; making a choice between differing values; to finding a rational resolution of the controversy (Bridges, 1979). Humanities and science teachers canvassed by Levinson and Turner (2001) gave a variety of justifications for teaching social and ethical aspects of biomedical science, with ‘sensitivity’ and decision-making being the most frequent. These reasons reflect the two extremes of Bridges’ (1979) hierarchy of purposes and imply some opportunity for discussion. Members of the research team have encountered many instances where the potential for discussion and analysis of socio-scientific issues has not been fully exploited in science classrooms, resulting in some cases in amorphous discussion or rapid decisionmaking (Ratcliffe & Grace, 2003). Teachers seem to make limited opportunities for pupils to engage in critical analysis of a socio-scientific issue. However, research evidence of innovative practice has provided some understanding of pupils’ use of values, beliefs, and scientific knowledge when dealing with socio-scientific issues in science lessons and how these relate to the pedagogical issues (Solomon, 1992; Gayford, 1993; Ratcliffe, 1997 & 1999). Such case study research highlights the need for an emphasis on the process of analysis of an issue. The research team considered critical peer group discussion as an important activity within the ‘collapsed day’ with the need for appropriate support by teachers. ‘Collapsed days’ give clear opportunities for collaboration between science and humanities teachers in supporting pupils’ learning. Throughout this project,

M. RATCLIFFE, R. HARRIS, J. MCWHIRTER

79

‘humanities’ was treated broadly as comprising English, history, geography, RE (Religious Education), citizenship, and PSHE (Personal, Social and Health Education). Cross-curricular collaboration in dealing with socio-scientific issues currently seems rare. In one example, Huckle (2001, p. 158) describes how geography and English departments in secondary schools co-operated in engaging with the genetically modified food debate, through pupils’ evaluation of information on relevant websites. It is of concern that science departments were not engaged in this initiative. Examination of media reports and social issues are more prominent in humanities curricula, which can result in discussion of socio-scientific issues without clear consideration of the underpinning science. There is limited research evidence to address factors in cross-curricular collaboration on socio-scientific issues. For example, Kerr (1999, p. 9) highlights the many gaps in our knowledge and understanding of citizenship education, such as pupils’ development of social knowledge and the relationship between pupils’ knowledge, attitudes, and beliefs. Principles of a holistic cross-curricular approach, in which the multi-faceted socio-scientific issue was supported by appropriate pedagogical strategies, underpinned the design of the ‘collapsed day’ programme. In particular, teaching strategies were encouraged which support pupil discussion AND ethical reasoning – i.e. allow pupils to engage with the complexity of the issue and to recognise and be able to apply the process of ethical reasoning in other contexts. Given the limited research base, the research questions focused particularly on the processes and outcomes of cross-curricular collaboration: • What are the learning outcomes for pupils from a ‘collapsed day’ on social aspects of genetics? • What are the gains and barriers for teaching and learning in cross-curricular collaboration? • What are the opportunities and barriers for teachers in their planning for and delivery of ‘collapsed days’ in principle and in practice? In answering these questions we sought to establish the feasibility of collapsed days as a method of effective engagement with a socio-scientific issue and the extent and value of cross-curricular collaboration as a possible teaching approach. This paper, for reasons of space, concentrates mainly on the first two questions. 2. METHODS The project had three phases: exploration of the feasibility in principle of ‘collapsed days’ for biomedical science; development of a collapsed day programme, in conjunction with participating teachers; evaluation of implementation of the programme in eight schools. This paper concentrates on the third phase. The research team designed an outline programme which was then shared with teachers and developed further, particularly in exemplifying and supporting teaching strategies. The framework shown in Figure 1 was designed to address key questions and have outcomes synthesised through 14-16 year old pupils engaged in team.

80

COLLABORATION IN TEACHING SOCIAL ASPECTS OF GENETICS

Activity Introduction Team working

Stimulus (video, keynote speaker, performance piece etc.)

Science – What is possible? Investigative activity How do we know? Evaluation of media reports Exploring viewpoints

Reflective peer group discussion

Purpose To provide information about the overall aims of the day. Team building. Identify pupils’ initial views of genetic issues. Engage pupils’ interest in a human dilemma involving genetic disorders. Identify key questions about the science and its impact on individuals and society. Gain better understanding of genes, genetic crosses, genetic engineering. Understanding of processes and practices of science.

Identify individual views on genetic testing and share these. Recognise the diversity of views and some of the issues that these might raise. Understand principles of ethical reasoning and decision making. Practise using ethical reasoning tools.

Teacher’s role Introduce the key aims.

Manage activity. Help pupils to clarify views. Organise the stimulus.

Manage pupils’ discussion.

Explain science concepts. Manage activity. Explain nature of science and media reporting. Introduce issue.

Manage whole class and peer group discussion.

Introduce particular genetic issue. Ethical reasoning activity Explain ethical principles. Manage peer group ethical analysis. Synthesise and present Help pupils synthesise Can we? Should we? Synthesis activities, e.g. arguments related to: elements. drama, debate, radio/TV What is possible? Manage peer group report, poster How should we decide? discussions and presentations. Figure 1 Framework for a collapsed day programme to consider implications of advancements in genetics How should we decide?

work. The research team devised activities or adapted existing resources to produce a full programme. Six schools were recruited to work with the research team during a development day in which the framework, activities, and teaching strategies were exemplified and discussed.In order to build in cross-curricular planning from the

M. RATCLIFFE, R. HARRIS, J. MCWHIRTER

81

outset, three teachers from each participating school attended, including at least one science teacher and at least one ‘humanities’ teacher. By insisting that teachers of different disciplines attended the development day, it was expected that the resulting collaboration in initial planning would be built upon by these teachers collaboratively taking the lead in organisation within the school. In addition, two schools undertook to deliver the ‘collapsed day’ programme without the benefit of participating in the development day.The expectation of cross-curricular planning was reinforced with these two schools through suggestions from the research team of how planning could be developed. The inclusion of these two schools allowed some exploration of issues of teacher development and ownership A number of research instruments were used to collect data about the implemented programme and its impact on pupils and teachers. Pupils’ reactions to the collapsed day were canvassed in three ways in each school: i. Two classes were followed during the event and field notes were made by two researchers, focusing on pupils’ engagement in the activities and the nature of teachers’ actions. ii. Pupils in these two classes completed a questionnaire which sought their views on learning, interest, and participation. iii. Focus group discussions were held with two small groups of pupils and explored pupils’ views on their learning, interest, and participation, and added further detail to the evidence from observation and pupil questionnaires. Questionnaires were also administered to all participating teachers, focusing on pupils’ learning and motivation, cross-curricular collaboration, and management and logistics. Also, in each school in informal discussion, teachers shared their immediate reactions to delivery of the programme with the research team. Focus group discussions were transcribed and coded reflexively for major themes in pupils’ responses. A qualitative data software package (NUD*IST NVivo) was used to assist in the mechanics of coding to allow exploration of the extent to which themes were common to different groups. For each school a member of the research team used all relevant data, including materials given to pupils, to build up an extensive document describing pupils’ and teachers’ experiences of the programme in that school. Another member of the research team, who observed the programme, read through and verified, or amended following discussion, the summary of that school’s experience. These extensive portraits, coding of focus group discussions, and quantitative analysis of questionnaires formed the evidence base for the evaluation. 3. RESULTS Across the eight schools delivering the programme, there was considerable variation in adaptation of the programme and its implementation. The two schools whose teachers did not attend the development day incorporated more elements of the original design but still had quite different programmes. All schools planned to use

82

COLLABORATION IN TEACHING SOCIAL ASPECTS OF GENETICS

some type of stimulus to start the event, had at least one activity which explored knowledge of genetics necessary to understand the issue, and expected pupils to engage in discussion and/or debate in sharing their opinions. All but one school planned to use ethical analysis. However, even these common activities took different final forms both in the schools’ plans and in implementation. Only one school explicitly explored the processes of science. Pupil questionnaires and focus groups allowed identification of perceptions of the focus of the event and of detail of learning. Pupils in all schools recognised the event as being about aspects of advancements in genetics. In five of the eight schools, pupils also recognised the ethical dimension as prominent. Pupils at the start of the event were exposed to controversy, and expectations were raised that they would consider that controversy during the day. From experience of the day as a whole, pupils expressed their views of the main aspects they had learnt (Table 1). In all but one school, learning about genetics was perceived as one of the most highly rated learning gains. However, pupils had difficulty pinpointing exactly what it was they had learnt about genetics: [Focus group School 2] P1: I think it is important if you want to build your family to know if you have got the gene or if you are a carrier. I knew about it, but not in depth like this. I thought these were quite rare cases but it is not now. It is a bit scary. P1: I did not know much about it before, but I was against it. You know this genetic modification on crops and stuff, but today did help to stabilize what I thought. [Focus group School 4] P4:How genetic engineering can solve lots of problems cos I always thought it was a bad thing but it can like solve world hunger and that can’t it, I didn’t really know it was any good. [Focus group School 5] P2: The questions in the booklet they made you think like about cystic fibrosis. P1:And how to improve it by genetically changing your genes and stuff with gene therapy.

A notable feature of such discussions is the value judgements which are apparent in considering the genetics learnt (many utterances being coded, as above, as containing value laden or emotive comments). Thus, although pupils showed an emphasis on learning genetics, this was not divorced from the social implications. These extracts give an idea of the range of concepts which were addressed in different schools. It is interesting to examine what scientific background was being introduced or expected to be used. The concepts addressed in most schools included: chromosome; recessive and dominant genes; genetic crosses; DNA; inherited disorders - cystic fibrosis and Friedreich’s ataxia; genetic testing; embryo selection. In most schools, the process of genetic engineering was not explained in any detail. Rather an understanding of genetic modification was based on the principle of genes as the vehicle for determining an organism’s characteristics and capable of being changed in some way. This research project has raised an unresolved question – What level of scientific understanding is needed to discuss the implications of

M. RATCLIFFE, R. HARRIS, J. MCWHIRTER

83

advancements in genetics, particularly genetic engineering? Some appreciation of the level of understanding used may be gleaned from pupils’ input into ‘synthesis’ activities – those in the latter half of the programme drawing different elements together. Table 1 Pupils’ perceptions of what they learnt School 1 2 3 4 5 6

7

8

Learnt

n=40 n=25 n=32 n=14 n=37 n=33 n=19 n=41

Genetics

#* 1.9

* 2.0 2.4

#* 2.5

* 2.1

* 1.9

My own views

2.1

# 2.1 2.3

2.4

2.0

2.3

How to express views in different ways 2.5

2.4

2.5

1.9

2.3

How to decide what's right and wrong

2.3

2.3

2.5

2.0

People's rights and responsibilities

2.1

2.0

2.8

How people can deal with a complex issue

2.2

2.4

The processes of science – HOW it's done

2.4

How to work as part of a team 2.4

* 2.0

* 1.7

2.3

# 2.0

2.6

2.5

2.3

2.2

* 2.2

2.2

* 2.1

2.4

2.1

2.2

2.7

1.7

2.3

2.4

2.0

2.2

2.6

2.1

2.4

3.0

2.5

2.7

2.5

2.6

2.6

2.1

3.2

2.3

2.3

2.7

3.2

1.9

NUMBERS ARE THE MEAN FOR RESPONSES TO EACH CLOSED QUESTION ON THE QUESTIONNAIRE. 4 point scale used: 1= a lot; 4= a little Low means show greater learning. * Indicates learning gains which were mentioned with high frequency (> 30%) in open questions on the questionnaire. # indicates learning gains which were strongly supported in focus group discussions. Figures in bold-face show the 3 mostly highly rated learning in each school.

The framework presented to teachers allowed for flexibility in the way in which pupils might synthesise their learning outcomes, using two key questions – What is possible? How should we decide? Two main approaches to synthesis were adopted by schools: a debate involving half or full year groups; small groups working on specific tasks, each with a tangible outcome which could be displayed. Debate was used in three schools. For example, pupils in school 6 in half-year groups considered the motion ‘This house believes reproduction should be left to chance.’ In the debates, it was difficult to judge from observation the nature of the

84

COLLABORATION IN TEACHING SOCIAL ASPECTS OF GENETICS

learning outcomes for many of the pupils. Although many pupils enjoyed the debate, several commented on limited opportunities for active learning. In three other schools tangible products resulted from small groups of pupils working on particular issues. For example, in school 1 each class was given one of the following tasks: design of a message about genetic engineering to go on a T shirt; production of a TV debate involving pupils acting out roles of people taking different positions on human cloning; production of a powerpoint presentation showing arguments for and against genetic engineering. Pupils were encouraged to develop clear arguments and to use terminology clearly and correctly. Pupils engaged in their given activities with enthusiasm for the 90 minutes, and showed very little off-task behaviour. Teachers in questionnaire responses also supported the enthusiasm and sustained attention which pupils gave to the tasks. The presentations resulting from tangible products, as in the debates, were strong on pupils giving their opinions. Specialist terminology, such as genetic engineering, cloning, and cystic fibrosis, was used comfortably but without explicit discussion of the concepts. The nature of discussion both within pupils’ presentations and during group discussion seemed similar to that observed in previous research (Solomon, 1992; Ratcliffe, 1997); that is, scientific evidence was drawn into discussions with low frequency, but familiarity with science terms and concepts underpinned discussions. The fact that genetic engineering, for example, was used as an unproblematic term in pupils’ discussions suggests a sufficient understanding to engage with the issue. However, the exact nature of this ‘sufficient’ understanding is open to debate and worth further study. In the synthesis activities and in focus group discussions, pupils were able to show evidence of bringing together different views on issues involving genetics, i.e. supporting a view that pupils were able to come to an informed stance on the social impact of advancements in genetics, including recognition of their own views and those of other people. Having a focus on a tangible product seemed to promote active engagement by a wide range of pupils, providing opportunities for differentiation and acknowledgement of different learning styles. Pupils appreciated the opportunity to share perspectives and engage in discussion, with peer group discussion amongst the three most highly rated activities in each school for enjoyment and learning. It is worth examining the activities of peer group discussion and ethical analysis in more detail. These were areas where potential existed for effective cross-curricular collaboration between teachers with complementary skills, and for support for pupils’ learning. Within each school there were opportunities for pupils to develop opinions and share them with other pupils. The following two extracts are typical: In both classes observed, there was mixed response to the scenarios – some groups were able to carry out a discussion with minimal support; others needed substantial contact with a teacher to keep them on task and focused. A group in one class near the observer were overheard. They initially did get engaged with the problem – however the discussion ran out of steam fairly quickly and pupils then engaged in

M. RATCLIFFE, R. HARRIS, J. MCWHIRTER

85

off-task talk. Teachers circulated to the different groups but there was a tendency to stay with one group for a considerable period of time. [school 5 field notes] In groups of about 10 pupils had to brainstorm their ideas. The teacher moved from group to group to prompt and get pupils discussing ideas. In this class the pupils had little disagreement about their responses and little was made of inconsistencies within their thinking. For example, pupils were unhappy with the idea that the person sitting next to them might be genetically engineered but didn’t see a problem if there were no genetic mishaps in society. [school 2 field notes]

There was a sense from observations of the group discussions that scaffolding which encouraged the challenging of different viewpoints could have encouraged a more critical exchange of views. It appeared that there were some missed learning opportunities within the discussions. One limiting factor, perhaps inevitably, was time. In several schools, the time allocated for discussion was shorter than might have been needed to give due consideration to the complexity of the issue under discussion. There was perhaps an underlying, and in a few cases justifiable, fear that pupils would not sustain group discussion for any length of time. Teachers’ prior experience in supporting peer group discussion was unknown. However, it might be expected that what was seen reflected teachers’ normal practice or interpretation of the task rather than any development during cross-curricular planning. Thus, our expectations that humanities teachers might support peer group discussion more effectively than science teachers or that cross-curricular planning and implementation might result in support for peer group discussion were not borne out. Similar issues emerged in activities where pupils were engaged in ethical analysis but in these cases apparently there were more missed opportunities for scaffolding pupils’ learning. The process of using Goals, Rights and Responsibilities (GRR) as one method of ethical analysis was new to all teachers participating in the project. In this approach, an ethical dilemma involving a number of individuals is posed, for example: ‘Pat and Rick wish to have children. Rick is a carrier of sickle cell anaemia. Should Pat, his partner, be screened to see if she is a carrier? Her son, Tom, by a first marriage, is a healthy 4 year old.’ Rather than answer the question from their own feelings, groups of pupils consider the Goals, Rights and Responsibilities of one person in the scenario. Groups’ results are pooled and the conflicts between GRRs for different people exposed, showing the potential complexity of the decision-making. Those schools addressing the ethics of the genetics issue all chose to use the GRR approach. However, there were different interpretations and different levels of expertise. This is perhaps best exemplified by contrasts within the same schools: Pupils had to focus most of their discussion on a number of different scenarios, regarding GRR. In one observed class this was a poorly focused session, whilst in the other observed class some skilful chairing of discussion kept the momentum going. It seemed clear from observation that the staff had not fully understood the GRR issue and so didn’t fully explain the

86

COLLABORATION IN TEACHING SOCIAL ASPECTS OF GENETICS distinction between them, how they might come into conflict with each other, which then needs to be addressed someway. Pupils were able to produce some thoughtful responses, but little time was given to unpacking the thinking behind the decision-making processes of the pupils. [school 1 field notes] In one observed class, groups of pupils were given 10 minutes to complete this task. Pupils were given little explanation of G, R, R and it was not made clear to them how they would feed back their ideas. Many worked sensitively on this task. 10 minutes was given over to feedback to the class, though the quality of feedback was variable. In another class pupils were given 45 minutes for the task, had an example explained and were able to work in smaller groups. Once they were clear about G, R, R they were able to work thoughtfully. This class were able to discuss the nature of G, R, R well during feedback. [school 2 field notes]

The second extract shows a comparatively rare event – the teacher reflecting with the pupils on the ethical principles being used. The combination of giving the activity considerable time and drawing on the (RE) teacher’s own expertise allowed for clarification of the nature of ethics. Pupils’ level of understanding increased as a result, allowing pupils in focus group discussion to articulate that they had learnt about ethics: Messing about with genetics, is it right or wrong, is it morally right, if it is allowed. Ethical decisions which I did not know anything about before.[school 2 focus group] Where ethical analysis was led by a teacher with expertise in both ethics and managing discussion, and there was sufficient time to explore principles, the level of pupils’ critical engagement was high. Although having a task on ethical analysis allowed pupils an introduction to some ethical principles, the pockets of expertise within the participating schools were not always shared. For both peer group discussion and ethical analysis, the activity seemed for most teachers NOT to be part of their repertoire of teaching strategies, the exception being RE teachers. In most cases cross-curricular collaboration did not extend to an effective sharing of teaching strategies, i.e. there was not transfer of expertise between teachers of different disciplines. It may be unrealistic to expect crosscurricular planning of one event to allow not only sharing of the nature of activities undertaken in different disciplines, but also development of expertise in using processes undertaken in disciplines other than the teacher’s own. Prior experience in cross-curricular collaboration and the extent to which teachers had been involved in collapsed days in the past seemed to affect the focus of planning. Thus, teachers for whom both managing a collapsed day and cross-curricular collaboration were new, perhaps inevitably focussed on aspects of organisation rather than on sharing teaching expertise. However, 94% of teachers indicated that they valued the opportunity to share teaching expertise with colleagues in other departments, with 58% identifying something specific they had learnt from another curriculum area.

M. RATCLIFFE, R. HARRIS, J. MCWHIRTER

87

Although 22% of respondents saw no barriers to cross-curricular collaboration, just over half (52%) identified the time necessary for planning as the major drawback. 4. CONCLUSIONS AND IMPLICATIONS Two strands have been present in this project: - a ‘collapsed day’ as a vehicle for pupils’ engagement with a socio-scientific issue at a holistic level; - the nature of effective cross-curricular collaboration in supporting such an event. Pupils and teachers were positive about the collapsed day as providing a good opportunity for considering social aspects of genetics. There was evidence that pupils started to develop informed views on genetics’ issues, drawing on many of the facets expected in consideration of such issues, e.g. concepts of genetics; ethical aspects of decision-making; making explicit and sharing personal views. This was shown in pupils’ arguments presented as powerpoint presentations, in debates, or in role-plays where pupils were able to show some considerations for and against a genetic advancement, e.g. cloning, drawing on both scientific principles, and their value positions. Synthesis of the evaluation evidence, discussed above, suggests that the following aspects of the collapsed day are important in promoting pupils’ engagement in considering social and ethical aspects of biomedical science: • The study of one issue in depth of intrinsic interest to the pupils; • A novel stimulus which raises questions about the social and ethical applications of genetics; • Opportunities for pupils to voice and share their views, with this being most effective in small groups with a structure which supports critical discussion; • Ethical analysis in which pupils extend their appreciation of the moral dilemmas the issue raises and ways of addressing such complexity; • An activity centred around the construction of a tangible product, allowing pupils to synthesise their views actively. • Pupils working in teams as a feature to reinforce active learning and critical discussion The programme in design and implementation provided opportunities for these features to occur. However, the project has shown that expertise among both science and humanities teachers is patchy concerning the ability to consider issues holistically. RE teachers seemed best able to support effective consideration of the social aspects of issues. Teachers' differing expertise was not always shared effectively. For the participating schools, the experience of planning, sharing, and delivery was perceived as demanding of time, resources, and expertise. Nonetheless, the experience was regarded as positive in encouraging cross-curricular collaboration, with most schools indicating the likelihood of repeating the event after learning from this initial experience. However, it is not clear if these teachers, if they develop the programme further, will continue to focus on the content and tasks which they expect pupils to cover, or on the best methods of supporting learning though the latter would encourage the development of further clarity in the

88

COLLABORATION IN TEACHING SOCIAL ASPECTS OF GENETICS

teachers’ role and sharing of their expertise. The experience of this project suggests that the cross-curricular collaboration necessary in planning and delivering a ‘collapsed day’ programme allows some recognition of differing expertise but limited development of individual teacher’s skills. Considerable further professional development of both science and humanities teachers seems necessary to address socio-scientific issues fully, wherever they appear in the curriculum, as ‘collapsed days’ or in other ways. This professional development may be facilitated by bringing science and humanities teachers together in a structured programme which focuses on the means of supporting peer group discussion and ethical analysis. REFERENCES Bridges, D. (1979). Education, Democracy and Discussion. Windsor: NFER Publishing Company. Gayford, C. (1993). Discussion-based group work related to environmental issues in science classes with 15 year old pupils in England. International Journal of Science Education 15, 5, 521-529. Huckle, J. (2001). Towards ecological citizenship. In D. Lambert & P. Machon (Eds.) Citizenship through secondary geography (pp. 144-160). London: Routledge/Falmer. Kerr, D. (1999). Re-examining citizenship education: the case of England. Slough: NFER. Levinson, R. & Turner, S. (2001). Valuable Lessons: The teaching of social and ethical issues in the school curriculum, arising from developments in biomedical research – a research study of teachers. London: The Wellcome Trust. Ratcliffe, M. (1997). Pupil decision-making about socio-scientific issues, within the science curriculum. International Journal of Science Education 19, 2, 167-182. Ratcliffe, M. (1999). Evaluation of abilities in interpreting media reports of scientific research. International Journal of Science Education 21, 10, 1085-1099. Ratcliffe, M. & Grace M. (2003). Science Education and Citizenship Buckingham: Open University Press. Solomon J. (1992). The classroom discussion of science-based social issues presented on television: knowledge, attitudes & values. International Journal of Science Education 14, 4, 431-444. Warnock, M. (2001). An Intelligent Person’s Guide to Ethics. London: Duckbacks.

SCHOOL INNOVATION IN SCIENCE: CHANGE, CULTURE, COMPLEXITY

RUSSEL TYTLER Deakin University, Australia

ABSTRACT The School Innovation in Science (SIS) initiative has developed and evaluated a model to improve science teaching and learning across a school system. The model involves a framework for describing effective teaching and learning, and a strategy that allows schools flexibility to develop their practice to suit local conditions and to maintain ownership of the change process. SIS has proved successful in improving science teaching and learning in primary and secondary schools. Evidence of variations in the nature and extent of the change is used to argue that the process is essentially cultural in nature, and that change occurs at different levels within a school. Processes supporting change thus need to be flexible and responsive.

1. INTRODUCTION School Innovation in Science (SIS) is the largest school science initiative of its kind in Australia in decades. The project has been a major part of a set of initiatives developed by the Victorian Department of Education and Training (DE&T). During 2000 to 2002, the Deakin University-based research team worked with more than 200 primary and secondary schools to develop and trial a model for improving science teaching and learning in schools. The model has two major features: • The SIS Components, which represent a framework of effective science teaching and learning (Figure 1), and • The SIS Strategy, which is the process by which schools can improve their science teaching and learning (Figure 2). The SIS Strategy provides flexibility for schools and teachers to plan and implement initiatives based on the particular needs of the school within an overall framework provided by the SIS Components. School science teams are supported to identify and capitalise on their strengths and experience. Tests of student learning, attitudes, and perceptions have been used to monitor student progress and outcomes, and changes in classroom practice have been monitored by a teacher interview and verification process. The change process has been monitored using field notes, interviews, questionnaires, and reporting protocols.

89 K. Boersma et al. (eds.), Research and the Quality of Science Education, 89—105. © 2005 Springer. Printed in the Netherlands.

90

SCHOOL INNOVATION IN SCIENCE

In classrooms that effectively support student learning and engagement in science:

1. Students are encouraged to engage actively with ideas and evidence Students are encouraged to express their ideas and to question evidence in investigations and in public science issues. Their input influences the course of lessons. They are encouraged and supported to take some responsibility for science investigations and for their own learning. 2. Students are challenged to develop meaningful understandings Students are challenged and supported to develop deeper level understanding of major science ideas and to connect and extend ideas across lessons and contexts. They are challenged to develop higher order thinking and to think laterally in solving science-based problems. 3. Science is linked with students’ lives and interests Student interests and concerns are acknowledged in framing learning sequences. Links between students’ interests, science knowledge, and the real world are constantly emphasised 4. Students’ individual learning needs and preferences are catered for A range of strategies is used to monitor and respond to students’ different learning needs and preferences, and to their social and personal needs. There is a focused and sympathetic response to the range of ideas, interests, and abilities of students. 5. Assessment is embedded within the science learning strategy Monitoring of student learning is varied and continuous, focuses on significant science understandings, and contributes to planning at a number of levels. A range of styles of assessment tasks is used to reflect different aspects of science and types of understanding 6. The nature of science is represented in its different aspects Science is presented as a significant human enterprise with varied investigative traditions and constantly evolving understandings, which also has important social, personal and technological dimensions. The successes and limitations of science are acknowledged and discussed. 7. The classroom is linked with the broader community. A variety of links are made between the classroom program and the local and broader community. These links emphasise the broad relevance and social and cultural implications of science, and frame the learning of science within a wider setting. 8. Learning technologies are exploited for their learning potentialities Learning technologies are used strategically for increasing the effectiveness of, and student control over, learning in science. Students use information and communication technology (ICT) in a variety of ways that reflect their use by professional scientists.

Figure 1. The SIS Components of effective teaching and learning in science.

R. TYTLER

91

Many writers over the last decade or more have decried the predominance of short term workshops that traditionally count as teacher professional development in science. Numerous studies have shown that these ‘one-shot’ professional development events are ineffective in promoting changes in teacher and school practices (e.g. Hoban, 1992). Their ineffectiveness is related to the lack of followthrough, the lack of connection with school priorities or the direct needs and concerns of participants, and the lack of long term and systematic planning (Webb, 1993). Interview-based studies (Loughran & Ingvarson, 1993; Paige, 1994) indicate that teachers of science identify longer term professional development as being uniquely significant in changing their professional practice. Many writers (e.g. Hargreaves, 1994; Hall & Hord, 2001) have emphasised that change requires of teachers that they ground new ideas in their own personal experience. Joyce and Showers (1995), drawing on research from a large number of studies, argue strongly for the need to site professional development within the school context. They discuss professional development within a framework of cultural change and argue the need for social support, as people practice teaching strategies that are new to their repertoire or as they implement the difficult areas of a curriculum change. Contemporary large scale reform projects in a number of countries have tended to incorporate these principles (Beeth et al., 2003). While the primary focus of SIS is on changes in teacher classroom practice leading to improved student outcomes, the science team in the school is conceived as being the engine of change. Project support structures and advice have been chiefly at the science team level. This approach is consistent with a number of contemporary initiatives in Australian education. The support structures include an SIS Coordinator in each school, time release partly funded by the school, regional consultants and support networks, leadership training, and professional development for the science team to support the teaching and learning focus. The SIS Strategy is not resource-based, but is designed to allow flexibility for schools to adapt to local conditions and develop a sense of ownership. The SIS Components, for which considerable interpretive material has been generated, inform on the areas of focus and are a reference point for discussions about teaching and learning and to clarify goals and initiatives. (The process of developing the SIS Components has been described by Tytler, 2003 and Tytler, Waldrip & Griffiths, 2004.) The core of the SIS Strategy is the generation of an action plan that focuses on teaching and learning and takes into account particular school conditions and goals. The main steps in developing the action plan are (1) auditing science in the school: a range of information is collected from student tests and surveys, teacher interviews, analyses of the school curriculum and resources, and science team processes; (2) reviewing and prioritising: analysis of the key issues arising from the audit and identification of initiatives and goals; (3) developing and writing the action plan: the action plan specifies initiatives/actions to be taken and details how this is to be done, by whom, over what time scale, and how teachers are to be monitored.

92

SCHOOL INNOVATION IN SCIENCE

Teaching and learning framework The SIS Components

Infrastructure Support Network support Consultants, workshops

Support Materials Handbook, website

Professional Development For leaders and teachers

Research instruments Teacher, student review

Engagement Understanding Student lives Differentiation Assessment

Auditing Science in the School

Developing an Action Plan

Implementing Change

Improving Student Outcomes

Nature of science Community

Supporting Actions within Schools

ICT Committing dissemina organisational support

Managing PD

Supporting individuals & groups

Monitoring & evaluating

Reporting & dissemminating

Figure 2: The SIS Strategy This paper describes the various ways in which schools’ experiences of change have been researched, and the key elements and outcomes of the change process identified. The questions I address are: 1. 2. 3. 4.

What elements of the SIS strategy are effective in supporting change? What is the nature and extent of the changes that occur as a result of SIS? What local factors influence the success of the change process? How can we best characterise the process of change within a school?

These questions are critical for a number of reasons. First, SIS is currently being extended to more than 200 further schools (see below). The project will continue to provide infrastructure support, but it will not have the funding it enjoyed in the research years. The question of how best to target support has thus become critically important. Secondly, the extension of SIS into other curriculum areas has made it important to develop a clear view on just what are the critical change elements, and how they translate from science across to other settings. Thirdly, Australia is currently engaged in a debate concerning the best model to use for a national initiative in science, with some arguing for a model that is much more strongly

R. TYTLER

93

grounded in the development of a national curriculum resource. This argument has yet to be resolved. 2. METHODS The research methodologies in which the Project is engaged include experimental research, case studies, document analysis, and action research. The data collection and analysis has occurred at the levels of individual students (up to 29,600 students), teachers and classrooms (up to 2200 teachers), and the whole school (up to 220 coordinators and principals). The data collection processes relevant to the major issues addressed by this paper are detailed below: –

–

–

– –

Development and validation of the SIS Strategy: regular progress notes from SIS Consultants and SIS Coordinators, interviews with selected coordinators, focus group discussions with the project team and consultants, and SIS Coordinator questionnaires; Nature of school initiatives and experience: structured interim and final reports from schools, field notes and documents relating to regional workshops, and SIS Coordinator questionnaires and interviews; Nature and extent of the change in school science: a variety of questionnaires for coordinators, teachers, and principals, field notes, and consultant/researcher focus group discussions; Factors affecting change: consultant/researcher focus group discussions; Nature of the change process: interviews with selected coordinators and action research involving a small number of SIS schools to identify conditions for change.

Analysis of this material occurred throughout the life of the project, increasing our understanding of aspects of the change process, and providing feedback to schools in the form of advice and materials for principals, coordinators, and teachers. The results described below reflect the analyses carried out mainly at the end of the research phase of the project, after three years of research effort.

3. VALIDATION OF THE SIS STRATEGY Critical elements of the strategy were to (a) provide a language and process to promote a discourse centred on teaching and learning, and (b) challenge and support teachers to examine their practice and take ownership of the change process. A questionnaire given to Coordinators asked them to rate aspects of the strategy in terms of its usefulness (the rating scale was: 4 – Of critical importance; 3 – Very useful; 2 – Somewhat useful; 1 – Not very useful). All aspects were valued, but especially valued were the SIS Components (mean 3.4) and Component Mapping (3.1), staff meetings (3.4), and the action planning process (3.6). Infrastructure

94

SCHOOL INNOVATION IN SCIENCE

support areas which were particularly valued included the leadership training workshop (mean 3.4), regional workshops (3.2), the SIS Handbook (3.2), consultant support (3.7), and time release (3.8). (In the mature phase of the project 0.5 days per week is the recommended release time for coordinators in large schools.) These findings and associated comments were incorporated in planning for the extension of SIS in 2003 to provide advice to regions about support structures and the level and nature of teacher support needed within a school. The component mapping exercise was a powerful innovation. In this exercise SIS Coordinators interviewed each teacher to reach an agreed teaching and learning profile based on the SIS Components, . An excerpt from the component map is shown in Figure 3. Word descriptors are used to represent four different levels of exemplification of the components (some are divided into sub-components for clarity), and each teacher’s profile is constructed during an interview with the SIS Coordinator who clarifies and probes. The exercise caused teachers to think about what they had been doing in science and what they wanted to do in the future. SIS Coordinators valued the process for the direction it gave to the project: The teaching and learning review exercise … identified teacher strengths and areas that they would like to improve on … allowed teachers to identify and be open about their limitations and expertise … encouraged a more thoughtful approach to teaching and learning … encouraged the development of a shared vision of science. (From a review meeting of SIS Coordinators)

4. NATURE OF SCHOOL INITIATIVES AND EXPERIENCE Case stories of school initiatives and change issues were developed from school reports, workshop presentations, field notes, and interviews with SIS Coordinators. Many of these case stories that have been generated (Tytler & Nakos, 2003) illustrate the richness and variety of experiences encouraged by this approach, as well as some of the difficulties associated with implementing change. The initiatives can be organized into four broad types. •

Teaching and learning initiatives include catering for individual learning styles by using a greater range of teaching strategies, developing more student-centred approaches, developing more investigative approaches to practical work and promoting the use of higher order thinking through open-ended and problem solving tasks, and developing approaches that relate science to the real world and increase awareness of the role of science in contemporary society.

R. TYTLER

95

Students are encouraged and supported to express their ideas, question evidence, and raise issues Students often Student are Students are Shared and contribute to encouraged frequently open class structured to give short, encouraged discussion is discussions of clear and the basis for activities and responses to supported to all science targeted science express lessons. questions. concepts. opinions, to Students are The teacher raise issues, encouraged focuses on and to and supported clear question to express explanations evidence in their ideas of class and opinions, procedures discussions. to question and science evidence, and ideas. to raise issues about science and its applications. Student input (questions, ideas and expressions of interest) influences the course of lessons. Science Student Student Lesson ideas are input (e.g. response and sequences are presented items, structured planned to formally and stories, discussion is allow for are largely questions) sought to help flexibility clarify and text- or often leads based on worksheetmotivate to further student input based with students as part activities or (e.g. some limited extended of the questions, presentation of class discussion items, science ideas. discussion. during a interests that lesson. may extend into further activities or even lessons). Figure 3: Excerpt from the Component Map, for Component 1 ‘Active engagement with ideas and evidence’

96 •

•

•

SCHOOL INNOVATION IN SCIENCE Curriculum planning and organisation initiatives were an immediate focus for many primary schools for which physical and chemical science strands had previously been neglected. Many secondary schools focused on improving the documentation of teaching sequences and activities, including the use of the school intranet. Team planning became a key focus for teachers to develop a common view of curriculum and of teaching and learning strategies. Community initiatives focused on creating science learning experiences beyond the classroom. Community partnerships involved teachers and students working in association with local industry, in parks, on excursions or programs of visits, or with scientists. For example, a primary school worked in a new local wetlands park, the aim of which is to restore environmental conditions to be similar to those prior to European settlement. Secondary school students became involved in a wine production process in partnership with a local wine industry, in which they engaged in plant propagation and in chemical applications such as fermentation, microbiology, soil analysis, and weather studies. Information and communications technologies (ICT): Primary and secondary schools focused on embedding ICT during the planning process; they supported students to develop further their confidence and ICT skills. Activities were planned for students to collect and analyse data using data loggers and spreadsheets. Both primary and secondary students commonly used peripherals such as digital cameras, computer microscopes, and video cameras.

Of particular interest in the analysis has been the insight generated into the different cultures operating in primary and secondary schools. The circumstance of science in primary and secondary schools is very different in a number of respects, including the background of teachers and the very different histories of curriculum provision and organisation. Consequently, the project followed somewhat different pathways in primary and secondary schools, including: – the emphasis in secondary schools on classroom teaching strategies which focus on student engagement and individual learners and on different aspects of science; – the emphasis of primary schools on teacher knowledge and confidence, on meaningful understandings for students, and on the status and profile of science. The focus on individual learning and student engagement perhaps reflects an acknowledgment in secondary schools that teacher-centred, transmissive methods have predominated and need rethinking. This interpretation is consistent with numerous anecdotal reports from SIS Coordinators in secondary schools of the nature of the change agenda. The focus on meaningful understanding in primary schools most likely reflects the lack of confidence in science knowledge and a tendency within primary schools, again evidenced by observations from within the

R. TYTLER

97

project, to focus on activities without appropriate linking to significant science understandings.

5. NATURE AND EXTENT OF THE CHANGES DUE TO SIS There is compelling anecdotal evidence, supported by case descriptions, of significant change in the practice of science in both primary and secondary schools. A questionnaire used in primary schools has shown a doubling, on average, of the time spent on science. Interviews with principals and the questionnaires to which principals, coordinators and teachers responded, all attest to a variety of changes. In November 2002, SIS Coordinators and teachers were also asked to record the extent of their agreement or disagreement with statements concerning the overall success of SIS in a number of specified areas. Table 1 gives the percentage of SIS Coordinators and teachers in primary and secondary schools who agreed or strongly agreed with statements concerning the extent of the change. The results generally show agreement at levels of more than 80% for coordinators and 75% for teachers, with those in primary schools being particularly positive. Table 1: Percentage of primary and secondary school SIS Coordinators and teachers agreeing or strongly agreeing with statements concerning the success of the SIS Project in their school The project has been successful to date in: a. Increasing the profile of science in the school b. Improving the organisation and planning of science curriculum in the school c. Improving the way science is taught in classrooms d. Improving processes for assessing students’ science learning e. Increasing teachers’ enjoyment of teaching science f. Improving science learning outcomes for students g. Improving students’ attitude to science

Coordinators Prim. Sec. 79.6 80.5 100 92.7

Teachers Prim. Sec. 97.3 76.7 92.2 77.5

97.9

92.7

86.5

75.3

77.1

85.4

73.1

57.8

95.8

87.8

84.8

61.5

98.0

73.2

90.2

69.0

95.9

75.0

89.2

57.2

Change in teacher classroom practice Table 1 provides evidence of significant change in teacher classroom practice. There was also considerable anecdotal evidence, at workshops and through journals, reports, and consultant observations, that substantial change in classroom teaching and learning was occurring in some schools at least. A range of measures were used to establish the nature and the extent of the change, and to provide validity checks through triangulation. Some of these are described below.

98

SCHOOL INNOVATION IN SCIENCE

A profile of teacher classroom practice was established through the component mapping process, described above (Figure 3); by this means changes in practice were tracked. In interviews teachers stated their position (based on a maximum score of 4) on each of the Components, in terms of the degree of exemplification of the Component as represented in their teaching and learning practice (Figure 4). The mean scores in Figure 4, calculated across all Components, are generated with different cohorts of teachers; the score for 3 years is less certain than the scores for shorter periods, given the smaller number of teachers who were in the project for this longer period

3.0

Mean component score

2.8

2.6 Mean primary score Mean secondary score 2.4

2.2

2.0 Beginning of project

After 1 year

After 2 years

After 3 years

Point in time

Figure 4: Changes in mean Component Map scores over three years Improvement in student learning and attitude Table 1 indicates that a large majority of teachers believed that the quality of student learning had improved. A considerable amount of research focused directly on student learning and attitudinal outcomes. Student scores on standardised tests and on surveys of nine attitude constructs were linked to the component map scores by comparing results for high-SIS and low-SIS teachers. The student achievement linkage proved to be complex and provided different patterns in each year for primary and secondary schools, but overall there was a significant link between student outcomes and teacher Component Map scores. Attitude surveys of upper primary and secondary students showed a clear and significant link between each of nine attitudinal constructs (e.g. enjoyment, motivation) and teacher Component Map scores.

R. TYTLER

99

Changes in science team practices In teacher responses to an open question concerning perceived changes in science in the school, at the end of the second year of the project, the changes claimed were substantial, almost completely positive, and indicated interesting differences between primary and secondary school teachers. Secondary teachers focused on the science team as the ‘engine’ of change; learning how to work collaboratively and purposefully was a major benefit for secondary schools. Improved curriculum and classroom strategies were also a major perceived outcome. Primary teachers focused on the greater amount of science, the increased profile for science, better resources and resource management, and better organization of the science area in general. In the November 2002 questionnaires, SIS Coordinators and teachers were asked to indicate at which level the science team in the school was operating for each of the operational aspects listed (Table 2). They were asked to select from one of the following: 1 – A low level, 2 – A fairly low level, 3 – A moderate level, 4 – A high level, or 5 – A very high level. They were also asked to make separate judgments about this for both the current and the pre-SIS situation. Table 2: Percentage of Phase 1&2 Primary and Secondary school SIS Coordinators judging the science team to be operating at high or very high level Prim. Prim. Sec. Sec.. N=48 N=48 N=41 N=41 The science team in our school: PreCurrent PreCurrent project project a. Regularly discusses science 2 64 15 78 teaching and learning issues b. Has a shared vision of the purpose 4 89 9 68 and direction of science in the school. c. Has a shared view of effective 7 81 9 65 classroom teaching and learning in science d. Is focused on improving student 10 87 26 74 learning outcomes in science e. Is committed to ensuring that 20 94 28 90 students find\science interesting and relevant f. Has an agreed process for 2 44 22 46 assessment of student learning in science g. Plans together effectively 14 85 12 68 h. Has a coherent staff professional 18 73 13 59 development program focused on teaching and learning

100

SCHOOL INNOVATION IN SCIENCE

i. Support each other in teaching and learning strategies j. Promotes science effectively within the school community

Prim. N=48 27 5

Prim. N=48 83 79

Sec. N=41 22 8

Sec.. N=41 85 65

Table 2 shows the percentage of Coordinators who judged the science team in their school to be operating at a high or a very high level prior to the Project and currently (data were used from schools involved for two or three years). It was found that SIS Coordinators believed that science in their schools had undergone quite dramatic changes as a result of their planning and working together in SIS. These results are indicative of a strong cultural shift in the way science teams worked together, and the way teachers related to each other professionally, and they vindicate the emphasis placed within the SIS Strategy on the science team as the engine of change. Questionnaire and interview responses indicated that principals strongly held the view that teachers of science were operating together more effectively. A number of secondary teachers commented that the science faculty was seen, as a result of the project, to be leading innovation and change within the school External judgments of schools’ improvement At the end of schools years 2000 and 2001, consultants met to rank participating schools in terms of their ‘SIS-ness’, i.e. to determine the extent to which schools had embraced the SIS Strategy and Components. Descriptors were generated to describe different degrees of SIS-ness, and the main factors that determined this were identified. A rating scale was generated, as set out below.

Score 4

Score 3

Score 2 Score 1

The SIS approach is embedded across the school and there is clear and strategic commitment to the change process, including monitoring, a sense of building a firm basis for ongoing change, and a clear focus on teaching and learning. There is substantial commitment to the SIS approach, but not as much progress was made, or there were not so many initiatives, or not all teachers were involved. There was superficial change which was not focused on the SIS Components. There was very little commitment to change.

The results of this exercise are shown in Table 3 (note that some consultants used half scores).

R. TYTLER

101

Table 3: SIS-ness profiles of schools Rating SIS-ness: % of schools (N=115) 3.5 / 4.0 23 2.5 / 3.0

45

1.5 / 2.0

26

< 1.5

6

The data indicate that the majority of schools have undergone substantial change. About one quarter were judged to have achieved embedded change. There is a substantial minority of schools, however, which have not achieved significant progress. An important part of the analysis has been the identification of factors that act as barriers to improvement. Data relevant to this are reported below.

6. THE PROCESS OF CHANGE An important part of the research has concerned the nature of the change process itself, understanding the factors that make a difference to schools, including the aspects of the Strategy, and gaining a deeper understanding of school structures that are undergoing change. Factors affecting the school change process The research team has worked to identify factors determining the outcome of the project in schools, to incorporate these factors in advising schools, and to refine the support structures that are put in place. The identification process involved an analysis by a review meeting of the research team with consultants active in advising and monitoring schools across the state. In the analysis, a set of factors were identified that are particularly critical in determining a school’s success in improving science teaching and learning (Table 4). The nature of the support structures and issues of science team culture were discussed above. The emphasis on the strategic nature of change and the importance of local ownership and control immediately places the locus of control on local leadership. The role of coordinators as leaders became a major focus within the project and is one of the major factors influencing the success of the project in schools. In the project's developmental phase, evidence concerning the management of the change process was collected through questionnaires, field notes, and interviews with selected SIS Coordinators and principals.

102

Leadership

School culture:

Access to support and resources:

SCHOOL INNOVATION IN SCIENCE Table 4: Critical success factors for SIS Coordinator: status within school, degree of organisation, leadership qualities. Principal: leadership commitment; and actions related to support and commitment A culture of change existing in the school A positive attitude and willingness to try things The ability to share ideas and be open with each other concerning their classroom practice External support and prompting from consultants, Networks: other schools to share ideas, available PD, Access to physical resources Time, CRT* support, direction and project materials/advice

*Casual Relief Teacher In a focus group discussion of the project team (most of whom acted as consultants), field notes and observations concerning the role of SIS Coordinators were used as the basis for refining our understanding of how SIS Coordinators worked to support the change process successfully. Descriptions of effective Coordinator actions were generated, refined, and grouped into categories representing important dimensions of strategic action to support change. The dimensions include team building and encouraging a common agenda, supporting groups of staff working on initiatives, supporting individual teachers, encouraging innovation and involvement, allowing individuals to focus and work differently, and working with unenthusiastic teachers. During the second year of the project this analysis coalesced into a ‘Leading Change’ program which was developed for SIS Coordinators; this program has attracted widespread approval. Levels at which change occurs Within the guidelines provided by the SIS Components and Strategy, schools have differed in their areas of focus and in the way they have gone about supporting the change process. During 2002 the SIS team worked closely with a small number of schools and teachers to investigate in more depth, the impact SIS had made on teaching and learning, and the type of support that would best lead to improvement. It became clear that the change process in SIS schools needed to be seen as occurring at different levels within the school, with each level requiring particular modes of support. Experience with these schools showed clearly that, although there may have been substantial change in the school's policy and profile, or at the science team level in broad planning, curriculum direction, and discussion of teaching and learning issues, there was considerable variation in the extent to which this had impacted on unit planning processes and on teachers’ practices. The real engine of change in teaching and learning seemed to be the unit planning group. It is in close

R. TYTLER

103

planning that teachers really needed to talk about their beliefs and knowledge, and examine their understanding of student learning. Thus, in the further promotion and refinement of the SIS Strategy, it became clear that if change is to be embedded, attention must be paid to the ways in which quality conversations at school and science team levels are translated into unit planning and classroom practice. We have identified different levels at which change can occur, to develop a view of the way planning at these levels can interact effectively: – the place of science in the school (profile and status within the school community, support of leadership team); – whole science team processes (shared discourse and vision, common understandings, effective communication and planning, proportion of staff actively involved, enthusiasm, commitment to student outcomes); – unit planning at the year level (effective planning processes, degree of collaboration and sharing of ideas, agreement on purposes and teaching and learning principles); – individual teacher practice at the individual strategy level (change in practice, variety of strategies, understandings and commitments, confidence and attitudes). Monitoring changes in the school at these different levels provides needed information that is important for developing more detailed advice about effective management of the change process.

7. CONCLUSION AND IMPLICATIONS – UNDERSTANDING THE CHANGE PROCESS The first three research questions considered in this paper have been answered in the preceding sections. The SIS Strategy has led to a wide range of initiatives and has been successful in promoting change at a number of levels. Change has been evident in the focus on significant teaching and learning initiatives, and in the coherence of the accounts regarding the nature and extent of the changes at both classroom and school levels. The many instruments, including student tests, mapping of teacher practice, and questionnaires, have provided more focused evidence of substantial change. These and more detailed analyses have explored the nature of this change. The findings support two major propositions concerning the change process. Change should be seen as inherently cultural in nature. This is evident from the very different experiences of primary and secondary schools, at the level of teacher knowledge and commitments, core presumptions about purposes, and curriculum arrangements. It is also evident in the very different responses and outcomes in different schools, depending on interactions between the different levels of school operations and local circumstances. A change project must ground itself in the reality that schools have very different histories, populations, and

104

SCHOOL INNOVATION IN SCIENCE

needs. At the moment there is a debate occurring in Australia that concerns the effectiveness of resource-led reform. Our view and our evidence suggests that curriculum resources should be flexible and responsive to the different cultural traditions and circumstances within schools. Change within schools occurs at multiple levels. Support structures and processes must target these different levels, and attention must be given to the ways in which the levels interact. The SIS Strategy has been progressively developed to reflect this; monitoring of change must also reflect it. SIS schools have been successful to different degrees in preparing for and/or achieving fundamental changes. Identification of factors that impinge on the success of the change process has been an important element in determining what advice to offer schools and the government, regarding the most effective ways of ensuring a supportive environment for improvement within schools. Analysis of the different data sets has led to a deeper understanding of the different levels at which science is planned and supported in schools. The SIS Strategy, in allowing local control of the details of the change process, has provided the flexibility that schools need to manage change at these different levels.

ACKNOWLEDGMENTS The research described in this paper was undertaken as part of the Science in Schools Research Project, funded by the Victorian Department of Education and Training. The development and refining of the processes described in this paper has been shared by members of the Deakin University-based SIS research project team: Annette Gough, Brian Sharpley, Michele Griffiths, Sophie Nakos, Robin Matthews, Geoff Beeson, Bruce Waldrip, Jeff Northfield, Pat Armstrong, Gillian Milne (project manager). ENDNOTE A ‘case story’ is used in this case to describe a narrative account of the major features of a school undergoing change, drawn from a number of data sources. It is not have the methodological complexity of a ‘case study’ and it does not claim to be more than a partial view, but it does attempt a more complex narrative than a ‘vignette’ which is usually taken to refer to one event.

REFERENCES Beeth, M., Duit, R., Prenzel, M., Ostermeier, C., Tytler, R. & Wickman, P-O. (2003). Quality Development Projects in Science Education. In D. Psillos, P. Kariotoglou, V. Tselfes, G.

R. TYTLER

105

Fassoulopoulos, E. Hatzikraniotis & M. Kallery (Eds.), Science Education research in the knowledge based society. Dordrecht, The Netherlands: Kluwer Academic Publishers. Hall, G.E. & Hord, S.M. (2001). Implementing Change: Patterns, Principles, and Potholes. Boston: Allyn & Bacon. Hargreaves, A. (1994). Changing teachers, changing times: Teachers’ work and culture in the postmodern age. London: Cassell. Hoban, G. (1992). Teaching and report writing in primary science: Case studies of an intervention program. Research in Science Education, 22, 194-203. Joyce, B. & Showers, B. (1995). Student achievement through staff development: Fundamentals of school renewal (2nd ed.). New York: Longman. Loughran, J. & Ingvarson, L. (1993). Science teachers’ views of professional development. Research in Science Education, 23, 174-182. Paige, K. (1994). Factors perceived to have enabled 25 women to develop expertise to teach primary science. Research in Science Education, 24, 246-252. Tytler, R. & Nakos, S. (2003). School Innovation in Science: Transformative initiatives in Victorian secondary schools. Australian Science Teachers’ Journal, 49(4), 18-27. Tytler, R. (2003). A window for a purpose: Developing a framework for describing effective science teaching and learning. Research in Science Education, 30(3), 273-298. Tytler, R., Waldrip, B. & Griffiths, M. (2004). Windows into practice: Constructing effective science teaching and learning in a school change initiative. International Journal of Science Education, 26(2), 171-194. Webb, C. (1993). Teacher perceptions of professional development needs and the implementation of the K-6 Science and Technology syllabus. Research in Science Education, 23, 327-336.

WAYS OF USING ‘EVERYDAY LIFE’ IN THE SCIENCE CLASSROOM

MARIA ANDRÉE Stockholm Institute of Education, the Swedish National Graduate School in Science and Technology Education Research, Sweden

ABSTRACT Connecting science to students’ everyday life experiences is an important theme in science education discourse. The aim of this article is to explore in what ways ‘everyday life’ is used in the science classroom and what problems are solved through the use of ‘everyday life’. The research approach is ethnographic. Data was gathered through participant observation during one semester, in two Swedish science classes. Results show that ‘everyday life’ is brought into the classroom and made into school tasks within different types of activities; enculturation into science, education of scientifically literate citizens and making science interesting. The results underscore the importance of understanding the use of ‘everyday life’ in science classrooms as embedded in science classroom practice.

1. INTRODUCTION A washing machine can illustrate dispersion by centrifugal force; clothes show how to distinguish natural and synthetic fibres; plastics aid the study of oil derivatives; lemon juice and red cabbage bring acids to life, while television helps explain how electromagnetic waves work. (López 2000, p. 13)

The quote above is an illustrative example of a common view that things from everyday life could and should be used to illustrate different scientific principles in science education. Linking science to everyday life has been an important theme in science education discourse among both researchers and practicing teachers. The most frequent argument today is that it is a way to make science relevant (Campbell & Lubben, 2000). The use of everyday life is argued to be an important pedagogical tool for motivating students. In some discussions on scientific literacy, it is also proclaimed to be a way of educating scientifically literate citizens (e.g. Campbell & Lubben, 2000; Giachardi, 1994; Harlen, 2002). But what does it mean to use everyday life in the science classroom? In what ways is everyday life used? What everyday life are we talking about? In a constructivist perspective analyzing everyday life problems has been seen as a matter of understanding concepts well. Andersson (2001) writes that the meanings of scientific concepts are deepened if they are applied to everyday phenomena and that solving everyday life problems scientifically is a way to challenge students’ 107 K. Boersma et al. (eds.), Research and the Quality of Science Education, 107—116. © 2005 Springer. Printed in the Netherlands.

108

WAYS OF USING ‘EVERYDAY LIFE’ IN THE SCIENCE CLASSROOM

everyday conceptions. Aikenhead and Cobern (Aikenhead, 1996; Cobern & Aikenhead, 1998) have a somewhat different approach to everyday life and science. They describe science and everyday life as different cultural milieus with different languages, values, and norms. They write about students’ experiences in science education in terms of border-crossing experiences. Students are described as crossing cultural borders of everyday subcultures, such as home culture and/or youth culture to the cultures of school, science education, and science. The cultural bordercrossing is argued to be an important aspect of learning science. Szybek (1999) makes a similar phenomenological description. He writes about the interaction between different stages of events in biology education and argues that the interaction between a stage of everyday non-scientific life and a stage of science results in a stage of school science. Szybek (2002) writes that the aim of science education is to translate everyday problems to scientific problems so they can be solved using scientific techniques and ways of reasoning. Learning is an aspect of collective activities that cannot be separated from our participation in daily practices (Lave, 1993; Roth, 1998). Science is dealt with in science education, and expectations on and views of science are part of the cultures constituted in the science classroom. Through participation in daily activities in the science classroom, students are enculturated into particular ways of acting (Driver et al., 1994; Wolcott, 1994): ways of knowing (Crawford, Kelly & Brown, 2000), ways of talking science (Jewitt & Scott, 2002; Lemke, 2001; Ogborn et al., 1996), experimenting (Beach, 1999; Bergquist & Säljö, 1994; Delamont, Benyon & Atkinson, 1988), writing science (Knain, 2003), and so on. The act of referring to something talked about as ‘everyday life’ is another such way of acting that students learn through participation in science classroom practice. Linking science to something called ‘everyday life’ is part of the daily practice of science classrooms. In science education discourse, ‘everyday life’ does not refer to the activities in which we daily engage in school. ‘Everyday life’ in the science classroom is, rather, something outside of the classroom. Science is to be related to someone’s ‘everyday life’ in the ‘real world’, i.e. a world that is not school. Even if it is not presupposed whose ‘everyday life’ it is, it is presupposed that it is not the everyday life of the science classroom. In science education discourse ‘everyday life’ is brought into the science classroom, by someone from somewhere else; it could be by a teacher, a student, or a piece of text. Wanting to bring something into the classroom, which is not necessarily there, is a common objective for those advocating the use of ‘everyday life’ in science education. ‘Everyday life’ is, as well as ‘science’, brought into the science classroom and dealt with in certain ways in certain activities for certain purposes. The aim of this study is to explore ways of using ‘everyday life’ in the science classroom. The research questions addressed are: In what different ways is ‘everyday life’ used in the science classroom, when brought into the classroom by a teacher, a student, or a piece of text? What problems are solved using ‘everyday life’, and what are the objectives for solving those problems? What different norms and values are constituted through the use of ‘everyday life’? And, finally, how can the concepts of border-crossing and translation help us make the ways of using ‘everyday life’ comprehensible?

M. ANDRÉE

109

2. DESIGN OF THE STUDY The research approach is ethnographic. The strength and underlying purpose of an ethnographic study is that it enables a researcher to say something about what particular people do in particular circumstances (Wolcott 1999). Here, it enables me to study how ‘everyday life’ is used as a part of the daily activities in a particular science classroom. Two science classes, grade six and seven, were studied in a Swedish midsized compulsory school during one semester (fall 2002). Both classes were taught by one teacher; here called “Ann”. Ann is a mathematics and science teacher with four years of teaching experience. During fieldwork a variety of data was collected by me through participant observation. Data include field notes, audiotape recordings, teaching materials, and some student work. The analyses can be described in terms of what Glaser and Strauss (1967) named the “constant comparative method”. This involves a constant comparison of incidents in empirical data that eventually generates the theoretical properties of a category. First, incidents of using ‘everyday life’ were compared by searching for differences and similarities. These are incidents where ‘everyday life’, as something outside of the classroom, was brought into the science classroom. The result of this analysis is described very briefly as different ways of linking science to ‘everyday life’ (e.g. giving examples from ‘everyday life’). Second, these different ways of using ‘everyday life’ were analyzed in terms of the problems which are solved and the objectives used for solving the problems (e.g. examples from ‘everyday life’ are used to illustrate scientific principles). Then, the problems and objectives were compared, and three different activities emerged. 3. USING ‘EVERYDAY LIFE’ IN THE SCIENCE CLASSROOM The different ways of linking science to ‘everyday life’ resulting from the first analysis are: giving examples from ‘everyday life, making computations on ‘everyday’ phenomena, analyzing and categorizing ‘everyday life’ objects and phenomena, using ‘everyday life’ artifacts in laboratory work, and asking questions on how to cope with ‘everyday’ problems at home and in society at large. When giving examples, making computations on, or analyzing and categorizing ‘everyday life’ objects and phenomena, students are involved in solving scientific problems and allowed to develop scientific ways of experiencing and relating to the world. Students compute and analyze situations in ’everyday life’, e.g. by asking, “how much is my muscle power enlarged by a car jack when I’m elevating a car to change tires” or by analyzing the process of making fruit syrup. Students describe and categorize ‘everyday life’ phenomena, e.g. by describing different substances such as tea and müsli as mixtures or solutions. Giving examples from ‘everyday life’ that fit a particular scientific category is similar to categorization, e.g. questions on examples of mixtures or on how pressure is used in ‘everyday life’. These actions are categorized as part of an activity labeled enculturation into science. When working with questions on household problems, large-scale societal challenges (e.g. environmental issues), and the role of science in society, certain

110

WAYS OF USING ‘EVERYDAY LIFE’ IN THE SCIENCE CLASSROOM

ways of acting in ‘everyday life’ are formulated. These actions are categorized as part of an activity labeled education of scientifically literate citizens. When using ‘everyday life’ artifacts in laboratory work, such as Pepsi, caramel coloring, waste baskets, and plastics cups, the science class is made fun and exciting to students. These actions are categorized as part of an activity labeled making science interesting. The activities described have different norms and values that make a particular way of using ‘everyday life’ rational; they can be constituted in parallel during the same lesson or even the same exercise. I elaborate some characteristics of the three activities below. Enculturation into science Most incidents, where ‘everyday life’ is used in the science classroom studied, are part of the enculturation into science. The foremost characteristic of this activity is that embodiment or visualization of scientific concepts or relations is superior in solving ‘everyday life’ problems. The following example is a question posed on a written examination in grade six: You got a stain of dirt on your shirt. What should you try first to remove it? If your first attempt to clean your shirt failed, what could you try then?

One student answered that you take wash powder and wash the shirt. When doing laundry, this is what most people would do. Few would try plain water first and then an oil-dissolving solvent such as acetone second. But water and acetone is the answer that the teacher expects. The question is supposed to be an embodiment of water solubility. Students have to transform the task of washing a shirt to a question of water solubility in order to get credit for their answers. When discussing ‘everyday’ problems with me, the teacher says that ‘everyday life’ questions are good to test which students have developed a more complex understanding. For her, they function as a way to sort students who excel in science from those who just pass. These questions on ‘everyday life’ are not about formulating ways of acting in everyday life, but rather about formulating what scientific aspects are relevant in a particular ‘everyday life’ problem. The ‘everyday life’ used may be purely hypothetical and even surreal. There are some computational tasks that one might classify as surreal. The following is a worksheet exercise asking for the pressure on a diver: A free diver goes down to the depth of 50 m in water. How deep would he have to dive in acetone to be affected by the same pressure?

Students are to calculate and compare pressure in water and acetone. The exercise is very appropriate in a theoretical context. It highlights the relation between pressure and density, and it gives students a possibility to envision two different situations that they can use for computation. If the question is taken for real in an ‘everyday’ context, it appears absurd. Acetone is a toxic and inflammable solvent. That the diver is not only a diver but a free diver makes the question even more absurd in an ‘everyday’ perspective. The purpose of the exercise is to illustrate a scientific relation, and for that, the surreal might be more useful than the real ‘everyday life’.

M. ANDRÉE

111

In this activity, usefulness of ‘everyday life’ examples can be measured in terms of how well they embody a particular concept. Education of scientifically literate citizens Three different types of education of scientifically literate citizens are discerned in the empirical analyses of classroom data1: First, there are examples of learning to cope with ‘everyday life’ that involve questions and recommendations concerning how to deal with certain ‘everyday life’ issues, e.g. Therefore never let a spray can lie in the sun and never use it near open fire! (Original emphasis, Physics textbook grade 7)

There are other, similar recommendations concerning how to act when dealing with potentially dangerous things, such as how to handle solvents and how to save a person from a hole in the ice. These recommendations are later reformulated as questions on worksheets and written exams. Second, there are examples of teaching about political processes and discussions concerning environmental issues. Examples are: a teacher-led discussion on the meaning of Agenda 21 in an introductory biology course, worksheet questions on pros and cons concerning pesticides, and worksheet questions concerning human impact on different ecosystems. Third, there are examples of attempts to develop students’ faith in science and technology by formulating the necessity of the scientific and technological enterprise for maintaining high standards of living. This is expressed in a textbook where students are encouraged to recognize their dependence on science and sciencedriven technologies: When you hear the word ‘chemistry’, you might think of boiling liquids and smoke, puffing out of flasks, and test tubes in a laboratory. In actual fact, chemistry has been around us in our everyday life since time immemorial. The ability to use different materials has played a great part in our development. […] The chemists’ discoveries give us, among other things, better medicines, new materials, and better yields in agriculture. (Chemistry textbook, grade 6)

On the same page there is a picture depicting a mother with her two children in a supermarket, and students are asked to give examples of the substances in the pictures that are chemically produced. The text on this page expresses that chemistry is something students should appreciate and value for its importance to ‘everyday life’. The text is also critical of ignorant use of dangerous substances in the past, and it claims that chemists have better knowledge today, even though they still need to know more. Making science interesting In the activity of making science interesting, ‘everyday life’ is used a pedagogical tool for motivating students and making science classes more fun and interesting. In 1

In historical and contemporary science education discourse, there is, however, a much greater diversity in the meanings ascribed to scientific literacy. For an overview, see DeBoer (2000).

112

WAYS OF USING ‘EVERYDAY LIFE’ IN THE SCIENCE CLASSROOM

some laboratory lessons ‘everyday’ artifacts are used. A common feature of these lessons is excitement and play. In one laboratory session students are to do something like a magic trick: fill a glass with water, put a paper on top of the glass, and turn it around. Students are not surprised when the paper remains on the brim of the glass and keeps the water inside the glass, even if they don’t know how to explain it. But students are very excited and play around with the artifacts. When learning about filtration and distillation, Pepsi and caramel coloring are used to illustrate laboratory processes. Students filter Pepsi and distill caramelcolored water. They are excited working with Pepsi and want to taste it again and again. When working with caramel coloring, students compare colors; some proclaim proudly that they have the most beautiful color. Through laboratory work, ‘everyday’ artifacts are made into chemicals. Using artifacts is not just a way to make science class interesting, but also a way of making the world chemical (c.f. the introductory quote from López 2000). 4. CONTEXTUAL DIFFICULTIES In the activities of enculturation into science and making science interesting, ‘everyday life’ does not necessarily function as contextual support for students who feel uncertain about how to solve a problem in the science classroom. The following is an excerpt from a lesson on pressure: Teacher: Yes, if we’re going to use this brick to press flowers now, then in what way could it be suitable for us to put it? Robert: Like that. Teacher: Like that? There are three alternatives, Robert. Robert: The one in the middle, the one in the middle. [A brick lying on its side.] Teacher: If you are to press one flower Robert, in what way should you put it. Robert: The one in the middle. Teacher: Why that? Robert: Because it’s the flattest one. Students laugh. Simon: The one at the bottom is the flattest one. Robert: Yeah the one at the bottom. Teacher: Would you want a high or a low pressure if you’re pressing something? Robert: I’d want… high pressure. Teacher: Yes high pressure. Robert: Then it has to be the top one. [A brick standing on its end.] Teacher: The top one, all right. (Tape recorded transcription, grade 7)

After some discussion Robert agrees with the teacher that the brick should stand up on its end for the pressure to be as high as possible. His first suggestion to use “the flattest one” is reasonable if you truly would want to press flowers, since a brick on its end could easily fall and also because most flowers are larger than the end of a brick. His first answer is an example of practical reasoning in relation to a practical problem. For the teacher though, the pressing of flowers is a hypothetical problem. The intended task is to use the concept of pressure to analyze the problem of pressing flowers. The example functions as a way to embody the instruction on

M. ANDRÉE

113

pressure in order to make the concept of pressure less abstract and more comprehensible. The discussion illustrates the risk of misinterpreting an ‘everyday’ context as more valid than the theoretical issue at stake. When ‘everyday life’ problems are dealt with, answers that are valid and relevant in ‘everyday life’ contexts may not be valid in the science classroom. Another question posed on an examination is about making fruit syrup: You are going to make orange syrup. You press the oranges, pour water on the fruit juice, and add sugar. Now you have a “mishmash” containing orange juice, water, and sugar on the bottom. What should you do to make it fruit syrup that you can pour into a bottle? Explain how you do it and why. (Original emphasis.)

None of the twenty-two students in grade six could answer this question correctly, according to the teacher. Most students answered that they would boil the “mishmash”, but none said that they would filter it after boiling. One student wrote that “you add preservatives”, which may signal that he had actually taken part in the making of fruit syrup. However, when the question is put in science, as opposed to home economics, students are to see and mention only what is relevant from a specific science perspective. In a science perspective, making fruit syrup is first about boiling the ingredients for the sugar to dissolve, and second, to separate pulp from syrup through filtration. Adding preservatives or not is irrelevant. Answering these questions requires an awareness of how ‘everyday life’ examples are to be handled in the science classroom. Students must ask what scientific concept a particular ‘everyday life’ question is supposed to be an embodiment of, in order to translate it correctly. When students worked with Pepsi in the science classroom, confusion arose concerning how Pepsi should be represented in the laboratory report: Is Pepsi laboratory material, or what is it? Students know that Pepsi in the practice of science education is not what it is to them in their everyday lives. Pepsi is dealt with differently in the science classroom. It is not a waste to make it undrinkable but rather in line with the purpose of the lesson. 5. DISCUSSION Three different activities emerged in the analyses of ways of using ‘everyday life’ in the science classroom: enculturation into science, education of scientifically literate citizens, and making science interesting. In all three, ‘everyday life’ is brought into the classroom and transformed into school tasks. In enculturation into science and making science interesting, ‘everyday life’ does not necessarily function as contextual support. When ‘everyday life’ problems are dealt with, answers that are valid and relevant in ‘everyday life’ contexts may not be valid in the science classroom. Szybek (2002) and Cobern and Aikenhead (1998) describe the science classroom as a space for translation and a subculture for border-crossings between science and everyday life. However, they limit the role of the science classroom to this. In order to make sense of the results, we need to recognize school science practice as a

114

WAYS OF USING ‘EVERYDAY LIFE’ IN THE SCIENCE CLASSROOM

practice of its own which is not reducible to other practices (cf. Carlgren, 1999), even if it is largely influenced and related to other practices: science-related as well as non-science-related, in and out of school. The classroom examples studied, like pressing flowers and diving in acetone, are examples where ‘everyday life’ problems are not just translated to scientifically solvable problems. The issue of pressing flowers is rather transmuted in science classroom practice to a school science task where students are to identify what scientific aspects of pressing flowers are relevant in this particular course at this particular moment. If you would want to press flowers using a brick and reflect upon that scientifically, you would need to consider both stability and pressure. In science class, however, the problem is discussed only as an illustration of the relation between pressure and area. In the science classroom a particular school science practice emerges. Gutiérrez, Baquedano-Lopez, and Tejeda (1999) suggest that what is characteristic for education is not so much border-crossing between cultures, but rather hybridization. They describe learning practices as immanently hybrid, meaning polycontextual, multivoiced, and multiscripted. This means that conflicts and tensions are unavoidable and intrinsic to all cultural practices. People participate in different open-ended subcultures to which they bring experiences and values created through participation in other cultural settings. In this sense all cultures are hybrid. Hybrid cultures are also constituted in the sense that classroom activities very much take place within cultures of schooling, at the same time as they are attributed in science education to cultures of scientists (cf. Brown, Collins & Duguid 1989). Many activities of the science classroom are not activities of practicing scientists, and many of them do not necessarily make sense in out-of-school science practices. Roth (2003) illustrates this in a study on graphing ‘in captivity’ (i.e. educational settings) and graphing ‘in the wild’ (i.e. out-of-school science practices when solving authentic problems). His study highlights the situatedness of graphing in both scientific and educational practices. Similarly, activities students undertake in the classroom attributed to ‘everyday life’ may not make sense in ‘everyday life’. Lave’s (1988) studies on mathematical reasoning in ‘everyday life’ show that math activities take form differently in different situations. Mathematics while shopping or dieting seemed to be structured in relation to dilemmas that motivated those activities rather than in relation to mathematical algorithms. In different school subjects, different school subject cultures are constituted (c.f. Ensign, 1997). The school subject cultures have implications for how tasks are formulated and dealt with in the classroom. Säljö and Wyndhamn (1993) show that the frequency of correct solutions to a task of determining the postage for a letter is different in mathematics and social studies. In social studies most students solve the task by reading the table for postage rates. In mathematics there was a tendency for students to solve the task as a computational task; students had learned to look past the context of the task and just work out the numbers. Interesting, though, is that the ‘everyday’ way of solving the task is correct also in the mathematics classroom. In the science classroom studied, ‘everyday’ answers are not always correct (e.g. pressing flowers, washing a shirt), and one might expect that answers valid to the question of making fruit syrup will be somewhat different on a home economics exam than on the science exam.

M. ANDRÉE

115

The results of this study underscore the situated character of how ‘everyday life’ is used in the science classroom. When ‘everyday life’ problems are brought into the science classroom, their context is inevitably transmuted; they become classroom tasks and part of school culture. It is not sufficient to describe the use of ‘everyday life’ in terms of border-crossing experiences or translations between science and ‘everyday life’, because science classroom practice is structured also in relation to the activity system of schooling. The use of ‘everyday life’ is embedded in science classroom practice and shaped by particular goals, desires, demands, and traditions that are constituted within that practice. REFERENCES Aikenhead, G. S. (1996). Science Education: Border Crossings into the Subculture of Science. Studies in Science Education, 27, 1-52. Andersson, B. (2001). Elevers tänkande och skolans naturvetenskap. Forskningsresultat som ger nya idéer. Skolverket. Liber: Stockholm. Beach, D. (1999). Alienation and Fetish in Science Education. Scandinavian Journal of Educational Research, 43 (2), 157-172. Bergqvist, K. & Säljö, R. (1994). Conceptually blindfolded in the optics lab. Dilemmas of inductive learning. European Journal of Psychology of Education, 9 (2), 149-158. Brown, J. S., Collins, A. & Duguid, P. (1989). Situated Cognition and the Culture of Learning. Educational Researcher 18 (1), 32 – 42. Campbell, B. & Lubben, F. (2000). Learning science through contexts: helping pupils make sense of everyday situations. International Journal of Science Education, 22 (3), 239252. Carlgren, I. (1999). Pedagogiska verksamheter som miljöer för lärande. In I. Carlgren (Ed.), Miljöer för lärande. (pp. 9-28). Studentlitteratur: Lund. Cobern, W. & Aikenhead, G. S. (1998). Cultural Aspects of Learning Science. In Fraser, B.J. & Tobin, K.G. (Eds.), The International handbook of Science Education. Kluwer Academic Publishers: Dordrecht. Crawford, T., Kelly, G. & Brown, C. (2000). Ways of Knowing beyond Facts and Laws of Science: An Ethnographic Investigation of Student Engagement in Scientific Process. Journal of Research in Science Teaching, 37 (3), 237-258. De Boer, G. (2000). Scientific Literacy: Another look at Its Historical and Contemporary Meanings and Its Relationship to Science Education Reform. Journal of Research in Science Teaching, 37(6), 582-601. Delamont, S., Beynon, J. & Atkinson, P. (1988). In the Beginning was the Bunsen: the foundations of secondary school science. Qualitative Studies in Education, 1(4), 315328. Driver, R., Asoko, H., Leach, J., Mortimer, E. & Scott, P. (1994). Constructing scientific knowledge in the classroom. Educational Researcher, 23 (7), 5-12. Ensign, J. (1997). Ritualizing Sacredness in Math: Profaneness in Language Arts and Social Studies. Urban Review, 29 (4), 253-61. Giachardi, D. (1994). Relevance and the accessibility: the role of science education. School Science Review, 75 (273), 7-14. Glaser, B.G. & Strauss, A. L. (1967). The Discovery of Grounded Theory. Strategies for Qualitative Research. Aldine de Guyter: New York.

116

WAYS OF USING ‘EVERYDAY LIFE’ IN THE SCIENCE CLASSROOM

Gutiérrez, K. D., Baquedano-Lopez, P. & Tejeda, C. (1999). Rethinking Diversity: Hybridity and Hybrid Language Practices in the Third Space. Mind, Culture, and Activity, 6(4), 286-303. Harlen, W. (2002). Links to everyday life: the roots of scientific literacy. Primary Science Review 71, 8-10. Jewitt, C. & Scott, P. (2002). Meaning making in science classrooms: a joint perspective drawing on multimodal and sociocultural theoretical approaches. Paper presented at ISCRAT, Language, action and communication in science education symposium. Knain, E. (2003). Identity & genre literacy in high-school students’ experimental reports. Paper presented at the 4th ESERA Conference, Noordwijkerhout, the Netherlands. Lave, J. (1988). Cognition in practice: mind, mathematics and culture in everyday life. Cambridge University Press: Cambridge. Lave, J. (1993). The practice of learning. In S. Chaiklin & J. Lave (Eds.), Understanding practice. Perspectives on activity and context. (pp. 3-32). Cambridge University Press: Cambridge. Lemke, J. L. (2001). Articulating Communities: Sociocultural Perspectives on Science Education. Journal of Research in Science Teaching, 38 (3), 296-316. López, A. (2000). Science Teaching’s quantum leap. The UNESCO Courier, May 2000, 1315. Ogborn, J., Kress, G., Martin, I. & McGillicuddy, K. (1996). Explaining Science in the Classroom. Open University Press: Buckingham/Philadelphia. Roth, W.-M. (2003). Toward an Anthropology of Graphing. Semiotic and Activity-Theoretic Perspectives. Kluwer Academic Publishers: Dordrecht/Boston/London. Roth, W.-M. (1998). Designing communities. Kluwer Academic Publishers: Dordrecht/Boston/London. Szybek, P. (2002). Science Education – An Event Staged on Two Stages Simultaneously. Science & Education 11, 525-555. Szybek, P. (1999). Staging Science. Some Aspects of the Production and Distribution of Science Knowledge. Doctoral Dissertation Lund University. Department of Education: Lund. Säljö, R. & Wyndhamn, J. (1993). Solving everyday problems in the formal setting. An empirical study of the school as context for thought. In S. Chaiklin & J. Lave (Eds.), Understanding practice. Perspectives on activity and context (pp. 327-342). Cambridge University Press: Cambridge. Wolcott, H. F. (1999). Ethnography: a way of seeing. Sage Publications: Walnut Creek/London/New Delhi. Wolcott, H. F. (1994). Transforming Qualitative Data. Description, Analysis, and Interpretation. Sage Publications: Thousand Oaks/London/New Delhi.

PART 3 Science teacher education

SCIENCE TEACHER EDUCATION: ISSUES AND PROPOSALS DIMITRIS PSILLOS¹ , ANNA SPYRTOU², PETROS KARIOTOGLOU² ¹Aristotle University of Thessaloniki, Greece ² University of Western Macedonia, Greece ABSTRACT Research in science teacher thinking and constructivist pedagogy calls for an expanded knowledge base of teaching, and raising the issue of teaching and understanding of such knowledge by students during teacher education. In the present paper we discuss certain recent studies concerning teachers’ knowledge base; besides we present and discuss a framework for developing and investigating courses in science teacher education; finally, in the third part, we present aspects of a case study illustrating the suggested framework.

1. INTRODUCTION Influenced by the conception of teaching as a thinking profession, teacher education researchers have displayed great interest in the basis of teachers’ knowledge and cognition (Clark & Peterson, 1986, Gess-Newsome & Lederman, 1999). Moreover, in the field of science education, research into students’ conceptions of natural phenomena influenced researchers interests in science teachers’ conceptions about scientific concepts and phenomena, as well as about teaching and learning science (Cochran & Jones, 1998, Hewson, Kerby & Cook, 1995). Researchers investigating the character of teachers’ knowledge have advocated a broad conception of the expert teacher knowledge base, suggesting that such knowledge is grounded in acts of pedagogical reasoning (Van Driel, Beijaard & Verloop, 2001). From the perspective of pedagogy, constructivist approaches, as the practices of teaching for student learning with understanding, commonly call for a greatly expanded knowledge base for teaching. How an extensive knowledge of teaching, can be developed at all, and what courses are favourable to it during the brief period allotted to teacher preparation, are critical research and development issues (Hewson et. al., 1999). In this context the main purposes of the present paper are to discuss recent studies concerning teachers’ knowledge base and to present a framework for developing and investigating courses in science teacher education, including scientific and pedagogical knowledge.

119 K. Boersma et al. (eds.), Research and the Quality of Science Education, 119—128. © 2005 Springer. Printed in the Netherlands.

120

SCIENCE TEACHER EDUCATION: ISSUES AND PROPOSALS

2. SCIENCE TEACHERS KNOWLEDGE AND VIEWS ON SCIENCE AND SCIENCE TEACHING Central issues in teachers’ knowledge base are the importance of the subject that teachers teach and their views on teaching and learning science. Knowledge of subject matter is an area that only recently has drawn the interest of researchers who have started to investigate the complex issues related to the development of it by science teachers. One consistent, striking result from several studies is that many student teachers are deficient in their understanding of important aspects of scientific knowledge that they learn to teach, despite having previously completed a number of scientific courses (De Jong, Korthagen & Wubbels, 1998). Specifically, primary teachers hold conceptions about physical phenomena and scientific concepts similar to those held by school children, although to a lesser degree and expressed in a more sophisticated language (Cochran & Jones, 1998). To some extent this applies to novice secondary teachers, particularly when they are questioned outside their major subject. Certain studies suggest that the subject matter knowledge structures of prospective teachers are often vague and fragmented, and in some cases it has been noted that student teachers are unable to present their subject matter knowledge in a coherent manner (Gess-Newsome, 1999). Other studies all over the world, consistently point out that teachers hold a variety of conceptions on teaching and learning science (Gao & Watkins, 2002, Koballa et al., 2000). These can be merged into two broad orientations (MarenticPozarnik, 2002). In the first, called didactic/reproductive, teaching is regarded as a process of transmitting knowledge and learning as a process of absorbing scientific content. In the second, called facilitative/transformative, teaching is the process of facilitating learning, which involves the construction or transformation of knowledge by students, leading possibly to conceptual change. It is remarkable that student teachers’ views on the teaching of science are largely determined by their learning experiences in scientific course during schooling and even during teacher education. Student teachers seem in practice to pay scarce attention to academic theories they are told about, such as constructivist approaches. This may be an explanation for the contradiction between exposed facilitative-constructivist views and underlying didactic practices in actual teaching, or even in planning instruction (De Jong, Korthagen & Wubbels, 1998). It appears that teachers’ beliefs and conceptions on teaching and learning act as a filter in relation to the learning of new approaches, with the result that these are frequently rejected either in whole or in part (Gunstone et al., 1993). However, there is ample evidence to suggest that science teachers have difficulties in developing constructivist views; in teaching they perform in terms of an expository model (Stofflet & Stoddart, 1994). For example, studies have pointed out that while students following a research base course appeared to have understood constructivist strategies, few of them challenged their initial conceptions, falling into the didactic/reproductive orientation (Mintrop, 2001). Yet learning a variety of teaching approaches (and the theoretical positions underlying them) can make a substantial

D. PSILLOS, A. SPYRTOUR, P. KARIOTOGLOU

121

contribution to the development of a teacher’s professional ability to teach science (Joyce, Galhoun & Hopkins, 1997). The more representations and strategies teachers have at their disposal within a certain domain, and the better they understand their students’ learning process in the same domain, the more effectively they teach in that domain by adopting constructivist methods. 3. DEVELOPING TEACHING LEARNING SEQUENCES FOR SCIENCE TEACHER EDUCATION Teacher education in general, and pre-service teacher education in particular, should be regarded as an enterprise in which teachers learn about what to teach and how to teach it in a coherent program. A sound basis is necessary for making a student teacher an inquirer and a reflective practitioner who is capable of learning with and from others in a life long process and of moving smoothly from pre-service teacher education to ongoing professional development in the course of his/her career (Hewson et al., 1999). Such a situation seems rather ideal. Pre-service teacher education is often described as being delivered in the form of isolated components (Northfield, 1998). Both the fragmentary nature of courses and the differences and tensions between pedagogies in various courses, especially content courses and courses such as didactics of science, result in student teachers claiming little gain from university education apart from their teaching practice. The development of programs in which such tensions can be resolved is a critical issue that draws the attention of researchers. As the links between pedagogical knowledge and content knowledge appear to be rather loose in graduate student teachers’ minds, an improved teacher education program would draw on a sound cognitive basis of research on teacher knowledge and cognitions (Northfield, 1998). In this context, beyond existing ordinary programs, a growing number of science education researchers have been developing and investigating the design and effectiveness of research-based proposals aimed at providing appropriate conditions for learning, instead of telling student teachers what they ought to do. In line with a developmental perspective, it is envisioned that this will lead to teachers and student teachers beginning to be transformed from practitioners and students into teacherlearners capable of conceptualising and controlling their own learning, not only in terms of scientific but also in regard to pedagogical knowledge. A pre-eminent goal of research based approaches is to create science teacher education leading to a coherent understanding and the integration of scientific and pedagogical knowledge. Towards this end, which is the focus of the present paper, we distinguish two kinds of works: namely, programs that have rather broad aims and attempt to link several courses on subject matter and pedagogy over several years, and specific medium scale courses combining targeted instruction on aspects of science and pedagogy, particularly conceptual change strategies (e.g. Hewson et al. 1999, Stofflet & Stoddart, 1994). In line with studies in science education, we consider these small-scale courses in science teacher education as innovative teaching-learning sequence (TLS) that focuses on the potential construction of fruitful links between the designed teaching

122

SCIENCE TEACHER EDUCATION: ISSUES AND PROPOSALS

and expected student learning (Lijnse, 1995). A TLS is often both a research process, bringing research and teaching closer in several contexts, and a product, like a traditional curriculum unit package that includes well researched teaching/learning activities and possible students’ learning pathways (for a research review, see Méheut & Psillos, 2004). It is at this level that targeted TLSs can contribute substantially to a deep understanding of teacher learning and understanding of both scientific and pedagogical knowledge in given contexts, in analogy with science education, despite some possible reservations that such research is rather limited in scope. A review of recent studies in science teacher education shows that several TLSs focus on the learning of scientific and pedagogical knowledge and their combination, mainly from a constructivist perspective. Works concerning scientific content investigate the thesis that learning of scientific topics in a constructivist manner may provide practical experiences out of which students can develop their understanding of constructivist models and specifically of conceptual change strategies (Kruger, Placio & Summers, 1991). In this respect, a shared assumption is that a coherent understanding of scientific knowledge provides a basis for the development of pedagogical knowledge related to teaching and learning science. Other studies advance the hypothesis that the learning of subject-specific teaching strategies, as an important part of teachers’ pedagogical knowledge, would involve the interlacing of scientific content and instructional methodology with the simultaneous provision of information to teachers on pupils’ views (Stofflet & Stoddart, 1994). However, there is disagreement among researchers whether instructional strategies form part of general pedagogical knowledge, or form an integral part of pedagogical content knowledge; different views have implications on the teaching of instructional strategies to student teachers (Morine-Dershimer & Kent, 1999, Smith, 1999 The design and effectiveness of TLS in science teacher education appears at present to be an open issue which warrants further theoretical discussion and empirical investigation. A few published studies have a model-based perspective, while others involve implicit assumptions and decisions that affect, to a considerable degree, the design and development of the corresponding teaching approaches which are not widely reported and may not even be clearly presented. One point to consider is that the scientific content in a number of published TLSs is clearly described and transformed to adapt to student teachers’ conceptions, whereas the pedagogical knowledge to be taught from a constructivist perspective is rather vaguely articulated (Din Yan Yip, 2001). In this context, we suggest that theoretical works referring to TLS for the learning of science by students may provide insights and powerful tools for developing TLS in science teacher education, if they are extended to include pedagogical knowledge. At the theoretical level, the “educational reconstruction” model developed by Kattmann et al., (1995), provides a framework for designing and validating TLS that is characterised by an emphasis on the analysis of both scientific knowledge and students’ conceptions. We argue that “educational reconstruction” can be extended and applied to science teacher education, providing a framework for designing and validating TLS in an integrated perspective that includes both scientific and

D. PSILLOS, A. SPYRTOUR, P. KARIOTOGLOU

123

pedagogical knowledge. In its original form, educational reconstruction attempts to combine a hermeneutic approach to scientific knowledge with constructivist approaches to teaching and learning. Educational reconstruction holds that clarification of science subject matter is a key issue when instruction in a particular science topic is to be developed. This is a process leading to the construction of core ideas of the content to be taught. The educational reconstruction model closely links considerations of the science concept structure with analysis of the educational significance of the content in question, as well as with empirical studies on students’ learning processes and interests (Duit et al., 1999). We suggest that such design principles may be adopted, not only in terms of the scientific knowledge but also of the pedagogical knowledge. This implies that clarification of pedagogical knowledge is a key issue if instruction in, say, constructivism is to be developed. Such a process leads to the construction of the core pedagogical ideas to be taught taking into account both epistemic dimensions and context and applications. Student teachers’ conceptions about teaching and learning science are considered in adapting and reconstructing the pedagogical content structure to their views, which are dominated by the didactic/reproductive model.

Figure 1: An adapted model for designing teaching-learning sequences in science teacher education The main features of the adapted “education reconstruction” model for designing TLS in science teacher education is illustrated in Figure 1. Briefly, the top line concerns the scientific knowledge, and the bottom one refers to the pedagogical knowledge. The construction of instruction is depicted in the middle line with four

124

SCIENCE TEACHER EDUCATION: ISSUES AND PROPOSALS

boxes relating both the pedagogical and the scientific knowledge. This process takes place when a particular interlaced content structure for instruction has to be developed; it is transformed in order to adapt the student teachers’ point of view, more specifically to their pre-instructional conceptions and their learning pathways during instruction. The science content structure, the pedagogical content structure, and student teachers’ conceptions about scientific and pedagogical concepts and phenomena are seen as being equally important parameters in the process of educational reconstruction. The model involves a non-linear design and construction of instruction. Information from one of the components influences the activities and the interpretation of the results of the other components; their interlacing, in a cycling dynamic process in which reflection on the practices during application of instruction, gives rise to new insights concerning the integration of scientific and pedagogical knowledge. Underlying the model is the assumption that knowledge is actively constructed by individual students, and that it involves social interactions in certain material settings. Scientific and pedagogical knowledge are viewed as tentative social constructions. The results of the analysis of both pedagogical and scientific knowledge, as well as preliminary ideas about the construction of an integrated instruction, play an important role in planning empirical studies on teaching and learning scientific and pedagogical knowledge. The results of empirical studies influence the processes of educational analysis, scientific and pedagogical knowledge transformation, and even the setting of goals for the specific sequence. 4. A STUDY OF A TEACHING LEARNING SEQUENCE In this section we present a brief retrospective account of the development of a TLS in terms of the adapted educational reconstruction model. i) Context. This TLS has been applied in the sixth semester (out of eight) at the Department of Primary Education, Aristotle University of Thessaloniki. The student teachers were prospective all-subject primary education teachers whom had taken courses in foundation studies (e.g. sociology, psychology), pedagogy, and discipline studies (e.g. science, mathematics, language), and whom already had some practical experience in classrooms followed by a laboratory-based course in Didactics of Science. The TLS integrated the teaching of scientific knowledge (energy content) and the teaching of the pedagogical knowledge (teaching strategies) within a constructivist framework. ii) Analysis and Transformation of scientific knowledge. Analysis of the research literature and university and school textbooks pointed out that the concept of energy constitutes a unifying concept in science. Preliminary empirical studies of both student teachers’ and pupils’ conceptions suggested that, while students are able to relate the concept of energy with life and movement, they find it difficult to comprehend basic features of energy, e.g. energy storage and energy conservation, in line with those found in the literature. Analysis and empirical investigations suggested that energy provides an appropriate scientific content in which students can be involved in true construction of knowledge. An educational reconstruction of the energy concept was deemed appropriate; the TLS was based on a qualitative

D. PSILLOS, A. SPYRTOUR, P. KARIOTOGLOU

125

treatment of five energy characteristics: storage, transformation, transfer, degradation, and conservation. However, in retrospect we may note that instead of an in-depth study, only scant observation of students’ learning pathways in energy took place. iii)Analysis and Transformation of pedagogical knowledge. Subject-specific teaching strategies were chosen as appropriate content for pedagogical knowledge (Smith, 1999). Analysis of the research literature in (science) education revealed broad conceptualisations of expository, discovery, and constructivist strategies, but a lack of specific unified modelling in terms of teaching-learning activities comprehensible to students. Initial questionnaires were addressed to the students, and in-depth learning process studies were carried out concerning the evolution of student views on teaching and learning science. Both the initial and the learning process studies found that students’ initial didactic/reproductive teaching conceptions and their alternative ideas on the scientific content seemed to be two essential components of their difficulty to learn constructivist views. Indeed, it became evident that these two components were highly interdependent. In addition, the results suggested that, while constructivist strategies were broadly understood, their differences with expository and with discovery strategies, particularly, needed to be clearly identified. Following these results, innovative unit models of these strategies (lasting from one to two hours) were developed and adapted for students. Such units were reconstructed in order to enact theoretical assumptions and avoid ambivalent terms concerning teaching strategies. The strategies were described on the basis of syntax and reaction principles. As argued by Joyce, Galhoun, & Hopkins (1997), syntax refers to the type and the structure of activities performed by both teacher and students in one teaching hour, while reaction principles refer to the type of teacher reactions to whatever his/her pupils do (Spyrtou, Kariotoglou & Psillos, 2002). iv) Construction of instruction. In terms of the model, the final form of the TLS has emerged as a product of dynamic interrelations between the above components and reflections on applications (Figure 1). Besides understanding energy, one main goal of the TLS is to render students able to design constructivist teaching units, develop clear criteria when choosing the type of teaching strategy, and discern the constructivist from the expository and discovery strategies. The achievement of these goals is pursued within an integrated constructivist teaching framework involving both the scientific and the pedagogical content (Spyrtou & Kariotoglou, 2001). Through the teaching of energy, we aim for students to understand that learning does not involve only addition or extension of their previous knowledge, e.g. as characteristics of transformation and transfer, but that it also involves a conceptual change process, e.g. as with storage, degradation, and conservation. We should not hesitate to mention that we do not want students to reject their initial teaching conceptions but to extend them through experiencing and reflecting on constructivist ones. We note that the modules on teaching strategies provided the conceptual space for reflecting on the learning practices applied during the scientific modules. This TLS comprises 11 modules out of which 5 are applied for teaching the energy content and 6 are used for teaching about expository, discovery, and (particularly) constructivist strategies (Spyrtou & Kariotoglou, 2001). As presented

126

SCIENCE TEACHER EDUCATION: ISSUES AND PROPOSALS

elsewhere, results suggest that the TLS was reasonably successful in facilitating students’ planning of strategies (Spyrtou, Kariotoglou & Psillos, 2002). Moreover, the TLS provided a tool for investigating their learning pathways, for example by revealing that the distinction between discovery and constructivist models was quite difficult for these students (Psillos, Spyrtou & Kariotoglou, 2002). In retrospect, we consider that such a distinction was not pursued in depth in applying and investigating this TLS. 5. CONCLUSIONS It appears that expert (science) teachers develop gradually integrative schemes influencing their practice, which are referred to with many concepts such as practical knowledge, implicit and subjective theories, and pedagogical content knowledge (e.g. De Jong, 2003, Van Driel, Beijaard & Verloop, 2001). However, student teachers seem to relate to a less degree subject matter views with pedagogical knowledge. We consider that TLS in general, and specifically the suggested adapted educational reconstruction model, may provide powerful tools for investigating in depth the intertwining of pedagogical and scientific knowledge by the student teachers, and for designing model- based courses that lead to their integration. REFERENCES Clark, C. & Peterson, P. (1986). Teachers’ Thought Processes. In Wittrock M.C. (Ed.), 4rth Handbook of research on teaching (pp.255-296). New York: Macmillan. Marentic Pozarnik, B. (2002). Professional Development of Teachers as a (Re)construction of their Conceptions and Teacher’s Role. Paper presented at the 6th ESERA Summer School. Aug, 25-31, Radovlijkca, Slovenia. Cochran, K. & Jones, L. (1998). The Subject matter knowledge of Preservice Science Teachers In B.J. Fraser and K.G. Tobin (Eds.), International Handbook of Science Education (pp 707-717). Dordrecht: Kluwer. De Jong, (2003). Exploring Science Teachers’ Pedagogical Content Knowledge. In D. Psillos, P. Kariotoglou, V. Tselfes, E. Hatzikraniotis, G. Fassoloupolos & M. Kallery, (Eds.), Science Education Research in the Knowledge Based Society (pp.373-381). Dordrecht: Kluwer.. De Jong, O., Korthagen, F. & Wubbels, T. (1998). Research on Science Teacher Education in Europe: Teacher Thinking and Conceptual Change. In B.J. Fraser and K.G. Tobin (Eds.), International Handbook of Science Education (pp.745-758 ). Dordrecht: Kluwer. Din Yan Yip, (2001). Promoting the development of a conceptual change model of science instruction in prospective secondary biology teachers. International Journal of Science Education, 23(7), 755-770. Gao, L. & Watkins, D.A. (2002). Conceptions of teaching held by school science teachers in P. R. China: identification and cross-cultural comparisons. International Journal of Science Education, 24(1), 61-79

D. PSILLOS, A. SPYRTOUR, P. KARIOTOGLOU

127

Gess-Newsome, J. (1999). Secondary teachers’ knowledge and beliefs about subject matter and their impact on instruction. In J. Gess-Newsome & N. G. Lederman (Eds.), Examining pedagogical content knowledge, (pp. 51-94). Dordrecht: Kluwer. Gess-Newsome, J. & Lederman, N. G. (Eds.), Examining pedagogical content knowledge. Dordrecht: Kluwer. Gunstone, R. and Slattery, M., Baird, J. & Northfield, J. (1993). A Case Study of Development in Pre-service Science Teachers. Science Education, 77(1), 47-73. Joyce, B., Galhoun, E. & Hopkins, D. (1997). Models of learning-tools for teaching. Buckingham-Philadelphia: Open University Press. Hewson, P., Kerby, H. & Cook, P. (1995). Determining the conceptions of teaching science held by experienced high school science teachers. Journal of Research in Science Teaching, 32(5), 503-520. Hewson, P.W., Tabachnick, B.R., Zeichner, K.M. & Lemberg, J. (1999). Educating Prospective Teachers of Biology: Findings, Limitations, and Recommendations. Science Education, 83(3), 373-384. Kattmann U., Duit R., Gropengieber, H.& Komorek, M., (1995). A model of Educational Reconstruction. Paper presented at The NARST annual meeting. San Francisco. Koballa, T., Gräber, W., Coleman D. & Kemp, A. (2000). Prospective gymnasium teachers’ conceptions of chemistry learning and teaching. International Journal of Science Education, 22(2), 209-224 Duit R., Roth, W-M, Komorek, M. & Wilbers, J. (1998) Studies on educational reconstruction of chaos theory. Research in Science Education 27, Research in Science Education 27, 339-357 Kruger, C., Palacio, D. & Summers, M. (1991). Understanding energy. Primary School Teachers and Science (PSTS) Project. Oxford: Oxford University Department of Educational Studies. Lijnse P.-L. (1995). “Developmental Research” as a Way to an Empirically Based “Didactical Structure” of Science, Science Education 79(2,), 189-199. Meheuet, M. & Psillos D. (2004). Teaching – learning sequences: aims and tools for science education research. International Journal of Science Education, Special Issue (forthcoming). Morine-Dercshimer, G. & Kent, D. T (1999). The Complex Nature and Sources of Teachers’ Pedagogical Content Knowledge. In J. Gess-Newsome & N. G. Lederman (Eds.), Examining pedagogical content knowledge, (pp. 21-50). Dordrecht: Kluwer. Mintrop H. (2001). Educating Students to Teach in a constructivist Way – Can It All Be Done? Teachers College Record, 103(2), 207-239. Nortfield J. (1998). Teacher educators and the Practice of Science Teacher Education. International Handbook of Science Education (pp. 707-717). Dordrecht: Kluwer. Psillos, D., Spyrtou A. & Kariotoglou P. (2002). Investigating the complexity of teacher’s conceptions on science teaching: issues and tools. Invited workshop for the 6th ESERA Summerschool. Aug, 25-31, Slovenia. Smith, D.C. (1999). Changing our teaching: The role of pedagogical content knowledge in elementary science. In J. Gess-Newsome & N.G. Lederman (Eds.), Examining pedagogical content knowledge (pp. 163-197). Dordrecht: Kluwer.. Spyrtou, A. & Kariotoglou, P. (2001) Interlacing content and methodology in educating primary student teachers. In M. Bandiera, S. Caravita, E. Torracca, M. Vicentini (Eds.), Research Education in Europe: The Picture Expands, (pp.651-658). Rome: Litoflash..

128

SCIENCE TEACHER EDUCATION: ISSUES AND PROPOSALS

Spyrtou, A., Kariotoglou, P. Psillos, D. (2002). A 3-D approach to investigate the development of lesson planning. Paper presented at the Third Panellenic Conference, Didactics of Science & Application of New Technologies in Education. May 2-5, Heraklio, Crete. Stofflett, R. & Stoddart, T. (1994). The Ability to Understand and Use Conceptual Change Pedagogy as a Function of Prior Content Learning Experience. Journal of Research in Science Teaching, 31(1), 31-51. Van Driel, J.H., Beijaard, D. & Verloop, N. (2001). Professional Development and Reform in Science Education: The role of Teachers’ Practical Knowledge. Journal of Research in Science Teaching, 38(2), 137-158.

OUTCOMES OF PROFESSIONAL DEVELOPMENT IN PRIMARY SCIENCE: DEVELOPING A CONCEPTUAL FRAMEWORK

PAUL DENLEY, KEITH BISHOP University of Bath, UK

ABSTRACT In a climate of continuing change, the continuing professional development of science teachers is an important issue, but one which is subject to resource constraints. It is vital that professional development is as effective as possible. This paper describes an attempt to examine the outcomes of such activities and to try to apply an existing categorisation system for framing these outcomes in terms of their impact at different levels on pupils, teachers, and schools. Data come from a number of large professional development projects for teachers in primary schools, the projects were funded by an independent educational charitable trust over a period of six years. Analysis of the data confirms that many of the categories proposed in the early 1990s are still applicable today but that new ones are needed to extend the framework, perhaps to reflect a changed context for science in primary schools. An attempt is made to show the relationship between categories identified.

1. THE CHANGING CONTEXT OF PROFESSIONAL DEVELOPMENT In recent years in the field of school improvement (see, for example, Stoll & Fink, 1996), the quality of teaching has become increasingly recognised as a key factor in raising standards in schools. In turn, the quality of teaching is related not only to the calibre of teachers recruited and their initial training, but also to their on-going professional development. This is especially true in a context of massive educational change through the introduction of new government initiatives and the promotion of research-based practice. The implication for and the changing context of professional development of teachers has been explored in some depth by, for example, Watson and Fullan (1992), Guskey & Huberman (1995), Craft (1996), and Loucks-Horsley et al. (1998). Craft identified a range of weaknesses inherent in much professional development practice at the time of writing. She argued that professional development tended to be geared to individual needs rather than school needs, that provision was largely in the form of courses, attendance was voluntary, and there was little acknowledgement that participants would almost certainly have different starting points. This sort of professional development also disrupted teaching and appeared to have limited impact on practice with little or no dissemination thereafter. Dissatisfaction with this kind of provision led to the 129 K. Boersma et al. (eds.), Research and the Quality of Science Education, 129—140. © 2005 Springer. Printed in the Netherlands.

130

OUTCOMES OF PROFESSIONAL DEVELOPMENT IN PRIMARY SCIENCE

promotion of school-based approaches with teachers learning from one another. More recently the situation in England has been characterised by the introduction of new educational initiatives and priorities with their own associated professional development programmes (see, for example, DfES, 2003). The emphasis here is on institutional needs rather that those of individuals, and attendance is often required or at least expected. The demonstration of impact is still a key issue with clearly defined sets of performance indicators linked to national assessment programmes. The picture of provision is currently fragmented with numerous agencies and independent bodies, in addition to higher education institutions and local education authorities themselves, offering a variety of courses and activities. Accompanying such diversification, however, are concerns about the suitability, and impact of what is available, questions regarding its effect on teacher practice and its long-term sustainability (see Ingvarson, 1988; Craft, 1996; Guskey, 2000; Adey, 2004). There is also the need to address concerns with the balance between the needs of the individual teacher and the needs of the institution; to move to a more holistic view of ‘continuing professional development’ rather than ‘INSET’ (‘INSErvice Training’). Hargreaves and Fullan (2000) suggest that “professional learning is not to be found in a choice between school-based and course-based modes of provision, but in an active integration of and synergy between the two”. However, there are few research-based models for effective professional development (Bell & Gilbert, 1996; Adey, 2004) and, when resources are limited, it is important that they are used to maximum effect. 2. THE ASTRAZENECA SCIENCE TEACHING TRUST (AZSTT) To mark the Millennium, the pharmaceutical company, AstraZeneca PLC, established the AstraZeneca Science Teaching Trust (AZSTT) through a £20m endowment. An initial priority for the Trust was to provide support for professional development programmes to raise the confidence and competence of primary teachers in science. Trustees invited proposals from appropriately qualified ‘providers’ who would design and run their own projects with clusters of about twenty primary schools and with funding of about £90 000 (€136 500) per provider. (Details of projects can be found on the Trust’s website http://www.azteachscience.co.uk.) The programmes have been designed by the ‘providers’ themselves (usually staff from university education departments working with local education authority staff) allowing their own distinctive approaches and ways of working (see for example, Lloyd et al., 2000; Crebbin, 2001; Jarvis and Pell, 2001; Rodrigues, 2003). All of the Trust’s projects were required to conduct their own internal evaluations, but the Trustees also commissioned from the University of Bath an independent evaluation which has given a valuable opportunity to look across a range of approaches.

P. DENLEY, K. BISHOP

131

Data collection and analysis As part of the external evaluation, data have been collected since 1997. Data sources include internal evaluation reports from providers to the Trust, interviews with the providers themselves, and, through visits to a sample of project schools, interviews with science co-ordinators and head teachers. The school interviews explored perceptions of the aims of the project, potential benefits in the short and the longer term, how the project would impact on practice as teachers of science and as subject leaders, how success might be judged, and what evidence would be used to justify any claims. In order to build a broader picture, a questionnaire survey was administered in 1998, which was designed to validate the interview data and identify emergent themes. A second survey in 2003 was distributed to a sample of schools within all of the current and past projects The latter was more clearly focussed on identifying examples of outcomes relating to the typology discussed in this paper. Qualitative analysis software was used on the questionnaire survey data, allowing us to explore responses from head teachers and science co-ordinators within and across projects. The most recent questionnaire survey has provided confirmation of earlier work (Denley and Bishop, 1999) and extended the data set to a wider range of projects. 3. EVALUATION AND THE NATURE OF PROFESSIONAL DEVELOPMENT OUTCOMES The AZSTT Trustees are looking for evidence of impact through both the internal and external evaluation of the professional development programmes supported. We have pointed out that attribution of impact relating directly to provider input is difficult to establish as there are many key variables. Furthermore, it was evident that in the early stages not all providers had outcomes sufficiently well defined to allow a reliable assessment of potential impact. There was also not an appropriate conceptual framework to enable the outcomes to be discussed in terms of their effect on teachers’ practice and children’s achievement or of their sustainability in the longer term. The ‘Level of Use’ scale (Hall & Loucks, 1977) could have provided a research tool to identify impact but it was quite narrowly focussed. The typology generated by Kinder and Harland (1991) in their analysis of primary science INSET undertaken in the late 1980s, however, seemed to have greater potential, providing us with a research-based frame of reference which could be used to compare the nature of the outcomes reported by the providers with those derived from our own evaluation. All projects were required to define at the outset their own set of performance indicators to be used in their internal evaluation. Our purpose here is not to comment on how well those indicators have been met, but rather to examine the outcomes and to try to make some sense of what has been achieved. Kinder and Harland generated a typology of nine categories of INSET outcomes (Figure 1):

132

OUTCOMES OF PROFESSIONAL DEVELOPMENT IN PRIMARY SCIENCE

1 ‘Material and provisionary’ “the procurement of physical resources and service as a result of participation” 2 ‘Informational’ “the state of being briefed or cognizant of the background facts and news about curriculum and management developments” 3 ‘New awareness’ “a changed perception about one or more aspects of primary science accruing from the school’s initial involvement” 4 ‘Value congruence’ “the personalised versions of curriculum and classroom management which inform a primary practitioner’s teaching and how far these come to coincide with INSET messages about ‘good practice’” 5 ‘Affective’ “acknowledge that there is an ‘emotional’ experience inherent in any learning situation .... an increased confidence to undertake science in the classroom” 6 ‘Motivational and attitudinal’ “enhanced motivation to undertake science” 7 ‘Knowledge and skills’ “increased knowledge and skills outcomes denote deeper levels of understanding, critical reflexivity and theoretical rationales” 8 ‘Institutional-strategic’ “collective impact of groups of teachers - in this case, whole schools and that such a corporate outcome can have a constructive influence on teachers’ efforts to change their own practice” 9 ‘Impact on classroom practice’ “any changes or developments in teachers’ classroom delivery of science that can be attributed to the various components of the ESG scheme” Figure 1. A typology of INSET outcomes (Kinder and Harland, 1991) The construction of this framework was supported by evidence drawn from the accounts of science coordinators, teachers, head teachers, and advisory teachers in five case study schools. We believed the typology might be a particularly useful

P. DENLEY, K. BISHOP

133

frame of reference for us to use, as the professional development provision supported by the AZSTT initiative has some features in common with that described by Kinder and Harland. We have tried to examine the match between project outcomes and Kinder and Harland’s framework, to identify any differences and to extend the typology if necessary. 4. CATEGORISING OUTCOMES FROM OUR STUDY The following are some examples of outcomes in each of Kinder and Harland’s nine categories from AZSTT projects. It is clear that there were ‘Material and provisionary’ outcomes from all projects, as the budget included a £1000 grant to be spent on books, equipment, and other resources. The significance of this varied, but it stimulated teacher enthusiasm and gave the science co-ordinator a stronger incentive to provide subject leadership. The knowledge and experience of all the project teams led to clear ‘Informational’ outcomes from all the projects. Kinder and Harland distinguish between this category and the ‘Knowledge and skills’ category which might be more explicit and more directly linked to the desired outcomes of the project. These less formal outcomes are more concerned with building a broader foundation for the teaching of the subject. It was apparent that information about primary science teaching was disseminated through informal discussions with team members or other project teachers. Similarly, there was universal agreement from all schools that involvement with the project had resulted in ‘New awareness’ outcomes. Projects adopted different approaches to working with schools and in encouraging them to work together, but it is clear that the “changed perceptions” alluded to by Kinder and Harland came both directly from members of the project teams and from contact with practice in other schools. In some cases it was more a ‘new vision’ than ‘new awareness’ – a practical demonstration that ideas could actually work in the classroom. Kinder and Harland assert that ‘New awareness’ outcomes in themselves are no guarantee that changes in practice will follow without the presence of their next category – ‘Value congruence’. These outcomes were apparent in responses from science co-ordinators interviewed and from the questionnaire data. It would perhaps be truer to say that the projects resulted in a greater degree of congruence rather than to suggest complete congruence. In a latter study, Harland and Kinder (1992) found that there was an increased likelihood of impact on practice when there was value congruence. Kinder and Harland define their category of ‘Affective’ quite narrowly terms of the emotional dimension, particularly in influencing confidence. Enhancing teacher confidence was seen to be a major aim of all projects; thus, this outcome was clearly recognised by the vast majority of teachers taking part and was seen as a major benefit of involvement. ‘Motivational and attitudinal’ outcomes were evident through raising of/restoring the status of science, particularly at a time when science came under threat from competing national strategies to promote literacy and numeracy. Kinder

134

OUTCOMES OF PROFESSIONAL DEVELOPMENT IN PRIMARY SCIENCE

and Harland point out that motivation and positive attitudes may be short-lived without a follow-up programme “… to monitor and support subsequent classroom practice”. Kinder and Harland’s category of ‘Knowledge and skills’ recognises that there are several different kinds of knowledge (citing Eraut, 1994). Although this category was reflected in all projects, there were differences in emphasis in relation to conceptual and procedural aspects. The major ‘Institutional-strategic’ outcomes related to the revision and development of schemes of work for science but there were others: more inventive and effective science displays in classrooms and more conversations about science in schools. One area where we have tried to look for evidence is how well the projects have been able to ‘embed’ themselves into the culture of the school. One way in which w did this was by asking the question, if the participating teachers were to leave, what would be left behind? This notion of embedding is hard to define in terms of concrete outcomes, but they could be categorised here. Kinder and Harland recognise the difficulty of validating outcomes relating to ‘Impact on classroom practice’ – teachers say that more investigative work is going on but is it? – and what about the quality? Headteachers’ comments do confirm what teachers say about impact on practice. Several projects explored ways of teachers working together with other teachers or project team members in real classroom situations. These strategies of team-teaching, peer support, modelling, and coaching were seen to be particularly valuable in having a direct impact on classroom practice in a way that was easy to validate. This would link with the association between strategies such as coaching and changes in practice (see Joyce and Showers, 1984 & 1995; Adey, 2004). 5. EXTENDING THE TYPOLOGY In the first instance, our analysis naturally led us to look for conformity between Kinder and Harland’s typology and the outcomes we had identified through our evaluation. It was apparent, however, that some outcomes did not fit.. We therefore present four additional categories: ‘Impact on pupil response’ Kinder and Harland state that impact on classroom practice (i.e. what teachers teach) “represents the ultimate goal of all the earlier outcome types”. However, we suggest that the ultimate goal ought to be seen in terms of a more direct impact on pupils. We use the word ‘response’ rather than achievement, performance, or attainment, or even learning, in order to be more inclusive. Impact on pupils was seen as taking different forms in many of the projects. For example, several projects administered pre- and post-tests to gain measures of pupil attitudes towards science (e.g. Jarvis and Pell, 2001 & 2002) focusing on confidence, interest, readiness to ask questions, motivation, and enjoyment. Some assessed pupil progress through collective assessment techniques such as floor books (McMahon and Davies, 2004) and poster displays or exhibitions. Others reported pupils’ enhanced readiness to discuss

P. DENLEY, K. BISHOP

135

science outside the confines of the science classroom. One project identified a range of indicators that offered evidence of pupils’ improved capability as learners (Bianchi, 2003). In our view, the sustained evidence of pupil enhancement in science learning from these projects should be a significant indicator in determining the level of impact a programme of professional development attains. ‘Leadership and management’ Kinder and Harland recognise that their categorisation was shaped by the principal aim to impact on classroom practice, and that it might have been different with a different aim. The AZSTT projects have mostly focussed on the development of the science co-ordinator. Thus, improving his or her competence and confidence have been overt goals in project design. Outcomes relating to these teachers as subject leaders and managers are frequently stated as major instances of impact of the projects. Jarvis et al (2003) report as an outcome of their AZSTT funded project, the enhancement of primary teachers’ subject knowledge, whilst Rodrigues (2003) claims teachers improved their pedagogical content knowledge; but it is clear that outcomes in this area are not just to do with the enlargement of knowledge about science. Recent projects we have evaluated include strong emphases of partnership teaching and co-teaching as mechanisms to develop leadership and management skills. Thus, teachers have enhanced their skills in how to work with their colleagues to improve their practice, citing confidence to give input at staff meetings and running school-based sessions, as well as offering in-class support for other colleagues. ‘Collateral’ Kinder and Harland’s ‘Institutional-strategic’ category is seen solely in terms of institutional impact in the context of science. We have detected ‘spin-off’ outcomes from what has happened in science to other subject areas or to whole school aspects of organisation and management. Thus, whether intended or not, the projects have had outcomes beyond that focal subject area. Examples relate to other subject areas where other subject co-ordinators have been able to take approaches used by the science co-ordinator into their own subject, or where generic messages about running school-based professional development can be applied to other fields. Crebbin (2001) describes ‘partnership teaching’ as the central concept underpinning the project devised by his AZSTT funded project. In our project evaluation it was evident that schools embracing this approach to professional development found ways to encourage teachers in other subjects to work together. In another project, the technique of co-teaching, between pre-service teachers specialising in science and classroom teachers, has ‘spun-off’ to other subjects. ‘Individual-strategic’ Without wishing to add categories unnecessarily, we also found some outcomes which related to individual teachers’ strategic planning that seemed to be mid-way between ‘Institutional-strategic’ outcomes and those concerned with ‘Impact on practice’ in the classroom. Short- and medium-term planning documents are key instruments providing direct evidence of change. They translate the scheme of work

136

OUTCOMES OF PROFESSIONAL DEVELOPMENT IN PRIMARY SCIENCE

into teaching activities, show how investigative work is being organised, and how resources (existing and new) are being used. They may indeed be more important than the scheme of work, which in itself may describe curriculum rhetoric rather than reality. 6. RELATIONSHIPS BETWEEN CATEGORIES Kinder and Harland are at pains to stress the tentative nature of their typology and are even less firm about their attempt to suggest an ordering of their categories. They do suggest a tentative hierarchical model (see Figure 2), whilst at the same time recognising that the relationship is probably more complex.

I N S E T 3rd Order

PROVISIONARY: INFORMATION : NEW AWARENESS

2nd Order

MOTIVATION :

1st Order

VALUE CONGRUENCE : KNOWLEDGE & SKILLS

I M P A C T

AFFECTIVE :

O N

INSTITUTIONAL

P R A C T I C E

Figure 2. A tentative hierarchy of outcomes (Kinder and Harland, 1991) We would agree that the complexity and inter-dependence of relationships renders any attempt to show a strictly positional (particularly a hierarchical) relationship problematic. For example, the relationship between confidence (an ‘Affective’ outcome) and competence (in ‘Knowledge and skills’) is hard to unravel – is one confident as a result of being competent? – or is confidence a pre-requisite for developing competence? The hierarchy cannot deal adequately with the time dimension – some outcomes are short-term but may be necessary pre-requisites for longer-term impact. Perhaps the ultimate goal of impact on pupils’ learning requires ‘scaffolding’ with a firm framework of ‘lower order’ outcomes.

P. DENLEY, K. BISHOP

137

7. DISCUSSION Our intention in this paper has been to examine our data regarding outcomes of the AZSTT pilot projects using Kinder and Harland’s framework. We have shown agreement with and support for their categorisation, but there are clearly differences in emphasis and balance. We have also found it necessary to add some additional categories. A changed context for continuing professional development (CPD) in England in the intervening years could explain the less explicit emphasis on ‘Informational’ and ‘New awareness’ outcomes and the need to introduce our ‘Impact on pupil response’ category to reflect the emergence of these outcomes (particularly those to do with pupil performance), as being the ultimate goal, certainly for external agencies. Although ‘Knowledge and skills’ outcomes are still important, there is perhaps a shift away from personal knowledge of science concepts and towards the translation of personal knowledge into appropriate classroom science activities. We would suggest that the Kinder and Harland typology (perhaps extended with some additional categories as we have done) has more potential in helping those trying to examine professional development outcomes than the Hall and Loucks (1977) ‘Levels of Use’ or Joyce and Showers (1980) ‘levels of impact’ which are often cited in discussions of the effectiveness of professional development. Kinder and Harland intended to use the Joyce and Showers model to provide a conceptual framework for their study but found that “the nature and range of outcomes were more complex and broad ranging” than the model could accommodate. We find ourselves in agreement with criticisms of this framework (e.g. Craft, 1996) on the grounds of its oversimplification of outcomes and inappropriateness for some CPD activities. Joyce and Showers were focussing quite narrowly on the acquisition and use of new teaching strategies. The programme Kinder and Harland were evaluating had a broader professional development agenda in supporting primary science. This is continued in the AZSTT projects which are also about much more than introducing new teaching methods. Our study also suggests that presenting categories of outcome in a progression or hierarchy is also problematic for at least two reasons. First, the relationships are not linear and interdependence is more complex than a simple sequence or hierarchy is able to show; and second, the hierarchy suggests an end-point or ultimate goal which may therefore undervalue ‘lower’ categories of nonetheless important outcomes. Perhaps a concept map could provide an alternative form of representation (see Figure 3).

138

OUTCOMES OF PROFESSIONAL DEVELOPMENT IN PRIMARY SCIENCE

Figure 3. Mapping CPD outcomes

P. DENLEY, K. BISHOP

139

This representation is offered even more tentatively than Kinder and Harland offered their hierarchy, but it has been included to provoke some debate about its improvement or other alternative ways of representing these inter-relationships. The advantage of the concept map is that it not only allows positional relationships to be established and displayed, but it also requires the nature of those relationships to be defined. This map is not complete and still oversimplifies the complexities it attempts to represent, but it does force some thinking about the interdependence of different categories of outcome without being too hierarchical. In conclusion, we have found it useful to think about analysing the outcomes from these AZSTT CPD projects using a typology/categorisation system. Although there was some degree of diversity, all these projects were in the context of primary science and were planned and delivered within a framework containing some predetermined elements. We are interested to know whether the extended typology can be applied more generally to other CPD activities or projects, and whether it can comprehensively cover the wide range of intended outcomes. REFERENCES Adey, P. (2004). The professional development of teachers: practice and theory. Dordrecht: Kluwer. Bell, B. & Gilbert, J. (1996). Teacher development: a model for science education. London: Falmer Press. Bianchi, L. (2003). Better Learners. Primary Science Review, 80, pp 22-24. Craft, A. (1996). Continuing Professional Development. London: Routledge/Open University. Crebbin, C. (2001). Partnership teaching in primary science. Primary Science Review, 70, pp 22-25. Denley, P. & Bishop, K. (1999). Making Sense of Professional Development in Primary Science. Paper presented to ‘The Challenge of Change’ conference held at the University of Durham, 7-9 July 1999. Department of Education and Science (1985). Science 5-16: A statement of policy, London: HMSO. Department for Education and Skills (2003). The Primary National Strategy, http://www.standards.dfes.gov.uk/primary (6 April 2004). Eraut, M. (1994). Developing professional knowledge and competence. London: Falmer Press. Guskey, T. (2000). Evaluating professional development. London: Sage. Guskey, T. & Huberman, M. (Eds.) (1995). Professional Development in Education: New Paradigms and Practices. New York: Teachers’ College Press. Hall, G. & Loucks, S (1997). A developmental model for determining whether the treatment is actually implemented. American Educational Research Journal, 14 (3), 263-276. Harland, J. & Kinder, K. (1992). Mathematics and science courses for primary teachers. Slough: NFER. Harland, J. & Kinder, K. (1997). Teachers’ continuing professional development: framing a model of outcomes. British Journal of In-service Education, 23 (1), 71-84.

140

OUTCOMES OF PROFESSIONAL DEVELOPMENT IN PRIMARY SCIENCE

Hargreaves, A. & Fullan, M. (2000). Mentoring in the new millennium. Theory into Practice, 39 (1), pp 50-56. Ingvarson, L. (1988). Factors affecting the impact of inservice courses for teachers: implications for policy. Teaching and Teacher Education, 4 (2) pp 139-155l. IPSE (1988). Initiatives in Primary Science: An Evaluation Report. Hatfield: Association for Science Education. Jarvis, T. & Pell, A. (2001). Developing attitude to science scales for use with children of ages from five to eleven years. International Journal of Science Education, 23 (8), pp 847-862. Jarvis, T. & Pell, A. (2002). Changes in primary boys’ and girls’ attitudes to school and science during a two-year science in-service programme. The Curriculum Journal, 13 (1), pp 43-69. Jarvis, T., Pell, A. & McKeon, F. (2003). Changes in primary teachers’ science knowledge and understanding during a two year in-service programme. Research in Science & Technological Education, 21 (1), pp 17-42. Joyce, B. & Showers, B. (1980). Improving in-service training: the messages of research. Educational Leadership, 37 (5), pp 379-85. Joyce, B. & Showers, B. (1984). Transfer of training: the contribution of ‘coaching’. In D. Hopkins & M. Wideen (Eds.) Alternative Perspectives on School Improvement. Lewes: Falmer. Joyce, B. & Showers, B. (1995). Student achievement through staff development. New York: Longman. Kinder, K. & Harland, J. (1991). The Impact of INSET: The Case of Primary Science. Slough: NFER. Lloyd, J., Braund, M., Crebbin, C. & Phipps, R. (2000). Primary Teachers' Confidence About and Understanding of Process Skills. Teacher Development, 4 (3). Loucks-Horsley, S., Hewson, P., Love, N. & Stiles, K. (1998). Designing professional development for teachers of science and mathematics. Thousand Oaks, CA, Corwin. McMahon, K. & Davies, D. (in press). Assessment for enquiry: Supporting teaching and learning in primary science, Science Education International. Rodrigues, S. (2003). Partnership in Primary Science Project: Developing a community of practice to encourage the development of pedagogical content knowledge. Science Education International, 14 (2), pp 2-11. Rodrigues, S., Marks, A. & Steel, P. (2003). Developing science and ICT pedagogical content knowledge: a model of continuing professional development. Innovations in Education and Teaching International, 40 (4), pp 386-394. Stoll, L. & Fink, D (1996). Changing our schools: linking school effectiveness and school improvement. Buckingham: Open University Press. Watson, N. & Fullan, M. (1992). Beyond school district-university partnerships. In M. Fullan & A. Hargreaves, (Eds.) Teacher Development and Educational Change. Lewes: Falmer.

CHEMISTRY TEACHERS RESEARCH THEIR OWN WORK: TWO CASE STUDIES RACHEL MAMLOK-NAAMAN, OSHRIT NAVON, MIRIAM CARMELI, AVI HOFSTEIN The Weizmann Institute Of Science, Israel

ABSTRACT Ten high-school chemistry teachers and two staff members from the Science Teaching Department of the Weizmann Institute of Science who served as coordinators participated in a one-year professional development program aimed at enhancing the teaching and learning of chemistry using Action Research methodology. In Action Research, teachers research their own practice of teaching. The program involved monthly meetings throughout the year at the Science Teaching Department. Here we present two case studies which will serve as examples of the program. In the first study, teachers investigated their students’ misconceptions about the electrical conductivity of metals and ionic materials. The second study focused on the behavior of non-science-oriented students and their attitudes toward chemistry studies. The program included an evaluation of the process that teachers underwent while doing their classroom research; the evaluation was done by the workshop coordinators. Based on the findings of these two studies, we may conclude that involving teachers in an intensive workshop dealing with various aspects of teaching and with investigating their own work, provides teachers with tools for systematically diagnosing students’ learning difficulties and the ability to change their instruction accordingly. Moreover, the workshop experience supported an environment of collegiality and enabled teachers to collaborate with professional researchers and other teachers.

1. THEORETICAL BACKGROUND Action Research is an inquiry in which teachers research their own work and their students' learning in the classroom (Feldman & Minstrel, 2000). According to Feldman (1996), the primary goal of Action Research is not to generate new knowledge, but rather to improve and change classroom practices. The process of Action Research can be described as a cycle of planning, implementation, observation, and reflection. Implementing changes and improving classroom practices is an iterative process (Kemmis & Mctaggart, 1988; O’Hanlon, 1996; Zuber–Skerritt, 1996). Each cycle of Action Research is repeated, and all cycles together form a spiral. Lewis and Munn (1987) indicated three main reasons for conducting teacher-based research: (1) to try to determine what is actually going on, (2) to monitor and thereby formatively influence the direction of new developments, and (3) to evaluate what is already taking place. 141 K. Boersma et al. (eds.), Research and the Quality of Science Education, 141—155. © 2005 Spriner. Printed in the Netherlands.

142

CHEMISTRY TEACHERS RESEARCH THEIR OWN WORK

Loucks-Horsley, Hewson, Love, & Stiles (1998) wrote that Action research has evolved in the education community into an ongoing process of systematic study in which teachers examine their own teaching and students' learning through descriptive reporting, purposeful conversation, collegial sharing and critical reflection for the purpose of improving classroom practice. (p. 95) The use of Action Research as a strategy for professional development is based on the following assumptions (Loucks-Horsley et al., 1998, p. 97): • • •

Teachers are intelligent, inquiring individuals with important expertise and experiences that are central to the improvement of education practice. By contributing to or formulating their own questions and by collecting data to answer these questions, teachers grow professionally. Teachers are motivated to use more effective practices when they are continuously investigating the results of their actions in the classroom. For Action Research to be an effective means of helping teachers to reflect on their practice, we must provide them with opportunities to engage in life-long professional development (Hofstein, 2001). These opportunities will provide them with an environment of support, collegiality, and a chance to collaborate with professional researchers and other teachers who teach the same or related subjects, in a milieu that encourages teachers’ reflection on their classroom practice and on the results of their research efforts. According to Holly (1991), collaboration is now seen as a major form of professional development. Indeed, this collaborative inquiry should be conducted by professionals acting as reflective practitioners (Schon, 1983). When teachers reflect critically on their experiences, they scrutinize them and improve their ability to teach and understand their students’ learning difficulties (Obaya, 2003). Van Zee (1998) discussed the meaning of the term “teacher researcher”, the rationale for preparing teachers to do research as they learn to do it, and suggested ways to educate teachers as researchers. Typically, teachers who are inexperienced in Action Research need support and training regarding its methodology, procedures, and activities. This includes designing tools, collecting data, analyzing and interpreting the results, and finally, applying the findings in the science classroom. Engaging in professional development provides teachers with an opportunity to share the results of their classroom research and related pedagogical activities with fellow teachers who will later provide them with feedback and other ideas. Having a long and varied history, Action Research was first introduced by Kurt Lewin in the 1940s. However, only in the last 10 years have we discovered the potential that this strategy has in our effort to bring about changes in the science classroom. In recent years, Action Research has been widely used as a tool for the professional development of teachers in all stages of their career, including their pre-

R. MAMLOK-NAAMAN, O. NAVON, M. CARMELI, A. HOFSTEIN

143

service preparation. The following are a few examples from the literature about programs for the professional development of teachers regarding Action Research, and for prospective teachers. Korthagen (1985) described how a Dutch teachers’ school of education developed a program on how to prepare future teachers to reflect on their experiences as a means of directing their own growth in the profession. Gipe and Richards (1992) conducted a study in which they examined the relationship between future teachers' reflections and the advancement of their teaching abilities in any early placement field. The analysis of data that was collected over one semester from various journals and multiple observations indicated that teacher preparation programs should foster reflective thinking as an important aspect of improving practitioners’ teaching abilities. Gore and Zeichner (1991) stressed the importance of Action Research in the framework of different conceptions of reflective teaching, for example, regarding the social view of reflection that underlies the University of Wisconsin-Madison elementary teacher education program. Their study was conducted by one supervisor with eighteen student teachers from 1988–89 and was analyzed for evidence of reflective thinking, which was favored. In this paper we describe a one-year professional development program aimed at enhancing the professional skills of teachers using Action Research methodology. The program’s objectives were as follows: • • •

To enhance the professional development of the participants through their experience with Action Research. To encourage the creation of a professional community of chemistry teachers, and a leading-teachers team at school. To establish a leading-teachers team that will perform Action Research with teachers 2. THE PROGRAM Population Ten chemistry teachers participated in the program, led by two staff members from the Science Teaching Department at the Weizmann Institute of Science in Israel, who served as coordinators. The participants were high-school chemistry teachers. They met for a four-hour meeting each month, for a period of 30 weeks. Table 1 presents the background of each teacher. The participants were chemistry teachers with at least 12 years of teaching experience; all had a teaching diploma in chemistry. All participating teachers had participated in courses and programs for chemistry teachers in previous years. Nevertheless, they lacked experience in doing research and were not acquainted with qualitative research in general and with Action Research in particular.

144

CHEMISTRY TEACHERS RESEARCH THEIR OWN WORK Table 1: The background of each teacher

Teacher’s name

Education

Alice

B.Sc. in 17 years Chemistry and Computer Science B.Sc. in 14 years Chemistry and M.Sc in Science Teaching

Debra

Sara

Eva

Lea

Linda

Sima

Sara

B.Sc. in Psychology and Chemistry Practical engineer B.Sc and M.Sc in Chemistry and B.Sc. in Biology B.Sc and M.Sc. in Chemistry

Teaching experience

20 years

Educational Roles

Principal’s assistent

Coordinator of Course for chemistry teachers chemistry teachers’ and the coordinators coordinator of social activities in school Coordinator of Course for chemistry teachers chemistry teachers’ coordinators

28 years – 9 in Israel

Chemistry teacher

27 years

Chemistry teacher

B.Sc. in 20 years Chemical engineering and M.Sc in Education B.Sc. and M.Sc. 15 years in Chemistry and B.Sc. in Computer Science B.Sc. and M.Sc. 22 years in Chemistry

Debra

B.Sc. in Chemistry

12 years

Orit

B.Sc. and M.Sc. 25 years in Chemistry

Participation in workshops and programs Advanced study programs for chemistry teachers

Course for chemistry teachers’ coordinators

Advanced study programs for chemistry teachers Coordinator of Course for chemistry teachers chemistry teachers’ coordinators

Chemistry teacher

Advanced study programs for chemistry teachers

Regional high school chemistry consultant Chemistry teacher

Course for chemistry teachers’ coordinators Advanced study programs for chemistry teachers Course for chemistry teachers’ coordinators

Chemistry teacher

R. MAMLOK-NAAMAN, O. NAVON, M. CARMELI, A. HOFSTEIN

145

The Content of the Workshop The program consisted of two main components: A The workshop: the coordinators and the members of the group met once a month. The syllabus of the workshop included the following subjects: • The principles of Action Research • A qualitative research approach • Methodology: (1) The rationale for choosing a research subject; (2) Defining a good research question for Action Research; (3) Research tools and data collection (questionnaires, interviews, and observations); (4) Data analysis. • Reflection at each stage • Portfolios, including all the above. The workshop coordinators discussed with the participants, the various stages of their classroom-based Action Research (see Figure 1).

2. Planning

3. Data Collecting and Analyzing

1. Identification of the problem and the research question

4. Implementing

6. Evaluating and Reflecting

5. Data Collecting and Analyzing

Figure 1: Various stages of Action Research B

The teachers as researchers: the teachers researched their own teaching in their classes where they worked on subjects that were interesting and relevant for them. After each workshop meeting, the participants met with their colleagues at school and shared with them the topics and subjects discussed in the workshop. In this way, they involved the whole team of chemistry teachers at their school in the “Action Research” process.

In this paper we will elaborate on two studies that will serve as examples of the program. In the first study, teachers investigated their students’ misconceptions about the electrical conductivity of metals and ionic materials. The second study focused on the behavior of non-science-oriented students and their attitudes toward chemistry studies.

146

CHEMISTRY TEACHERS RESEARCH THEIR OWN WORK 3. DESCRIPTION OF TWO CASE STUDIES – THE FIRST STUDY

The first study dealt with students’ understanding of electrical conductivity. We chose to present this particular study of two chemistry teachers, Sara and Debra, since the data they collected and their findings regarding learning about electrical conductivity stressed the fact that electrical conductivity is a problematic subject for high school students. Moreover, their study showed how the Action Research approach may help teachers to cope with the cognitive aspects of learning. Sarah and Debra were experienced chemistry teachers from a regional high school in a small town in the center of Israel. They taught 10th grade students who had finished junior high school studies at the same school. The team of chemistry teachers in their school consisted of 5 teachers, with Sarah serving as the chemistry coordinator. In the past, Sarah had participated in leadership courses as well as in long-term in-service professional development workshops. She was a member of several high level curricular committees, regarding high school chemistry studies. Both Sarah and Debra had been concerned for sometime about their students’ difficulties in understanding electrical conductivity and their students' misconceptions regarding metals and ionic solutions. There are several references to misconceptions about electrical conductivity and electrochemistry in the literature. Garnett & Treagust (1992) used semi-structured interviews to investigate students’ understanding of electrochemistry. Ozkaya (2002) referred to a previous study of prospective teachers which found that students from different countries and possessing different levels of knowledge had common misconceptions about electrochemistry. In examining senior secondary and tertiary level chemistry students, Coll & Taylor (2001) found 20 alternative conceptions of chemical bonding. During discussions on the electrical conductivity of copper, they found that students held alternative conceptions about electrical conductivity. Niaz (2002) reported on a research study in which a teaching strategy was utilized that might facilitate conceptual change in freshman students` understanding of electrochemistry. In the study it was found that providing students with the correct response along with alternative responses (teaching experiments) created a conflict situation that was conducive towards reaching an equilibration of students’ cognitive structures. It was concluded that the `teaching experiments` facilitated students` understanding of electrochemistry. Sarah and Debra tried to determine the reasons for students’ difficulties in understanding electrical conductivity. They realized that while planning their instruction, they had assumed that their students had previous knowledge about the subject, because those issues had already been taught in junior high school (7th–9th grade). The junior high school science curriculum includes a course dealing with electrical conductivity, namely “electricity and chemistry”. It consists of basic concepts of electricity and chemistry, e.g., electrical flow, electrical charge, electrical charges in a material, and the conductivity in an ionic solution and in metals. Both teachers decided to introduce a change in their teaching procedure to encourage a change in their students’ knowledge of electrical conductivity. They

R. MAMLOK-NAAMAN, O. NAVON, M. CARMELI, A. HOFSTEIN

147

decided to create some tools to investigate students’ previous knowledge; they consequently changed their teaching strategies according to their findings. The research questions selected for investigation were as follows: • •

What are the misconceptions of students who study the electrical conductivity of metals and ionic materials? How can teachers cope with their students’ misconceptions?

During the workshop meetings, teachers reflected on their work and received feedback from their colleagues. Research Tools Chosen by the Teachers The research tools chosen by Sara and Debra consisted of interviews with students, along with an achievement test. Sara and Debra developed the interviews as well as the achievement test; they also conducted the interviews and the achievement test. Interviews The interviews (before and after the teaching process) were semi-structured and consisted of specific questions aimed at determining what students understood in using a model of an electrical circuit. The interviewer was allowed to add or omit questions according to the interviewee's answers. Students were asked about two circuits – one consisting of a lamp and electrodes, and the other one consisting of beaker, a lamp and electrodes. In the first experiment, they put a piece of metal between the electrodes. In the second experiment, they poured on the electrodes an ionic solution. The following questions serve as examples of the questions used by the teachers: • I am closing the electrical circuit with a copper strip. Can you describe the molecular structure of the copper? • Where do the electrons come from? • Do you think that the battery stores the electrons? • Were there any electrons in the copper before I closed the circuit? • If I were a magician and I could paint the electrons of the copper in red, where could I find them? • If you could “accompany” the electrons that come out of the battery as you mentioned, what would happen to them? An Achievement Test The effect of the changes in teaching on students’ knowledge was evaluated by using a paper and pencil achievement test after the teaching process was completed. The test consisted of three different parts: in the first part students had to answer if the statement was true or false and to explain their answers. The following question serves as an example: • Ionic materials do not conduct in the solid phase, since in this phase there are no ions. True/False

148

CHEMISTRY TEACHERS RESEARCH THEIR OWN WORK

The second part consisted of open-ended questions. The last question dealt with the materials that students were asked about in the interviews, and they had to explain their answers. The question was as follows: “Copper and copper Chloride: 1. Are they solid at room temperature? 2. Can they conduct electricity at room temperature?” Explain! Results An analysis of the interviews revealed misconceptions that were typical of many students who were involved in this study. More specifically, students thought that: • The particles that are charge carriers are electrical particles, atoms, and protons. • There is a flow of energy. • The battery is the source of the electrons. Below are quotations of students’ answers to questions in the interviews: The electrons come from the battery. The battery stores the electrons. Yes! Protons and electrons as well. I do not know where I can find the electrons, I think that they stay in the metal itself. The electrons would pass through the tire, reach the lamp, continue through the electrode to the solution, pass through the solution and continue with the tire to the battery.

• • • •

The charge carriers are: Electrons, protons, copper, and chlorine (the components of copper chloride). The teachers concluded that students have difficulties with the following: Understanding the 'accumulation of atoms' concept. Integrating the macroscopic and microscopic worlds. The fact that the flow carriers in the ionic solution are not electrons. The fact that each compound that consists of a metal may conduct electricity under certain circumstances.

Based on the results of the interviews, the teachers decided to make some changes in their instructions. The experience gained in the professional development courses for chemistry teachers helped them understand the importance of using different and varied teaching approaches during the teaching process (Harrison & Treagust, 2000; Harrison & Treagust, 1996; Hoffman & Krajcik, 1999). Thus, they decided to integrate these ideas in their teaching:i • Models for demonstrating the particulate nature of matter. • Videos and slides, such as structured observations of the video “the metals”.

R. MAMLOK-NAAMAN, O. NAVON, M. CARMELI, A. HOFSTEIN •

•

149

Educational computer programs describing the structure, bonding, and properties of compounds, developed by the Chemistry Group, Science Teaching Department, The Weizmann Institute of Science, Rehovot, Israel.(http://stwww.weizmann.ac.il/g-chem/.htm) Computer animations.

After completing the instructional unit, the teachers interviewed the same students and administered an achievement test to the whole class. Based on the achievement test results and the analysis of the repeated interviews, the teachers reported that: • Most of the students were able to distinguish between the electrical conductivity of metals and the electrical conductivity of ionic solutions and between the characteristics of copper as a metal and copper chloride as an ionic solution. • The students learned how to explain the relationship between the macroscopic and microscopic worlds. For example, the teacher asked students to draw the model of the microscopic structure of copper and of the ionic solution of copper chloride. One of the students drew the models as presented in Figure 2 and in Figure 3. Figure 2 presents a model of the microscopic structure of copper; Figure 3, a model of particles in the ionic solution of copper chloride.

Figure 2: A model of the microscopic structure of copper, drawn by a student

Figure 3: A model of the particles in the ionic solution of copper chloride, drawn by a student However, both Sarah and Debra mentioned that there were still a small number of students who were “stuck” with the notion that electrons are the only pulse carriers. Sarah, the senior teacher, claimed that she should investigate further her students’ knowledge and understanding in each of the topics that she taught although

150

CHEMISTRY TEACHERS RESEARCH THEIR OWN WORK

this would necessitate a change in her lesson plans and in the pace of teaching a certain subject 4. DESCRIPTION OF TWO CASE STUDIES – THE SECOND STUDY The second study dealt with the behavior of 10th grade non-science-oriented students and their attitudes toward chemistry studies in a high school in the central part of Israel. The attitude of the students influenced their behavior in class and their achievements in chemistry. Linda, Orna, and Eva, all experienced chemistry teachers, (see Table 1), taught chemistry for 10th grade students both in the science and non-science-oriented classes. They faced problems in the non-science-oriented classes, discussed it in team meetings, decided to try to solve these problems and to reflect on their work through an Action Research process. In one non-science-oriented class consisting of thirty-eight students, the traditional curriculum became irrelevant because it included very few issues that were meaningful to the students. The teachers made great efforts to simplify it and to attract students, but they still had severe discipline problems, and students' achievements remained low. Then teachers decided to make two main changes: (1) to divide one of the classes into two groups (the principal approved trying this approach in one class); (2) to teach the subject matter according to an STS approach. The STS approach is aimed at non-science-oriented students, namely students who do not choose to major in any scientific discipline (Bybee & Trowbridge, 1996). It is an effort to produce an informed citizenry capable of making crucial decisions about current problems and issues, and as a result, taking personal actions. Moreover, STS provides the teacher with a wide range of teaching techniques, enabling diversifying the classroom learning environment. Consequently, the student’s motivation to learn science increases and creativity is enhanced (Tobin, Tippins & Gallard, 1994; Hofstein & Walberg, 1995; Hofstein, Mamlok and Carmeli, 1997; Bodzin & Mamlok, 2000). As mentioned above, the teachers divided one of the classes into two groups and used the STS approach in their chemistry lessons. They used the process of Action Research to investigate the change that occurred in this class. For their Action Research investigation they selected the following research question: • Will the interest and motivation of non-science-oriented students increase due to changes in the chemistry curriculum? Research Tools The research tools that the teachers used consisted of classroom observations and student interviews. Observations Out of the three chemistry teachers, two teachers taught in class and the third observed five sessions of each class. The focus of the observations was on students’ behavior concerning their: (1) active participation in the classroom and their ability to express their ideas, (2) active listening, (3) way of preparing homework, and (4)

R. MAMLOK-NAAMAN, O. NAVON, M. CARMELI, A. HOFSTEIN

151

motivation to write notes in their journals. In addition, the observer took notes of special events taking place in the classroom. Interviews The teachers interviewed seven students who represented three different kinds of student levels low achievers, intermediate achievers, and high achievers. The short interviews, conducted toward the end of the year, consisted of the three questions below. Note that each question is followed by an example of a student answer. The interviews were audio-recorded and the content analyzed. (T = Teacher, S = Student): i) T: How do you feel about the changes made in our class? S: I think that the changes improved the atmosphere in class; it becomes more pleasant for us and also for the teacher. We better understand the material and the teacher is less busy. ii) T: What was special in the chemistry studies this year? S: We learned chemistry in a different way. The content was new for us, and there were new experiments. I even made a small project by myself. iii) T: Were your learning habits influenced by the changes made by your teacher? S: For me personally the change was meaningful, I am not now falling asleep during the lessons, I can draw conclusions while performing experiments, and I can listen better to the teacher during lessons. Data analysis was conducted during workshop meetings by participants with guidance of the workshop coordinators, using various qualitative research techniques.

Results Based on observations, the teachers concluded that there was a meaningful change in students’ behavior in the chemistry classroom and in their attitudes toward science, more specifically: • The class was active and about 60% of the students took notes in their journals and did not hurry to leave the classroom when the lesson was over. • Students’ interest in the experiments and laboratory demonstrations increased. • Students started to ask more questions. Students’ interviews conducted by the teachers validated the observed findings. Some students claimed that in the past they had never experienced such a serious attitude toward them on the part of the teachers. A typical remarks was: “Are we so important to you? You’re devoting so much time to us.”

152

CHEMISTRY TEACHERS RESEARCH THEIR OWN WORK

It was also found that students performed experiments at home, and also looked for enrichment materials. They said that their learning habits had changed, from studying only just before an examination to continuous study. 5. CONCLUSIONS At the completion of the program, the workshop coordinators asked the 10 participating high school chemistry teachers to reflect on their experience. The teachers reflected on both workshop meetings and the Action Research they had performed in their classes. Most teacher participants were aware of the support they had received from the workshop coordinators and their colleagues. They claimed that the workshop contributed to their self-image and to their instruction in class. As an example of their professionalism, some participants mentioned that they had presented their studies at the annual conference of chemistry teachers in Israel. The analysis of the interviews revealed that through Action Research teachers experienced a new dimension of professional development. Regarding the contribution of Action Research, three main areas emerged: (1) implementation of change through Action Research, (2) having a sense of being part of a professional community, and (3) having contacts with academic experts. (1) Implementation of change through Action Research: I certainly think that I will use Action Research tools in my teaching. This year I taught chemistry by the inquiry approach and the students were already used to the group that they were in last year. Moreover, I assessed the change in my teaching strategy, using the tools that I had experienced within the Action Research workshop. (2) Having a sense of being part of a professional community: During the workshop meetings we consulted each other; we maintained contact through email and exchanged information and ideas. For instance, the discussions in the workshop helped me define my Action Research question. I presented my project in front of the group and I received meaningful feedback. (3) Having contacts with academic experts: During our Action Research workshop, we established closer contact with the academic staff on a professional basis. I felt that that we can contact the experts and consult about our problems whenever we need to. The results of our study showed that teachers had undergone a new process in their professional development. They got new insights regarding their teaching and were able to improve and promote their classroom instruction. It strengthened their teamwork at school and encouraged collaboration between themselves and their colleagues. They supported each other in their work at school in general, as well as in their Action Research study. All the teachers were enthusiastic about the fruitful discussions during the workshop, and the lectures that they had attended. The

R. MAMLOK-NAAMAN, O. NAVON, M. CARMELI, A. HOFSTEIN

153

teachers stressed the contribution of the workshop to their work in class and the importance of the support they received during the workshop. The lectures that these teachers attended at the workshop enabled them to undergo a conceptual change and to realize that a reflective study has its own value and is indeed beneficial to their work (Elkis & Ralle, 2002). As Joyce and Showers (1983) suggested, teachers are interested in improving and enriching their teaching methods, and Action Research, in our case, has been a new experience for those teachers who participated in the workshop. Note that teachers do not usually take part in qualitative reflective research. They are, if at all, involved in positivistic studies in which they serve as experimental or control groups in an external study conducted by science education experts. Thus, teachers who participate in an Action Research project have to be highly motivated and dedicated to this kind of work which is both demanding and time consuming. To sum up, the participating teachers were more satisfied with their teaching, and had closer contact with academic institutions on a professional basis. In addition, they became more concerned about improving their teaching, and they learned how to share their ideas and experiences with their colleagues. From our experience with this program, Action Research is a powerful tool for improving the professional development of teachers. ENDNOTE For more details see Mamlok-Naaman, Navon, Carmeli & Hofstein, (2004). Teachers research their students’ understanding of electrical conductivity, The Australian Journal of Education in Chemistry, The Royal Australian Chemical Institute, Chemical Education Division, in press. REFERENCES Bodzin, A.M., & Mamlok, R. (2000). STS Simulations engaging students with issues-based scenarios. The Science Teacher, 8(1), 36-39. Bybee, R.W. & Trowbridge, L.W. (1996). Teaching Secondary School Science: Strategies for Developing Scientific Literacy. Englewood Cliffs, NJ: Prentice-Hall. Coll, R.K., & Taylor, T. (2001). Alternative conceptions of chemical bonding held by upper secondary and tertiary students. Research in Science & Technological Education, 19(2), 171-191. Elkis, I., & Ralle, B. (˫2002). Participatory Action Research within chemical education. Proceeding of the 16th Symposium on Chemical Education held at the University of Dortmund, 22-24, May 2002, Shaker Verlag, Aachen. Feldman, A. (1996). Enhancing the practice of physics teachers: Mechanisms for generation and sharing knowledge and understanding in collaborative Action Research. Journal of Research in Science Teaching, 33, 513-540. Feldman, A., & Minstrel, J. (2000). Action Research as a research methodology for study of teaching and learning science. In A. E. Kelly & R. A. Lesh (Eds.), Handbook of Research Design in Mathematics and Science Education (pp. 429-455).

154

CHEMISTRY TEACHERS RESEARCH THEIR OWN WORK

Garnett, P.J., & Treagust, D.F. (1992). Conceptual difficulties experienced by senior high school students of electrochemistry: Electrochemical (galvanic) and electrolytic cells. Journal of Research in Science Teaching, 29(2), 1079-1099. Gipe, J.P., & Richards, J.C. (1992). Reflective thinking and growth in novices’ teaching abilities. Journal of Educational Research, 86(2), 52-57. Gore, J.M., & Zeichner, K.M. (1991). Action Research and reflective teaching in preservice teacher education: A case study from the United States. Teaching & Teacher Education, 7(2), 119-136. Harrison, A.G., & Treagust, D.F. (2000). Learning about atoms, molecules, and chemical bonds: A case study of multiple model use in grade 11 chemistry. Science Education, 84, 352-381. Harrison, A.G., & Treagust, D.F. (1996). Secondary students’ mental models of atoms and molecules: Implications for teaching science. Science Education, 80, 509-534. Hoffman, J.L., & Krajcik, J.S. (1999). Assessing the nature of learning science content understandings as a result of utilizing on–line resources. Paper Presented at the meeting of the National Association for Research in Science Teaching, Boston, MA, U.S.A. Hofstein, A. (2001, April). Action Research: Involving Classroom-Related Studies and Professional Development. Paper presented at IOSTE conference, Paralimni, Cyprus. Hofstein, A. & Walberg, H.J. (1995). Instructional Strategies. In: B.J. Fraser & H.J. Walberg (Eds.). Improving Science Education (pp. 70-89). The National Society for the Study of Education. Hofstein, A., Mamlok, R., & Carmeli, M. (1997). Science teachers as curriculum developers of science and technology for all. Science Education International, 8(2), 26-36. Holly, P. (1991). Action research: The missing linking the creation of schools as centers of inquiry. In A. Liberman & L. Millaer (Eds.). Staff Development for Education in the 90’s: New demands, New Realities, New perspectives. (pp.133-157). New York: Teachers College Press. Kemmis, S., & Mctaggart, R. (1988). The Action Research Planner. (Eds.) Geelong, Victoria, BC, Canada: Deakin University Press. Korthagen, F.A.J. (1985). Reflective teaching and preservice teacher education in the Netherlands. Journal of Teacher Education, 36(5), 11-15. Lewis, I., & Munn, P. (1987). So you want to do research! A guide for teachers on how to formulate research questions. Edinburgh: Scottish Council for Research in Education. Loucks-Horsley, S., Hewson, P. W., Love, N., & Stiles, K. E. (1998). Designing Professional Development for Teachers of Science and Mathematics. Thousand Oaks, CA: Corwin Press. Mamlok-Naaman, R., Navon, O., Carmeli, M., & Hofstein, A. (2003). Teachers research their students’ understanding of electrical conductivity, The Australian Journal of Education in Chemistry, The Royal Australian Chemical Institute, Chemical Education Division, in press National Research Council. (1996). National Science Education Standards. Washington, D.C.: National Academy Press. Niaz, M. (2002). Facilitating conceptual change in students’ understanding of electrochemistry. International Journal of Science Education, 24(4), 425-439. Obaya, O. (2003) Action Research: Creating a context for science teaching and learning. Science Education International, 14(1), 37-47. O’Hanlon, C. (1996). Professional Development through Action Research in Educational Settings’. (Ed.). Washington, DC: Falmer.

R. MAMLOK-NAAMAN, O. NAVON, M. CARMELI, A. HOFSTEIN

155

Ozkaya, A.R. (2002). Conceptual difficulties experienced by prospective teachers in electrochemistry: Half-cell potential, cell potential, and chemical and electrochemical equilibrium in galvantic cells. Journal of Chemical Education, 79(6), 735-738. Schon, D.A. (1983). The Reflective Practitioner. New York: Basic Books, Inc. Tobin, K., Tippins, D.J. & Gallard, A.J. (1994). Research on instructional strategies for teaching science. In: D.L. (Ed.). Handbook of Research on Science Teaching and Learning (pp. 62-63). New York: Macmillan Publishing Company. Van Zee, E. H. (1998) Preparing teachers as researchers in courses on methods of teaching science. Journal of Research in Science Teaching, 35(7), 791-809. Zuber-Skerritt, O.(1996). New Directions in Action Research. London: Falmer.

THE RELATIONSHIPS BETWEEN PRIMARY TEACHERS’ ATTITUDES AND COGNITION DURING A TWO YEAR SCIENCE IN-SERVICE PROGRAMME

TINA JARVIS, ANTHONY PELL University of Leicester, UK

ABSTRACT Teachers’ confidence and attitudes towards science teaching and science understanding were tested before and after a major in-service programme in 31 schools. The 70 teachers' attitudes were assessed using a 49-item Likert-scale test. Science understanding was measured by multi-choice and open-ended questions. Data on pupils’ attitudes and cognition was also collected. After in-service, overall teachers’ initial confidence about science teaching had improved significantly. The majority of teachers, but not all, had developed satisfactory levels of understanding and more positive attitudes. Teachers responded to the in-service programme in different ways. Four teacher types were identified: high attainers who improved attitudes and confidence; teachers with limited science knowledge who found the course difficult but made improvements; unaffected professionals who were already working well and for whom the course had little effect; and disaffected teachers who showed low levels of confidence and competence throughout. Pupil cognition and attitudinal differences related to these types were found.

1. INTRODUCTION In many countries primary teachers’ background knowledge in science is very variable with the effect that they lack confidence and competence in teaching science (Goodrum, Hackling & Rennie, 2001). Science knowledge is a significant factor that influences primary teachers’ confidence in teaching science (Harlen & Holroyd, 1997). Teachers with low confidence cope by teaching only the minimum required; stressing aspects they do feel more confident in, such as biology; using prescriptive texts and work cards; underplaying questioning and discussion; and only doing very simple practical work. When these coping strategies become the norm, pupils’ attainment will be limited (Osborne & Simon, 1996). It is also likely that pupils’ attitudes will be detrimentally effected. She & Fisher (2002) found that pupils’ attitudes towards science were influenced by teachers’ behaviour in the classroom, such as asking challenging questions, encouraging, and praising. In addition, they found lower secondary pupils’ higher attitudinal scores were associated with higher cognitive scores. While Simpson & Oliver (1990) also found a strong pupil attitude-achievement relationship in their longitudinal study of pupils grade 6-10, they did not find a 157 K. Boersma et al. (eds.), Research and the Quality of Science Education, 157—168. © 2005 Springer. Printed in the Netherlands.

158

PRIMARY TEACHERS’ATTITUDES AND COGNITION

relationship between teacher affect and student affect. However, Gallagher (1994) found that pupils’ perception that a teacher finds the subject matter interesting may enhance pupils’ interest. If there is a close relationship between teacher knowledge and attitudes which in turn effects pupils’ understanding and attitudes, in-service to improve teachers’ cognition should influence the other factors. This research set out to explore the changing relationship between teachers’ attitudes and cognition with those of their pupils over a two year in-service programme. The research questions to be addressed were: • What are the attitudinal and cognitive profiles of teachers selected for the inservice course? • Do the attitudinal and cognitive profiles change after the in-service course? • Do all teachers respond to the in-service course in the same way? • Do changes in pupils’ attitudes and attainment show any correspondence to changes in teachers? 2. SETTING AND PARTICIPANTS Thirty one inner-city schools took part in a 6 month in-service course focusing on Developing and Assessing Investigations. The majority of the schools were considered to have weaknesses in science, as shown by national science tests and/or inspection reports. Thirty-nine teachers from sixteen schools took a 10-day course between January and July 1999, with an additional thirty-one teachers from all the schools taking a similar course between January to July 2000. Progress of the teachers and their 1878 pupils was monitored. 3. IN-SERVICE CONTENT Virtually all the schools had identified the development of classroom investigations as a problem area. School inspections indicated that pupils needed to be more independent to set up their own investigations, as well as to be enabled to explain their findings. Head teachers, co-ordinators, and local authority education advisors considered that teachers were reluctant to provide open-ended investigations because they lacked confidence and knowledge in science. The 10-day course (spread over 6 months) on Developing and Assessing Investigations was designed to address these concerns. It covered strategies to develop open-ended investigations in the areas of electricity; melting, evaporation and dissolving, and friction. These topics were chosen as they are particularly difficult for primary teachers (Kruger & Summers, 1989; Kruger et al., 1990; Webb, 1992). A constructivist approach was taken in which teachers were helped to assess their own knowledge. Then, experiences were provided to help them accept the current scientific view though active, collaborative learning using an approach outlined by Summers (1992). Tutor visits supported classroom follow-up activities.

T. JARVIS, A. PELL

159

4. THE TEACHERS’ INSTRUMENTS Teachers were asked to complete confidence, attitude, and science understanding tests at the beginning of the course. The same tests were repeated in June 2000, 12 months after the end of the first course and at the end of the second course. The design of these instruments was influenced by the need to match them to the pupils’ instruments so that the relationship between teachers' and pupils' attitudes and cognition could be examined. The confidence and attitude questionnaire had 4 parts. 1. Personal information was collected to explore influences related to gender, experience, and subject responsibility. 2. Confidence scales about teaching science as well as English, mathematics, and information technology covered the class currently being taught, and with respect to early years and older primary children (adapted from Hargreaves et al., 1996). 3. Confidence scales focused on delivering different aspects of the Primary Science National Curriculum required in English schools followed. This five-point scale measured confidence towards different science content, investigations and pedagogy and was also developed from Hargreaves et al. (1996). 4. A science attitude scale of 49 items with 0.96 reliability (Pell & Jarvis, 2003) took up the remainder of the questionnaire. This attitude scale has six sub-scales. Three explore teachers’ views about effective science teaching in the classroom: • Investigative, pupil-centred science scale probes the value teachers put on encouraging pupil initiative, interest and wonder (reliability 0.89); • Classroom management science scale focuses on the value given to systematic, structured approaches to the learning (reliability 0.83); and • General scientific method scale centres on views of empirical, pupilparticipative science (reliability 0.83). A further three sub-scales probe teachers’ views of preparation and professional aims: • In-service improvement scale reflects teachers' attitudes towards the worth of inservice education with an emphasis on the use of human and physical resources (reliability 0.86); • Theoretically-grounded science teaching scale indicates the extent to which teachers feel they should operate from a child-centred, constructivist process (reliability 0.86); and • Testing which focuses on attitudes to formative assessment (reliability 0.65). The cognitive test investigated elements of primary teachers' subject knowledge and process skills required in the primary pupils' Science National Curriculum. The test included four extended questions on electricity; melting, dissolving and evaporation, forces, and investigations. Substantial intercorrelation between three sub-scales of Understanding Electricity, Understanding Change of State, Understanding Forces, and one item on investigations gave an overall measure of

160

PRIMARY TEACHERS’ATTITUDES AND COGNITION

attainment which had a reliability of 0.63 before in-service and 0.76 after in-service (Jarvis, Pell & McKeon, 2003). 5. ANALYSIS The research design allowed for paired pre- and post-test measures on the affective and cognitive instruments. This also matched the design adopted for the associated pupil research study. Despite the fact that the first cohort completed their tests a year after the end of their main in-service programme, the overall attitude and cognition profiles of the two cohorts were very similar (Jarvis, Pell & McKeon 2003). Consequently, for the focus of this paper, the scores were pooled to allow for straightforward pre-test/post-test analyses. In the analysis stage, appropriate parametric and non-parametric statistics were employed to identify significant changes and effect sizes. Residual gain analyses for both attitude and cognitive tests identified individual teacher outcomes. Similarities between teachers were explored by cluster analysis. Pupil results for each cluster were then examined with respect to the changes shown by their teachers. 6. RESULTS Attitudinal and Cognitive Profiles of Teachers Before and after In-Service Confidence was assessed in the first part of the questionnaire. Teachers were asked to rate their confidence, on a five-point scale, in teaching English, mathematics, science, and information technology. On the pre-test, project teachers were more uncertain when teaching both science and information technology than when teaching English. This is consistent with earlier work (Wragg et al., 1989). On the post-test, confidence in teaching science in their own classes improved significantly (with a large effect size equivalent to a standard deviation) and no longer differed from their confidence in teaching English. Self-rated competence in teaching the National Primary Science Curriculum section asked teachers to rate their competence in teaching about life processes, materials, and physical processes. They were also asked to rate their competence in using different pedagogical strategies such as planning investigations and promoting questioning. A factor analysis of the competence items showed the presence of one dominating factor on the pre- and post-tests, accounting for 55% and 63%, respectively, of the variance of all the items. This ‘competence in science rating’ scale uses all the items to give a reliability of 0.95. Before in-service, there was only a significant difference in teachers' confidence across the three content areas (Friedman statistic χ2=37.2, df=2, p<0.0001). As many researchers have found, teachers were most confident when teaching about life processes (Council for Science & Technology, 2000). The project teachers also lacked confidence in guiding children's discoveries in investigations and in eliciting

T. JARVIS, A. PELL

161

pupils' ideas. They showed the least confidence in planning a course of lessons according to the required National Curriculum criteria. After in-service, the response to every item on the Competence in Teaching National Curriculum Science scale was significantly more positive showing an overall increase from a mean of 3.12 to 3.84 (Wilcoxon paired test, p<1%). However, the pattern of greater confidence in biological processes and lower confidence in guided discovery, pupil questioning, and long-term planning remained. Attitudes towards science teaching and in-service showed small increases in mean scores/item in five of the six sub-scales after in-service. The 'Testing' subscale showed a decrease. The subset of 38 teachers who returned data for both tests showed highly significant gains (p=0.002) on the rating of 'classroom management science teaching' from a pre-test score of 3.38 to post-test score of 3.60. This was particularly helpful because the teachers needed to appreciate that potentially disruptive children could be managed during practical investigations. The decline in attitudes towards formative testing was of concern, as the teachers may not have appreciated the importance of formative assessment as an element of the constructivist approach. Therefore, the value of assessment needed to be more overt. Cognitive scores after in-service showed significant overall gains in understanding of investigations and concepts covered in the in-service (Table 1). However, different teachers’ gains varied. As Summers & Kruger (1994) found, while the majority of teachers had developed scientifically acceptable ideas, others made progress but still had misconceptions. In addition, some teachers did not develop a good understanding of variables and their control. However, these teachers were not aware of these problems and felt confident and enthusiastic about teaching science. Despite the fact that this was a substantial in-service programme, there is a need to provide further long-term follow up support in at least some cases as suggested by Joyce & Showers (1988). While no significant inter-correlation between overall gains in cognitive and attitude scores were found, cognition and attitude to different aspects of science education appeared to be related, alongside other factors. For example, the majority of the 33 teachers with complete data sets, improved cognitive scores significantly alongside improved attitudes towards science classroom management Differences between teachers showed great variation, while the overall scores for the attitude scales before and after in-service showed limited changes. The composite post-test scores of all the attitude scales correlate with pre-test attitude scores at 0.42 (p<5%) and give a regression equation for the post-test score of: Post-test attitude score = 0.46 Pre-test score + 102.5.

162

PRIMARY TEACHERS’ATTITUDES AND COGNITION

Table 1. In-service teachers’ cognitive scores on pre- and post-tests Pre-test Post-test N Mean Std. dev. N Mean Std. dev. Cognitive sub-scales Electricity 68 5.40 3.24 46 10.20** 3.82 Change of State 68 8.98 3.07 46 11.46** 3.29 Forces 68 16.21 6.83 46 21.15** 5.39 Investigations item

68

1.65

1.43

46

2.35 *

1.43

68 32.22 11.44 46 45.15** 11.40 Overall gains in science understanding ** p<1%, *p<5%, sig. improvement, paired t-test/Wilcoxon test for N=46 on both tests Differences Between Teachers and Their Effect on Pupils When this equation is used for every teacher's pre-test score, a generalised score is predicted from the effect of in-service. This residual gain varied among teachers from –39.6 to + 25.7. The correlation between pre- and post-test cognitive scores is 0.43 (p<1%) and the regression equation for the gain in cognitive scores is: Post-test cognitive score = 0.41 Pre-test score + 31.4 Individual teacher’s cognitive residual gains varied from –21.5 to +26.8. In order to explore the reasons behind this range of residual scores, factors such as gender, administrative responsibility, and length of teaching experience were examined. Before in-service, female teachers were more positive than male teachers about teaching investigative, pupil-centred science, but this difference disappeared after in-service. While both men and women showed significant increases in cognitive scores, female teachers’ self-perception of their competence in teaching science increased significantly. Level of qualifications, experience, and responsibility also had an effect. For example, teachers without responsibility showed the greatest post in-service cognitive scores. This may be because these teachers were more likely to be recently qualified teachers with a better science grounding, and/or because they had more time to follow up work from the course in their classrooms. Some of these factors may also interrelate with gender differences as this population reflected national patterns where comparatively few women hold posts of responsibility and far more have over 15 years experience (Department for Education & Skills, 2002). It was clear that no one factor explained the differences in teachers’ responses to the in-service programme. Similarities and differences were then explored by cluster analysis using the pre-test and post-test cognitive scores, the overall attitude to effective science teaching scores, the measures of confidence in teaching science, and competence in putting National Curriculum science into practice. Only 33 of the

T. JARVIS, A. PELL

163

teachers supplied a full set of scores for this analysis. Characteristics of each cluster type were established by looking for significant differences between mean cluster scores and the rest of the sample. Tutors’ reports on progress during the course and observations in school were also taken into account. Table 2 summarises the four broad types of teacher that emerged. (One teacher could not be classified by this grouping.) The cluster mean scores grouped into three categories: high, average, and low. Generally, high and low classifications are at the 5% significance level at least. Table 2: Classification of teacher types according score range (high, average, or low) Disaffected Limited (N=5) Cognitive Development (N=13)

Enthusiastic (N=5)

Unaffected Professionals (N=9)

Cognitive scores Pre-test Post-test

Low Average

Low Low but improved

High§ High§ but improved

High High§ slight improvement

Overall science teaching attitudes Pre-test Post-test

Average Low

Average Average

Average High

Average Average

Low Low but improved

Average Average but improved

Average High

High Average

Low Low but improved

Average Average but improved

Average High

High Average

Competence in NC science teaching Pre-test Post-test Confidence in teaching science Pre-test Post-test

High and Low cells are at 5% significance at least with the exception of cells marked § Relationship between pupils’ performance and teacher type was investigated. The pupils’ attitude test had five scales and covered children’s attitudes to school, science at home and school, as well as science's value to society (Pell & Jarvis, 2001). Pupils also had a cognitive test that covered the same conceptual ideas as their teachers. (Copies of both instruments are available www.azteachscience.co.uk ) Analysis of the pupil attitudinal data shows that before in-service the pupils had a general positive disposition towards going to school and a positive view of the value of science in society. In contrast, as boys and girls got older their enthusiasm

164

PRIMARY TEACHERS’ATTITUDES AND COGNITION

for school science fell with the decline being especially noticeable for girls. Related to this was an age-related science 'difficulty' effect indicating that the pupils, girls in particular, felt science got easier. The science in-service appeared to have influenced whole school attitudes, not just attitudes to science. The pupils' attitudes to school, in classes of teachers attending the in-service, did not deteriorate as the academic year progressed, unlike those in three control schools. There was an indication that the pupils enjoyed independent practical work more. The in-service also appeared to break the relationship between loss of enthusiasm for science as it was seen as easy, hopefully reflecting the fact that the teachers were able to give challenges to the children which captured their interest (Jarvis & Pell, 2002). The pupils’ overall mean cognitive score showed an increase after in-service. While there was an expected increase over the school year, the project pupils’ overall understanding in science showed highly significant increases which were not matched to the same degree by a control sample of 77 pupils. However, the overall positive cognitive and attitudinal effects mask the fact that class results varied greatly. Before in-service there was evidence that pupils of teachers with better subject knowledge had higher cognitive scores. These pupils also had higher attitude scores, apparently confirming a relationship between teachers’ good subject knowledge and better pupil attitudes that has been suggested by other researchers (Osborne & Simon, 1996). However, this relationship was not so clear after in-service. The response of the classes’ varied for the different teacher types. A direct comparison between teacher types and pupil cognitive data was not appropriate. This is partly because year groups for each teacher type varied considerably with ‘Disaffected’ teachers only teaching the youngest classes and ‘Enthusiastically Fired’ teachers only teaching Years 4 and 5. Pupils’ cognitive scores were also strongly dependent upon year group. In addition, there was the complication that young pupils had a simpler version of the cognitive test. Consequently, the most valuable measure was a residual gain score controlling for the effect of year group. The regression equation is: Post-test pupil cognitive score =14.19 + 0.57 Pre-test cognitive score +9.84 Y6 + 2.18 Y 5 Membership of Years 5 or 6 is recognised by a corresponding score of '1' in the equation. There was no distinction between Years 3 and 4, nor were there any gender effects. It was decided to use the Year 3-6 test data only, although this meant that only one ‘Disaffected’ teacher would be represented in the comparison. The examination of these scores showed that while all mean cognitive scores showed an increase, not all residual gain scores for the four teacher types were positive. Disaffected teachers made small gains from a low base level of cognition, competence, and confidence. However, attitudes to investigative science- and theory-grounded education dropped. This was a small group of Year 2 and 3 classes. (All other teachers taught Year 3 and above.) Over the period of the in-service these pupils’ attitudes dropped, but there was attainment gain in the one Year 3 class. This

T. JARVIS, A. PELL

165

group of teachers had poor attendance and did limited follow-up classroom work. In two out of five cases, absence was due to medical reasons. Absence of the remaining teachers could be lack of agreement with the aims of the in-service, as indicated by the drop in attitudes to investigative science, or poor school support to give teachers time for classroom-based work. On the other hand, they may not have been motivated to attend because they did not appreciate the value of developing correct science ideas with very young children. Watts & Walsh (1997) suggest there is a particular pedagogical cultural clash for Early Years teachers who focus on a childcentred integrated topic approach, which they feel is in conflict with whole class subject-specific strategies that are typified by science. Teachers with Limited Cognitive Development started from a low cognitive base and made significant improvements. Their attitudes were stable and slightly below average. Confidence and self-rated competence increased significantly but remained at the average level of the whole cohort. These teachers were keen but appeared to find that the day-to-day pressures of school demands limited their follow-up work. Their pupils’ attitudes improved from a low base, which could be due to the teachers’ increased self-confidence. However, the pupils did not gain cognitively as much as other pupils, which may reflect limited follow-up classroom work. These teachers probably needed continued long-term classroom support from tutors, as well as further opportunities to take time out of the classroom to address continuing science misconceptions. Enthusiastically Fired teachers started with average or above baseline cognitive scores and showed increases in all affective scores. In-service appeared to add to an already high cognitive base as well as firing their enthusiasm. Pupils showed little change in attitudes and cognition related to the in-service. This may have been because the general level of teaching was good before the in-service and because all but one did follow-up work that focused on their schools rather than their classrooms. The in-service content and length was probably appropriate for the personal development of these teachers and their schools. Unaffected Professionals also started with high cognitive levels which improved slightly while their attitudes remained steady. Their high relative confidence levels enjoyed at pre-test disappeared after in-service, while all other groups’ scores rose. Though their pupils showed little attitudinal change, they did show a significant cognitive gain, unlike the pupils of their Enthusiastically Fired colleagues. One explanation could be that some of these teachers were teaching Year 6 classes and, as is common (OFSTED, 1999), worked intensively with them in preparation for the science National Standard Assessment Tasks. 7. CONCLUSION The implications of the research reported here are dependent upon the reliability and validity of the measures. Intensive piloting of the attitudinal instruments and the judicious use of item selection for the cognitive measures resulted in a range of

166

PRIMARY TEACHERS’ATTITUDES AND COGNITION

reliabilities from 0.75 to 0.96. The weakest measure, which was not used in the cluster analysis, dealt with attitudes to testing. An overall measure of validation of the instruments is given by only a 3% mis-classification by the statistical cluster software. Pre-test/post-test designs do suffer from attrition in sample numbers, but this was partly unavoidable due to movement in the teacher population. In-service had a mainly positive effect on teachers’ attitudes and cognition, but not always. The amount and type of effect appeared to depend on the interplay of teachers’ personal factors. In turn, pupils’ attitudes to science and cognition were influenced by their teachers and the changes brought about in their teachers’ through in-service. It appears that the common ‘one course fits all’ approach is unlikely to be as effective or as economic as focused differentiated courses. This general finding for these inner-city primary schools is likely to be as applicable in secondary schools in the UK and for in-service in other countries. In England the new National and Regional Science Learning Centres provide an opportunity to provide and research a variety of targeted, differentiated in-service. This research found two main types of teacher with regard to in-service requirements, depending on their initial level of cognition and confidence. While many teachers with a high cognitive base were relatively unaffected by the inservice, some were fired with confidence and enthusiasm. Such enthusiasm is important in order to retain able teachers in the teaching profession, an issue which is of particularly acute concern in schools, like the ones in this programme, that serve pupils from disadvantaged backgrounds. The in-service content and length appeared appropriate for this group of teachers, but the addition of other innovative pedagogical strategies might have enthused more of the Unaffected Professionals. Teachers with initial low attainment and confidence need very substantial long-term in-service with school support for classroom-based work. This will need to be longer than the 6 months in-service of this programme, in order to address persistent science misconceptions and to help teachers apply their new found ideas in the classroom so that positive pupil attitudes and cognition can be fostered. The inservice was not so successful with all low attaining teachers who also had a very low confidence base. The issue of their self-confidence needs addressing. It may also help if these Early Years teachers have support for seeing the value of science for early years’ children and help for presenting it effectively in a traditional Early Years integrated child-centred approach. ACKNOWLEDGEMENTS This major in-service project was funded by the AstraZeneca Science Teaching Trust.

T. JARVIS, A. PELL

167

REFERENCES Council for Science and Technology (CST) (2000). A Report on Supporting and Developing the Profession of Science Teaching in Primary and Secondary Schools. London: CST. Department for Education & Skills (DfES) (2002). Statistics of Education: School Workforce in England. London: HMSO. Gallagher S. A. (1994). Middle school classroom predictors of science persistence. Journal of Research in Science Teaching, 31(7), 721-734. Goodrum, D., Hackling, M. & Rennie, L. (2001). The Status and Quality of Teaching and Learning of Science in Australian Schools. Canberra: Department of Education, Training and Youth Affairs. Hargreaves, L., Comber, C. & Galton, M. W. (1996). The National Curriculum: can small schools deliver? Confidence and competence levels of teachers in small rural primary schools. British Educational Research Journal, 22, 89-99. Harlen, W. & Holroyd, C. (1997). Primary teachers' understanding of concepts of science: Impact on confidence and teaching. International Journal of Science Education, 19(1), 93-105. Jarvis, T. & Pell, A. (2002). Changes in primary boys’ and girls’ attitudes to school and science during a two-year science in-service programme. The Curriculum Journal, 13(1), 43-69. Jarvis, T., Pell, A. & McKeon, F. (2003). Changes in primary teachers’ science knowledge and understanding during a two year in-service programme. Research in Science and Technological Education, 21(1), 17-42. Joyce, B. & Showers, B. (1988). Student Achievement through Staff Development. New York: Longman. Kruger, C. & Summers, M. (1989). An investigation of some primary teachers' understandings of changes in materials. School Science Review, 71, 17-27. Kruger, C., Summers, M. & Palacio, D. (1990). An investigation of some English primary school teachers' understanding of the concepts force and gravity. British Educational Research Journal, 16(4), 383-397. OFSTED (Office for Standards in Education) (1999). Primary Education 1994-98: A Review of Primary Schools in England. London: OFSTED. Osborne, J. & Simon, S. (1996). Primary science: past and future directions. Studies in Science Education, 26, 99-147. Pell, T. & Jarvis, T. (2001). Developing attitude to science scales for use with children of ages from five to eleven years. International Journal of Science Education, 23(8), 847-862. Pell, A. & Jarvis, T. (2003). Developing attitude to science education scales for use with primary teachers. International Journal of Science 25(10), 1273-1295. She, H. & Fisher, D. (2002). Teacher communication behavior and its association with students’ cognitive and attitudinal outcomes in science in Taiwan. Journal of Research in Science Teaching, 39(1), 63-78. Simpson, R. D. & Oliver J. S. (1990). A summary of major influences on attitude toward and achievement in science among adolescent students Science Education, 74(1), 1-18. Summers, M. (1992). Improving primary teachers' understanding of science concepts - theory into practice. International Journal of Science Education, 14(1), 25-40. Summers, M. & Kruger, C. (1994). A longitudinal study of a constructivist approach to improving primary school teachers’ subject matter knowledge in science. Teaching & Teacher Education, 10(5), 499-519.

168

PRIMARY TEACHERS’ATTITUDES AND COGNITION

Watts, M. & Walsh, A (1997). Affecting primary science: a case from the early years. Early Child Development and Care. 129, 51-61. Webb, P. (1992). Primary science teachers' understandings of electrical current. International Journal of Science Education, 14(4), 423-429. Wragg, E. C., Bennett, S. N. & Carre, C. G. (1989). Primary Teachers and the National Curriculum. Research Papers in Education, 4(3), 17-37.

TEACHING CONCEPTS IN CONTEXTS: DESIGNING A CHEMISTRY TEACHER COURSE IN A CURRICULUM INNOVATION MACHIEL STOLK, ASTRID BULTE, ONNO DE JONG, ALBERT PILOT Utrecht University, The Netherlands

ABSTRACT This paper focuses on the professional development of school chemistry teachers in the context of curriculum reform in The Netherlands. An important aim of this reform is the implementation of teaching chemistry concepts in contexts, which requires substantial changes in current teaching practice. The aim of our research was to develop an empirically validated course design and design principles for courses, on teaching concepts in contexts. A developmental research approach was used with several cycles of an in-depth case study. We describe the design and evaluation results of the first and second cycles. The conclusions are formulated as design principles for a third cycle of this type of teacher in-service course.

1. INTRODUCTION It is widely acknowledged that teachers play a key role in curriculum innovations by interpreting new curriculum documents and enacting them in practice. It has also been accepted that the implementation of curricula should be accompanied by appropriate teacher professional development which takes into account teachers' knowledge, beliefs, and intentions. There is a growing research interest in this type of professional development (Marx et al., 1998; Van Driel et al., 2001). While many studies focus on deriving general design principles and strategies from specific professional development projects (cf. Loucks-Horsley et al., 1998), the present inquiry focuses on the design of a in-service science teacher course and the intended and realized outcomes of the course. This research was conducted in the context of an innovation of the secondary school chemistry curriculum in the Netherlands, for which a governmental committee has proposed the implementation of teaching concepts in contexts. This teaching approach is also found elsewhere, for example, the Salters' project (Campbell et al., 1994)) in the UK and ´Chemie im Kontext´ in Germany (Nentwig et al., 2002). Instead of simply offering teachers new ‘teacher-proof’ curriculum materials, the Dutch committee proposes, inspired by the German project, to involve teachers in the innovation by calling upon their willingness and ability to (re)design 169 K. Boersma et al. (eds.), Research and the Quality of Science Education, 169—180. © 2005 Springer. Printed in the Netherlands.

170

TEACHING CONCEPTS IN CONTEXTS

curriculum materials. For Dutch chemistry teachers, using contexts to teach concepts requires substantial changes in their practice, and to make such an approach successful, they must be able to recognise the fruitfulness of the innovation with respect to their current practice. They need to acquire knowledge about the innovation for underpinning their intention to change. So far, with respect to chemistry teachers, research in this area is scarce. Consequently, our study deals with this early stage of chemistry teacher professional development. 2. BACKGROUND Teachers do not tend to risk changing their own practice which is rooted in the beliefs and practical knowledge they have accumulated during their years of teaching. To enable them to broaden their perspectives and see the fruitfulness of teaching concepts in contexts, teachers should become actively involved in the process. It is essential that they acquire a sense of ownership of a new curriculum. This requires in-service teacher courses which include multiple elements like the use and adoption of teaching materials, opportunities to try new curriculum ideas, reflection by exchanging practical experiences with peers, and a safe and supportive course environment (Van Driel et al., 2001). We explain these elements below (Stolk et al., 2001). (a) The use of innovative teaching materials is essential to make the general curriculum innovation goals more accessible for teachers. It also creates a crucial opportunity for teachers to learn from these materials because they are addressed at the level of their practical knowledge. (b) Reflection. Because of the tacit character of teachers’ beliefs and knowledge, it is very important that teachers are stimulated to exchange experiences and reflect on them. According to Korthagen (2001), reflection promotes teachers’ awareness of their own behaviour, creates alternative methods of action, and extends their practical knowledge. (c) In teacher networks, participants learn more effectively than they do individually; such networks also reduce experienced teachers’ reluctance to change. External conditions, like available time and clarity of learning goals, and internal conditions like teachers’ personal expectations, contribute to creating a collaborative, open-minded environment (Adams, 2000). Teaching chemistry concepts in contexts In traditional chemistry education usually refers to an example of a practical application of chemical concepts already studied. Along with the curriculum demand, such concept-led applications become a kind of justification for learning chemical concepts. A traditional approach as this will not enhance students’ motivation to learn chemistry topics (cf. Van Berkel, De Vos, Verdonk & Pilot, 2000). In this research project, the curriculum materials developed include a context-

M. STOLK, A. BULTE, O. DE JONG, A. PILOT

171

related justification before introducing chemical concepts, and a context-related inquiry where the chemical concepts are applied. We consider a detailed explanation of context-based science approaches to teaching and learning chemical concepts beyond the scope of this paper (for discussion of the topic, see e.g. Westbroek et al., 2004). Research question. The following research question guided the present in-depth study: What is an adequate structure for a teacher course through which chemistry teachers intend to teach chemical concepts in contexts? This intention is further operationalized into three levels: (i) appreciate the usefulness of teaching concepts in contexts, (ii) understand how concepts in contexts can be incorporated in innovative teaching materials, and (iii) appreciate and understand how concepts can be taught in contexts in the teachers’ own practice. 3. RESEARCH DESIGN In this design study the methodology of developmental research was applied (Lijnse, 1995) so that the design of the in-service teacher course consisted of step-wise, argued expectations. These expectations describe what will happen as a result of the planned course activities, and they describe why we think certain effects of the course should arise. Evaluation of the teacher course consisted of comparing our expectations with the realized process during enactment of the course. Data were obtained by video- and audiotapes of course meetings, collecting worksheets, and interviewing teachers. After evaluation, empirical findings, as well as new theoretical considerations, led to an improved structure of the course, making possible a further design cycle. This process resulted in an empirically validated course design, the structure of which can be useful for other, similar courses. In this paper we describe the design and evaluation of the first and second cycles of our teacher course, which led to design-principles for the third and final cycle of our study. 4. FIRST CYCLE The in-service course was directed to adoption and (re)design of context-based approaches. We used a new context-based student module which was developed and taught previously by two teacher-designers. From their experience it was clear that the module had shortcomings which needed improvement. Teacher activities during the in-service course were redesigning the module, teaching the redesigned module, further improving the module, and on the basis of these experiences, constructing a similar, but new module.

172

TEACHING CONCEPTS IN CONTEXTS

The student module The student module was structured in three parts. Each part has a specific instructional function (De Vos, Bulte and Pilot, 2002): In part 1 (context-related justification) students conduct an introductory experiment on the water uptake of a disposable diaper. The instructional function of this experiment is to raise students’ curiosity and a need to explain their results. In part 2 (chemical concepts) students search for explanations in their textbook chapter about organic chemistry: its nomenclature and the structure-property relations of water-absorbing polymers. The instructional function of studying these chemical concepts is to satisfy students’ curiosity and to fulfil their need. In part 3 (context-related inquiry) students conduct some practical assignments, related to the module’s theme, about other properties and applications of superabsorbing polymers. The instructional function of conducting these projects is to make students aware of the fruitfulness of studying these chemical concepts. The teacher course The structure of the in-service teacher course was based on an adaptation of the learning cycle strategy (Abraham, 1998) with three main phases: exploration, invention, and application. We give a short description of the argumentation. (I) Motivation & orientation phase. After a general explication of the module, the teachers needed to conduct the introductory student experiment. We expected them to become motivated to use the module and to understand the stimulating value of the experiment for their students. After an explanation about the shortcomings in the module design concerning the connection of the three parts (instructional functions), the teachers had to discuss the module and redesign it to improve its shortcomings according to their own insight. We expected that teachers would feel co-ownership of the new (redesigned) module. (II) Invention & reflection phase. Teachers had to use the materials in their schools. We expected them to acquire experience while teaching their own redesigned module, and on the basis of their experiences reflect in their peer group on the usefulness of their own designs. As a consequence, we expected that they would be able and motivated to formulate suggestions for improving the module. (III) Application phase. After reflection, teachers were asked to redesign a chapter of their textbook by using the instructional functions as design criteria. We expected them to use their acquired knowledge to accomplish this activity. Research method Participants: Six chemistry teachers participated in the in-service course. They were recruited by advertising in a journal for teachers and by personal communication. Four teachers had more than twenty-five years of teaching experience; the other two had between five and ten years experience; all teachers had a master’s degree in chemistry. The module was taught to students in grade 10 (ages 16-17), at senior high school. Course context: The course covered five three-hour instructional meetings and teaching practice with their own classes from October 2001 to April 2002. For each course meeting, a scenario was written that included the purposes and program,

M. STOLK, A. BULTE, O. DE JONG, A. PILOT

173

teaching tasks for use in school, guidelines for the teacher trainer, and our expectations about the meeting (connected with the relevant phase). The evaluation questions based on these expectations are summarized in Table 1. After the course, the teachers were interviewed about their perceptions of the course objectives and how they valued the course. Data collection and analysis: A multi-method approach of data collection was chosen. Data were collected at specific moments which were closely associated with the design. We used transcriptions of audio records of presentations and discussions, observation notes of the meetings, and assignment forms used by the teachers. The post-course interviews were used to evaluate the course and to confirm some of our interpretations of the data collected during the meetings. All data were analysed independently by two researchers who focused on similarities and differences between the expected and realized outcomes for each of the five phases. The individual analyses were compared and discussed, when necessary, to reach consensus. Findings Table 1 briefly describes the answers to the evaluation questions. Table 1. Structure of course in first cycle and the corresponding evaluation questions based upon the expectations (see text) and the results of the evaluation. Structure (phase) Motivation & Orientation

Evaluation question

•

•

• Invention& reflection

•

•

Application

•

Answers

Did teachers become motivated to use module, and did they understand the value of the introductory experiment? Did teachers understand the instructional functions and the shortcomings of the module? Did teachers become coowner of their redesigned module? Did teachers acquire experiences with their own redesigned module and reflect upon the usefulness of their redesign? Could they formulate (more) suggestions for design?

•

Yes, teachers were enthusiastic, and understood the value of the experiment.

•

No, the functions were not understood (too abstract), and teachers did not recognize shortcomings No, co-ownership

Were teachers able to prepare a new design based on the instructional functions and experiences with the module?

•

• •

Partly, they taught the original module with its shortcomings which they only recognized after teaching

•

No, they asked for a restructured chemical concepts, but could not formulate alternatives No, since the previous steps were already too ambitious a cascade-effect occurred.

174

TEACHING CONCEPTS IN CONTEXTS

(I) Motivation & Orientation phase. After a general explanation, conducting the introductory experiment caused considerable enthusiasm for the module among teachers. All teachers recognized the motivational value of the introductory experiment. However, most teachers did not understand the presented instructional functions, and they could not relate these functions to shortcomings in the design of the module. As a consequence, none of the teachers redesigned the material, and subsequently, they did not feel co-ownership of the module. (II) Invention & Reflection phase. After teachers taught the module, they reported having experienced difficulties in motivating their students to explain the results of the introductory experiment by examining relevant parts of their chemistry textbook; they also reported having difficulty in motivating their students to apply their acquired knowledge in the practical assignments. Thus, it was only after teaching the module, that teachers recognized its shortcomings. Because of the teaching difficulties encountered, teachers were now motivated to redesign the module. They suggested improvements in the connection between the introductory student experiment and the chemical concepts. They also wished to restructure the chemical concepts but did not see how this could be done. (III) Application phase. Teachers were motivated to redesign a particular chapter of their traditional textbook about the structure and properties of water, but they were not able to redesign this chapter according to the instructional functions of the module. Contrary to our expectations, this task was too ambitious for them. After the course when interviewed, teachers indicated some confusion about the course objectives: was this course about improving the student module or about teacher learning? Furthermore, they also mentioned that, in their opinion, the functions of parts of the module were too abstract to be used in practice. 5. CONCLUSIONS AND IMPLICATIONS FROM THE FIRST CYCLE With respect to the three levels of the research question, we can conclude that: (i) Teachers showed an initial appreciation of this context-based approach. (ii) However, it proved to be rather difficult for teachers to develop new ideas about how to connect the concepts in their classroom practice. (iii) Asking teachers to (re)design context-based modules without a more specific preparation connected to classroom experience appears to be too ambitious. This implies that a thorough revision is needed of the course design and the student module. Before asking teachers to construct their own context-based module, they need a more detailed illustration of how such a context–based approach can become operational in their teaching practice. 6. SECOND CYCLE It was decided that in the redesigned student module the instructional functions of the context-based approach some key characteristics of a context-based approach

M. STOLK, A. BULTE, O. DE JONG, A. PILOT

175

must be illustrated in more detail. So, the connection between contexts and chemical concepts was reconstructed and described by three key characteristics: A, B and C. • Characteristic A was an improved connection between the introductory student experiment and the students’ search for explanatory concepts. The introductory experiment was redesigned so that students needed to formulate questions about what they wanted to know. • Characteristic B was the restructured presentation of chemical concepts about water absorbing materials using structure-property relations as an overarching chemical theme. The text was structured in a reverse way compared to a traditional textbook approach: instead of the usual scaling up (from alkanes to networks of polymers) it zoomed in on a specific example. • Characteristic C was the necessity to use chemical concepts in new practical assignments. Characteristics A, B and C formed the basis of teachers' orientation in the second cycle. The revision of the teacher course was guided by an approach based on the model of Gal’Perin for internalization of mental actions, as adapted by Terlouw (2001) for a teacher professional development context. This model offered us a clear perspective on the difficulties we experienced in the first cycle. We now needed to find a solution in which more emphasis would be put on teachers’ classroom experiences with a context-based chemistry module. Then they could begin to formulate ideas for (re)designing context-based modules. Gal’Perin’s model offered heuristic guidelines for the design of a course for teachers: a sequence of creating an initial orientation base before teaching, acquiring teaching experiences, completing the orientation base afterwards, and applying the orientation base for a new product. We expected that this integrated and embedded knowledge would enable teachers to acquire an in-depth understanding of key characteristics of the context-based approach. The structure of the new course now consisted of six phases (see table 2). Phase 0. According to Gal’Perin’s model, the recruitment procedure should be included in our evaluation. This procedure in this phase was similar to what had been done in the first cycle. We expected that the information given teachers about the module and the objectives of the in-service course were concise and to the point. In the first meeting, the teachers had to conduct the introductory student experiment. As in the first cycle, they were expected to become motivated to participate in the course, and to understand the stimulating value of the experiment for their students. Phase 1. Regarding characteristic A, teachers were asked to formulate questions they could expect their students to ask after the introductory experiment. They also needed to check whether such questions could be answered by using the texts about structure-property relations and to devise strategies to handle these questions. Regarding characteristic B, teachers were asked to conduct an alternative introductory student experiment and again to formulate questions which could be expected from students at the end of the experiment. Finally, they should check whether the questions could be answered by a leading, overarching chemical theme.

176

TEACHING CONCEPTS IN CONTEXTS

Regarding characteristic C, teachers were asked to find out which chemical concepts, especially which molecular structures, could be relevant for the student inquiry projects. Phase 2. Teachers had to use the materials at school. They were expected to acquire experience, concerning the three characteristics, by teaching the module. Phase 3. Teachers were expected to exchange teaching experiences and to reflect on them. From their reflection, they were asked to formulate general guidelines about how to teach these characteristics, resulting in a more in-depth understanding of the context-based approach. This phase completed teachers' orientation base as defined by characteristics A, B and C. Phase 4. We expected teachers to apply their knowledge gained from the orientation base so that they could make a sketch of a new context-based module. Teachers were asked to search for a topic they themselves wanted to structure according to the characteristics of the student module. After choosing a topic and an accompanying introductory experiment, they had to carry out the experiment, and think of questions students might ask. With respect to the theme of structureproperty relations, teachers had to define an overarching chemical theme and search for chemical concepts feasible in this theme. Teachers were also expected to select, from a pre-structured collage of small experiments, some experiments which might fit within the theme, and they were expected to formulate design principles for a new context-based module. Phase 5. The teachers were expected to reflect on the accomplished learning results in order to reach a higher level of understanding, which in turn should lead them to attaining a higher level of abstraction of their knowledge about a contextbased approach. Research method During this cycle the applied research method (recruitment, course context and data collection and analysis) was almost identical to the first cycle. Four (new) chemistry teachers participated, one of them had more than twenty-five years of teaching experience; the others between four and fifteen years. All teachers had a master’s degree in chemistry. The five three–hour course meetings and teaching practice with their own classes took place from October 2002 until April 2003. Findings We directed our evaluation only to those parts of the course that led to important recommendations for the third cycle. The evaluation questions of this second cycle and the results are summarized in Table 2. Phase 0. In this cycle the activities in the first meeting had already led to confusion about the objectives of the course because teachers thought they were involved in a process of improving the details of a new innovative module. They reported in the evaluation that they were the subject of an inquiry. As in the first cycle, the introductory experiment had made teachers enthusiastic about using this approach.

M. STOLK, A. BULTE, O. DE JONG, A. PILOT

177

Table 2. Course design with sequence of activities (second cycle), the corresponding evaluation questions based on expectations (see text), and the results of the evaluation with respect to the three characteristics (X: not included in the evaluation). Structure (phase)

Evaluation question

0. Create conditions for teacher learning: - clarify learning goals - motivate to participate

•

1. Create preorientation base - understand underlying contextconcept approach in modules - orientation how to use approach in classrooms

Answers A

B

C

Did the recruitment procedure clarify the learning goals? Did introductory experiment motivate teachers?

•

Yes, but mortality occurred later in course; confused ideas about what this course was about…

•

Yes, teachers were motivated

•

Did course activities lead to understanding of the approach?

•

Did course activities give teachers directions for classroom use?

Yes, teachers formulated student questions, understood relevance No, too much emphasis on understanding instead of use

2. Acquire concrete teaching experiences with approach

•

Did the teachers acquire experiences?

No, most teachers did not teach the introduction as planned

X

3. Create orientation base afterwards - acquire more indepth and reflected understanding of context-concept approach

•

Did the classroom experiences lead to a more in-depth understanding of context-concept approach?

Partly, all teachers formulated general guidelines in encouraging students to ask questions

X

4. Apply orientation to design problem ideas for new context-based modules

•

Did course activities enable teachers to formulate new ideas?

X

X

5. Create metaorientation base to discuss criteria for good/bad examples, raise level of abstraction & new learning goals

•

Did activities stimulate alternative approaches, raising level of abstraction and inviting teachers to learn more?

X

X

•

No understanding; illdefined activity; too high level of abstraction X

No understanding.

No, see remarks A.

No, most teachers did not teach the practical assignments as planned Yes, the use of final versions of students’ reports effective: teachers wanted to know classroom strategies. X

X

178

TEACHING CONCEPTS IN CONTEXTS

Phase 1. Regarding characteristic A, teachers were able to think of questions their students might ask, and they understood the relevance of these questions. However, due to a lack of time, no attention was given to the use of these ‘student’ questions in practice, and no understanding of this evolved. The activities concerning characteristic B did not produce an understanding about the overarching chemical theme. This characteristic seemed to be described in terms which were too abstract. Regarding characteristic C, the corresponding course activity did not produce an understanding about the use of chemical concepts in the practical assignments. Phase 2. Teachers used the module, although most of them did not make use of the questions their students asked during the introductory experiment. Most teachers reported that they did not teach the third part of the module because of lack of time. Phase 3. Regarding characteristic A, it became clear that nearly all teachers did not pay attention to the student questions in class. One teacher reported that her students had written down several good questions. Teachers were very surprised about the quality of these questions; they explicitly mentioned the usefulness and their appreciation of the questions. All teachers were able to formulate a set of guidelines regarding ways of encouraging students to ask adequate questions. Concerning characteristic C, a initial orientation base was not achieved. Final versions of student reports were analyzed on the use of structure–property relations. The teachers were surprised that students of one teacher had indeed used these relations and wanted to know from their colleague how he had guided his students. The teachers wanted to know what classroom strategies were used to arrive at such student results. Phases 4 & 5 were not evaluated because too many participants had dropped out, partly for personal reasons and partly because teachers did not find the course relevant in their (already overloaded) teaching practice. Conclusions and implications from the second cycle As in the first cycle, the recruitment procedure led to confusion about the course objectives with the result that, apparently, the course did not build properly on teachers’ prior expectations and beliefs. This led leading to less motivation to participate in the course. However, the course structure, intentionally using classroom experiences to create an orientation base afterwards (phase 3) was successful for both characteristic A and C. Teachers were able to formulate guidelines about classroom practice, which were based upon real classroom experiences. Compared to the first cycle, we achieved an improvement concerning two levels in our research question: teachers achieved (ii) some understanding of a context-based approach, and (iii) the competency to formulate practical guidelines. The activities directed towards guidelines for teacher actions in classroom practice proved to be successful. The rather abstract characteristic B, the overarching chemistry theme connecting the introductory experiment with the practical assignments, proved difficult to understand for teachers. The structure of the student module was still not good enough to give teachers a clear perspective of how structure–property relations could be used for practical assignments. However, applying the orientation base for the

M. STOLK, A. BULTE, O. DE JONG, A. PILOT

179

design of a new module (phase 3 -> phase 4), seemed to be beyond teachers' perspective at this time. The conclusions of the first and second cycles can be formulated as design principles that need specific attention in this type of teacher course. They will be used in the third cycle of this study to redesign the course structure: • Define and communicate the course objectives clearly, and the (transparent) roles for the teachers, teacher trainer, and researcher; and build on the prior expectations and beliefs of teachers. • Maintain the overall structure of the course in the six phases, as formulated. The sequence – create an initial orientation base, acquire experiences, and complete the orientation base afterwards – is an important design principle for this type of teacher course. • Reduce the level of abstraction (removal of characteristic B), and plan course activities in such a way that they are directed to form guidelines for teachers’ actions in their classroom practice. • The requirements for the student modules to be used in this type of teacher training are high: characteristics of a concept-based approach should be clear, e.g. there should be a transparent connection among the introductory experiment and the chemical concepts that should give an answer to student questions which are evoked, and the use of theory in the practical assignments. • Make the design of new context-based modules relevant from a teacher’s point of view REFERENCES Abraham, M. (1998). The learning cycle approach as a strategy for instruction in science. In B.J. Fraser & K.G. Tobin (Eds.), International Handbook of Science Education. Dordrecht: Kluwer Academic Publishers. Adams, J.E. (2000). Taking Charge of Curriculum: Teacher Networks and Curriculum Reform. New York: Teacher College Press. De Vos, W., Bulte, A.M.W. & Pilot, A. (2002). Chemistry Curricula for General Education: Analysis and Elements of a Design. In: J.K. Gilbert, O. de Jong, R. Justi, D.F. Treagust & J.H. van Driel (Eds.), Chemical Education: Towards Research-based Practice. Dordrecht: Kluwer Academic Publishers, pp. 101-124. Cambell, B., Lazonby, J., Millar, R., Nicholson, P., Ramsden, J. & Waddington, D. (1994). Science: The Salters’ Approach – A case study of the process of large scale curriculum development. Science Education, 78 (5), 415-447. Korthagen, F.A.J. (2001). A reflection on reflection. In: F.A.J. Korthagen, Linking practice and theory. The pedagogy of realistic teacher education. London: Erlbaum Publishers. Lijnse, P.L. (1995). Developmental research as a way to an empirically based ‘didactical structure’ of science. Science Education, 79 (2), 189-199. Loucks-Horsley, S., Hewson, P., Love, N. & Stiles, K. (1998). Designing professional development for teachers of science and mathematics. Thousand Oaks: Corwin Press. Marx, R., Freeman, J., Krajick, J. & Blumenfeld, P. (1998). Professional development of science teachers. In B.J. Fraser & K.G. Tobin (Eds.). International Handbook of Science Education. Dordrecht: Kluwer Academic Publishers.

180

TEACHING CONCEPTS IN CONTEXTS

Nentwig, P., Parchmann, I., Demuth, R., Graesel, C. & Ralle, B. (2002). Chemie im Kontext, from situated learning in relevant contexts to systematic development of chemical concepts. Paper presented at the second IPN_YSEG Symposium on context-based curricula, October 10- 13th, Kiel, Germany. Stolk, M.J., De Jong, O. & Pilot, A. (2000). Professional development of teachers in a process of chemistry curriculum innovation. In: R.H Evans, A. Møller Andersen, & H. Sørensen (Eds.), Bridging Research Methodology and Research Aims. Proceedings of the 5th European Science Education Summerschool (Gilleleje, Denmark. September 6-13, 2000). The Danish University of Education, Copenhagen. Terlouw, C. (2001). Een integrale aanpak van ICT scholing: opzet en ervaringen [an integral approach of ICT professional development: design and experiences]. In A. Pilot & H. Purper (Eds.), Leren in een informatiewereld [Learning in an information society]. Alphen aan den Rijn: Kluwer. Van Berkel, B., De Vos, W., Verdonk, A. H. & Pilot, A. (2000). Normal Science Education and its Dangers: The case of School Chemistry. Science and Education, 9, 123-159. Van Driel, J., Beijaard, D. & Verloop, N. (2001). Professional development and reform in science education: the role of teachers’ practical knowledge. Journal of Research in Science Teaching, 38(2), 137-158. Westbroek, H.B., Klaassen, C.W.J.M., Bulte, A.M.W. & Pilot, A. (2004). Characteristics Of Meaningful Chemistry Education. In: K. Boersma, O. de Jong, H. Eijkelhof, & M. Goedhart (Eds.). Research and the quality of science education. Dordrecht: Kluwer Academic Publishers.

EPISTEMOLOGICAL THOUGHT AND ROLEPLAYING: IMPACT ON PRE-SERVICE TEACHERS' OPINIONS ON MOBILE PHONE RISKS

VIRGINIE ALBE, LAURENCE SIMONNEAUX Ecole Nationale de Formation Agronomique, France

ABSTRACT The purpose of this case study was to evaluate the changes in opinion of a group of future teachers on the danger of mobile telephones. The opinions were determined before and after epistemological work and role-playing in a form of a legal suit. This activity was part of pre-service teachers' initial training in a one-week module dedicated to prepare them to lead debates on controversial scientific issues. The fifteen participants were future physical science and biology teachers. A comparison of pre- and post-test evaluations revealed that these pre-service teachers were less sure of the risk involved in using mobile telephones following the activity and that the role-playing affected their epistemological interpretation of the research results.

1. INTRODUCTION The need for scientific education has been emphasised in several countries. It has been recommended, for example, that students should be prepared in the scientific culture they need in order to make informed decisions on scientific issues. In France, science teaching should enable students to “take part in citizens’ choices concerning issues that involve science" (Bulletin Officiel, 1999). Various authors have suggested introducing case studies on current controversies in science teaching for the purpose of educating citizens (Grace & Ratcliffe, 2002; Kortland, 2001; Lewis & Leach, 2001; Ryder 2001). Some of them justify this scientific education for action (Jenkins, 1994; Osborne, 1997; Zoller, 1982; Désautels et al., 1995), while others have argued that it is important to train students to understand the nature of science (Millar & Wynne, 1988; Sadler et al. 2002). We agree with these positions and consider that the teaching of socio-scientific issues is crucial as there are many controversial scientific developments at the present time. Such issues concern, for instance, questions raised by biotechnology, BSE, food safety, the greenhouse effect, mobile telephones, and the ecological and economic repercussions of agricultural practices. As emphasised by Legardez et Alpe (2001), these issues are socially relevant for three reasons: 181 K. Boersma et al. (eds.), Research and the Quality of Science Education, 181—191. © 2005 Springer. Printed in the Netherlands.

182

EPISTEMOLOGICAL THOUGHT AND ROLE-PLAYING

they are controversial and lead to debates on the production of scientific knowledge; - they are controversial and lead to debates in which the actors in the didactic situation, students and teachers, cannot avoid being involved since such issues are crucial in the social and media environments they experience; - they are controversial in the classroom, but teachers do not feel capable of dealing with them. These issues are characterised by a lack of consensus among researchers, particularly on health risks and environmental impact. In our opinion, the challenge is to train informed people in research methods, applications, and their possible repercussions, so that they are capable of taking rational decisions in cases where facts are uncertain and of participating in debates. We consider ourselves to be participants in the educational trend of studying the interactions between science, technology and society. The teaching of controversial issues helps to educate people for citizenship, and therefore it is important to exploit situations in which the declarations of various researchers, institutions and journalists are debated and examined for educational purposes (Kolstø, 2001; Simonneaux, 2001). Teaching of these issues places uncertainty and risk at the centre of the teaching-learning process. Moreover, this frontier science, also known as science-in-the-making, shows how scientific research is integrated into social and commercial activity (Aikenhead, 1994; Jenkins, 1992). These are disturbing situations since the dominant trend is to consider science as an authority which cannot be discussed. Science and technology teachers feel responsible for teaching facts, but at the same time, they do not think they have the required competence to deal with social and ethical questions or to manage debates. The main result from a large scale survey done in England and Wales is summed up as follows: “half of all science teachers interviewed feel that teaching science should be 'value free' ” (Levinson & Turner, 2001). However, these issues provide lively discussions in the classroom and are increasingly introduced in science teaching programs. During the implementation of the new programs in French agricultural teaching curricula, teachers are advised to give particular attention to raising their students' social awareness as citizens. Concerning the teaching options put forward, debates are particularly recommended (Ministère de l’Agriculture et de la Pêche, 2000). Thus, teachers are being called upon to commit themselves to pedagogical practices with which they are not familiar and which are based on controversial scientific issues involving economic, political, environmental, cultural and ethical aspects. In a previous study (Albe & Simonneaux, 2002) on the declared intentions of teachers with respect to the teaching of controversial scientific issues, we showed that, a priori, teachers were not necessarily reluctant to deal with social problems related to the development of techno-sciences, but that this brought into question their epistemology and teaching practices. The teaching of controversial scientific issues is thus a complex phenomenon. We think that one way to study this process is to enquire into science teachers' opinions in the context of a socio-scientific issue characterised by uncertainty and controversy. In this paper, we have investigated two research aims. One is to ascertain how the opinions of future science teachers on the danger of mobile -

V. ALBE, L. SIMONNEAUX

183

telephones change following a role-playing situation. The second is to determine their epistemological interpretations of the research results on this controversial issue. 2. METHODS We offered teachers undergoing initial training in our institute the possibility of learning how to teach by means of controversial scientific issues. Fifteen future science teachers who had no professional experience agreed to follow this one-week course. Thirteen of this group were training to be teachers of physical science (in France, physics and chemistry are part of a single teaching discipline), and two of them were training to teach biology and ecology (one discipline). During the training session, we were inspired by a module developed by Hind et al. (2001) on mobile telephones: Assessing Data Quality: Mobile Phones – Health Risk or Scare? The module involves a simulated legal suit in which a telecommunications employee suffering from cancer, brings a suit against his employer on the basis that use of a mobile telephone caused his illness. After being introduced to the role playing situation, the future teachers were divided into two groups: the experts for the defence lawyers and those for the plaintiff's lawyers. Each group included six physical science teachers and a biology teacher, while one teacher played the role of judge. In a pre-test, they were asked to write their answers to the questions: In your opinion, are mobile phones dangerous for human health? Why? Then, eight extracts of research results from Hind et al. (2001) were distributed to participants so that they could prepare the trial. These extracts were from current research papers dealing with the occurrence of disease in animals, epidemiological investigations, and tests of memory (see appendix). Each group was asked to specify the epistemological status of the research results and to establish whether each of the research projects: - had tried to identify a physiological mechanism by which telephone microwaves could negatively affect the tissue of living organisms; - had tried to identify a correlation between the occurrence of health problems and emissions from mobile telephones; - had indicated whether there might be an increased health risk due to the use of mobile telephones; - was only of limited value because the relationship between the results obtained on animal tissue and cancer in humans was not necessarily valid; - was of limited value as other research groups had not managed to repeat the results in similar experiments; - was of limited value because the size of the sample used meant that it was not statistically relevant or sufficiently reliable. Each group wrote out the three best proofs according to which, mobile telephones are, or are not, dangerous to health. They also wrote down the list of arguments they intended to develop to convince the other group during the role-play, and the questions they intended to ask the other group. Following a preliminary

184

EPISTEMOLOGICAL THOUGHT AND ROLE-PLAYING

study of the documents, the 'trial' took place. During the 'trial' different expert witnesses were called to debate by the 'judge'. This phase lasted 45 minutes. In a post-test after the role-play, the participants were asked to re-evaluate their opinion on the first two questions (In your opinion, are mobile phones dangerous for human health? Why?) and to specify which condition(s) might lead them to change their opinion. Following this, use of such role play in classroom situations was discussed. 3. RESULTS As the activity progressed, the future teachers became less and less sure of themselves as to whether mobile telephones are in fact dangerous for human beings. During the pre-test, most of the future teachers considered that mobile telephones were dangerous to for health, while during During the post-test, most of the group future teachers said that they were not sure. Table 1. Opinions of future science teachers on the danger of mobile telephones before and after the role play AFTER THE DEBATE YES

BEFORE THE DEBATE

YES

T8 T9 T10 T11 T12 T15

YES with uncertainty

T4

NO

T6 T13

NO with uncertainty

YES with NO uncertainty

UNCER TAIN T8 T9

T10 T11 T12 T15 T4 T6 T13

T1 T5

UNCERTAIN T2 T3 T7 T14

NO with uncertainty

T1 T5 T2 T3 T7 T14

V. ALBE, L. SIMONNEAUX

185

Some future teachers justified their arguments on the grounds of conceptual errors in physics, while others had strong prejudices which shored up their case and which were not questioned either by a study of the research results or by the discussion: - infrared and ultraviolet rays cause an increase in mutation rates, hence this must be true for microwaves of mobile telephones as well; - we live without any problems in a magnetic environment, hence the magnetic waves of mobile telephones should not be a problem; - when you play sport, the body's temperature rises to 40°C without any problem, hence a possible increase in brain temperature should not be a problem either. After the role play exercise, others based their arguments on an analysis of the research results. For one future teacher, an unconditional trust in the positive benefits of science was enough to justify that mobile telephones were not dangerous. Fourteen out of fifteen future teachers had an epistemological discussion on the conditions which might cause them to change their mind: - if long-term studies were to be conducted; - if several scientific studies revealed the same results; - if studies were made on a large statistical population; - if significant results were available for humans… One future teacher referred to the socio-economic effect on research results: - I would change my mind if I had access to reliable research results and if we were told the truth. As long as money stops us from knowing the truth, there is no way we can be sure. The others did not refer at all to any impact that the type of research funding might have on research results even if it came from telephone companies. The roles taken by individuals in the role play affected the epistemological interpretation of the research results. 'Experts' for the defence and for the plaintiff did not base their proofs or arguments on the same research results; they also used the results differently to develop their opinion on the risk of using mobile telephones. The group defending the case that mobile telephones are dangerous (group A) based its arguments on the results of three research projects when stating their three best proofs of this risk and the arguments they intended to develop against the other side. Their proofs were also their arguments. The research was done by De Pomeroi and his research team (increase in the cell division of nematode larvae exposed to microwaves), the Lai team (increase in the corticotrophin stress hormone in rats exposed to microwaves), and the Repacholi team (development of lymphoma in transgenic mice exposed to microwaves). (See appendix for descriptions of these research projects.)

186

EPISTEMOLOGICAL THOUGHT AND ROLE-PLAYING

Table 2. Use of the research projects by future science teachers defending the case that mobile telephones are dangerous (group A) or are not dangerous (group B) Emphasis on results Emphasis on epistemology RESEARCH PROJECTS (products of science) (process of science) 1 : De Pomeroi and team Group A Group B 2 : Lai and Team Group A Group B 3 : Tattersal and team Group A Group A & Group B 4 : Preece and team 5 : Repacholi and team Group B 6 : Brooks Air Force Base Group B 7 : Hardell and team Group B 8 : Carlo and team Group B Group B Those who were defending the point of view that mobile telephones are not dangerous to health (group B) based their arguments on a greater amount of research. Of the 8 texts provided, 7 were used by this group of future teachers who mainly focused their arguments on epistemological considerations of the research extracts. Their three best proofs referred to the following: - the non-validity of transferring results obtained on animal tissue to humans (research done by De Pomeroi, Lai, Tattersal, Repacholi teams and the experts at the Brooks Air Force Base), - the necessity of studying larger making studies of bigger samples for a longer period of time, and - the fact that "no particular relationship has yet been established between brain cancers and the use of mobile telephones" (research done by Hardell and Carlo). Moreover, the future teachers in group B questioned the validity of the results of Carlo's research, and pointed out its limitations and weaknesses. For instance, the number of people (30) with a particular type of cancer called neurocytoma did not seem to them to be especially significant. Of these 30 people, 12 had used mobile telephones, but there was no indication of the way in which the telephones had been used or the frequency of use. In addition, this group emphasised that this was a particular tumour, and wondered whether there was a "causal relationship" between the use of mobile telephones and the appearance of a tumour in the brain - What percentage of people suffering of tumours uses mobile phones? This percentage may be smaller than the 40% in the study. Finally, the Carlo team had done research on wireless technology – a possible competitor of mobile telephones. The analysis conducted on both groups showed how they were immediately influenced in their epistemological interpretation of the research results by the roles they took in the exercise. When asked to specify the epistemological status of each research extract, each group of future teachers developed opposing points of view depending on the position they would have to defend as to the risk of telephones to human health. In particular, the group which defended the case that mobile

V. ALBE, L. SIMONNEAUX

187

telephones are dangerous considered that 7 out of 8 research projects seemed to indicate that there might be an increased health risk for people using mobile telephones, whereas the group defending the opposing point of view felt that there were no research grounds for believing this. The future teachers in the latter group moreover considered that the research done by the teams of Lai, Tattersal, Repacholi and the experts at the Brooks Air Force Base were only of limited value because the relationship between the results obtained on animal tissue and cancer in humans is not necessarily valid. However, this point was not admitted by the opposing group. Both groups did agree that the research done by the De Pomeroi team was not necessarily valid, but this did not prevent the group defending the argument, that mobile telephones are dangerous, from basing its case on the results of the De Pomeroi research when quoting proofs and giving arguments they intended to develop against the opposing party. Moreover, the two groups of future teachers disagreed with respect to 5 research projects aiming to identify a physiological mechanism by which telephone microwaves might have negative effects on living organic tissue. The group defending the case that mobile telephones are dangerous disagreed with the research done by Tattersal, Repacholi, and experts at Brooks Air Force Base, while the group defending the point of view that mobile telephones are not dangerous disagreed with the research conducted by Preece and Hardell. 4. CONCLUSION AND IMPLICATIONS Most of the future teachers voted in favour of the motion that mobile telephones are dangerous. The comparison of pre- and post-test evaluations revealed that following the role-playing situation, the future teachers were much less certain as to the risk of mobile telephones to human health. Some of them felt that their proofs indicated conceptual errors in physics, others had strong ideas which were not brought into question either by a study of the research results or by the debate, whereas some, following the simulated court legal suit, based their arguments on an analysis of research results. Almost all of the future teachers used epistemological arguments when identifying the conditions which might lead them to changing their views, and almost none of them considered the social aspects. In preparing its arguments as to the health risk of using mobile telephones, one group referred to the impact that the type of funding might have on research results. The roles assumed in the exercise affected the epistemological interpretation of the research results. The future teachers did not grant the research results the same epistemological status, nor did they use the same results for proofs or arguments; they also used the results differently to develop their position with respect to the possible risk of mobile telephones. Those who defended the point of view that mobile telephones are not dangerous for human health based their arguments on a greater amount of research. One of the challenges of this study was to train teachers to reflect on their teaching practices, in particular by enabling them to distance themselves from their

188

EPISTEMOLOGICAL THOUGHT AND ROLE-PLAYING

personal opinions, to reflect critically on the training activity, and to focus on the teaching of new and vital issues related to knowledge and society. After the role-play exercise, future teachers specified the modifications they would apply to this activity to implement it in schools. They stressed the necessity to work on the notion of uncertainty with pupils and they also stressed the importance of having their pupils do preliminary bibliographical research before they engage in the role play activity. Pupils' bibliographical research would be done as a multidisciplinary approach, possibly on the Internet. According to the future teachers, different notions could also be treated before the role play or in another class. For example, in biology class, cancer development, mutagen factors' actions on cells, use of animals in laboratory research, and functioning of the nervous system could be discussed. In physics class, there could be discussions on electromagnetic waves, wave-matter interactions, wave energy, microwaves; in mathematics teachers could point out notions of statistics. These future teachers expressed that they were willing to implement similar role-play exercises with their future pupils on various socially controversial scientific questions. REFERENCES Aikenhead, G. (1994). The social contract of science. In J. Solomon & G. Aikenhead (Eds.), STS Education. International Perspectives on Reform, New York: Teachers College Press, 11-20. Albe, V. and Simonneaux, L. (2002). L’enseignement des questions scientifiques socialement vives dans l’enseignement agricole : quelles sont les intentions des enseignants ? Aster, 34, 131-156. Bulletin Officiel (1999). L’enseignement des sciences au lycée. Numéro hors-série 6. Désautels, J., Gagné, B., Gauthier, R., Larochelle, M., and Lessard, N. (1995). La participation des scientifiques québécois au projet HUGO – pour ou contre ? Rapport de recherche. Université Laval, Québec. Grace, M.M. and Ratcliffe, M. (2002). The science and values that young people draw upon to make decisions about biological conservation issues. International Journal of Science Education, 24 (11), 1157-1169. Hind, A., Leach, J., Ryder, J. and Prideaux, N. (2001). Teaching about the nature of scientific knowledge and investigation on AS/A level science courses. Leeds: CSSME. Jenkins, E.W. (1992). School science education : towards a reconstruction. Journal of Curriculum Studies, 24 (3), 229-246. Jenkins, E.W. (1994). Public understanding of science and science education for action. Journal of Curriculum Studies, 26 (6), 601-611. Kolstø, S. D. (2001). Scientific literacy for citizenship: Tools for dealing with the science dimension of controversial issues. Science Education, 85 (3), 291-310. Kortland (2001). A problem posing approach to teaching decision making about the waste issue. Utrecht : CD-β Press. Legardez, A. and Alpe, Y. (2001). La construction des objets d'enseignements scolaires sur des questions socialement vives: problématisation, stratégies didactiques et circulations des savoirs. Paper presented at the 4th Congress AECSE in Lille, September 5-8, 2001. Levinson, R. and Turner, S. (2001). Valuable lessons engaging with the social context of science in schools. Recommendations and summary of research findings. London : The Wellcome Trust.

V. ALBE, L. SIMONNEAUX

189

Lewis, J. and Leach, J. (2001). Reasoning about socio-scientific issues in the science classroom. A paper presented at the conference of the European Science Education Research Association in Thessaloniki. Millar, R. and Wynne, B. (1988). Public understanding of science: from contents to processes. International Journal of Science Education, 10 (4), 388-398. Ministère de l’Agriculture et de la Pêche (2000). Note de service n° 2000-2072 du 18 Juillet 2000. Osborne, J. (1997). Science education for the future – the road ahead ? Paper presented at the first conference of European Science Education Research Association, September 1997, Rome. Ryder, J. (2001). Identifying science understanding for functional scientific literacy. Studies in Science Education, 36 (1), 1-44.

Sadler, T.D., Chambers, F. W. and Zeidler, D. L. (2002). Investigating the Crossroads of Socioscientific Issues, the Nature of Science, and Critical thinking. Paper presented at the National Association for Research in Science Teaching Annual Meeting in New Orleans, April 7-10, 2002. Simonneaux, L. (2001). Role-play or debate to promote students' argumentation and justification on an issue in animal transgenesis, International Journal of Science Education, 23 (9), 903-927. Zoller, U. (1982). Decision-making in future science and technology curricula. European Journal of Science Education, 4 (1), 11-17.

190

EPISTEMOLOGICAL THOUGHT AND ROLE-PLAYING

APPENDIX Research project 1: David De Pomeroi and team – University of Nottingham The team of researchers beamed microwaves at tiny nematode worms. These were chosen because their cell biology is simple and well understood. In one experiment the team found that larvae exposed to an overnight dose of microwaves wriggled less and grew 5% faster than those in a control group. This suggests that microwaves may be speeding up cell division. The group now plans to investigate whether a similar effect can be observed in mammalian cells, a finding that would raise fears about a possible link with cancer. Research project 2: Henry Lai – University of Washington, Seattle A team in Seattle has researched the effect of microwaves on levels of stress in rats. They have found evidence that exposure to microwaves in rats causes the production of corticotrophin releasing factor, a stress hormone that disrupts neurotransmitters in the brain; neurotransmitters are involved in memory and alertness. The rats showed an increased tendency to binge on alcohol and took longer to learn the location of a submerged platform in cloudy water. Research project 3: John Tattersal and colleagues – Defence Evaluation and Research Agency This team exposed slices of rat brain to microwaves. They found that the exposure reduced electrical activity and weakened response to stimuli. The brain slices were taken from the hippocampus, a part of the brain with a role in learning. However Tattersall has indicated that he thinks the hippocampus is too deeply buried within the brain to be affected by mobile phones. His more recent research has shown that nerve cell synapses may become more receptive to changes linked to memory when they are exposed to microwaves. Research project 4: Alan Preece and colleagues – University of Bristol The group used a device that mimicked the emissions of mobile phones to test the responses of volunteers asked to recall words and pictures shown on a screen. The microwave emissions had no effect on recall but when response time was tested by asking the volunteers to press a button matching a ‘Yes’ or ‘No’ image on screen, those with the headsets switched on showed about a 4% improvement in response times. The effect was seen in two separate groups of volunteers. Research project 5: Michael Repacholi and research group – Royal Adelaide Hospital The group spent 18 months exposing mice to emissions that mimicked those of mobile phones. The group used mice that had been genetically engineered to increase their susceptibility to lymphoma in order to make the experiment more sensitive. They found that twice as many mice exposed to radiation developed lymphomas as those in the control group.

V. ALBE, L. SIMONNEAUX

191

Research project 6: Brooks Air Force Base – San Antonio, Texas Experts used mice genetically engineered to be susceptible to breast tumours in a study similar to Repacholi’s. They exposed the mice to radiation for 20 hours a day for 18 months. There was no increase in the rate of tumours in these mice. Research project 7: Lennart Hardell – Orebro Medical Centre A study was made of 209 people with brain tumours and a control group of 425 without brain tumours. They found that mobile phone users were no more likely to develop tumours than non-users. Of those with tumours however, mobile phone users were 2.5 times more likely to develop tumours close to their ‘phone ear’ than the non-users. There were only 13 mobile phone users with tumours in the study group so the result may not be statistically significant. Research project 8: George Carlo – Wireless Technology Research, Washington DC Here researchers studied 450 people with brain tumours and a control group of 425. There was no general link between brain cancers and mobile phone use. However, the study identified a smaller group of 30 people who had a particular form of brain tumour called a neurocytoma; 40% of this group were mobile phone users. This compared with the control group without neurocytomas, in which only 18% were mobile phone users. This result is statistically significant.

PART 4 Teaching-learning sequences in science education

TEACHING-LEARNING SEQUENCES TOOLS FOR LEARNING AND/OR RESEARCH

MARTINE MÉHEUT Créteil IUFM and LDSP-Paris 7 University, France

ABSTRACT What can contribute to the 'value' of a piece of research about an innovative teaching-learning sequence from a research point of view and a teacher’s point of view? We will try to demonstrate that such values from these two perspectives are different, but not contradictory, and that they can be sought in the same research work. Two aspects will be developed and illustrated. The first aspect is about 'a priori' justification. We will propose a general framework which can help to make the principles underlying the design of a sequence clear, and so situate various teaching-learning sequences concerning the same domain of knowledge. Such a framework can be useful both for researchers to make their choices and hypotheses more explicit and for teachers to select one approach over another. The second aspect is about 'a posteriori' or 'empirical' validation. Referring to various pieces of research work, we will discuss the limits of usual 'comparative' approaches and will focus on more 'internal', 'descriptive' approaches. We will argue that describing cognitive pathways of learners through teaching-learning situations constitutes a fruitful tool, both for researchers to validate some of the choices or hypotheses underlying the design of the learning situations and for teachers to feel more comfortable with such innovative teaching-learning sequences.

1. INTRODUCTION Coming back to the seventies and the early eighties, we can remember the importance of research about students’ (mis)conceptions and spontaneous, common ways of reasoning. A question then arose: how to take into account such pieces of information for teaching? Trying to give answers to this important question, many teaching-learning sequences (TLS) have been developed and experimented with in classrooms. Is it now possible to identify general frameworks which could be used by researchers to develop such sequences? What are we doing when experimenting with such sequences? What kinds of results can we seek? These are two questions I would like to discuss now in light of numerous pieces of research work which were developed over a period of about twenty years. These 195 K. Boersma et al. (eds.), Research and the Quality of Science Education, 195—207. © 2005 Springer. Printed in the Netherlands.

196

TEACHING-LEARNING SEQUENCES TOOLS

questions formed the subjects of an International Symposium in Paris (Méheut & Psillos, 2000) and of a workshop during the ESERA Conference in Thessaloniki (Psillos & Méheut, 2001). During these meetings contributions were presented and discussed which have been published in a special issue of the International Journal of Science Education . 2. SOME APPROACHES IN DESIGNING TEACHING-LEARNING SEQUENCES In order to characterize various approaches, let us start with a very simple model of teaching-learning situations. This first model (Figure 1) implies four components: teacher, learners, material world, and knowledge to be developed.

Scientific Knowledge e p i s t e m i c

TEACHER -----------------------------------------------STUDENTS pedagogical

d i m e n s i o n

dimension

Material World

Figure 1. A « didactical rhombus » to describe the design of a TLS This pictorial representation allows us to organize various considerations we can put into play when designing a TLS. The vertical axis represents an epistemic dimension, i.e. how knowledge works with respect to the material world. Along this axis, we can find assumptions about scientific methods, processes of elaboration, and validation of scientific knowledge. The horizontal axis represents the pedagogical dimension. We find along this axis choices about a teacher’s role, types of interactions between teacher and students, and close to the vertex “students”, we can place what is expected about interactions among students. Using this framework, we can characterize two prototypical approaches in designing teachinglearning sequences.

M. MÉHEUT

197

Picturing “cognitive-conflict” approaches (see for instance Nussbaum & Novick, 1982; Driver & Erickson, 1983; Driver & Bell, 1986; Nussbaum, 1989; Dewey & Dykstra, 1992; Ravanis & Papamichael, 1995) makes clear that great importance is given in such approaches to learners, to their conceptions and ways of reasoning, and to the confrontation with answers from the material world (Figure 2). It also reveals the weakness of two important components of a teaching-learning situation, that is, the part played by the teacher and the knowledge to be developed.

Scientific Knowledge ?

TEACHER ?

STUDENTS Conceptions

Cognitive Conflict Material World

Figure 2. Figuring cognitive conflict approaches It leads us to ask such questions as: What are the outcomes of such cognitive conflicts? What knowledge can be produced by learners in such situations? What can be the part of the teacher in such an approach, and with what kinds of interventions? On the other hand, we can characterize rather “epistemic” approaches (Figure 3) which were developed in France, for example, in the eighties and nineties (see Martinand, 1983; Lemeignan & Weil-Barais, 1988 & 1992; Tsoumpelis, 1993; Robardet, 1995). In such approaches, most attention is paid to the knowledge to be developed relative to the physical world, to historical genesis of this knowledge, and/or to possible artificial genesis; all of this without paying great attention to teachers and learners.

198

TEACHING-LEARNING SEQUENCES TOOLS

Scientific Knowledge

TEACHER ?

Epistemic problems

STUDENTS ?

Material World

Figure 3. Figuring epistemic approaches We can formulate here other questions: How much will learners get involved in such “epistemic” problems? How can they solve them? Will the expected knowledge be produced? What can be the part played by the teacher? With which interventions? Both kinds of considerations, which are rather pedagogical and epistemological, can be interlaced into what we will call “integrated constructivist” approaches, many examples of which can be found in literature published in the nineties. These pieces of research reflect various points of view about the processes of developing knowledge; some of them give great importance to contradictions (see for example Kaminski, 1991; Chauvet, 1996; Viennot & Rainson, 1999) as a source of motivation for learning, while others work more from analogies (Schwedes & Schmidt, 1992; Arnold & Millar, 1996; Duit et al., 1998; Komorek et al., 2001). We can also notice an important trend using a “modelling” point of view (Méheut & Chomat, 1990; Tiberghien, Psillos & Koumaras, 1995; Kariotoglou, Koumaras & Psillos, 1995; Méheut, 1997; Gilbert & Boulter, 1998). In such “integrated constructivist” approaches, attention is paid not only to the epistemic dimension, but also to the “pedagogical” one. Great attention is also paid to students, their conceptions, their own ways of reasoning, and (sometimes) motivational dimensions of the teaching-learning situations. Some support can be found in general methodological frameworks, such as for example the "problem-posing approach" as presented by Piet Lijnse in this volume or the framework of “Educational Reconstruction” presented by Kattmann et al.

M. MÉHEUT

199

(1995). Let us also consider the framework of “Ingenierie Didactique” as developed by math education researchers (Artigue, 1988). Three main dimensions for a priori analyses are suggested: - an "epistemological dimension" analysing the contents to be learnt, the problems they can answer, their historical genesis; - a "psycho-cognitive" dimension analysing the students’ cognitive characteristics; - a "didactic" dimension analysing the constraints due to the functioning of the teaching institution (programs, timetables, and so on). These a priori analyses are interlaced in order to define accurately "problems" to be managed by students and to anticipate the elaboration of knowledge by students through these "problems". A comparison of the cognitive itineraries actually observed with those predicted can validate or challenge the hypotheses involved in the building up of learning situations. According to "Educational Reconstruction" as to "Ingenierie Didactique", great attention is paid to the analysis of scientific knowledge put into play on one side, and to the learning difficulties, the learners’ conceptions and ways of reasoning in the domain on the other side. Moreover, it is noticeable that the “Ingenierie Didactique” framework includes institutional constraints, whereas in “Educational Reconstruction” framework motivational aspects and social significance of the knowledge are emphasized. It may be remarked that we have not found many pieces of research work which propose and analyze, precisely, interventions of the teacher during a given TLS. More attention has been paid again to this vertex of our "didactical rhombus" by a few researchers (Dumas-Carré & Weil-Barais, 1998; Leach & Scott, 2002) referring to a Vygotskian approach, but this dimension seems to be considered thus far in a quite general way that is independent of other dimensions of a given TLS. 3. SOME WAYS OF VALIDATING TEACHING-LEARNING SEQUENCES We can now look at different points of view about validation of a TeachingLearning Sequence: the first one can be said to be an “external” or “comparative” evaluation; the second one is rather “internal”. Both points of view can be considered as complementary. Comparative evaluation In most cases, TLSs are evaluated by a pre-test/post-test procedure (see for example Minstrell, 1992; Nikolopoulou, 1993; Ravanis & Papamichael, 1995; Kariotoglou, Koumaras & Psillos, 1995; Psillos, 1998; Chang & Barufaldi, 1999) in an attempt to compare the effects of a research TLS with those of usual teaching.

200

TEACHING-LEARNING SEQUENCES TOOLS

Post-test

Aims1

Sequence

Students 1

1

Aims0

Sequence

Pr é-test

0

Students 0

Figure 4. Comparative evaluation Two types of questions can be formulated. The first one refers to the aims: We compare two ways of teaching with respect to our own aims, but are the aims of usual teaching the same as ours? What would happen if we evaluate both sequences with respect to the aims of usual teaching? We probably obtain better results than usual by teaching with respect to our aims: What are the factors of success? Could one be a “placebo” effect linked to the motivation of the teacher? What are the choices, the components of the teachinglearning situations which are important and can “explain” the success? Internal validation Another approach is to analyse the outcomes of our TLS with respect to our aims. This can be done in various ways (see for example Dekkers, 1993; Mortimer, 1993; Andersson and Bach, 1996; Asoko, 1996; Boohan, 1996), not only using pretest/post-test procedures, but also observing the “learning pathways” of learners throughout the teaching-learning sequence (for example, see Duit, Goldberg & Niedderer, 1992 (Part 3); Arnold & Millar, 1996; Petri & Niedderer, 1998; Welzel, 1998; Aufschnaiter & Welzel, 1999; Psillos & Kariotoglou, 1999). We can find an interesting contribution to such an approach in the framework of “Ingenierie Didactique” (Artigue, 1988). In such an approach, teaching-learning situations are elaborated; taken into account are previous analyses of knowledge to be developed and what is known about learners’ difficulties, conceptions, and ways

M. MÉHEUT

201

of reasoning in the domain. Expected cognitive pathways of students through the proposed learning situations are described. This makes it possible to validate the design of the sequence by observing the learning pathways the students actually develop, and by comparing these actual learning pathways with the expected ones. We will see further how such an approach can give the opportunity of discovering unexpected steps in the possible learning pathways, and to put to the test “local” or sometimes more general hypotheses underlying the design of learning situations. 4. AN EXAMPLE: DESIGNING AND VALIDATING TEACHING-LEARNING SEQUENCES ABOUT PARTICLE MODELS We will now illustrate some aspects of previous considerations by referring to the design and the validation of teaching-learning sequences that we have developed about particle models (Méheut & Chomat, 1990; Méheut, 1997). Designing teaching-learning sequences about particle models We can consider the design of these sequences as rooted in an “integrated constructivist” point of view.

Scientific Knowledge

TEACHER ?

Modelling

STUDENTS Conceptions Ways of reasoning

Material World

Figure 5. Figuring the design of TLS about particle models

202

TEACHING-LEARNING SEQUENCES TOOLS

After taking into account difficulties and misconceptions of students in this field, we returned to the historical development of atomic models. Interweaving both points of view (Figure 5), we proposed a teaching-learning process which consists of putting students in the position of developing particle models by using such models first to explain, then to predict physical events. We tried to put into play the following characteristic features of particle models of matter: their rational rather than their empirical origins, their status as “instruments for thinking” rather than “observable reality”, and their character of mechanical analogy (Méheut, 1998). The aim was to develop models as cognitive tools in order to unify descriptions and then to predict physical phenomena, the models becoming more and more precise in relation with the questions. As a first step, we intended students to interpret physical phenomena as changes of the spatial organization of immutable particles. The kinetic aspects were scarcely brought into play (Méheut & Chomat, 1990). As a second step, we tried to put students in the position of taking into account the kinetic aspects of particulate models in order to explain and predict some thermoelastic properties of gases. To do that, we developed a computer simulation based on the kinetic theory of gases. The activities that we proposed were the elaboration of simulations of several phenomena and the use of these simulations to develop explanations or predictions (Méheut, 1997). Validating these teaching-learning sequences in a research perspective We will now focus our attention on experimenting with the second sequence. 5. SOME METHODOLOGICAL ASPECTS The experiment included several stages. The first stage consisted of interviews of five pairs of students; these interviews were divided into two sequences of nearly three quarters of an hour each, and they were tape-recorded. The second stage was the implementation of a nine hour learning sequence in sixteen second year classes in French secondary schools. We gathered data all along the sequence by collecting written work: nine sheets for each student. The analysis concerned the work of ten randomly-selected students out of each class (i.e. 160 students). In a third stage, two years after we experimented with this learning sequence, we gathered some additional information. We did this for two purposes. The first was to ascertain and clarify some results obtained by the analysis of the data gathered during the interviews and the classroom sequence; the second was to assess the long-term effects of this learning process. Some results The analysis of the data collected during interviews provides information about the way students took into account the variables of the model by developing explanations and predictions in relation to the various phenomena and questions. This allows us to draw up the learning pathways that students followed throughout the teaching-learning situations. Some key steps appear in these learning pathways:

M. MÉHEUT

203

-

Considering not only the action of one sample of gas, but the actions of two samples of gas and comparing them to explain events due to differences of pressure; - Passing from a static point of view (pressure as compression) to a dynamic one (pressure as the result of particle impacts); - Interpreting an increase of temperature as an increase in the speed of particles; - Taking into account not only the frequency but also the “force” of particle impacts to explain and predict pressure phenomena. The analysis of the data collected after the classroom sequences gives information about the short-term and the longer-term effectiveness of this learning sequence. For example, in order to assess the efficiency of the particle model built by students as the outcome of this sequence, we gave students experiments that were a little different from those they simulated during the sequence. For both experiments, the number of students reasoning in a particulate way was a little more than half. Among these, the greatest number (34%) compared the frequencies of impacts. The predictions were correct for a great majority of students (more than 80%). Two years later, students who attended this learning sequence used a particle model more than students who did not attend it (about 20% more). Analysis of the data provided evidence about the hypotheses underlying the choice of phenomena and of the questions. For instance, in choosing the questions, we formulated the hypothesis that students would take into account and compare actions exerted by two systems more for a stopping of an action than for a shifting of it. The analysis of the data are in accordance with this first hypothesis. In choosing the phenomena, we hypothesized that phenomena related to temperature would be more problematic than elastic properties when temperature is not put into play. This second hypothesis can be seen as preliminary to a third one, that the model will seem more useful to students for explaining thermoelastic properties, so then they will use it more than for (studying) elastic properties. The analysis of the data is in accordance with the second hypothesis, but we didn't observe the expected effect related to the third hypothesis. We now consider that the questions we chose were not sufficient for students to feel the need of such a model and use it more than alternative phenomenological types of explanation. We have to remember that a major quality of the particulate model is its unifying power. Establishing this characteristic on a secure basis is a long process! 6. TO CONCLUDE Looking back at the design of these sequences, we consider that it fits in with the “ingenierie didactique” methodological framework (Artigue, 1988). Two components of this framework, the psycho-cognitive and the epistemological ones were developed and integrated into a process not far from an “educational reconstruction” process (Kattman et al., 1995). Referring to an “Ingenierie Didactique” methodology makes it appear that one component is poorly explained, taking into account the institutional constraints. Reference to an “Educational Reconstruction” model makes it clear that, if the cognitive and epistemological

204

TEACHING-LEARNING SEQUENCES TOOLS

meaning of knowledge to be developed by students is precisely situated, more attention should be paid to the pragmatic and social significance of problems and questions to be solved. Experimenting with these sequences, we obtained two kinds of results: the first ones contribute to the characterization, the appreciation, the evaluation, and the global pedagogical effectiveness of the sequences; the second ones provide information about the cognitive pathways learners followed along the sequences. Such information made it possible to put to the test precise hypotheses underlying the design of teaching-learning situations, and to improve the design of these situations. It should be noted that these approaches have quickly diffused in several ways. They are now referred to in the national curriculum, and the information we collected about possible learning pathways is now considered as a piece of "professional knowledge". As such it is used in teacher training to help teachers manage such an "integrated constructivist" approach (Morge, 2003).

REFERENCES Andersson, B. & Bach, F. (1996). Developing new teaching sequences in science: the example of "Gases and their properties". In G. Welford, J. Osborne & P. Scott (Eds.), Research in science education in Europe: current issues and themes (pp.7-21). London: The Falmer Press. Arnold, M. & Millar, R. (1996). Exploring the use of analogy in the teaching of heat, temperature and thermal equilibrium. In G. Welford, J. Osborne & P. Scott (Eds.), Research in science education in Europe: current issues and themes (pp.22-35). London: The Falmer Press. Artigue, M. (1988). Ingéniérie didactique. Recherches en didactique des Mathématiques, 9, 281-308. Asoko, H. (1996). Developing scientific concepts in the primary classroom: teaching about electric circuits. In G. Welford, J. Osborne & P. Scott (Eds.), Research in science education in Europe: current issues and themes (pp.36-49). London: The Falmer Press. Aufschnaiter, S. & Welzel, M. (1999). Individual learning processes: A research programme with focus on the complexity of situated cognition. In M. Bandiera et al. (Eds.), Research in science education in Europe (pp.209-215). Dordrecht: Kluwer Academic Publishers. Boohan, R. (1996). Using a picture language to teach about processes of change. In G. Welford, J. Osborne & P. Scott (Eds.), Research in science education in Europe: current issues and themes (pp.85-99). London: The Falmer Press. Chang, C.-Y. & Barufaldi, J.-P. (1999). The use of a problem-solving-based instructional model in initiating change in students’ achievement and alternative frameworks. International Journal of Science Education, 21, 373-388. Chauvet, F. (1996). Teaching colour: design and evaluation of a sequence. European Journal of Teacher Education 19, 119-134. Dekkers, P. (1993). Effectiveness of practical work in the remediation of alternative conceptions of force with students in Botswana. In P.L. Lijnse et al. (Ed.), European

M. MÉHEUT

205

research in science education: Proceedings of the first PhD Summerschool (pp. 233241). Utrecht: CDβ Press. Dewey, I. & Dykstra, Jr. (1992). Studying conceptual change: constructing new understandings. In R. Duit, F. Goldberg & H. Niedderer (Eds.), Research in physics learning: theoretical issues and empirical studies (pp.40-58). Kiel: IPN. Driver, R. & Bell, B. (1986). Students' thinking and the learning of science: a constructivist view. The school science review, 67, 443-456. Driver, R. & Erickson, G. (1983). Theories in action: some theoretical and empirical issues in the study of students' conceptual frameworks in science. Studies in Science Education, 10, 37-60. Duit, R., Goldberg, F. & Niedderer, H. (1992). Research in physics learning: theoretical issues and empirical studies. Kiel: IPN. Duit, R., Roth, W-M, Komorek, M. & Wilbers, J. (1998). Conceptual change cum discourse analysis to understand cognition in a unit on chaotic systems: towards an integrative perspective on learning in science. International Journal of Science Education, 20, 10591073. Dumas-Carré, A. & Weil-Barais, A. (1998). Tutelle et médiation dans l’éducation scientifique. Bern, Peter Lang. Gilbert, J.K. & Boulter, C. (1998). Learning science through models and modelling. In B.J. Fraser and K.G. Tobin (Eds.), International Handbook of Science Education (pp.53-67). Dordrecht, Kluwer academic Press. Kaminski, W. (1991). Optique élémentaire en classe de quatrième: raisons et impact sur les maîtres d'une maquette d'enseignement. Thèse de doctorat, Université Paris 7. Kariotoglou, P., Koumaras, P. & Psillos, D. (1995). Différenciation conceptuelle: un enseignement d'hydrostatique fondé sur le développement et la contradiction des conceptions des élèves. Didaskalia, 7, 63-90. Kattmann, U., Duit, R., Gropengieber, H. & Komorek, M. (1995). A model of Educational Reconstruction. Paper presented at The NARST annual meeting. San Francisco. Komorek, M. , Stavrou, D. & Duit, R. (2001). Nonlinear Physics in Upper Physics Classes: Educational Reconstruction as a Frame for Development and Research in a Study of Teaching and Learning Basic Ideas of Nonlinearity. In D. Psillos et al. (Eds.), Proceedings of the Third International Conference on Science Education Research in a Knowledge Based Society (pp.483-485). Thessaloniki: Art of Text. Leach, J. & Scott, P. (2002). Designing and evaluating science teaching sequences: An approach drawing upon the concept of learning demand and a social constructivist perspective on learning. Studies in Science Education, 38, 115-142. Lemeignan, G. & Weil-Barais, A. (1988). Etude de quelques activités de modélisation. In G. Vergnaud, G. Brousseau, M. Hulin (Eds.), Didactique et acquisition des connaissances scientifiques (pp.229-244). Grenoble: La Pensée Sauvage. Lemeignan, G. & Weil-Barais, A. (1992). L'apprentissage de la modélisation dans l'enseignement de l'énergie. In Equipe INRP-LIREST (Eds.), Enseignement et apprentissage de la modélisation en sciences (pp.171-232). Paris: INRP. Méheut, M. & Chomat, A. (1990). The bounds of children atomism; an attempt to make children build up a particulate model of matter. In P.L. Lijnse et al. (Eds.), Relating macroscopic phenomena to microscopic particles. (pp.266-282). Utrecht: CDβ Press. Méheut, M. (1997). Designing a learning sequence about a pre-quantitative model of gases: the parts played by questions and by a computer-simulation. International Journal of Science Education, 19, 647-660.

206

TEACHING-LEARNING SEQUENCES TOOLS

Méheut, M. (1998). Designing learning sequences about pre-quantitative particle models. In A. Tiberghien, E.-L. Jossem & J. Barojas (Eds.), Connecting Research in Physics Education with Teacher Education http://www.physics.ohio-state.edu/~jossem/ICPE/BOOKS.html. Méheut, M. & Psillos, D. (org.) (2000) Designing and validating teaching-learning sequences in a research perspective. Paris. Minstrell, J. (1992). Facets of students' knowledge and relevant instruction. In R. Duit, F. Goldberg and H. Niedderer (Eds.), Research in physics learning: theoretical issues and empirical studies (pp.110-128). Kiel: IPN. Morge, L. (2003). Les connaissances professionnelles locales: le cas d’une séance sur le modèle particulaire. Didaskalia, 23, 101-132. Mortimer, E.F. (1993). The evolution of students' explanations for physical state of matter as a change in their conceptual profile. In P.L. Lijnse et al. (Eds.), European research in science education: Proceedings of the first PhD Summerschool (pp.281-287). Utrecht: CDβ Press. Nikolopoulou, K. (1993). An investigation into the effect of I.T. on pupils' understanding of some science concepts and processes. In P.L. Lijnse et al. (Eds.), European research in science education: proceedings of the first PhD Summerschool (pp.206-214). Utrecht: CDβ Press. Nussbaum, J. & Novick, S. (1982). Alternative frameworks, conceptual conflict and accommodation: toward a principled teaching strategy. Instructional Science 11, 183200. Nussbaum, J. (1989). Classroom conceptual change: philosophical perspectives. International Journal of Science Education 11, 530-540. Petri, J. & Niedderer, H. (1998). A learning pathway in high-school level quantum atomic physics. International Journal of Science Education,20, 1075-1088. Psillos, D. (1998). Teaching introductory electricity. In A. Tiberghien, E.-L. Jossem & J. Barojas (Eds.), Connecting Research in Physics Education with Teacher Education. http://www.physics.ohiostate.edu/~jossem/ICPE/BOOKS.html. Psillos, D. & Kariotoglou, P. (1999). Teaching fluids: intended knowledge and students’ actual conceptual evolution. International Journal of Science Education 21, 17-38. Psillos, D. & Méheut, M. (coord.) (2001). Teaching-learning sequences as a means for linking research to development. In D. Psillos et al. (Eds.), Proceedings of the Third International Conference on Science Education Research in the Knowledge Based Society (pp.226-241). Thessaloniki, Art of Text. Ravanis, K. & Papamichael, Y. (1995). Procédures didactiques de déstabilisation du système de représentations spontanées des élèves pour la propagation de la lumière. Didaskalia 7, 43-61. Robardet, G. (1995). Situations problèmes et modélisation; enseignement en lycée d'un modèle newtonien de mécanique. Didaskalia, 7, 131-143. Schwedes, H. & Schmidt, D. (1992). Conceptual change and theoretical comments. In R. Duit, F. Goldberg and H. Niedderer (Eds.), Research in physics learning: theoretical issues and empirical studies (pp.188-202). Kiel: IPN. Tiberghien, A., Psillos, D. & Koumaras, P. (1995). Physics instruction from epistemological and didactical basis. Instructional Science,22, 423-444. Tsoumpelis, L. (1993). Explications et modèles dans des situations a-didactiques en sciences physiques: le cas de la concentration molaire. Thèse de doctorat. Université Lyon 1.

M. MÉHEUT

207

Viennot, L. & Rainson, S. (1999). Design and evaluation of a research-based teaching sequence: the superposition of electric fields. International Journal of Science Education, 21, 1-16. Welzel, M. (1998) Emergence of complex cognition during a unit on static electricity. International Journal of Science Education, 20, 1107-1118.

DESIGNING AND EVALUATING SHORT SCIENCE TEACHING SEQUENCES: IMPROVING STUDENT LEARNING JOHN LEACH, JAUME AMETLLER, ANDY HIND, JENNY LEWIS, PHILIP SCOTT The University of Leeds, UK

ABSTRACT This paper reports a study designed to provide evidence about the feasibility of designing short teaching sequences, based on insights from research and scholarship on teaching and learning science, which are measurably better at promoting conceptual understanding amongst students than the teaching approaches usually used by their schools. The research team worked in collaboration with a group of 9 teachers (3 biology, 3 chemistry, 3 physics) to design, implement, and evaluate 3 teaching sequences for use with students aged 1115. The physics and biology teaching sequences were also implemented by other teachers (11 and 5 respectively) not involved in their design. Teachers implemented the physics and biology teaching sequences in ways broadly consistent with the planned approach. In all cases where a valid comparison can be made, students’ responses to diagnostic questions requiring the use of conceptual models to construct explanations were significantly better following the designed teaching sequences, than the responses of comparable students following the school’s usual teaching approach. The significance of these findings for research in science education, and for policy and practice relating to science teaching, are discussed.

1. THE PROBLEM ADDRESSED IN THIS PAPER The literature on students’ learning of scientific concepts is extensive (Pfundt & Duit, 2001). However, the impact of this research on the practices of day-to-day science teaching has not been great (Duit & Treagust, 1998). Furthermore, some are sceptical as to whether teaching based on information about students’ existing knowledge leads to gains in students’ understanding (e.g. Matthews, 1997). Although there are some studies in the literature that do provide evidence of improvements in student learning against specified goals, following researchinformed teaching interventions (for example, Brown & Clement, 1991; Tiberghien, 2000; Viennot & Rainson, 1999), such studies generally say rather little about the role of the teacher in implementing the teaching. Furthermore, the teacher in these studies has often worked very closely, over an extended period of time, with the research team. There is little or no evidence that teachers less closely involved with the research process can replicate the improvements in student learning. 209 K. Boersma et al. (eds.), Research and the Quality of Science Education, 209—220. © 2005 Springer. Printed in the Netherlands.

210

DESIGNING AND EVALUATING SHORT SCIENCE TEACHING SEQUENCES

The study reported in this paper was designed to provide evidence about the feasibility of improving student learning against specified curriculum goals, when the design of the teaching is informed by insights from research on students’ learning. The study consists of two phases. During the Development phase, groups of teachers and researchers worked together to design, implement, and evaluate short teaching sequences. During the Transfer phase, two of the sequences were implemented by teachers not involved in the design of the teaching sequence. An important aspect of the evaluation of both phases addressed the extent to which students following the designed teaching approach attained a richer understanding of the target conceptual content, compared to other students of the same ability following the school’s usual approach. 2. DESIGN AND METHODOLOGY Three short teaching sequences were designed, implemented, and evaluated. Each sequence was prepared by a group of three teachers and university-based researchers working together. The teaching sequences were aimed primarily at pupils aged between 11 and 14 and lasted for around 6 hours. The schemes focused upon introductory ideas about plant nutrition, the process of modelling change in terms of a simple particle model of matter, and introductory ideas about electric circuits. These areas were selected on the grounds that there is a significant body of empirical research on students’ learning in each area, together with studies describing the design and evaluation of teaching approaches. The overall shaping of the teaching sequences was informed by a social constructivist perspective on learning (Driver et al., 1994; Leach & Scott, 2003), with particular attention being given to the different communicative approaches (Mortimer & Scott, 2002; Mortimer & Scott, 2003) to be taken by the teacher in promoting learning. In addition, an analysis of the particular learning demands (Leach & Scott, 2002) was made for each of the topic areas, drawing on research evidence about students’ learning in those areas; instructional activities were planned to address those learning demands. Each participating teacher then implemented the teaching sequence with at least one class. The implementation of the teaching sequences was evaluated using multiple data sources. Students’ learning against specified goals was measured by comparing responses to diagnostic questions set prior to teaching, immediately after teaching, and after a delay of several weeks.1 Students of classes who had followed the school’s regular teaching approach were evaluated with the same instruments in order to provide baseline information. A pre-test was used to establish the comparability of the case study and the baseline groups. Pre-test and post-test questionnaires were not identical, as in some

1

The results of the delayed post-test are not, however, reported as there was evidence in some groups that further relevant teaching had taken place.

J. LEACH, J. AMETLLER, A. HIND, J. LEWIS, P. SCOTT

211

cases it was judged to be inappropriate to include questions in the pre-test addressing technical content not yet encountered by students. The questions tend to be in two parts. The first part involves students in making a prediction of some kind (the behaviour of lamps in a simple circuit, the mass of a solution on dissolving). This is followed by an opportunity for students to explain their prediction. Students’ responses were coded according to whether they made a correct prediction or not, and the extent to which their explanation drew upon the target conceptual content of the teaching sequence. Responses were coded by one researcher, and a subset of responses was independently coded to assess inter-rater reliability. Modifications to coding were then made to the whole data set where appropriate. All the teachers were interviewed after the implementation of the new sequences. In addition, video- and audio-recordings of all classes were used to analyse the ‘staging’ of the lessons by each teacher. This analysis was made in terms of four classes of communicative approach, derived from categorising the teacherstudent interactions along each of two dimensions: interactive/non-interactive and dialogic/authoritative (Mortimer & Scott, 2003). The video and audio records were also used to make a record of the sequence in which scientific ideas were introduced during the teaching, to establish the extent to which the teacher followed the planned teaching sequence. The results of these analyses will be reported elsewhere, though they are mentioned here where relevant in explaining findings about student learning. Sample Nine teachers worked with us on the Development Phase of the project. All were at the early or middle stages of their careers, with only 1 holding a significant middle management position in a science faculty. Six are female and 3 are male. The teachers were selected on the grounds that they are viewed by us and their peers as being enthusiastic and able practitioners, whilst having no special training in science education research. Their schools are located in a variety of different communities in the North of England, ranging between inner-city multicultural, and affluent suburban. The 11 teachers involved in the Transfer Phase (5 biology, 6 physics), and their schools, had a similar profile to those of the Development Phase. These teachers did not receive any specific training on using the sequences beyond a brief introduction to the new material. They volunteered to participate and do not have any special relationship with the research team or the University of Leeds. Information about the sample is presented in Table 1:

212

DESIGNING AND EVALUATING SHORT SCIENCE TEACHING SEQUENCES Table 1: Sample information

Development Phase

Biology

Chemistry

Physics

BDV BDC BDS CDL CDA CDS1 CDS2 PDA PDD1 P2D2 PDS

Case Study Class1 n= 27 28 29 28 20 18 17 20 29 28 23

Baseline Class n= 27 30 26 n.a.2 n.a.2 n.a.2 n.a.2 20 22 22 26

Code

Biology Transfer Phase

Code

Physics

BTC BTF BTP BTT BTL PTM PTJ PTL PTR PTT PTK

Case Study Class1 n= 26 26 29 22 27 24 26 26 25 21 26

Baseline Class n= 21 28 28 29 19 29 n.a.2 23 23 23 23

n refers to the number of students responding to the diagnostic questions after the implementation (post-test). There might be some differences between pre-test and post-test (see tables 2 and 3). No baseline information was available for the chemistry teaching sequence. None of the 3 schools usually addressed the modelling of physical and chemical change in any one teaching unit; rather, the teaching was spread across several units.

The design of the teaching sequences The design of the three teaching sequences followed the same steps: 1. Content analysis of the topic, paying special attention to the requirements of the official curriculum for the intended schooling level. 2. Review of the literature on teaching and learning the contents detailed in the first step, focusing on well established research on the patterns of students’ reasoning. 3. Identification of the learning demands (Leach & Scott, 2002) by assessing the gaps between students’ conceptualisation of a specific concept, as found in the research, and the school science perspective on the same concept. 4. Explicit teaching goals of the sequence can be drawn from the learning demands identified in step 3. This provides a way of operationalising a large corpus of research into the design of the teaching sequence by choosing learning goals that focus on helping students overcome the learning demands posed by the topic as analysed in the first step. 5. Design of teaching activities to address the teaching goals of the sequence. The activities have to be designed keeping in mind the whole structure of the lesson and the sequence, and the practicalities of its implementation in the schools. The resulting sequence has to be coherent, workable, and feasible within the time normally available to the topic. Explicit attention was paid to the kinds of interaction between teacher and students as teaching activities were in progress. Thus, in some situations the intention might be that the teacher introduces new ideas. At other times an

J. LEACH, J. AMETLLER, A. HIND, J. LEWIS, P. SCOTT

213

activity might be planned with the central aim of providing an opportunity for students to talk through the ideas for themselves with the teacher probing and supporting that discussion. Each situation benefits from a specific kind of communicative approach. These different kinds of talk were explicitly referred to and highlighted in the teaching scheme with a set of simple icons developed with the teachers of the development phase. 6. The teaching sequence began and ended with students completing diagnostic questions. University-based staff introduced relevant findings from research and scholarship in science education to the design process. In addition, both school- and university-based staff made contributions based upon professional knowledge about developing science lessons that would be effective and well-received by students. The chemistry teaching sequence was the first to be produced and implemented. On implementation, it was found to have significant weaknesses, mainly concerning the extent to which teaching activities had been developed and embedded into workable whole lessons that were motivating for students. It was therefore decided not to use it in the transfer phase, and it will not be discussed in this paper. Drawing upon our experience of the shortcomings of the chemistry teaching sequence, the design and presentation of the physics and biology teaching sequences was modified considerably, and more time was spent to ensure that teaching and learning activities were embedded into workable whole lessons. Furthermore, more explicit guidance about the communicative approach was built into the physics and biology teaching sequences than had been used in the chemistry sequence. 3. FINDINGS In this section we present findings from the pre- and post-tests completed by students following the designed teaching sequences, compared with baseline information from students in comparable classes in the same schools following the school’s usual programme of instruction (where such information was available). Findings from both the Development and Transfer Phases are reported. Data are presented about the number and percentage of students making correct and incorrect predictions, and the number and percentage using an explanatory model consistent with its presentation in the teaching, or any of the following: using an incomplete but consistent model, not using the model (or using a model inconsistent with that introduced in the teaching), or providing no/other responses. The results of the students’ answers have been averaged across the total number of questions included in each test. The teaching sequences, and associated diagnostic questions, can be found at: http://www.education.leeds.ac.uk/research/scienceed/epse_teach_resources.htm Analysis of the video records of the lessons indicates that the conceptual content of both the physics and biology teaching sequences was followed quite closely by

214

DESIGNING AND EVALUATING SHORT SCIENCE TEACHING SEQUENCES

teachers and, furthermore, some features of the planned communicative approach were used by them. Evaluation of the biology teaching sequence Table 2 shows students’ results in pre- and post-tests, where information is available. Table 2: Evaluation of students’ learning following implementation of the biology teaching sequence

On the basis of the information in Table 2, we make the following claims: There is no strong evidence for different levels of understanding of plant nutrition between students in case study and associated baseline groups prior to

•

J. LEACH, J. AMETLLER, A. HIND, J. LEWIS, P. SCOTT

•

•

•

•

•

215

teaching (χ2; p>0.01)2. This claim applies to all case studies in both Development and Transfer Phases (though pre-test baseline data are not available for the BDS and BTL case studies). There is no evidence that students in the case study groups were better at making predictions about plant nutrition than those in the baseline groups following teaching (χ2; p>0.01 in all cases). However, there is evidence that in all cases except one, (BTP) students in the case study groups immediately after teaching drew upon a scientifically consistent model of plant nutrition more often than did students in the baseline groups (χ2; p<0.001 in all cases but BTL). The results for the development phase case studies show an increase of the percentage of consistent answers between baseline and case studies ranging from 23 to 38.4%. The percentage of Baseline groups' answers drawing on no aspects of the model ranged between 23.7% and 35.9% higher than their case study groups. In BDV, only 38.9% of students gave explanations that used no aspects of the taught model of plant nutrition, compared with 66.3% of students in the baseline group. Only 7.6% of the responses by the baseline group drew on most aspects of an appropriate model compared with 30.6% from the developmental group. In BDC, on average, 35.3% of the baseline group gave explanations that used no aspects of an appropriate scientific model compared with 11.6% in the developmental case study group. Only 25.9% of the responses by the baseline case study group drew on most aspects of an appropriate model compared with 64.3% from the developmental group. Only 10.3% of students in BDS gave an explanation that used no aspects of an appropriate scientific model, compared with 46.2% of the baseline group. Only 38.5% of the responses by the baseline group drew on most aspects of an appropriate model compared with 74.1% from the case study group. The same trend can be seen for the Transfer Phase case studies, though the difference between students’ results in the baseline and case study groups are smaller than for the Development Phase. For case study BTP, there is evidence that the baseline group students were significantly better at constructing explanations of plant nutrition based upon concepts about photosynthesis in the pre-test, and that the groups were not therefore comparable prior to teaching.

2

In order to calculate χ2 students’ responses for each question were totalled. This process provided samples large enough to be considered independent. In the case of BDV, the null hypothesis is ‘There is no difference between the way in which students use a model of plant nutrition in the baseline and case study groups prior to teaching. Using the following data, the probability that the null hypothesis is true is 0.3519: Pre-test answers BDV observed BDV expected Baseline observed Baseline expected Totals

Consistent 19 21 23 21 42

Inconsistent 15 17.5 20 17.5 35

Other 47 42.5 38 42.5 85

Totals 81 81 162

216

•

DESIGNING AND EVALUATING SHORT SCIENCE TEACHING SEQUENCES Although the case study students in BTP achieved results similar to those of other Transfer Case Study groups, their results were not as good as those of the baseline students. Post-test responses from students in the remaining case study groups, where it is possible to make a comparison, were between 9.4% and 13.1% higher than students in the corresponding baseline group. However, the number of students whose responses drew partially upon models introduced in teaching was slightly higher than in the Development Phase. In the Development Phase case studies, the percentage difference between experimental and baseline ranged from –14.7% to 4.5%, the range was between 5.3% and 22.7% for the Transfer Phase. These results suggest that the increase in the use of the scientific model between experimental and baseline groups in the Transfer Phase is similar to that encountered in the Development Phase, though the latter were significantly more successful at using a complete, consistent model. Post-implementation interviews with the biology teachers in the Development Phase suggested that all three reacted positively to their experience of implementing the teaching sequence. All the teachers stated that they felt that the teaching sequence had enabled them to address curriculum content in a better way than their usual approach.

Evaluation of the physics teaching sequence Table 3 shows students’ results in the pre- and post-tests, where information is available.

•

•

•

On the basis of the information in Table 3, we make the following claims: There is no strong evidence for different levels of understanding of electric circuits between students in case study and baseline groups for both development and transfer phase schools prior to teaching (χ2; p>0.01 in all cases for explanations and predictions, with the exception of predictions in PDD1 and PDD2). There is evidence that students in the case study groups were better at making predictions about the behaviour of electric circuits than those in the baseline groups after teaching (χ2; p<0.01 in all cases except for PDS and PTR). Nevertheless, these results should be treated cautiously as the percentage of correct answers was above 70% in all groups, and the percentage increases observed were quite small (between 6% to 17 %). There is evidence that students in each of the case study groups, immediately after teaching, drew upon a scientifically consistent model of the behaviour of electric circuits more often than students in the baseline groups (χ2; p<0.001 for all groups). Even though all groups’ predictions improved by a similar amount after teaching, it was striking that students’ explanations in the three baseline groups made use of fewer elements of the taught model after teaching than in the pre-test. By contrast, the case study students’ explanations made a wider use of elements of the model after teaching.

J. LEACH, J. AMETLLER, A. HIND, J. LEWIS, P. SCOTT

217

Table 3: Evaluation of students’ learning following implementation of the physics teaching sequence

•

•

Explanations coded as Consistent account for less than 1% of the answers of the baseline group students after teaching, while ranging from 4.3% to 28.3% of the development case groups students’ answers, and 3.9% to 12.7% of the transfer phase answers. In addition, the percentage of answers coded as Incomplete increased by between 18.9% and 37.1% in the Development Phase case studies, and between 13.5% and 25.1 % for the Transfer Phase case studies. These results are remarkably consistent across both phases, with a slightly better performance from the Development Phase case study groups. In common with the biology teaching sequence, post-implementation interviews with the physics teachers in the Development Phase suggested that all three reacted positively to their experience of implementing the teaching sequence, stating that it had enabled them to address curriculum content in a better way than their usual approach.

218

DESIGNING AND EVALUATING SHORT SCIENCE TEACHING SEQUENCES 4. DISCUSSION

Building insights from research into teaching materials to improve learning In all the 13 case studies with a comparable baseline (Tables 2 and 3), the performance of students following the designed teaching sequence was significantly better than the performance of baseline students. This leads us to conclude that it is possible to design short science teaching sequences, informed by research and scholarship on teaching and learning science, which can be used by teachers who may or may not have been involved in the design of the sequences, and which can result in significant improvements in students’ conceptual understanding in science. The teaching sequences designed in this project were informed by empirical information about students’ likely pre-instructional understanding in the given concept area (as well as insights about some of the conceptual difficulties that many students encounter as a result of teaching), and by insights about the nature of effective classroom communication between teacher and students. They were also informed by rich professional knowledge about teaching. We have written elsewhere about how such insights were used in the design of teaching (Leach & Scott, 2002). It is important to note, however, that our strategy for improving student learning did not involve extended periods of training for teachers on the theoretical underpinnings drawn upon to inform the teaching. Rather, we attempted to guide teachers’ actions by building insights about students’ likely responses to content, and a communicative approach, into materials in a format that is very familiar to UK teachers. We do not believe that it is possible to control teachers’ actions by scripting lessons. We view teaching expertise as involving a complex mixture of skills, including skills in the treatment of content and skills in developing relationships with students as individuals and groups. The knowledge drawn upon for teaching is often tacit and cannot therefore be scripted in advance (Bransford et al., 2000). Our findings in relation to the chemistry teaching sequence also suggest that attempts to make teachers use an unfamiliar approach, without working the approach through in an example, does not enhance teaching skills. Rather than viewing the teaching sequences as scripts, we view them as planning maps which indicate to teachers some of the critical details (Viennot, 2003) of conceptual content in a topic area and ways of dealing with that content which are embedded in a worked example. Teachers are left with choices to make, and the capacity to ‘make the teaching their own’, while central aspects of conceptual content and communicative approach are clearly highlighted for them. It is, of course, possible that the differences in student learning reported between the case study and control groups are due to factors other than the extent to which the teaching was informed by insights and evidence from research on teaching and learning science. We think that there are two main possibilities. Firstly, testing bias: it is likely that our test instruments were biased towards the content of the designed teaching, compared to the school’s normal teaching. However, our test instruments do address conceptual content identified in the English national

J. LEACH, J. AMETLLER, A. HIND, J. LEWIS, P. SCOTT

219

curriculum. To that extent we would defend the use of the instruments to provide evidence of improvements in student learning against specific curriculum goals. The second possibility is that the students and teachers in case study and baseline groups were not comparable. The pre-test data do not provide evidence to support this hypothesis. However, it is possible that the teachers with whom we worked may well achieve better results amongst their students against stated curriculum goals than the teachers teaching the baseline groups. However, we are encouraged by the fact that in all of the case studies where a comparison can validly be made, students following the designed teaching sequences perform significantly better on diagnostic questions that test for conceptual understanding than do their peers following the school’s usual teaching. In addition, there are no significant differences in students’ performance on diagnostic questions requiring factual recall. We are also encouraged by the enthusiasm of the teachers we have interviewed, to adopt the physics and biology teaching sequences, because they are viewed as better at meeting curriculum content objectives and more enjoyable for students, than the usual approaches used. Our findings pose a challenge to Matthews’ (1997) claim that there is no evidence that science teaching informed by insights from research on science learning is better at promoting conceptual understanding amongst students. Rather, our findings us to suggest that science curriculum development and accompanying professional development programmes for science teachers, focussing on the teaching of key scientific concepts and informed by research, can result in improvements in students’ conceptual understanding. Furthermore, a potentially successful method of disseminating, to teachers, the results of research on science teaching and learning involves transforming those findings into workable practices for the classroom. The potential of this method is reinforced by the fact that transfer case teachers obtained significant improvements in their classes without any teacher training. Assessing science learning In this study, responses requiring factual recall were no better from students following the designed teaching sequences, than the responses from baseline students. However, the same students’ responses were measurably better on questions requiring the use of conceptual models to construct explanations compared to baseline students. The external assessment questions used for students aged 14 and 16 in England are widely criticised as testing factual recall rather than underlying conceptual understanding. Our findings suggest that students following the designed teaching sequences were just as capable as students following schools’ normal teaching programmes at completing such questions, and were also better at questions requiring the use of conceptual models. Our findings lead us to believe that the use of such assessment questions may result in national assessment data over-estimating students’ understanding. Furthermore, the use of such assessment questions may

220

DESIGNING AND EVALUATING SHORT SCIENCE TEACHING SEQUENCES

well result in teachers under-estimating their pupils’ capacity to undertake work requiring them to use scientific conceptual models. ACKNOWLEDGEMENTS The Evidence-based Practice in Science Education Research Network was funded through the Economic and Social Research Council’s Teaching and Learning Research Programme. The authors thank other members of the EPSE network for valuable discussions on the work reported in this paper. REFERENCES Bransford, J. D., Brown, A. L. & Cocking, R. R. (2000). How people learn: brain, mind, experience, and school. Washington, D. C.: National Academy Press. Driver, R., Leach, J., Scott, P. & Wood-Robinson, C. (1994). Young people’s understanding of science concepts: implications of cross-age studies for curriculum planning. Studies in Science Education, 24, 75-100. Duit, R. & Treagust, D. (1998). Learning in Science: From Behaviourism to Social Constructivism and Beyond. In B. Fraser & K. Tobin (Eds.), International Handbook of Science Education. Dordrecht, NC: Kluwer Academic Publishers. Leach, J. & Scott, P. (2002). Designing and evaluating science teaching sequences: An approach drawing upon the concept of learning demand and a social constructivist perspective on learning. Studies in Science Education, 38, 115-142. Leach, J. & Scott, P. (2003). Learning science in the classroom: Drawing on individual and social perspectives. Science and Education, 12(1), 91-113. Matthews, M. (1997). Introductory Comments on Philosophy and Constructivism in Science Education. Science and Education, 6(1), 5-14. Mortimer, E.F. & Scott, P.H. (2002). Discursive activity on the social plane of high school science classrooms: a tool for analysing and planning teaching interactions. Paper presented at the AERA Annual Meeting as part of the BERA invited symposium: Developments in Sociocultural and Activity Theory Analyses of Learning in School, New Orleans, USA. Mortimer, E.F. & Scott, P.H. (2003). Meaning making in science classrooms. Milton Keynes: Open University Press. Pfundt, H. & Duit, R. (2001). Bibliography: Students’ Alternative Frameworks and Science Education (fifth ed.). IPN: Kiel. Viennot, L. (2003). Teaching physics. Dordrecht. NL: Kluwer Academic.

DISCUSSING A RESEARCH PROGRAMME FOR THE IMPROVEMENT OF SCIENCE TEACHING BJÖRN ANDERSSON, FRANK BACH, MATS HAGMAN, CLAS OLANDER, ANITA WALLIN Göteborg University, Sweden

ABSTRACT A research programme for the improvement of science teaching is described, exemplified, and discussed. Briefly, the idea of the programme is that researchers in science education and teachers in schools should work together to design teaching sequences and to assess how they function in practice. Research results concerning pupils’ everyday conceptions, as well as analyses of the conceptual structure of a given area and the reasons for teaching it, play an important role when working on a design. The most important product of the design phase is a detailed guide for teachers, which we look upon as a tool for further knowledge-building. In our paper we suggest that the idea of domain-specific theories is worth examining and developing. It might contribute to strengthening science education as an autonomous discipline.

1. PROBLEM Research programmes are motivated and conducted according to various aims, e.g. developing and testing a theory, developing and applying a method, or attempting to solve a practical problem. Research is said to be theory-driven, method-driven, or problem-driven. A common conception of the relation between research driven by theory and that driven by problems is that the former is more important and a prerequisite for the latter. Acquired theoretical insights are assumed to generate applied research that can lead to solutions of practical problems. In an analysis made by the National Research Council in the USA, however, this model is judged to be too limiting when it comes to setting up an educational research agenda (Bransford, Brown & Cocking, 2000). They emphasise research based on practical problems as a significant alternative model. These problems are made the subject of research, generating both results of practical use and contributions to the development of educational science. Current examples are the various design experiments completed or still going on in the USA (Kelley, 2003) and similar work in Europe, e.g. 'developmental research' (Lijnse, 1995) and 'teaching-learning sequences' (Méheut & Psillos, 2004). Our research programme is problem-driven. We wish to help solve an urgent school problem, namely the lack of scientific understanding among the majority of pupils evident from assessments and other investigations in Sweden and abroad. If 221 K. Boersma et al. (eds.), Research and the Quality of Science Education, 221—230. © 2005 Springer. Printed in the Netherlands.

222

DISCUSSING A RESEARCH PROGRAMME

one does not understand, one tends to lose interest, and this is probably a contributory factor in the diminishing recruitment to science courses. An obstacle to progress is that results from science education research that could help to improve teaching do not reach the teachers. 2. OVERVIEW OF THE RESEARCH PROGRAMME In Sweden there is no well developed tradition of systematically making use of the practical experience of teachers, while some countries base much of their school development on this (Stigler & Hiebert, 1999). Nor can it be said that we use research results in a systematic way to improve practice. Such results are expressed in scientific papers, and are often at a high level of abstraction. It takes time to understand them, and therefore they are difficult for a teacher to put into practice. Lijnse (2000) even goes so far as to declare that this is an impossible task for the teacher. This probably influences how teachers perceive the legitimacy of educational science. The step from research results to practice is thus far from trivial. However, we think that it is possible to utilise both teachers’ experience and research results to improve science teaching. Research-based changes in teaching practice cannot be forced upon teachers, but may be achieved by researchers working together with practising teachers (Baird & Northfield, 1992). An element of our research programme is, therefore, the cooperation between teachers and researchers. The strategy we have selected involves developing teaching sequences during a 'design phase', followed by a 'trial phase' in which the function of the sequence is studied in practice. The results obtained are used to improve the design. The design process is creative work that does not necessarily follow a definite plan, but there are systematic elements. A number of aspects based on available research results and well-tried experience need to be taken into account and analysed during the work. These aspects are summarised on the periphery of the circle in Figure 1. Allowing the aspects to interact can generate new ideas and new insights. Our experience is that interactions between knowledge of pupils’ everyday conceptions and their difficulties in understanding (the pupil's’ 'starting-point'), on the one hand, and insights into the scientific content ('the character of the area'), on the other, can be productive. The main result of the design phase is a set of goals and a draft for a teaching sequence. In addition to descriptions of lessons, it may comprise texts for pupils to read, problems to discuss, computer simulations, and so forth. All this is written down in the form of a teacher guide, which also includes background information, e.g. about the character of the area, reasons for teaching it, and what is known about the pupils' starting point. We look upon the guide as a tool for further knowledge building. When revised and published, it will be a major result of our work. The next step is to try the sequence under regular school conditions, evaluate it, and use the results achieved to revise the original draft. The process can be repeated several times. There are a number of foci that can be set up for the research during the trial

B. ANDERSSON, F. BACH, M. HAGMAN,. C. ORLANDER, A. WALLIN

223

phase. One is what the pupil has experienced and what s/he has learnt about the given area. There are several studies which give us reason to feel a certain optimism about this important aspect they show that pupils can achieve good long-term conceptual understanding (e.g. Klaassen, 1995; Lijnse, 1995; Tiberghien, 1997; Viennot & Rainson, 1999; Bach, 2001; Wallin, Hagman & Olander, 2001). Another focus is to study different interactions when they are taking place, e.g. between pupils in different groups and between teachers and pupils. Our interest is studying content-specific aspects by trying to create situations that arouse interest and invite further conceptual development, e.g. small-group problem-solving, through discussions. A third focus is the teacher and his/her experiences of the teacher guide, the teaching, and the coaching that has occurred. An interesting perspective is how the teacher’s competence in teaching the content area in question might develop in the course of the work. Relatively few studies about this are reported (e.g. Summers & Kruger, 1994; van Driel, Verloop & de Vos, 1999; Jones, Carter & Rua, 1999).

teachingsequence

goals

teaching

evaluation

TRIAL PHASE

DESIGN PHASE

reasons for teaching the area theories & results of learning and teaching

character of the area

teaching praxis

selection of content

pupil's starting-point

curriculum & syllabuses

external circumstances

teacher's starting-point

Figure 1. Aspects of the development and study of teaching sequences

224

DISCUSSING A RESEARCH PROGRAMME

So far, we have completed three teaching sequences. The areas are 'a qualitative particle model for gases' (Andersson & Bach, 1996), 'the theory of evolution' (Hagman, Olander, & Wallin, 2002; Wallin, Hagman & Olander, 2001) and 'geometrical optics' (Bach, 2001). 3. THEORETICAL CONSIDERATIONS General theoretical platform Our general theoretical platform currently comprises the following features: A constructivist view of learning and knowing. We take a constructivist view of the knower-known relation as described by Furth (1969). The educational relevance of constructivism has been well expressed by Ogborn (1997, p. 131), e.g. that the design of teaching should give high priority to making sense to pupils, capitalising on and using what they know, and addressing difficulties that may arise from how they imagine things to be. Communication is necessary for the development of knowledge. Another theoretical approach is the sociocultural one. A representative of this approach who is interested in science teaching is Lemke (1990). He makes a fair number of recommendations for teaching, stressing small group discussion and science writing. We agree with the recommendations made by Ogborn and Lemke. One may think that they are neither new nor original but belong to long established pedagogical wisdom, but the fact that they emanate from various theoretical approaches enriches their meaning in different ways. Increased awareness of learning and teaching through formative assessment. Some relatively extensive scientific documentation shows that if formative assessment is done consciously and systematically, teaching and learning can be improved (Black & Wiliam, 1998). We therefore stress formative assessment. Emphasis on understanding and applying basic concepts and theories. A person who understands tends to be interested. He or she has also achieved independence in relation to the thing understood, compared to knowledge that has just been memorised. Problems with general statements Although the recommendations above are good ones, they are not sufficient for solving our problem. Take, for example, teaching about light and its properties. Teachers and researchers ask themselves questions like: – How do pupils conceive light, and what does this mean for instruction? – How can one motivate pupils to discuss and write about light and its properties? Answering these and similar questions is not a trivial task. No general theoretical approaches succeed on their own to accomplish this task. The answers must be sought in combination with content-specific research. This is the case for any area of science teaching. We agree with Lijnse (2000) that this kind of work is 'the forgotten

B. ANDERSSON, F. BACH, M. HAGMAN,. C. ORLANDER, A. WALLIN

225

dimension of science education research'. We wish to go one step further by asserting that the knowledge that is generated through content-specific research both defines and motivates its own knowledge domain within the educational sciences, namely 'didactics of science subjects'. Domain-specific theories – a new way of looking at teaching sequences From the problems with general statements, it is perhaps not such a big step to a new way of looking at the significance of developing teaching sequences. It is not only a question of applying general theories, but it is also one of theory construction. We try to work as follows: If research results and analyses about students' preconceptions and learning possibilities are available in the area of interest, we make a synthesis of this knowledge in the form of a list of conditions which we think will improve students' learning with understanding. This list, which is also informed by our understanding of the scientific content, is regarded as a tentative domain-specific theory which complements general theoretical statements. The conditions stated in the theory are concretized when the teaching sequence is designed. Thus, developing a teaching sequence is not primarily a question of finding a good recipe for teaching but a scientific knowledge-building process. A similar view is put forward by Cobb et al. (2003): Design experiments are conducted to develop theories, not merely to empirically tune 'what works'. These theories are relatively humble in that they target domain-specific learning processes.

The idea of domain-specific theories is new to science education. We think it is worth examining and developing. 4. A DOMAIN-SPECIFIC THEORY FOR TEACHING GEOMETRICAL OPTICS We now give some examples of how our research programme has been applied in the area of geometrical optics; we begin with the domain-specific theory. On the basis of our understanding of the scientific content and of the research literature on pupils' conceptions of light and its properties (see e.g. Driver et al., 1994, pp. 41-45 and 128-133), this is how the theory is formulated: If the following aspects are taken into consideration when teaching geometrical optics, pupils' prospects of developing understanding will improve: • From the beginning, a need for the key idea of optics is created, namely, that light exists and propagates along straight lines between sources and effects. • From the beginning, pupils are offered opportunities to use the key idea of optics as a tool to explain real world phenomena, such as the size and shape of shadows and illuminated areas. • The teaching clarifies that light propagating between sources and effects cannot be seen.

226

DISCUSSING A RESEARCH PROGRAMME • Only after having established the key idea of optics, is seeing explained by light entering the eye from the object seen. • Only after having established the key idea of optics and having explained seeing, are techniques for showing the path of light introduced, such as blowing smoke or letting light from a slit interact with a sheet of paper. If these techniques are already introduced during the very first lessons, pupils may easily get the impression that one can see light propagating in space, i.e. that seeing is a separate ability which does not depend on light entering the eye but rather on the eye looking, sending out glances, etc. • Teaching about image formation takes as a point of departure the idea, that when light which diverges from a point P on an object and due to refraction or reflection meets again in another point P1, an image appears of P in P1, and the corresponding idea of virtual images. Only after these ideas have been carefully discussed are conventional geometrical techniques for constructing images introduced.

5. RESULTS FROM THE DESIGN PHASE The design work is very much informed by the domain-specific theory, both when it is a question of sequencing and of designing activities, problems, and text for students. Also, elements of general theories play a role, e.g. we stress small group discussions and invent problems that might stimulate debate about different conceptions. During the design phase, both products of a conceptual and a practical nature are developed and are then described in the teacher guide. We think that the status of such inventions should be upgraded and looked upon as important results that might help improve teaching and learning. An example is the map of the structure of geometrical optics shown in Figure 2. Its construction has been informed both by our understanding of geometrical optics and of students' conceptions and learning possibilities. The knowledge area may be said to have a theoretical core, namely, the idea that light moves along straight lines (cf. the first point in the domain-specific theory). This idea exists as an element in everything that has to do with geometrical optics. The next layer in the structure comprises some theoretical propositions, all of which are based on the idea that light moves along straight lines in a given medium, namely, the laws of reflection and refraction, the conditions for image formation, and the refraction of white light into colours. These propositions enrich the theory and make it possible to understand and explain phenomena in the world. Some examples are given in the outermost layer. The structure according to Figure 2 is a tool for reflections and decisions. Suppose the teacher wants his/her pupils to practice theoretical reasoning. S/he discovers that they find it hard to understand the conditions for image formation. S/he then looks for other examples and perhaps chooses to follow the line 'linear propagation – the law of refraction – burning glass'. on a more general level, Figure 2 provides an overall picture of geometrical optics, linking together theory and phenomena. If the picture is clear in the teacher’s mind, s/he does not lose sight of the important relation 'theory-observation' when planning and performing teaching.

B. ANDERSSON, F. BACH, M. HAGMAN,. C. ORLANDER, A. WALLIN

227

Figure 2. A representation of the conceptual structure of geometrical optics 6. RESULTS FROM THE TRIAL PHASE Various constraints make it is necessary to prioritise among the possibilities of making investigations. We give the highest priority to finding out the pupils’ longterm retention, particularly of their conceptual understanding. If this is poor, then the designed teaching does not contribute to solving the problem that we formulated at the beginning. In one experiment pupils in grades 8 and 9 (13 groups of pupils) were taught geometrical optics by five teachers according to a sequence designed in the way indicated above (Bach, 2001; Andersson & Bach, 2003). We used 15 problems, most of them requiring a written explanation, for pre- and post-tests, the latter given half a year after teaching. As an example, the student is asked if anything happens between an observer's eyes and the object observed. The answers are categorized according to the models proposed by students, e.g. 'the eyes send out rays that are reflected by the object and back to the eyes' or 'sunlight is reflected by the object into the eyes'. To be able to compare different teaching groups, we have allocated one point to each acceptable answer to a problem. We can then observe an increase from pre- to post-test in the number of acceptable answers per pupil, varying between 0.7 for the poorest and 6.5 for the best group. We note that the teacher guide is understood in different ways. The teacher who regards him/herself as an unobtrusive supervisor tends to achieve poorer results than the one who adopts a more active teacher role, but we do not think that the data available from lessons and interviews constitute any secure basis for this assertion.

228

DISCUSSING A RESEARCH PROGRAMME

One interesting observation is that a teacher who taught the sequence three times in succession with different groups achieved better and better results. Dialogues during classroom visits also pointed to the fact that he understood the intentions and content of the teaching sequence increasingly well. This is an example of the importance of the teacher in the achievement of good results, and a reminder that one should not automatically make drastic changes in a teaching sequence that has perhaps not produced such good results the first time it was tested. 7. DISCUSSION Teaching that is done in a number of classes is a complex phenomenon and, depending on what questions to which one wants answers various methodological problems will arise. One question is whether the completed teaching sequence constitutes a substantial improvement in learning compared with the 'prevailing teaching praxis', a concept of which the meaning is not self-evident but can be defined to a certain extent by analysing popular teaching material. As far as our sequence in geometrical optics is concerned, however, we have partly been able to use the same test problems as in the national assessment. It is then evident that in many respects our pupils answer markedly better than the national sample. This points to the possibility of integrating national assessments into research projects that are aimed at improving teaching. Another question is what aspects of the teaching are particularly important to learning? What is the significance of the domain-specific theory or particular aspects of it? How important for learning are general elements such as small group discussions and elicitation of pupils' ideas? In principle, this problem can be tackled by comparing two teaching sequences that only differ with respect to one of several of the factors just stated. This type of study is both technically and practically demanding. Generally speaking, one can say that the more the studied variable is refined, the weaker its effect is likely to be, compared with the total teaching environment which may render any positive differences insignificant. Another possibility of getting at the critical aspects of learning is to follow individual pupils throughout the teaching sequence. It is then a question of not only focussing on the pupil, e.g. by repeated interviews. The important thing is how the pupil interacts with different influential factors, such as the teacher’s practice, discussions with classmates and teacher, study of texts, and so on. One difficulty is that this influence may have a delayed effect. Thus, verifying a domain-specific theory by experiments is not an easy matter. This, by the way, is also the case for general theories of teaching and learning. Perhaps the best one can attain is a demonstration that clever applications of both a domain-specific and a general theory lead to significantly better learning than does conventional teaching. This is not a small achievement. To this we want to add our experience, that clarifying domain-specific conditions for improved learning with the help of research results concerning students' conceptions and learning possibilities, leads to many new design ideas. Finally, we note a growing interest in the design of teaching, both in the U.S.

B. ANDERSSON, F. BACH, M. HAGMAN,. C. ORLANDER, A. WALLIN

229

(Kelley, 2003) and in Europe (Psillos & Méheut, 2001; Méheut & Psillos, in press). In Sweden the government demands considerably more of the kind of research that turns directly to the school world and emphasises especially the need of research in didactics of subject matter. This work, including developing domain-specific theories, is probably a way to strengthen science education as an autonomous discipline, which is in the interest of both teachers and researchers. ACKNOWLEDGEMENT This paper has been written within a project financed by the Swedish Research Council. REFERENCES Andersson, B., & Bach, F. (1996). Developing new teaching sequences in science: The example of 'gases and their properties'. In G. Welford, J. Osborne & P. Scott (Eds.), Research in science education in Europe: current issues and themes (pp. 7-21). London: The Falmer Press. Andersson, B., & Bach, F. (2003). Att undervisa i geometrisk optik – kunskapsbas och undervisningsförslag. Ämnesdidaktik i praktiken nr 6. Mölndal: Göteborgs Universitet, Inst. för pepdagogik och didaktik. Bach, F. (2001). Om ljuset i tillvaron. Göteborg Studies in Educational Sciences, 162. Göteborg: Acta Universitatis Gothoburgensis. Baird, J. R. & Northfield, J. R. (eds.) (1992). Learning from the Peel Experience. Melbourne: University of Monash. Black, P., & Wiliam, D. (1998). Inside the black box: Raising standards through classroom assessment. Phi Delta Kappan, 80(2), 139-48. Bransford, J. D., Brown, A. L., & Cocking, R. C. (Eds.). (2000). How people learn. Brain, mind, experience, and school. Washington, D. C.: National Academy Press. Cobb, P., Confrey, J., diSessa, A., Lehrer, R., & Schauble, L. (2003). Design experiments in educational research. Educational Researcher, 32(1), pp.9-13. Driel, J. H. v., Verloop, N., & de Vos, W. (1999). Developing science teachers´ pedagogical content knowledge. Journal of Research in Science Teaching, 35(6), 673-696. Driver, R., Squires, A., Rushworth, P., & Wood-Robinson, V. (1994). Making sense of secondary science – research into children's ideas. London: Routledge. Furth, H. (1969) Piaget and knowledge. Englewood Cliffs, N. J.: Prentice-Hall. Hagman, M., Olander, C., & Wallin, A. (2002) Research-based teaching about biological evolution. In J. Lewis, A. Magro & L. Simonneaux (Eds.), Biology Education for the Real World. Student – Teacher – Citizen (pp. 105-119). Proceedings of the IVth ERIDOB Conference. Toulouse: Ecole National de Formation Agronomique. Jones, M. G., Carter, G., & Rua, M. J. (1999). Children´s concepts: Tools for transforming science teachers´ knowledge. Science Education, 83(5), 545-557. Kelley, A. (2003). Theme issue: The role of design in educational research. Educational Researcher, 32(1), pp.3-4. Klaassen, C. W. J. M. (1995). A problem-posing approach to teaching the topic of radioactivity. Utrecht: CD-β Press.

230

DISCUSSING A RESEARCH PROGRAMME

Lemke, J. L. (1990). Talking Science. Norwood, N. J.: Abelx Publishing Corporation. Lijnse, P. (1995) “Developmental research' as a way to an empirically based “didactical structure' of science. Science Education, 79(2), 189-199. Lijnse, P. (2000). Didactics of science: the forgotten dimension of science education research. In R. Millar, J. Leach & J. Osborne (Eds.), Improving science education. The contribution of research (pp. 308-326). Buckingham: Open University Press. Méheut, M., & Psillos, D. (2004). Teaching-learning sequences. Aims and tools for science education. International Journal of Science Education,26, 515-535. Ogborn, J. (1997). Constructivist metaphors of learning science. Science & Education, 6(1-2), 121-133. Psillos, D., & Méheut, M. (2001). Teaching-learning sequences as a means for linking research to development. In D. Psillos, P. Kariotoglou, V. Tselfes, G. Bisdikian, G. Fassoulopoulos, E. Hatzikraniotis, & M. Kallery (Eds.), Proceedings of the third international conference on science education research in the knowledge based society, Vol. 1 (p. 226). Thessaloniki: Aristotle University of Thessaloniki, Dept of Primary Education. Stigler, J.W., & Hiebert, J. (1999). The teaching gap. Best ideas from the world’s teachers for improving education in the classroom. New York: The Free Press. Summers, M., Kruger, C. (1994). A longitudinal study of a constructivist approach to improving primary school teachers' subject matter knowledge in science. Teaching & Teacher Education, 10 (5), 499-519. Tiberghien, A (1997). Construction of prototypical situations in teaching the concept of energy. In G. Welford, J. Osborne & P. Scott (Eds.) Research in Science Education in Europe: Current Issues and Themes (pp. 269-282). London: Falmer. Viennot, L., & Rainson, S. (1999). Design and evaluation of a research-based teaching sequence: the superposition of electric field. International Journal of Science Education, 21(1), 1-16. Wallin, A., Hagman, M., & Olander, C. (2001). Teaching and learning about the biological evolution: Conceptual understanding before, during and after teaching. In I. GarcíaRodeja Gayoso, J. Díaz de Bustamante, U. Harms, & M.P. Jiménez Aleixandre, Proceedings from III Conference of European Researchers in Didactic of Biology (ERIDOB) (pp. 127-139) Spain: Universidade de Santiago de Compostela.Universidade de Santiago de Compostela, Spain.

"SCIENTIFIC COMMUNICATION": AN INSTRUCTIONAL PROGRAM FOR HIGH-ORDER LEARNING SKILLS AND ITS IMPACT ON STUDENTS’ PERFORMANCE ZAHAVA SCHERZ, ORNIT SPEKTOR-LEVY, BAT SHEVA EYLON The Weizmann Institute of Science, Israël

ABSTRACT In this paper we describe an instructional model for the acquisition of high order learning skills (HOLS) and the program "Scientific Communication", which supports its application in a junior high school (JHS) science and technology curriculum. The model emphasizes explicit and spiral instruction of learning skills, and a continuous demand for their implementation in various contexts and tasks. We describe a study that assessed the impact of our instructional model on students' performances. Students (N=447) from five different JHSs participated in the study: One group (N=334) studied the program “Scientific Communication”, and the other (N=113) did not study learning skills through any formal program. The results show superior performance of the first group over the second in the following ways: the ability to describe and explicate the practice of learning skills; three aspects of the actual performances of a complex task: knowledge, learning skills, and the quality of products; and reports by students on the skills that they had acquired. The results also indicate that high and average achieving students gained the most from the program. We concluded that the contribution of the program “Scientific Communication” to students’ performances of learning skills indicates the potential of its underlying instructional model in achieving its goals.

1. INTRODUCTION Nowadays science teachers have to face an endless struggle between the vast number of topics that must be taught and the need to develop independent learners who master a variety of high order skills, namely, learning skills, thinking skills, and inquiry and problem solving skills. A major goal of science instruction in school is to prepare all students for life in a world of rapid scientific and technological change, rather than to prepare a small number of students for a highly specialized scientific career. The shift is toward placing curriculum content in more ecologically valid contexts by making it more inquiry-based (Linn, Songer & Eylon, 1996; Bybee, 1997) and by urging the adoption of measures to assess students in ways which tap their ability to engage in inquiry activities rather than memorization of content knowledge per se (Bol & Strage, 1996). This shift in goals, has led to reforms in science and technology education in many countries (National Science 231 K. Boersma et al. (eds.), Research and the Quality of Science Education, 231—243. © 2005 Springer. Printed in the Netherlands.

232

AN INSTRUCTIONAL PROGRAM FOR HIGH-ORDER LEARNING SKILLS

Education Standards, 1996; The National Curriculum of England – Science, 1999; DeBoer, 2000). These reforms reflect the overall tendency to encourage students to integrate into their lives what they learn in the science classroom and to replace traditional frontal instruction with instruction that is more student-centered and active (Linn, Songer & Eylon, 1996; Bybee, 1997). In order for students to learn independently, they must be able to implement high order skills successfully (Berliner, 1992; Campbell et al., 2000). However, some researchers and educators claim that students' capabilities and skills develop spontaneously throughout various learning experiences in the course of their studies. Others claim that skill development can be achieved only through explicit, guided, and well-planned learning opportunities. Most studies do not support the first approach of spontaneous acquisition of skills; in fact, such studies show that students find it difficult to acquire skills by themselves and thus need supervision and direct instruction (Shamos, 1995; Castello, & Monereo, 1999). Thus, it is necessary to equip teachers with instructional methods and materials that foster adequate skills development. In this article we focus mainly on high-order learning skills (HOLS): any HOLS is composed of skills and sub-skills. The performance of each of these skills and sub-skills and the ability to integrate them in complex learning tasks determine the level of one's learning capabilities. For example, oral presentation is a HOLS that requires the implementation of various skills and sub-skills, like gathering information from different sources, information organization, and summarizing. Integrating well all these skills into a final product of oral presentation constitutes evidence of high order learning skills. In this article we refer to the following high order learning skills: information retrieval, scientific reading, scientific writing, listening and observing, information representation, and knowledge presentation. We designed a model for HOLS instruction, and developed learning materials, the program “Scientific Communication”, for the junior high school (JHS) level (grades 7-9). In this article we also present a study of the effect of our instructional model on the performances and achievements of students in learning tasks. 2. THE “SCIENTIFIC COMMUNICATION” PROGRAM In 1993 a new reform took place in Israeli Junior High Schools (JHS): a new subject was introduced, “Science and Technology” (S&T), that was followed by a new syllabus which emphasized skills learning. In accordance with that reform, we developed a new program: "Scientific communication", for teaching HOLS. Each of these skills (HOLS) can be further divided into sub-skills; for example, scientific writing may refer to knowing how to write a scientific essay, how to write a scientific report, how to compose an abstract, and so on (Figure 1).

Z. SCHERZ, O. SPEKTOR-LEVY, B.S. EYLON

233

High Order Learning Skills “Scientific Communication”

Information Retrieval

Scientific Reading

Listening & Observing

Scientific Writing

Information Representation

Table

Article

Textbook

Knowledge Presentation

Graph

Lecture Scheme

Report

Demonstration

Library

Model

Electronic Video

Scientific Essay

Poster

Abstract

Database

Multimedia Report

Experts

Oral Presentation

Figure 1: “Scientific Communication”: high-order learning skills and sub-skills The “Scientific Communication” program is based on a general model we propose for high order skills instruction in S&T studies. The model is characterized by the following: Framework: The framework consists of many generic activities that support the instruction and practice of different skills. The generic activities can be used in conjunction with any given content in S&T. Integration: Teachers are supposed to tailor the generic activities into specific scientific content areas and to adjust the development of skills to the level and abilities of the students. Spiral instruction: Throughout the school years, students learn and practice high order skills in S&T. Each year they are also introduced in depth to different skills and sub-skills. They are given the opportunity to practice these skills several times in the course of their studies.

234

AN INSTRUCTIONAL PROGRAM FOR HIGH-ORDER LEARNING SKILLS

Modularity & flexibility: Each teacher can plan a sequence of skills instruction, using the generic activities. The teacher has the option to decide what skills will be developed, the content, the timing, and the level of complexity. The program “Scientific Communication” was developed with the following main goals: (1) to enhance the performance of HOLS by students, (2) to furnish teachers with instructional materials and activities that can be implemented and integrated in a variety of scientific subjects, and (3) to design flexible instructional materials suitable for different levels of students and to meet the different needs of the class and the teacher. In order to put into practice these characteristics and goals, “Scientific Communication” was designed as a didactic learning package that includes an activity booklet for the student, an extended guide for the teacher, interfaces consisting of relevant texts and articles for several main topics of the S&T curriculum, and assessment tasks. Our model of skills’ instruction and the program “Scientific Communication” emphasize what we consider as central and essential aspects of skills instruction: explicit instruction, practicing in various topics, implementing in complex tasks, and applying spiral instruction. 3. THE STUDY Rational and main questions The main goal of the study was to examine the influence of the instructional model and the program “Scientific Communication” on JHS students. The questions generated by the study were as follows: 1. How do students describe their practice of learning skill? 2. How did the program “Scientific Communication” influence students’ performances and achievements? 3. What skills do the students report they acquired during their S&T studies? 4. METHODOLOGY The study followed students through the 7th & 8th grades. It focused mainly on students who studied learning skills according to our program, and it also monitored a comparison group of students who did not study learning skills through any formal program. Sample The student population consisted of 447 students from five different JHS (urban schools) divided into two groups: (1) The "Scientific Communication" group (SC group) consisted of 334 students (174 boys and 160 girls) who learned the program “Scientific Communication”, integrated within the subject “Science and technology for JHS”. These students were taught by experienced science teachers who had

Z. SCHERZ, O. SPEKTOR-LEVY, B.S. EYLON

235

previously participated in a workshop on the "Scientific Communication" program (minimum of 16 hours). (2) The comparison group (Comp. group) consisted of 113 students (47 boys and 66 girls) who did not study learning skills through any formal instruction. Methods and tools In order to evaluate the equivalency of the treatment and the comparison groups, two indicators which might have a strong influence on students' performances were used: (1) Students’ prior general knowledge of scientific topics was examined by a multiple-choice questionnaire (11 questions). The questions were selected from the published form of the international standards test TIMMS (TIMMS, 1999) which was designed for 8th grade students internationally. Internal reliability was measured using the Kuder-Richardson formula 20, α =0.82. (2) Students’ prior academic achievement level was based on a ranking provided by the school science teachers and science coordinators. Students were ranked according to three levels: high-achievers, average-achievers, and low-achievers. Students’ performances and descriptions of their practice of learning skills were examined by the following: A ‘Learning situations’ questionnaire was used which consisted of open questions that presented different learning tasks. Students were asked to describe how they would accomplish these tasks. The learning tasks will be outlined along with the results. A Complex performance task that was designed to assess students' HOLSs such as gathering information, analyzing data, information representation, and knowledge presentation. None of the two groups had experienced tasks of a similar kind during their previous school studies. A detailed description of the complex task will be presented along with the results. A ‘Skills and capabilities’ questionnaire presented an open question in which students were asked to indicate what skills and capabilities they had acquired during their S&T studies. 5. RESULTS Analysis of students’ prior general knowledge of scientific topics and their prior academic achievement level revealed that there were no significant group differences regarding the two indicators (Table 1). Therefore, analysis of variance (rather than covariance) was carried out in the subsequent analyses. How do students describe their practice of learning skills? At the beginning of the 7th grade and at the end of 8th grade, a questionnaire was administered to students in order to assess their ability to describe and explicate their learning skills. Two learning situations were presented, and students were asked to describe how they would accomplish these tasks:

236

AN INSTRUCTIONAL PROGRAM FOR HIGH-ORDER LEARNING SKILLS Table 1: Indicators of prior general knowledge and academic achievements of the study groups

(a) Students’ prior general knowledge of scientific topics N Average score Comparison group 105 74.8 SC group 217 72.1 T320= 1.01 ; p = 0.3155

Std Error 2.2 1.5

(b) Rank of academic achievement level N Average level Comparison group 107 2.3 SC group 296 2.3 T401=0.5; p = 0.6201

Std Error 0.06 0.04

Task 1 "Scanning an article": Students were asked to describe how they can decide in five minutes whether a ten page popular scientific article is relevant for a certain task. Task 2 "Navigation in the library": Students were asked to describe in detail what they would do in order to find three reliable sources of information in the library which dealt with a specific subject in science. The task had to be accomplished without any help from the librarian. The students' responses to the learning tasks were analyzed; the responses were graded according to the number of meaningful phrases or keywords that indicated acquaintance with learning skills. The answers were categorized and coded according to criteria that were validated by two science teaching researchers, and one expert in qualitative research. For example, for Task 1 we considered phrases like: I will check: Who is the author? Who is the publisher? What is the date it was published? Is there a bibliography? I will look at figures, graphs, illustrations. I will read the abstract. I will skim the article. In Task 2 we considered keywords/phrases like: I will search in the catalogue. I will find the Dewey classification number. I will look at the index. I will check the list of contents. I will read in a lexicon. I will search in scientific journals. I will look for a textbook. I will consult a scientific encyclopedia. Table 1 shows the analysis of the students' answers. The results indicate that students in the SC group who studied "Scientific Communication" improved their answers and used more meaningful keywords and phrases to describe their practice of learning skills. Such an increase did not occur in the answers of students from the comparison group. In Task 1 “Scanning an article“, the SC group used significantly more keywords (average of 1.96) than the comparison group (average of 1.2) in the post questionnaire. Also in Task 2 “Navigation in the library”, students from the SC group used significantly more keywords (average of 1.88) than the comparison group on the post-questionnaire (average of 1.42). We also found that in the post-

Z. SCHERZ, O. SPEKTOR-LEVY, B.S. EYLON

237

test, a higher percentage of SC students could mention three keywords or more to describe their task process. In both tasks, there was no difference between the two groups on the pre-test (Table 2). Table 2: No. of students (in %) that used keywords to describe learning tasks and the average number of keyword for each group Task 1: "Scanning an article" % of students No. of Comp. Pre SC Pre Comp. Post keywords (N=34) (N=106) (N=39) 0 17.65 36.79 20.51 1 58.82 31.13 43.59 2 23.53 19.81 33.33 •3 Average no. (Std Dev)

12.27 1.06 (0.65)

2.56

1.09 (1.07) 1.2 (0.86) P<0.0001

Task 2: "Navigation in the library" % of students No. of Comp. Pre SC Pre Comp. Post keywords (N=52) (N=155) (N=45) 0 7.69 5.81 13.33 1 46.1 49.03 42.22 2 30.77 28.39 33.33 •3 Average no. (Std Dev)

15.38 1.54 (0.85)

16.77

11.11

1.6 (0.92) 1.42 (0.86) *P<0.01

SC Post (N=133) 15.04 26.32 27.82 30.83 1.96 (1.45)*

SC Post (N=130) 6.92 35.38 26.15 31.53 1.88 (1.05)*

The data presented above was based on students’ declarations and verbal descriptions only. Students’ achievements and performances in learning skills were indicated through a complex performance task that will be described next. How did the program “Scientific Communication” influence students’ performances and achievements? At the end of the 8th grade, all students were asked to accomplish a complex performance task: “Update Report”. This task was related to the topics: materials and earth sciences (the atmosphere, air pollution, and the mutual influence of man and environment). The task demanded implementation of learning skills, such as searching for information from a variety of sources (scientific books and articles,

238

AN INSTRUCTIONAL PROGRAM FOR HIGH-ORDER LEARNING SKILLS

professional internet sites, experts, etc.), analyzing and evaluating the data, writing a report, preparing illustrations, and presenting knowledge. As part of this task students were asked to take the role of an expert in atmospheric sciences and to prepare a report about one of the causes of air pollution. In order to prepare the task, the students had to read about air pollution, choose one of the pollutants or associated phenomena, and prepare a presentation. The presentation had to include an oral report accompanied by a visual presentation (up to five slides/transparencies to be presented in a documentary TV program) that showed relevant pictures, figures, graphs, tables, and so on. The text of the report (up to three pages ) and the visual presentation were handed to the teacher. Detailed instructions for completing the task provided students with necessary scaffolding and support. The students were also informed about the criteria of assessment. Table 3 illustrates the outlines of the scoring rubric that was used to analyze the task. For each category, detailed criteria were developed to determine the specific level of performance on a scale of 0-5. The rubric was developed through interaction with science education experts and several pilot experiences with JHS students from schools that did not participate in the study. Table 3: Outline of the scoring rubric concerning learning skills performance Category General criteria Information: The student gathered information from a variety of variety of sources sources like scientific books and articles, internet sites, experts, and government reports. Information: The student selected relevant information concerning the reliable & relevant pollutant. The information is reliable and professional. Text of the report The text is clear and understandable. It was written by the student himself - in his own words. The text includes information from different sources. The text is a max. of three pages. Visual The presentation includes graphs, tables, schemes, etc. illustrations that represent data concerning the pollutant and its concentration in air. The illustrations are clear and correctly designed. The illustrations were placed only within the slides and not in the “broadcasting text”. Slides / The slides are clear. The text is easy to read. The color transparencies of the text is in contrast with the background. The illustrations are the right size. Special effects are used. No more than five slides are used. Bibliography The list of references is detailed and includes all the components. The list is edited in alphabetical order. A sample of 206 students who participated in the research accomplished the complex task "Update Report" (112 students from the SC group and 94 students from the comparison group). Students' products were assessed by an experienced

Z. SCHERZ, O. SPEKTOR-LEVY, B.S. EYLON

239

teacher who was 'blind' regarding the trial conditions and the study hypotheses, and by one of the conductors of the study. The teacher’s judgments according to the scoring rubrics were compared with the researcher’s judgment and were found to have a good degree of agreement (Fleiss, 1981; Agresti, 1990): Weighted kappa (KW) = 0.448 (SE = 0.03; Z = 12.94). Whenever a disagreement was detected, the teacher’s judgment was considered, as the researcher was not blind to the hypotheses of the study. Table 4: "Update Report": Students’ mean scores (on a scale of 0 – 5). • P< 0.005; ** P< 0.0005 Categories

Knowledge

Learning Skills

Products & outcomes

Criteria -Information about the pollutant -Main concepts & terms -Chemical processes -The air as a mixture of gases -Change of the phases of matter -Effects of air pollution on man & environment -Ways to decrease pollution -Information: variety of sources, reliable & relevant -Text: clear & professional -Representation of quantitative data -Computerized presentation -Text: comprehensive, edited according to requirements -Analysis of data & conclusions -Presentation: edited according to requirements -Integration of text & presentation -Compiled bibliography

Std Error

P

2.7

0.12

*

2.7

0.13

**

0.11

**

SC Group N=112

Std Error

Comp. Group N=94

3.3

0.11

3.3

0.11

2.7

0.11

2.1

The results of students' products in the complex performance task were categorized into three groups: knowledge, learning skills, and products & outcomes. An analysis of students’ performances indicated that the SC group which studied the program “Scientific Communication” performed significantly better than the comparison group. Specifically, they attained a higher level of knowledge and much better learning skills and products (Table 4)

240

AN INSTRUCTIONAL PROGRAM FOR HIGH-ORDER LEARNING SKILLS

The SC Students were divided into three sub-groups according to their achievements on the pre-test of 'General knowledge in scientific topics'. The sample was divided into the following categories: high-achievers, average-achievers, and low-achievers. According to the analysis and the results presented in Table 5, it is apparent that the average- and high-achievers gained most from the instruction of learning skills using the “Scientific Communication” program. Within the low-achiever groups, no differences were detected between those who learned the program and those who did not Table 5: “Update Report” – students’ mean scores (and Std Error) of high, average, and low-achievers (on a scale of 0–5), N=206; * P < 0.05; ** P < 0.001

DOMAINS & ITEMS

Knowledge Ways to decrease pollution

Low achievers SC Comp. 2.8 (0.33) 2.4 (0.19) 3 (0.4)

2.2 (0.33)

Medium achievers SC 3.4 (0.22)

High achievers

Comp. 2.8 (0.22)

SC 3.4 (0.13)

Comp. 3.2 (0.19)

3.6 (0.33)* 2.4 (O.38)

3.7 (0.22)

3.4 (0.33)

3.4 (0.16)* 2.6 (0.25)

Learning skills

2.8 (0.18) 2.6 (0.19)

3.4 (0.24)

Computerized presentation

2.7 (0.4)

1.7 (0.33)

4 (0.32)** 2.2 (0.37) 3.5 (0.26)** 1.6 (0.44)

Products & outcomes

2.2 (0.2)

2 (0.17)

Presentation: edited according to requirements

2.9 (0.42) 1.5 (0.32)

Integration of text & presentation

2 (0.37)

1.7 (0.3)

2.9 (0.23)

2.9 (0.25)

2.3 (0.21)

2.8 (0.14)* 2.1 (0.22)

3.4 (0.34)* 2.1 (0.38) 3.3 (0.26)** 1.4 (0.41)

3.3 (0.33)*

2 (0.34)

3.2 (0.25)** 1.4 (0.41)

What skills do students report they acquired during the S&T studies? At the end of the 8th grade, students were asked to indicate in an open question which skills and capabilities they had acquired during their school science and technology studies. The skills that were mentioned were listed and categorized. The results show that students who studied the program “Scientific Communication” indicated they had acquired a larger variety of skills and capabilities. Moreover, each of the skills was mentioned by a significantly higher percentage of the SC students than in the comparison group (Figure 2).

Z. SCHERZ, O. SPEKTOR-LEVY, B.S. EYLON

241

Figure 2: Students' indications of skills they had acquired in S&T during the 7th and 8th grades 6. DISCUSSION The results of this study demonstrate the importance of the explicit and spiral instruction of learning skills in science classes in JHS. The contribution to students’ learning was demonstrated not only at the declarative level (based on students’ answers in the questionnaires), but also at the practical level in the context of carrying out a complex performance task. More specifically, we found that students who studied learning skills through the program “Scientific Communication” could describe in detail the actions that they had conducted while implementing learning skills in learning situations, and they used more meaningful phrases and keywords which reflected their acquaintance with the skills. These results may indicate that students who experienced our instructional model can explicate and describe better their implementation of learning skills. Students’ performances and achievements were assessed by a complex performance task that required the implementation of several learning skills.

242

AN INSTRUCTIONAL PROGRAM FOR HIGH-ORDER LEARNING SKILLS

Importantly, students who learned “Scientific Communication” achieved much better scores in three main categories that were measured: (a) knowledge, (b) performance of learning skills, and (c) products & outcomes. However, the program “Scientific Communication” seems to affect mainly high and average-achievers, with very little impact on low-achievers. The differences in performance, between students who acquired learning skills through explicit and spiral instruction of skills and those who did not, have important implications. The well-planned instruction of skills through a well-defined model of skills’ instruction is not widespread. On the contrary, science and technology curricula often declare the development of skills as a central goal but provide minimal guidance regarding how to teach the skills in class. There is a hidden assumption that the acquisition of skills happens spontaneously. Our study shows that learning skills development may occur to some extent without any formal instruction. However, formal and systematic teaching can make a significant difference, as indicated by the finding that the program “Scientific Communication”, based on our instructional model, improved students' capabilities. Such an enhancement is an important step toward independent learning. We can conclude that the development of learning skills may occur to some extent through experience or ordinary S&T studies even without formal training. However, formal training is necessary in order to make a significant difference. We showed that explicit and spiral instruction of skills, integrated into scientific contents, along with a continuous demand for implementation in various contexts and tasks, promotes a meaningful development of HOLS. This model of learning skills instruction can be applied to the instruction of other high order skills like thinking skills, inquiry and problem-solving skills. This general model has the potential to enable teachers and educators to promote skills instruction that is integrated into scientific content. Thus, its application can lead to realization of the central goals of science education. REFERENCES Agresti, A. (1990). Categorical Data Analysis. New York: Wiley. Berliner, D. C. (1992). Redesigning classroom activities for the future. Educational Technology, 32, 7-13. Bol, L. & Strage, A. (1996). The contradiction between teacher's instructional goals and their assessment practices in high school biology courses. Science Education 80 (2), 145-163. Bybee, R. W. (1997). Achieving Scientific Literacy: From Purpose to Practice. Portsmouth, NH: Heinemann. Campbell, B., Kaunda, L., Allie, S., Buffler, A. & Lubben, F. (2000). The communication of laboratory investigations by university entrants. Journal of Research in Science Teaching, 37 (8), 839-853. Castello, M. & Monereo, C. (1999) Teaching learning strategies in compulsory secondary education. 8th European Conference for Research on Learning and Instruction, Sweden.

Z. SCHERZ, O. SPEKTOR-LEVY, B.S. EYLON

243

DeBoer, G. E. (2000). Scientific literacy: Another look at its historical and contemporary meaning and its relationship to science education reform. Journal of Research in Science Teaching, 37, (6), 582-601. Fleiss, J. L. (1981). Statistical Methods for Rates and Proportions. New York: Wiley. Linn, M. C., Songer, N. B, & Eylon, B. S. (1996). Shifts and convergences in science learning and instruction. In R. Calfee & D. Berliner (Eds.), Handbook of Educational Psychology. New York: Macmillan. National Curriculum of England (1999). http://www.nc.uk.net/index.html National Science Education Standards, (1996) National Academic Press, USA. Shamos, M. H. (1995) The Myth of Scientific Literacy, Rutgers Univ.. Press. Spektor-levy, O., & Scherz, Z. (1999) Scientific Communication. The Weizmann Institute for Science and The Ministry of Education. Rehovot, Israel. TIMMS, (1999). http://isc.bc.edu/timss1999i/pdf/t99science_items.pdf

PART 5 Teaching the nature of science

LEARNING ABOUT THE NATURE OF SCIENTIFIC KNOWLEDGE: THE IMITATING-SCIENCE PROJECT

STEIN DANKERT KOLSTØ, IDAR MESTAD University of Bergen, Norway

ABSTRACT This paper reports a small-scale curriculum project aimed at teaching about the nature of science at lower secondary school. The main idea of this project is to stimulate students' learning of science as a process by involving the students in reflections based on personal experience of an open-ended investigation. The paper describes how it is possible to include publication and argumentation of methods and results in a school experiment, as well as findings related to changes in students' epistemological thinking. The students wrote about how researchers conduct research prior to and after the project. Analysis of these texts showed that more students included the idea of testing hypotheses in the post-study texts than in the pre-study texts. Students also expressed more awareness of what researchers might do to enhance the quality of their research. We found that students tended to use words like "facts" and "proofs" in the prestudy texts. In the post-study texts, however, more students emphasised that research findings do not represent final answers but the researcher's argument-based conclusions.

1. INTRODUCTION "The Imitating-Science Project" aimed to stimulate the development of students' epistemological thinking. The focus of the project was science as a process. In general, science involves processes like examination of literature and generation of hypotheses, empirical data, and factual claims. Science as a process, however, also includes communal aspects. Inherent in the learning objectives of the ImitatingScience Project is therefore knowledge about social processes in science, e.g. publication and critical evaluation of research reports. In this study the imitation of scientific research is not done only to facilitate students' learning of how to make scientific investigations, but to increase their knowledge of communal aspects of science like publication and argumentation. Researchers who have studied curriculum projects using practical work or "authentic school science" have often found less change in students' ideas about science as a process than expected (Bell et al., 2003; Leach & Paulsen, 1999). There are indications, however, that involving students in reflection on the nature of scientific inquiry throughout a project might lead to better results (Gess-Newsome, 2002; Lederman, 1992). In this project it was also assumed that including a focus on publication and argumentation would make it easier for students to understand the interpretative nature of science. We hoped this would facilitate a conceptual change 247 K. Boersma et al. (eds.), Research and the Quality of Science Education, 247—257. © 2005 Springer. Printed in the Netherlands.

248

LEARNING ABOUT THE NATURE OF SCIENTIFIC KNOWLEDGE

away from the identification of measurements and research findings as facts, often found among students (Lederman, 1992), towards a more constructivist view of scientific knowledge. 2. THEORETICAL FRAMEWORK The theoretical framework of this study is threefold. First, our understanding of the concept scientific literacy is the basis of the study. Socio-scientific issues which students encounter through, for example the media, often involve disputed scientific arguments, contested frontier science, and real or perceived expert disagreement. In this way, science-related public debate mirrors professional scientific debate to some extent. Adequate interpretation of arguments and factual claims require knowledge of processes involved in developing scientific knowledge. This view is supported by a number of qualitative case studies of lay people's interaction with science and scientists (see Ryder, 2001 for an informative overview). These studies indicate that knowledge of the nature of science and scientific knowledge are at least as important as scientific content knowledge for lay peoples' assessment of science-based arguments (Kolstø, 2001; Ryder, 2001). Also, there is a growing recognition that "social and institutional processes are critically important in determining what comes to be agreed as reliable public scientific knowledge" (Leach & Paulsen, 1999, p. 135). Our curriculum project is therefore based on the view that it is relevant and important to increase students' knowledge about science as a process as well as improving their skills in participating in evaluation and argumentation related to empirical data and scientific knowledge claims. Second, our constructivist view of science influenced both the design of the curriculum and the analysis of data. This view emphasises interpretation and argumentation related to the production of empirical data and scientific knowledge claims. It also emphasises the difference between two kinds of science (Cole, 1992; Latour, 1987; Ziman, 1968). One kind, denoted as 'core science' by Cole, is characterised by a stable consensus within the scientific community. This is science where the disputes, at the initial stages of the research, have settled and now occurs as facts in textbooks. The other kind, denoted as 'frontier science' by Cole, is science in the process of being researched. At this stage of the production of scientific knowledge, hypotheses are being developed and scrutinised, and results from studies are presented to colleagues and discussed with them (Ziman, 1968). Subjective and unreliable frontier science is transformed into core science, or refused, for example as unreliable, through different social processes characterised by publication, evaluation, and argumentation. Such a constructivist view of science inspired us to include an open debate on methods, interpretations, and conclusions in the project in order to imitate these aspects of science. We also maintain that the main purpose of scientific research is the development of theory. Investigative science should therefore involve the question, "why does this happen?" (Leach & Paulsen, 1999, p. 61), even if it can be argued that authentic school science could include other aspects of science as practised in modern society. Third, the didactical design of the Imitating-Science Project is inspired by Dewey's (1913) emphasis on the need to base students' learning on practical problem

S. DANKERT KOLSTØ, I. MESTAD

249

solving activities in order to achieve a genuine understanding of theoretical concepts. However, the practical activities are not expected to develop students' theoretical understanding automatically. The teacher's role is to build on the interest created by the practical activity and to challenge the students to try to understand the underlying principles involved. Inspired by Dewey, we chose to let students design, perform, make reports, and discuss results based on their own hypotheses, about a problem put forward by the teacher. 3. THE DESIGN OF THE IMITATING-SCIENCE PROJECT The implementation of the curriculum presented here involved two Norwegian science classes, 9th and 10th grades, at two different schools. The teacher, who is the second author of this article, was doing this project as part of his master's degree in Science Education. He taught both classes for the duration of the project period. That semester he was also a supply teacher in one of the classes, and the year prior to the project he was a regular teacher at one of the participating schools. The basic idea in the Imitating-Science Project was to let students perform an open-ended investigation in small 'research' groups consisting of three to four students. To trigger students' interest in research and to focus their attention on the topic, the project started with a class discussion of three scientific claims from recent newspaper headlines. The discussion focussed on the nature of the research behind such headlines, and on the reliability of research-based claims. Following the discussion, students were informed about the project and the learning objectives. Then they were given the general research question for the experiments: "Why do people walk around in circles in open spaces when visibility is low due to fog or snow?" All the research groups should work on the same research question, but they had to develop their own hypotheses. This was important in order to imitate the process where several research teams work on a problem, trying to develop an adequate model for a specific aspect of the physical world. In the TIMMS study (Kind et al., 1999), Norwegian students often employed unstructured experimenting without planning when faced with open-ended practical problems. Based on an idea proposed by Frost (1995), we therefore introduced an electronic framework (see Mestad, 2003) for structured planning to scaffold the planning process. The scaffolding framework contained headlines, sentence starters, and short explanations in order to help students plan and understand concepts like hypothesis, variable, and conclusion. It was also designed to stimulate reflection on items like testability of hypotheses and methodological reliability. The framework was transformed into a plan for an experiment through student collaborative work in front of a computer. The teacher checked each group's plan and approved it using the following criteria: if it was completed, if the necessary equipment was obtainable, and if the suggested method did not involve any safety problems. At this stage the teacher did not discuss the methods suggested by the students. After printing the plans, the research groups carried out their experiment by selecting, collecting and interpreting data, drawing conclusions, and writing an electronic report. The discussion of results passed through two stages. First, the

250

LEARNING ABOUT THE NATURE OF SCIENTIFIC KNOWLEDGE

research groups at each school presented their hypotheses, methods, and results orally to their science class. Each presentation was followed by a class discussion of weak and strong aspects of the different methods used. The teacher wrote all the methodological points mentioned on the blackboard: strong aspects first, weak aspects below, and sorted them horizontally according to the six different research groups involved. The driving force behind this discussion was the challenge of making a joint report presenting the different experiments and a tentative conclusion to the research question. In order for students to get their research included, it was important for them to find weak points in other groups' reports and be able to defend their own group's report. Through voting, it was decided which group should be responsible for writing the joint report. Second, the joint reports from the participating classes were published in a closed learning management system (Luvit). Then the groups in both classes started an on-line discussion, commenting on and criticising the different hypotheses, experimental methods, and conclusions used. During the on-line discussions, the teacher served as a consultant who sometimes challenged groups of students to examine certain aspects of a report. Throughout the planning process and the experimentation phase, the teacher often started the lessons with a short discussion about the purpose of the project and the evolving ideas of the characteristics of scientific research. The final lesson in the project was devoted totally to a teacher-led class debate of science as a process and the similarities and differences between authentic scientific research and the process that students had gone through in the project. By referring to their own open experiments and reliability discussions, it was hoped that concepts like hypothesis, reliability, community of scientists, and so forth could be backed up by meaningful associations. The discussion also touched upon the question of the reliability of scientific knowledge and knowledge claims in general in order to stimulate the students' evolving epistemological thinking. 4. EVALUATION METHOD The students involved in the project were fourteen to fifteen years old. The two science classes were mixed-ability classes with twenty and twenty-eight students, respectively, from mixed socio-economic backgrounds. The analysis presented is based on the short texts written in the class of twenty-eight students. (The first texts were written after a class discussion on three newspaper headlines.) Students were given ten minutes and a sheet of paper with the question: "How does a researcher work?" They were given this task both before the open experiment started and after the project was completed. In total the data consist of twenty-three pre-study texts and twenty-three post-study texts, all between one third to a full hand-written page. The analysis of the texts was based on the following three research questions: 1. 2. 3.

What are pupils' ideas about processes involved in research? What are pupils' ideas about what researchers do to increase the quality of their research? What are pupils' views about the reliability of research outcomes?

S. DANKERT KOLSTØ, I. MESTAD

251

The analysis was qualitative. Descriptive categories were inductively generated from inspection of data (Merriam, 1998), but they were influenced by our views and our prior knowledge about the characteristics of scientific research. The analyses were done through code and retrieval processes based on the constant comparative method (Strauss & Corbin, 1990) and supported by the computer software Atlas.ti. A main aim of the analysis was to identify similarities and differences between students' ideas prior to and after the project. It is not our claim that the efficiency of the Imitating-Science Project might be judged on the basis of this small-scale qualitative study. Nevertheless, we did hope to gain insight into new ideas that students developed during the project, and also to generate hypotheses about weak and strong points in the curriculum to inform further development. Finally, we wanted to generate 'working hypotheses' (Cronbach, 1975) for other teachers, about how to use the curriculum successfully. Concerning the validity of the findings, it should be noted that the students were informed that the pre- and post-study texts were for research purposes and would not be given marks. However, an issue of concern is the individualistic phrasing of students' answers to the question in the texts. This might have led to lower emphasis on social processes and community-related aspects in students' texts. Another issue is the selections that students make in writing their texts. They do not necessarily write down all their thoughts, but only what is foremost in their minds, or what they believe is most important in their teacher's opinion. Finally, the students' texts are seldom written in clear and exact language. This probably reflects the students' immature and unclear thoughts about research and the nature of scientific knowledge. When categorising students' statements, we therefore tried to keep the number of categories low, e.g. we chose not to distinguish between different versions of constructivism when identifying students' epistemological thinking. The trustworthiness of the analysis and the findings are based on the disclosure of the theoretical framework and the process of the analysis. In addition, the analysis and findings have been checked and criticised by an experienced researcher in the field. 5. THE STUDENTS' IDEAS ABOUT PROCESSES INVOLVED IN RESEARCH Our first research question concerned students' views about the main processes involved in scientific research. Most students chose to write about this issue in their texts. In the pre-study texts, many students only described a researcher as someone who "makes inquiries" (n=8) or who "solves problems" (n=6). Several students (n=4) described a researcher as someone who collects and analyses data in order to find answers to problems. Erik wrote, for instance: I think researchers make inquiries and analyse and establish a conclusion. Some students (n=5) emphasised the development and testing of hypotheses, or of "theories" as the students called them. However, this idea became very evident in the post-study texts (n=14+5). In fact all other categories more or less disappeared from the students' texts and nineteen out of twenty students, who wrote about main

252

LEARNING ABOUT THE NATURE OF SCIENTIFIC KNOWLEDGE

activities in research, wrote about testing hypotheses. In addition, some students also included discussions with colleagues in their texts. Berit wrote: They [researchers] make hypotheses and then they test these in different ways (many different). Then they find a conclusion. Then they discuss with other researchers and criticise each other's work. In the post-study texts students' language relating to "hypotheses" was more precise, and the concept "hypothesis" was often used explicitly and in adequate ways. An overview of the findings is given in Table 1. Table 1: Main research processes mentioned by the students in their pre-and poststudy texts Main process in research Pre-study texts Post-study texts "Make inquiries" "Solve problems" Collecting and analysing data Testing hypotheses Several ideas including testing hypothesis The issue is missing / unclear Total

8 6 4 5 0

1 0 0 14 5

0 23

4 23

Thus, the general tendency was a shift away from focussing on a passive collection of data and drawing conclusions, towards active experimenting based on a hypothesis generated by the researcher. 6. THE STUDENTS' IDEAS ABOUT WHAT RESEARCHERS DO TO INCREASE THE QUALITY OF THEIR RESEARCH The second research question concerns characteristics of scientific research which are related to the need for reliable conclusions. In the pre-study texts only a minority of the students (n=7) mentioned strategies scientists and the scientific community use in order to increase the reliability of research findings. The most frequently mentioned strategies were the use of different methods (n=5) and co-operation with other researchers (n=2). In the post-study texts a majority (n=15) of the students mentioned strategies related to reliability; most of them mentioned two or more strategies. Several qualitatively new strategies also emerged. Students wrote about the examination of results by other researchers (n=6), the use of replications (n=4), and comparisons with other researchers' findings (n=4). The use of different methods (n=5) and the need for accuracy and patience (n=3) were also mentioned by some students. All strategies found are listed in Table 2.

S. DANKERT KOLSTØ, I. MESTAD

253

Table 2: Strategies related to reliability of findings found in the students' texts. The numbers indicate how many times the strategies were found Strategy Pre-study Post-study texts texts Examination of results by other researchers Use different methods The use of replications Compare with other researchers' findings Accuracy and patience Control of variables Use several different persons in the test Co-operate with other researchers Total

1 5 0 0 1 0 1 2 10

6 5 4 4 3 2 2 2 27

The general impression from reading the texts was an increased awareness of the importance of strategies to strengthen reliability. It is also interesting to note that strategies involving other scientists were rare in the pre-study texts but appeared quite frequently in the post-study texts. Thus, ideas related to social processes in science seem to have developed during the project. 7. THE STUDENTS' VIEWS ABOUT THE RELIABILITY OF FINDINGS Our third research question concerned students' views on the reliability of research findings. Here we looked for statements in which signs of students' epistemological thinking were possible to trace. Some students expressed the idea that research provides conclusive answers because it is based on measurements and observations. Anette for example wrote that a researcher ... maybe makes an inquiry to gain knowledge about what other people think is right, and then he must test it somehow so that he can prove that "it is like this". We labelled such statements 'positivistic'. In other texts students expressed the view that research results are based on the researcher's testing of different hypotheses and that his/her findings have to be discussed and might have to be revised. This view, which we called 'critical', implies that research findings do not represent final answers, but rather the researcher's argument-based conclusions. Sigrid for instance wrote: Arrive at a reasoned conclusion. Discuss their opinion with other researchers. If necessary make a new conclusion/hypothesis. If new hypothesis -> "start again". Finally, a few students expressed the view that research results, or the presentation of results, might be influenced by institutions with money to fund research. This view we labelled 'opportunistic'. Table 3 provides an overview of the categorisation of the different texts.

254

LEARNING ABOUT THE NATURE OF SCIENTIFIC KNOWLEDGE

Table 3: The table shows the number of pre- and post-study texts found to express the different three main views on the reliability of research findings. Category Pre-study texts Post-study texts Positivistic 9 6 Critical 2 9 Opportunistic 2 0 The issue is missing 10 8 Total 23 23 In the pre-study texts positivistic ideas were predominant, while critical views were predominant in the post-study texts. Two students expressed opportunistic views in their pre-study texts, but these views were absent from their post-study texts. We can only speculate on whether the absence of opportunistic views in the post-study texts indicates that the project increased some students' beliefs in science being neutral and objective. Comparison between pre- and post-study texts revealed that four students expressed positivistic views both in pre- and post-study texts. However, five students who had expressed positivistic views in their pre-texts expressed critical views in their post-texts. None of the students changed from having critical views to having positivist views. Thus, the Imitating-Science Project probably made an impact on some students' epistemological thinking. Interestingly, our general impression from the analysis is that prior to the project, the students wrote about findings, proofs, and facts, while after the project the students tended to view research results as arguable. 8. DISCUSSION The main idea in this presentation has been to propose the use of open-ended experiments that include social aspects like publication and mutual criticism as a way of teaching about science as a process. By including the whole process of research in one project, we hoped to avoid the "atomised" teaching of the nature of science associated with the so-called "process approach". An important question to ask, however, is to what extent the Imitating-Science Project managed to imitate or illustrate main characteristics of science as a process and to develop students' understanding of these characteristics. Development and testing of hypotheses were emphasised in the planning framework. The analysis of students' texts also indicated that the important role of hypotheses in science became evident to some participating students. Today it is widely accepted that a plurality of methods is used in science. This project allowed students to design their own methods without being corrected by the teacher. However, the planning framework used put restrictions on the general design of the experiment. The implicit message was that research is based on the hypothetical-deductive method. Our analysis indicates that this implicit message had an impact on students' thinking. In order to illustrate the use of different methods in

S. DANKERT KOLSTØ, I. MESTAD

255

science, it is therefore recommended to present students with different frameworks throughout their science education. A clear weakness in this project was the lack of emphasis on theory. The students did not build their hypotheses on textbook science nor on prior research, and the project did not emphasise the need to relate conclusions to existing theories in order to show how results might add to theory development. Also, the complex relationship between theory and observations, e.g. the idea that observation is theory-laden, was not given very much attention. In order not to overload an Imitating-Science Project with learning objectives, it is probably wise to design separate curricula for such supplementary issues, and refer to such teaching activities during the project. In fact, Ntombela (1999) gives the advice not to link the problem in an investigation to a particular syllabus topic, as "this will be tantamount to overburdening the investigation with theory" (p.127). In this project we also tried to illustrate social processes in science through practical activities. However, only a few such processes were included. Publication was one of these, but only in a closed learning management system. One social process not included in the Imitating-Science Project was peer review prior to publication. The peer review process is important in science. It could be argued that the process was illustrated to some extent by the process which led to the joint reports, as these mainly emphasised those results which the class thought to be adequate and reliable. In a school setting the use of peers to judge a student's report as not worthy of publishing is not advisable, and even letting the teacher refuse publication might lead to a very unpleasant experience. Nevertheless, the teacher might review reports and demand improvements prior to publication. The role of the community of scientists, apart from peer reviewers, was sought, as illustrated through report-based criticism and argumentation during intra- and inter-class discussions. Science as a process includes other aspects than those included in this project. Examples here are how new researchers might take advantage of criticism of former studies and the process of consensus formation. Obviously it is impossible to illustrate all aspects of science in one project. A consequence of the above discussion, however, is recognition of some of the limitations of the project. If students are to develop an adequate understanding of science as a process, a prerequisite is the teacher's awareness of such limitations. 9. CONCLUDING REMARKS It is our conclusion that, even if there is room for improvement, the ImitatingScience Project had an impact on the students' thinking. In fact we also observed that most of the time the students enjoyed participating in the project and engaged actively in the different activities. We believe that a main reason for this engagement was that students had to be creative and make their own designs, evaluations, and decisions because of the open-ended character of their investigations. However, as Dewey (1913) emphasised, practical activities are not objectives in themselves. We believe that the most essential factor in enhancing student learning

256

LEARNING ABOUT THE NATURE OF SCIENTIFIC KNOWLEDGE

was their ongoing reflections on science as a process during the project, and the class discussion at the end of it. This finding echoes the results of earlier studies that also found the frequent use of higher-level questioning to be crucial (Lederman, 1992; Gess-Newsome, 2002). In a trial version of the project, experience-based reflection on science as a process was postponed until the end of the project. In that project, which was conducted by the same teacher, it was obvious that most of the students interpreted the answer to the research question as the main learning objective. In the project reported in this paper the teacher reminded the students of the main learning objective at the outset of most lessons. Thus, the class had several short discussions on the nature of science throughout the project. In the implementation of this project, students did not seem to confuse the learning objectives with their conclusions to the research question. In the introduction we emphasised the relevance of knowledge about science as a process for students' examination of socio-scientific controversies. This argument implies that teaching about science as a process needs to be followed by projects where current controversial socio-scientific issues are examined. This is paramount if we want students to gain competence in using knowledge about science as a process to perform thoughtful assessment of contested science-based arguments. REFERENCES Bell, R. L., Blair, L. M., Crawford, B. A., & Lederman, N. (2003). Just do it? Impact of a science apprenticeship program on high school students' understanding of the nature of science and scientific inquiry. Journal or Research in Science Teaching, 40(5), 487-509. Cole, S. (1992). Making Science. Between Nature and Society. Cambridge, Massachusetts: Harvard University Press. Cronbach, L. J. (1975). Beyond the two disciplines of scientific psychology. American Psychologist, 30(February), 116-127. Dewey, J. (1913). Interest and Effort in Education. Boston: Houghton Mifflin. Frost, R. (1995). The IT in Secondary Science Book. London: IT in Science. Gess-Newsome, J. (2002). The use and impact of explicit instruction about the nature of science and scientific inquiry in a elementary science methods course. Science & Education, 11(1), 55-67. Kind, P. M., Kjærnsli, M., Lie, S., & Turmo, A. (1999). Hva i all verden gjør elevene i realfag? Praktiske oppgaver i matematikk og naturfag. (What on earth do science and mathematics students do? Practical tasks in mathematics and science). Oslo: Department of Teacher Education and School Development. Kolstø, S. D. (2001). Scientific literacy for citizenship: tools for dealing with the science dimension of controversial socio-scientific issues. Science Education, 85(3), 291-310. Latour, B. (1987). Science in Action: How to Follow Scientists and Engineers through Society. Milton Keynes: Open University Press. Leach, J., & Paulsen, A. C. (Eds.). (1999). Practical Work in Science Education - Recent Research Studies. Dordrecht: Kluwer Academic Publishers. Lederman, N. (1992). Students' and teachers' conceptions of the nature of science: a review of the research. Journal of Research in Science Teaching, 29(4), 331-359. Merriam, S. B. (1998). Qualitative Research and Case Study Applications in Education. San Francisco: Jossey-Bass.

S. DANKERT KOLSTØ, I. MESTAD

257

Mestad, I. (2003). Planning framework, [WorldWideWeb]. Available: http://www.uib.no/people/pprsk/Dankert/Handouts/ Ntombela, G. M. (1999). A marriage of inconvenience? School science practical work and the nature of science. In J. Leach & A. C. Paulsen (Eds.), Practical Work in Science Education - Recent Research Studies. Dordrecht: Kluwer Academic Publishers. Ryder, J. (2001). Identifying science understanding for functional scientific literacy. Studies in Science Education, 35, 1-44. Strauss, A. L., & Corbin, J. (1990). Basics of Qualitative Research: Grounded Theory Procedures and Techniques. Newbury Park, California: Sage. Ziman, J. (1968). Public Knowledge: An Essay Concerning the Social Dimensions of Science. Cambridge: Cambridge university Press.

THE EFFECT OF USING DRAMA IN SCIENCE TEACHING ON STUDENTS' CONCEPTIONS OF THE NATURE OF SCIENCE SAOUMA BOUJAOUDE¹, SUHA SOWWAN², FOUAD. ABD-EL-KHALICK³ ¹American University of Beirut, Lebanon ¹University of Illinois at Urbana-Champaign, USA

ABSTRACT This study investigated the effect of using drama as a supporting learning strategy on students’ conceptions of the nature of science (NOS). Participants were 32 grade 10 and 11 students from a private all-girls’ school in Beirut, Lebanon. Fourteen students chose to participate in the extracurricular drama activity. The remaining 18 students were considered the control group and required only to attend the culminating performances. The drama group met for 36 hours over the course of 12 weeks to write scripts about the development of the concept of light using the work of four scientists. Data sources included open-ended questions about the tentative, empirical, and theory-laden NOS, group discussions, interviews, and researchers’ field notes and reflections. Results showed that the drama group students exhibited more informed views than the control group about the targeted aspects of NOS.

1. INTRODUCTION AND RATIONALE Unlike some typical methods of science teaching such as lecturing, use of drama to portray the lives of scientists may help students achieve meaningful learning – especially if they are required to research, write, edit, perform, and reflect on performances – and provide them with a more authentic sense of science and how it works, rather than giving them feelings of detachment, a state often experienced by science students (Bailey & Watson, 1998). It was expected that use of drama would portray the dynamics of the Nature of Science (NOS), particularly because textbooks and teachers often present science as a final product, ignoring its developmental nature (Anderson, 1987), and further, that drama would help to dismiss the myth of the “scientific method,” which is presented to students at the beginning of almost every science course (Gibbs & Lawson, 1992). By examining scientists’ lives and work and playing their roles, students might come to realize that scientists fail as much as they succeed, that an algorithmic or prescribed way for doing science is not readily available, and that science and scientists are not totally objective (McComas, 1996). 259 K. Boersma et al. (eds.), Research and the Quality of Science Education, 259—267. © 2005 Springer. Printed in the Netherlands.

260

THE EFFECT OF USING DRAMA IN SCIENCE TEACHING

In this study NOS refers to the values and assumptions inherent to science and the development of scientific knowledge (Lederman, 1992). While disagreements about a specific definition for NOS prevail among philosophers, historians, and sociologists of science, there is a level of generality regarding some aspects of NOS at which virtually no disagreements exist. These aspects are emphasized in current reform documents in science education (e.g., AAAS, 1990). The NOS aspects emphasized in this study were that science is tentative (subject to change), empirical (based on and/or derived from observing the natural world), and theory-laden (scientific knowledge and practices are influenced by scientists’ backgrounds, knowledge, training, theoretical commitments, and assumptions). Despite theoretical relevance, research on using drama in the science classroom is presently sparsely reported in the literature (e.g. Bailey & Watson 1998, Budzinsky, 1995). Consequently, the present study set out to investigate the effect of using drama on students’ conceptions of NOS. The study was guided by the following question: What is the influence, if any, of engagement with drama-related activities on students’ conceptions of the tentative, empirical, and theory-laden NOS? 2. METHOD Participants and Procedures Participants were 32 grade 10 and 11 students from a private girls’ school in Beirut, Lebanon. Fourteen students participated in the extracurricular drama activity while the remaining 18 were considered a control group and were only required to attend the culminating performances. The drama group met for three hours on Friday mornings for 12 weeks. Participants worked in heterogeneous groups that included high, low, and average achievers. Before the study was implemented, participants were instructed about writing a play. Two language teachers (Arabic and English) with experience in drama supervised the writing/editing part of the activity along with one of the researchers. Participants were divided into four groups and were asked to choose one from a list of scientists whose work they were studying or had recently studied in science. The scientists selected were Archimedes, Al-Hasan Ibn Al-Haitham, Newton, and Edison, the development of the concept of light in the work of these scientists was the focus of the drama activity. Participants were asked to pay attention to, and reflect on, NOS and scientific knowledge when writing the scripts. They were provided a set of questions to consider, such as, how open to change were scientific theories and knowledge, how “objective” was 'their' scientist, and how much influence did a scientist’s background have on his work. All drama students were expected to engage in conducting research, writing, editing scripts, providing feedback to each other, performing the play, and reflecting on their work. Following script writing, a researcher led discussions in each of the drama groups in order to integrate explicit and reflective elements about NOS into students’ drama activities (Abd-El-Khalick, Bell & Lederman, 1998). In these discussions, students were explicitly asked the same questions about NOS that were supposed to guide their writing of the scripts. Following the performance of the play, all 14 drama group students were involved in

S. BOUJAOUDE, S. SOWWAN, F. ABD-EL-KHALICK

261

a discussion about the NOS aspects that seemed to cut across the work of the four scientists portrayed in the play. All discussions were audiotaped and transcribed. Also, one of the researchers kept field notes and reflections on each discussion session. Instruments Three open-ended questions modeled after Lederman and O’Malley (1990) were used to assess student views of the three target NOS aspects. Open-ended questions ensure that respondents are given the opportunity to explicate their own NOS views rather than impose a certain conception of NOS on them (Lederman, Abd-ElKhalick, Bell, & Schwartz, 2002). The questions were pilot-tested with 23 grade 10 (non-participant) students in the same school, modified accordingly, and then administered to all participants both prior to and at the conclusion of the study. Following each administration of these questions, a group of seven (4 drama and 3 control group) participants were individually interviewed. They were asked to explain their responses to the NOS questions along with answering follow-up and probing questions. All interviews were audiotaped and transcribed for analysis. The NOS questions were: (1) Do you think that your grandchildren will be studying the same scientific knowledge that you are studying now? Why or why not? (2) Although a physician told you it would take you a week to heal, you healed in two days, just like your horoscope said. Which is more scientific, the physician’s or the horoscope’s claim? Explain by stating what makes a statement scientific or nonscientific. (3) Although scientists generally use the same sets of data, they come up with different explanations. For example, Newton said that gravity affected an object’s rate of falling, but Aristotle said it was the object’s weight. How would you explain these differences? Data Analysis Three categories were used to rate students’ responses: Naïve, transitional, and informed. (1) Students categorized as having naïve NOS conceptions acknowledged the possibility of scientific knowledge changing, but attributed change solely to the incremental addition of new knowledge without the possibility of rejecting existing claims. They considered a statement as “scientific” based on its source (e.g., a physician) and did not refer to empirical observations or experiments as being necessary for making a claim “scientific.” These students did not recognize the influence of scientists’ backgrounds or theoretical commitments to their work; these students often attributed differences in interpreting the same set of data to scientists’ use of different methods, their entitlement to their own opinion (without regard to evidence), or the fact that some scientists were “right” while others were “wrong.” (2) Students categorized as having transitional (or partially informed) NOS conceptions articulated somewhat informed views but were unable to explicate, justify, or defend those views. For instance, they believed that existing scientific knowledge could change through “evolution” without further explication as to what would “evolve” and why. To them, statements were scientific because they were “facts” or based on facts. They could not, however, articulate the nature or origin of

262

THE EFFECT OF USING DRAMA IN SCIENCE TEACHING

these facts (e.g., experimental or observational data). They referred to the influence of scientists’ “time” or environment as factors that influenced observations. However, they still seemed to believe that differences in interpreting data were instances of erroneous science as compared to being integral to scientific research. (3) Students categorized as having informed NOS conceptions believed that scientific knowledge was tentative, yet durable. They indicated that scientific statements were those based on evidence, observation, or experiment rather than on superstitions or opinions lacking evidentiary support. Finally, they acknowledged the influence of scientists’ theoretical backgrounds (e.g., prior knowledge and expectations) on their observations and interpretation of data. Student responses to each NOS question were examined separately, using data from questionnaires, interviews, and group and whole discussions. The researchers initially rated student responses independently and then compared and contrasted their ratings in order to establish inter-rater reliability, which was initially found to be 0.80. Differences were resolved by consulting the data or by consensus. Percentages of responses falling under each category were then calculated. Finally, all data were analyzed for emergent themes. Using these data, a profile for each group member’s NOS conceptions was prepared in the form of a spreadsheet and compared to those of students in the other group. The individualized spreadsheets created a concise reference that kept track of students’ progress in terms of their NOS views. 3. RESULTS Pre-Drama NOS Conceptions Initially, a majority of participants in both groups held naïve conceptions of the target NOS aspects. Tables 1 and 2 present the percentage of participants in the three categories. In the following sections, the letters “D” and “C” refer to participants in the drama and control groups, respectively. The numbers following these letters are used to refer to individual participants. Drama group (n=14)

Category

Tentative NOS (Question 1) %Pre % ǻ* Post

Empirical NOS NOS (Question 2) % % ǻ Pre Post

Theory-laden NOS (Question 3) % % ǻ Pre Post

Naïve

79

21

–58

29

29

0

50

14

–36

Transitional

14

43

+29

21

14

–7

14

14

0

Informed

7

36

+29

50

57

+7

36

72

+36

*Note. ǻ = %post – %pre or the change in the percentage of participants under this category.

S. BOUJAOUDE, S. SOWWAN, F. ABD-EL-KHALICK

263

Table 2. Control Group’s Pre and Post-Drama Views of the Tentative, Empirical, and Theory-laden NOS Control group (n = 18) Tentative NOS (Question 1) % % ǻ Pre Post

Empirical NOS (Question 2) % % ǻ Pre Post

Theory-laden NOS (Question 3) % % ǻ Pre Post

Naïve

89

83

–6

55

50

–5

73

67

–6

Transitional

11

11

0

17

10

–7

21

17

–4

Informed

0

6

+6

28

40

+12

6

16

+10

Category

*Note. ǻ = %post – %pre or the change in the percentage of participants under this category. Tentative NOS. The larger majority of the drama (79%) and control (89%) group participants ascribed naïve conceptions to this NOS aspect. They believed that knowledge produced by scientists was final and not liable to change. Knowledge grows by additions to what we already know to be “true”: I think that my grandchildren will not only be studying the same materials in science, but also some new things because science may develop in the coming years and because new technology help scientists in discovering new facts (C8, pretest).

On the other hand, 14% of the drama and 11% of the control group participants started out with what could be considered a transitional view of the tentative NOS. These participants acknowledged that existing scientific claims might change but only because of scientists’ opinions or mistakes. Thus, change occurs because scientific knowledge “is being corrected”, beyond which what we know will not change: I think that my grandchildren will be studying the same materials in science that I am now studying because science is something to be continued. Everything was invented or discovered and scientists are now “correcting” it or adding something new to it. (D4, pretest)

Empirical NOS. When asked to evaluate the scientific status of a statement made by a physician versus that made by an astrologer, 97% of all participants indicated that the former was scientific and the latter superstitious. Participants, however, differed in their justifications of this judgment. Some justifications reflected informed views of the empirical NOS. Indeed, 50% of the drama and 28% of the control group participants argued that the reliance of physicians on evidence renders their statements “scientific”: Astrologers rely in their writings or sayings on chance and lies. They don’t rely on logic or studying evidence and proof. When looking at doctors’ and scientists’ work, we find

264

THE EFFECT OF USING DRAMA IN SCIENCE TEACHING that they rely on events and well-trusted, dependable studies and experiments. A person should therefore let his mind control his actions and not be convinced with anything other than facts, evidence, and proofs (D12, pretest).

Also, 21% of the drama and 17% of the control group participants provided partially informed (transitional) responses, whereby they acknowledged that a physician’s statement was the scientific of the two, but could not provide articulate justifications: “The horoscope is only superstitions and everybody says something in horoscopes. In addition, I don’t believe in horoscopes” (C8, pretest). The justifications provided by the remaining participants appealed to the authority of science: “The doctor [is scientific] because he studied science, but the astrologer didn’t” (C9, pre-interview). These students (29% of the drama and 55% of the control group), however, did not refer to the use of observational or experimental evidence as a basis for the status of scientific knowledge which they believed was derived from, for example, authoritative texts: “The doctor is right most of the time because he knows that what he’s saying is due to what he . . . learned in books that he has studied at the university” (D9, pre-interview). Theory-laden NOS. When asked to account for differences in scientists’ interpretations of the same data, 36% of the drama and 6% of the control group participants expressed informed conceptions of this NOS aspect. They acknowledged that scientists’ theoretical backgrounds and training influence their inferences: I think that differences are due to the way scientists think and how impressed they are with that experiment, and I think it is due to their knowledge and experiences. (D4, preinterview)

Concurrently, 14% of the drama and 22% of the control group participants provided transitional or partially informed conceptions, whereby scientists living in different times could come to different conclusions starting with the same set of data: “Because every scientist . . . lives in a different environment and at a different age because the age affects every scientist in his thinking” (D9, pretest). Students who held more naïve views of this NOS aspect differed in their explanations as to why scientists come to different conclusions starting from the same set of data. Half the participants in the drama group argued that some scientists are simply more “scientific” than others. In a sense, scientists differ in their inferences because some do “good” science while others “do” bad science: Every one in this world has his point of view, and no one can make him change it even the scientists. Maybe Aristotle depended on observation only with no experiments. When Newton saw the apple falling to the ground he made many experiment to prove his point of view. He threw some bodies upwards and saw that they are always attracted to the ground, and this is how he proved his theory about gravitation and this is because he relied on experiments and proofs, but Aristotle didn’t. (D13, pretest)

A majority of the control group participants (73%) held naïve views of the theoryladen NOS. Most participants (56%) argued that differences in interpretations were because “every scientist is entitled to his point of view” (C14, pretest). The data did not seem to play a role in this conception of how scientists interpret evidence. The

S. BOUJAOUDE, S. SOWWAN, F. ABD-EL-KHALICK

265

remaining students (17%) indicated that scientists reach different conclusions because “Every scientist has different views about the data they have, despite the fact that they have the same information, but maybe Aristotle didn’t give this explanation enough study to reach the result that Newton reached” (C7, pretest). Post-Drama NOS Conceptions As is evident in Tables 1 and 2, several changes were observed in participants’ views of the target NOS aspects. There were some positive shifts in the control group students’ views of the tentative (+6%), empirical (+12%), and theory-laden (+10%) NOS. These changes, however, were not substantial. Similarly, minor positive changes were evident in the drama group participants’ views of the empirical NOS (+7%). By comparison, substantial changes were evident in the drama group students’ views of the tentative (+58%) and theory-laden (+36%) NOS. The nature of these changes is explicated in the following sections. Tentative NOS. The larger majority of all participants started out with naïve conceptions of the tentative NOS. Following engagement with the drama activities, 29% more students in the drama group exhibited informed views of this aspect, compared to only 6% more students in the control group: I don’t think that my grandchildren will be studying all the facts, theories, and laws we are now studying. Maybe some will change and others will remain the same. But I believe that science is always changing . . . Science wasn’t the same in all eras. There were always scientists who suggested new theories and rejected the old ones. (D10, posttest)

Additionally, the conceptions of another 29% of the drama group students shifted from the naïve to the transitional category. While recognizing that scientific knowledge might change, these students still believed that some aspects of this knowledge would never change: I believe that our grandchildren will use the same basics, which we are using now and have been used before . . . when a scientist, e.g., wants to do an experiment it will be possible that he might use different tools but he will depend on the basics we are depending on now. In addition to this he might find out that some need partial correction and some are correct. (D12, posttest)

The posttest responses of 21% of the drama students were naïve and virtually identical to the views they expressed at the outset of the study: “My grandchildren will be studying the facts, theories, and laws that I am studying now, but they definitely will be studying some other new things that we may not even know now” (D11, posttest). Finally, the posttest responses of 58% of the drama students revealed favorable changes in terms of the tentative NOS. In contrast, the majority of the control group students (72%) still ascribed to an absolutist view of scientific knowledge. They believed that scientific claims would not change “because science is a fact” (C18, posttest). Empirical NOS. Minimal change was observed in both drama and control group students’ views of this NOS aspect (Tables 1 and 2). The views of only one student

266

THE EFFECT OF USING DRAMA IN SCIENCE TEACHING

shifted from the transitional to the informed category. No change was evident in the views of the 29% who started out with naïve views of the empirical NOS. These students still considered authority to be the basis of scientific status: “A statement is considered scientific if it were (...) taken from a source of total knowledge about the subject” (D8, posttest). In the control group, the views of one student shifted from the naïve to the informed category, while the views of another shifted from the transitional to the informed category. Theory-laden NOS. About 36% more students in the drama group articulated informed views of the theory-laden NOS. This change was substantial especially because all students who experienced it initially held naïve views. When explaining why scientists reach different conclusions starting with the same set of data, one student noted, “I think that they had different backgrounds and because they had different aims and thoughts before they made this observation. Maybe each one looked at this phenomenon in his own way and with his own background” (D4, posttest). By comparison, one of the control group students shifted from a naïve to an informed conception, while another shifted from a transitional to an informed conception. 4. DISCUSSION AND CONCLUSIONS The present results support using drama activities to bring about positive changes in student views of some NOS aspects. A majority of students started out with naïve conceptions of the three target NOS aspects. The drama group students showed most favorable changes in their views of the theory-laden and tentative NOS. These students’ intensive contact, with the selected scientists’ work, personal life stories, and underlying motives and aims, might have contributed to their appreciation of the influence of scientists’ backgrounds and theoretical commitments to their “scientific” activities. At the end of the study, it was clear that more students in the drama group did not ascribe to outdated views of the target NOS aspects. Tracking the historical development of ideas about a single topic (light in the present case) through the works of several scientists seems to have helped the drama students develop a sense of the tentative NOS, whereby certain “established” claims are sometimes discarded and replaced with different claims, partly because of new evidence and technology, but also because of theoretical advances and the resultant reinterpretation of extant evidence. The empirical NOS was the aspect that showed the least change. This result could be, at least partially, explained by the fact that the explicit prompts designed to help the drama group students discuss and reflect on the target NOS aspects were more focused on the tentative and theory-laden than on the empirical NOS. The present results indicate that using drama activities to portray scientists’ work and lives, coupled with structured reflection opportunities, could substantially contribute to helping students internalize more informed views of selected NOS aspects. The study indicated that focusing on the development of a single topic or phenomenon (e.g., light) can facilitate the achievement of this goal because focus

S. BOUJAOUDE, S. SOWWAN, F. ABD-EL-KHALICK

267

reduces ambiguity and helps students monitor changes in the “scientific” conceptions of the target topic or phenomenon over time. Moreover, the discussion/reflection sessions, especially the whole group sessions that took place after the final performance, served a very pivotal role in influencing students’ NOS views. These discussions and reflections were guided by explicit prompts and questions targeting specific NOS aspects. Students were encouraged and guided to analyze and discuss their views, and provide support for their positions from their researching, writing, editing, and performing experiences. These sessions created a context that made the discussions explicit enough for students to grasp the intricacies of some NOS aspects (see Abd-El-Khalick, 1998; Abd-El-Khalick et al., 1998). REFERENCES Abd-El-Khalick (1998). The influence of history of science courses on students’ conceptions of the nature of science. Unpublished doctoral dissertation, Oregon State University, Oregon. Abd-El-Khalick, F., Bell, R., & Lederman, N. G. (1998). The nature of science and instructional practices: Making the unnatural natural. Science Education, 82, 417-436. American Association for the Advancement of Science. (1990). Science for all Americans. New York: Oxford University Press. Anderson, C. (1987, May). The role of education in the academic disciplines in teacher preparation. Paper presented at the Rutgers Invitational Symposium on Education: The Graduate Preparation of Teachers. New Brunswick, NJ. Bailey, S., & Watson, R. (1998). Establishing basis for ecological understanding in younger pupils: A pupil evaluation of a strategy based on drama/role play. Science Education, 20, 139-152. Budzinsky, F. (1995). Chemistry on stage: A strategy for integrating science and dramatic arts. School Science and Mathematics, 95, 406-410. Gibbs, A., & Lawson, A. (1992). The nature of scientific thinking as reflected by the work of biologists and by biological textbooks. The American Biology Teacher, 54, 137-152. Lederman, N., & O’Malley, M., (1990). Students’ perceptions of tentativeness in science: Development, use, and sources of change. Science Education, 74, 225-239. Lederman, N. (1992). Students’ and teachers’ conceptions of the nature of science: A review of the research. Journal of Research in Science Teaching, 29, 331-359. Lederman, N. G., Abd-El-Khalick, F., Bell, R. L., & Schwartz, R. (2002). Views of nature of science questionnaire (VNOS): Toward valid and meaningful assessment of learners’ conceptions of nature of science. Journal of Research in Science Teaching, 39(6), 497521. McComas, W. (1996). Ten myths of science: Reexamining what we think we know about the nature of science. School Science and Mathematics, 96, 10-15.Table 1. Drama Group’s Pre and Post-Drama Views of the Tentative, Empirical, and Theory-laden NOS

THE RELEVANCE OF TEACHING ABOUT THE "NATURE OF SCIENCE" TO STUDENTS OF THE HEALTH SCIENCES

SVERRE PETTERSEN Akershus University College, Norway

ABSTRACT This paper argues for the significance of teaching about the “nature of science” to students of Health Sciences in Norway. The national Health Sciences’ curricula contain the core subjects of biological sciences as well as the philosophy of science and research methods. Biological science research has a large influence on the evolvement of professional knowledge in the Health Sciences. However, it is likely that Health Science graduates become involved in occupations in which they are exposed to lay health claims, pseudoscience, and comparative-alternative medicine counselling. The epistemologies of normal science and alternative-comparative medicine are largely different. In a scientific evaluation-test of health claims, most of the tested Health Science students failed. In a questionnaire, many students expressed quite “ambivalent relationships” with the aims of scientific research, and their consideration of “what counts as reliable knowledge” was to some extent non-scientific. Most students expressed positive attitudes towards the use of comparative-alternative medical treatments. For students to be able to achieve skills to critically evaluate health claims, teaching about the “nature of science” might be significant, especially within the core subjects of the Health Sciences.

1. INTRODUCTION In reforms taking place in the late 1980's and early 1990's, education in the Health Sciences (HS) in Norway implemented a new course of study that included The philosophy of science and research methods (PS/RM) as a part of the core curriculum together with the Biological Sciences (BS). The HS in Norway include, for example, nursing, physiotherapy, social educator (caring and nursing requirements of mentally retarded persons integrated in society), and radiography. The BS curriculum consists of human biology subjects including physiology, microbiology, and pathology, and its importance is indisputable for HS professional studies. In the field of nursing and physiotherapy, background knowledge of physiology and anatomy is of uttermost importance for proper action (Roskell, Hewison & Wildman, 1998). Social educators must know the possible mental and physical expressions of common human genetic syndromes, and radiographers have to know the benefits and risks of x-ray exposure to human tissue. However, since high-school science is not a compulsory entrance requirement for most HS programmes, interest in science does not seem to be the main reason for students 269 K. Boersma et al. (eds.), Research and the Quality of Science Education, 269—282. © 2005 Springer. Printed in the Netherlands.

270

THE RELEVANCE OF TEACHING ABOUT THE ‘NATURE OF SCIENCE’

choosing this type of education; no doubt, it is quite the contrary (Kersten, Bakewell & Meyer, 1991). Some students without senior high school qualifications for university college studies can get entrance to the HS after evaluation of their previous health work practice. In the HS in Norway, physicians, human biology researchers, and pharmacists are commonly used as teachers of the BS. They are often professionals lacking pedagogical training, and many do not have HS institutions as their main place of work (Pettersen, 2003a). To be able to understand and apply new health research findings, knowledge in the BS is important (Trnobranski, 1996; Pettersen & Solberg, 2003). However, it has been claimed that social and behavioural sciences are increasingly displacing the BS in international health education (Trnobranski, 1996). In addition to the continuous decline of the number of BS lessons taught in the HS, investigators have also called attention to the risk of omitting the teaching of the epistemology of science and science research methodology in the HS. This is a concern since HS graduates are likely to become involved in occupations in which they are frequently exposed to reports of scientific research, lay health claims, pseudoscience,1 and comparativealternative medicine2 (CAM) counselling. When graduate HS students do not have satisfactory skills for recognizing “true” scientific health messages and discriminating them from pseudo- and non-scientific messages, the scientific foundation of professional HS education and work practice could be undermined. One of the teaching goals of PS/RM addresses science epistemology: (students should) acquire scientific insight and scientific methods in order to read research material and utilize research findings in their work (ODIN, 2003). In principle, this teaching goal might be equivalent to the teaching goals of the “nature of science”, NOS (Efflin, 1999; Hogan, 2000). In broadest terms, the meaning of the concept NOS is those ideas someone has about science, rather then a person's scientific knowledge: -

how the body of public knowledge called science has been established and is added to; what our grounds are for considering it reliable knowledge; how the agreement which characterizes much of science is maintained (Driver et al. 1996).

The latter involves an understanding of the social organization and practices of science (social context), whereby knowledge claims are transmitted into public knowledge, and of the influence of science on the wider culture – and vice versa. Issues surrounding the application of scientific knowledge in practical situations are an important focus, as the lack of consensus about these invites a re-evaluation of 1

Pseudoscience could be characterized by the use of theories which cannot be tested, the use of ad hoc hypotheses and the selective use of data presenting anecdotes and myths as evidence (Giuffre, 1997). 2 Complementary medicine refers to 'not scientifically proven' therapies that are given in addition to conventional therapy (e.g. herb tea with antibiotics for pneumonia, and therapeutic touch), whilst alternative medicine often consists of therapies or remedies that are used alone in place of conventional therapy, e.g. zone therapy, healing, and homeopathy (Cassileth, 1999).

S. PETTERSEN

271

claims about the status of particular kinds of knowledge. A related issue is about the purpose of scientific work (in seeking explanation) and the boundaries of its areas of interest. Another, quite distinct argument, that an understanding of the NOS is useful, underpins the so-called “process approach” to science (Millar & Driver, 1987). This identifies the NOS with a method of inquiry, through claims, for example, that the essential characteristic of education in science is that it introduces students to the methods of science. In this view, science is portrayed as a powerful, and quite general, method of inquiry that can be learnt and then used in a wide range of other contexts, both scientific and non-scientific (Driver et al., 1996). Therefore, to achieve the goal of the PS/RM subject, it might be important to teach HS students about the NOS. Student knowledge about the NOS and their critical thinking skills are connected (Sadier, Chambers & Zeidler, 2002). During the last fifteen years, student achievement of “critical reflection skills” has been a superficial teaching goal in HS curricula in Norway (Pettersen, 2003a). According to Watson and Glaser (1980), critical thinking includes attitudes of inquiry that involve an ability to recognize the existence of problems and an acceptance of the general need for evidence in support of what is asserted to be true – which probably reflects the aim that the knowledge taught to students of the HS should essentially be based on scientific evidence. HS students' capability to request important and necessary scientific information on health claims (e.g. newspaper health news, and CAM information and counselling), especially when the claims invoke a BS-based methodology, can probably be telling reflections of what HS educators have taught them – or failed to teach them – about science epistemology and about the NOS (Norris, 1994; Korpan et al., 1997; Pettersen & Solberg, 2003). The following research question emerged as a result of examining the analyzed data of a national survey of nursing, physiotherapy, social educator, and radiography student samples, and of PS/RM and BS teacher samples: What research findings might legitimize the teaching about the NOS to HS students in Norway? 2. MATERIAL AND METHODS

The questionnaire development, administration procedure, and response rate The number of items on the questionnaires was 254 for HS students, 105 for BS teachers, and 129 for PS/RM teachers. In addition to questions about 3rd Year HS students’ age, gender, and pre-college background, the student questionnaire also contained: (1) four highly scientifically-deficient health news briefs whose aim was to measure respondents' skills in scientifically evaluating health claims, (2) Likertscale statements about their attitudes towards science research, “what counts as reliable knowledge”, and CAM, respectively, (3) Likert-scale statements about respondents' “beliefs about paranormal phenomena”, and (4) Likert-scale questions

272

THE RELEVANCE OF TEACHING ABOUT THE ‘NATURE OF SCIENCE’

about their NOS-topic-related learning experiences. The two teacher questionnaires also contained descriptive questions (age, gender, and qualifications), while Likertscale statements and questions exploring the teachers’ CAM attitudes and NOS teaching strategies predominated in the questionnaires. The PS/RM teacher questionnaire had a special focus on epistemology teaching, while the BS teacher questionnaire had a focus on the teachers’ experience with teaching the biological sciences in HS-education. The three questionnaires were slightly altered after a pilot study analysis. During the spring of 2001, all Norwegian colleges in the four HS categories were invited to participate in this study. The eligible number of colleges was as follows: nursing, 31; physiotherapy, 4; social educator, 8; and radiography, 4. The number of responding students was: nursing, n = 317; physiotherapy, n = 63; social educator, n = 59; and radiography, n = 34. Thus, a total of N = 473 HS student participants were sampled from 16, 3, 5, and 2 of the eligible number of colleges, respectively. The response rate of the HS student questionnaire was rather low: only 20-33% returned it after two reminders. The total number of responding PS/RM and BS teachers in the four HS areas was N = 47, and N = 59, sampled from 15, 4, 8 and 3 of the eligible number of colleges. In this paper, only results from BS teachers of physiology (N = 35) are presented and discussed. The response rate of the PS/RM and BS teacher questionnaires was higher than for the HS student questionnaire; 41-88% and 3371%, respectively, were returned after two reminders. Because of the non-probabilistic nature of the sampling (invited participants, not randomly selected), and the low response rates on the questionnaires, extensive use of sample statistics and result generalizations have been avoided. 3. RESULTS AND ARGUMENTATION

HS students’ minor pre-university college qualification in the philosophy of science and research methods In Norway's compulsory senior high school education, there is probably no subject or topic that addresses either the teaching of the philosophy of science or research methods. Therefore, it is possible that most students registered for the HS do not know the ways in which the evolvement of science knowledge can be distinguished from other sources of human knowledge. Students might have some conceptions of the social context of (biological) science research (where and by whom), but most likely they do not know about the science processes (how) which largely define the validity standards of scientific knowledge (Millar & Driver, 1987) and/or the “hard” road to consensus in the scientific community (Kolstoe, 2000). Teaching about science processes is intrinsic to teaching about the NOS; it might be important to take this approach when teaching the philosophy of science.

S. PETTERSEN

273

HS students’ minor qualification in senior high school science Immersion in second and third year senior high school science (biology, chemistry, and physics courses) is not a mandatory entrance requirement for HS education in Norway. About one third of the physiotherapy students in the study, one fifth of the radiography students, one sixth of the nursing students, and one tenth of the social educator students have taken advanced high school courses in biology, chemistry, and physics. Of the participating nursing and social educator students, 10% do not even have the formal senior high school qualifications for university college studies (Pettersen, 2003b). The importance of the biological sciences in the HS During the last twenty years, the teaching of the BS has declined in favour of the social and behavioural sciences (Trnobranski, 1996; Pettersen, 2003b). However, the number of science subject ECTS credits in a bachelor study programme (180 credits) of the four areas of HS education studied is still substantial: nursing, 45 credits; physiotherapy, 72 credits; social educator, 21 credits; radiography, 108 credits (Pettersen, 2003b). Of the participating students, 42% of nursing, 33% of physiotherapy, 30% the radiography, and 19% of the social educator students “agreed” (4+5 and 1+2 Likert scale answers collapsed) to the statement about science knowledge and skills being more important than social science knowledge and skills, but most students were “undecided” (42-55%) on this point. However, the majority of the PS/RM and physiology teachers “agreed” to this statement (54% and 57%, respectively). Most physiology teachers “agreed” that the BS should receive more hours of teaching (71%), whilst this was not the case for PS/RM teachers (16%) and HS students (7-35%). Among the 47 PS/RM teachers in the study, only two had an academic qualification in science, the others were qualified in the health, social, or the behavioural sciences (Pettersen, 2003b). These results probably demonstrate many HS students’ problematic relationship with science: whilst BS knowledge seems to be important for student work practice, such knowledge achievement might seem to be very uninteresting to the non-science HS students. More hours available to teach the BS might simultaneously lead to curriculum inclusion of additional (and more complicated?) BS topics, which may not be desirable to these HS students. However, physiology teachers assert this increase to be necessary in order to improve the quality of HS students. The huge and steady flow of health claims in society makes it just as important to teach HS students critical evaluation skills. By teaching students about the NOS, such skills could be achieved. HS students’ lack of scientific evaluation skills Students’ ability to request scientific information (organized by topic code) for scientifically deficient health news reports could be a measurable outcome of the teaching goal of the PS/RM subject mentioned above. In a hierarchical manner, an evaluative scientific report might contain information relating to the following six scientific topic code category descriptions of the research: Social context,

274

THE RELEVANCE OF TEACHING ABOUT THE ‘NATURE OF SCIENCE’

Theory/Agent, Method(s), Data/Statistics, Related research, and Relevance (Korpan et al., 1997). However, 33-47% of the HS students studied (N = 473) did not request any of these scientific topic codes in four highly scientifically-deficient news briefs (Figure 1; for methodological details see Pettersen & Solberg, 2003). THERAPEUTIC TOUCH RELIEVES PAIN Using Therapeutic Touch is an important treatment of migraine. British scientists have reported that people suffering from migraine receiving the Therapeutic Touch treatment experienced a noticeable pain relief. Intensive and durational headaches are problematic for these people. The Migraine Association has hailed this important research finding. (Pettersen & Solberg, 2003) Figure 1. An example of a fictitious, scientific topic code-deficient news brief 28-33% of the students requested information in only one of the scientific topic areas in each of the four news briefs, 17-23% requested two areas, and 7-13% requested three; just 0.4-3% of the HS students requested information in four of the six possible scientific topic areas. HS students, who had immersed themselves in science during their senior high school period requested more scientific areas than HS students who had not (not significant) (Pettersen & Solberg, 2003). The mode was just one scientific topic request across all four news briefs, while one third of the students made no requests. This led to the suspicion that the teaching of critical evaluation of published health claims had probably not been emphasized in the four HS programmes studied. This seemed especially the case when the publications invoked a natural science-based methodology. The students probably did not know what to ask for, aside from information in the three scientific topic code categories, Theory/Agent, Method, and Data/Statistics, which were most frequently requested (but only by one third of the HS-students). Few HS students requested information about Social Context and Related Research. These findings may reflect a lack of sensitivity to the fact that scientific research takes place within a social community which can influence the selection of research questions, the interpretation of results, and the acceptance of research findings and theory. The rather disappointing results of this student evaluation test probably indicate that teaching of NOS topics is relevant to the HS student respondents. HS students’ ”ambivalent relationship” with science research Table 1 shows the six-statement construct that was developed to gain information on HS students’ “attitudes towards science research”. The analysis and the rather low Cronbach alpha = 0.56 of this construct probably showed that many students have an “ambivalent relationship” to science research activities.

S. PETTERSEN

275

Table 1. Health Science students’ (N = 473) attitudes towards science research activities Statements % “Agree” “Scientists are insensitive and remote persons.” 5 “Science research activities do not express closeness to being human.” “The aim of science research is to improve human life.”

15

“Products of science research cause human cultural progress.”

14

“In the future, science researchers will be able to find explanations to the most serious and fatal human diseases.” “In the future, science researc human diseases.”hers will be able to develop cures for the most serious and fatal

79

36

83

The number of 4 (“agree”) and 5 (“strongly agree”) Likert scale answers to each of the six statements were collapsed and recoded into a new variable named “agree”. To explore the PS/RM teachers' emphasis on social and natural science topics in their teaching, a Likert scale question was provided (1 = include to a very small extent; 5 = include to a very large extent). The analysis showed that social science topics seemed to be significantly (p < 0.01) more emphasized than natural science topics [3.90 ± 1.13 versus 3.17 ± 1.09 when teaching the PS (n = 30), and 3.74 ± 1.24 versus 2.84 ± 1.21 (n = 31) when teaching RM]. This finding was also confirmed in subsequent interviews of PS/RM teachers (N = 27). However, the teaching of social science seemed to be synonymous with the teaching of qualitative research methodology in which most of the explored HS teachers of the PS/RM might have the greater experience (Pettersen, unpublished results). The teaching of the NOS to HS student respondents could supply them with a fairer and more nuanced “picture” of scientific research activities, and possibly make them more capable of analyzing and evaluating health research reports. HS students’ “non-scientific” and “undecided” views of “what counts as reliable knowledge” Discussions of science epistemology must involve the mention of the NOS (Bell, Lederman & Abd-El-Khalick, 2000). In Norway, science epistemology did not seem to have high ”status” among PS/RM teachers of HS programmes (Pettersen, 2003b). Dissemination of any epistemology to HS students is probably ideologically founded (Caldwell, 1997). Educational ideologies express themselves, for example, by what kind of “stuff” you plan to teach, how you mention it, and what “stuff” you consciously delete. Although the kind of knowledge to rely on might be strongly dependent on the context in which it is applied, the HS student respondent answers to the five Likert scale statements showed that some students have rather nonscientific conceptions of what counts as reliable knowledge (the percentage of students who “agreed” – by collapsing 4+5 Likert scale answers – are marked in

276

THE RELEVANCE OF TEACHING ABOUT THE ‘NATURE OF SCIENCE’

brackets): “natural science research” (23%), “elderly people’s experience in life” (24%), “reliable knowledge varies in time, cultures and contexts” (12%), “spiritual experiences” (3%), and “intuition”(3%). The dispersion of “undecided”-answers were 50%, 54%, 35%, 31%, 36%, respectively, which probably indicates strong student uncertainty on this matter. That just one fifth of the HS students seemed to rely on science research, is interesting. Whether this attitude results from ideologies expressed by teachers of the PS/RM subject is just a matter of speculation. The “undecidedness” or uncertainty of most HS student respondents on “what counts as reliable knowledge”, and their low confidence in science, probably calls for more emphasis on teaching about the NOS in the HS. HS students’ and HS teachers’ positive attitudes towards CAM therapies Table 2 shows the percentage of the responding HS students and of the PS/RM and physiology teachers of the four HS programmes who had a positive attitude towards 15 CAM treatments. However, many respondents were not familiar with some of these CAM treatments (note the high frequencies of m, which are collapsed “missing” and “I don’t know” answers). Both the HS students and the PS/RM teachers were quite positive towards CAM, whilst the physiology teachers seemed not to be as positive. These CAM attitude differences could probably be linked to different epistemological views evolved through the participants’ own higher education (Pettersen & Bye, 2003). The epistemologies of science and CAM are largely different (Sampson, 1995). In regression analysis, HS student beliefs in the following 9 paranormal phenomena: prophecies, star sign, horoscope, mascot, séances, ghost, clairvoyance, mental contact with the dead, and telepathy (Cronbach alpha = 0.79 for the construct), strongly predicted (p < 0.01) their attitudes towards the CAM treatments (Pettersen, unpublished results). The HS student and PS/RM teacher answers to the following two Likert scale statements: “CAM should be taught to students of the Health Sciences” and “The professional leadership of Health Science educations should be more receptive to include CAM in the curricula” showed that the HS students were slightly more “agreed” on both statements than the PS/RM teachers (56% and 59% versus 40% and 57%). The question of incorporating CAM into scientific, evidence-based programmes and practices probably requires much wider debate: HS students might not learn CAM simply as 'techniques', whilst their teachers need to think carefully about how they approach the different epistemological perspectives of CAM and conventional medicine. This provides an ideal opportunity to discuss these differences when teaching the HS students about the NOS. NOS topics possibly related to the teaching goal of PS/RM is not strongly emphasized in the HS

S. PETTERSEN

277

Table 2: The fraction (%) of the responding HS-students’ and HS-teachers’ positive attitude towards 15 alternative-complementary medicine (CAM) treatments CAMtreatments

Nursing (N = 317)

m

7

% pos 82

m

1

% pos 91

m

3

% pos 97

84

40

80

14

60

24

Zone therapy

85

23

58

4

85

Light therapy

83

66

94

13

Homeopathic medicine

76

53

45

Aroma therapy

74

22

Herbal medicine

71

Kinesiology

HS-TEACHERS PS/RM Physiology (N = 47) (N = 35)

m

0

% pos 95

m

5

% pos 85

27

12

88

22

31

9

6

85

1

54

12

39

2

63

10

69

8

79

19

39

12

12

65

10

54

6

68

10

38

1

50

5

67

5

81

3

50

9

19

3

55

48

17

81

12

63

4

55

16

36

4

54

157

29

32

73

33

38

18

38

23

15

8

Osteopathic medicine

39

228

58

30

59

42

29

27

31

31

25

11

Magnet therapy

36

162

28

34

29

35

30

11

16

28

21

7

Hydrotherapy of colon

38

87

25

23

44

20

30

4

4

23

4

7

Megavitamin doses

30

194

28

38

61

41

8

21

26

28

16

4

Colour therapy

32

141

21

35

37

21

22

11

16

22

4

8

Healing

23

39

8

14

33

10

17

4

18

13

6

1

Prayer, faith healing

17

45

37

9

16

10

9

1

36

11

3

3

% pos 92

Therapeutic touch

Acupuncture

m

HS-STUDENTS PhysioSocial Radiography therapy Educator (N = 63) (N = 63) (N = 59)

1

278

THE RELEVANCE OF TEACHING ABOUT THE ‘NATURE OF SCIENCE’

Table 3. HS-students, HS-teachers of The philosophy of science (PS), research methods (RM), and physiology to five Likert scale valued questions concerning their learning and teaching experiences with five typical “nature of science”(NOS) related topi “To what extent have you in HS-classes learned/taught:” I that scientists often present different views and models of how to solve a problem % HS-students Subject of:

N

-

±

+

Nursing PS Physiotherapy “ Social Ed. “ Radiography “ Nursing RM Physiotherapy “ Social Ed. “ Radiography “ PS - teachers RM - teachers Physiology teachers

317 63 59 34 317 63 59 34 32 34 35

33 24 19 46 32 26 21 49 23 27 40

26 42 35 30 30 41 37 30 29 23 26

41 34 46 24 38 33 42 21 48 50 34

II how to judge the academic qualification of authors, who claim to be researchers %

m

-

11 4 1 1 11 4 2 1 1 2 0

%

± 38 31 21 58 38 20 23 55 38 29 40

+

29 37 38 27 25 33 29 15 34 35 29

IV that the recent scientific research results are not immediately accepted as “facts” by the scientific community % HS-students Subject of:

III that scientists and scientific research projects are not value-free

m 12 4 1 1 9 3 3 1 0 2 0

33 32 41 15 37 47 48 30 28 36 31

-

± 50 39 36 67 52 34 35 61 31 21 63

25 49 40 27 24 48 35 30 16 32 23

V that scientists often speak of the recent research findings with some reservation

%

N

-

±

+

m

-

±

+

m

1

PS

Nursing

317

44

27

29

41

25

34

14

“ “ “ RM

Physiotherapy Social Ed. Radiography Nursing

63 59 34 317

32 35 65 46

39 41 16 26

29 4 39 24 1 31 19 2 56 28 15 44

31 31 25 25

30 38 19 31

4 1 2 13

63 59 34 32 34 35

27 33 72 28 44 37

34 39 13 22 18 26

39 28 15 50 38 37

4 2 2 0 2 0

27 26 28 32 36 14

36 39 19 26 22 26

4 2 2 1 3 0

Physiotherapy “ Social Ed. “ Radiography “ PS - teachers RM - teachers Physiology teachers

37 35 53 42 42 60

+

m

25 12 24 6 24 18 30 9 53 47 14

15 4 1 1 11 4 2 1 0 2 0

S. PETTERSEN

279

The number of 1 + 2 (“to a very small extent” and “to a small extent“) and 4 + 5 (“to a large extent” and “to a very large extent”) Likert scale-valued answers were collapsed into two new “dichotomised” variables, which are symbolized with (−) and (+), respectively. The symbol (±) means the number of 3-valued answers (“neither/nor”), and m is the number of “missing ” and “ I don’t know” answers collapsed. The fraction (%) of respondents who expressed (−) is outlined in bold Table 3 shows to what extent the responding HS students and the PS/RM and physiology teachers might have learned and taught about typical NOS-related topics, respectively. Knowledge of these topics might be relevant in order to reach the goal of the PS/RM subject mentioned above. Briefly, under half of the responding HS students reported learning about these NOS topics extensively, and about the same proportion of PS/RM teachers might have emphasized the teaching of such topics. Even fewer physiology teachers expressed that they had included these NOS topics in their teaching. Whilst the number of hours available to teach physiology might be (very) limited, these teachers might consider teaching about science didactical issues to have the lowest priority – or not to be their obligation at all. On the other hand, the science teachers involved in the teaching of BS in the HS (e.g. there are very few pedagogically-trained physicians and human biology researchers teaching in the HS programmes) might not be as familiar with the NOS concept as science educators possibly are. Twenty HS teachers of physiology were asked the question in subsequent interviews: “What do you think of when you hear the words; 'the nature of science' (NOS)?” However, none of the interviewed teachers conveyed anything substantial to this concept. Apparently most physiology teacher informants had barely heard of this “phrase” before. Consequently, the NOS concept might be exclusively associated with the discourse within the science education research community (Pettersen, 2003b). When asking the responding PS/RM and physiology teachers about the emphasis they put on teaching HS students about pseudoscience, 52% of the PS teachers, 57% of the RM teachers, and 63% of the physiology teachers expressed they had not done it extensively (1+2 and 4+5 Likert scale answers collapsed, respectively), whilst about one third of the respondents indicated that they had done so. Many health claims associated with CAM treatments are pseudoscientific (Johnson, 1999). An aim of teaching about the NOS could be to improve the receivers’ ability to reveal pseudoscience. The validity standards of CAM and scientific evidence-based treatments and cures might be very different, and the HS teachers should be aware of this, especially if they mention CAM in their student teaching. 4. CONCLUSION AND IMPLICATIONS Although it is probably not appropriate to generalize the results of these rather small non-probabilistic HS student and teacher samples to the national population, the results might demonstrate some ”trends” and ”patterns”, which could be valuable to further research into the field of the NOS in HS education. However, results of

280

THE RELEVANCE OF TEACHING ABOUT THE ‘NATURE OF SCIENCE’

subsequent interviews of PS/RM and BS teachers (N = 27 and N = 24, respectively) probably strengthen the validity of the study (Pettersen, 2003b). In summary, the research results presented in this paper point to the following compressed arguments for teaching about the NOS to students of the HS: (1) most of the responding HS students lack scientific evaluation skills in analyzing texts making health claims, (2) many students have a rather diffuse relationship with science research activities, (3) students’ uncertainty, about what counts as reliable knowledge, might have resulted from the minor emphasis put on NOS-related topics, especially when teaching the PS/RM subject, and (4) the minor emphasis put on teaching about the NOS in HS subjects (like physiology) might be an explanation for the majority of the responding HS students having a positive attitude towards non-scientifically based CAM treatments. The will among many responding HS students and PS/RM teachers to incorporate CAM teaching in HS programmes which are mainly scientific evidence-based requires wider debate: What is evidence of a health claim, what counts as warrant for that evidence, and what are the standards of validity? These are all epistemologically linked questions, and for the debaters of such questions in the HS, it could be useful to have knowledge about the NOS. To achieve critical health literacy, teaching approaches should build an individual’s capacity to distinguish fact from opinion and to analyze health information carefully (Nutbeam, 1999). Teaching about the NOS could probably bridge the skills of science and critical health literacy – especially for students of the HS. REFERENCES Bell, R.L., Lederman, N.G., and Abd-El-Khalick, F. (2000) Developing and acting upon one’s conception of the nature of science: A follow-up study. Journal of research in science teaching, 37 (6), 563-581. Caldwell, K. (1997). Ideological influences on curriculum development in nurse education. Nurse Education Today 17 (2), 140-144. Cassileth, B.R. (1999). Evaluating complementary and alternative therapies for cancer patients. Cancer Journal of Clinicians, 49 (6), 362-375. Driver R., Leach J., Millar R., and Scott P. (1996). Young People's Images of Science Open University Press UK. Eflin, J.T., Glennan, S. & Reisch, G. (1999). The nature of science: A perspective from the philosophy of science. Journal of Research in Science Teaching, 36 (1), 107-116. Giuffre, M. (1997). Science, bad science, and pseudoscience. Journal of PeriAnesthesia Nursing, 12 (6), 434-438. Hogan, Kathleen (2000). Exploring a process view of students` knowledge about the nature of science. Science Education, 84 (1), 51-70. Johnson, M. (1999). Observation on positivism and pseudoscience in qualitative nursing research. Journal of Advanced Nursing, 30 (1), 67-73.

S. PETTERSEN

281

Kersten, J., Bakewell, K., and Meyer, D. (1991). Motivating factors in a student’s choice of nursing as a career. Journal of Nursing Education, 30 (1), 30-33. Kolstoe, S.D. (2000). Consensus projects: teaching science for citizenship. International Journal of Science Education, 22 (6), 645-664. Korpan, C. A., Bisanz, G. L., Bisanz, J. & Henderson, J. (1997). Assessing literacy in science: Evaluation of scientific news briefs. Science Education, 81(5), 515532. Millar, R. & Driver, R. (1987). ”Beyond processes. Studies in Science Education, 14, 33-62. Norris, S.P. (1994). Interpreting pragmatic meaning when reading popular reports of science. Journal of research in science teaching, 31(9) 947 – 967. Nutbeam, D. (1999). Literacies across the lifespan: Health literacy. Literacy and Numeracy Studies, 9 (2), 47 – 55. ODIN (2003) The nursing, physiotherapy, social educator, and radiography curricula). (http://odin.dep.no/ufd/norsk/utdanning/hogreutdanning/regelverk/045061140002/indexdok000-b-n-a.html) (available in English). Last update: 01/15/04. Pettersen, S. (2003a). Er også naturfagdidaktikk godt for helsen? In: B. Bungum, & D. Jorde, (red.): Naturfagdidaktikk. Perspektiver – Forskning – Utvikling (pp. 273-288). Gyldendal Akademisk, Norway. (non-English). Pettersen, S. (2003b). Exploring the “status“ of the biological sciences in Norwegian Health Sciences. In: E.K. Henriksen & M. Oedegaard (reds.): Naturfagenes didaktikk - en disiplin i forandring? Hoeyskoleforlaget, Kristiansand, Norway. (in press) Pettersen, S. & Solberg, J. (2003). Students of Health Sciences’ Evaluation of Media reports of Health Research: A Norwegian Study. In: J. Lewis, A. Magro, and L. Simonneaux, (eds.): Biology education for the real world. Student – teacher – citizen. Proceedings of the IVth ERIDOB Conference, Enfa, Toulouse-Auzeville, France, (pp. 293-307). Pettersen, S. & Bye, A. (2003). Complementary-Alternative Medicine: Comparing attitudes of nursing students and their teachers. Journal of Health Communication (submitted Feb. 2004; in review). Roskell, C., Hewison, A. & Wildman, S. (1998). The theory-practice gap and physiotherapy in the UK. Insights from the nursing experience. Physiotherapy Theory Prac, 14, 223-233. Sadier, T.D., Chambers, F. and Zeidler, D.L. (2002). Investigating the crossroads of socio-scientific issues, the nature of science, and critical thinking. Paper presented at the Annual Meeting of the National Association for Research in Science Teaching (New Orleans, LA, April 6-10, 2002). ED466401. SE066464. EDRS. Sampson, W.(1995). Anti-science trends in the rise of the alternative medicine movement. In P.R. Gross, N. Levett, and M.W. Lewis, (eds.), Flight from Science and Reason. John Hopkins University Press, USA. (pp. 188-197). Trnobranski, P. (1996). Biological sciences in Project 2000: an exploration of status. Journal of Advanced Nursing, 23 (6), 1071-1079.

282

THE RELEVANCE OF TEACHING ABOUT THE ‘NATURE OF SCIENCE’

Watson, G. & Glaser, E.M. (1980). Watson-Glaser critical thinking appraisal manual. San Antonio: The Psychological Corporation: Harcourt Brace Jovanovich, Inc.

TEACHING ABOUT THE EPISTEMOLOGY OF SCIENCE IN SCHOOL SCIENCE CLASSROOMS: CASE STUDIES OF TEACHERS' EXPERIENCES

JIM RYDER, ANDY HIND, JOHN LEACH The University of Leeds, UK

ABSTRACT There is evidence that many science teachers have limited expertise in teaching the epistemology of science (the ways in which knowledge claims in science are developed and justified). We examine the classroom talk of seven teachers as they use published lesson resources to teach about the development of scientific models in two concept areas (cell membrane structure and electromagnetism). Our aim is to provide recommendations for the content and form of professional development activities likely to support teachers' effective uptake of these, and similar, teaching resources. We first provide a characterisation of the content of science-related classroom talk. Two distinctive lines of talk related to conceptual development can be identified in each of the Cell Membranes lessons, and an additional line of talk in each of the lessons focuses on the epistemology of science. Handling these distinctive classroom conversations was a new pedagogical challenge for these teachers. We then identify features of classroom talk likely to constrain or promote student learning about the epistemology of science. Several teachers supported student learning by making explicit statements about what students were intended to learn about the epistemology of science. Teachers also made links to other lessons to exemplify epistemic issues in a variety of science concept areas. The paper ends with a discussion of the design of continuing professional development activities to support teachers in introducing epistemic ideas in the science curriculum.

1. INTRODUCTION Teaching about the epistemology of science is gaining an increasing emphasis within many school science curricula (AAAS, 1995; Matthews, 1994). However, classroom-based studies have shown that there is currently no broadly established body of professional knowledge and expertise within the teaching community related to teaching about the epistemology of science (Lederman at al., 2003; Roth & Lucas, 1997). It is likely that science teachers responding to the increasing emphasis on epistemic issues within curricula will rely heavily on published resources, at least initially. We examine how a small group of teachers used published lesson resources for teaching about a single aspect of the epistemology of science: the development of theoretical models in the context of cell membrane structure and electromagnetism. Our analysis addresses two research issues. We first provide a characterisation of the content of science-related classroom talk. Given the 283 K. Boersma et al. (eds.), Research and the Quality of Science Education, 283—293. © 2005 Springer. Printed in the Netherlands.

284

TEACHING ABOUT THE EPISTEMOLOGY OF SCIENCE

epistemic focus of the lessons, we were particularly interested in the balance between talk about science concepts and talk about the epistemology of science. We then identify features of classroom talk likely to constrain or promote student learning about the epistemology of science. Our findings lead to recommendations for the content and form of professional development activities likely to support teachers' effective uptake of these teaching resources. Our approach reflects the view that there is much that the teacher must do in order to take published teaching resources and enact them in the classroom, i.e. teaching is a situated practice (Lave & Wenger, 1991). As a result, rather than examining teachers' espoused knowledge about the epistemology of science, or statements about what they wanted students to learn from the lessons, we focus on teachers' classroom talk. We recognise that, whilst teacher action is influenced by personal knowledge and intentions, factors specific to the school and classroom context also have a significant impact on the 'craft' of teaching. Brown and McIntyre (1993) refer to such factors as classroom conditions that guide teacher action, e.g. pupil attitudes and abilities, time available in class, issues raised by pupils and teacher fatigue. The study reported here is part of a larger study in which six different teaching activities were designed and evaluated in the classroom (Hind, 2002; Leach, Hind, & Ryder, 2003). 2. STUDY DESIGN The teachers whose lessons we examine in this paper volunteered to be involved in the project, and were already known to the authors as experienced and effective high school science teachers. Seven teachers were involved (four Cell Membranes lessons; three Electromagnetism lessons). Teachers are referred to in this paper using pseudonyms. None of the teachers had previously undertaken professional development activities associated with teaching about the nature of theoretical models in science. By focusing on teachers' first time use of published teaching resources we hope to identify development needs that are most critical for these teachers. Examining the use of each of these teaching resources in 3-4 different classroom contexts, allows for a range of contingent factors to come into play, reflecting our view of teaching as a situated practice. The lessons involve students aged 16-18 following specialist Advanced level science courses in the UK. These are academic courses focusing primarily on subject matter knowledge within a specific science discipline. Whilst aspects of the epistemology of science are identified as general learning aims for these courses, such issues are not emphasised in the ‘knowledge to be taught’ within the curriculum. Each class was attended by 10-20 students, the teacher, and at least one of the authors. Whole class teacher/student talk was audio recorded. We also recorded some student group discussions (typically 2-4 students) to examine the impact of whole class talk on a sample of students; the student groups were randomly selected. In addition we draw upon statements during post-lesson interviews with two of the

J. RYDER, A. HIND, J. LEACH

285

teachers. Interview questions addressed the teacher's perceptions of the effectiveness of the lesson and the challenges faced in teaching the lesson for the first time. The published materials consist of resources for use in classrooms and guidance notes for teachers (Hind, Leach, & Ryder, 2000). The Electromagnetism lesson follows the development of James Clerk Maxwell’s magnetic vortex model of electromagnetic interactions, focusing on his use of abstract ideas and analogies. The first activity addresses the distinction between observable phenomena/objects and abstract ideas in science. The second activity begins with a teacher presentation of the historical development of ideas about electromagnetism involving Faraday and Maxwell. This presentation addresses the abstract nature of the vortex model, the role of analogy, and the fruitfulness of the model in unifying theoretical ideas. Students are then given questions about this historical episode to discuss in pairs, followed by a whole class discussion. In the final activity students work with the kinetic model of gases to highlight the usefulness of models in making testable predictions that generate further research. The Cell Membranes lesson presents a history of the development of models of cell membrane structure. The lesson begins with an examination of early data concerning the surface area of lipids in the cell wall. At this point in history, no detailed model of membrane structure had been proposed. In the next activity two models are presented (Danielli-Davson and Robertson), along with electronmicrograph evidence of a double layer cell membrane structure. At this point both models are equally well supported by all the available data. Historically these models went beyond available data and their development involved informed conjecture about the arrangement of lipids in the cell wall. In the final activity students are presented with additional evidence and a new model (SingerNicholson). This model corresponds with contemporary accounts of the structure of cell membranes (the fluid mosaic model). At this point only one of the models supports all the evidence. Using this sequence of historical snapshots, the lesson aims to communicate how scientific models often 'go beyond' the available data, and that scientists may use conjecture and creativity in developing explanatory models. 3. ANALYSIS Our first aim was to identify scientific knowledge that featured in the classroom talk. We were particularly interested in the extent to which talk about the epistemology of science was contextualised using science concepts. Two of the authors listened to the audio tapes of whole class and student group discussions independently, and noted explicit talk about the epistemology of science and talk about science content. The two authors then met and compared their notes, with a view to developing a comprehensive list of explicit talk about the epistemology of science and science content. These quotations were then transcribed, and the sets of notes and transcripts were analysed to characterise teacher talk. We also considered the time devoted to each aspect of classroom talk. Our second aim was to identify features of classroom talk likely to constrain or promote student learning about the epistemology of science. We examined student

286

TEACHING ABOUT THE EPISTEMOLOGY OF SCIENCE

learning arising from these lessons (Leach et al., 2003). However, because of the small number of students and teachers involved in these case studies, we did not attempt to correlate student learning with features of science-related talk in each classroom. Rather, our approach was to consider whether the content of sciencerelated talk was consistent with our intentions as designers of these lessons (and therefore more likely, in principle, to support student learning). Guidance notes accompanying the lesson materials asked teachers to make epistemic learning aims explicit. Teachers were also asked to make links to other science concept areas to further contextualise ideas about the nature of theoretical models in science. We listened to the audio tapes and noted incidents when teachers did or did not follow these expectations. In some cases teachers described the reasons for their classroom actions in post-lesson interviews. 4. FINDINGS Our analysis identified three distinct areas of science-related talk (Table 1). Talk in each of these areas tended to run through a lesson, reappearing at different points in the classroom discourse. As a result, we describe these ongoing strands of discourse as 'lines of talk' (Scott, 1998). Table 1. A characterisation of science-related classroom talk in the Cell Membrane and Electromagnetism lessons Internal conceptual line of talk

Extended conceptual line of talk

Epistemic line of talk

Cell Membranes

The sequence of models of membrane structure: DanielliDavson; Robertson; Singer-Nicholson

The fluid mosaic model of cell membrane structure

The interplay between data and models in the development of models of cell structure

Electromagnetism (Faraday-Maxwell historical episode)

Maxwell's magnetic vortex model of electromagnetic interaction

[did not feature in the classroom talk]

e.g. the role of analogies in the development of Maxwell's theory of electromagnetism

J. RYDER, A. HIND, J. LEACH

287

The internal conceptual line begins and ends within the lesson itself. For example, Maxwell's magnetic vortex model of electromagnetic interactions had not been discussed in any previous lesson conducted by these teachers, and they were unlikely to extend students' conceptual development of this model in the future. By contrast the extended conceptual line has a broader significance within the science course. For example, classroom talk in the Cell Membranes lesson referred to a model of cell membrane structure that had been introduced in earlier science lessons: the fluid mosaic model. However, this model did not feature in the resources for the Cell Membranes lesson, although an equivalent model (SingerNicholson) was introduced at the end of the lesson. Whilst such talk featured only briefly in the Cell Membrane lessons, it did have a significant impact on students' engagement with epistemic issues (as exemplified below). By contrast, we did not identify an extended line of conceptual talk in the Electromagnetism lesson. This highlights the importance of examining teaching about the epistemology of science within specific science content areas. A third line of talk related to the epistemology of science. For the Cell Membranes lesson such talk focused on the interplay between data and model development as new data became available. In the Electromagnetism lesson, epistemic lines of talk included the distinction between observable phenomena/objects and abstract ideas, the role of analogies in the development of theories, and the predictive power of theoretical models. Much of this talk was set in the context of concepts such as the magnetic vortex model or cell membrane structure. Nevertheless, the emphasis was on developing students' views about the epistemology of science. Of course, discourse related to epistemic issues reflects the design and focus of the lesson materials. However, the time devoted to epistemic lines of talk was much less than that addressing science concepts. The second aim of our analysis was to identify features of classroom talk likely to constrain or promote student learning about the epistemology of science. In the analysis we focus on three features of classroom talk. The first feature emerged from our analysis of the content of science-related classroom talk: recognising the impact of broader conceptual learning upon students’ ability to engage with the epistemic aims of the teaching materials. The remaining two features were identified as key issues in the teacher guidance notes for the lesson resources: making epistemic learning aims explicit in the classroom; and making epistemic links to learning in other concept areas. Here we examine the extent to which these teachers enacted these suggestions within the classroom. Excerpts of classroom talk are used either to exemplify typical classroom interactions, or in some cases to illustrate the ways in which a particular feature of classroom talk can either promote or constrain student learning. Recognising the impact of broader conceptual development within these lessons The extended conceptual line of talk within the Cell Membranes lesson did not feature prominently within science-related classroom discourse. Nevertheless, we found that in some instances such talk is likely to have a significant impact on student learning about the epistemology of science. Below we describe two episodes from Alison's lesson that illustrate how such talk can constrain student learning. A

288

TEACHING ABOUT THE EPISTEMOLOGY OF SCIENCE

third episode from Claire's lesson shows this talk being handled by the teacher to promote student learning. At the beginning of Alison's lesson, students mention the fluid mosaic model of cell membrane structure. This model features elsewhere in their science course and is a standard model introduced in text books, at this level in which lipids are free to move within the cell wall. The class is discussing data suggesting a 2:1 ratio between the surface area occupied by lipids extracted from a cell and the surface area of that cell. As yet none of the three models of membrane structure has been presented. Student: Teacher: Student:

Let's say if the lipids were in action, would that ratio [2:1] stick, would it still…? If the lipid was? In action, as in binding, moving, you know. When the cell membrane, we say is fluid mosaic isn't it.

In response the teacher clarifies how the data were gathered. She does not make any reference to the fluid mosaic model. However, it is clear that discussion of this model in other lessons (the 'extended conceptual line of talk') is influencing this student's engagement with the activity. Later in Alison's lesson, students are discussing in groups whether electronmicrograph evidence of a double lipid layer supports or undermines either of two models (Danielli-Davson, Robertson). One discussion begins with a student asking: 'Which one is the right one? Is it that one?'. This consideration of 'the right answer' continues a little later when another student raises the fluid mosaic model explicitly: 'these are free to move, the lipids, I've read that you know, that's what they call the fluid mosaic model'. This gets in the way of the epistemic purpose of this discussion: how the electronmicrograph evidence relates to the Danielli-Davson and Robertson models. As the teacher is talking with another group at this point, the teacher is not aware of this discussion amongst the students, and they do not raise it during the subsequent whole class discussion. We can contrast the above episode with discussions in Claire's Cell Membrane lesson, during the second activity mentioned above. The same challenge is apparent, but here the teacher follows an effective strategy. Student thinking about the fluid mosaic model becomes apparent as Claire is talking to one student group, although the students do not mention the model explicitly. Students: Teacher: Student 1: Teacher: Student 2: Teacher:

They knew that that was wrong… These [lipids] can move over there. But at this stage do they know anything about it [lipids] being able to move about? I don't know. I think they knew about the lipids being moved. Do they know about that though? I don’t think so. Look on our time line. What evidence is available so far?

J. RYDER, A. HIND, J. LEACH

289

The time line is provided in the published resources. It shows that evidence about the movement of lipids within cell membranes did not appear until after the development of the Danielli-Davson and Robertson models. Claire also refers back to the fluid mosaic model at two further points in the lesson during whole class discussions. In Alison's lesson the extended conceptual line was raised by students, but not addressed explicitly by the teacher. Claire's lesson was very different. She recognised students' references to the fluid mosaic model and worked to sustain this extended conceptual line of talk throughout the lesson. In doing so, she contrasts this with the internal conceptual line – as represented on the time line provided with the published resources. By raising each explicitly with the class, she is encouraging them to recognise, but set aside, their thinking about the fluid mosaic model. This helped students to focus on the epistemic issue of how models and evidence interacted in the historical development of models of the cell membrane. Making epistemic learning aims explicit in the classroom A common feature of teacher talk in many of the lessons was explicit reference to the epistemic learning aims of the lesson. For example, near the end of her Cell Membrane lesson Judith asks: Teacher:

What have we learnt today? Not particularly the facts about the membrane. We've learnt that 'when scientists produce theoretical models they use their imagination and creativity to think about data in new ways…'

Daniel's Electromagnetism lesson provides a similar episode. Students have been discussing the Faraday-Maxwell historical episode in small groups. Daniel begins the whole class review by stating explicitly to the class his reactions to what he has heard of their discussions. In particular, he recognises that some students have been struggling with the conceptual basis of Maxwell's vortex model: Teacher:

In talking with one or two of you it's quite clear that there are some parts of the story in a way that you're not a hundred percent secure in (…) but don't let that cloud the issue. I mean the issue here is…

He then goes on to emphasise the epistemic learning aims for this part of the lesson. Elsewhere in the lesson he describes the epistemic line as 'the thread of what we are doing'. In the post-lesson interview Daniel reflects on his treatment of this activity: They're seeing it more as trying to develop an understanding of the model rather than an understanding of what the model is for (…) One or two of the lads, like [John], he'll just get wrapped up in the model to the point where he'll lose track of what the purpose of the lesson is.

290

TEACHING ABOUT THE EPISTEMOLOGY OF SCIENCE

Here Daniel identifies two distinct learning challenges in this part of the lesson: understanding the conceptual model of Maxwell's vortices, and learning about the role of analogies in developing scientific understanding – an epistemic learning aim. Daniel recognises that some students are struggling with the 'internal' conceptual line ('trying to develop an understanding of the model'). However, he is concerned to establish the grounds for the epistemic line of the lesson ('understanding of what the model is for…what the purpose of the lesson is'). Making epistemic links to learning in other concept areas In their Cell Membranes lessons Judith and Claire made extensive links to other science lessons. For example, Claire referred to an earlier lesson when the class had watched a video on the development of ideas about the structure of DNA. In that lesson she had highlighted how the scientists involved had used creative thinking in addition to evidence in order to develop models of DNA. She also referred to a lesson to come later in that week when the class was to look at how substances are transported in the phloem of plants. By contrast we saw very few such links made by either Alison or Clive. Of the three teachers following the Electromagnetism lesson, Daniel made the most extensive use of links between lessons to sustain the epistemic line of the lesson. In particular, he encouraged his students to identify other models that they had encountered in their science education: Teacher Student 1 Teacher Student 2 Teacher

Student 3 Teacher

Let's think of all the work that we've done in the last year what other models have we used (…) Photoelectric. (…) we've got this billiard ball idea of a photon of energy. We use that model (…) Any others? Waves behaving like particles (…) so that was a good model that we used (...) We tend to use models all the time to try and explain things. Can anyone think of any other models? The atom with electrons around. Excellent (…) or energy levels, spectra, excitation (...)

In the interview Daniel reflected on his intentions for this part of the lesson: I wanted them at that point to realise that they'd already gone through many opportunities to look at modeling situations already. The interesting thing is, again in conversations with them, how many of them also related it to modeling situations in chemistry and biology (…) they were looking at it as a process that wasn't just across physics but was across the sciences and I think that, I think, is a good outcome. These reflections show that one of Daniel's intended learning aims for the lesson is that his students recognise that ideas about modeling run across concept areas and disciplines ('a good outcome').

J. RYDER, A. HIND, J. LEACH

291

5. DISCUSSION In the majority of lessons, far less time was devoted to the epistemic line than to the conceptual lines. This was not consistent with the guidance given in the lesson materials: how might this be explained? It is certainly true that the technical content of the models addressed in the lessons was challenging for students and needed to be taught. However, only one of the teachers (Daniel) drew upon his expertise to develop an extended epistemic line throughout the lesson, making links between the conceptual content of the model of electromagnetism, and other models encountered by the students in other parts of the curriculum. The data available to us in this study do not allow us to comment on why the other teachers did not develop such an extended epistemic line throughout their lessons, though previous studies offer insights on this aspect. Surveys and interview studies of science teachers’ epistemological knowledge (Lederman, 1994, 1999) suggest that many have limited knowledge that is not easily articulated in specific contexts. In particular, some teachers appear to have a naïve correspondence theory about the relationship of scientific models and the material world. If this is the case for some of the teachers in this study, it would hardly be surprising to find that their classroom talk did not have an extended epistemic line, even with the support of teaching materials. However, teachers who have the expertise to articulate epistemological ideas relatively clearly and appropriately in response to survey or interview questions, may not have the expertise to draw upon such ideas during classroom talk. Given that most science teachers in the UK have relatively little experience of addressing epistemic aims in their teaching, it is unlikely that most have well developed pedagogical content knowledge (Shulman, 1987) in the area of teaching the epistemology of science. They are unlikely to show the fluency of action (Brown & McIntyre, 1993; Dreyfus & Dreyfus, 1986) when teaching about the epistemology of science that might characterise their teaching of other scientific content. When working on the cell membranes lesson, we noted several occasions when the extended conceptual line (i.e. students’ prior knowledge of contemporary models of the structure of cell membranes) made an impact upon their engagement with the epistemic aims of the teaching. One response to this difficulty would be to introduce epistemic content to students through examples that have very little conceptual content. Indeed, teaching materials of this kind have been published (e.g. Lederman & Abd El Khalick, 1998). Such materials can be a useful way of opening up discussions about the epistemology of science. However, there is a growing body of evidence that advanced science students already have a well-developed profile of epistemological ideas, though they may not deploy them appropriately according to the specific conceptual context (e.g. Brickhouse, Dagher, Letts, & Shipman, 2000; Leach, Millar, Ryder, & Séré, 2000). Therefore, it is our view that teaching approaches for advanced students need to address explicitly the issues of appropriateness and context. Learning about the development of theoretical models in science involves recognising how ideas and evidence interact in a range of different content areas, such as membrane structure and electromagnetism.

292

TEACHING ABOUT THE EPISTEMOLOGY OF SCIENCE 6. PROFESSIONAL DEVELOPMENT ACTIVITIES

An obvious focus for professional development is upon teachers' understandings about the role of data in the development of theoretical models (Matthews, 1994; McComas, 2000). Of course, having a clear understanding of the epistemic content that the lessons aim to teach is of critical importance. However, such understanding in itself is not sufficient to enable teachers to address epistemic ideas in the classroom because it is necessary for them to transform and communicate such knowledge for the purpose of teaching. We can imagine that it would be possible to present teachers with a small number of ‘thinking tools’, to use in reflecting upon and talking about their teaching, which address epistemic aims and thereby enable them to develop their expertise. An example of such a thinking tool is the idea that science lessons with an epistemic aim need to address explicitly the technical details of any models used (internal conceptual line), the epistemic aims of the lesson (epistemic line), and the way in which previously taught content might interfere with the epistemic content of the lesson (extended conceptual line). Teachers might use thinking tools to reflect critically upon video presentations of the teaching of lessons with an epistemic aim, enabling them to develop a representational repertoire of epistemic links for use in the classroom (Shulman, 1987). The teachers in our study would have benefited greatly from the chance soon after the lessons to reflect as a group on their implementation of these resources, followed by a chance to use the activities a second time. This would have provided opportunity for teachers to exchange experiences about 'what pupils say in response to…', as well as explanations, stories, and so on that they had used in the classroom. Providing opportunities for teachers to engage in cycles of 'in-class' and 'out-ofclass' reflections will support teachers in developing the 'discursive resources' essential for the effective use of published 'material' teaching resources. Such an approach recognises the significance of 'classroom conditions' on teachers' classroom actions (Brown & McIntyre, 1993). For example, recognising how different conceptual lines of talk might interact in a lesson will depend on a detailed knowledge of the science course followed by a particular class. Similarly, the epistemic links, which teachers might make, depend on their knowledge of previous science learning activities. Thus, the specific ways in which broader conceptual and epistemic lines of talk feature within a lesson cannot be identified outside of school/classroom contexts. Focusing development activities on teachers' own classroom experiences also allows consideration of the often very legitimate reasons for why teachers do not follow suggested strategies in their classrooms, e.g. concerns about course assessment, perceptions of more senior colleagues, or behaviour management in the classroom. We would urge those developing published teaching resources in this area to consider how these might be supported by teacher development initiatives that enable teachers to reflect on their classroom experiences with other teachers. Such activities are likely to enhance the effective uptake (and hopefully future revision) of published materials.

J. RYDER, A. HIND, J. LEACH

293

REFERENCES American Association for the Advancement of Science (AAAS). (1995). Science for all Americans. American Association for the Advancement of Science, Oxford University Press. Brickhouse, N. W., Dagher, Z. R., Letts, W. J., & Shipman, H. L. (2000). Diversity of students' views about evidence, theory, and the interface between science and religion in an Astronomy course. Journal of Research in Science Teaching, 37, 4 340-362. Brown, S., & McIntyre, D. (1993). Making sense of teaching. Buckingham, Philadelphia: Open University Press. Dreyfus, H. L., & Dreyfus, S. E. (1986). Mind over Machine. New York: Free Press. Hind, A. (2002). Teaching and learning about the nature of scientific theoretical models as part of an AS level programme. Unpublished Masters thesis, University of Leeds, Leeds, UK. Hind, A., Leach, J., & Ryder, J. (2000). Nuffield Teaching About Science: Resources for AS/A level. Retrieved June, 2004, from

http://www.nuffieldcurriculumcentre.org/go/minisite/Post16TeachingAboutScience/Introd uction Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. Cambridge, UK: Cambridge University Press. Leach, Hind, A., & Ryder, J. (2003). Designing and evaluating short teaching interventions about the epistemology of science in high school. Science Education, 87, 831-848. Leach, Millar, R., Ryder, J., & Séré, M. G. (2000). Epistemological understanding in science learning: the consistency of representations across contexts. Learning and Instruction, 10(6), 497-527. Lederman. (1994). Students' and Teachers' conceptions of the Nature of Science: A review of the research. Journal of research in science teaching, 29, 4 331-359. Lederman. (1999). Teachers' understanding of the nature of science and classroom practice: factors that facilitate or impede the relationship. Journal of Research in Science Teaching, 36(8), 916-929. Lederman, & Abd El Khalick, F. (1998). Avoiding de-natured science: activities that promote understandings of the nature of science. In W. F. McComas (Ed.), The nature of science in science education: rationales and strategies (pp. 83126). Dordrecht/Boston/London: Kluwer. Lederman, Lederman, J. S., Khishfe, R., Druger, E., Gnoffo, G., & Tantoco, C. (2003, March 23-26). Project ICAN: A multi-layered model of professional development. Paper presented at the National Association for Research in Science Teaching, Philadelphia, PA. Matthews, M. R. (1994). Science teaching: The role of History and Philosophy of Science. Routledge. McComas, W. F. (Ed.). (2000). The nature of science in science education: Rationales and strategies. Dordrecht/Boston/London: Kluwer. Roth, W. M., & Lucas, K. B. (1997). From "Truth" to "Invented Reality": A discourse analysis of high school physics students' talk about scientific. Journal of Research in Science Teaching, 34, 2 145179. Scott, P. H. (1998). Teacher talk and meaning making in science classrooms: A Vygotskian analysis and review. Studies in Science Education, 32, 45-80. Shulman, L. S. (1987). Knowledge and Teaching: Foundations of the new reform. Harvard Educational Review, 57(1), 1 1-21 February.

PART 6 Models, modelling and analogies in science education

A THREE-PHASE DESIGN FOR PRODUCTIVE USE OF ANALOGY IN THE TEACHING OF ENTROPY

WOLTER KAPER, MARTIN GOEDHART Universiteit van Amsterdam, The Netherlands

ABSTRACT Gentner has described analogy as a mapping of terms from a base (better known) domain to a target domain. She asserts that use of analogy can lead to new conclusions in the target domain. This 'structure mapping' theory, though useful, does not yet describe the process of analogical reasoning. We will argue that an analogy can be used productively in a process that has two phases: first, constructing the analogy using existing knowledge of base and target domains, and second, extrapolating the analogy within the target domain. In the first phase object mapping is motivated by the recognition of mappable relations. In the second phase, the productive use of the analogy can involve creation of both new terms and relations, as a result of mapping existing terms and relations from the base domain. If analogies are to be understood critically, then a third phase might be the testing of new relations against learners’ experience. This three-phase process description of analogy has been tried out in a teaching experiment that aimed at an understanding of entropy, by an analogy to falling water. We conclude that this three-phase description is useful.

1. INTRODUCTION The ability to think in analogies is believed to be central to human cognition. It is thought to play an important role in scientific reasoning, as well as in guided learning processes (Kurtz, Miao & Gentner, 2001; Baker & Lawson, 2001; Glynn & Takahashi, 1998). Research into the use of analogies in teaching can be roughly divided into two categories: according to whether the analogy is primarily seen as a tool used by the teacher to convey new concepts or information, or as a tool used by the students in drawing conclusions that are new to them. We would regard Glynn’s (1989, 1998) “teaching with analogies” model and Zeitoun's (1984) “general model of analogy teaching” primarily as tools for the teacher in conveying concepts or information. In testing the effect of analogy, correct “recall” is the primary measure of success (Glynn & Takahashi, 1998; Arnold & Millar, 1996). Alternatively, theories developed by Gentner (1983) and Gick & Holyoak (1983), as well as Clement's (1993) and Brown’s (1993) ”bridging analogies”, are about analogies as tools in the hands of the learner for solving problems, or reaching new conclusions about the target concept. 297 K. Boersma et al. (eds.), Research and the Quality of Science Education, 297—308. © 2005 Springer. Printed in the Netherlands.

298

PRODUCTIVE USE OF ANALOGY IN THE TEACHING OF ENTROPY

Most authors from the last mentioned group describe analogy as a relation between “situations”, or representations of situations, while in earlier publications Gentner (1983) clearly describes analogy as a relation between two propositional networks both of which already contain generalized statements about, for instance, electricity or the planetary system. In Gentner’s later publications (Gentner & Holyoak, 1997; Kurtz, Miao & Gentner, 2001) concrete situations return as a basis and target for analogical reasoning. We agree that both definitions (relation between situations or between networks of propositions) may be of use in education, but they both apply to different phenomena. In this work we define analogy as: a one-to-one mapping of terms between two domains of experiences that are each already described by networks of (generalized) propositions, the base domain, and the target domain, about which students want to draw conclusions. We chose this definition because the learning process that we were aiming at seems to demand working with generalised propositions. Another example seems to be the teaching of laws of electricity, starting from experiences with flowing water, where it was noted that first the base domain needs to be understood in terms of generalised propositions1, before successful mapping to the target domain can be realised (Haeberlen & Schwedes, 1999; Gentner & Gentner, 1983). We therefore chose Gentner's (1983) “structure mapping” theory as our point of departure. In the original version of this theory, the process in which an analogy is used to draw new conclusions could not be adequately described. The theory only describes the structure of a finished analogy, not the development of analogical reasoning in the course of time. More recently, it was asserted that: The process of analogical thinking can be usefully decomposed into several basic constituent processes. One or more relevant analogs stored in memory must be accessed. A familiar analog must be mapped to the target analog to identify systematic correspondences between the two (...). The resulting mapping allows analogical inferences to be made about the target analog (...). These inferences need to be evaluated and possibly adapted to fit the unique requirements of the target. (Gentner & Holyoak, 1997).

However, this “decomposition” does not yet give us a time sequence for use in education. Kurtz, Miao & Gentner (2001) recommend the detailed study of the processes that make up analogical reasoning as a direction for future research. In this study, we test the following three-phase description of the process of analogical reasoning, for its usefulness in organising a learning process: 0) Learners study base and target domains separately for sometime. 1

For example, “water is divided at branching points” and “the water is not jammed in our water circuits”.

W. KAPER, M. GOEDHART

299

1) Learners begin to notice homonymous relations in their descriptions of both domains. This mapping of relations is expected to motivate the mapping of corresponding terms in such relations. The mapped terms are not homonymous, and they represent different concepts.2 2) Some terms or relations in the base domain are noticed as not having corresponding terms or relations in the target domain. Learners then create such corresponding terms and relations, by way of hypothesis. 3) The usefulness of the new terms or relations is evaluated against experience. The analogical reasoning starts at phase 1, while phase 0 is a condition for phase 1. This description involves several of the “processes” listed by Gentner & Holyoak (1997). The aim of the present study is to gain detailed information about these processes when they are organised by a teacher in an educational setting. Particularly, we want to know: − Can these three phases be used to organise a process of analogical reasoning in a group of students, so that students can actively contribute to each of the phases? − If yes, what are their contributions, and can we find an understanding of the target concept in them? 2. METHOD AND MATERIALS Method An assignment sequence was prepared that aimed at understanding entropy, using an analogy between work done by falling water, work done by an expanding gas, and work done by a heat engine. The ‘water’ analogy was first given by Carnot (1824) and has been developed for use in education by Herrmann (1989). However, in Carnot’s setup the analogy is presented to students in a more finished form than in our case. Our assignment sequence was structured according to the three phases mentioned above. Researcher's expectations about each of the assignments were explicated beforehand. Four first year undergraduate students of chemistry, while taking a first course on thermodynamics, volunteered to participate in the experiment. They received their tutorial sessions from the researcher, working in two pairs on the experimental assignments. Their discourse was audio taped. The teacher-researcher strived to refrain from ‘giving away’ steps that students were expected to take themselves. However, he did ask critical questions if he noticed possible inconsistencies in students’ answers.

2

Examples are: atom and solar system, or water and electricity. Such radical differences between domains is what distinguishes an analogy from a straight, law-like generalization.

300

PRODUCTIVE USE OF ANALOGY IN THE TEACHING OF ENTROPY

Analysis of conversations between students and between students and teacher aimed at describing students' contributions to each of the three phases. Differences between the expected and realised processes were noted. Materials: the assignment sequence The three phases proposed as a working hypothesis were used to construct those parts of the assignment sequence that aimed at a first concept of entropy. A peculiarity of this case-study should be mentioned here: we used two different base domains (work by water because of differing height, work by a gas because of differing pressure) as a means for understanding a third domain: work by a steam engine, because of differing temperature. In general we do not regard the use of two separate base domains as necessary for analogical reasoning, but we thought it profitable in our case. At first, this was based on the idea that, after generalizing over two quite different cases, one has a stronger base for leaping into the unknown. The assignment sequence was constructed in six parts that are characterised by the three phases and by the use of each of the three domains (Table 1). First, the two base domains are studied separately for some time. There were assignments for each of the domains, accompanied by illustrations given in Figure 1.

Figure 1. Drawings presented to the students to illustrate the three domains

W. KAPER, M. GOEDHART

assignment block a b c d e f

301

Table 1. Construction of assignment sequence domains involved phase A: falling water B: expanding gas A, B C: steam engine and A, B A, B, C C

0, separate study 0, separate study 1, recognising the analogy: A...B 1, recognising the analogy: (A+B) ... C 2, extending the analogy (A+B)→ C 3, checking hypothesized relations

Then assignments asked for the mapping of terms between the domains of falling water and expanding gas. An exemplary assignment is shown in Table 2: Table 2. Assignment designed for phase 1: recognizing / constructing the analogy 11-5a In 11-1 the height of the water surface was involved. Which quantity from 11-4 shows the most similarities to the height from 11-1, regarding the role that both play in your answers. b Make a list of these similarities. Formulate them (also) in words. Note: 11-1, 11-4 and 11-5 are assignment numbers.

This question, in Gentner's terms, invites students to map the term ‘water height’ to another term in the second domain. It is expected that this mapping will be guided by a perceived similarity in already-known relations. Students are expected to begin their mapping, by naming and formulating these relations. We expect the named relations to be homonymous in both domains, while we expect the mapped terms to be heteronymous (differently named). In particular, we expect water height to be mapped to gas pressure, because both determine the work done by their difference between two systems. The italicized part could be called an expected equal name of both expectedly mapped relations: w = ∆m. g(h2 - h1) and w = -∆V(p2 - p1). Following assignments asked for the mapping of other quantities and terms, for instance water mass was expected to be mapped to gas volume. Assignments asked students to analyse the experimental setups and subdivide them into various named systems. Names for these systems were also expected to be mapped. So, for instance, 'water-turbine' was to be mapped onto 'gas piston'. After this, the third domain was introduced: the steam engine. It was expected that some terms could be mapped immediately to the target domain, for instance it was expected that water height and gas pressure would be mapped to temperature. However, a term corresponding to water mass or gas volume was expected to be missing. Assignments were designed to make students aware of the missing term, and to let them invent this term, by analogy (Table 3).

302

PRODUCTIVE USE OF ANALOGY IN THE TEACHING OF ENTROPY

Table 3. Assignments designed for phase 2, extending the analogy by hypothesis 11-7e Do you recognise a quantity that could play the corresponding role of mass (from assignment 11-1) and of volume (from assignment 11-4)? f If you see such a quantity: check if all similarities between mass and volume that you mentioned in 11-5 also hold for this quantity. g If you do not recognise such a quantity, then call it X temporarily, and write down which properties X should have, in order to let it correspond to mass (from 11-1) and volume (from 11-4). h Can you now derive some equations in which X itself does not appear, but which are based on the assumption that an X with these properties exists? Try it. i On which known or supposed properties of X do your predicted formulas depend?

It was expected that students would know some terms which partly satisfy homonymous relations found earlier between mass and volume. It was also expected that none of these terms would satisfy all of these relations, and that students would find this dissatisfying (f). As a solution, it was proposed in the following assignment to start working with the missing term, by just assuming that it existed (g and h). Working with this unknown term should make it possible to derive relations for the work done and the efficiency of the steam engine (i). Of the expected relations, some do not contain the unknown X, but their correctness still depends on the existence of such an X, because X is used in the derivation. It was expected that students would realise that their predictions depended on their analogically satisfying, but still possibly untrue, assumptions (i). It was hoped that this would create a need for an experimental check, as planned in phase 3. 3. RESULTS First, we shall summarize our observations regarding the first phases of our scheme. Then, we will illustrate our method by showing one piece of dialogue and discussing it in detail to draw a conclusion about phase 2 "extending the analogy". We finish by again summarizing about the third phase.3 Phase 0: separate study of each domain Concerning the separate study of both base domains, we found that most of the expected relations were available to these four students by the end of this phase. Although students noticed and could formulate the analogy during the assignments designed for phase 0, they did not in fact do so – a spontaneous use of analogy did not arise.

3

The 31-page paper distributed at the ESERA conference 2003 contained student dialogue analyses to support all conclusions presented here. It is available on request.

W. KAPER, M. GOEDHART

303

Phase 1: recognising or constructing an analogy In the formulation of expectations, we used both the words "recognizing" and "constructing". With hindsight, the assignments can be seen as ambiguous, asking sometimes for recognition of an analogy that the teacher/researcher then saw as at least partly obvious. At other times students were asked to construct, that is to make choices that could have been made in other ways. One of the similarities between water height and gas pressure, expected by the teacher, is the efficiency of the gas engine, ε=(p1-p2)/p1, which depends on the difference in pressure between both gases, just like the efficiency of the water wheel, ε=(h1-h2)/h1, depends on the difference in height between both water levels. While students discussed their answers, it turned out that some had formulated this similarity, but another student objected, “it depends on how you've understood the efficiency in this assignment”. He had analysed the two-gases situation differently. Instead of distinguishing three systems, he had distinguished just two, called “energy-source” (gas 1) and “task” (gas 2 + piston + flywheel)4, and he concluded that all energy delivered by the source was used for the task, giving an efficiency of 100%. This efficiency did not depend on the difference in pressure at all! It took a few minutes before the teacher realised that a choice is involved here: many times in thermodynamics it is useful to divide the world into two systems. But when you want to discuss efficiency, a division into at least three systems must be chosen, where one of the three energy changes is considered a ‘waste’. This way of dividing the world is not self-evident, and is not dictated by nature, although good reasons can be given for it. This means, that our analogy is less something to be 'recognised' (by comparing 'situations'), though at that time it was expected. Instead, it is more something to be constructed, by comparing and modifying two deliberately chosen descriptions of situations. Phase 2: extending the analogy, by way of hypothesis. For students to be able to extend an analogy, we first assumed that they had to experience some terms in the base domains as not being suitably mapped to the third domain. In other words, a suitable term to which to map was yet missing in the third domain, but would students agree with this? They were expected to come up with some candidates, and the assignment cited before (11-7f) asks them to double-check these. One possible candidate, heat capacity, was expected by the teacher. Students came up with two more: amount of liquid in the steam cylinder, and the heat exchanged between reservoir and cylinder. Each of these three candidates was defended by them, each student was motivated to mention homonymous relations. The teacher required that the new term should satisfy relations corresponding to all 4 According to the assignment text, in all three cases work was done on a flywheel, or on a massive object "such as a flywheel".

304

PRODUCTIVE USE OF ANALOGY IN THE TEACHING OF ENTROPY

homonymous relations previously found by mapping between the two base domains. This requirement turned out to be necessary to show these candidates what was not satisfactory. This implies that for our purpose the use of two base domains is necessary. If we had used just one base domain, then each of the student-suggested mapping candidates would have had to be accepted. The fact that some relations are not mapped would just prove that analogies are only partial. In our setup, using two base domains, the selection of mappable and non-mappable relations had already been done when mapping between the two base domains. Therefore, it was now possible to ask students if we could stick to their earlier selection of mappable relations. After students have exhausted their creativity in using their existing knowledge, they try the suggested approach using 'X', as shown in the following dialogue (Table 4). Te Teacher, L, J and M students L: Oh you (J) assume the efficiency, just like eh... (J has written (∆T1 - ∆T2) / ∆T1 for the efficiency.) Te: L: J: Te: J: Te: J: Te: J:

Well, he doesn't assume it, I think, he can derive it if it's alright. No... you just copied it from those other assignments. Yes, but now you can those X's, everywhere. Yes, you have corresponding assumptions in each case. Yes, right. And these assumptions you can use… Yes, so this formula should be right if the rest is right also. Yes, well, explain that will you? Err... you can set those... ǻX1 is equal to -ǻX2, because we have: ǻXtotal = 0. Then you would, then you would have here: ǻX1, minus, or, what's it called, you could put here (before ǻT1) X1 and here, and here. J now has:

∆ - ∆ η= X1 T1 X1 T2 X 1∆T 1

Te: L: M: J:

You must not explain it to me, but to your unbelieving partners. Who says I'm unbelieving? We, doubting Thomases… No but then you have... here the energy that’s transferred here (points to an interaction line in the scheme made for assignment 11-7a), plus the energy that's transferred here. That's err..

W. KAPER, M. GOEDHART

305

The assignment to derive relations without X , based on the assumption that an X with predicted properties (by analogy to mass and volume) exists, is first answered in the easiest way: by looking at the calculations previously made and simply looking for a formula that does not contain the quantities analogous to X. The formula for the efficiency satisfies this (L in lines 1 and 6). Then in this formula the one quantity that does appear (h or p) is replaced by its already known corresponding quantity, the temperature (J in l. 3). This gives:

η=

T1 − T2 ∆T1 − ∆T2 or, according to J: η = ∆T1 T1

L calls this procedure “copying”, indicating that he experiences this as a superficial trick to get an answer (l. 6). The teacher sees this ‘copying’ as legitimate, because it is a way of using the analogy and to extend its use into predicting a new relation in the new domain, by substituting (thus ‘mapping’) the corresponding term belonging to the new domain into a relation that is homonymous in both base domains. However, the teacher asks for more because “he doesn't assume this formula, he can derive it, if it's all right” (l.5), and J starts explaining his derivation (l.7-25). Instead of deriving it, he starts from the formula itself and derives:

η=

X 1∆T1 − X 1∆T2 (by multiplying above and below with the same term) X 1∆T1

Now by using the idea that X should be a conserved quantity (∆X1 = - ∆X2 in l.14) and that X must be multiplied by T to get an energy (E =T X, earlier arrived at by mapping E = gh ∆m, or E = -p ∆V), he explains that the efficiency above is consistent with

η=

∆E1 − ∆E2 ∆E1

J. formulates this equation in words (lines 23-25). Starting from the “copied” relation, he has now shown that it is consistent with the agreed definition of efficiency, if several other relations arrived at by analogical reasoning are also assumed. Phase 3: Testing candidate-inferences against experience. In order to test the predicted efficiency formula, students proposed to measure the temperatures, then calculate the expected efficiency, ε = (Th – Tl) / Th, and finally “to

306

PRODUCTIVE USE OF ANALOGY IN THE TEACHING OF ENTROPY

think up another way to test the efficiency”. However, for some minutes they showed no progress in designing this “other way”. The teacher then confirmed “you need to have another definition of the efficiency” and wrote ε = ∆Etask / ∆Esource on the blackboard, a definition that they had used before. Students then immediately came up with proposals for measuring the velocity change of the flywheel, and the temperature change and heat capacity of the hot reservoir. This completed students’ design of an experiment that would test their predictions. Students said in advance that agreement between measured and predicted efficiency would “prove” that “∆X1 + ∆X2=0”, which is one of the relations expected by analogy. This use of the verb “to prove” is not according to accepted theories of logic. Students were aware that they had used the assumed conservation property of X in their derivation of the efficiency formula. Therefore, if the efficiency formula turns out to be correct, the various assumptions about X are not falsified; therefore we can say that they are corroborated, but not proved. This third phase, in conjunction with the second, seems to elicit a need for discussing the reliability of conclusions reached. It would be worth trying if, by doing so, students could come to appreciate the distinction between proof and corroboration at this point in their learning. 4. CONCLUSIONS AND IMPLICATIONS As a general conclusion, we can state that the three phases were in fact realised in this educational experiment for this small group of students. Analogical reasoning had led these students to expect a quantity with some of the more important characteristics of entropy, and to connect this expectation to experience. We conclude that the teaching aim of the assignment sequence was reached. However, we can not generalise on the basis of one case; other groups of students may act differently. Most of our observations might be peculiar to the subject that we were trying to teach. For instance, our experience that the second phase of extending the analogy would not have succeeded without first generalizing over two base domains before jumping to the (target) domain of heat engines is presumably not generalizable to analogies in general. On the other hand, each time the expected reasoning involves the creation of a new term, the problem of students proposing known terms instead of creating a new term can reoccur, necessitating a basis for criticism. The observation that, in our case, students did not see or recognise the analogy but ‘constructed’ it – or in other words, that the analogy does not involve a mapping of ‘objects’ between ‘situations’ but a mapping of terms between deliberately chosen descriptions of situations – might be a candidate for generalization. It could be generally true for analogies in Gentner’s sense of the term. For instance, the water analogy as a means for drawing conclusions about electricity seems to ask for a deliberate analysis of both domains (cf. Gentner & Gentner, 1983; Haeberlen & Schwedes, 1999).

W. KAPER, M. GOEDHART

307

Our students did not spontaneously recognise the intended analogy. This has been noticed before, for example by Arnold and Millar (1996) who found that 3 in 94 students spontaneously recognised an intended analogy. In the teaching of thermodynamics, complaints are sometimes heard that entropy is a very abstract idea which is difficult to connect to experiences. We have shown that students were able to make a connection to experience. They could, with a bit of help, propose an experiment that would test their hypotheses about ‘X’. A reasoning by analogy had made it possible to generate these hypotheses, without which the proposed experiment would not have had meaning. This is a second candidate for generalization: possibly the real usefulness of analogies, in research as well as in education, is in generating proposals for meaningful experiments that would be difficult to arrive at by other means. An implication for teaching would mean that analogies can be used to generate hypotheses about invisible entities (like entropy) that can be tested. In this way it would make such entities more concrete. Along the same line, analogies can be used to let students experience part of the scientific process in which such invisible entities are created. REFERENCES Arnold, M. & Millar, R. (1996). Learning the scientific "story": a case study in the teaching and learning of elementary thermodynamics. Science education 80 (3), 249-281. Baker, W.P. & Lawson, A.E. (2001). Complex instructional analogies and theoretical concept acquisition in college genetics. Science Education 85 (6), 665-683. Brown, D.E. (1993). Refocusing core intuitions: a concretizing role for analogy in conceptual change. Journal of Research in Science Teaching 30 (10), 1273-1290. Carnot, S. (1824). Reflexions sur la puissance motrice du feu et sur les machines propres à déveloper cette puissance. Edition critique par R. Fox, 1978, Paris: J. Vrin. Clement, J. (1993). Using bridging analogies and anchoring intuitions to deal with students’ preconceptions in physics. Journal of Research in Science Teaching 30 (10), 1241-1257. Gentner, D. (1983). Structure-mapping: A theoretical framework for analogy. Cognitive Science 7, 155-170. Gentner D. & Gentner, D.R. (1983). Flowing waters or teeming crowds: mental models of electricity. In D. Gentner & A.L. Stevens (Eds.), Mental models (pp.99-129). Hillsdale N.J.: Erlbaum. Gentner, D. & Holyoak, K.J. (1997). Reasoning and learning by analogy: Introduction. American Psychologist 52 (1), 32-34. Gick, M.L. & Holyoak, K.J. (1983). Schema induction and analogical transfer. Cognitive psychology 15, 1-38. Glynn, S.M. (1989). The Teaching-with-Analogies (TWA) model: Explaining concepts in expository text. In K.D. Muth (Ed.), Children’s comprehension of narrative and expository text: research into practice (pp.185-204), Newark DeI.: International Reading Association. Glynn, S.M. & Takahashi, T. (1998). Learning from analogy-enhanced science text. Journal of Research in Science Teaching 35 (10), 1129-1149.

308

PRODUCTIVE USE OF ANALOGY IN THE TEACHING OF ENTROPY

Haeberlen, S. & Schwedes, H. (1999). Learning-processes in analogy-based instruction about electricity: learning to understand the water-model. In M. Komorek et al., Research in Science Education: Past, Present, and Future, proceedings of the second international conference of the ESERA, (vol. 1, pp.198-201). Kiel: IPN. Online at: http://www.ipn.uni-kiel.de/projekte/esera/book/eserbook.htm Herrmann, F. (1989). Physik, Unterrichtshilfen. Teil 1, Auflage November 1989. Karlsruhe: Institut für Didaktik der Physik der Universität Karlsruhe. Kurtz, K.J., Miao, C.-H. & Gentner, D. (2001). Learning by analogical bootstrapping. The Journal of the Learning Sciences 10 (4), 417-446. Zeitoun, H.H. (1984). Teaching scientific analogies: a proposed model. Research in Science & Technological Education 2 (2), 107-125.

DYNAMIC ASSESSMENTS OF PRESERVICE TEACHERS’ KNOWLEDGE OF MODELS AND MODELLING

BARBARA CRAWFORD¹, MICHAEL CULLIN² ¹Cornell University, USA, ²Lock Haven University, USA. ABSTRACT The authors are concerned with identifying and developing preservice teachers’ understandings and use of scientific models related to the nature of science and scientific inquiry. Empirical research suggests that teachers possess uninformed and/or alternative views of aspects of scientific work, in particular of the role of models and modelling in science. In this study we focus on a particular kind of scientific model: models based on mathematical equations and depicting multiple processes. Participants included graduate students and advanced undergraduates in a teacher preparation program for biology, earth and space science, physics, and chemistry in a large university in the U.S.A. The purpose of this paper is to present several assessments used to track our preservice teachers’ understandings, as they engaged in building computer models of pond ecosystems. These assessments, developed for research purposes, include 1) an open-ended questionnaire; 2) a semi-structured interview protocol used in combination with the computer models constructed by preservice teachers, and 3) a process map to track pair conversations and activities. We consider these as dynamic assessments, designed for use with the non-static work of teachers learning to build and test computer models of natural phenomena. Strengths and limitations of these assessments are discussed.

1. INTRODUCTION AND PROBLEM STATEMENT How does one enhance and assess teachers’ understandings of the nature and use of scientific modelling? The authors have grappled with this problem during the last several years (Crawford & Cullin, 2002; Cullin & Crawford, 2003). To enhance our preservice teachers’ understandings, we first created a context for situating our preservice teachers squarely in the act of model building. Second, we utilized various ways to uncover and track their understandings of the nature of scientific models and modelling in science. We addressed the question, how can we assess preservice teachers’ knowledge of modelling in science in order to monitor changes? The assessments described in this paper were developed to ascertain progress made by our preservice teachers, as they endeavored to understand the complexity of models. We used several kinds of assessments to provide a rich view of their modelling understandings: 1) an open-ended Questionnaire to determine quickly the initial and later understandings of models; 2) a semi-structured Interview to probe the depth of their understandings; and 3) a Process Map analysis of the processvideotape to depict how the participants built their models. For the purposes of this paper, we describe these assessments and discuss the strengths and limitations, in 309 K. Boersma et al. (eds.), Research and the Quality of Science Education, 309—323. © 2005 Springer. Printed in the Netherlands.

310

DYNAMIC ASSESSMENTS OF PRESERVICE TEACHERS KNOWLEDGE

light of previous research on views of modelling. We also highlight our design of the interactive learning environment, which enabled our preservice teachers to actively engage in modelling tasks. The vision of science education reform in many countries and the U.S.A. (American Association for the Advancement of Science [AAAS], 1989 & 1993; National Research Council [NRC], 1996) requires students to be knowledgeable in varied aspects of scientific inquiry and the nature of science, including the role of models and modelling. Models guide scientists in building explanations, interpretations, understanding, and discovery, and they enable scientists to generate predictions (Jungck & Calley, 1985). It seems intuitive that teachers themselves need a robust understanding of how models are built and used by scientists in order to support their students in developing knowledge. It is evident from empirical studies that prospective and practicing science teachers possess uninformed and/or alternative views of models and modelling. Teachers may recognize the usefulness of models as pedagogical tools, but they fail to recognize the true power of scientific models – the function of idea testing (De Jong & van Driel, 2001; Harrison, 2001a & 2001b; Justi & Gilbert, 2002; Smit & Finegold, 1995; and van Driel & Verloop, 1999). Therefore, there is a clear need to develop teachers’ knowledge of key aspects of models and modelling. School students can utilize simulation models to help develop understandings of important science concepts, such as diurnal cycling and predator-prey relationships. Simulation models can also provide an opportunity to learn how scientists use model-based reasoning and computational technologies in order to investigate complex phenomena. Finally, school students can learn how to use such technologies to make sense of the natural world. Modelling provides opportunities for students to demonstrate important thinking strategies (Stratford, 1996), learn science subject matter (Harrison & Treagust, 1996), and learn about science inquiry (Schwarz & White, 1998; Wisnudel-Spitulnik, Krajcik & Soloway, 1999). A major assumption of our work is that teachers need an adequate grasp of subject matter and the purpose and nature of scientific models in order to teach their own students (Justi & Gilbert, 2001). It is important for teachers to know how to do modelling and know about modelling (Hodson, 1993). There are only a few studies focused on preservice teachers’ understandings of models and views of using scientific models in classrooms (e.g. van Driel & Verloop, 1999). To address this problem, we created a context for situating preservice teachers as model builders. In our previous work we determined that our secondary science preservice teachers had fragmented knowledge of the purpose and nature of scientific models (Crawford & Cullin, 2002; Cullin & Crawford, 2003). During the last several years we have designed several variations of instructional modules to address this problem. Each of the modules used the modelling software, Model-It, and an authentic investigation. These earlier authentic projects included designing TerraAqua Columns and investigating the water quality of a nearby stream (Cullin & Crawford, 2003). The current research focused on a revision of an instructional module, in this case, one that was designed to explore the ecosystems of two nearby ponds. This study is a formative evaluation of the quality (in terms of validity, practicality,

B. CRAWFORD, M. CULLIN

311

effectiveness) of an intervention, in which we used dynamic assessments to track change. We posed the following questions; 1) What were prospective science teachers’ understandings of scientific models and modelling; and in what ways, if any, did these understandings change after engaging in modelling tasks? 2) In what ways did our assessments depict our prospective teachers’ understandings of models and modelling? In order to determine accurately the nature and extent of our preservice teachers’ knowledge structures, we developed a number of assessments. In our earlier studies we suspected change, but our assessments appeared too coarse to detect small changes in our preservice teachers’ knowledge structures. We based the development of the current assessments on the modelling literature. This paper describes what we were able to learn from these different assessments and the strengths and limitations of the assessments. 2. THEORETICAL FOUNDATION We used the work of Grosslight, Unger, Jay & Smith (1991) as a primary resource for development of the initial questionnaires and the semi-structured interview protocol. In their study Grosslight, et al. used a scoring scheme to describe modelling understandings of students (grades 7 and 12) and of experts. Three general levels of models and modelling understandings emerged from analyses of interviews. Level I modelers have a simple copy theory epistemology, whereby one believes that the purpose of a model is to be like the real thing. Level II modelers distinguish between ideas and/or the purposes motivating the model and the model itself, and realize that the purpose of the model dictates some aspect of the form of the model. Level III modelers are considered “expert-like”, involving an understanding that the target (phenomenon or object) is explained through examination and/or manipulation of the model. One implication of the Grosslight, et al. study included modification of the interview to further test descriptions of the three levels of understandings. Our work is also informed by other studies including: De Jong & van Driel, 2001; Harrison, 2001; Justi & Gilbert, 2001; Smit & Finegold, 1995; and van Driel & Verloop, 1999. In particular, van Driel and Verloop (1999) reported that inservice teachers they studied, used various criteria for deciding what qualifies as a model, and that they rarely mentioned the important role played by models in making predictions. Harrison (2000) reported that only 2 of 25 in-service teachers he interviewed expressed the belief that models could be used as thinking tools. Justi and Gilbert (2002) reported that the sample of teachers they interviewed did not generally acknowledge a need for considering the scope and limitations of models in the presentation of models to students. Smit and Finegold (1995) examined the conceptions of the origin, nature, and functions of scientific models possessed by 196 final-year prospective physical science teachers. They concluded that the preservice teachers’ knowledge of modelling was limited. The theoretical foundation for the design of this study is based on socialconstructivist views of learning and teaching. Following Vygotsky’s theoretical

312

DYNAMIC ASSESSMENTS OF PRESERVICE TEACHERS KNOWLEDGE

framework, social interaction plays a fundamental role in cognition (1978). We developed our modelling context based on an assumption that constructivist views of learning are viable. Aligned with a social-constructivist view of learning, we believe that learning is dependent on building on the learner’s previous experiences in collaboration with others. We extended our work in a previous project, Analyzing Software Scaffolds in Educational Settings for Science (ASSESS), by investigating technology-rich, supportive tools called scaffolds (Reiser, 2002). Our assessments focused on aspects of preservice teachers’ practical knowledge of modelling. 3. METHOD Participants The study took place in a teacher preparation program for secondary biology, physics, chemistry, and earth and space science teachers in a large state university in the eastern part of the U.S.A. The course, offered during the spring semester of 2002, focused on scientific inquiry using a backdrop of technology. The 17 preservice teachers included the following disciplines: biology (7), physics (3), chemistry (2), earth and space (4), and elementary education (1). Of the 17 preservice teachers, 16 who agreed to be part of the study completed the Preinstructional Questionnaire used in our previous research (Crawford & Cullin, 2002). We selected a sample of eight, who represented the mix of disciplines and initial modelling understandings. To select the sample, we used the Pre-Instructional Questionnaires. The authors and a third science educator scored the Questionnaires using the Grosslight, et al. classification scheme to determine if each preservice teacher initially possessed level I, II, or III modelling understandings. Collectively, we agreed that all of the preservice teachers possessed either Level I or II modelling understandings. We believed that at the beginning of the study none of our preservice teachers held an expert or Level III modelling understanding. Using the scores of the Pre-instructional Questionnaires we created the following four pairs for researching the building of the models: Pair 1 – Level I modelers; Pair 2 and Pair 3 – mixed (one Level-I and one Level-II modeler), and Pair 4 – Level II modelers. We believed the collaborative pairing would foster more robust model building and allow participants to negotiate understandings. Designing the Modelling Context In designing the module we utilized a Project-Based Science approach (Krajcik, Blumenfeld, Marx & Soloway, 1994; Marx, Blumenfeld, Krajcik & Soloway, 1997), highlighted by the use of the simulation modelling software, Model-It. Central to the modelling tasks was a field study of two ponds located within a few miles of the university. The ill-structured, problem-based context for our instruction involved the study of these ponds. Our intent was to engage our preservice teachers in the cognitive work of building computer models to answer the Driving Question, what would happen to the fish if you cut down all the trees around a pond? We utilized the software, Model-It, developed at the University of Michigan's Center for Highly Interactive

B. CRAWFORD, M. CULLIN

313

Computing in Education [Hi-ce] (http:// www.hi-ce.org). The software uses three basic components: objects, the actual physical entities of the phenomenon under study; variables, either qualitative or quantitative descriptions of the objects; and relationships that describe how variables affect one another (Jackson, Krajcik & Soloway, 2000). To support students in model construction, Model-It scaffolds the learner in transitioning from what s/he already knows of the world to building a computerized model representation. The process of modelling the pond ecosystem consisted of identification of critical objects, variables, and relationships. The final task involved identifying the correct causal assumptions. (See Figure 1 for a screenshot of the Model-It software.)

Figure 1. Relationship Editor of Model-It 4. ASSESSMENTS In the following section we highlight three of our assessments: 1) an open-ended Questionnaire, 2) a semi-structured Interview, and 3) a Process Map for analyzing the process-video. The open-ended Questionnaire and the semi-structured Interview were modified from the Grosslight, et al. (1991) instrument. These assessments served to reveal our preservice science teachers’ understandings of scientific models and modelling, and in what ways they changed during the modelling tasks. The Questionnaire was identical to the one used in a previous study (Crawford & Cullin, 2002). The questions were as follows: 1) What is a scientific model? 2) What is the purpose of a scientific model? 3) When making a model, what do you have to keep in mind or think about? 4) How close does a model have to be to the thing itself?

314

DYNAMIC ASSESSMENTS OF PRESERVICE TEACHERS KNOWLEDGE 5) Would a scientist ever change a model? If so, why? If not, why not? 6) Can a scientist have more than one model for the same thing? If so, why? If not, why not? 7) Is teaching about models important in your area of science? Why or why not? 8) Do you intend to teach students about models and modelling? Why or why not

The Interview protocol was semi-structured and developed in collaboration with researchers at the University of Michigan, U. S. A. The interviewer used questions that followed the questionnaire, but which also varied, based in part on each participant’s responses to the questionnaire. The purpose of the Interview was for clarification and to provide an opportunity for in-depth probing into participants’ understandings. The Pre-instruction Interviews were audio taped, and the Postinstruction Interviews were both audio taped and video taped. An important aspect of the Post-instruction Interviews was that these were context-rich. The interviewer (the second author) encouraged participants to reference their own pond models, which they had built. During the interviews he probed preservice science teachers’ ideas, as the participants discussed the models they had built and tested (Creswell, 1998). Developing the Matrix of Modelling Dimensions To assess our preservice teachers’ modelling understandings we used a set of a priori modelling understandings developed from the literature. In this way we expanded, as well as refined, the three-level Grosslight, et al. (1991) rating system. Using the literature-based features of scientific modelling we created a matrix depicting various levels of these understandings. We referred to the modelling features as Dimensions. Our Dimensions are similar in nature to what Justi and Gilbert (2003) referred to as “Aspects”. Justi and Gilbert’s seven Aspects of notions of a model included: the nature of a model, the use to which it can be put, the entities of which it consists, its relative uniqueness, the time span over which it is used, its status in the making of predictions, and the basis for the accreditation of its existence and use. We assumed that our preservice teachers’ modelling understandings would be of varying levels depending on the aspect or, in our case, the dimension. The five by four matrix depicts five inter-related dimensions of modelling understandings: a) Purpose of Models, b) Designing and Creating Models, c) Changing a Model, d) Multiple Models, and e) Validating and Testing Models; and four levels of understandings (limited, pre-scientific, emerging scientific, and scientific) for each dimension (Table 1). We arranged responses to the same items on the pre- and post-questionnaires in tables (Miles & Huberman, 1994). Process Map of Analysis of the Process-video In order to capture the conversations of the pairs as they built their models, we used a data collection method called process-video (Krajcik, Simmons & Lunetta, 1988).

B. CRAWFORD, M. CULLIN

315

A video camera was set up behind the heads of each pair of participants and was focused on the computer screen. An external lapel microphone was fixed to the bottom of the computer monitor in such as a way as to capture clearly the preservice teachers’ conversations as they worked on building their models. We employed this process-video technique to capture pair conversations during the two 1-hour modelbuilding sessions, because we were interested in the ways in which preservice science teachers go about constructing models. The video was divided into segments that we termed, episodes; these segments were tagged using the various components of model building. Displays were then created to show the episodes and the modelling tasks along the two axes. These displays provided graphical representation useful for discerning patterns across participants; we refer to this graphical representation as a Process Map. An example of a Process Map appears in Figure 2. In this display, number 5 (#5) represents the testing mode. The model is associated with testing and involves setting up the test, running the test, changing the rate of the relationship, and evaluating the results. One can clearly see from studying Figure 2 that there is a pattern depicting a non-iterative approach to the task of building a pond model. It is apparent that Jackie and Marvin did a minimum of testing and revising their model in their first session. The graphical representation offers an opportunity to compare modelling activities over time, from one modelling session to the next. An excerpt of the narrative analysis characterizing the modelling activities of Jackie and Marvin follows: 1st Modelling Experience With Model-It Jackie and Marvin defined numerous objects and variables following the general pattern of first defining all objects, then defining the variables for those objects. This pattern is clearly seen in the Process Map above. These preservice teachers did not appear to be focused on the driving question for most of the session. This was evidenced by their failure to create a “fish” object. The only testing they did during this session was for pairs of variables that were directly connected. In other words, they tested the relationships they had previously built that contained no interconnectedness to other variables. In addition to the overall analysis of the modelling behaviors, we analyzed the talk during a specific Episode. Following is an example of a transcription of the recording of Marvin’s talk during Episode 23. M: What I don’t understand, it’s tied into what Mark said… where is the data at? That’s what I want to see, when we get to the relationship thing. I don’t have any data to base these relationships on… I don’t even know… I’m only making these relationships, because of what I know about ponds and my own background… just personal experiences, playing around in a pond as a kid. I don’t have empirical… I’m just using intuition of what I’ve read and watched on TV… stuff like that.

316

DYNAMIC ASSESSMENTS OF PRESERVICE TEACHERS KNOWLEDGE Table 1. Matrix of Modelling Dimensions developed from the literature. Limited

Dimension

PURPOSE OF MODELS

I

Teaching purposes. Used as an aid in making an explanation to someone else.

DESIGNING AND CREATING MODELS

CHANGING A MODEL

Models are not changed.

PreScientific II

Emerging Scientific III

Scientific

A model is used to think with. It is something to help visualization, while user is thinking about phenomenon. Used as aid in formulating an explanation.

Model is used in place of target. Test unsafe/potentially destructive things.

A model is a research tool that is used to obtain information about a target that cannot be observed directly. A model bears certain analogies to the target, thus enabling the researcher to derive hypotheses from the model that may be tested while studying the target. Testing these hypotheses produces new information about the target. (van Driel & Verloop, 1999)

Connection between modeler’s ideas and the model what the modeler thinks rather than what they are trying to get across.

Get the model to behave like the target. This would result in different relationships being built into the model.

A model is developed through an iterative process, in which empirical data, with respect to the target, may lead to a revision of the model, while in a following step the model is tested by further study of the target. (van Driel & Verloop, 1999)

A model is changed primarily when new discoveries are made.

A model is changed when it does not behave like the modeler wants it to.

Models are temporary in nature. (Smit & Finegold) A model is changed when its behavior is not in agreement with observations of the target.

IV

B. CRAWFORD, M. CULLIN

317

MULTIPLE MODELS FOR THE SAME THING

Different models are the result of different learning modalities, educational levels, audiences, or forms.

Different models result from different modeler’s ideas OR from focusing on different aspects of the target.

Different modeler’s ideas represent competing models or theories for explaining the target phenomenon.

Different models for the same phenomenon result from different assumptions about the target or addressing different aspects of the target. (Grosslight et al, 1991)

VALIDATING / TESTING MODELS

No reference.

Models are validated by the scientific community (an external authority).

Models are validated by comparing the behavior of the model with the behavior of the target.

Models can be checked or verified by comparing the results obtained by manipulating the model with the observations obtained in the real world. (Grosslight et al, 1991)

Jackie/Marvin 1st Use Activity During Each Episode 10 Activity

8 6 4 2 0 0

10

20

30

40

50

8-create object 7-define variable 6-build relationship 5-test model 4-revise model 3-discussion 2-question instructor 1-other

Episode Number

Figure 2. Example of a process map for one pair’s first modelling session.

318

DYNAMIC ASSESSMENTS OF PRESERVICE TEACHERS KNOWLEDGE

The following is our interpretation of that talk: Episode 23, 1st use Marvin expressed his concern over how to arrive at relationships. He recognized the fact that relationships need to be based on how he and his partner thought things should be related. He added that he wanted data upon which to base the relationships. However, Marvin failed to recognize that modelling involves building a model based on what one thinks to be the relationship among variables. 5. USE OF ASSESSMENTS TO DEPICT UNDERSTANDINGS As in our earlier work we found that, despite nearly completing a major in a scientific discipline (either biology, physics, chemistry, or earth and space science), our preservice teachers, prior to the study, had had very little experience in actually building and thinking about scientific models. Following the model-building module, the pre- and post-instructional assessments suggested that our preservice teachers developed enhanced understandings of some of the dimensions of modelling. (Detailed results are described in a later paper.) Using the Dimensions Matrix An example of one of the participant’s change from pre- to post-instruction is illustrated in Table 2. Although there is evidence of movement towards more sophisticated understandings, this participant did not fully achieve a scientific level of understanding of any one Dimension of modelling. When looking across the eight preservice teachers, Figure 3 gives an overview of changes for one of the Dimensions. This Dimension involves understanding about how models change (models are temporary in nature). As shown in this composite view (Figure 3), there is clear evidence of enhanced knowledge across the participants. Although we did detect many positive changes, preservice teachers tenaciously held on to some scientifically uninformed views. In particular, we noted little change in understanding the relationship of the model to the target. Using the Process Maps When comparing the Process Maps across pairs of participants, we were able to determine that first, there was typically more testing and revising during the second use of Model-It than in the first session; second, pairs varied in the amount of iterative testing/revising they carried out in the first session. For example, one pair (Kate/Matt) did substantially more iterative testing/revising in the first session than did other groups (e.g. Jackie/Marvin). On the negative side, surprisingly few insights were revealed about gaining modelling knowledge from using process-video. However, much information was revealed about pairs’ subject matter knowledge –

B. CRAWFORD, M. CULLIN

319

that of the pond ecosystems. Surprisingly, we found that even our secondary biology majors demonstrated areas of inaccurate and shallow subject matter knowledge. Pre-module Understandings Level

1

2

3

Post-module Understandings 4

1

2

3

4

DIMENSION

PURPOSE OF MODELS

√

√

DESIGNING AND CREATING MODELS CHANGING A MODEL

√ √

√

MULTIPLE MODELS FOR THE SAME THING VALIDATING/TESTING MODELS Level 1 – limited Level 2 – pre-scientific Level 3 – emerging scientific Level 4 – scientific

√

√

√

√

√

Table 2. Example of changes in dimensions ratings for one participant using pre and post Interview data

Figure 3. Changes in understandings across one of the Dimensions of modelling

320

DYNAMIC ASSESSMENTS OF PRESERVICE TEACHERS KNOWLEDGE

Fine-tuned assessments tracked small changes The expansion and refinement of the Grosslight, et al. (1991) levels, using our Matrix of Modelling Dimensions, enabled us to fine-tune our assessments and track small changes across participants. After immersing our preservice science teachers in an active modelling experience and engaging them in reflection on the experience, our preservice teachers’ awareness of the usefulness of models in making predictions was notably enhanced. One of the most interesting aspects of our study is the type of information gained from each kind of assessment. When we used only the Pre-Instructional Questionnaire as an assessment, we determined that none of our preservice teachers held expert understandings prior to the module. An important point is that after analyzing the Pre-Instructional Interviews, two of the preservice teachers appeared to possess initially, what Grosslight, et al. (1991) would have classified as a Level III understanding of how scientists use models. For example, the analyses of the PreInstructional Questionnaires had revealed that one preservice teacher, Marvin, held a very limited view of models. Specifically, Marvin viewed a model as an exact replica of the target. Marvin imagined building a model space station here on earth that might eventually be built on Mars. When using our Matrix of Modelling Dimensions, we determined through the Pre-instructional Interview that Marvin, instead, held a more advanced view of scientific models, consistent with an engineering approach. This contrasted with our initial determination of a Level I using the Grosslight, et al. levels. The two authors and a third science educator independently arrived at this (same) conclusion. This situation suggests that the use of a single assessment instrument may give an inaccurate or an incomplete picture of a teacher’s knowledge structure. 6. DISCUSSION Strengths of Assessments We claim there are several positive aspects of the assessments we developed. These include: 1) Matrix of Dimensions depicts subtle changes. These changes are more finedtuned than the Grosslight, et al. levels, and the Matrix enables a researcher to detect understandings related to various aspects of scientific models. 2) The context-rich nature of the assessments. The post-instructional Interviews were based primarily on the modelling experiences gained during the module. We believe that the modelling experience itself (doing modelling) provided the needed context for accurate determination of the extent of our preservice teachers’ modelling understandings. 3) Process-video tracks thinking-in-action during modelling. The process-video data allowed us to uniquely uncover what the modelers actually did and thought. We were able to determine how they selected their modelling activities, and we could compare these to how scientists build and test models. This analysis has the potential to inform a researcher about the way different people go about the tasks of

B. CRAWFORD, M. CULLIN

321

modelling. The Process Map display has the potential to give yet another view of a modeler's complex knowledge structure. Limitations of Assessments Although we claim there are many positive aspects to our assessments, we acknowledge that there are limitations as well. 1) The four by five Matrix of Dimensions is more difficult to use than the three levels of Grosslight, et al. We are in the process of doing reliability testing and clarifying each of the four levels. 2) Interpreting each of the assessments is time consuming, when compared to scoring a Likert-style questionnaire. In particular, viewing and interpreting the process-video takes a great deal of time and requires simultaneous use of a computer and video recorder. 3) The Questionnaires used alone were misleading. 7. CONCLUSION AND IMPLICATIONS In addressing the first research question, 1) What were prospective science teachers’ understandings of scientific models and modelling; and in what ways, if any, did these understandings change after engaging in modelling tasks?, we determined that active engagement of preservice teachers in the integrated activities of model building, authentic inquiry, and collaboration, all contributed to enhanced conceptions of modelling. Although we cannot claim a causal result, we can assume that the modelling experiences, in total, positively influenced these preservice teachers’ conceptions of modelling. Related to the second question, 2) In what ways did our assessments depict our prospective teachers’ understandings of models and modelling, it is clear that the different assessments shed different light on our understandings of our preservice teachers’ knowledge of models and modelling. Implications of this study include those for research design and science teacher development. We acknowledge that we would change certain aspects of the research design in future studies to allow us to better track individual knowledge structures in a reliable and valid way. Related to science teacher education, the lead instructor of the module (second author) recognized several missed opportunities, during which he could have given supporting prompts as pairs were actively engaged in model construction. Use of the process-video allowed recognition of some of these missed opportunities. However, this realization occurred long after the module had ended. In summary, we argue that the use of multiple assessments, used in conjunction with engaging teachers in the active process of model building, provides opportunity for powerful and dynamic research tools. We refer to these as dynamic context-rich assessments. From our experiences and the data reported in this paper, we believe the use of single instruments (i.e. written questionnaire) may limit accurate determination of understandings. We claim that the in-depth interviews, linked with the act of building models of real-world phenomena, provided a rich opportunity for

322

DYNAMIC ASSESSMENTS OF PRESERVICE TEACHERS KNOWLEDGE

exploring teachers’ understandings in robust ways. At the same time we recognize the limitations and the context-dependent nature of these assessments (Shavelson, Baxter & Pine, 1991) and the issues of transferability of knowledge. REFERENCES American Association for the Advancement of Science (1989 & 1993). Science for all Americans. Washington, D.C.: American Association for the Advancement of Science. Crawford, B. A. & Cullin, M. J. (2002, April 7-10). Engaging preservice science teachers in building, testing, and teaching about models. Paper presented at the Annual Meeting of the National Association for Research in Science Teaching, New Orleans, LA. Creswell, J. W. (1998). Qualitative inquiry and research design. Thousand Oaks, CA: Sage Publishers. Cullin, M. & Crawford, B. A. (2003). Using technology to support preservice science teachers in learning and teaching about scientific models. Contemporary Issues in Technology and Teacher Education [Online serial], 2(4). Available: http://www.citejournal.org/vol2/iss4/science/article1.cfm De Jong, O. & Van Driel, J. H.. (2001). Developing pre-service teachers' content knowledge and PCK of models and modelling. Paper presented at the National Association for Research in Science Teaching Annual Meeting, St. Louis, MO. Gilbert, J. K. (1993). Models and Modelling in Science Education. Hatfield, UK: Association for Science Education. Grosslight, L., Unger, C., Jay, E. & Smith, C. (1991). Understanding models and their use in science: Conceptions of middle and high school students and experts. Journal of Research in Science Teaching, 28(9), 799-822. Harrison, A. G. (2000). A typology of school science models. International Journal of Science Education,22, 1011-1026. Harrison, A. G. & Treagust, D.F. (1996). Secondary students’ mental models of atom and molecules: Implications for teaching chemistry. Science Education, 80, 509-534. Hodson, D. (1993). Re-thinking old ways: toward a more critical approach to practical work in school science. Studies in Science Education (22), 85-142. Jackson, S. L., Krajcik, J. S. & Soloway, E. (2000). Model-It: A design retrospective. In M. J. Jacobson and R. B. Kozma (Eds.), Innovations in Science and Mathematics Education : Advanced Designs for Technologies of Learning. Mahwah, N.J.: Erlbaum. Jungck, J. & Calley, J. (1985). Strategic simulations and post-socratic pedagogy: Constructing computer software to develop long-term inference through experimental inquiry. American Biology Teacher, 47, 11-15. Justi, R. S. & Gilbert, J. K. (2001). Teachers’ views on models and modelling in science education. Paper presented at the Annual Meeting of the National Association of Research in Science Teaching, St. Louis, MI. Justi, R. S. & Gilbert, J. K. (2002). Science teachers’ knowledge about and attitudes towards the use of models and modelling in learning science. International Journal of Science Education, 24, 1273-1292. Krajcik, J. S., Blumenfeld, P., Marx, R. & Soloway, E. (1994). A collaborative model for helping middle grade teachers learn project-based instruction. The Elementary School Journal, 94, 517-538.

B. CRAWFORD, M. CULLIN

323

Krajcik, J. S., Simmons, E. R. & Lunetta, V. N. (1988). A research strategy for the dynamic study of students' conception and problem solving strategies using science software. Journal of Research in Science Teaching, 25, 147-155. Marx, R., Blumenfeld, P., Krajcik, J. S. & Soloway, E. (1997). Enacting project-based science. The Elementary School Journal, 97(4), 341-358. Miles, M. B. & Huberman, A. M. (1994). Qualitative data analysis: an expanded sourcebook. Thousand Oaks, CA: Sage Publishers. National Research Council. (1996). National Science Education Standards. Washington, D.C.: National Academy Press. Reiser, B. (2002). Characterizing and evaluating software scaffolds for scientific inquiry. An interactive poster session presented at the annual meeting of the American Educational Research Association, New Orleans, LA. April 1-5, 2002. Schwarz, C. & White, B. (1998, April 13-17). Fostering middle school students' understanding of scientific modelling. Paper presented at the Annual Meeting of the American Educational Research Association, San Diego, CA. Shavelson, R. J., Baxter, G. P. & Pine, J. (1991). Performance assessments: Political rhetoric and measurement reliability. Educational Researcher, 21, 22-27. Smit, J. J. & Finegold, M. (1995). Models in physics: Perceptions held by final-year preservice physical science teachers studying at South African Universities. International Journal of Science Education, 19, 621-634. Stratford, S. (1996). Investigating processes and products of secondary science students using dynamic modelling software. Unpublished doctoral dissertation. University of Michigan. Van Driel, J. H. & Verloop, N. (1999). Teachers' knowledge of models and modelling in science. International Journal of Science Education, 21(11), 1141-1153. Vygotsky, L. S. (1978). Mind in society: The Development of Higher Psychological Processes (M. Cole, V. John-Steiner, S. Scriber, & E Souberman, Eds. and trans.). Cambridge, MA: Harvard University Press. Wisnudel-Spitulnik, M., Kracjik, J. & Soloway, E. (1999). Construction of models to promote scientific understanding. In W. Feurzeig & N. Roberts (Eds.), Modelling and simulation in science and mathematics (pp. 70-94). New York: Springer-Verlag.

INVESTIGATING TEACHERS’ IDEAS ABOUT MODELS AND MODELLING – SOME ISSUES OF AUTHENTICITY

ROSÁRIA JUSTI¹, JOHN K. GILBERT² ¹Federal de Minas Gerais, Brazil ²The University of Reading, UK ABSTRACT ‘Models and modelling’ has made an increased contribution to research in science education in recent years. Almost all the papers published discuss either the ideas expressed by teachers/students or the implications of such ideas for the practice of science teaching/learning. Here we focus on the research instruments that we have developed in the last six years to investigate teachers’ ideas about the theme. The paper discusses the strengths and limitations of the instruments and their influence on the ‘authenticity’ of the knowledge so gained.

1. INTRODUCTION The importance of models in science emerges from the recognition that they are non-unique partial representations of an object, an event, a process, or an idea, that are used for specific purposes (Gilbert, Boulter & Rutherford, 1998). Scientific knowledge is developed by the dynamic process of modelling phenomena. Therefore, in order to learn science in a comprehensive and contextualised way, students should learn not only the main scientific/historical models, but also their scope and limitations and issues concerning their development. Moreover, students should develop their ability to create, express, and test their own models. A distinction has to be established in science education between curricular models – simplified versions of scientific or historical models that are taught to students – and teaching models – representations that are created with the specific purpose of facilitating students’ understanding of such models (Gilbert & Boulter, 1995). Science teachers must have specific knowledge (Shulman, 1987) in order to guide students in the learning of science from such a perspective. Thus: 1. Teachers’ subject content knowledge should include a comprehensive understanding of the curricular models that they are required to teach. This will include an understanding of the entities of which they were constructed and the cause-effect relationships operating within them. They must understand the scope and limitations of each of these models: the purposes to which they can be put and the quality of the explanations to which they give rise. Such understanding should not be taken for granted, given the widespread occurrence of misconceptions of all kinds amongst teachers (Gilbert & Watts, 1983). 325 K. Boersma et al. (eds.), Research and the Quality of Science Education, 325—335. © 2005 Springer. Printed in the Netherlands.

326

INVESTIGATION TEACHERS’ IDEAS ABOUT MODELS AND MODELLING

2. Teachers’ subject content knowledge should include a comprehensive understanding of the nature of what a model is per se (Justi & Gilbert, 2002a). 3. Teachers’ curricular knowledge should include when, how, and why both the general idea of models and the natures of specific scientific/historical models should be introduced into the curriculum. In other words, teachers should be able to develop and/or change existing curricular models related to the topics that have to be taught in their classes. 4. Teachers’ pedagogical content knowledge should include: their ability to develop good teaching models, their ability to conduct modelling activities in their classes, their understanding of how their students construct their own mental models, and how they should deal with the resulting expressed models in class (Gilbert et al., 1998). The recognition of the role of models and modelling in science education is fairly recent (e.g. Gilbert & Osborne, 1980; Norman, 1983; Clement, 1989; Mayer, 1989; Nersessian, 1999; Harrison & Treagust, 2000). Consequently, many science teachers all over the world have not been explicitly equipped with such knowledge and skills. In our view, the initial step in this process has to be the investigation of their current knowledge about models and modelling. Our involvement with such an investigation has resulted in a dynamic process of proposition, testing, and changing of a series of research instruments, as well as in the reframing of our own ideas on models and modelling. In the first project that we jointly conducted in this area, a semi-structured interview was used with Brazilian science teachers. Due to the time consumed in conducting and analysing the interviews, we then developed a ‘closed item’ (a Likert-type) questionnaire for application to larger samples. The results obtained with this questionnaire suggested that respondents were sometimes influenced by the precise wording used. Combining the results of these two enquiries, we subsequently developed an ‘open item’ questionnaire that was applied to samples of Brazilian, British, Dutch, and Finnish science teachers, its application being followed by interviews in some cases. A comparison of the analyses of both the questionnaires and the sets of interviews has enabled us to address issues of validity and reliability (authenticity) in such studies. In this paper, we detail the process of proposing, testing, and changing each of the instruments or parts of the instruments. The aim of the paper is not to discuss the results obtained from each instrument, but to address issues of ‘authenticity’ in the outcomes produced by the different methods of enquiry. 2. AUTHENTICITY IN QUALITATIVE ENQUIRY The enquiries that we conducted were essentially qualitative in nature: we wanted to identify teachers’ personal meanings for the words ‘model’ and ‘modelling’. We were therefore concerned to ensure the authenticity of the data that we collected. It had to show internal validity, reporting their understanding of ‘model and modelling’ as they saw it (LeCompte & Preissle, 1993). This means that such data should show:

R. JUSTI, J. K. GILBERT

327

–

fairness: it should be a complete and balanced representation of that understanding; – ontological authenticity: it should provide a fresh and sophisticated representation of that understanding; – educative authenticity: it should generate a new appreciation of that understanding; – catalytic authenticity: it should give rise to specific courses of action in the education of science teachers; – tactical authenticity: it should benefit the teachers involved. (Cohen, Manion & Morrison, 2000). It is in the nature of qualitative research that ensuring external validity would be problematic. We would be unable to show that the results of a particular enquiry could be generalised to a wider population. However, we could hope to show comparability, that the people from whom we collected data were like other teachers, and transferability, such that the data might have been collected from other teachers. There were two other aspects of validity to which we would have to pay special attention: – content validity: the research instruments used would have to fairly and comprehensively cover all aspects of ‘models and modelling’; – ecological validity: the research instruments should collect data in as natural a social setting as possible. The other facet of authenticity, reliability – the ability to collect similar data over time and from different respondents – is also problematic in qualitative research. The best that can be done is to ensure that the choice of respondents, the status positions of the researcher and respondents, the social circumstances in which the research took place, and the precise methods of data collection and analysis, all remained as constant as possible (LeCompte & Preissle, 1993). 3. FIRST INSTRUMENT 39 Brazilian science teachers from different school levels, from primary school to university, were involved in the first study. We investigated their knowledge of the nature of models per se (Justi & Gilbert, 2003), of the nature of modelling and of the education of modellers (Justi & Gilbert, 2002a), and of the use of models and modelling in science teaching (Justi & Gilbert, 2002b). A semi-structured interview methodology was used. Such a format was chosen so that all the teachers could be asked in some detail about key aspects of the nature of ‘model’ that were contained in the questions. It would allow the interviewer (RJ) several degrees of freedom. First, it would be possible to modify the sequence and wording of questions or to add secondary questions in order to probe for a deeper understanding of teachers’ views. Second, each teacher would be given an opportunity both to talk about a specific point in which s/he was interested and to pose questions that s/he wished to answer subsequently (Cohen, Manion &

328

INVESTIGATION TEACHERS’ IDEAS ABOUT MODELS AND MODELLING

Morrison, 2000; Merriam, 1988). The questions were piloted with five high school science teachers. The final version was published in Justi and Gilbert (2003). The interview’s questions were initially based on the work of Grosslight et al. (1991), which focused on the identification of people’s beliefs about the structure and purposes of models. However, we were not only interested in how teachers defined the nature of ‘model’, but also in how, if at all, these interpretations changed when they were applied to specific instances. We thus used a set of questions that could give us a comprehensive view of how people recognise a model as such, of what they understand about a given model, and of how they believe a model to be produced. For instance, in questions focusing on the recognition of something as being a model, we included different models of a given object or phenomenon (e.g. a drawing of a solution of KMnO4 in water, a drawing of molecular models representing the dissolution of KMnO4 in water, and a system where small white balls were kept in motion by stirring with a stick whilst some coloured balls were introduced into it). Here we were alert to potential issues arising from the distinction between modes of representation – communication systems which may be used for the representation of a model, i.e. a model may be expressed in a concrete, verbal, visual, or mathematical form (Twyman, 1985) – and attributes of representation – types of information, i.e. a model may be static or dynamic, deterministic or stochastic, qualitative or quantitative (Mirham, 1972). The instances were presented to the teachers in three ways: as a concrete object that could be handled; as writing or a drawing on a card; as a practical demonstration by the interviewer. They were not presented in a fixed order. When probing ideas concerned with modelling, we decided not to ask directly what teachers thought about the process. We asked them instead to produce a model of a process that might be unknown to them (how soft drink machines work) and a model of a scientific idea (the occurrence of a chemical reaction). They were questioned about several aspects of the processes involved, e.g. how they went about the process of modelling; the possible use of multiple representations; the explanatory and predictive capabilities of their models. The interviews were conducted in Portuguese, transcribed in full, and translated into English. Our main aim when starting the analysis of the transcribed interviews was to organise the teachers’ ideas creatively. This was done so that they could be discussed in order to detect possible patterns within and across them. Given the amount of data obtained, we decided to do this with the help of the Q.S.R. NUD*IST Vivo® qualitative data analysis package. Such a decision resulted not only in saving a lot of time in indexing the data, but also in being able to refine our indexing system. The initial system was based on the examples about which the teachers were questioned. As soon as the process of analysis was started, it was changed in order to allow for the cross-indexing of all those parts of the text relevant to a discussion of the research question. From the reading of each interview transcription, categories and sub-categories concerned with different issues emerged and were modified. As soon as a new category or sub-category was created, all the previously analysed interviews were re-analysed in order to check whether some of their parts could be categorised in a different way. After the analysis of all the transcripts, the whole system of categories and sub-categories was revised. By so

R. JUSTI, J. K. GILBERT

329

doing, we refined the system, making changes that provided a more effective forum for a discussion of interesting aspects of the teachers’ ideas. At this stage we established the ideas of ‘Aspects’ and ‘Categories of Meaning’ on which we based our discussion of teachers’ ideas. For instance, we defined “Nature” as an ‘Aspect’ of teachers’ understanding of models. For it, we defined four ‘Categories of Meaning’: a model is “a reproduction of something”; “a representation of the whole of something”; “a representation of part of something”; “a mental image”. The whole system of ‘Aspects’ and ‘Categories of Meaning’, as well as the analysis of the data from its use, is presented in Justi and Gilbert (2003). The definition of the categorisation system was one of the main results of that project. It allowed us to discuss the similarities, particularities, scope, and limitations of knowledge of teachers from different school levels, as well as to propose some changes in science teachers’ education. Moreover, it enabled the following to take place: (i) the development of later instruments; (ii) the analysis of the data gathered in later research projects; and (iii) the definition of levels of understanding of models (Justi & Gilbert, in preparation). Despite the satisfactory results obtained, we decided not to use this semistructured interview in later studies. This was because each interview took around one hour, which might discourage teachers from participating in such enquiries. Moreover, even after both the definition of the system of aspects and categories and the use of the software to analyse data, the analysis of each interview was a very time-consuming process. Such factors led us to conclude that such a methodology could only be used in investigating a small sample of teachers. As, at that time, we were interested in probing the knowledge of larger samples of teachers, a new instrument had to be developed. 4. SECOND INSTRUMENT We opted for a Likert-type questionnaire, in which teachers would only have to read each sentence and judge it according to their beliefs – a process that should not consume a lot of time. We decided that the system of categorisation resulting from the first study should be the basis of the ‘stem’ sentences in the questionnaire. For instance, from the ‘Categories of Meaning’ for the ‘Aspect’ “Nature”, we proposed the following sentences as the initial ones of the questionnaire: ƒ“A model is a copy of something.” ƒ“A model is very alike something in every way except for its size.” ƒ“A model represents the whole of something”, “A model represents only part of something.” ƒ“A model is a mental image of something.” Due to the comprehensiveness of the system of categories, 74 sentences were included in the first version of the ‘Views on Models and Modelling’ questionnaire (VOMM A). The questionnaire was answered by two distinct samples: 74 Brazilian chemistry teachers from both the medium level of schools (for 15-17 year old students) and

330

INVESTIGATION TEACHERS’ IDEAS ABOUT MODELS AND MODELLING

from a university (Viana & Justi, 2001), and 19 British pre-service science teachers. As the initial sample was similar to that in the first study, we could do a content validation of the questionnaire by comparison with the interview study. Teachers in such a sample were divided into seven sub-groups according to their teaching service (from less than one year to more than 20 years). This allowed us to identify patterns of ideas in specific sub-groups. No statistical treatment of the data in that study was undertaken because it aimed only to produce a general characterisation of those teachers’ ideas. However, for each sentence were produced tables showing the percentage of teachers from each sub-group that gave each answer. From the data gathered in this study it was possible not only to identify general ideas expressed by each sub-group of teachers, but also to establish some relationships between answers for different, but content-related, questions. It emerged that each sub-group of teachers gave very coherent answers. However, when the questionnaires completed by specific teachers, whose actual teaching practices were been investigated at the same time in another research project, were analysed, it was realised that, in questions concerning PCK, some gave answers that were not in agreement with their teaching practices. In other words, they answered according to the ‘image’ that they wanted to present to the researchers. For instance, most of such teachers agreed with sentences like “Students learn how to produce models in science classes” and “Teachers should use the discussion of students’ models as a way to achieve understanding of the models produced by scientists”. However, none of their teaching practices included modelling activities. Such results made us think that respondents were sometimes influenced by both the precise wording used and their wish to avoid showing the reality of what happened in their classes. Finally, many teachers who had initially agreed to participate in the study withdrew when they received the questionnaire. The reason was always the same: ‘I do not have the time to answer so many questions’. This, despite our attempts to be make VOMM A ‘user-friendly’. We came to two conclusions. First, the number of questions should be decreased so teachers would be more willing in answer the questionnaire. Second, whilst a Likert scale questionnaire allows teachers to express the extent to which they agree with a given idea, it does not allow them to express their own ideas (Cohen et al., 2000). We then started to think about a different kind of written instrument that would address these challenges. 5. THIRD INSTRUMENT We decided to produce an open-ended questionnaire, but as this would take more time to answer than a Likert-type questionnaire, we decided to pose only a few questions. We tried to extract, from our system of ‘Aspects’ and ‘Categories of Meaning’ (Justi & Gilbert, 2003), the most relevant elements by which to characterise teachers’ subject content knowledge. We also decided to ask teachers about those aspects we thought should be an important part of their content knowledge and whose characterisation had been impossible with the previous instrument. These were: the identification of different modes of representation, the

R. JUSTI, J. K. GILBERT

331

relationship between historical and scientific models, and the role of models in the development of scientific knowledge. We also decided to pose both direct and indirect questions, the latter being based on examples. Such examples would involve both teaching and historical models that should be very well known to science teachers. For instance, teachers had to identify parts of some phrases with which they disagreed and to justify their ideas. Two of the sentences were: “The model of the atom proposed by Dalton was not used in science after Thomson later proposed a different model” and “Ball-and-stick models are very useful because they show us what a given structure is”. The initial version of this questionnaire (VOMM B) was piloted with a small sample of Brazilian chemistry student teachers who were interviewed about their understanding of the questions. By taking into account their suggestions, we produced another version, VOMM C. This was then used to investigate 63 science student teachers divided into four samples: Brazilian, British, Dutch, and Finnish. These teachers answered the questionnaire in periods that varied from 30 to 60 minutes. The analysis of the data was conducted by using our system of ‘Aspects’ and ‘Categories of Meaning’. From the results, we built tables showing which student teacher expressed each idea. For the Dutch teachers, the analysis was conducted independently by two researchers and the results obtained by each were compared. When necessary, discussion took place to reach an agreement. However, no significant differences were found which points to a good reliability for the analysis. Brazilian and Dutch student teachers were also interviewed in order to validate our analysis of the results. They were asked to explain and justify their answers. When it was necessary, some questions were added to clarify ambiguities. The comparison between the ideas expressed in both situations indicated that our interpretation of the ideas expressed in the VOMM C questionnaire were congruent to those expressed during the interviews. In order to evaluate the extent to which each question was really motivating teachers to express their ideas about a given aspect, we also produced a table showing how many teachers had expressed a specific idea in each question. By comparing this table with the defined aims for each question, we could identify the effectiveness of questions in probing a specific idea. The analysis of this table showed us that the most effective questions were those that asked directly for a specific answer and those that involved very simple examples. Questions related to historical models (e.g. the one that quoted the models of the atom proposed by Dalton and Thomson) were among the less effective ones; most of the teachers said they did not know about such models! This was very surprising to us because the historical examples used were very important in the development of science and (we think) should have been well known to science teachers. These results imply that the VOMM C questionnaire cannot be used with teachers who may not have an adequate knowledge of the history of science. Finally, the smaller number of questions seems to have only partially solved the problem of teachers’ unwillingness to answer the whole questionnaire. Some of them, mainly the British and Finnish teachers, tended to answer with a few words and/or left lots of questions blank. This points to another relevant aspect in

332

INVESTIGATION TEACHERS’ IDEAS ABOUT MODELS AND MODELLING

conducting educational research: the selection of the sample. It is essential that teachers – and students – participating in a research study are strongly motivated to do so. 6. AN EVALUATION OF THE AUTHENTICITY OF THE THREE INSTRUMENTS Table 1 contains a summary evaluation of the three instruments used against the multiple criteria for ‘authenticity’. The following conclusions may be tentatively drawn: a. The use of an interview approach (the first instrument) with as large and diverse a sample as possible provided the bedrock of ideas about ‘aspects’ and ‘categories’ on which the research programme was founded. However, the resource requirements of such a study limited its scope. b. Whilst the second instrument enabled much more data to be collected, the inability to obtain reasons for the responses selected made its interpretation dependent on the initial instrument. c. The third instrument provided more explanations for answers given, thus reducing the dependency on the first instrument. However, the success of this approach was very dependent on the precise examples shown. d. All the instruments suffered from two weaknesses. First, they were not set within the context of an enquiry into respondents’ understanding of the ‘nature of science’ per se, which might be expected to provide a background to their responses. Second, what the teachers responded during the enquiry was not necessarily reflected in what they said or inferred in the classroom. 7. DISCUSSION What instrument or combination of instruments should we use to better and/or more effectively investigate teachers’ content knowledge about models and modelling? The answer depends on two related factors: the purposes of the study and the sample used for the enquiry. The first instrument was very successful in producing a characterisation of teachers’ content knowledge about models and modelling. The outcomes were comprehensive pictures that allowed us both to discuss many aspects of pivotal importance and to construct new knowledge about the subject. An absolute majority of the teachers seemed very motivated to participate in the study.

R. JUSTI, J. K. GILBERT

333

Table 1. Summary evaluation of the three instruments used against the multiple criteria for ‘authenticity’ First Second Third Criterion instrument instrument instrument Good: interview Doubtful: only Good: derived used Poor: question some Fairness from initial wording and order aspects/categories instrument varied enquired into Doubtful: Doubtful: Good: aspects and Ontological responses very responses very categories sensitive to words sensitive to content authenticity identified used used Good: better Good: analysis by Doubtful: reasons Educative understanding teacher type for only some authenticity obtained possible answers given Good: refined Catalytic Good: guidance for Good: as earlier guidance for authenticity themes in INSET instruments specific groups Good: as for Good: as for other Tactical Good: see above initial instrument two instruments authenticity Content Good: derived Good: as for Doubtful: limited validity from literature initial instrument coverage only Good: measures Ecological As for initial As for other taken. Poor: not in validity instrument instruments classroom Good: many Good: many measures taken measures taken Poor: question Good: all Reliability order and wording Poor: data analysis measures taken measures as for first varied. Subjective data analysis instrument inevitable They thought seriously about each question and expressed their own ideas – exactly what we needed for an authentic enquiry. However, as the number of teachers investigated was high (39) and as the duration of each interview was around one hour, we generated a huge amount of qualitative data whose analysis was very time-consuming. The way that the semi-structured interview was conducted allowed us to gather insightful data. A series of characteristics of the interviewer are of pivotal importance in getting an understanding of teachers’ ideas (Cohen et al., 2000; Kvale, 1996). The interviewer must: know the subject matter, be an expert in interaction and communication, and be able to establish an atmosphere so that the interviewee could feel secure to express his/her ideas. As a consequence, the instrument must be used by a well-educated and well-trained interviewer.

334

INVESTIGATION TEACHERS’ IDEAS ABOUT MODELS AND MODELLING

Probing the knowledge of larger samples of teachers is a task that cannot be undertaken with a ‘comprehensive interview’ approach. If it is only possible to investigate some components of ‘models and modelling’, it is better to use open questionnaires (like VOMM C) than closed questionnaires (like VOMM A). When teachers are motivated to answer the questionnaire, they will express their own ideas and doubts, instead of only agreeing (maybe only partially) with ideas expressed by a researcher – as we observed in using VOMM A. Improving VOMM C is not a simple task. If we substitute more simple examples for the historical examples that are not known by most of the teachers, we run the risk of producing statements that do not stimulate them to discuss their ideas. If we substitute general statements for the historical examples, the risk is again that teachers would not express their own ideas. Perhaps we should try to reach a balance between discussing interesting instances and relevant general ideas. However, it must be a situation in which teachers use their own words to express their ideas. In this case, the limitations to authenticity would depend on an individual’s capacity for written expression. The emergence of so many questions corroborates the common-sense knowledge that it is impossible to produce a perfect research instrument to probe teachers’ knowledge. Perhaps a combination of instruments is the best way forward. However, turning the spotlight on questions like these may help researchers to become aware of how the strengths and limitations of each instrument developed influences the knowledge base that may be built for science education. REFERENCES Clement, J. (1989). Learning via model construction and criticism. In J. A. Glover, R. R. Ronning and C. R. Reynolds (Eds.), Handbook of Creativity (pp. 341-381). New York: Plenum. Cohen, L., Manion, L., & Morrison, K. (2000). Research Methods in Education, 5th ed. London: Routledge Falmer. Gilbert, J.K. & Boulter, C.J. (1995). Stretching models too far. Paper presented at the Annual Conference of the American Educational Research Association, San Francisco, 18-22 April. Gilbert, J., Boulter, C., & Rutherford, M. (1998). Models in explanations, Part 1: Horses for courses? International Journal of Science Education, 20(1), 83-97. Gilbert, J.K. & Osborne, R.J. (1980). The Use of Models in Science and Science Teaching. European Journal of Science Education, 2(1), 3-13. Gilbert, J.K. & Watts, D.M. (1983). Concepts, misconceptions, and alternative conceptions: changing perspectives in science education. Studies in Science Education, 10, 61-98. Grosslight, L., Unger, C., Jay, E., & Smith, C. (1991). Understanding models and their use in science: conceptions of middle and high school students and experts. Journal of Research in Science Teaching, 28, 799-822. Harrison, A.G. & Treagust, D.F. (2000). A Typology of School Science Models. International Journal of Science Education, 22(9), 1011-1026.

R. JUSTI, J. K. GILBERT

335

Justi, R. & Gilbert, J. (2002a). Modelling, teachers’ views on the nature of modelling, implications for the education of modellers. International Journal of Science Education, 24(4), 369-387. Justi, R. & Gilbert, J. (2002b). Science teachers’ knowledge about and attitudes towards the use of models and modelling in learning science. International Journal of Science Education, 24(12), 1273-1292. Justi, R. & Gilbert, J. (2003). Teachers’ views on the nature of models. International Journal of Science Education, 25(11), 1369-1386. Justi, R. & Gilbert, J. (in preparation). On the Notion of ‘Level’ in the Understanding of Models. Kvale, S. (1996). Interviews. London: Sage Publications. LeCompte, M. & Preissle, J. (1993). Ethnography and Qualitative Design in Educational Research. London: Academic Press. Mayer, R.E. (1989). Models for Understanding. Review of Educational Research, 59(1), 4364. Merriam, S.B. (1988). Case Study Research in Education – A Qualitative Approach. San Francisco and London: Jossey-Bass. Mirham, G. (1972). The modelling process. IEEE Transactions on Systems, Man and Cybernetics, SMC-2, 621-629. Nersessian, N.J. (1999). Model-Based Reasoning in Conceptual Change. In L. Magnani, N.J. Nersessian and P. Thagard (Eds.) Model-Based Reasoning in Scientific Discovery. (pp. 5-22) New York: Kluwer and Plenum Publishers. Norman, D.A. (1983). Some Observations on Mental Models. In D. Gentner and A.L. Stevens (Eds.), Mental Models (pp. 7-14). Hillsdale, New Jersey: Lawrence Erlbaum. Shulman, L.S. (1987). Knowledge and Teaching: Foundations of the New Reform. Harvard Educational Review, 57(1), 1-22. Twyman, M. (1985). Using pictorial language: a discussion of the dimensions of the problem. In: T. M. Duffy (Ed.), Designing Usable Texts (pp. 245-312). London: Academic Press. Viana, A.P.P. and Justi, R. (2001). Investigação sobre Concepções de Professores de Química Relacionadas a Modelos [Investigation on Chemistry Teachers’ Ideas about Models]. Research report. Belo Horizonte, Brazil: Department of Chemistry, Universidade Federal de Minas Gerais.

INVESTIGATION OF EFFECTS AND STABILITY IN TEACHING MODEL COMPETENCE

SILKE MIKELSKIS-SEIFERT¹, ANTJE LEISNER² ¹Leibniz-Institute for Science Education, Germany ²University of Potsdam, Germany ABSTRACT A concept for a curriculum concerning the particle structure of matter, which aimed at the development of students’ thinking about models, was developed for grades 9 and 10. Research results in this field indicate that thorough discussions concerning epistemology, models, and reality are necessary in class in order to develop an appropriate understanding of the micro-world (Mikelskis-Seifert, 2002). Accordingly, a learning environment was constructed in which the focus was on developing an understanding of the world of experiences and of the world of models. In this case, students had to distinguish systematically between these worlds (Seifert & Fischler, 2001). When teaching and learning about models, a further objective was that students develop metaconcepts regarding particle conception. As part of an empirical study conducted with 120 students from 8th grade, effects of this approach were analysed; the aim of the analysis was to measure the development of an appropriate understanding of models. We were also interested in the transferability of our approach to an introductory class. The results of this evaluation are presented here.

1. INTRODUCTION Problems of teaching and learning the particle structure of matter The use of particle models is an important issue in science education. Particles and atoms are central examples of modelling scientific phenomena in school. However, many empirical studies show that traditional teaching approaches are rather inefficient. Results of such studies indicate that most students have a far from adequate and comprehensive understanding (for example: Kircher, 1986; Ben-Zvi, 1986; Andersson, 1990; Griffiths & Preston, 1992; Duit, 1992; Fischler et al., 1997; Fischler, 1997). The transition into the micro-world is dominated by the macroscopic thinking of students, and the transfer of macroscopic attributes to submicroscopic objects is the central problem of learning about particles and atoms (Seifert & Fischler, 2001). In the opposite direction, the emergence of new macroscopic properties from combining atomic objects is far from being adequately understood. The results of further studies show that the model nature of particles and atoms is also not really understood (i.e. Harrison & Treagust, 1996; Treagust et al., 2002). A common misconception is that students place their models of micro-objects on the same level of reality as e.g. cars or books. In this project, we focus on the question concerning whether an adequate understanding of particle model of matter can be developed via a special design 337 K. Boersma et al. (eds.), Research and the Quality of Science Education, 337—351. © 2005 Springer. Printed in the Netherlands.

338

EFFECTS AND STABILITY IN TEACHING MODEL COMPETENCE

intervention. Such an intervention is based on the reflective interplay of doing experiments and modelling. Consequences for teaching and learning the discrete structure of matter All didactic approaches have to face these problems, especially the dilemma of visualization. Some researchers try to avoid any visualization of the micro-world (Buck, 1990); some do not use macroscopic analogies of the micro-world. Other researchers discuss hybrid models between a macro- and micro-world (Justi & Gilbert, 1998). However, students are confronted every day with a lot of colourful pictures of micro-objects in the media and also in common schoolbooks. In any case, we have to deal with the meaning of all these visualizations and discuss them with students. In this regard, one major teaching problem is the difficulty in achieving an acceptable understanding of the particle world; that is understanding that most of the macroscopic properties of matter with which we are so familiar are lacking in the micro-world. As a consequence, teachers should more strongly emphasize the differences that exist between a scientist’s understanding of the world of experiences and the world of models presented to students. In our teaching approach students should be supported in the development of a metaconceptual awareness, and in this case students should be able to differentiate between the world of experiences and the world of models. Thus, the students also become acquainted with the specific characteristics of these two worlds. Our hypothesis is that students with increased metaconceptual awareness can overcome their misconceptions about the micro world. In another research project, it was found that metaconception is accompanied by a comprehensive, precise, and coherent knowledge (Fischler & Peuckert, 1999). In other words: the transfer of macroscopic properties to particles is done almost exclusively by those students who have a naive-realistic conception about particles. 2. THE APPROACH OF LEARNING ABOUT MODELS Main objective and basic idea of learning about models In order to develop an appropriate understanding of the particle structure of matter, it is necessary to focus on the specific nature of the submicroscopic world and the modelling of the micro-world (Mikelskis-Seifert, 2002.) The following teaching activities that concern learning about models are therefore integrated into the curriculum:

S. MIKELSKIS-SEIFERT, A. LEISNER

339

1) discussing and reflecting on the nature of (particle) models and the process of modelling; 2) illustrating the construction of various particle models to describe and interpret submicroscopic phenomena; 3) evaluating models for their effectiveness and their limitations; 4) systematic separation of models and experiences when considering experiences and models. This means that it is necessary to distinguish stringently between the world of experiences and the world of models. The basic idea of this approach is the emphasis on an elaborate metaconceptual reflection with regard to perceptions in the world of experiences and in the world of models (Figure 1). An awareness of the existence of these two worlds and their distinction needs to be developed and maintained.

Fig. 1: The basic idea underlying this approach We have constructed a “level system of multiple representations“ to differentiate the worlds. Accordingly, this system contains two areas: the world of daily experiences and the world of models. Within these areas, different sections are distinguished, which in the following representation are called levels. The arrangement of the levels, however, does not express any hierarchy (see Figure 2). The superordinate level of metaconceptual reasoning reflects explicit argumentations about individual levels or combinations of them. The key ideas of such a lesson are: explicit distinction of both worlds; emphasis on continuous model construction and therefore on the hypothetical character of models; use and critical assessment of alternative modelling. Here the development of metaconcepts concerning particle conceptions presents an important aim; the other aim is to address common risks of representing submicroscopic objects.

340

EFFECTS AND STABILITY IN TEACHING MODEL COMPETENCE

Fig. 2: The level system of multiple representations We expect that students will develop a model competence by participating in such a learning environment. In our case model competence means that students have an adequate understanding about the nature of particle models and that they can use particle models to explain macroscopic phenomena in various situations. This is to say that if we could observe in students’ argumentations an appropriate microscopic thinking and adequate thinking in models, as well as a renunciation of macroscopic thinking in the micro-world, then we could say that students have a model competence. Implementation during a project week In order to accomplish an intensive debate concerning models and related issues, learning about models was conducted during an interdisciplinary project week. During this week all students (about 120) of the four grade 8 classes from the same school participated. They worked in groups with the help of worksheets and received support from 8 teachers. The main objective of the interdisciplinary project week was the development of an awareness of the specific character of models. With this foundation, it was then possible to develop further a model competency during additional teaching in science. In conjunction with this, thinking in models needed to be encouraged, for example, through the deliberate and stringent separation of real and modelled phenomena. In addition, students should attain a notion of the structure and the modelling of the micro-world.

S. MIKELSKIS-SEIFERT, A. LEISNER

341

Our project week on particle models was structured in the following four phases: “introduction”, “transition into the micro-world”, “investigations of submicroscopic phenomena in learning units”, and “reflection about modelling the micro-world”. First phase – introduction: The interdisciplinary project week began with the phase of sensitization to various aspects of thinking in models and of the model notion. For example, students were introduced to the characteristics of models with the help of self-made models (Figure 3). Thus, they would also come to know the problems with regard to models (such as for example: problems of modelling and the art of model building by constructing models of objects). Further, a black-box approach to the modelling of reality played a central role in this phase. The aim was that students would be astonished that with the same reality diverse model constructors could come up with different results. This was the decisive point in the following discussion with regard to the limited ability of models to make declarations and the necessity to gain new results through new, expanded methods.

342

EFFECTS AND STABILITY IN TEACHING MODEL COMPETENCE

Fig. 3: Examples of self-made models of the students Second phase – transition into the micro-world: In this phase, students’ conceptions about the micro-world were elicited and the systematic distinction between the world of experiences and the world of models was introduced. A box contained various objects, e.g. a test-tube filled with sand, a test-tube filled with a translucent liquid, and a blackened filled test-tube. The objective of this box was to discover the limits of what students perceived. With

S. MIKELSKIS-SEIFERT, A. LEISNER

343

regard to the experiences collected with the objects in the box, it was resolved during discussions in class that perceivable bodies could be described with the help of attributes like colour, form, and smell; they were thus identifiable as a result of their specific characteristics. However, it was difficult to describe the contents of the blackened filled test-tube. Those tubes were special black boxes; models were necessary to obtain hypotheses about the contents. Furthermore, the systematic distinction between the world of experiences and the world of models was introduced using a poster (Figure 4). A preliminary particle model was elaborated under the strong guidance of a teacher.

Fig. 4: Poster as a tool for analysis of a micro-world in the two worlds Third phase – investigations of submicroscopic phenomena in learning units: During this phase, the class was divided into small groups (three to four persons per group). With the help of several worksheets, each group examined different phenomena concerning crystallization, diffusion, evaporation, and volume increase of a liquid when heated. The task was to interpret the observations based on models. During the investigation of these different phenomena, it was necessary for students to modify their particle models. Each group presented their results and usually a thorough discussion occurred. In this phase the particle model was developed further, step by step. Fourth phase – reflection about modelling of the micro world:After the learning units, students were asked to think and report about the modelling they had carried out. The students were also asked to reflect on the ideas of models they had developed and then to apply them to new phenomena.

344

EFFECTS AND STABILITY IN TEACHING MODEL COMPETENCE 3. METHODS

Design of investigation Effects of the project week (approximately 120 students participating) were investigated by means of the following design (Figure 5). At the beginning of the year, a pre-test was administrated which included a questionnaire on the idea of particles. Concept maps were also used to measure students’ conceptions about the micro-world.

Fig. 5: Design of the empirical study (4 classes of grade 8) The post-test, which was conducted directly following project week, used the same instruments. Selected students were interviewed; in the interviews special attention was paid to questions relating to the transfer. Moreover, students were surveyed 6 months later using the same questionnaire and the same concept map. This was done in order for us to be able to determine the long-term effects of the project week intervention. Methods of investigation In order to investigate students’ conceptions concerning the particle structure of matter, as well as changes in their conceptions as a result of participating in project week, we utilised a particle questionnaire called Berlin Particle Concept Inventory (BPCI; see Mikelskis-Seifert, 2002), based on research results in the field of the particle structure of matter. The instrument consisted of several scales of the different areas of particle conception, problem solving tasks, and a concept map. In

S. MIKELSKIS-SEIFERT, A. LEISNER

345

this case, students’ conceptions could be divided into three categories: microscopic thinking, macroscopic thinking, and model understanding. The following parts of aspects (partial constructs) existed within these categories: forces between particles, movement of particles, distances between particles, no direct application of macroscopic attributes and behaviour, and model understanding (for 6 scales of the different areas of particle conception, see fig. 6). This assortment was adopted and used to evaluate BPCI as a consequence of satisfactory results of analyses for reliability. A variety of empirical procedures, e.g. variety and correlation analyses, latent-class analysis, and quantitative and qualitative assessment based on modal maps, served to evaluate data which were gathered to examine students’ conceptions. Selected students were interviewed after a period of three months to obtain information concerning transfer effects.

Fig. 6: Scales used for investigating the students’ conceptions and their change over time 4. SELECTED RESULTS OF THE EVALUATION Analysis of learning effects in the different scales Can 8th grade students develop an appropriate understanding of particle models as well as a metaconceptional way of thinking as a result of explicitly dealing with worlds of experience and models? With the help of the results of the analysis of scales, we intend to illustrate how one can be successful in learning about models. We first present the learning effects in the six scales of the different areas of particle conception and then the result of the latent-class analysis.

346

EFFECTS AND STABILITY IN TEACHING MODEL COMPETENCE

In parts of the pre-test, it became evident that students had already learned concepts about the particle structure of matter (Figure 7). As a result, students already had a relatively adequate knowledge concerning the forces and distances between particles, or of the movement of particles. In comparison to these microscopic attributes, conceptions of the macroscopic attributes and of the behaviour of the particle model were not suitable nor did a model of understanding exist. After the project week, the desired positive effects became evident in all areas of the particle conception and were highly significant. The learning effects with regard to the thinking about models suggested that students advanced from a naiverealistic conception to a hypothetical-realistic position. Moreover, the results of the analysis of scales elucidated that students were more careful not to apply macroscopic characteristics directly to the smallest particles. The same pertained to ascribing macroscopic behaviour.

Fig. 7: Students’ conceptions and their change as a result of teaching (vertical axis represents students’ average score in the scales) In addition, desired and significant changes in conceptions could be observed in the area of microscopic thinking. The long-term test illustrates how the learning effects were sustained. The differences in the average value which can be seen here are not significant; thus, one can assume that the concepts remained stable.

S. MIKELSKIS-SEIFERT, A. LEISNER

347

The result of the latent-class analysis One possibility to investigate a developed model competency is to calculate the data while taking the profile of answers into consideration. According to our definition of model competency, students have this competency when they show an adequate thinking and argumentation in all three areas (in macroscopic thinking, microscopic thinking, and in thinking in models). This reflects that they answered the items of the six scales according to expectations and, consequently, achieved high scores. Thus, it is the goal of the latent-class analysis of scale values to identify such a profile of answers (Figure 8). The neutral axis in this diagram is now located at two and signifies uncertain argumentation. The parameter 4 then denotes the optimal maximum value in terms of concepts. The profile of answers illustrated here by the line with circles characterises the class of students who have no model competence with regard to the particle structure. The class represented by the line with triangles includes students with uncertain argumentations. Students who have excelled in providing correct argumentation with regard to the intended model competence are illustrated here by the line with squares. Since it was in our research interest to determine the conditions for the development of such a competence empirically, we conducted further analyses. For example, we needed to clarify how many students went through the desired process of development (Figure 9). As expected, only a small number of students were found in the class of model competence during the pre-test. Most students presented only inappropriate argumentation and, therefore, did not have model competency in accordance with our classification. During the post-test, the distribution shifted in the direction of the class of model competence. This means that the development of a modelling competency can be observed. The number of students who moved into the class of model competence correlates with the number of students who left the class of no model competence. As the results of the long-term test show, this effect can be sustained. This result of the latent-class analysis is in accordance with the results of other methods of evaluation Since four student groups were analyzed for the effectiveness of learning about models, it is now interesting to inquire into the types of changes in conception that took place in these student groups. In trying to answer this question, we will only illustrate the results of the average increase in the individual student groups who improved in an exemplary manner (Figure 10)

348

EFFECTS AND STABILITY IN TEACHING MODEL COMPETENCE

Fig. 8: Typical profile of students’ answers about the micro-world

Fig. 9: Diagram of students’ distribution on the 3 latent classes

S. MIKELSKIS-SEIFERT, A. LEISNER

349

Fig. 10: Value comparison of the four students’ groups No significant differences in parameters appeared while implementing this type of data analysis. Thus, similar results of learning can be observed in all four student groups although the student groups were taught by different teachers. The result of student group 8d which reached the maximum value of three is especially interesting because this means that all students of this student group can be classified as having model competency. 5. CONCLUSIONS The following conclusions can be drawn from the analysis of students’ conceptions and the changes in their conceptions as a result of the project week: 1) Considerable improvements in model thinking by students could be observed. These desired positive changes in thinking about models could be attained by students, presumably by dealing explicitly with the worlds of experiences and models. 2) After the project week, students did not transfer inadequate macroscopic attributes and macroscopic behaviour to the smallest particles. 3) Students had a more comprehensive knowledge in their microscopic thinking, which included appropriate argumentation with regard to forces, distances, and movement.

350

EFFECTS AND STABILITY IN TEACHING MODEL COMPETENCE

4) The learning effects suggest that it is necessary to discuss and reflect on the nature of utilized models as well as on the process of already modelling during introduction of a particle model in secondary science education. Furthermore, we can assume that during an interdisciplinary project week the majority of students will develop a stable model competency in the area of the micro-world. An additional empirical study conducted with 10th grade students was performed; in this study teaching about models was done in the same way. For comparison between learning of 8th and 10th graders, the same pre-/post-test design was applied to the group of 10th graders. If one compares, then the following was observed: during the pre-test, students of compulsory courses demonstrated a higher proficiency in almost all areas. The reason for this was that the 10th grade students had already had a classical introduction to particles. The situation changed in the post-test on which younger students attained similar and sometimes even higher marks in partial constructs. Here is where the area “abstention from macroscopic thinking” stands out. Students without a traditional introduction were able to abstain more easily from such thinking in the micro-world. Other differences between average values in sub-areas were not significant. Generally, learning about models supports developing an adequate understanding of the submicroscopic world. However, it seems that teaching about models should start as soon as possible. After a traditional introduction to the microworld, it is difficult to change students’ conceptions and help them develop an understanding of the nature of particle structure. REFERENCES Anderson, B. R. (1990). Pupils’ conceptions of matter and its transformations (age 12-16). In P. L. Lijnse, P. Licht, W. de Vos, & A. J. Waarlo (Eds.), Relating macroscopic phenomena to microscopic particles: a central problem in secondary science education. Utrecht: CD-β Press, 12-35. Ben-Zvi, R., Eylon, B.-S. & Silberstein, J. (1986). Is an atom of copper malleable? In Journal of Chemical Education, 63, 1, 64-66. Buck, P. (1990). Jumping to the Atoms. In P. L. Lijnse, P. Licht, W. de Vos, & A. J. Waarlo (Eds.), Relating macroscopic phenomena to microscopic particles: a central problem in secondary science education. Utrecht: CD-β Press, 212-219. Buck, P. (1981). Eine Unterrichtseinheit über die Natur der Atome. Chimica didactica 7 (1981) 1, 5-24. Buck, P., Johnson, P., Fischler, H., Peuckert, J. & Seifert, S. (2001). The Need for and the Role of Metacognition in Teaching and Learning the Particle Model. In H. Behrendt, H. Dahncke, R. Duit, W. Gräber, M. Komorek, A. Kross & P. Reiska (Eds.), Research in Science Education – Past, Present, and Future. Dordrecht: Kluwer Academic Publishers, 225-234. Duit, R. (1992). Teilchen- und Atomvorstellungen. In H. Fischler (Ed.), Quantenphysik in der Schule. Kiel: IPN, 204-214. Fischler, H., Lichtfeldt, M. & Peuckert, J. (1997). Die Teilchenstruktur der Materie im Physikunterricht der Sekundarstufe I (Teil 1): Kann Forschung den didaktischen

S. MIKELSKIS-SEIFERT, A. LEISNER

351

Wirrwarr beenden? Didaktik der Physik, Vorträge der Frühjahrstagung der Deutschen Physikalischen Gesellschaft, 572-577. Fischler, H. (1997). Was versteht man in der Physik unter „Teilchen“? Naturwissenschaften im Unterricht, Physik, Teilchen, 41, 8, 9-11. Fischler, H. & Peuckert, J. (1999). Stabililty of students´ conceptions concerning particle models. In M. Komorek, H. Behrendt, H. Dahncke, R. Duit, W. Gräber & A. Kross (Eds.), Proceedings of the Second International Conference of the European Science Education Research Association, Vol. 1. Kiel: IPN, 396-398. Griffith, A., Preston, K. (1992). Grade-12 Students’ Misconceptions Relating to Fundamental Characteristics of Atoms and Molecules. Journal of Research in Science Teaching, 29, 611-628. Harrison, A. & Treagust, D. (1996). Secondary students’ mental models of atoms and molecules: Implications for teaching chemistry. Science Education, 80, 5, 509-534. Justi, R. & Gilbert, J. (1998). Historical models in teaching about the atom: Bridging a gap in science education. Paper presented at: AERA Conferences, San Diego. Kircher, E. (1986). Vorstellungen über Atome. Naturwissenschaften im Unterricht – Physik/Chemie, 34, 13, 34-37 Mikelskis-Seifert, S. (2002). Die Entwicklung von Metakonzepten zur Teilchenvorstellung bei Schülern. Untersuchung eines Unterrichts über Modelle mithilfe eines Systems multipler Repräsentationsebenen. Logos Verlag Berlin. Peuckert, J., Fischler, H. & Seifert, S. (1999). Stabilität und Ausprägung kognitiver Strukturen zum Atombegriff. In R. Brechel, R. (Hrsg.), Zur Didaktik der Physik und Chemie – Vorträge auf der Tagung der Physik/Chemie in Essen, Alsbach/Bergstraße, 358-360. Seifert, S. & Fischler, H. (2001). Can Students Develop Meta Concepts for Particle Representation? Presenting an Approach for Introducing the Particle Model. In D. Psillos, P. Kariotoglou, V. Tselfes, G. Bidsdikian, G. Fassoulopoulos, E. Hatzikraniotis & M. Kallery (Eds.), Proceedings of the Third International Conference of the European Science Education Research Association, Vol. 2, 2001. ART OF TEXT publications: Thessaloniki, 624-626. Treagust, D. F., Chittleborough, G. & Maniala, T. L. (2002). Students’ understanding of the role of scientific models in learning science. International Journal of Science Education, 24, 357-368. Vollebregt, M. J. (1998). A Problem Posing Approach to Teaching an Initial Particle Model. Utrecht: CD-ß Press.

USING MULTIPLE ANALOGIES: CASE STUDY OF A CHEMISTRY TEACHER’S PREPARATIONS, PRESENTATIONS AND REFLECTIONS

ALLAN HARRISON¹, ONNO DE JONG² ¹ Central Queensland University, Australia ² Utrecht University, The Netherlands

ABSTRACT The use of analogies by teachers is influenced by their existing knowledge base, especially their pedagogical content knowledge (PCK). With respect to teaching with multiple analogies, little is known about the relationship between teachers’ classroom practice and their PCK before and after teaching. This study explores that relationship. An expert chemistry teacher was the subject of this study, and three lessons on chemical equilibrium for Grade-12 students were observed. The teacher was interviewed about his teaching intentions, and a reflective post-teaching interview conducted. The analysis indicates a number of relevant correspondences and differences between the teacher’s intentions and his classroom practice. After teaching, the teacher appeared to be aware of the relevant correspondences, but was not aware of the differences, especially the absence of his intended attention to the limitations of specific analogies, and the absence of his intended check of students’ understanding of links between an analogy and its target. These results underline the need to pay attention to specific aspects of teaching with analogies in the context of science teacher education.

1. INTRODUCTION The past 20 years has seen a growing interest in the teaching and learning of science using analogies, especially at secondary school level (Duit, 1991; Gabel & Samuel, 1986; Treagust, Harrison & Venville, 1998). There is good reason for this interest because analogies can function as powerful tools facilitating students’ understanding of science topics. While most studies of analogy in school science concern physics and biology topics, this study focuses on chemical equilibrium. More precisely, the study concentrates on the conceptions and actions of an expert chemistry teacher as he taught this difficult topic to Grade-12 students. 2. BACKGROUND Analogies and their use in school science. An analogy can be considered as a relation between structures and functions from two conceptual domains; that is, 353 K. Boersma et al. (eds.), Research and the Quality of Science Education, 353—364. © 2005 Spinger. Printed in the Netherlands.

354

USING MULTIPLE ANALOGIES

similarities exist between the analog domain (a familiar object or event) and the target domain (in our case, a science concept) (Duit, 1991). An analogy expresses a comparison and is created by mapping similar features from the analog onto the target. Every analogy breaks down somewhere because there are always analog features that do not correspond with target features (Glynn, 1989), and this characteristic restricts the extent of every analogy. Curtis and Reigeluth (1984) have classified analogies under three headings based on each analogy’s degree of elaboration. The first type is simple analogy, for example, ‘assembling a car is like the mechanism of a chemical reaction’. The second type, enriched analogy, includes the grounds for the likeness, for example ‘assembling a car is like the mechanism of a chemical reaction, because both cases proceed step by step’. The third type, extended analogy, consists of multiple simple and/or multiple enriched analogies. Duit (1991) sees constructivist advantages in analogy use. Analogies give students the opportunity to relate features of daily-life objects or events to similar features of an abstract science topic, and they also may provide mental visualizations which, in turn, enhance student motivation. He warns that analogy use does not guarantee learning outcomes, for instance, students need to be able to ‘see’ the analogy and teachers need to assess the relevance of students’ prior knowledge. Most studies of concept learning by analogy focus on the teacher’s use of analogies in their classroom practice (e.g. Dagher, 1995; Jarman, 1996; Treagust, Duit, Joslin & Lindauer, 1992). A minority of studies also examine student learning when analogies are used (e.g. Dupin & Johsua, 1989; Harrison & Treagust, 1993) and only a few studies are reported in chemistry education. Thiele and Treagust (1994) observed four chemistry teachers’ classroom practices and found that they used planned analogies to explain abstract concepts to the whole class, and inserted spontaneous analogies when groups of students demonstrated a lack of understanding. Later, Harrison and Treagust (2000) described students’ conceptual changes when multiple analogies were used to teach atoms and molecules in Grade11 chemistry. The effectiveness of analogy use seems to be related to the teacher’s explanatory ability, the students’ familiarity with the analog, and the ability of both to map between similar features. Gentner (1983) recommends analogies whose surface similarities provide easy student access to the analogy, and also develop conceptual relationships. No analogy will cover all features of a target, but multiple analogies can address more features and in different ways. Thus, multiple analogies seem to be more effective than single analogies, especially when the target concept is complex and abstract (Thagard, 1992). Teaching chemical equilibrium. With these ideas in mind, this study reports an expert chemistry teacher’s use of analogies when teaching chemical equilibrium. This topic is central to chemical education, and is considered complex because it includes important sub-topics such as reversible reactions, reaction rates, chemical kinetics, and the dynamic nature of equilibria. Many students misunderstand chemical equilibrium believing that the forward reaction finishes before the reverse reaction commences, and that at equilibrium the reaction stops and ‘nothing

A. HARRISON, O. DE JONG

355

happens’ (Van Driel & Gräber, 2002) and they do not visualise particle collisions when learning about the factors influencing reaction rate (Justi, 2002). Interest in chemistry teachers’ knowledge of how to teach difficult curriculum topics is growing (De Jong, Veal & Van Driel, 2002), but studies that explore chemistry teachers’ intentions, actions, and reflections, with regard to particular topics, are scarce. Consequently, the following research questions guided the present case study: (1) What are the teacher’s intentions when choosing analogies to develop students’ understanding of chemical equilibrium? (2) How are these analogies presented in the classroom? (3) How does the teacher reflect on his/her use of analogies? 3. RESEARCH METHOD Participants. The teacher, whom we called Neil, had taught chemistry for 18 years at senior high schools in Australia. Neil supports his school’s view on teaching, which is ‘teaching-for-understanding’. The class involved in the study consisted of 11 Grade 12 students (3 female, 8 male, average age 17 years). This is their second year of senior chemistry, and Neil was very familiar with the class. Classroom context. The lessons that we studied occurred over two days. On the first day, a double lesson (80 minutes) was given, starting with a recapitulation of the particulate nature of chemical reactions and factors that influence reaction rate, followed by activation energy, reaction profile diagrams, and the conditions for chemical equilibrium. The next day, a single lesson (40 minutes) elaborated the concept of dynamic equilibrium. No teacher demonstrations or student practical work were included in the lessons. The topics were presented and discussed in an interactive way with Neil and his students asking many questions. Data collection and analysis. Research data were collected as follows: prelesson interviews explored Neil’s intentions, the three lessons were observed and audio taped, and post-lesson interviews explored Neil’s reflections. The teacher interviews were conducted by both authors and audio taped; all tapes were transcribed verbatim. The data were analyzed from an interpretative phenomenological perspective (Smith, 1995) aimed at producing trustworthy interpretations. To enhance rigor, all of the transcripts and observation notes were independently read and interpreted by both authors. Individual analyses were compared and discussed, when necessary, to reach consensus (Janesick, 1994). In the larger study, of which this paper is part, students were interviewed and their understandings were compared with Neil’s intentions (Harrison & de Jong, in revision).

356

USING MULTIPLE ANALOGIES 4. FINDINGS

During the three lessons, Neil discussed 10 analogies (see Table 1), one of which was proposed by a student (Table 1, item 9). Five analogies are reported here in detail (Table 1, items 1, 7, 8, 9, 10) because they were given much more attention in the lessons than the other analogies. Table 1. Analogs, in order of appearance in the lessons on chemical equilibrium Analog (familiar situation)

Target (science concept)

1. School dance (shorter version)

Chemical reaction, reaction rate

2. Skier surmounts a slope to go downhill

Activation energy

3. Making an air trip including transfers

Reaction path, intermediates

4. Assembling an aircraft or car in stages

Reaction path, intermediates

5. Balancing on a see-saw

Physical equilibrium

6. Being alternatively sane and insane

Physical equilibrium

7. School dance (elaborated version)

Chemical equilibrium conditions

8. Excess sugar in a cup of tea

Dynamic nature of equilibrium

9. Simmering curry in a pot (lid in place)

Dynamic nature of equilibrium

10. Busy highway

Dynamic nature of equilibrium

Chemistry Lesson 1 and 2 (double lesson) Pre-lesson interview. Neil announced that he was concluding reaction rates and that he would use this concept as a foundation for chemical equilibrium. He intended to use analogies and wanted to impress on his students that an analogy is always artificial, and he wanted them to find the flaws (i.e., where the analogies break down) because, in his opinion, all analogies have limitations. The planned analogy for bridging reaction rate and equilibrium was the ‘school dance’ scenario. In the previous lesson, he introduced this analogy to discuss chemical reactions that go to completion. The analog was a school dance in the school gymnasium (hall or ‘gym’) where everyone is blind-folded. Boys have stubble on their chins and girls have their hair in pony-tails. Feeling the head of a potential partner is the only way boys and girls can pair off. Boy-girl meetings represent compatible atom collisions, and when the couple goes to a side room (the “commitment room”) and commit to each other, this represents a chemical reaction and bonding occurs. In the coming lesson, Neil wanted to elaborate the analogy with respect to reaction rate and the effects of concentration, temperature, and surface area. He also intended to raise a problem: in the model, there are 500 boys and 500 girls in the hall to start with, but the commitment room can only hold 250 couples. Only when one of the 250 couples in

A. HARRISON, O. DE JONG

357

the commitment room splits, can a new couple enter and commit. This introduces the idea of simultaneous forward and reverse reactions with the same rate. According to Neil, teenagers, who easily make and break relationships, understand the dynamics of this analogy. When asked about the functions of the ‘school dance’ model, Neil responded by saying that he tries to present something that the students can visualise, that can help them understand what is happening in the submicroscopic world and that entertains and maintains their interest. Neil then talked at length about the dynamics of teaching and learning analogies, especially in the case of lower achievers, whom he would help by asking them to retell the analog and relate it to the target in their own words. He concluded that he intended using the analog as a lead-in to equilibrium, because he felt the students had not previously encountered reversible reactions. He also told how he likes to introduce analogies without telling the students where he is going. He felt that if he started by announcing the concept he wanted to explain “they will tune out”. He enjoyed “keeping them wondering”. (a) The ‘school dance’ analogy (shorter version). Neil began by recapitulating the ‘school dance’ analogy, and its links to chemical reactions and the factors that influence reaction rate, namely, concentration, temperature, and surface area (see Table 2). He also addressed the grounds for the likeness, but scarcely indicated the analogy’s limitations. For the surface area factor only, Neil indicated that the analogy of ‘extent of body contact’ had broken down, but he did not discuss this issue further, despite his stated intentions in the pre-lesson interview. Neil recapitulated the analogy in a very interactive way, checking that the students shared his version of the story before developing it further, although he did not ask any lower achieving student to tell the whole story in his or her own words, again despite his stated intentions. (b) Intermezzo with more analogies. After the ‘school dance’, Neil explained more kinetic features, namely, activation energy in terms of an up- and down-hill skier, and reaction path and intermediate states/products in terms of an air trip, including transfers, and the assembly of a model aircraft or car in stages. In all cases, he used enriched analogies. Then, he talked briefly about physical equilibrium in terms of balancing on a see-saw and being alternately mentally sane and insane. Here, he used only simple analogies. (c) The ‘school dance’ analogy (elaborated version). In the second part of the double lesson, Neil elaborated the ‘school dance’ story by making a distinction between the larger hall and the smaller commitment room (see pre-lesson interview). The following discussion then took place. Neil

Let’s go to the commitment room. Now, we have a slight problem. We have 1000 students in the gym.... I may have to stand at the door as the teacher supervising this dance, and count. I know the commitment room can take 500 people, 500 is 250 couples.... The 251st couple comes to the door…. I make an announcement: “if anyone in the commitment room has

358

St. Neil

St. Neil

St. Neil

USING MULTIPLE ANALOGIES found someone by chance that they don’t particularly want to spend the rest of their life with, one couple may go out into the gym and search again”. One couple? One couple ... the other couple, they come into the commitment room because I can allow 250 couples. So, someone else comes to the door, and I say: “anyone else picked unwisely, fine you can leave. You’re off the hook, someone else can come in”. Now, I think you can see how this can happen simultaneously .... Now, for this to work, what are my conditions? You can only have so many in the room. Yeah, I can’t have people coming in and out of the gym, the gym is sealed ... I’ve got to have my system closed from the world…. What is the rate of people per minute committing as opposed to people uncommitting? It’s the same. Another thing that defines equilibrium is not only, it’s a closed system, but the rate of the forward reaction equals the rate of the reverse reaction ... Table 2. The ‘school dance’ analog for chemical reaction, reaction rate, and chemical equilibrium conditions

Analog (familiar situation) School dance (shorter version)

Target (science concept) Chemical reaction

Dancing students in the gym hall

Moving and colliding particles, reactants

Couples in the side (‘commitment’) room

Chemical bond, products

School dance (shorter version)

Reaction rate and influencing factors

Rate of boys colliding with girls

Reaction rate

Number of students in the gym hall

Concentration effect on reaction rate

Speed of dancing students

Temperature effect on reaction rate

Extent of body contact

Surface area effect on reaction rate

School dance (elaborated version)

Chemical equilibrium conditions

Couples going in and out of the side room

Simultaneous forward and reverse reaction

Couples going in and out at the same time

Rate forward reaction = rate reverse reaction

Gym hall doors are sealed

Reaction system is closed

This episode shows that Neil highlighted three chemical equilibrium conditions: (a) forward and reverse reaction run simultaneously, (b) with the same rate, (c) in a closed system (see Table 2). He addressed the grounds for the similarities, but he did not indicate the analogy’s limitations. Again, lower achieving students were not asked to show their understanding by rephrasing the analogies in their own words

A. HARRISON, O. DE JONG

359

Post-lesson interview. When asked to comment on what he thought went well and satisfied his expectations, Neil stated that he was pleased with the students’ recollection of the ‘school dance’ in the shorter version, and he was also happy with the students’ recognition that reversible reactions can occur simultaneously with the same rate. However, he spontaneously added that he had the impression that he too quickly dealt with the dynamic aspect and, for that reason, felt that he had not explicated this feature sufficiently. As a result, in the next lesson he planned to spend more time talking about dynamic aspects, probably by focusing first on physical equilibrium in a solution or in gases. Neil stated that he thinks his students can understand a physical process better than a chemical one because they are much more familiar with them. He also was keen to access his students’ prior knowledge, especially their everyday experiences. He believed that some students were not able to understand the chemical nature of reversible reaction in the first instance, but as soon as they thought about the commitment room, the concept of reversible reactions made sense to them. For that reason, analogies are very important to him. Neil chose the ‘school dance’ analogy because he sees it as a “working analogy”. Chemistry Lesson 3 (single lesson) Pre-lesson interview. In the reflective interview on lessons 1 and 2, Neil had already suggested introducing the features of physical equilibrium. Looking forward to lesson 3, he wanted to first recapitulate the three conditions for chemical equilibrium and then apply them to physical equilibrium. He declared his intention to examine the case of excess sugar in a cup of tea. Neil also intended to use the ‘busy highway’ analogy, where the rate of cars entering the highway equals the rate of cars leaving the highway. He said that he frequently uses this analogy when teaching the topic of equilibrium between evaporation and condensation. He enjoys this kind of analogy because the students are just earning their driver’s licenses. He felt that his students were familiar with the difficulties of merging into and out of traffic. He once used a similar analogy, the ‘peak hour train’ analogy, but found that this analogy did not work because few of his students used the public transport system. The analogy was relevant to him, but not to his students, so he changed the analogy. (a) The ‘excess sugar in a cup of tea’ analogy. Neil began the lesson by quizzing students verbally on the conditions for chemical equilibrium. In order, students indicated that the system must be closed, the rate of the forward reaction is the same as the reverse reaction, and the processes are dynamic. At this point, a student asked Neil to elaborate what was meant by ‘dynamic’. Neil immediately launched the story of ‘excess sugar in a cup of tea’, but at the end, he also admitted to his students an important limitation for understanding the analogy: the dynamic process of sugar simultaneously solidifying and dissolving at the same rate cannot be observed. (b) ‘Simmering curry in a pot’ analogy. Neil was concluding the ‘excess sugar in the cup of tea’ episode, when a student offered a variation on the analogy by asking:

360 St.

Neil

USING MULTIPLE ANALOGIES Is that happening when you’ve got like food in a pot and you’ve got a lid on, and when some evaporates at the same time some is condensing and dropping down at the same time? Ok, very good!

Neil accepted this analogy, replaced the term ‘food’ by simmering curry, and indicated that the proposed analogy’s feature of the rate of water evaporation equals the rate of water condensing is more easily related to daily experiences because there are more observable cues than in his ‘excess sugar in a cup of tea’ analogy. (c) Teaching the ‘busy highway’ analogy. After these two analogies, discussion about dynamic equilibrium continued. During the discussion, Neil introduced a new analogy: Neil St. Neil

St. Neil

St. Neil

St. Neil St. Neil

Who’s got their licenses? We’ve got our Ls [learner driving license]. If … your street runs into a major road [Neil draws on the board] so here you are, at the STOP sign lined up ready to do a right hand turn into traffic. Now … at 5[am], there’s a couple of early birds getting out to start jobs … At 8 o’clock in the morning, and this is the major road into the city? Packed. Bumper to bumper … Let’s now go to the really bizarre situation … you don’t obey the STOP sign, you don’t look for traffic, you simply go out. Right in peak hour what would happen to you? You’d hit a car … You’d cause complete traffic chaos, right? Now, I bring up that example, because it demonstrates to some degree, why vapor pressure happens. Why we get condensation happening after evaporation … suddenly a molecule evaporates, it hits another molecule. Now that molecule may even go back [into liquid], that can happen. Or it causes this molecule to hit that molecule to hit that one, to hit that one to knock it back. The very dynamic situation comes to a point where there’s just too much traffic on the road, for another car to get in [8am]. So for every car that goes on the road, another car has to come off. I suppose you could always think of a car park. If this was a car park I can come in but there comes a point where the car park is totally full of cars. Well that happens when like you get water condensing, a big drop, drops down. If you want to take my analogy, there might be little car parks on the side of the road with some cars parking for a while until it gets really full until a truck comes, loads them on and takes them away to be dropped. But it wouldn’t go back on the road. No, it‘d have to go … Down the service station I should have actually drawn it out here, taken off back to where they came, and just remember: analogies have their limitations.

A. HARRISON, O. DE JONG

361

The ‘busy highway’ analogy is now breaking down, but Neil did not discuss why the analogy broke down. Post-lesson interview. When asked about the analogies used in the lesson, Neil did not comment on the ‘excess sugar in the tea’ story, probably because that went to plan (see pre-lesson interview). However, he noted that halfway through the ‘busy highway’ analogy, he actually thought that a ‘car-park’ analog might be even better. He said that by describing the highway situation at 4am, he would be able to get into the car-park situation. He thought it might get the message across. When asked, “explain why you were thinking about it”, he replied that he was aware that his students can see when a car-park gets full and they cannot go in. They have to wait for a car to come out before another can go in. He thought it was more structured. Later he remarked that he would not change direction in the middle of an explanation. When asked if he used ad hoc analogies, he said he did, but sometimes had “fallen flat on my face”. 5. DISCUSSION AND CONCLUSIONS Neil used several categories of analogies. First, by introducing and elaborating the ‘school dance’, he provided a set of multiple enriched analogies between one general analog (‘school dance’) and different targets (chemical reaction, reaction rate and influencing factors, chemical equilibrium). Second, Neil also used multiple enriched analogies between different analogs (‘sugar in teacup’, ‘busy highway’, and ‘curry in a pot’) and the same target (dynamic nature of equilibrium). He repeated this strategy when linking the analogs of ‘air trip’ and ‘assembling an aircraft’ to the target of reaction paths and intermediates. Third, Neil also used multiple simple analogies, namely, when relating two different analogs (‘see-saw’ and ‘sane/insane’) to one target (physical equilibrium). Finally, he linked one particular analog (‘up- and down-hill skier’) with one target (activation energy) in one enriched analogy. By using this range of analogies, Neil showed a high level of attention to student interest and their conceptual difficulties. Neil’s stories were carefully rehearsed, and he made conscious efforts to marry difficult concepts to everyday stories that were familiar to most of his students. Neil’s working analogy is evidence of his conscious need to hold student interest, and, as the pre- and post-interviews show, he had a clear vision of the three equilibrium conditions that the students needed to understand. But Neil was not finished yet: in the third lesson he grasped the opportunity to develop the ‘excess sugar in tea’ analogy when asked what he means by ‘dynamic’. The ‘simmering curry in a pot’ suggestion received a similar response. Everything seems so natural, but Neil’s planning shows that his marriage of content and pedagogy was planned and honed over many years. In conclusion, we claim that this is an instance of pedagogical content knowledge (PCK) in action; or, as Cochran, deRuiter, and King (1993) better call it, pedagogical content knowing. He knew what he wanted to explain and had the tools to explain the concepts.

362

USING MULTIPLE ANALOGIES

When Neil introduced his analogies, he did not always tell the students where he was going, as in the cases of the elaborated ‘school dance’ and the ‘busy highway’. This strategy reflected his pedagogical intention of “keeping them wondering” and accords with his teaching intentions. From the lesson observations, it can be concluded that his approach motivated his students. Nevertheless, we want to comment on two aspects of this strategy. First, we wonder if this strategy is also effective when teaching less involved concepts or lower achieving students? Second, by introducing analogies ‘out of the blue’, where the target remains ‘hidden’ for some time, we wonder if students are able to construct links between the analog and the target by themselves? In sum, the delayed explanation of links between the analog and the target may have the advantage of evoking curiosity among students, but it also may disadvantage students who cannot map the analogy by themselves. After the final lesson, Neil reflected on the use of the ‘busy highway’ story by stating that halfway through the analogy, he thought that a ‘car-park’ analog might be better. His reflection that students can see when a car-park gets full, and his conclusion that the car-park analog is more structured is indicative of his continuous search for the ‘most effective story’. He did not change his explanation midstream because he knew that this would confuse some students. Neil’s actions reinforce the recommendation that teachers should avoid ad hoc analogies in favor of carefully prepared or tested analogies (cf. Treagust et al., 1998). If analogies are used to promote understandings, the shared and unshared attributes must be carefully mapped with students (Duit, 1991). In the pre-lesson interview, Neil declared that all analogies break down somewhere, yet he infrequently mentioned or discussed the limitation of his analogies with his students. Another disparity was also evident. Neil intended inviting lower achieving students to rephrase the analogies in their own words, but did not use this strategy. In general, disparities between teachers’ intended and actual behavior are not uncommon and are often related to the pressure of unexpected questions and management situations that arise in class (Treagust et al., 1998). However, Neil was an expert chemistry teacher, and we did not observe any pressures or management problems. Also, in the post-lesson interviews, Neil did not raise or comment on either disparity. Perhaps Neil was not sufficiently convinced of the importance of these intentions, to put them into practice and to reflect on them. In conclusion, these findings suggest a renewed need to pay more attention to negotiating the limitations of analogies with students, and inviting lower achievers to retell analogies in their own words. Finally, the interviews and the observed classroom discussions provided an opportunity to examine a teacher’s preparations, actions, and reflections with respect to the use of multiple analogies when teaching chemical equilibrium. Additional research into student understandings and the use of teacher presented analogies will be helpful in enhancing the use of multiple analogies in the chemistry classroom.

A. HARRISON, O. DE JONG

363

REFERENCES Cochran, K., deRuiter, J., & King, R. (1993). Pedagogical content knowing: an integrative model for teacher preparation. Journal of Teacher Education, 44, 263-272. Curtis, R.V., & Reigeluth, C.M. (1984). The use of analogies in written text. Instructional Science, 13, 99-117. Dagher, Z.R. (1995). Analysis of analogies used by teachers. Journal of Research in Science Education, 32, 259-270. De Jong, O., Veal, W. & Van Driel, J.H. (2002). Exploring chemistry teachers’ knowledge base. In J.K. Gilbert, O. De Jong, R. Justi, D.F. Treagust & J.H. Van Driel (Eds.), Chemical Education: Towards Research-based Practice (pp. 369-390). Dordrecht NL: Kluwer Academic Publishers. Duit, R. (1991). On the role of analogies and metaphors in learning science. Science Education, 75, 649-672. Dupin, J.J., & Johsua, S. (1989). Analogies and 'modeling analogies' in teaching: some examples in basic electricity. Science Education, 73, 207-224. Gabel, D.L., & Samuel, K. V. 91986). High school students’ ability to solve molarity problems and their analog counterparts. Journal of Research in Science Education, 23, 165-176. Gentner, D. (1983). Structure mapping; a theoretical framework for analogy. Cognitive Science, 7, 155-170. Glynn, S.M. (1989). The teaching with analogies model: explaining concepts in expository texts. In K. D. Muth (Ed.), Children’s Comprehension of Narrative and Expository texts: Research into Practice (pp. 185-204). Neward, DE: International Reading Association. Harrison, A.G., & de Jong, O. (in revision). Exploring the use of multiple analogical models when teaching and learning chemical equilibrium. Journal of Research in Science Teaching. Harrison A.G., & Treagust, D.F. (1993). Teaching with analogies: A case study in grade 10 optics. Journal of Research in Science Teaching, 30, 1291-1307. Harrison, A.G., & Treagust, D.F. (2000). Learning about atoms, molecules and chemical bonds: a case-study of multiple model use in grade-11 chemistry. Science Education.84, 352-381. Janesick, V.J. (1994). The dance of qualitative research design. In N.K. Denzin & Y.S. Lincoln (Eds.), Handbook of Qualitative Research Design (pp.209-219). Thousand Oaks, CA: Sage. Jarman, R. (1996). Students teachers' use of analogies in science instruction. International Journal of Science Education, 18, 869-880. Justi, R. (2002). Teaching and learning chemical kinetics. In J.K. Gilbert, O. De Jong, R. Justi, D.F. Treagust & J.H. Van Driel (Eds.), Chemical Education: Towards Researchbased Practice (pp. 293-315). Dordrecht NL: Kluwer Academic Publishers. Smith, J.A. (1995). Semi-structured interviewing and qualitative analysis. In J. A. Smith, R. Harre & L. Van Langenhove (Eds.), Rethinking Methods in Psychology (pp. 9-26). Thousand Oaks, CA: Sage. Thagard, P. (1992). Analogy, explanation, and education. Journal of Research in Science Teaching, 29, 537-544. Thiele, R.B., & Treagust, D.F. (1994). An interpretive examination of high school chemistry teachers' analogical explanations. Journal of Research in Science Teaching, 31, 227-242.

364

USING MULTIPLE ANALOGIES

Treagust, D.F., Duit, R., Joslin, P., & Lindauer, I. (1992). Science teachers' use of analogies: observations from classroom practice. International Journal of Science Education, 14, 413-422. Treagust, D.F., Harrison, A.G., & Venville, G. (1998). Teaching science effectively with analogies: An approach for pre-service and in-service teacher education. Journal of Science Teacher Education. 9, 85-101. Van Driel, J.H., & Graber, W. (2002). The teaching and learning of chemical equilibrium. In J.K. Gilbert, O. De Jong, R. Justi, D.F. Treagust & J.H. Van Driel (Eds.), Chemical Education: Towards Research-based Practice (pp. 271-293). Dordrecht NL: Kluwer Academic Publishers.

PART 7 Discourse and argumentation in science education

THE ROLE OF ARGUMENT IN SCIENCE EDUCATION

JONATHAN OSBORNE University of London, UK

ABSTRACT This paper makes the case for argument in science education drawing on a range of research efforts in the field. The specific research reported here took place over two years between 1999 and 2001 in junior high schools in the greater London area. The research was conducted in two phases. In phase 1, working with a group of 12 science teachers, the main emphasis was to develop sets of materials and strategies to support argumentation in the classroom, and to support and assess teachers’ development with teaching argumentation. In phase 2 of the project, the focus of this paper, teachers taught the experimental groups a minimum of nine lessons that involved socio-scientific or scientific argumentation. In addition, these teachers taught similar lessons to a comparison group at the beginning and end of the year. The focus of this research was to assess the progression in student capabilities with argumentation. For this purpose, data were collected from 33 lessons by videotaping two groups of four students in each class engaging in argumentation. Using a framework for evaluating the nature of the discourse and its quality developed from Toulmin’s argument pattern, the findings show that there was an improvement in the quality of students’ argumentation.

1. INTRODUCTION This paper, a summary of the ideas presented in a keynote address at the ESERA conference1, addresses three questions: • • •

Why does argument matter in science education? What do we know about how it can be taught? How can we assist teachers to develop their practice? 2. WHY DOES ARGUMENT MATTER IN SCIENCE EDUCATION?

In answering the first of these questions, I and my co-workers, Shirley Simon of the Institute of Education and Sibel Erduran from King’s College, have a particular conception of scientific literacy which is associated with two particular features that we believe to be missing from the landscape of contemporary science education. These are (a) opportunities to inspect, and engage with, the arguments that lead to the construction of scientific explanations, and (b) opportunities to evaluate 1

To capture some of the essence of the talk, the style is, therefore, somewhat colloquial.

367 K. Boersma et al. (eds.), Research and the Quality of Science Education, 367—380. © 2005 Springer. Printed in the Netherlands.

368

THE ROLE OF ARGUMENT IN SCIENCE EDUCATION

critically plural alternatives. In short, what we subscribe to is a critical notion of scientific literacy. Embedded in that view is a premise that: …‘to know science’ is a statement that one knows not only what a phenomenon is, but also how it relates to other events, why it is important and how this particular view of the world came to be (Osborne, 2000).

Put simply, too much of science education presents a set of disembodied information masquerading as knowledge. Information is characterised by the fact that it is second hand consisting of other people’s interpretation of experience. It may be about people, events, objects, and may deal with the abstract and theoretical concepts in science (Wells, 1999). Whether it can be remembered depends on the extent to which it can be infused by the receiver with meaning and assimilated into the individual's existing body of knowledge. Addicted, as many are, to simplistic conceptions in which education is seen as a process of information transmission (Reddy, 1979), my view is that that we are failing to offer students the opportunity to truly know science. Rather, it is as if the grand picture of science education were presented as a 1000 piece jigsaw, forgetting that its complexity ensures that most do not stay the course to ever see the whole picture. Knowledge, in contrast, is something which is constructed through a process of justifying beliefs through reasoning, conjecturing, evaluating evidence, and considering counter-arguments. It is these processes, as Wells (1999) would argue, that ‘constitute the activity of knowing’. That science teachers may lack aspects of such knowledge is evidenced by the difficulty they have when confronted with instances of the epistemic question – how do you know, for instance, that the following fundamental tenets and beliefs developed by school science are true? • • • • •

Day and Night are caused by a spinning Earth We live at the bottom of a sea of air We look like our parents because every cell carries a coded chemical blueprint of ourselves Plants take in carbon dioxide from the air All matter is made of atoms

My informal research would suggest that less than 10% of teachers of science are able to provide one of the two crucial pieces of evidence that demonstrate that the Earth spins. This is not so much a failure on the part of these teachers, but a failure on the part of their own education. Yet appreciation of the strange vision that science offers of the material world is not self-evident; science requires that what we offer students, at least some of the time, is the opportunity to engage in the process of scientific argument – to consider evidence, to appreciate that not all evidence is equally significant, and to evaluate arguments and propose counter-arguments. Lest we forget, there are four potential learning goals to any science lesson (Duschl & Erduran, 1996). Goals may be conceptual in nature in an attempt to

J. OSBORNE

369

achieve the understanding of some scientific idea; they may be cognitive to develop students’ ability to reason; they may be epistemic to examine the reasons for justified belief; and they may be social to develop students’ ability to work collaboratively or to enhance their motivation to learn. Sadly, in too many science lesson, it is the first of these – the conceptual – which predominates. Such an emphasis is not so much a consequence of teachers’ choice but more a product of an assessment system which prioritises ontology at the expense of epistemology. Engaging in argumentation is, then, one means of restoring a balanced set of learning goals to the learning of science. 3. WHAT DO WE KNOW ABOUT HOW IT CAN BE TAUGHT? We turn then to the second question, what do we know about how argumentation can be taught? Fortunately, there is now a body of emerging ideas which offer windows into appropriate forms of practice. There are, for instance, the wealth of Concept Cartoons produced by Brenda Keogh and Stuart Naylor (2000). Such simple but effective ideas lay before students alternative explanations of physical phenomena. The cartoons naturally engage students in the process of conjecturing, reasoning, and counter-argument, which are the substance of knowledge construction. In addition, there are materials that offer pragmatic examples of how the nature of science might be approached in the classroom (Lederman & Abd-el-Khalick, 1998; Solomon, 1991). One interesting example invites students to conjecture about the possible events that may have led to the discovery of these tracks on an archaeological dig (Figure 1). Students are first shown Figure 1a and then invited to conjecture about what may have been the cause of such tracks.

Fig 1.Tracks exposed from an archeological dig This activity also raises the issue of inference and how theories go beyond the data. Progressively, as more information is revealed to the student (Figures 1b and 1c), theory revision and counter arguments are demanded of the student, and, in

370

THE ROLE OF ARGUMENT IN SCIENCE EDUCATION

addition, the consideration that not all science is a hypothetico-deductive interpretation from experimental data. There are also a body of materials that we have developed in our own work on argumentation (Osborne, Erduran & Simon, 2004; J.F. Osborne et al. 2001). In one activity, for instance, students are offered two possible graphical records of the variation in temperature with time as ice is heated to form steam. One of these is the correct graph; the other is the common misconception. Students are provided with a set of key points which they must use to construct an argument about what they believe is scientifically correct. In this task, our data show students engaging constructively in debate about which representation is correct and discussing the problem in order to generate a common understanding. A crucial point emerging from the study of argumentation in these contexts is that students need a resource for arguing, or to put it more colloquially – ‘you can’t bootstrap if you don’t have boots’.2 For in the context of science, critical evaluation of evidence can only take place if there is evidence to evaluate and, lacking such evidence, students will find it difficult to either generate an argument or consider a counter-argument. A more extended rationale for including argumentation in school science has been developed elsewhere (Driver, Newton & Osborne, 2000; Newton, Driver & Osborne, 1999). Those arguments led to the development of a research project, funded by the UK Economic and Social Research Council, which had three objectives. These were: (i)

identify the pedagogical strategies necessary to promote ‘argument’ skills in young people in science lessons; (ii) trial the pedagogical strategies and determine the extent to which their implementation enhances teachers’ pedagogic practice with ‘argument’; (iii) determine the extent to which lessons that follow these pedagogical strategies lead to enhanced quality in pupils’ arguments. This paper primarily considers the data and results which answer the third objective. Suffice to say that in answering the first objective, our work led to the development of nine generic frameworks that would support argumentation in the classroom. The one feature that all these frameworks share is the presentation of plural alternatives requiring students to consider and evaluate the evidence and arguments for each. Scientific literacy depends as much on the ability to refute and recognise poor scientific arguments as it does on the ability to reproduce the correct scientific view. Moreover, research would suggest that knowing why the wrong answer is wrong is as important to conceptual understanding as knowing why the right answer is right (Alverman, Qian & Hynd, 1995; Hynd & Alvermann, 1986). As for the second objective, we worked with teachers exploring their ideas about argumentation and supporting their efforts to improve their practice over the 2

Apparently used first by the current Governor of California in one of his election speeches.

J. OSBORNE

371

period of a year. The data that we have obtained of the teachers' use of argument show that, for more than half of the teachers, there was a significant improvement in their practice, with argumentation in the classroom. Theoretical Background Our conception of argument draws on the ideas of Stephen Toulmin’s work The Uses of Argument (Toulmin, 1958). Toulmin’s achievement was to release the study of argument from the narrow confines of philosophy with its singular vision of the deductive, inductive, or adductive nature of argument, to propose a more generic form of argument which is commonly used in everyday life. Toulmin saw argument as a rational activity in which claims were advanced and supported by data related by warrants to the principal claim. Such claims often had qualifiers and the warrants were underpinned by theoretical presuppositions or backings which were often never mentioned. Moreover, dialogic activity relied on the ability to advance rebuttals of another’s argument. Put simply in the words of the great French philosopher, Gaston Bachelard (1940), 'Two people must first contradict each other if they really wish to understand each other. Truth is the child of argument, not of fond affinity.' Using such a model, we have sought to identify, in students’ discourse, their claims, data, and warrants which form the elements of their argument. Essentially our question of interest has been, what is the quality of the argumentation accepting that it is possible to have a complex argument which may rest on fallacious data or warrants. Our approach to answering this question has been to classify arguments into three groups: • Simple claims • Arguments with justifications • Arguments with justification and rebuttals. Whilst some would argue that claims without justification are essentially nonarguments, we believe that they are the first stage of engaging in argumentation and establishing difference. Arguments with justifications, i.e. data and warrants, are superior as they recognise that beliefs must have warrants or evidence. Episodes with rebuttals are, however, of better quality as they force evaluation of the quality of the substance of an argument. Moreover, as Kuhn (1991) argues, the ability to use rebuttals is ‘the most complex skill’ as an individual must ‘integrate an original and an alternative theory, arguing that the original theory is more correct.’ Thus, rebuttals are an essential element of arguments of better quality and demonstrate a higher-level capability with argumentation. In our work, we found three types of arguments with rebuttals: arguments with weak rebuttals where the counterargument was not self-evident; arguments with clear rebuttals; and arguments with multiple rebuttals. This then led us to propose a framework for the quality of argument in terms of a set of five levels of argumentation (Table 1) as follows:

372

THE ROLE OF ARGUMENT IN SCIENCE EDUCATION

Table 1: Analytical Framework used for assessing the quality of argumentation

Level 1:

Level 1 argumentation consists of arguments that are a simple claim v a counter claim or a claim v claim

Level 2:

Level 2 argumentation has arguments consisting of claims with either data, warrants, or backings but do not contain any rebuttals.

Level 3:

Level 3 argumentation has arguments with a series of claims or counter claims with either data, warrants or backings with the occasional weak rebuttal.

Level 4:

Level 4 argumentation shows arguments with a claim with a clearly identifiable rebuttal. Such an argument may have several claims and counter claims as well but this is not necessary.

Level 5:

Level 5 argumentation displays an extended argument with more than one rebuttal.

The first phase of our work enabled us to identify six teachers who had developed their practice and ability with argumentation and to: a)

video and audio record a lesson of a science-based argument activity with the same class and, wherever possible, the same groups of children age 12. b) video and audio record a second lesson of a socio-scientific argument with the same class and, wherever possible, the same groups of children age 12. c) video and audio record the same teacher teaching the socio-scientific lesson to a similar group so students for comparison purposes. All of these audio records of the lessons were transcribed and then analysed using the framework for evaluating the quality of argumentation that we had developed. First, we looked at the transcripts to examine the nature of the discourse that was a feature of these lessons. The results are shown in Table 2. The rows in this table shows the type of discourse as a percentage of the whole lesson – that is how much consists of claims advanced by students, ground to support those claims, and non-argumentative discourse and how much of it is contributed by the teacher. The columns show the breakdown by type of lesson.

J. OSBORNE

373

Table 2: Percentages of group discourse of an argumentative nature

LESSON

Zoo Zoo Science Science Leisure 1 2 Centre Lesson Control

Leisure Centre Control

%

%

%

%

%

%

4

3

4

5

4

3

Grounds

28

24

11

12

22

26

NonArgument

8

8

15

9

13

10

Teacher

59

64

71

75

61

61

Type of Claims Discourse

The major feature of the transcripts is, perhaps inevitably, the dominance of the discourse by the teacher. The next major feature is that this dominance occurs more in the science lessons than it does in the either of the socio-scientific lessons. This should be compared with the findings from earlier work and observation of the discourse in science lessons (Newton et al., 1999) which found that deliberative discourse formed less than 5% of the classroom interactions. Hence, these data would suggest that engaging in these kinds of argumentation activities, significantly changes the nature of the discourse in science classrooms, opening up the space for dialogic discussion. In scanning the transcripts we identified 135 oppositional episodes in 43 discussion groups in 23 lessons. Analysing these produced the distribution of levels of argumentation shown in Figure 2. The first positive feature to emerge from this data is that looking broadly at the pattern, there has been an improvement, although it is not significant. Our view is that developing such skills is a long term process. Skills at argumentation are habits of mind acquired through experience and repetitive practice. If our hypothesis is correct, empirical evidence to support it would require more extensive research conducted over a longer period of time. The second feature of the data is shown by Figure 3 which compares the levels of argument achieved by these students when arguing in a scientific context at the beginning of the year with that achieved at the end of the year.

374

THE ROLE OF ARGUMENT IN SCIENCE EDUCATION

40% Zoo+ Sci lesson

35% 30%

LC + Sci Lesson

25% 20% 15% 10% 5% 0% 1

2

3

4

5

Argument Level

Figure 2: Levels of argumentation achieved by experimental groups, pre (n=69) and post-intervention (n=66).

50% 40% 30%

Science Beg Science End

20% 10% 0% 1

2

3

4

5

Levels of Argum ent

Figure 3: A comparison of levels of argumentation achieved by experimental groups in a scientific context at the beginning of the year (n=31) with that achieved at the end of the year (n=28).

J. OSBORNE

375

The figure shows that, likewise, that there has been an improvement but that the mean of the distribution is more heavily weighted towards the lower levels of argumentation. This feature becomes clearer when we compare directly the levels achieved in all the argumentation lessons in a socio-scientific context with the levels achieved in a scientific context (Figure 4). 40% 35% 30% 25% Zoo Exp & LC Expt

20%

Science Lessons

15% 10% 5% 0% 1

2

3

4

5

Levels of Argument

Figure 4: Comparison of levels of argumentation (socio-scientific context) with those achieved in a scientific context These findings suggest that argumentation in a scientific context is more difficult than that in a socio-scientific context. As socio-scientific arguments draw on a range of scientific knowledge, ethics, and values which children of age 12 at least have in some naïve form, this finding is, perhaps, not surprising. In contrast, deciding whether a coat on a snowman will make the snowman melt faster or slower can only be resolved by a specific knowledge of science. Other Research Other significant contributions to work in this domain is the research of Zohar & Nemet (2002) exploring the effect of fostering students' knowledge and argumentation skills through dilemmas in human genetics. Using a similar Toulminbased framework and a somewhat different method of scoring the analysis revealed two major findings concerning biological knowledge: (a) that following instruction the frequency of student references to correct, specific biological knowledge in constructing arguments increased; and (b) that students in the experimental group scored significantly higher than students in the comparison group in a test of genetic knowledge. These results led Zohar and Nemet (ibid.) to conclude ‘that integrating

376

THE ROLE OF ARGUMENT IN SCIENCE EDUCATION

explicit teaching of argumentation into the teaching of dilemmas in human genetics enhances performance in both biological knowledge and3 argumentation.' A second significant contribution is an interesting analysis of our data conducted by Von Aufschnaiter (2004) of the University of Bremen. Using her four level schema for analyzing cognitive operations in terms of a set of entities, such as objects, aspects, events, principles, and connections, she concluded that: 1

Students can only engage in argumentation when they find something known to them in the task (or in the statements offered to them).

2

Students understand tasks/statements offered at their own level of “prior” knowledge (no matter whether this is “right” or “wrong”).

3

Argumentation helps students to improve what they already know.

4

Argumentation does not have a direct impact on students developing new understanding in a sense that the new understanding does not emerge within the argumentation directly (or within scaffolding, see above). But argumentation seems to have a double function: it supports students’ improvement of thinking (by developing an idea more readily or having a different, but well known, aspect offered by another) to scaffold the basis of further learning. In addition, argumentation may help students to discover those aspects they had never thought about. However, as they can only engage at the level of their actual understanding, these aspects need to be very near to their actual knowledge.

Von Aufschnaiter’s work provides some important clues to the way in which argumentation can support and develop conceptual goals as well as cognitive and epistemic ones. A third contribution is another piece of work conducted by Schwarz et al. (2003). This detailed study of the effect of argumentation with 120 fifth grade students found that: …in general, all measures of individual arguments steadily increased along the successive argumentative activities… They were better in the sense that they were gradually less one sided and more compounded. Also, the reasons invoked were more relevant to the standpoint claimed and more acceptable. More reasons supporting alternative arguments were raised. Finally the reasons invoked were more abstract.

3

Emphasis added.

J. OSBORNE

377

In summary, both our own work and that of others point to the view that argumentation has a role not only in developing a better understanding of the epistemic basis of science but also in achieving a better conceptual understanding of science itself. 4. WHY IS IT SO DIFFICULT FOR SCIENCE TEACHERS AND HOW CAN WE ASSIST THEM TO DEVELOP THEIR PRACTICE? Argument and discussion are not normal features of many science classrooms because science education is one of the last authoritarian socio-intellectual discourses still to be found in our schools (Ravetz, 2002). Changing that practice requires a determined endeavour and recognising the messages of research on what constitutes effective professional development. As Guskey and Huberman (1995), have argued: A good deal of what passes for ‘professional development’ in schools is a joke ...radically under-resourced, brief, not sustained, designed for ‘one size fits all’, imposed rather than owned, lacking any intellectual coherence, treated as a special add-on event rather than part of a natural process, and trapped in the constraints of the bureaucratic system... Consequently, we have developed a set of materials that will support the professional development of teachers who wish to improve their skill and expertise with argumentation in the classroom. These materials, the IDEAs materials, have addressed the following aspects in 6 half-day workshops. 1. 2. 3. 4. 5. 6.

An Introduction to Argument Managing Small Group Discussions Teaching Argument Resources for Argumentation Evaluating Argument Modelling Argument

In developing these materials we recognised, following the work of Joyce & Showers (1988), that it was important to have video exemplars of what constitutes effective practice in the teaching of argumentation. Using the teachers who had worked with us on the research project, we have video taped a range of their lessons from which we have extracted key features to illustrate how to introduce argument, how to support small group discussion, and how to model argument to students. With such extracts we have been able to point to such features as the need to structure the activity, the need to introduce and use appropriate language, and the need to encourage counter-argument. Teachers then have a much better model of what they need to do themselves to encourage argumentation in the classroom.

378

THE ROLE OF ARGUMENT IN SCIENCE EDUCATION 5. CONCLUSIONS

The case made here is that argumentation in the classroom is not a bolt-on extra but a core feature of the school science curriculum. Additional evidence to support this position comes from a recent study of 377 students in 22 junior high school classes by Nolen (2003). She found that classrooms that focused on understanding and independent thinking using discourse-based activities, positively predicted students' self-reported satisfaction with learning. Such activities offer a vehicle for engaging in a form of learning which uses collaborative, horizontal structures where communication is aimed at generating understanding and student engagement. This contrasts strongly with the ‘assembly-line instruction’ model which still permeates our schools where the participation structures are hierarchical, where communication is often reduced to quizzing learners (Rogoff et al., 2003), and where there is significant evidence of negative attitudes towards science (Osborne, Simon & Collins, 2003). However finally, and perhaps more fundamentally, we should not forget that the rationality of science and of our contemporary culture is secured by its commitment to evidence. When we fail to present the evidence for scientific beliefs to children, and when we fail to give them the opportunity to consider and inspect the arguments on which those beliefs rest, science education becomes an oxymoron – a contradiction in terms and self defeating. For, as Horton (1971) so eloquently pointed out, it makes the epistemic basis of belief reside solely in the authority of the source and, as a means of education, is no better than that of the oral transmission of knowledge which was a feature of medieval life and of less-developed societies today. My argument is that 21st Century Science education requires more than a simple reworking of the curriculum content – replacing the blast furnace with examples of nanotechnology, lasers, or information technology. It requires us to ask an additional set of much harder questions which are not only what knowledge should we teach, but also why do we value it so much? Why do we believe it to be true, and why was it so hard won? That, I would argue, requires placing argument and evidence to the fore. An education for scientific literacy must see the fostering of rationality as its core value. Put simply, it is not for a set of narrow professional concerns that science education matters; rather, it is that science and argument matter to education in general as they are the foundation of rational thought and the critical spirit that we must seek to impart to our students. REFERENCES Alverman, D. E., Qian, G. & Hynd, C. E. (1995). Effects of interactive discussion and text type on learning counterintuitive science concepts. Journal of Educational Research, 88, 146-154. Bachelard, G. (1940). The Philosophy of No. Paris: Paris University Press. Driver, R., Newton, P. & Osborne, J. (2000). Establishing the norms of scientific argumentation in classrooms. Science Education, 84(3), 287-312.

J. OSBORNE

379

Duschl, R. A. & Erduran, S. (1996). Modelling the Growth of Scientific Knowledge. Paper presented at the First European Conference on Science Education, Leeds, UK. Guskey, T. R. & Huberman, M. (1995). Professional Development in Education: New Paradigms and Practises. New York: Teachers' College Press. Horton, R. (1971). African Traditional Thought and Western Science. In M. D. Young (Ed.), Knowledge and Control (pp. 208-266). London: Colin-MacMillan. Hynd, C. & Alvermann, D. E. (1986). The Role of Refutation Text in Overcoming Difficulty with Science Concepts. Journal of Reading, 29(5), 440-446. Joyce, B. & Showers, B. (1988). Student Achievement Through Staff Development. White Plains, NY: Longman. Kuhn, D. (1991). The Skills of Argument. Cambridge: Cambridge University Press. Lederman, N. & Abd-el-Khalick, F. (1998). Avoiding the De-natured Science: Activities that promote understandings of the Nature of Science. In W. F. McComas (Ed.), The Nature of Science in Science Education: Rationales and Strategies (pp. 83-126). Dordrecht: Kluwer. Naylor, S. & Keogh, B. (2000). Concept Cartoons in Education. Sandbach: Millgate House Publishers. Newton, P., Driver, R. & Osborne, J. (1999). The Place of Argumentation in the Pedagogy of School Science. International Journal of Science Education, 21(5), 553-576. Nolen, S. B. (2003). Learning Environment, Motivation and Achievement in High School Science. Journal of Research in Science Teaching (40), 4. Osborne, J. F. (2000). Science for Citizenship. In M. Monk & J. F. Osborne (Eds.), Good Practice in Science Teaching: What Research Has to Say (pp. 225-240). Buckingham: Open University Press. Osborne, J. F., Erduran, S. & Simon, S. (2004). The IDEAS Project. London: King's College London. Osborne, J. F., Erduran, S., Simon, S. & Monk, M. (2001). Enhancing the Quality of Argument in School Science. School Science Review, 82(301), 63-70. Osborne, J. F., Simon, S. & Collins, S. (2003). Attitudes towards Science: A Review of the Literature and its Implications. International Journal of Science Education, 25(9), 1049– 1079. Ravetz, J. (2002). Reflections on the new tasks for science education. Unpublished Evidence submitted to the House of Commons Committee for Science and Technology. Reddy, M. (1979). The conduit metaphor. In A. Ortony (Ed.), Metaphor and Thought. New York: Cambridge University Press. Rogoff, B., Paradise, R., Mejía Arauz, R., Correa-Chávez, M. & Angelillo, C. (2003). Firsthand Learning Through Intent Participation. Annu. Rev. Psychol., 54, 175-203. Schwarz, B. B., Neuman, Y., Gil, J. & Ilya, M. (2003). Construction of Collective and Individual Knowledge in Argumentative Activity. Journal of the Learning Sciences, 12(2), 219-256. Solomon, J. (1991). Exploring the Nature of Science: Key Stage 3. Glasgow: Blackie. Toulmin, S. (1958). The Uses of Argument. Cambridge: Cambridge University Press. Von Aufschnaiter, C. (2004). Argumentation and Cognitive Processes in Science Education. Paper presented at the Annual Conference of the National Association for Research in Science Teaching, Vancouver. Wells, G. (1999). Dialogic Inquiry: Towards a sociocultural theory and practice of education. New York: Cambridge University Press.

380

THE ROLE OF ARGUMENT IN SCIENCE EDUCATION

Zohar, A. & Nemet, F. (2002). Fostering Students' Knowledge and Argumentation Skills Through Dilemmas in Human Genetics. Journal of Research in Science Teaching, 39(1), 35-62.

THE ROLE OF ARGUMENTATION IN DEVELOPING SCIENTIFIC LITERACY SIBEL ERDURAN¹, JONATHAN OSBORNE², SHIRLEY SIMON³ ¹University of Delaware, USA ²King’s College, University of London, UK ³Institute of Education, University of London, UK

ABSTRACT Recent approaches in educational research frame science learning in terms of the appropriation of discourse practices where argumentation plays a central role in the development of explanations and theories. The main objectives of the research reported in this paper were to (1) investigate the pedagogical strategies necessary to promote argumentation skills in students; (2) determine the extent to which the implementation of such strategies enhances teachers’ pedagogical practices with argumentation; and (3) examine the extent to which lessons which follow these pedagogical strategies lead to enhanced quality in students’ argumentation. Data collected from a set of lessons on scientific and socioscientific topics from twelve, year 8 schools in London are reported and discussed. These lessons were analysed using a framework based on Toulmin’s Argument Pattern. There were statistically significant differences in the quality of arguments generated in the classrooms of the project teachers who had participated in the training workshops. The strategies that we have adopted for working with teachers, and the frameworks to support argumentation will be discussed.

1. INTRODUCTION Decades after Joseph Schwab’s argument that science should be taught as an ‘enquiry into enquiry’ and almost a century since John Dewey advocated classroom learning as a student-centred process of enquiry, school science is still struggling to achieve such aspects of scientific literacy in the classroom. Take, for instance, the publication of the AAAS edited volume on inquiry (Minstrell & Van Zee, 2000), the recent release of Inquiry and the National Science Education Standards (National Research Council, 2000), and the inclusion of ‘scientific enquiry’ as a separate strand in the English and Welsh science national curriculum (Department for Education and Employment, 1999). These three works serve as signposts to an ideological commitment that teaching science needs to accomplish much more than simply detailing what we know. Of growing importance is the need to educate our pupils and citizens about how we know and why we believe, for example, science as a way of knowing (Driver, Leach, Millar & Scott, 1996; Erduran, 2001; Erduran & Osborne, in press). The shift requires a focus on (1) how evidence is used in science for the construction of explanations – that is, on the arguments that form the links between data and the theories that science has constructed; and (2), the development

381 K. Boersma et al. (eds.), Research and the Quality of Science Education, 381—394. © 2005 Springer. Printed in the Netherlands.

382

THE ROLE OF ARGUMENTATION

of criteria used in science to evaluate the selection of evidence and the construction of explanations. While consideration of the important roles language, conversation, and discussion have in science learning can be traced back several decades (Bruner, 1964; Lansdown, Blackwood & Brandwein, 1971), it was not until the 1980s that serious discussion of the role of language in science learning began (Gee, 1996; Lemke, 1990). More recently, the field has turned its attention to that discourse which addresses argumentation (Driver, Newton & Osborne, 2000). The case made here is that argumentation, i.e., the coordination of evidence and theory to support or refute an explanatory conclusion, model, or prediction (Suppe, 1998), is a critically important epistemic task and discourse process in science. Situating argumentation as a central element in the learning of sciences has two functions: one is as a heuristic to engage learners in the coordination of conceptual and epistemic goals, and the other is to make students’ scientific thinking and reasoning visible to enable formative assessment by teachers. From this perspective, epistemic goals are not additional extraneous aspects of science to be marginalized to single lessons or the periphery of the curriculum. Rather, striving for epistemic goals like developing, evaluating, and revising scientific arguments represents an essential element of any contemporary science education process. An important task for science education, therefore, is to develop children’s ability to understand and practice scientifically valid ways of arguing, and enable them to recognise not only the strengths of scientific argument, but also its limitations. Hence, the research presented in this paper, seeks to study how the teaching and learning of argumentation about scientific issues can be enhanced in science lessons. In so doing, our work builds on previous research into young people’s epistemologies of science (Driver et al, 1996) and the conduct of group discussion in science lessons (Alexopoulou & Driver, 1997; Newton, Driver & Osborne, 1999). 2. THEORETICAL BACKGROUND ON ARGUMENTATION Over the past few decades certain influential educational projects have all laid foundations for the work on argumentation in science lessons. These projects have promoted independent thinking, the importance of discourse in education and the significance of cooperative and collaborative group work (e.g., Cowie & Rudduck, 1990). In addition to these projects, a body of relatively unintegrated research concerning argumentative discourse in science education has begun to emerge (e.g., Boulter & Gilbert, 1995; Means & Voss, 1996; Mason, 1996; Herrenkohl & Guerra, 1998). Perhaps the most significant contribution to this literature has come from Kuhn (e.g., Kuhn, 1991) who explored the basic capacity of individuals to use reasoned argument. Kuhn investigated the responses of children and adults to questions concerning problematic social issues. She concluded that many children and adults (especially the less well educated) are very poor at the coordination of evidence (data) and theory (claim) that is essential to a valid argument. More recent work by Hogan and Maglienti (2001) exploring the differences between the reasoning ability of scientists, students, and non-scientists found, likewise, that the performance of the latter two groups were significantly limited.

S. ERDURAN, J. OSBORNE, S. SIMON

383

A significant problem confronting the development of argumentation in the science classroom is that it is fundamentally a dialogic event carried out among two or more individuals. Scott (1998), in a significant review of the nature of classroom discourse, shows how it lies on a continuum from ‘authoritative’, which is associated with closed questioning and IRE dialogue, to ‘dialogic’, which is associated with extended student contributions and uncertainty. However, the combination of the power relationship that exists between science teacher and student, and the rhetorical project of the science teacher, which seeks to establish the consensually agreed scientific world-view with the student, means that opportunities for dialogic discourse are minimized. Hence, introducing argumentation requires a shift in the normative nature of classroom discourse, and change has required that science teachers be convinced that argumentation is an essential component for the learning of science. In addition, teachers have required a range of pedagogical strategies that will both initiate and support argumentation if they are to adopt and integrate argumentation into the classroom. 3. RESEARCH PROGRAMME We believe that promoting the practice of ‘argument’ in science lessons requires the development of appropriate pedagogical strategies that offer practical guidance for teachers. Furthermore, the benefit of such guidance needs to be assessed through empirical studies. Our research was seeking, therefore to: (i)

identify the pedagogical strategies necessary to promote ‘argument’ skills in young people in science lessons; (ii) trial the pedagogical strategies and determine the extent to which their implementation enhances teachers’ pedagogic practice with ‘argument’; (iii) determine the extent to which lessons which follow these pedagogical strategies lead to enhanced quality in pupils’ arguments. In conducting our research we have worked initially with a group of 12 teachers to explore and develop their practice in initiating argumentation in the classroom, and then in the second year with a subset of 6 teachers to explore what effect such activities had on the classroom discourse and student use of argument. The research has been conducted in essentially two phases. In the first year (Sept. 1999 – Sept. 2000), we sought to focus on developing the skills of the teacher and the materials for use in argument-based lessons. To this end, we have video- and audio-recorded the teacher at the beginning of year 1 and year 2, and systematically analysed these transcripts to evaluate the characteristics of their approach to argumentation, to see if there is an identifiable measure of their progress. We have also taped and transcribed two groups of pupils in each class to develop a schema for evaluating the quality of their argumentation. During that time, the teachers also attended 6 half-day meetings, held at King's College London, to discuss and share pedagogical strategies for teaching such lessons and develop materials, and to develop their understanding of our theoretical perspective on argument. The training materials have subsequently been published with the financial support of the Nuffield Foundation (Osborne,

384

THE ROLE OF ARGUMENTATION

Erduran & Simon, 2004). In the second phase of the project (Sept 2000 – Sept 2001), we have worked with a subset of 6 teachers asking them to repeat the process. Support in this phase was reduced to three half-day meetings across the year and in situ feedback provided whenever a visit was made for the purpose of data collection. In addition, another set of classes, taught by the same teacher, was used as a control. The focus of our analysis in this second stage has been on the recordings and transcripts of the discussions by pupils to see if there was any improvement in the quality or quantity of argument. Following is a summary of the salient findings that have emerged from the work of the project and an exploration of their implications. 4. MATERIALS One of the features of this work has been to try and develop materials that could be used for supporting argumentation in the classroom. The essential precursor to initiating argument is the generation of difference or plural theoretical interpretations. Hence, a common framework for all the materials we have developed has taken the form of presenting competing theories to students for examination and discussion. These have been presented to pupils to read in small groups and then discuss. However, initiating argument also requires a resource or evidence to enable the construction of argument. Hence, commonly, competing theories have been accompanied by evidence which students are asked to use to decide whether the evidence presented supports theory 1, theory 2, both, or neither (an example is shown below). In addition, sessions with the teachers in the first year of work aimed to develop their theoretical understanding of argument and explored how argument could be supported in the classroom through the use of argument prompts. Fuller details can be found in Osborne, Erduran, and Simon (2004) and Osborne, Erduran, Simon, and Monk (2001). Example 1: Competing Theories A Theory 1: Light rays travel from our eyes onto the objects and enable us to see them. Theory 2: Light rays are produced by a source of light and reflect off objects into our eyes so we can see them. Which of the following pieces of evidence supports Theory 1, Theory 2, both or neither. Discuss. a. Light travels in straight lines b. We can still see at night when there is no sun c. Sunglasses are worn to protect our eyes d. If there is no light we cannot see a thing e. We ‘stare at’ people, ‘look daggers’ and ‘catch people’s eye’ f. You have to look at something to see it.

S. ERDURAN, J. OSBORNE, S. SIMON

385

5. DATA SOURCES The data sources were verbal conversations of teachers and students audio-taped in classes of year 8 (age 12-13) students. In year 1, we video-taped two lessons – one at the beginning of the year and one a year later. At this stage of our work the focus was on argumentation in socio-scientific context. Hence, the main task within these lessons was an exploration of arguments for and against the funding of a new zoo. Each lesson had 3 sections. At the onset, the teacher distributed a letter outlining the task, and there was a whole class discussion on the pros and cons of zoos. Then the students were put into groups and asked to come to some consensus about whether or not the zoo should be built. Finally, in the last phase of the lesson, the groups made presentations and shared their opinions with the rest of the class. As homework, students were typically asked to write a letter or compose a poster that would communicate their arguments. Needless to say, there was considerable variation among teachers in the detail of their implementation. The schools chosen for this work were located in the Greater London area and ranged from urban to suburban settings with mixed ethnic groups. Three schools were all-girls schools, one school was private, and 12 schools were public. Teachers wore microphones so as to capture their verbal contributions to the lesson as well as their interactions with students during the group format. In addition, two groups of four pupils were selected and their conversations recorded. In the second year of our work, a subset of six teachers was selected on the basis that they were individuals who were considered to have made more progress in their ability to facilitate and incorporate argumentation in their pedagogical practice. As well as recording the teachers’ second attempt at teaching the zoo lesson to use as a comparison with their first attempt a year previously, this phase sought to examine pupils’ ability to incorporate and use argumentation in two contexts: 1.

2.

a socio-scientific topic to compare the development of the experimental group with a control group using data drawn from the zoo lesson at the beginning of the year, and a lesson about siting a leisure centre in a nature reserve at the end of the year; and a scientific context to compare the development of the experimental group, using data drawn from a lesson at the beginning of the year and the end of the year.

Thus, in addition to the data collected from the lessons exploring arguments for and against the establishment of a new zoo at the beginning of year 2 (6 lessons, 11 groups1), data were collected from the same teachers teaching the same lesson to a control group (5 lessons, 9 groups); and from the same teachers implementing argument in a scientific context (6 lessons, 12 groups). In each of the lessons, wherever possible, a recording was made of each teacher and two selected groups of four pupils. In the intervening period, teachers taught a minimum of 8 lessons using 1

Due to a set of factors, such as changes in teachers timetable and occasional technical problems, a complete data set does not exist for all lessons.

386

THE ROLE OF ARGUMENTATION

arguments in a scientific context. Because of the contingent nature of individual schemes of work and school curricula, it was impossible to expect that all teachers would teach the same lessons. Thus, using a general set of frameworks that had been developed to support argumentation, teachers wrote their own lesson material to facilitate its use in ways that were appropriate to the content of their curricula. At the end of the year, another set of data was collected from the same group of 6 teachers who were teaching argumentation to the intervention class in a scientific context and in a socio-scientific context. Again, data were collected by audiotaping the teachers and video-taping the same set of four pupils, wherever possible (11 teacher tapes, 22 pupil videos). In addition for comparison purposes, a set of exactly similar data was collected from the control group on argumentation in a socio-scientific context (5 teacher audiotapes, 10 pupil videos). The control classes only implemented the zoo, leisure centre, and science lessons at the beginning and at the end of the school year. The teachers did not conduct any other argument-based lessons across the year, whereas the experimental classes did. 6. DATA ANALYSES The approach taken to the analysis of the teachers’ discourse was to use Toulmin’s (1958) model of argument as an analytical framework to identify the salient features of argument in the speech. This required an extended process of defining and elaborating how this framework should be interpreted and used. The following section illustrates our method of coding the transcripts using TAP as a guiding framework. In the case of the following example of pupil discourse: ‘Zoos are horrible, I am totally against zoos’ our focus would be on the substantive claim. In this case, the difficulty lies in the fact that both can be considered to be claims, i.e. ‘Zoos are horrible’ and ‘I am totally against zoos’ The question for the analysis then becomes which of these is the substantive claim and which is a subsidiary claim. Our general view is that there is inevitably a process of interpretation to be made and that some of that process is reliant on listening to the tape and hearing the force of the various statements. Part of this might be substantiated by Austin and Urmson’s (1976) distinction between locutionary statements, ones which have an explicit meaning, and perlocutionary statements, ones which have implicit meaning. The perlocutionary force, with which these statements are distinguished, is an aid to resolving that which is intended as the substantive claim. Here, our reading is that the emphasis lies on the second part of the statement because the task context demands a reference to a particular position (for or against zoos) and that this is therefore the substantive claim. In choosing to use TAP in this manner, we have developed a good reliability (more than 80 %) among the coders.

S. ERDURAN, J. OSBORNE, S. SIMON

387

As an example, consider the following case between the student and the teacher. S

I’ve got a con. If the animals are always walking about in the same places they might get angry and be dangerous.

T

Right, this is an anti, is it? behaviour.

So, being caged may alter their

The position represented by the student is ‘against zoos’ and is expressed as a claim in the phrase: “I’ve got a con.” The student further adds to this claim by saying that “if the animals are always walking about in the same places, they might get angry and be dangerous.” This elaboration, we consider as data to support his claim. The teacher subsequently interprets and justifies the choice for data by saying that “being caged may alter their behavior”. We regard the teacher’s contribution as the warrant for the argument being constructed. Such a co-construction of arguments between students and teachers was typical in all the transcripts we have studied in our project. Thus, our approach to the work was always to seek to identify, through either a careful reading of the transcript, or alternatively, listening to the tape, what constituted the claim. Once, the claim was established, the next step was the resolution of data, warrants, and backings. Our view here is that a necessary requirement of all arguments that transcend mere claims is that they are substantiated by data. Therefore, the next task is the identification of what constitutes the data for the argument which is often preceded by words such as ‘because’, ‘since’, or ‘as’. The warrant, if present, is then the phrase or substance of the discourse which relates the data to the claim. 7. ARGUMENTATION IN THE CLASSROOM Each teacher implemented the same activity one year apart with comparable students. The lessons were similar in structure in that there was an introduction, group discussions, group presentations, and finally assignment of homework in either case for both years. Typical transcript data on two teachers for the two years are summarized in Figure 1. The x-axis indicates the features of Toulmin’s argument pattern (TAP) that were used in different combinations. For example, CD indicates those instances where a claim (C) was coupled with data (D). CDWB indicates that there was a claim, data, warrant, and backing as part of one argument presented. The y-axis illustrates the frequency of instances that such permutations of TAP occurred within the transcript. In other words, we counted the number of times each kind of TAP occurred in the data across both years for each teacher. The figures seem to suggest several trends. First, there was argumentation discourse in the classroom across both years. In the figures we see specific examples of the extent to which each teacher’s class is involved collectively in the construction of the particular aspects of TAP. In other words, we can trace the nature of different permutations of TAP in their lessons. Second, it is apparent that the pattern of these features suggests that each of these teachers carries out/uses

388

THE ROLE OF ARGUMENTATION

argument in the same way across the two years. In other words, the trends across the use of different permutations of TAP are similar across two years. This would suggest that there is no common pattern and that the use of argumentation is teacher dependent – in other words, that there are no general patterns Overall, the figures illustrate the nature of progression of teachers across two years. Going from left to right on the x-axis, there is an increasing complexity in the way that TAP is constructed, i.e. the inclusion of warrants, backings, and rebuttals. Hence, counts on the right side of the charts indicate, in our view, an improvement in the nature of arguments in that they contain more components of TAP. Therefore, a shift, for example, from CD (claim-data) to CDW (claim-data-warrant) across two years represents an improvement in the arguments constructed in the class format. Using this approach to analysis for all the teachers, we have produced a profile of the discourse of argumentation for all the teachers across the two years. (Table 1 and 2).

Table 1. Collective reasoning in Sarah's class Year 1 vs Year 2 35

Frequency of instances

30

Year 1 Year 2

25

20

15

10

5

0 CD

CW

CDW

CDR

Feature of TAP

CDWB

CDWR

S. ERDURAN, J. OSBORNE, S. SIMON

389

Table 2. Collective reasoning in Thomas' class Year 1 vs. Year 2 35 Year 1 Year 2 30

Frequency of instances

25

20

15

10

5

0 CD

CDW

CDB

CDR

CDWB

CDWR

CDWBR

Features of TAP

The data show how the discourse of the classroom is dominated by arguments that contain fewer elements of TAP and are less elaborated. The important detail, nevertheless, is that there is a significant (p<0.01) improvement in the overall pattern of discourse between year 1 and year 2, with more elaborated arguments being used by some or all of the teachers. Closer analysis shows that this change is a result of the changes made by 8 of the twelve teachers and that for 4 teachers there was no significant change. 8. ASSESSING THE QUALITY OF ARGUMENTATION In seeking to answer our third research objective we have also focused on the discussions between pupils. In each class, two groups of 3 to 4 pupils were identified by the teacher, and their discussions were taped and transcribed. The transcripts were then searched to identify genuine episodes of oppositional analysis and dialogical argument. Opposition took many different forms and many arguments were co-constructed where students provided data or warrants for others’ claims. Transcripts of group discussions (2 groups per teacher) were examined to determine the number of episodes of explicit opposition in student discourse. In other words, the instances where students were clearly against each other were traced. Typically, these instances were identified through the use of words such as “but”, “I disagree with you”, “I don’t think so”, and so on. Once these episodes were characterized in the group format, they were re-examined for the interactions among the students in terms of who was opposing whom and who was elaborating on what idea or

390

THE ROLE OF ARGUMENTATION

reinforcing or repeating an idea. In this fashion, the pattern of interaction for each oppositional episode was recorded for two groups from each teacher's classroom. In establishing our framework, see Figure 1, we have drawn two major distinctions. The first is, does an argument contain any reasons, i.e. data, warrants, or backing, to substantiate its claim; transcending mere opinion and developing rational thought is reliant on the ability to justify and defend one’s beliefs. Hence, we see the simplest arguments are those consisting of a claim; the next level is arguments accompanied by data or warrants; followed by a third level of arguments consisting of claims, data, warrants, and rebuttals. Episodes with rebuttals are, however, of better quality than those without because individuals engaged in episodes without rebuttals remain epistemically unchallenged. The reasons for their belief are never questioned and are simply opposed by a counter claim that may be more or less persuasive, but which is not a substantive challenge to the original claim. At their worst, such arguments are reducible simply to the enunciation of contrasting belief systems. For instance, the confrontation between a creationist and a Darwinist without any attempt to rebut the data or the warrants of the other would have no potential to change the ideas and thinking of either, for the basis of their belief rests on the data and warrants they use as justification. Only arguments that rebut these components of argument can ever undermine the belief of another.

Level 1:

Level 1 argumentation consists of arguments that are a simple claim vs. a counter claim or a claim vs. claim

Level 2:

Level 2 argumentation has arguments consisting of claims with either data, warrants or backings but do not contain any rebuttals.

Level 3:

Level 3 argumentation has arguments with a series of claims or counter claims with either data, warrants or backings with the occasional weak rebuttal.

Level 4:

Level 4 argumentation shows arguments with a claim with a clearly identifiable rebuttal. Such an argument may have several claims and counter claims as well but this is not necessary.

Level 5:

Level 5 argumentation displays an extended argument with more than one rebuttal.

Figure 1. Analytical Framework used for assessing the quality of argumentation This schema of analysis enables us to make various comparisons of the performance of the different groups at argumentation. Figure 2 shows the distribution of arguments by level for all of the oppositional episodes currently analysed. In total, we have identified 183 oppositional episodes from 63 groups in

S. ERDURAN, J. OSBORNE, S. SIMON

391

33 lessons. Thus, in summary, there were on average approximately 3 oppositional episodes per group per lesson.

Argument Levels Pre and Post 45% 40% 35% 30% 25% 20% 15% 10% 5% 0%

Pre Post

1

2

3

4

5

Argument Level Figure 2. Chart showing numbers of each level of argumentation achieved in small group discussions This chart shows that the largest number of arguments emerging from the data both at the beginning and the end of the year was at level 2 (40% and 30% respectively). Encouragingly though, whereas at the beginning of the year only 39% of pupil arguments were at level 3 or above at the beginning of the year, by the end of the year, the corresponding figure was 55%. Whilst this change is not significant, it does show some positive development in the quality of argument. Moreover, the number of level 1 arguments was reduced from 21% to 15%. This finding is particularly encouraging as it suggests that only a small minority of pupils who developed arguments did not attempt to offer a rationale for their claims, and that the intervention has led to a diminishment in the number of such arguments. Level 1 arguments are also problematic in that it is these types of argument that have the most potential for argumentation which is confrontational and thus reinforcing the lay perception of ‘argument as war’ (Cohen, 1995), rather than argument as a process of collaborative brainstorming towards the establishment of ‘truth’ or better understanding. This method of analysis permits a number of comparisons of the performance of the groups. Results about the context of the lessons (science vs socioscientific) will be described in more detail in another paper (Osborne, Erduran & Simon, in press).

392

THE ROLE OF ARGUMENTATION 9. CONCLUSIONS

In this paper, we have presented some of the findings emerging from our work on developing argumentation in school science classrooms, its analysis, and the assessment of its quality. Methodologically, we think our work has made progress on several fronts. First, the work has sought to develop with teachers sets of materials that can be used in a structured and focused manner to facilitate argumentation in the classroom. As a result of this experience, we feel that we have gained some insights into the means of establishing a context which facilitates argumentation in the classroom. Hence, in the next phase of our work, we have developed training resources through the IDEAS project (Osborne, Erduran & Simon, 2004). Second, our work with teachers has led to a change in the practice of the majority of this group, leading us to believe that, despite the many obstacles and barriers posed by the demands to implement different and innovative practice, it is possible for science teachers to adapt, change, and develop their practice to one where there is a fundamental change in the nature of classroom discourse. Third, one of the many problems that bedevils work in this field is a reliable systematic methodology for (a) identifying argument and (b) assessing quality. Our adoption and use of Toulmin has also provided us with a method for studying discourse in the classroom which we present in more detail elsewhere (Erduran, Simon & Osborne, in press). We have also illustrated how we can apply this schema to sets of data obtained from teachers implementing argumentation in the classroom. These data sets do show evidence of positive improvement in the quality of student argumentation; however, this change has not been significant. This would suggest to us that developing the skill and ability to argue effectively is a long-term process – something which only comes with recurrent opportunities to engage in argumentation throughout the curriculum, rather than the limited period of our 9 month intervention. ACKNOWLEDGEMENTS This study was supported by the UK Economic and Social Science Research Council grant number R000237915. In addition, we would like to acknowledge the many teachers who have worked with us on this project and their efforts with this research. REFERENCES Alexopoulou, E. & Driver, R. (1997). Small group discussions in physics: peer interaction modes in pairs and fours. Journal of Research in Science Teaching, 33(10), 1099-1114. Austin, J. L. & Urmson, J. O. (1976). How to do things with words. (2nd ed.). London: Oxford University Press. Boulter, C.J. & Gilbert, J.K. (1995). Argument and Science Education. In P.J.M. Costello and S. Mitchell (Eds.) Competing and consensual voices: the theory and practice of argumentation. Clevedon. Multilingual Matters.

S. ERDURAN, J. OSBORNE, S. SIMON

393

Cowie, H. & Rudduck, J. (1990). Co-operative learning traditions and transitions. Volume three of Learning Together - Working Together. London. BP Educational Service. Cohen, D. (1995). Argument is War....and War is Hell: Philosophy, Education, and Metaphors for Argumentation. Informal Logic, 17(2), 177-188. Department for Education and Employment (1999). Science in the National Curriculum. London: HMSO. Driver, R., Leach, J., Millar, R. & Scott, P. (1996). Young People's Images of Science. Buckingham: Open University Press. Driver, R., Newton, P. & Osborne, J. (2000). Establishing the norms of scientific argumentation in classrooms. Science Education, 84(3), 287-312. Erduran, S. & Osborne, J. (in press). Developing arguments. In, S. Alsop, L. Bencze & E. Pedretti (Eds.), Analysing exemplary science teaching: Theoretical lenses and a spectrum of possibilities for practice. Open University Press. Erduran, S., Simon, S. & Osborne, J. (in press). TAPping into argumentation: Developments in the use of Toulmin's Argument Pattern in studying science discourse. Paper to appear in Science Education. Erduran, S. (2001). Philosophy of chemistry: An emerging field with implications for chemistry education. Science & Education, 10(6), pp. 581-593. Gee, J. (1996). Social Linguistics and Literacies (2nd ed.). London: Taylor. Giddens, A. (1990). The Consequences of Modernity. Cambridge: Polity Press. Herrenkohl, L.R. & Guerra, M.R. (1998). Participant Structures, Scientific Discourse, and Student Engagement in Fourth Grade. Cognition and Instruction, 16(4), 431-473. Kuhn, D. (1991). The Skills of Argument. Cambridge: Cambridge University Press. Lemke, J.L. (1990). Talking Science: Language, Learning and Values. Norwood, New Jersey: Ablex Publishing. Mason, L. (1996). An analysis of children's construction of new knowledge through their use of reasoning and arguing in classroom discussions. Qualitative Studies in Education, 9(4), 411-433. Means, M.L. & Voss, J.F. (1996). Who reasons well? Two studies of informal reasoning among children of different grade, ability, and knowledge levels. Cognition and Instruction, 14, 139-178. Minstrell, J. & Van Zee, E. (Eds.) (2000). Teaching in the Inquiry-based science classroom. Washington, DC: American Association for the Advancement of Science. National Research Council. (2000). Inquiry and the National Science Education Standards. Washington D.C.: National Academy Press. Newton, P., Driver, R. & Osborne, J. (1999). The Place of Argumentation in the Pedagogy of School Science. International Journal of Science Education, 21(5), 553-576. Osborne, J., Erduran, S. & Simon, S. (2004). Ideas, Evidence and Argument in Science. INSET manual, Resource Pack and Training Video. London: King's College London. Osborne, J., Erduran, S. & Simon, S. (in press). Enhancing the quality of argument in school science. Paper to appear in Journal of Research in Science Teaching. Osborne, J.F., Erduran, S., Simon, S. & Monk, M. (2001). Enhancing the Quality of Argument in School Science. School Science Review, 82(301), 63-70. Russell, T.L. (1983). Analysing arguments in science classroom discourse: can teachers’ questions distort scientific authority? Journal of Research in Science Teaching, 20, 2745. Scott, P. (1998). Teacher talk and meaning making in science classrooms: a Vygotskian analysis and review. Studies in Science Education, 32, 45-80.

394

THE ROLE OF ARGUMENTATION

Suppe, F. (1998). The structure of a scientific paper. Philosophy of Science, 65(3), 381-405. Toulmin, S. (1958). The Uses of Argument. Cambridge: Cambridge University Press.

MEANING MAKING IN HIGH SCHOOL SCIENCE CLASSROOMS: A FRAMEWORK FOR ANALYSING MEANING MAKING INTERACTIONS

PHIL SCOTT¹, EDUARDO MORTIMER² ¹University of Leeds, UK ²Universidade Federal de Minas Gerais, Brazil ABSTRACT In this paper, we introduce and exemplify aspects of a tool for analysing the various forms and functions of discursive interactions in high school science classrooms. This tool, or analytical framework, is based on a sociocultural view of teaching and learning, and consists of five linked aspects: Teaching purposes; Content of the classroom interactions; Communicative approach; Patterns of discourse; Teacher interventions. Here we focus attention on introducing and exemplifying how different teaching purposes can be addressed through combinations of communicative approach and patterns of discourse, as the scientific ‘story’ develops. In this way we demonstrate how the different aspects of the framework interrelate, providing a coherent basis for analysing classroom interactions. Finally we turn to the ways in which the framework has been used in planning science teaching and discuss how the framework is being used with science teachers in the context of professional development programmes, in both the UK and Brazil.

1. INTRODUCTION In recent years the influence of sociocultural psychology (see, for example, Bakhtin, 1986; Vygotsky, 1978 & 1987) on research in science education has been reflected in the gradual development of interest in studies of how meanings are developed through language and other modes of communication in the science classroom. Parallel to this new focus, and in some ways interwoven with it, is the so called ‘discursive turn’ in psychology (see, for example, Harré & Gillett, 1994) and an increasing interest in rhetoric (see, for example, Kuhn, 1992; Billig, 1996), which have highlighted, from different points of view, the importance of investigating classroom discourse and other rhetorical devices in science education (see, for example, Lemke, 1990; Sutton, 1992; Halliday & Martin, 1993; Ogborn et al., 1996; Roychoudhury & Roth, 1996; Mortimer, 1998; Scott, 1998; Kress et al., 2001). This ‘new direction’ for science education research (Duit & Treagust, 1998) signals a move away from studies focusing on individual student understandings of specific phenomena towards research into the ways in which understandings are developed in the social context of the science classroom. Sociocultural theory is one of the main traditions that has informed this research. From this perspective, concept is equated to meaning (Vygotsky, 1987) and the focus is on meaning and meaningmaking. Learning is viewed, not in terms of replacing old ideas with new ones, but as negotiating new meanings in a communicative process where different cultural 395 K. Boersma et al. (eds.), Research and the Quality of Science Education, 395—406. © 2005 Springer. Printed in the Netherlands.

396

MEANING MAKING IN HIGH SCHOOL SCIENCE CLASSROOMS

perspectives meet each other in a process of mutual growth. Discursive interaction constitutes the process of meaning-making. Despite this new emphasis on talk and interaction, relatively little is known about how teachers support students’ meaning making in science classrooms, how these interactions are enacted, and how different kinds of talk might support student learning. 2. BACKGROUND The problem addressed in this paper, is that of analysing the ways in which the teacher can act to guide meaning-making interactions in high school science classrooms to support student learning. In responding to the problem, we introduce a framework for analysing the speech genre (Bakhtin, 1986) of science classrooms. This framework is the product of an ongoing research programme conducted over a number of years (see, Mortimer, 1998; Scott, 1998; Mortimer & Scott, 2000; Mortimer & Scott, 2003). A detailed description of the development of the framework is set out elsewhere (Mortimer & Scott, 2003). Suffice it to say for the purposes of this article, the framework is based on a sociocultural perspective on teaching and learning (Mortimer & Scott, 2003) and has been developed through a series of detailed case studies. The case studies focus on the interactions and activities of high school science lessons, in England and Brazil, in which conceptually demanding science topics (such as ‘air pressure’ and ‘the particulate theory of matter’) were taught to students aged 14-16 years. The lesson sequences lasted for between 6 to 12 hours, and for each sequence video- and audio-recordings of both teacher-student and student-student interactions were collected. From the analysis of these data and from the insights gained from various aspects of sociocultural theory (for example, drawing on Bakhtin’s [1981] distinction between authoritative and internally persuasive discourse), the framework was developed through an iterative process of application and refinement. In this paper, we introduce and exemplify two central and linked aspects of the framework and discuss implications for the wider use of the framework, both as an instrument for analysis and planning. 3. A FRAMEWORK FOR ANALYSING MEANING-MAKING INTERACTIONS The analytical framework is based on five linked aspects, which focus on the role of the teacher, and are grouped in terms of teaching Focus, Approach and Action:

P.H. SCOTT, E.F. MORTIMER

397

Aspect of Analysis i. Focus

ii. Approach

iii. Action

1. TEACHING PURPOSES

3.

2. CONTENT

Communicative Approach

4. PATTERNS OF DISCOURSE

5. TEACHER INTERVENTIONS

Fig 1: The Analytical Framework: A tool for analysing science teaching interactions In the following sections we shall focus our attention on two aspects of analysis. These are the Communicative Approach and the Patterns of Discourse. 4. COMMUNICATIVE APPROACH The newly-developed concept of Communicative Approach is central to the framework, in providing a perspective on how the teacher works with students to develop ideas in the classroom. It focuses on questions such as whether or not the teacher interacts with students (taking turns in the discourse), and whether the teacher takes account of students’ ideas as the lesson proceeds. In developing this aspect of analysis, we have identified four fundamental classes of communicative approach, which are defined by characterising the talk between teacher and students along each of two dimensions, dialogic-authoritative and interactive-non interactive. The Dialogic-Authoritative Dimension As a teacher interacts with students in the classroom, their approach can be characterised along this dimension, which extends between two extreme positions: either the teacher hears what the student says from the student’s point of view, or the teacher hears what the student has to say only from the school science point of view. We refer to the first kind of interaction as involving a dialogic communicative approach. Here, the teacher attends to the students’ points of view as well as to the school science view. More than one voice is represented and there is an exploration or ‘interanimation’ (Bakhtin, 1981) of ideas. The second kind of interaction involves an authoritative communicative approach, where attention is focused on just one point of view: only one voice is heard, and there is no exploration of different ideas.

This distinction between authoritative and dialogic functions has been discussed by Wertsch (1991), and used by Mortimer (1998) in analysing discourse from a Brazilian classroom. It is based on the notions of authoritative and internally persuasive discourse, as outlined by Bakhtin (1987), and on the functional dualism of texts introduced by Lotman (1988; quoted by Wertsch, 1991, pp. 73-74).

398

MEANING MAKING IN HIGH SCHOOL SCIENCE CLASSROOMS

The Interactive-Non-Interactive Dimension An important feature of the distinction between dialogic and authoritative communicative approaches is that a sequence of talk can be dialogic or authoritative in nature, independent of whether it is uttered individually or between people. What makes talk functionally dialogic is that more than one point of view is represented and ideas are explored and developed, rather than whether it was produced by a group of people or by a solitary individual. This point leads us to the second dimension to consider in thinking about the Communicative Approach: that the talk can be interactive in the sense of allowing for the participation of more than one person, or non-interactive in the sense of excluding the participation of other people. Four Classes of Communicative Approach Combining the two dimensions, any sequence of classroom talk can be identified as being either interactive or non-interactive on the one hand, and dialogic or authoritative on the other (although, in practice, some classroom sequences entail aspects of both dialogic and authoritative talk). We can represent this combining of the two dimensions in the following way: INTERACTIVE DIALOGIC

AUTHORITATIVE

NON-INTERACTIVE

A. Interactive / Dialogic

B. Non-interactive / Dialogic

B.

C.

Interactive / Authoritative

Non-interactive/ Authoritative

Let us now consider how each of the two interactive communicative approaches might appear in practice in the classroom. Interactive/Authoritative Communicative Approach The following episode took place between a teacher and thirteen year-old students in an English high school. The transcript is taken from a series of lessons on Energy, and in the previous lesson the students had been working on a practical activity with electric bells. 1. Let’s just ignore the sparks… Teacher: Do you remember the electric bell? Students: Yes! [in chorus] Teacher: OK! Did any of you notice, did any of you actually hold onto the bell after it had...been working? What did you notice? Suzanne: Vibration Teacher: Well, the arm vibrated, yes. Sound. What else did you notice?

P.H. SCOTT, E.F. MORTIMER

399

Tom: It was loud. Teacher: That's not quite what I'm getting at. Remember the bell. There's the bell [holding up a bell in front of the class]. You did the experiment. If you held onto this bit here where the wires were [indicating], did you notice anything there? Jason: There were sparks there. Teacher: Heat, did you notice some heat? Jason: There were sparks from there. Teacher: There were? Jason: Sparks. Teacher: There were some sparks, yes. Let's just ignore the sparks a minute...some heat. There was a little bit of heat there with that one. In this sequence, it is clear that the teacher is focusing his attention exclusively on the production of heat by the electric bell. He reminds the students about the activities of the previous lesson, and asks, ‘What did you notice?’ Suzanne replies with a single word, ‘Vibration’. The teacher acknowledges her answer, ‘Well, the arm vibrated, yes. Sound’. It is clear, however, that this is not what the teacher wants to hear and he moves on, discounting in turn the students’ suggestions, ‘it was loud’ and ‘there were sparks’. This is an authoritative interaction, where the teacher’s sole aim is to arrive at the idea that the bell heats up as it is working. The teacher’s interventions are based on instructional questions for which he has in mind only one answer. If the students do not come up with the required answer, their suggestions are put to one side: ‘That’s not quite what I’m getting at’; ‘Let’s just ignore’. The students’ contributions are limited to single brief assertions ‘vibration’, ‘it was loud’, ‘there were sparks’, made in response to the teacher’s questions. This kind of interaction, where the teacher leads students through a sequence of instructional questions and answers with the aim of reaching one specific point of view, is typical of an interactive/authoritative communicative approach Interactive/Dialogic Communicative Approach The interactive/dialogic approach contrasts with authoritative interactions in that here the teacher listens to, and takes account of, the students’ points of view, even though these might be quite different from the school science perspective. In the following sequence a teacher is working with a group of fourteen year-old students, on the characteristic properties of solids, liquids and gases. The teacher asks the class whether ‘solids are hard’, an idea which has been suggested by one of the students: 2.Other people are desperate to say… Teacher: Solids are hard? Students: No, no. Soft! [together]

400

MEANING MAKING IN HIGH SCHOOL SCIENCE CLASSROOMS Teacher: Well, if you say ‘no’, put your hand up and tell me, give me an example, which would prove an exception to that… [the idea that solids are hard]. Suzanne: Powder’s a solid, but you can crush it. Teacher: Powder’s? Suzanne: a solid but you can still crush it. Teacher: Powders aren’t particularly hard, yes, if you’re talking about hard to the touch. Paul? [who has his hand up] Paul: It’s…cos…it’s [the powder] got a gas in between, so it’s hard. Teacher: So you think that all solids are hard? Paul: Yeah. Teacher: Other people are desperate to say that all solids aren’t hard. Martin? Martin: Er…fabric’s soft. Students:Yeah…yeah…[lots of muttering] Teacher: Wait. Just a minute. If you’re saying things, can you say it to the front, so that we can all share these ideas.

So the teacher and class explore the idea that ‘solids are hard’, with the students spontaneously offering points of view in responding to what others have said. The teacher helps sustain the discussion both in terms of enabling students to contribute (‘other people are desperate to say’; ‘can you say it to the front?’) and in making requests for points of substantive clarification (‘give me an example’; ‘so you think…?’). This kind of interaction, where the teacher seeks to elicit and explore the students’ ideas about a particular issue, is typical of an interactive/dialogic approach. Here the authoritative instructional questions of the previous example are replaced by ‘genuine’ questions, which probe students’ points of view. Non-Interactive Communicative Approaches Let us now turn our attention, more briefly, to the two classes of non-interactive communicative approach. On first consideration, the non-interactive/dialogic approach may appear to involve an unlikely combination of attributes (how can classroom talk be non-interactive and yet dialogic?). This approach sees the teacher in presentational mode (non-interactive), but this time explicitly considering and drawing attention to different points of view (dialogic). In the classroom the teacher might adopt this communicative approach in pulling together and presenting a range of students’ ideas or possibly in drawing attention to the differences between everyday and school science points of view. The non-interactive/authoritative approach involves the teacher in presenting a specific point of view. This approach is likely to take the form of a ‘lecture’ from the teacher which focuses exclusively on the school science account.

P.H. SCOTT, E.F. MORTIMER

401

5. PATTERNS OF DISCOURSE If we re-examine the two transcripts presented here, it is possible to identify distinctive patterns of discourse (a second aspect of the framework) in each. The I-R-E Pattern of Discourse This pattern of interaction is very common in classrooms and is played out in ‘patterns of three’ with utterances from teacher-student-teacher. It is referred to as a triadic ‘I-R-E’ interaction (Mehan, 1979), where: I stands for Initiation: normally through a question from the teacher, R stands for Response: from the student, E stands for Evaluation: by the teacher. Thus, in the ‘Let’s just ignore the sparks’ transcript, the teacher initiates the exchange by posing a question, ‘What did you notice?’ [Initiation]. Suzanne replies with a single word, ‘Vibration’ [Response]. The teacher acknowledges her answer, ‘Well, the arm vibrated, yes. Sound’ [Evaluation]. It is clear, however, that this is not what the teacher wanted to hear, and he moves on to his next question, ‘What else did you notice?’ [Initiation]. This pattern of initiation-response-evaluation is repeated throughout the episode until the teacher finds the response he is looking for. The I-R-E pattern of interaction is very distinctive and very common in high school classrooms, and most authoritative interactions are played out through an IR-E pattern. The I-R-F and I-R-F-R-F- Patterns of Discourse An alternative form of triadic discourse occurs when, instead of making an evaluation of the student’s response, the teacher gives the student feedback or elaborates on the student’s answer, so the student is supported in developing his/her own point of view. We refer to this pattern of interaction as an I-R-F (F standing for feedback), rather than an I-R-E form. This pattern of discourse can also occur in a chain of interactions, as an I-R-F-R-F-R- form, where the elaborative feedback (F) from the teacher is followed by a further response from the student (R), and so on. Thus, in the second transcript, ‘Other people are desperate to say’, the teacher starts with, ‘give me an example’ [Initiation]. Suzanne states, ‘Powder’s a solid…’ [Response] and the teacher repeats the comment to sustain the interaction ‘Powder’s?’ [Feedback]. Suzanne responds ‘a solid but you can still crush it’ [Response], and the teacher elaborates on her comment, ‘Powders aren’t particularly hard, yes, if you’re talking about hard to the touch’ [Feedback]. Paul then offers an alternative point of view, ‘It’s…cos… it’s got a gas in between, so it’s hard’ [Response], and the teacher asks for elaboration, ‘ So you think that all solids are hard?’ [Feedback]. By establishing this pattern of discourse, the teacher is able to explore the students’ ideas. In some of his elaboration, ‘So you think that all solids are hard?’ in order to clarify the point of view. In this way the teacher uses an I-R-F-R-F- pattern of discourse to support a dialogic interaction interventions, the teacher simply

402

MEANING MAKING IN HIGH SCHOOL SCIENCE CLASSROOMS

‘bounces-back’ the student’s words, ‘Powder’s?’, encouraging the student to continue and thereby helping to sustain the interaction. At other points the teacher asks for substantive points of. 6. COMMUNICATIVE APPROACH AND PATTERN OF DISCOURSE As exemplified in the preceding sections, we can identify an emerging relationship between classes of Communicative Approach and Patterns of Discourse. See Figure 3 below: Communicative Approach

Interactive/ Authoritative

Interactive/ Dialogic

often made via

often made via

Pattern of Discourse I-R-E triads I-R-F-R-F-R- chains Fig 3: Relationship between Communicative Approach and Pattern of Discourse In this way the link is made between two levels of the framework: Approach and Action. This allows us to begin to see how different communicative approaches might actually be played out by the teacher and students in the classroom. 7. DISCUSSION In this paper we have introduced and exemplified two aspects of a framework for capturing and characterising the talk of school science lessons. In this final section, we turn to considering what is distinctive about the framework and discuss how it might be drawn upon to inform thinking and action in relation to different aspects of science teaching and learning. Our first comments relate to the use of the framework as a tool for analysing the interactions of science lessons. Here, we believe that a particular strength of the framework lies in the way that its component aspects articulate with one another to provide an integrated analysis of classroom interactions. In particular, the framework brings together features of the discourse (such as the patterns of discourse and the communicative approach) alongside the content of that discourse. We have reported studies elsewhere (Mortimer & Scott, 2003) in which a detailed account of the development of the content of the talk of the social plane of the classroom, is integrated into the analysis of a sequence of lessons. The analysis of content is made in terms of whether the talk focuses on description, explanation, or generalisation, and whether it is empirically or theoretically based (see Mortimer & Scott, 2000). For example, in the episode ‘Let’s just ignore the sparks’, the talk is empirically based (focusing on directly observable properties) and descriptive in its content focus. By attending to content, the framework offers an analytical approach which goes beyond those linguistic studies of classroom discourse where analyses

P.H. SCOTT, E.F. MORTIMER

403

are carried out, and findings reported, solely in terms of discursive features (such as the mode of teacher questioning). The actual content of the discourse (whether on ‘volcanoes’ or ‘particle theory’) is not treated as a significant feature. In addition, we believe that the concept of Communicative Approach provides a fundamentally important way of thinking about, and analysing, classroom interactions. The key insight offered here is that classroom discourse can be analysed both along an authoritative-dialogic dimension and in terms of interactive and non-interactive approaches. We have found, through a number of subsequent studies (see, for example, Mortimer & Scott, 2003) that the resulting four classes of communicative approach map convincingly onto classroom interactions as researched in both Brazil and the UK. Furthermore, we consider that the relationship between communicative approach and pattern of discourse, which we discussed earlier, adds significantly to the integrated nature of the framework by making the link from approach to specific patterns of discourse. This integrated approach can be taken one step further in considering how a specific teaching purpose might be realised through a particular communicative approach and how this approach is played out in the classroom through a particular pattern of discourse (moving from Focus to Approach to Action in the framework). For example, the episode ‘Other people are desperate to say’ is taken from the start of a teaching sequence. Here the teacher addresses the teaching purpose of eliciting students’ views by way of an interactive/dialogic communicative approach, which she developed through I-R-FR-F-R- chains of discourse. Building on this line of reasoning, we believe that the framework successfully meets four criteria which might reasonably be applied to its evaluation. Firstly, the framework is generally applicable, insofar as it has provided valuable insights into all of the science lessons to which we, and others (see, for example, Viiri et al., 2003) have applied it. Secondly, it captures the key features of classroom interactions in drawing on new concepts (such as that of communicative approach) and extending existing ideas (such as moving beyond triadic discourse to demonstrate the importance of I-R-F-R-F-R- chains). Thirdly, the framework offers an integrated approach in which it is possible to see how the different parts of the analysis are inter-related. Finally, we believe that the framework is pitched at a workable level of detail which resonates with the practices and activities of real teachers and students in real classrooms. Just as the framework has much to offer in analysing the interactions of science lessons, we also believe that that it has great potential as a tool for planning lessons. In the UK at present, science lessons tend to be dominated by different kinds of practical activity, whether a demonstration by the teacher or an experimental activity to be carried out by the students, or a writing exercise, and so on. Planning science lessons is thus conceptualised in terms of what the students will be doing, and little or no attention is given to the nature of the talk around those activities (Leach & Scott, 2002). What we are suggesting here is an approach to planning in which explicit attention is paid to the nature of the talk in each phase of the lesson sequence, and how that links both to the teaching activities and to the teaching purposes.

404

MEANING MAKING IN HIGH SCHOOL SCIENCE CLASSROOMS

Lesson planning thus conceptualised, therefore involves identifying both teaching activities and communicative approaches, in order to address specific teaching purposes. Just as teachers are familiar with including a range of different kinds of teaching activities in a lesson sequence, then so too, there is the need to vary the communicative approach. If, for example, the teacher introduces the scientific point of view through an authoritative presentation, an important question in planning the next phase of the sequence is how the students can now be given the opportunity to talk through, and to explore, this point of view for themselves. Such a variation in communicative approach has been clearly illustrated in analyses, presented elsewhere (Mortimer & Scott, 2003), where a distinctive pattern or rhythm to the discourse emerged. We strongly maintain that in any teaching sequence, there should be variation in communicative approach, covering both dialogic/authoritative and interactive/non-interactive dimensions. Thus, there will be times in the development of a sequence of lessons, when the teacher needs to make an authoritative statement of the school science point of view. There will be other times when the teacher needs to allow time and space for the students to talk through and to use these scientific ideas for themselves. In this way, we believe that the rhythm of the teaching performance should be consequent upon changes in communicative approach. A group of colleagues at the University of Leeds, UK, has developed a number of teaching interventions as part of the ‘Evidence Based Practices in Science Education’ project. These interventions include explicit guidelines about the communicative approach to be used with different activities in addressing specific teaching purposes (Scott et al., 2001). Early findings indicate that the interventions have been successful in providing enhanced student learning gains, when compared with the gains achieved by parallel classes following the school’s normal curriculum (see Leach et al., 2003 in this book). The way in which science teaching is being conceptualised here is likely to differ significantly from existing practices. The nature of classroom talk is not something which tends to be explicitly considered by teachers; research studies point to the overwhelming dominance of triadic I-R-E discourse in science classrooms. If we are committed to recognising, and acting upon, the importance of the link between talking, thinking, and learning, then there is a considerable challenge to be addressed in considering how teachers might be supported in reflecting upon and developing their existing practices. We believe that a key starting point is to provide teachers with the tools which enable them to identify their existing ways of working with students and to see what other approaches are possible. We have found the framework to be extremely helpful in this respect whilst working with science teachers in both pre-service and in-service professional development contexts (Mortimer & Scott, 2003). It is particularly in these contexts that the workable level of detail referred to earlier assumes critical importance. Put simply, if the planning/analytical tool is too complicated and seems to miss the point of what happens in classrooms, then it will be of no instrumental value to teachers. In summary, we have made a case in this final section for the value of the framework, not only as a research tool for systematically analysing the interactions of science lessons, but also as a powerful basis for re-conceptualising approaches to

P.H. SCOTT, E.F. MORTIMER

405

thinking about the planning of science teaching. Furthermore we believe that the framework offers a workable set of tools which can be helpful to teachers in allowing them to reflect upon and to develop their teaching practices in professional development contexts. ENDNOTE In this section, we use the term dialogic as the opposing tendency to authoritative in characterising one of the dimensions of Communicative Approach. However, according to Bakhtin’s point of view, all discourse must be dialogic in form, including authoritative monologue (a non-interactive/authoritative communicative approach). We certainly agree that when a teacher makes a non-interactive authoritative presentation, then the meaning making process is dialogic in nature, as the students try to make sense of what is being said by laying down a set of their ‘own answering words’ to the words of the teacher. At the same time, and according to our own definition, we are clear that in authoritative discourse the teacher’s purpose is to focus the students’ full attention on just one meaning. It is in this sense that we have chosen to use the word ‘authoritative’ (whilst acknowledging the underlying dialogic nature of the interaction). Additionally, we have chosen the word ‘dialogic’ to contrast with an authoritative communicative approach, in order that we can draw upon the dialogic meaning of recognising others’ points of view. Thus, according to our definition, we are clear that in dialogic discourse the teacher attempts to take into account a range of students’, and others’, ideas.

REFERENCES Bakhtin, M.M. (1981). The dialogic imagination, Ed. by Michael Holquist, Trans. by Caryl Emerson & Michael Holquist. Austin: University of Texas Press. Bakhtin, M.M. (1986). Speech Genres & Other Late Essays, Ed. by Caryl Emerson & Michael Holquist, Trans. by Vern W. McGee. Austin: University of Texas Press. Billig, M. (1996). Arguing and thinking: A rhetorical approach to social psychology. Cambridge: Cambridge University Press Duit, R. & Treagust, D. (1998). Learning science: from behaviourism towards social constructivism and beyond. In: B.J. Fraser & K.G. Tobin (Eds.), International Handbook of Science Education, pp. 3-25. Dordrecht: Kluwer Academic Publishers. Halliday, M.A.K. & Martin, J.R. (1993). Writing Science. London: Falmer Press. Harré, R. & Gillett, G. (1994). The discursive mind. Thousand Oaks, California: Sage Publications, California. Kress, G., Jewitt, C., Ogborn, J. & Tsatsarelis, C. (2001). Multimodal teaching and learning: the rhetorics of the science classroom. London: Continuum. Kuhn, D. (1992). Thinking as argument. Harvard Educational Review, 62, 155-178. Leach, J.T. & Scott, P.H. (2002). Designing and evaluating science teaching sequences: an approach drawing upon the concept of learning demand and a social constructivist perspective on learning. Studies in Science Education, 38, 115-142. Leach, J., Ametller, J., Hind, A., Lewis, J. & Scott, P. (2003). Evidence-informed approaches to teaching science at junior high school level: outcomes in terms of student learning.

406

MEANING MAKING IN HIGH SCHOOL SCIENCE CLASSROOMS

Paper presented at the Annual Meeting of the National Association of Research in Science Teaching (NARST), Philadelphia, USA, March. Lemke, J.L. (1990). Talking Science. Language, Learning and Values. Norwood, New Jersey: Ablex Publishing Corporation. Lotman, Yu.M. (1988). Text within a text. Soviet Psychology, 26(3): 32-51. Mehan, H. (1979). Learning Lessons: Social organization in the classroom. Cambridge, MA: Harvard University Press. Mortimer, E.F. (1998). Multivoicedness and univocality in classroom discourse: an example from theory of matter. International Journal of Science Education, 20(1): 67-82. Mortimer, E.F. & Scott, P.H. (2000). Analysing discourse in the science classroom. In Leach, J., Millar, R. & Osborne, J. (Eds.) Improving Science Education: the contribution of research. Milton Keynes: Open University Press. Mortimer, E.F. & Scott, P.H. (2003). Meaning making in secondary science classrooms. Buckingham: Open University Press. Ogborn, J., Kress, G., Martins, I. & McGillicuddy, K. (1996). Explaining science in the classroom. Buckingham: Open University Press. Roychoudhury, A. & Roth, W.-M. (1996). Interactions in an open-inquiry physics laboratory. International Journal of Science Education, 18, No. 4, pp.423-445. Scott, P.H. (1998). Teacher talk and meaning making in science classrooms: A Vygotskian analysis and review. Studies in Science Education, 32: 45-80. Scott, P., Hind, A., Leach, J. & Lewis, J. (2001). Designing and implementing science teaching drawing upon research evidence about science teaching and learning. Paper presented at the European Science Education Research Association (ESERA), Third International Conference, Thessaloniki, Greece, August 21-25. Sutton, C. (1992). Words, science and learning. Buckingham: Open University Press. Viiri, J., Saari, H. & Sormunen, K. (2003). Describing the rhythm of science teacher talk. Paper presented at the European Science Education Research Association (ESERA), Fourth International Conference, Noordwijkerhout, The Netherlands, August 19-23, 2003. Vygotsky, L.S. (1978). Mind in Society: The development of higher psychological processes. Cambridge, MA: Harvard University Press. Vygotsky, L.S. (1987). Thinking and Speech. In The Collected Works of L.S. Vygotsky; Rieber, R.W.; Carton, A.S. (Eds.). Trans. by Minich, N. New York: Plenum Press. pp. 39-285. Wertsch, J.V. (1991). Voices of the mind: A sociocultural approach to mediated action. Harvester Wheatsheaf.

FROM A CAUSAL QUESTION TO STATING AND TESTING HYPOTHESES: EXPLORING THE DISCURSIVE ACTIVITY OF BIOLOGY STUDENTS

MARIDA ERGAZAKI, VASSILIKI ZOGZA University of Patras, Greece

ABSTRACT This paper aims at exploring the discursive activity of one group of second year biology students during their collaboration on a task of stating and testing hypotheses to answer a causal question. The specific task is a part of a didactic sequence that was developed in the context of genetic engineering considering aspects of situated-learning theory, with the aim of providing students the opportunity to ‘talk science’ with their peers as participants of a hypothetical gene cloning project. Our focus is set on certain cognitive aspects of peers’ discourse. Hence, this paper is concerned with the construction of arguments, particularly on the level of argumentative operations (e.g. claims, justifications, challenges) and the context-bound epistemic operations (e.g. abducting, appealing to instances) activated by peers in order to produce a joint answer to the task’s causal question. Furthermore, it is concerned with the development of the ‘if…and…then’ hypothetical-deductive reasoning pattern potentially involved in peers’ hypothesistesting process.

1. INTRODUCTION Recent research in science education focuses on the study of students’ argumentation in various contexts (Mason, 1996; Desautels & Larochelle, 1999; Jimenez, 2000; Driver et al., 2000; Simonneaux, 2000). Our study attempts to build on this body of research work by focusing on the construction of biology students’ arguments while interacting in the context of genetic engineering for the formulation of a hypothesis and the development of ‘if… and…then’ reasoning patterns to test their hypothesis. Stating hypotheses (tentative explanations to causal questions) constitutes, in general, a complex process of combining empirical evidence, previous knowledge, and intuition (Lawson, 1995). The role of argument in this process seems to be crucial. Partial scientific claims towards an explanatory framework do need to be well grounded in warranting structures that are built on reliable epistemic criteria (Driver et al., 2000). Furthermore, the possible formulation of more than one alternative hypothesis for the same causal question, activates a process of comparative evaluation of their explanatory efficacy to decide which one is the 407 K. Boersma et al. (eds.), Research and the Quality of Science Education, 407—417. © 2005 Springer. Printed in the Netherlands.

408

FROM A CAUSAL QUESTION TO STATING AND TESTING HYPOTHESES

fittest (Giere, 1991) and therefore should be experimentally tested. Argument is also a necessary tool when designing the experimental tests to be used in hypotheticaldeductive reasoning. Stating, justifying, and evaluating scientific claims remains a prerequisite for making predictions about the expected outcomes of the test, assuming the hypothesis’ validity, and for defining the conditions of the hypothesis’s rejection by comparing the expected and the potentially observed outcomes of the proposed test (Lawson, 1995). Our focus is set on the process of formulating and testing hypotheses and the employment of argumentative and epistemic tools in this process. Thus, the question framing our study is ‘how do collaborating students formulate and test hypotheses in the context of genetic engineering?’. More specifically, ’what kind of argumentative and epistemic operations do students activate in the process of stating and testing hypotheses?’ and ‘to what extent do they follow the hypothetical-deductive reasoning pattern to explore the validity of their hypotheses?’. In summary, the objective of this paper is to highlight students’ reasoning patterns on hypothesis stating and testing by analyzing both the argumentative process towards its construction and the resulting construct itself. 2. METHODS The task and the setting A didactic sequence of genetic engineering was developed on aspects of situatedlearning theory to provide biology students with an authentic context for practicing scientific reasoning and discourse. In a student-centered setting, peers collaborated in small groups to create joint answers to tasks embedded in a hypothetical gene cloning mission as its meaningful and purposeful steps. Peers who were supposed to be responsible for cloning a medically useful plant gene were faced with choices, predictions, experimental proposals, and stating/testing hypotheses. The teacher’s role was limited to introducing themes, giving hints, and conducting whole class discussions after the group work. The participants of the group discussion presented here are three female students who volunteered to be tape-recorded while interacting as they stated and tested tentative explanations of the cloned gene’s inability to synthesize its protein in bacteria (see Appendix). Lawson’s hypothesis-testing quizzes (Lawson, 1995) were taken into account for the task’s development, resulting in the insertion of a scaffolding device which explicitly requires that peers predict the expected outcomes of their experimental test regardless of the adequacy of their hypothesis. The task aims at encouraging students to practice scientific reasoning through argumentative discourse. It does this by engaging peers in a process of developing hypothetical-deductive reasoning to answer a causal question which derives its meaning and purpose from a hypothetical cloning mission, and which also challenges the application of peers’ background declarative and procedural knowledge.

M. ERGAZAKI, V. ZOGZA

409

An overview of the analysis process The peer group discussion was tape-recorded, transcribed and segmented to message units, each expressing a single idea in possibly more than one linguistic clauses (Kelly et al., 1998; Mason, 1996). Sequences of message units were then identified on the basis of the different levels at which they were carried out. Thus, we identified sequences on the levels of: • Constructing a joint answer to the task (on-task sequences); • Re-establishing intersubjectivity as a necessary condition for an effective shift to the previous level (repair sequences) (Roth, 1995); • Evaluating the constructed answer or part of it (meta-sequences); • Gaining hints and/or clarifications from the teacher (teacher-help sequences). The segmented discussion was coded for argumentative and epistemic operations and finally analyzed on the levels of the argument constructions and hypothesistesting reasoning pattern development. The analytic tools Concerning the argumentative operations, we mainly drew on the framework proposed by Pontecorvo and Girardet (1993), and additionally on that proposed by Resnick et al. (1993). The derived coding scheme, summarized in Table 1, incorporates typical claims and justifications, as well as non-typical structural elements of the argument (oppositions, concessions, challenges). So, it is considered adequate for approaching both the individual and the social aspect of students' argument constructions, namely adequate for identifying the contribution of the ‘other’ to the construction of one’s own arguments. For example, a justification request carried out through the socially oriented operation of challenge may be a condition that leads to providing grounds for a spontaneously non-justified claim, thus to constructing a new argument. Similarly, opposition to a stated claim may be the trigger for more complex, sufficient, or persuasive justification structures on which to ground the claim. Furthermore, this scheme may highlight the dynamic process of argument construction, since – considering argument as any justified claim, concession, or opposition – it becomes possible to probe all intermediate arguments formulated towards the final answer. Following these arguments, we can also reconstruct the argumentative patterns employed in peers’ discourse as persuasive strategies. After considering several coding schemes proposed for the domain-bound epistemic operations (Pontecorvo et al., 1993; Mason, 1996; Jimenez et al., 2000), we constructed a new scheme that emerged to a large extent from our own data. The derived scheme, presented in detail elsewhere (Ergazaki & Zogza, 2002), makes it possible to identify the ‘micro’ cognitive procedures in which students are engaged in the specific context in order to construct their reasoning strands. In the case of the specific task, categories like ‘abduction’, ‘prediction’, or ‘interpretation of outcomes’ may permit us to follow peers’ hypothetical-deductive reasoning, while others, like ‘referring to’ experimental handling or to background knowledge, may be useful in outlining the peer design process which moves students towards the

410

FROM A CAUSAL QUESTION TO STATING AND TESTING HYPOTHESES

experimental test of their hypothesis. Finally, the several kinds of ‘appeals’ incorporated in the scheme, make it possible to define which criteria (i.e. experimental goals, background knowledge, authority) do count among students as warranting tools throughout the process of hypothesis formulation and testing in the context of genetic engineering. Table 1. The coding scheme for the argumentative operations ARGUMENTATIVE

DEFINITION

OPERATIONS Claim

Any clause stating a position without necessarily constituting an answer to the task

Justification

Any clause providing grounds to a standpoint

Concession

Any clause admitting a point claimed by another peer (confirming a claim or a justification)

Opposition

Any clause denying a point claimed by another peer (rejecting a claim or a justification)

Challenge

Any clause requesting either justification or inquiry of specific issues 3 . RESULTS

An overview of peers’ discourse From the outset of the discourse, Fani (F) seems to have figured out one plausible explanation for the failure of the cloned plant gene to synthesize the MRT protein in bacteria. Thus, employing a series of challenges, she facilitates her peers to explore key issues in the formulation of the hypothesis, ‘failure cause: no mRNA splicing in bacteria’, which is finally contributed by Vasso (V). Nevertheless, the third peer of the group, Elsa (E), doubts the plausibility of the hypothesis by appealing to the fact that known eucaryotic proteins are indeed synthesized in bacteria. F, insisting on her standpoint, attempts to propose a hypothesis testing procedure, but she only comes up with a procedure ‘testing’ the datum of the MRT synthesis failure in bacteria: ‘if the cancer cells of the culture continue their division after the bacterial MRT protein is added, then the MRT protein is not actually produced in bacteria’. The proposed experiment is evaluated in a meta-sequence as inappropriate for testing the hypothesis. This becomes, in turn, the focus of a series of adversarial exchanges among peers. So, E doubts once more the plausibility of the hypothesis, based on the fact of eucaryotic proteins’ synthesis in bacteria. This time V is persuaded, while F grounds

M. ERGAZAKI, V. ZOGZA

411

her opposition by appealing to the handling of cDNA cloning. The dialogue is shifted again to the meta-level, which allows for a better definition of the cloning goal apart from protein synthesis. After requesting the teacher’s clarification on whether the failure of protein synthesis in bacteria represents a broad problem solved through cDNA cloning, peers develop another test for the stated hypothesis. F suggests the elimination of the proposed cause in order to examine the deriving consequences on the cause’s potential effect. Thus, she produces the strand ‘if the MRT protein is indeed synthesized in bacteria by an mRNA cloned in bacteria but already spliced elsewhere, then the cause of the previous failure of its synthesis must have been the inability of the mRNA splicing in bacteria’. So, once more the validity of the hypothesis is predicted upon the experimental outcomes and not vice versa. Following the scaffolding device of the task, peers use this last experimental test to develop a hypothetical-deductive reasoning pattern. The resulting reasoning strand is: ‘if the hypothesis is right, then the synthesis of a functional MRT protein is expected’. This ‘if…then’ pattern is enriched with information about the testing procedure: ‘if the hypothesis is right and the test of cloning spliced mRNA in bacteria is conducted, then the propagation of the cancer cells in the culture will be inhibited’. It is also worth noticing that the group does not proceed to comparisons between expected and ‘observed’ outcomes to reach a final conclusion about the rejection of the hypothesis. The dialogue is ending while peers summarize their reasoning strand to the teacher. Argumentative and epistemic operations The results of our analysis on the level of the argumentative and epistemic operations indicate that peers are engaged in a highly argumentative discourse using a rich set of epistemic tools (Table 2). Apart from activating the operation of justification (29 times) to support (20) claims directly, (7) oppositions and (2) concessions, peers also employ higher-order justification structures for (5) claims and (1) concession. These structures may be either ‘subsequent’ or ‘complex’. Subsequent warrants (Kelly, 1998) consist of a whole argument (a justified claim, opposition, or concession) in support of a premise, while complex ones may include more than one argument or several combinations of arguments and claims, oppositions, or concessions in support of a premise. The number of explicitly unjustified operations in the discourse is similar to the number of operations that were justified in various ways (35 in each case, unjustified and justified), showing peers’ tendency to leave things implicit while carrying out discussions on common ground. In fact, the character of the unjustified operations may account for the latter's non-warrantability. Confirmatory concessions, claims implicitly grounded in shared knowledge, common sense, and given data; or predicative counter-oppositions against oppositions to already thoroughly justified premises, may indeed be excluded from the justification rule without undermining the discourse’s argumentative character.

412

FROM A CAUSAL QUESTION TO STATING AND TESTING HYPOTHESES

Table 2. The epistemic operations activated by the peer-group Epistemic operation Times of activation in the discourse Appeal to - experimental handling - background knowledge - task data - instances

(2) (3) (3) (1)

- experimental handling - background knowledge - experimental goal

(9) (4) (2)

Refer to

Abduction

(2)

Prediction

(5)

Interpretation of outcomes

(1)

Task reframing

(1)

Recognition of assumption

(1)

Definition of concepts

(1)

Evaluation

(12)

The process of stating a hypothesis by the peer group is facilitated through the argumentative operation of challenge, not as a direct request for justification but as a request for exploring specific key issues by applying background knowledge. So, the contribution of challenge to the process of ‘abduction’ concerns its use as a scaffolding tool mobilizing epistemic operations like ‘refer to/appeal to background knowledge’ about introns, cDNA, and mRNA splicing. On the other hand, the process of reaching consensus on the plausibility of the thus formulated hypothesis is actually hindered by the employment of complex warranting structures (each consisting of two claims justified by ‘appealing to’ background knowledge or to experimental goals) based on the ‘counterfactual strategy’ (Pontecorvo & Girardet, 1993). Undermining the stated hypothesis by considering its implications as contradictory with real facts, is attempted repeatedly in the discourse: ‘if we accept the hypothesis ‘cause of protein-synthesis failure: cloned gene’s introns’, then we must accept that there is no way of producing eucaryotic proteins in procaryotic cells / that it is not possible for the recombinant DNA technology to function / that we shouldn’t have been given instructions to carry out the cloning procedure up to

M. ERGAZAKI, V. ZOGZA

413

this point using DNA; which we know that it is not really the case. Thus, the hypothesis cannot be accepted’. Nevertheless, the oversimplified generalization behind the counterfactual mechanism of these warrants is rebutted with a direct justification through the epistemic operation of ‘appealing to’ the handling of carrying out the cloning procedure with cDNA instead of DNA. It is worth noticing that background knowledge on cDNA when creatively applied to the construction of a simple argument, not only resolves peers’ disagreement on the hypothesis plausibility and unblocks the discourse towards the proposal of an experimental test, but also provides peers with a framework for disconnecting the cloning goal from the protein synthesis through the epistemic operation ‘definition of concepts’. The argument remains a significant tool in the process of testing the shared hypothesis through the epistemic operations of ‘referring/appealing to experimental handlings’, ‘evaluating’, ‘predicting’, and ‘interpreting outcomes’. Specifically, the initial proposal is rejected with a direct justification ‘evaluating’ negatively its effectiveness in testing the hypothesis. The next attempts are shaped by subsequent warranting structures in support of experimental handlings, like having the mRNA spliced elsewhere or adding the new protein to a culture of cancer cells. So, the idea of transferring mRNA into a system where the splicing process will be possible is warranted by an argument stressing the opportunity to remove the factor that possibly causes failure in protein synthesis in order to test the effect on the resulting protein. Similarly, the idea of having the new protein added to a culture of cancer cells is supported by another argument based on the assumption of the protein’s anticancer function and the possibility of interpreting the outcome of this in regard to the hypothesis validity. Peers develop their hypothesis-testing reasoning pattern through the epistemic operations of ‘prediction’ and ‘interpretation of outcomes’: the group predicts – although not deductively – the expected outcome and subsequently interprets it as the hypothesis confirmation. Hypotheses’ testing reasoning patterns The causal relationship of variables A (cause: no splicing in bacteria) and B (effect: no plant-protein synthesis in bacteria), is explored through an experimental intervention on A (cloning a spliced gene) and a subsequent examination of its implications on B (what happens to the MRT protein). In other words, peers’ reasoning, while proposing a test to establish or not a potentially causal relationship, is shaped by exploring the co-variation of the possible cause A and its effect B. Coming to a conclusion about the validity of the tested hypothesis leads to a further elaboration: ‘if the proposed experimental handling which aims at eliminating the ‘possible cause A – no splicing’, alters the ‘possible effect B – no protein synthesis’, then it may be concluded that the A is indeed a valid cause for the effect B’. The character of peers’ spontaneous, hypothesis-testing reasoning pattern is confirmatory. In fact, peers are interested in confirming the hypothesis despite the rejection-oriented scaffolding device embedded in the task. This is rather fallacious, since the confirmation of a hypothesis might be claimed only in the case that the hypothesis remains unrejected after many appropriately designed tests (Lawson, 1995). Reasoning in a confirmatory context, peers encounter the following pitfall:

414

FROM A CAUSAL QUESTION TO STATING AND TESTING HYPOTHESES

they claim the hypothesis validity based on the experimental outcomes of the proposed test, instead of predicting the latter on the basis of the former. So, they substitute the deductive reasoning ‘if hypothesis X is valid, then the expected outcomes of the proposed test are x’ with the inductive reasoning, ‘if the proposed test for hypothesis X gives the outcomes x, then the hypothesis X is valid’. The pitfall in question may be associated with the invalid assumption that the relationship ‘cause-effect’, and consequently the relationship ‘hypothesis-predicted testing outcomes’, are unidirectional. Thus, it indicates that peers seem to ignore the fact that the same hypothesis may be the source of more than one (different) prediction, as well as the fact that the same prediction may derive from different hypotheses. A second assumption possibly associated with peers’ fallacies is the one of absolute trust in the appropriateness and reliability of the proposed test. In other words, peers seem not to consider the possibility of an experimental outcome deriving from a bad experimental handling or a procedural mistake and not necessarily from the validity of the tested hypothesis itself. Finally, when attempting to adapt their reasoning to the scaffolding-device, peers shift from the problematic inductive pattern to the deductive one, but they do not proceed to the required predictions of those outcomes or to a rejection of the hypothesis. Instead, they remain consistent with their confirmatory context, since they only follow the first steps of proposing a test and predicting its outcomes, while leaving out of focus the critical step of defining the conditions of hypothesis rejection. 4.

DISCUSSION

Our analysis indicates that peers, being engaged in a symmetrical and mainly adversarial interaction, produced an argumentative discussion based on a rich set of epistemic tools to accomplish their collaborative goal of stating and testing a hypothesis for the cloned gene’s failure to synthesize its protein in bacteria. Peers managed to formulate the appropriate hypothesis by applying systematically their background knowledge concerning the introns of eucaryotic genes and the resulting need of mRNA splicing, the absence of splicing mechanisms in procaryotic cells, and also the cDNA synthesis from spliced mRNA. Combining, elaborating, and synthesizing ideas in a group discussion to come up with a commonly accepted explanation is much more demanding for peers than reproducing a ready-made statement in a typical ‘triadic dialogue’ led by the teacher (Lemke, 1990). The role of argumentation in this process is two-fold, since it is employed as a reasoning tool and also as a persuasive one. This is quite clear in peers’ adversarial exchanges while attempting to reach consensus on the hypothesis plausibility before proceeding jointly to its experimental test. Complex warranting structures synthesizing a counterfactual strategy are employed as a persuasive device against the stated hypothesis, while one direct justification that invokes background knowledge is contributed in favor of it. Peers’ commitment to the proving power of ‘real facts’, which is expressed in the counterfactual persuasive strategy, loses ground when confronted with valid knowledge resources. Thus, argument seems to

M. ERGAZAKI, V. ZOGZA

415

be more than a rhetorical device in peers’ discourse towards a joint hypothesis. Furthermore, it is worth noticing the social character of argument construction that is stressed by the role that argumentative operations like challenge and opposition, as well as epistemic operations like evaluation, do play in the discourse. The hypothesis-testing reasoning constructed in peers’ argumentative discourse does not follow the hypothetical-deductive reasoning pattern. It seems that peers prefer inductive to deductive reasoning, as well as hypothesis confirmation to hypothesis rejection. Thus, they infer the hypothesis validity based on the outcomes of the proposed experimental test, when they are actually requested to do the opposite. Moreover, when finally adapting their reasoning to the task’s scaffolding device, they shift to the deductive pattern, but they are still focused on confirming the hypothesis, since they do not define the required hypothesis-rejection conditions. This invalid preference needs to be explicitly stressed by the teacher so that peers can recognize this pitfall in hypothesis testing. Grasping the need to deductively predict instead of inductively infer is a significant task, since it might save students from fallacies, such as making hasty generalizations, when reasoning in scientific and in everyday contexts. It is also associated with understanding the key idea that confirming a hypothesis is a much more demanding task than rejecting one, since confirmation requires coming up with a series of different predictions based on the potential validity of the hypothesis as well as thinking of a series of appropriate experimental tests. Finally, a purposeful teaching goal might be to support peers in recognizing that, despite the multiple predictions and tests confirming a hypothesis, there is always the possibility that the next prediction’s test will break the confirming sequence irreversibly, dictating, therefore, rejection of the hypothesis. In summary, the implications of the differences between peers’ hypothesis-testing reasoning and the hypothetical-deductive pattern, may serve as a fruitful basis for improving peers’ hypothesis-testing skills. ACKNOWLEDGEMENTS The authors would like to thank G. Dimitriadis, Professor of Molecular Biology, and his students for their crucial contribution to this work. REFERENCES Brown, J. S., Collins, A., Duguid, P. (1989). Situated cognition and the culture of learning. Educational Researcher,18(1), 32-42. Desautels, J. & Larochelle, M. (1999). High school students' construal of socioethical issues in scientific controversies: An apercu. Paper presented at the 1999 annual meeting of the American Educational Research Association, Montreal, Quebec, April 1999 Driver R., Newton, P., Osborne, J. (2000). Establishing the norms of scientific argumentation in classrooms. Science Education, 84(3), 287 - 312. Ergazaki, M. &Zogza, V. (2002). Students’ reasoning while collaborating on a task in the context of genetic engineering. In Lewis J., Magro A., Simmoneaux L. (Eds.) Biology Education For The Real World (pp.171-183). Toulouse, France: Enfa

416

FROM A CAUSAL QUESTION TO STATING AND TESTING HYPOTHESES

Giere, R.N. (1991). Understanding Scientific Reasoning (3rd ed.). Forth Worth, TX: Holt, Rinehart & Winston. Jimenez-Aleixandre, M.P., Rodriguez, A.B. & Duschl, R.A. (2000).“Doing the Lesson” or “Doing Science”: Argument in High School Genetics. Science Education 84(6), 757-792. Kelly, G., Druker, S. Chen, C. (1998). Students’ reasoning about electricity: combining performance assessments with argumentation analysis. International Journal of Science Education, 20 (7), 849-871. Kuhn, D. (1993). Science as argument : Implications for teaching and learning scientific thinking. Science Education 77(3), 319-337. Lawson, A.E. (1995) Scientific thinking and the development of thinking. Wadsworth publishing company, Belmont, California. Lemke, J. (1990). Talking science: language, learning, and values. Norwood, New Jersey: Alex Publishing Corporation. Mason, L. (1996). An analysis of children construction of new knowledge through their use of reasoning and arguing in classrooms discussions. Qualitative studies in Education, 9(4), 411-433.. Pontecorvo, C. & Girardet, H. (1993). Arguing & reasoning in understanding historical topics. Cognition and Instruction, 11(3&4), 365-395. Resnick, L., Salmon, M., Zeitz, C., Wathen, S.H. & Holowchak, M. (1993). Reasoning in Conversation. Cognition and Instruction, 11(3&4), 347-364. Roth, W. – M. (1995). Authentic School Science – Knowing and Learning in Open-Inquiry Science Laboratories. Dordrecht, Netherlands: Kluwer Academic Publishers. Simonneaux, L. (2000). Comparison of the impact of a role-play and a conventional debate on pupils’ arguments on an issue in animal transgenesis. In: Gayoso, Bustamente, Harms, Jimenez-Aleixandre (Eds.) Proceedings of the III Conference of European Researchers in Didactic of Biology (pp. 291-311). Santiago de Compostella, Spain: Universidade De Santiago De Compostella Publicacions. Vygotsy, L.S. (1978). Mind in Society: The development of higher psychological processes. In M. Cole et al. (Eds. & Trans.). Cambridge, MA: Harvard University Press.

APPENDIX You have already accomplished the goal of completing successfully the cloning procedure of the MRT plant gene in bacterial cells. Nevertheless, your co-researchers have gained promising results from the continuing preliminary studies on the potential anti-cancer effect of MRT protein, and thus your team is assigned to the task of synthesizing this protein massively in bacterial cells. So, continuing to grow the cell culture having the MRT gene, you are surprised to realize that the MRT protein is not in fact synthesized in the bacterial culture. •

•

How could you explain this observation? Why is the MRT protein not produced in the bacteria by the cloned MRT gene? Can you give a tentative explanation for this? In other words, can you formulate a so-called hypothesis? To develop a testing-reasoning for your hypothesis, follow the next steps:

M. ERGAZAKI, V. ZOGZA

417

1. Propose an experimental procedure: how could you experimentally explore the validity of your hypothesis? 2. What are the outcomes of the proposed test that would show you that your hypothesis is probably right? 3. What are the outcomes of the proposed test that would show you that your hypothesis is probably wrong?

ARGUMENT CONSTRUCTION AND CHANGE WHILE WORKING ON A REAL ENVIRONMENT PROBLEM MARÍA PILAR JIMÉNEZ-ALEIXANDRE¹, CRISTINA PEREIRO-MUÑOZ² ¹University of Santiago de Compostela, Spain ²High School Castelao, Spain

ABSTRACT The process of collaborative construction of arguments about environmental management by 11th grade students working in small groups is studied. The question explored is the evolution of the students’ positions and arguments along a sequence shaped around an authentic – and real – problem: the impact of a drainpipe in a wetland of high ecological value; whether students kept their initial positions or changed them and the corresponding reasons. The collaborative construction is explored in terms of the dialogic voice (Mortimer & Scott, 2003). The participants were the 37 students in an 11th grade group and their teacher (the second author). The sessions were recorded in audio and video, and the data also include the students’ portfolios and essays. In this paper the transcriptions are analysed and the arguments represented using Toulmin’s (1958) layout. The analysis shows changes in the positions of 22 students, either radical, from positive to negative assessment, or shifts to balanced views. The causes for the changes and the co-construction of arguments are also discussed.

1. DIALOGIC ARGUMENTS: COMMUNICATION AND ARGUMENTATION IN SOCIO-SCIENTIFIC ISSUES The relevance of studying the communication system in the classroom, the classroom discourse, in order to understand learning processes is being acknowledged in educational research. Studies exploring processes in actual classrooms, as meaning construction by students, expand the knowledge gained with work about students’ ideas. One of the objectives of such studies is to make visible or “external” those processes that by nature are internal (as proposed in the cognitive apprenticeship perspective by Collins, Brown & Newman, 1989) and to make cognitive and metacognitive processes accessible through discourse analysis. The importance of communication has been highlighted by the sociocultural perspective which acknowledges the relationships among mental processes and the cultural context (Wertsch, 1991), following Vygotski who drew attention to the role of social interaction in cognitive development. Our study is framed in educational constructivism, viewing students as active participants in learning, and in the 419 K. Boersma et al. (eds.), Research and the Quality of Science Education, 419—431. © 2005 Springer. Printed in the Netherlands.

420

ARGUMENT CONSTRUCTION AND CHANGE

Science-Technology-Society (STS) perspective: scientific issues are viewed in a broader social context. We explore decision-making by students that is not only a technical issue (Aikenhead, 1985), but also requires consideration of values and social consequences. One of the classroom (and science) coexisting discourses is argumentation. By argumentation we mean the communication and evaluation of knowledge claims, the justification of claims by appeals to data, and the strategies for resolving issues opposing alternative positions. Like Toulmin (1958), we are interested in studying argumentation in natural conversation which is different from ideal argumentation in Logic. The focus is the dialogic argument in a context where students engage in decision making about a socio-scientific issue, environmental management in a wetland of high ecological value. Argumentation studies explore epistemic dimensions, processes in knowledge construction, and co-construction. By dialogic argument we understand: – Arguments are co-constructed in a collaborative effort among two or more participants. – Arguments are produced by one person, but take into account other participants’ statements, either to support or to contradict them. This means that we consider an argument dialogic even when a single speaker produces it, using the notion of Mortimer & Scott (2003) of dialogic voice as one also taking into account the listeners’ perspective. Driver, Newton, and Osborne (2000) also refer to dialogical or “multivoiced” arguments in social groups. The capacity to take different perspectives into consideration is thus considered as a criterion for argument quality. A favourable learning context for argumentation is solving authentic problems which have, among others, these features: they are relevant for the lives of the students; they involve methods related to scientific work, and they are ill-structured, having no straightforward “right” solution. The problem used for this teaching sequence is not only authentic but also real, and students could follow the controversy in the media, simultaneous with their debate. The question explored is the evolution of students’ positions and arguments along a sequence shaped around the impact of a drainpipe in a wetland, a problem chosen for its complex, controversial nature. In a previous paper (JiménezAleixandre & Pereiro, 2002), we analysed the content knowledge components in the warrants; here the focuses are the changes, in claims or in justifications, of individual and group arguments collaboratively constructed. The capacity of changing a position is important as it seems that exists a resistance to change (in students as well as in adults), and people tend to take into account mainly evidence which supports their opinion (see, for instance, Kuhn et al., 1995). The research question is: Do students’ arguments on environmental management evolve during the sequence? The question has been split into two: • Do students’ positions and arguments (as individuals or in groups) change along the sequence? • What are the causes of change, data or other people's arguments?

M.P. JIMÉNEZ-ALEIXANDRE, C. PEREIRO-MUĕOZ

421

2. METHODS, PARTICIPANTS AND EDUCATIONAL CONTEXT The study is framed in action research with the second author acting as a teacherresearcher in her classroom. It is research about education and for education with the teacher participating in the construction of knowledge by reflection on her practice. The context is a Biology and Geology course taught by its regular teacher. All the students (N = 37) of an 11th grade group from the evening shift (17–21 years) participated in the study. Students who enrol in the evening shift either work or are low achievers, thus explaining the age range which is higher than the standard 16-17 years. The sequence went on for 17 sessions in February/March 1998; it also included a field trip. The composition of groups changed twice, following the Jigsaw technique: in sessions 1 to 5 they were named Group I, Group II and so on; from session 6 to 12 “monograph groups” A to F (discussing plants, animals, landscape, technical features of the project, etc.); and from session 13 to 17 Jigsaw groups, J1 to J6, which put together information from monographs and wrote the reports. This means that it is not possible to follow the evolution in each group but only in the whole class or in individuals. The task: environmental management Students were asked to produce group reports assessing the suitability (or not) of a network of drainpipes crossing a wetland and collecting sewage. The wetland in question belonged to the basin of a river and contained a seasonal pond; it has the European status of “Area of Natural Interest”. Since it is close to an industrial area and a granite quarry, both of which drain their sewage into it, the wetland is heavily polluted. The small ponds are dry, and plants (including some rare insectivorous plants such as Drosera) and animals (for example, amphibian and migratory birds) are suffering the consequences of habitat destruction. The management plan involves a network of drainpipes that cross the wetland with the objective of collecting all industrial and domestic sewage. In the first draft of the project the main drainpipe was to surround (be outside) the wetland and the pond, but in the final project the layout was changed, and the drainpipe then crossed the wetland dividing it in two. The reason for the change was to reduce the cost. The students had to justify their positive or negative assessment and, if negative, offer an alternative proposal. The construction of the drainpipe had both positive (cleaning the polluted river and marshes) and negative (destruction of fragile habitats) outcomes, so a straightforward positive or negative assessment was out of the question. This dual nature was one of the reasons why we chose this problem, in the expectation that it would promote various positions, argumentation, and debate among students. Data collection and analysis Data collection involved recording the sessions in audio and video; participation of an external observer; collecting students’ portfolios and essays, and conducting interviews. Students’ conversations were transcribed and analysed using several

422

ARGUMENT CONSTRUCTION AND CHANGE

tools. For the purpose of this paper, Toulmin’s (1958) argument layout, which has been employed to analyse and represent the arguments, is relevant. The argument components, following Toulmin, are: a) data; b) claim; c) warrants, that is, reasons that justify the connection between data and claim; and d) backing (background knowledge) of a theoretical or general character to which warrants are related. Sometimes there are also modal qualifiers and rebuttal.

3. RESULTS: EVOLUTION OF STUDENTS’ ARGUMENTS The analysis focuses on the evolution of students’ individual arguments. We consider two levels of change: Change: involving a complete switch, either from a positive assessment to a negative one, or the reverse. In terms of argument components, it is a change in the claim. Evolution: not involving a reversal of position, but an elaboration about the assessment, for instance recognising the positive effects of the drainpipe in reducing pollution in the area in the context of a negative assessment, or acknowledging its negative effects in the ecosystem in the context of a positive one. In terms of argument components, it is a change in the warrants (justifications) and/or in the qualifiers. As sources of evidences for change we consider: 1) The transcriptions of students’ discussions in small groups or whole class debates (T). 2) The answers to an item included in the examination paper at the end of the term, “Explain the changes in your ideas about the construction of the drainpipe from the day when you first read the letter [the first activity in the sequence] to today, and which were the data relevant to your decision” (Ep). 3) The essays submitted after the examinations in which students were asked to give their opinions about the sequence and their work in small groups, and were not asked directly about changes (Es). The results for both levels of change are summarized in Table 1. At the beginning of the term there were 38 students, but one of them dropped out. All names are pseudonyms, but respect students' gender. The first column in Table 1 represents the different degrees of change; the second, the final assessment of each group expressed in the written report and in session 17. The shifts in students’ positions are summarized in column 3, distinguishing among (most) students who changed their views during the sequence and three students from group J3 (marked with *) who did it as a result of a debate with experts that took place after the reports were handled; the case of Bruno is discussed in detail in Jiménez, Pereiro & Aznar (2000). Column 4 summarizes the sources of data documenting the change: examination papers, individual essays, or transcriptions; and the fifth, the causes (discussed in the next section).

M.P. JIMÉNEZ-ALEIXANDRE, C. PEREIRO-MUĕOZ

423

Twenty-two of the 37 students, belonging to all the groups excepting J6, acknowledged a change in their positions.

Table 1 Changes in students’ positions. Legend: a) Assessment: N= negative; Nb = negative balanced; P = positive; Pb= Positive balanced; * change after report. b) Sources of data: T= Transcripts, Ep= Examination Paper, Es = Essays. c) Causes of changes: D= data, E = Experts opinion, C= Clarification with peers mentioned by students, (CT) Clarification evidenced in transcription, although not mentioned by students Level of Groups Students Sources Causes change assessment assessment J1 Dora P →N J1 Fito P →N J2 Anxos P J2 Negative: →N divert pipes to the J2 Cruz P West →N J2 Damián P →N J2 Flor P →N

Ep Ep

D: ecosystem damage (CT) + D: consequences

Ep Ep T + Ep Ep

(CT) + E + D: no recovery (CT) + D: field trip (CT) + E+D: irreplaceable sp D: interest of turf

J3 Bruno P J3 Positive: the → N* layout is adequate J3 Carlos P → N* J3 Fermín P → N*

T +Ep +Es Ep Ep + Es

(CT) + E + D: no recovery (CT) + D: impact ecosystem (CT) + E: debate with experts

J1 Negative: purifying plants close to factories

Change: N = 16

J4 Aldán P Ep Ep J4 Negative: a →N different J4 Fabián P Ep trajectory (layout) → N J4 Fran P →N

E + D: slides ecosystem D: alternative layout E + D: recovery time

424 Level of change

ARGUMENT CONSTRUCTION AND CHANGE Groups assessment

Students assessment

J5 Ana J5 Negative: a →N different purifying J5 Clara system →N J5 Denís →N J5 Fuco →N J1 J2

Evolution: N=6

No change: N = 15

P P P

Sources Causes

Ep Ep T +Ep +Es Ep

(CT) + D: dimensions pipes (CT) + D: site natural interest (CT) + E + D: turf, river (CT) + D: no recovery plan

T + Es Ep Ep + Es Ep + Es T T

C + D: layout (CT) + D: layout, turf C + D: Aesthetic values E + D: status area (CT) + E + D: alternatives C: alternative: not polluting

P

J1 Alfonso N → Nb J2 Begoña J3 N → Nb J3 Antia P J6 Negative: a → Pb different J3 Eva P trajectory (layout) → Pb J6 Antón N → Nb J6 Enrique N → Nb J1 3 students J2 1 student J3 2 students J4 4 students J5 1 student J6 4 students

Of these, 16 recognised that they initially thought that the network of pipes was to have a positive impact on the wetland, but they later changed this view to a negative assessment of the current project (which did not necessarily mean a rejection of the cleaning plan, but in most cases an objection to the pipe’s trajectory). None of the students acknowledging a reversal of their view changed from a negative assessment to a positive one. The six students, who evolved, modified a relatively simple acceptance or rejection to a more balanced view and included more qualifiers in their statements; four modified their negative assessment recognising the need for a project or discussing the benefits it would involve. The two students who modified their positive assessment expressed some caveats about its negative impact. Fifteen students, at least apparently, did not change their position. Concerning the data sources, all students involved in radical reversals acknowledged the change in the examination paper, and three also did so in the

M.P. JIMÉNEZ-ALEIXANDRE, C. PEREIRO-MUĕOZ

425

individual essay. As seen in Table 1, most of these changes cannot be clearly traced in the transcriptions because many students stated only their final position. Also not all of the students’ thoughts or reasoning were voiced, which points to the need for complementary data sources. In six cases the transcripts show students defending different views at different times. The data for students who shifted positions without reversing them range from only transcriptions or papers to a combination of two data sources. Excerpts from transcriptions, examination papers, and essays are reproduced here to illustrate the analysis. Change: Anxos, J2 (P to N) “In the beginning I thought that building the drainpipes would be good, but then I realised that it could seriously damage the ecosystem. The president of the ecologist association convinced me of the serious problems that the drainpipes network could produce: the trees removed to build it are not to be replaced... the pond and its surroundings would not recover its regular situation” (Examination paper). Damián, J2 (P to N) Transcription session 14, group J2 12 Anxos: The drainpipes. And: What do you think? It is positive or negative? 13 Begoña: Negative. 14 Damián: Negative? The drainpipes are positive. In the examination paper, Damián gives as a cause for his change that irreplaceable species would disappear. Fran, J4 (P to N) “When I read the letter I was in favour of the drainpipes being built in any place, because I did not know the environmental impact that it was going to produce; now I have another idea, I would like the drainpipe being built, but in a place where not so much nature would need to be destroyed. The relevant data were: – the extended recovery time for the site – the budget difference among the projected layout and the alternative one is not so wide – the species coming here for the summer wouldn’t come back – the devastation of plants and animals.” (Examination paper) Denís J5 (P to N) Transcription session 14, group J5 1 Ana: Here, the problems... we have to say whether we would built it [the drainpipe] or we wouldn’t ... 2 Belén: ... and why 6 Belén: The drainpipe, that it would be better not to build it 7 Denís: Not at all. The drainpipe must be built. (...) 423 Denís: I, too, was completely sure... 425 Denís: ... that, being in doubt, it should be built, but ...

426

ARGUMENT CONSTRUCTION AND CHANGE (Later, Denís gives reasons for not building the drainpipes with the projected layout: damages to birds, pollution from works, difficulties for recovery).

Evolution: Antia, J3 (positive to positive with qualifiers) “It (the drainpipe) would be good, because it would eliminate the pollution in the pond, but it does not reduce acoustic pollution and could even increase it. The pipes could even carry the pollution to other areas now clean that are also very important. Like river Miño, for it will flow into river Miño. (...) If we had to give a vote, my vote would be blank, given that on the one hand I agree, because the water in river Louro are very polluted, but on the other I wouldn’t like the marshes to disappear, because of the species and because they are so beautiful. So my opinion would be that they change a little the layout of the drainpipe.” (Individual essay) It is not easy to distinguish this “positive with qualifiers” assessment – with students saying on the one hand that it was good, but that it would be better to change the layout a little – from other “negative with qualifiers”, like the one reproduced from Fran, advocating an alternative layout. We see the positions as part of a continuum, rather than representing dichotomous views. So, in summary it can be said that in a majority of cases, students’ positions evolved to views of greater complexity. 4. RESULTS: CAUSES OF CHANGE The causes for change are summarized in the last column in Table 1. In the 16 cases where there was a change, students mentioned reasons for it; whereas in the 6 cases of evolution, causes were mentioned, but acknowledgement of the modification of position was not always made. Two types of causes of change were more frequently mentioned: new data (D), found in the documentation students were studying, in the field trip or provided by experts, and arguments from the experts’ debate (E). Data most frequently mentioned relate to the impact on the ecosystem, the status of natural interest of the site, or to some of its areas: turf, pond, trees, and the damage to the 140 species living (or migrating) there. Technical features of the project were also mentioned, such as the dimensions (e.g. the two metre diameter of the pipes), the difference between the initial and current budget, or the lack of recovery measures. In three cases during discussions with their peers, students mentioned not just the data, but their clarification (C) or weight, particularly concerning the importance of the ecosystem and the damage that building the drainpipe would produce. The transcriptions also show instances of clarification: students asking one another or offering an interpretation of information (CT). These clarifications occupied a substantial part of the debates in groups J2, J3 and J5. As the

M.P. JIMÉNEZ-ALEIXANDRE, C. PEREIRO-MUĕOZ

427

transcriptions show evidence of the positions of different members (and even of different teams) converging along a sequence, we believe that this cause was at least relevant in these three groups, although most students did not explicitly recognise it. Transcription session 15, group J2 136 Damián: Here it says 36,500 meter... That’s equivalent to ... 36 kilometre. 137 Cruz: Where? 140 Damián: You... you do not understand the units (conversion), now. 141 Cruz: There should be a comma 161 Cruz: ...it is really huge... 166 Anxos: All that in green (in the map)... that’s the drainpipe. 167 Begoña: All that green? 169 Damián: ...and about the width: How much? 175 Cruz: ... They have three hundred sixty millimetre... These above are fragments of a long discussion in group J2 about the longitude and width (in fact up to 2 metres diameter) of the pipes. In session 16 Damián asked for clarification with the map before deciding a position: Transcription session 13, group J3 83 Bruno: The drainpipes are a kind of... Does someone have it? (...) What are the drainpipes? 84 Antia: The drainpipes are... all this (in the map). All... the pipes together (...) the industries clean the water, then the water goes to the pipes The clarification goes on, with the interpretation of the different colour codes in the map. In session 15 the students discuss the meaning of the recovery measures: Transcription session 14, group J5 234 Denís: How do the drainpipes work? 235 Belén: I have no idea, ask her (Ana) 241 Ana: ..if the water is polluted, they close the lock gates... don’t let them go out. And then it goes to the pipes... it is a canal (...). A pipe, a pipe that takes the water that they want to divert to river Miño. Ana explains to Denís how the drainage and cleaning network will operate. In the next session they summarize again their discussion for Fuco, who was not in session 14. The arguments from external experts were expounded during a debate that the students held with the engineer who authored the project, and the president of the environmentalist group Erva (“grass” in Portuguese) who was the author of several pleas against the second layout. This debate took place after the sequence was completed and the reports written and reviewed, but before the examination paper and the final essay.

428

ARGUMENT CONSTRUCTION AND CHANGE

In summary, it can be said that the causes for these modifications relate to greater knowledge, more information, new data, but also to how these data were interpreted in the course of discursive interactions with other students or with the two external experts. This points to the question of whether the construction of arguments and its change was primarily an individual issue, or was it achieved as the result of a dialogic interaction. 5. DIALOGIC VOICE AND CO-CONSTRUCTED ARGUMENTS The results summarized in the two previous sections refer to individuals, because the redistribution of students in the teams makes it impossible to assess the change in each group. However, this may lead to beliefs such as, students’ arguments were individually constructed, or they achieved the change in the same individual way. On the contrary, the transcriptions of the six groups’ debates, particularly for groups J2, J3 and J5, point to collaborative construction; three students from groups J1, J3 and J6 even mentioned in their examination papers and essays, the importance of data clarification (or “weighing” as they referred to it) with their peers. The arguments could be considered dialogic in two cases (that may occur during the same discussion): (1) arguments collaboratively constructed among two or more participants in the group, and (2) arguments that take into account other participants’ positions, either to support or to refute them. Concerning the second case, the dialogic voice (as opposed to arguments not having a relationship to other participants’ statements) is apparent in the whole class discussion, more than within groups, as the hottest debates took place in a whole class setting. However, even in the small groups, most interactions were dialogic; an evolution was observed from the first sessions, when some no-dialogic interactions were occurring (for instance in group A) and the students were “running in parallel”, to the last sessions, when they listened to each other. Concerning the first case, it can be said that all arguments in the six groups were co-constructed, with participation of several persons in each group. As an instance, Figure 1 represents, in Toulmin’s argument layout (1958), the position of group J2 in session 16 at the end of the sequence. This argument has been co-constructed, with significant inputs (noted in the figure) from five of the six students. Thus, it is dialogical, and evidences an understanding of the complexity of the issue by using warrants both ecological and technical, one of them from the subsequent type (Kelly, Drucker & Chen, 1998), that is, complex arguments whose claims are in turn warrants for other arguments.

M.P. JIMÉNEZ-ALEIXANDRE, C. PEREIRO-MUĕOZ

429

Qualifier It is more expensive Enma, 157, Bego–a 173

Data ŠLayout ŠRiver Pollution ŠBuried pipes

Claim Should have a different layout

Warrant 2 (ecological) Claim This alternative could be ecologically more profitable in the long run

Warrant 1 (technical) Š(there is another layout) Šbetter divert it to the West

Anxos, 103, Bego–a 150

Backing (data source) ŠThat about the West is in that work...

Enma, 157, Bego–a 173, Dami‡n, 174

Warrant 2.1 (ecological) ŠNot only damages birds, it destroys the ecosystem

Warrant 2.2 situation Šthis layout crosses all the ecosystem Bego–a, 150

Cruz, 160

Cruz, 106

Figure 1 Co-constructed argument from group J2, session 16 6. CONCLUSIONS AND EDUCATIONAL IMPLICATIONS This paper focuses on the changes in claims, justifications, or qualifiers of individual and group arguments. The analysis shows that the positions of 22 students evolved along the sequence, in some cases shifting from a positive to a negative assessment, and in other cases refining the proposal from a simple negative to suggesting a precise new location of the pipes, increasing the complexity of their position, and/or balancing both positive and negative aspects of solutions (a criterion for assessing environmental reasoning, Hogan, 2002). The capacity of changing a position shows a development of reasoning skills among students. Concerning the causes of change, some of the students mentioned being confronted with data; others refer to the influence of experts’ arguments. Data clarification and instances of clarification were also mentioned in the transcription which, in our view, point to the importance of data interpretation in discursive

430

ARGUMENT CONSTRUCTION AND CHANGE

interactions with peers. The discussions and essays show how they strive to support their evaluation in the available data. Most of the interactions were dialogical, particularly in the last sessions, and the arguments from the six groups were collaboratively constructed, with participation of two, three, or more students. In our perspective, solving authentic problems offers favourable learning contexts for argumentation. The point that the student group studied belonged to the evening shift and that in academic terms these students were considered low achievers, could reinforce the suggestion that, given an appropriate task and classroom climate, secondary school students can engage in socio-scientific argumentation. The problem used for this teaching sequence was a real one, and students could see its reflection in the media. Since it was a case study, generalisation is not possible, but we suggest that the nature of the problem promoted high involvement of the students, pointing to the interest of using authentic – and, when possible, real – problems of social relevance in the classroom. ACKNOWLEDGEMENTS This work was supported by the Spanish MCYT, grant BSO2002-04073-C02-02, and was partly funded with European FEDER funding. REFERENCES Aikenhead, G. S. (1985) Collective Decision Making in the Social Context of Science. Science Education, 69: 453-475. Collins, A., Brown, J.S. & Newman, S. E (1989) Cognitive Apprenticeship: Teaching the Crafts of Reading, Writing and Mathematics. In L. Resnick (Ed.) Knowing, Learning and Instruction. Essays in Honor of Robert Glaser. (pp 453-494). Hillsdale, NJ: Lawrence Erlbaum,. Driver, R., Newton, P. & Osborne, J. (2000) Establishing the Norms of Scientific Argumentation in Classrooms. Science Education, 84: 287-312. Hogan, K. (2002) Small Groups’ Ecological Reasoning while making an Environmental Management decision. Journal of Research in Science Teaching, 39, 341-368. Jiménez-Aleixandre, M.P. & Pereiro-Muñoz, C. (2002) Knowledge producers or knowledge consumers? Argumentation and decision making about environmental management. International Journal of Science Education, 24, 1171-1190. Jiménez-Aleixandre, M.P. & Pereiro-Muñoz, C. & Aznar Cuadrado, V. (2000) Reasoning on Environmental Issues: an Empirical Study about Environmental Management in the 11th Grade. In H. Bayrhuber & J. Mayer (Eds.) Empirical Research on Environmental Management in Europe. (pp 67-75). Münster: Waxmann. Kelly, G. J., Drucker, S. & Chen, K. (1998) Students’ reasoning about electricity: Combining performance assessment with argumentation analysis. International Journal of Science Education, 20, 849-871. Kuhn, D., García-Milá, M., Zohar, A. & Andersen, C. (1995) Strategies of knowledge acquisition. Monographs of the Society for Research in Child Development. Chicago: University of Chicago Press.

M.P. JIMÉNEZ-ALEIXANDRE, C. PEREIRO-MUĕOZ

431

Mortimer, E.F. & Scott, P. H. (2003) Meaning Making in Secondary Science Classrooms. Maidenhead: Open University Press. Toulmin, S. (1958) The uses of argument. Cambridge: Cambridge University Press. Wertsch, J. (1991) Voices of the mind: a sociocultural approach to mediated action. Cambridge, MA: Harvard University Press.

PART 8 Teaching and learning scientific concepts

TEXTBOOKS AND THEIR AUTHORS: ANOTHER PERSPECTIVE ON THE DIFFICULTIES OF TEACHING AND LEARNING ELECTRICITY

RICHARD GUNSTONE, BRIAN MCKITTRICK, PAMELA MULHALL Monash University, Australia

ABSTRACT Despite extensive research, our understanding of the teaching and learning of direct current (DC) electricity remains poor. As part of a larger project focused on learning outcomes and analogies/models/metaphors appropriate at different levels of electricity learning, in this study we investigated the detailed forms of explanations and analogies/models/metaphors used in senior high school textbooks in Victoria, Australia, and the understandings of the writers of these textbooks. All 3 authors had inadequate understandings of models and analogies, there was great variation in author understanding of voltage (with one author having clearly inadequate understanding), and the approaches authors used in their books reflected these inadequacies. We suggest that this serious issue is not specific to the state of Victoria.

1. INTRODUCTION Many studies of students’ conceptions in the last 25 years have focused on introductory direct current (DC) electricity. These have involved young children (e.g. Cosgrove et al., 1985), high school physics students (e.g. Koumaras et al., 1997), university physics students (e.g. Viennot & Rainson, 1992), and science/physics teachers (e.g. Pordhan, & Bano, 2001). Recently we argued two reasons for this strong research concentration (Mulhall et al., 2001), [Edit: shifted reference slightly.] reasons initially stated in 1985 (Duit et al. 1985, p. 10) and still valid today. The first is that electricity in some form is seen as central in physics/science curricula at all levels of education. The second, central to the project from which this paper derives, is that the concepts of electricity are particularly difficult – these concepts are abstract and complex in ways that make their understanding both centrally dependent on analogies and metaphors and frequently problematic. Despite much research, on both student conceptions and approaches intended to change conceptions, our understandings of ways to develop concepts such as ‘voltage’ remain disappointingly poor. Two significant aspects of this poor progress 435 K. Boersma et al. (eds.), Research and the Quality of Science Education, 435—445. © 2005 Springer. Printed in the Netherlands.

436

DIFFICULTIES OF TEACHING AND LEARNING ELECTRICITY

are a lack of agreement about both detailed learning outcomes appropriate for different student ages/levels, and analogies/models/metaphors appropriate for different ages/levels – a stark contrast with another highly abstract area of science, particle theory (Mulhall et al., 2001, p. 583). We are currently seeking to establish informed consensus about appropriate different forms of learning outcomes and analogies/models/metaphors for different levels of students studying DC electricity. However, fundamental to such informed consensus is that those from whom the consensus is sought have good understandings of the concepts about which they are giving opinions. We have argued in some detail (Mulhall et al., 2001) the considerable difficulty of DC concepts per se. Physics textbooks (and their authors) have considerable influence here – these books shape the ways many students and teachers conceptualize these ideas. Preliminary analysis of some popular textbooks has revealed conceptual inadequacies (Mulhall et al., 2001). The purpose of the present study was to explore author and textbook understandings in greater detail. The research question underpinning the study is: How are the central concepts of direct current (DC) electricity understood by authors of senior high school physics textbooks and represented by these textbooks? 2. THE CONTEXT OF THE STUDY Our sample of books and authors is from the context in which we work, the Australian state of Victoria. Here the final years of school (Years 11 and 12) include a specialist physics course, with DC electricity in Year 11, as shown in Figure 1. This electricity content is not unusual for such a course. CENTRAL IDEAS The use of electricity underpins much of the structure of our lives. Safe and effective use of electricity is important for individuals and the community generally. Much of our present use can be explained by basic DC circuit theory. The area of study should include: • current, charge, voltage, energy and power in series and parallel circuits (including Q=It, W=VIt, P=VI); • Ohm’s law and resistance in series and parallel circuits (V-I graphs for ohmic conductors; RT=R!+R2 etc. • non-ohmic devices (awareness of their existence; examples of some V-I graphs); • cells, batteries and power supplies (including e.m.f. and internal resistance); • electric shocks (descriptive treatment of effects on humans, awareness of approximate dangerous quantities.

Figure 1: The DC electricity component of the curriculum in the context of the research

R. GUNSTONE, B. MCKITTRICK, P. MULHALL

437

Figure 1 also shows that the curriculum has a focus on context-based learning, in the manner of PLON (e.g. Eijkelhof & Kortland, 1988), although this has diminished during the decade of use of the curriculum (Hart, 2002). Two significant issues are not obvious from Figure 1. First, the absence of ‘electric field’ (which we see as most unfortunate) is more than an oversight – the evolution of the course indicates that ‘electric field’ has been specifically excluded. Second, and again evident from the evolution of the course, there has been for some time a general concern among many high school physics teachers that ‘voltage’ be seen in terms of difference (and indeed for a time there was debate about excluding the use of the symbol V and instead using ∆V in the course description). 3. THE CONTEXT OF THE STUDY – THE PROBLEMATIC NATURE OF ‘VOLTAGE’ In this paper we use the term ‘voltage’. This term is explicitly used in the course description for which the authors involved in this study were writing (see Figure 1), and is therefore used in their books. Hence, it has been necessary for us to use the term in our interviews and as one of our foci in the textbook analyses. However, our use in this study of ‘voltage’ is only because of this context. We regard ‘voltage’ as, at best, an unhelpful term in the (English language) study of DC electricity. We now outline our reasons for this. There is no clear agreed meaning for the term ‘voltage.’ We see unresolved issues in the multiple and variable ways that it is used in some textbooks, e.g. is ‘voltage’ concerned with energy or energy difference? with energy source or sink? with energy aspects associated with a particular point or difference between two points? We find it significant that many iconic English language physics textbooks do not make any mention of ‘voltage’ in the book index (e.g. the first 5 editions of PSSC Physics, Harvard Project Physics, and the 2001 edition of Halliday, Resnick & Walker). Our view is that the term ‘voltage’ is essentially a physics slang expression, derived from the word ‘volts’ and meaning ‘how many volts’. ‘Voltage’ has little conceptual status other than its association with ‘volts’. Our preference is to avoid use of the term altogether because its use cannot aid the development of understanding in electricity. However, as already indicated, this is not possible in the context of this research. 4. THE APPROACH OF THE STUDY Many different textbooks are used for Year 11 Physics in Victoria, with three widely used (and also used elsewhere in southeast Asia). These three and their authors were used in this study. Each book was carefully analyzed in terms of the ways it attempts to teach about DC concepts, particularly ‘current’ and ‘voltage’. These analyses focused on explanations used (form and sequence), intended development of

438

DIFFICULTIES OF TEACHING AND LEARNING ELECTRICITY

concepts through these explanations, and analogies/models/metaphors used in the explanations. Specific text/linguistic analyses were not used. After analyzing the DC electricity sections of the 3 books, we interviewed the authors of the sections. The two part interview, taking 2 - 3.5 hours in total, had several foci as outlined below. Part 1 (3 sections). Section A focused on interviewees as textbook writers (how they each saw student learning of physics, their perceptions of the place of texts in student learning, their ‘typicalness’ as a textbook writer, the logic of the structure of their book, the place of the nature of science/physics in their book). Section B concerned the electricity section of their book (perceptions of ease/difficulty of writing, concepts difficult to ‘convey’, how it would be changed if rewritten). Section C asked about the interviewee’s meanings for ‘models and analogies’ and the value of these for students learning electricity from their book, and differences in useful models and analogies for DC electricity at different levels of physics/science teaching. Part 2 (3 sections). Section A focused on electricity concepts. We asked authors how they would expect students, after using their book, to answer questions relating to a simple DC circuit (cell, switch, and one or two light globes). This circuit was then used to probe the interviewees’ understandings of ‘voltage’, potential difference, voltage drop, emf, and electric energy. Section B asked about specific matters in their book identified through our analysis (e.g. ways analogies or problems were used). Section C presented student statements about electricity, with interviewees asked to consider what each revealed about the understanding of the student who made it. This was to explore authors’ conceptual understandings in a different way. Data analysis Each interview was transcribed in full. It is clear from the outline of our interview approach that many sections could potentially give information about the authors’ conceptual understanding. Working both individually and collectively, we analyzed the transcript carefully for indications of the author’s understanding of current and ‘voltage’ (and related terms), and for their explanations of aspects of book content and sequence. For each textbook one of us initially analyzed the relevant parts of the book by section, under three headings: ‘Section description & comment/quotes’, ‘Comments/quotes on Current’, and ‘Comments/quotes on Voltage’. We then collectively considered the analysis, refined it, and formed conclusions about the book’s approach. 5. THE TEXTBOOK AUTHORS In accord with ethical practice, authors were interviewed after assurances that they would not be identified in reports of the research. Clearly this places restrictions on what and how we can report. In particular, we cannot report interview data and text

R. GUNSTONE, B. MCKITTRICK, P. MULHALL

439

analysis in ways that enable linking of individual interviews with a given book. The background of authors we describe is all a matter of public record; we cannot link this background with particular features of the data analysis. All 3 authors are trained high school physics teachers, with teaching experience ranging from 15 to 40 years. One had an engineering degree, the other two science degrees. All are qualified teachers. None had another career before teaching. One had recently completed a Ph.D., focused on contexts and physics curriculum/teaching/learning. One author has been successfully authoring physics texts for over 20 years (where ‘successful’ means high sales), another has been successfully authoring across the range of sciences for 20 years, and the third had prior experience only with writing children’s science reading books. Our prime focus in this paper is on authors’ understandings. In the next section we consider understandings of current and ‘voltage’ (and related terms). The authors are named A1 (Author 1), A2, A3. Then we very briefly comment on the analyses of the books (Book A, Book B, Book C). We emphasize the issue of anonymity – A1 may or may not be the author of Book A, and so on. 6. RESULTS – THE INTERVIEWS Summaries of authors’ views We first give very brief summaries of authors’ views of issues such as how students learn physics. This is important background for the greater detail we then give of understanding of current and ‘voltage’ (and related terms) as revealed during the interviews. There is more emphasis on ‘voltage’ because of the nature of these data. We note that A3 often answered interview questions as a physics teacher, not as a textbook author, even though our questions were framed for authors. We find this interesting, but do not understand what (if anything) it reveals. A2 was also different to the other two in that there were many times where s/he clearly and explicitly stopped and thought about a question – A2 engaged more closely and more reflectively with our interview. The authors’ views of the matters of background to conceptions of current and ‘voltage’ are given below. 1

2

None had well articulated and cohesive views of learning. Two (A1 & A2) drew quite strongly on their own experiences as physics learners. A1 made some reference to alternative conceptions. A2 saw language and models as important in learning “abstract ideas like electricity”, and the need to build on previous learning. A3 saw “real life contexts or situations” as central, and as coming from “current learning theories [and] personal experience”. None had any specific views about learning from text. There was variation in the authors’ views of the ease or difficulty of writing the electricity section of their books. A3’s views suggest there may have been little thought about student learning or possible learning problems. A1 saw this section as particularly difficult, not because of the concepts per se but because

440

3

DIFFICULTIES OF TEACHING AND LEARNING ELECTRICITY of his/her uncertainly about previous student learning that could be assumed. Both A1 and A2 made specific reference to problems resulting from the absence of electric field, and noted the difficulty of convincing a commercial publisher that something ‘not in the course’ should, for pedagogical reasons, be in a textbook. A1 and A3 both indicated that they did not see any concepts in this section as particularly difficult to write about. None were convincing when asked what they meant by the terms models and analogies. Author 1 saw models as physical things, “a replica of something that is used in practice”, and gave examples such as a working model of an engine. Analogies were seen as more abstract, with the commonly used water analogy for an electric circuit being given as an example. Author 2 said that “a model is almost something you can construct and an analogy is something it is like”. This author also indicated that “I’ve never thought of the question before.” At no time did Author 3 indicate what s/he meant when using these terms. When really pushed to indicate how s/he used these terms, the following was the response: A3: “difficult ones to define there. Yeah, um, ah … Both of them are sort of tied in there um I s’pose … you’re trying to use a simple little situation to ah help you explain a more complicated concept. Um, and your simplified um model there is um trying to represent abstract things that are in your concepts um and … they’re all geared around trying you know to help students to understand something that they can't really see all that well.... And analogies is a bit um harder to um define there. I don’t see them as being a great deal different there.... Um, I probably haven’t explained both of those two things all that well but probably … I don’t often see the requirement to sit down and (define?) those things.” [Laughs – we assume as an indication of some embarrassment]

We believe that there is strong evidence that Author 3 did not discriminate between models and analogies in his/her thinking. We now consider understandings of current, ‘voltage’ and related terms, as revealed by the interviews. Conceptions of current held by the authors Not unexpectedly, all authors had conceptions of current as charge movement. However, all 3 authors’ common use of “flow” with current contained no indications that any saw “flow” as metaphorical. There is one significant current-related issue for which we believe we see differences between authors, an inferential rather than directly observable difference. We describe this using the broad notions of ‘cause’ and ‘consequence.’ The difference is shown by differing responses to the first question in Part 2 Section A (“How would you expect students to explain” a particular circuit). A1 and A2

R. GUNSTONE, B. MCKITTRICK, P. MULHALL

441

immediately began with notions of energy as ‘cause’, and then clearly saw current as ‘consequence’. On the other hand, throughout the interview, A3 did not appear to have this same clear sense of cause and consequence when considering energy, current, potential difference, etc. in this simple circuit. To the extent that this sense of A3’s conceptions is reasonable, we would argue this represents a shortcoming in his/her conceptions of electric current. Conceptions of ‘voltage’ held by the authors As already noted, many sections of the interviews contributed relevant data. Here we summarize these data to present our view of each author’s understanding. Before giving the summaries, we reproduce extracts from the part of A3’s interview that directly asked about his/her understanding of ‘voltage’, to give both a sense of the interviews, and to illustrate some inadequacies in A3’s understanding. (I is the interviewer.) A3: Um, I see voltage as being – I stress this word there – an indication [A3’s emphasis] of the energy that the charges have got, um, and ah, so, you know, you can talk about the voltage that the battery has got there and the voltage that, ah, is used up by the globe so, um, I always, you know, explain it in that way, that it’s an indication [of] the energy that the charges have got flowing around the circuit. I: What about potential difference, do you see that as being different to voltage? A3: Um there’s 3 terms that often get mixed up and you see them used in different ways in books and that. Potential difference is one, voltage is the other, and emf is the other – those 3 things. Emf and potential difference are sort of the old traditional ones there and we tend to use voltage as being more familiar to students cos they can relate better to that than the other two terms. Um, at the senior level there, um I make use of and mention all 3 there ... if you look at [a] light globe ... um on one side of the globe there you’ve got the charges come in with a certain amount of energy, or electric potential energy if you like, and on the other side they emerge with less electrical potential energy ... there’s that difference in energy that the charges have got and I … refer to that as that’s the potential difference or electrical potential difference. Um, if we put a voltmeter across that globe there then the globe [sic] is really measuring the difference in electrical potential energy from one side to the other and it does a little subtraction and gives a reading. I: And is the electrical potential difference what you see as being ‘voltage’? A3: And I say we often call that, or just refer to it commonly as voltage, yeah. Um, emf is mainly the term used for, um, the voltage across a battery there with no load on it. ... I: So ‘voltage’ is not potential difference to you?

442

DIFFICULTIES OF TEACHING AND LEARNING ELECTRICITY A3: (Speaking slowly) They’re … fairly similar. (Speaking faster) I tend to use them in a fairly similar way. Cos I tell students that … we use the term emf and we’ve got potential difference there and voltage and I tend to tell them that you know, really all those things are sort of voltage if you want to use one sort of term, um, even though you know scientifically each of them on their own [has] got a more precise meaning, um, but I still um prefer to use the word voltage to refer to any of them or all of them.

The interview transcripts show differences between authors in conceptions of ‘voltage’, potential difference and emf. A1 is concise and direct in responses (and generally more concise and direct than for other parts of the interview), e.g. “Voltage to me is the measure of the amount of energy that each coulomb, if you are talking volt, that each charge has got as a result of its position in the circuit and the other charges that are around it.” There was similar conciseness and directness in his/her view of how s/he would expect students to explain the simple circuit after using his/her book – the view was clearly focused on energy (source and ‘drain’). This we take as a measure of this author's confidence in his/her answers, particularly given that not all the interview was as strong in this way. Despite this confidence, we see A1’s view of ‘voltage’ as having a conceptual status which, as noted above, we do not think helpful for understanding, and we see some possible uncertainty in his/her conception of potential difference. In this section of the interview it was extremely difficult to extract any elaboration of A1’s initial response. It may be that our inability to probe further successfully has led to an unreasonable view of A1’s ideas of ‘voltage’ and potential difference. A2 was more obviously reflective while framing responses. The non-verbal aspects in our transcript ([exhale], [pause], etc.) emphasize this, as do clear instances of thinking aloud in developing an answer. While not being as explicit as one might like about the idea of ‘voltage,’ A2 does indicate that the idea does not have for him/her the same conceptual status as potential difference. A2 does not approach these questions with the same confidence as A1 (e.g. “Electrical energy is another thing I have trouble with too, as to what that actually means”), something that we regard more positively than negatively, given the complexity in this conceptual area and the uncertainty that many textbooks and teachers show. A3 is, quite clearly, uncertain in much less helpful ways. It is, we argue, reasonable to suggest that A3 is rather confused in this area. Consider, for example, some statements in the above extract: - “I see voltage as being – I stress this word there – an indication [interviewee’s emphasis] ...” (we take the strong emphasis on “indication” here to indicate uncertainty). - “Emf and potential difference are sort of the old traditional ones there and we tend to use voltage as being more familiar to students cos they can relate better to that than the other two terms” (multiple points to take issue with in this quote).

R. GUNSTONE, B. MCKITTRICK, P. MULHALL

443

- “I tell students that … we use the term emf and we’ve got potential difference there and voltage and I tend to tell them that you know, really all those things are sort of voltage if you want to use one sort of term, um, even though you know scientifically each of them on their own [has] got a more precise meaning.” (Note: Despite our probing A3 gave no indication of what he/she saw as the “more precise meaning” of these terms.) We would not see the conceptions of A3 as being appropriate for a teacher of physics at this level, and thus certainly not acceptable for an author of a textbook at this level. 7. RESULTS – TEXTBOOK ANALYSES We now turn briefly to the analyses of the 3 textbooks (Books A – C). Space precludes the provision of detailed data here. We give a brief summary of significant aspects of our analyses, and again focus on the ways current and ‘voltage’ are treated. Book A: We see the treatment of current and ‘voltage’ as inadequate, and do not see how a student working just from this textbook could come to a satisfactory understanding. This judgment is supported by (i) no use is made of any analogies in the book (a point noted by the author in interview, and justified on the basis that teachers would use their own analogies; however there is no indication in the book that this is an expectation of the author); (ii) sections to introduce and explain concepts are very brief (for ‘voltage’, the introduction/explanation contains less than 200 words, 2 equations, and one diagram), and there is much use of qualifying words such as “tend to” which we see as indicators of conceptual uncertainty (e.g. “An electric current can be considered to be the rate at which charge moves through a conductor”); and (iii) there are conceptual problems with some of what is written (e.g. “Changes in potential (potential difference) across a component is [sic] often referred to as voltage”; “The greater the EMF of a battery the greater the current in any circuit connected to the battery”). Book B: The sequence and approach are generally reasonable. However, while the book claims a conceptual approach, the treatment of ‘voltage’ is little different from books that do not claim this approach. The treatment of current gives a good elaboration of matters leading to a conception of current. But there is a complete absence of such an approach for ‘voltage.’ The book makes explicit and generally helpful use of what it calls a “model” in the development of concepts, but this is an analogy. The value of the book is also diminished by its use of a single analogy for almost all the abstract concepts involved. Book C: The approach to current and ‘voltage’ focuses more on laying out essential facts known to physicists than on student development of conceptual understanding. In addition to some clearly incorrect statements about current (e.g. “…If, per second, 2 positive charges move to the right and 1 negative charge moves to the left past XY, then… the current is +3 units to the right because the right side

444

DIFFICULTIES OF TEACHING AND LEARNING ELECTRICITY

of XY is 3 units more positive than it was prior to the movement of the charge” – suggests that the conductor becomes more positive as you move in the direction of the [conventional] current!), the approach seems to be driven by the formula I = q/t. There is no attempt to provide a picture of what potential difference and voltage mean in terms of charged particles moving in a circuit. The only model mentioned is the atomic model of a metal conductor (comprising about 3 isolated sentences). Unlike most texts at this level, no model is used to explain the flow of charge. The major emphasis appears to be towards developing formulae for solving quantitative problems. 8. CONCLUSION Overall, we see a rather dismal picture in these data. The author interviews show surprisingly inadequate understandings of the nature of models and analogies (with A3 seemingly unable to differentiate these), and limited views of the ways students might learn (and might specifically learn from textbooks). The understandings of current of authors A1 and A2 were good, but that of A3 may not be. Conceptions of ‘voltage’ were varied with, again, A3’s being, we believe, inadequate. The approaches of the books were also extremely varied. Book A gives little attention to the development of concepts, and what is presented is clearly poor. Book B is confusing about models and analogies, and is diminished in value by its substantial adherence to a single analogy. The book gives a good conceptual treatment of the development of the idea of current, but does not repeat this for ‘voltage’. Book C focuses more on development of ability to solve problems than on conceptual understanding. This book is marred by some assertions about concepts that are misleading or incorrect., We see some consequences of the nature of each author’s conceptual understanding of current and ‘voltage’ in his/her book. It may be tempting to see these problems (in textbooks and levels of author understanding) as a problem specific to Victoria. However we have in other, as yet not as thorough, analyses of textbooks from different countries found most of the problems noted above (Mulhall et al., 2001, p. 582). Whether interviews with the authors of these books from other countries would also show the same range of problems with understanding remains to be seen, although we see no reason for believing this would not be the case. The more significant issue in these conclusions is the very strong reinforcement given to assertions made in the introduction to this paper about the problematic nature of the concepts of DC electricity. Here are experienced and successful textbook authors with at least some inadequacies of understanding, and aspects of conceptual confusion in their textbooks. That this had not impacted on the extent of use of these books in physics classrooms clearly suggests that teachers’ understandings are also often inadequate.

R. GUNSTONE, B. MCKITTRICK, P. MULHALL

445

ACKNOWLEDGEMENT This research was supported by Australian Research Council Grant A00104120. REFERENCES Cosgrove, M., Osborne, R. & Carr, M. (1985). Children's intuitive ideas on electric current and the modification of those ideas. In R. Duit, W. Jung & C. von Rhöneck (Eds.) Aspects of understanding electricity. Kiel, Germany: Schmidt & Klaunig. Duit, R., Jung, W. & von Rhöneck, C. (1985). Aspects of understanding electricity. Keil, Germany: Schmidt & Klaunig. Eijkelhof, H. & Kortland, K. (1988). Broadening the aims of physics education. In P. Fensham (Ed.) Development and dilemmas in science education. London: Falmer. Hart, C. (2002). Framing curriculum discursively: Theoretical perspectives on the experience of VCE Physics. International Journal of Science Education, 24 (10), 1055-1077. Koumaras P., Kariotoglou, P. & Psillos, D. (1997) Causal structures and counter-intuitive experiments in electricity. International Journal of Science Education, 19 (6), 617-630. Mulhall, P., McKittrick. B. & Gunstone, R. (2001). A perspective on the resolution of confusions in the teaching of electricity. Research in Science Education, 31 (4), 575-587. Pordhan, H., & Bano, Y. (2001) Science teachers' alternate conceptions about direct-currents. International Journal of Science Education, 23 (3), 301-318. Viennot, L. & Rainson, S. (1992). Students' reasoning about the superposition of electric fields. International Journal of Science Education, 14 (4), 475-487.

THE CONCEPT OF FORCE AS A CONSTITUTIVE ELEMENT OF UNDERSTANDING THE WORLD

KEES KLAASSEN Utrecht University, The Netherlands

ABSTRACT This paper concerns interpretation and constitutive elements of understanding the world, both of which are treated in relation to the concept of force. Studies are criticized in which students’ conceptions are formulated, without further clarification, in terms of the word ‘force’. From such reports it can neither be concluded what students believe nor how their beliefs relate to science. Instead, reasons or criteria for applying ‘force’ need to be made explicit. Those reasons concern the effects that forces produce, namely deviations from an influence-free state; they also concern their sources, as made explicit in laws from which, for a given situation, the forces acting in it can be derived. The general concept of force, thus associated with the two-tier explanatory strategy of specifying (1) influence-free states and (2) force laws to account for deviations from those states, is a constitutive element of understanding the world. Within the constraints set by this explanatory strategy, the concept of force can still be variously applied, both in everyday and in scientific explanations. The differences between these various applications are partly anchored in distinct explanatory interests.

1. INTRODUCTION There is quite some consensus about one cause of the ubiquitous finding that students do not learn as much as was hoped for. A student entering a science class is not a tabula rasa but already has conceptions: domain specific beliefs; more general ones concerning the nature of science, epistemology, or ontology; and attitudes other than beliefs, such as motives, interests, and affects. Opinions begin to diverge concerning the status of students’ prior conceptions and how to properly take them into account. Some claim that the prior conceptions ‘are often in stark contrast to the science conceptions to be learned’ and that ‘major restructuring of the already existing knowledge is necessary’ (Duit, 1999). Others argue that this view ‘overemphasizes the discontinuity between students and expert scientists’ and think of students’ prior conceptions as ‘resources for cognitive growth’ (Smith et al., 1993). One major focus of this paper is a question that ought to precede the further question if – or in what sense – students’ conceptions are impedient or expedient to their learning of science. This question is: what is the content of their conceptions? I raise this simple question, because I have doubts about the validity of a lot of 447 K. Boersma et al. (eds.), Research and the Quality of Science Education, 447—457. © 2005 Springer. Printed in the Netherlands.

448

THE CONCEPT OF FORCE AS A CONSTITUTIVE ELEMENT

research on students’ ideas. The answer to this question, with respect to the concept of force, introduces the other major focus of this paper: a discussion of the relations between common sense and science in terms of constitutive elements of understanding the world. 2. THE PROBLEM OF INTERPRETATION In early studies on children’s pre-instructional theories of motion (Halloun & Hestenes, 1985; Gunstone & Watts, 1985), it is reported that children seem to operate with basic intuitive rules such as: · Sustained motion needs a continuous force. · If an object is in motion, it has a force in the direction of its motion. · If there is no continuous supply of force, the force of an object wears out. · Forces can be imparted by agents and transferred from one object to another. Of course, children or lay people do not always frame their ideas in these exact words. But it is a small step from ‘If he wanted to keep moving along he would have to keep pushing’ (an example of what a child actually said, cf. Gunstone & Watts) to ‘Sustained motion needs a continuous force’. So, it is plausible to assume that the child would have assented to the latter sentence. If children and lay people can be said to hold the intuitive theory in this sense, what follows? According to many researchers the intuitive theory is ‘at variance with the principles of Newtonian mechanics’ (McCloskey, 1983). I agree that ‘Sustained motion needs a continuous force’ seems to be contradictory to Newton’s first law, and that ‘to have a force’ is meaningless in Newtonian mechanics. But does it follow that the intuitive theory contradicts Newtonian mechanics? I think not. Consider: S. Sustained motion needs a continuous force. Children and lay people would assent to S, while Newton would dissent from it. This only implies they have contradictory beliefs, however, if all parties understand S in the same way, i.e. if there is identity of meaning. But does students’ preinstructional concept of force, in particular, match the Newtonian one? Most researchers probably hold that it does not. My criticism begins when it is left unsettled what, then, students’ pre-instructional concept of force is, for this also leaves unsettled what children and lay people believe when they assent to S. Only after all this is settled, can one ask whether their belief contradicts any of Newton’s beliefs. The same criticism may apply to researchers who hold that at least some of students’ conceptions are in agreement with science. In the case of a spring that one holds compressed, both a student and a physicist assent to ‘The spring exerts a force on the hand’; and in the case of a book lying on a table, only the physicist assents to ‘The table exerts a force on the book’. On this basis nothing can be concluded, as yet, about beliefs. Neither does it follow that in the latter case student and physicist have contrary beliefs, nor that in the former their beliefs are in agreement (cf.

K. KLAASSEN

449

Klaassen & Lijnse, 1996, for further discussion, including an evaluation of the bridging strategy of Clement et al., 1989). The above exemplifies two related issues (cf. Davidson, 2001, essays 9 to 12). First, the interdependence of belief and meaning: what someone believes when she assents to some sentence depends on what the sentence means for her. Second, the associated problem of interpretation: if we only know which sentences someone assents to, and we cannot assume identity of meaning, then how are we to find out what her sentences mean and what she believes? My methodological criticism of a lot of research into children’s thinking is that this problem of interpretation, despite quite common implicit recognition, is hardly ever explicitly mentioned, let alone properly solved. Reports in which intuitive theories are formulated, without further clarification, in terms that are also in use in science, cannot be expected to have solved the problem. 3. TOWARD A SOLUTION OF THE PROBLEM OF INTERPRETATION In this section I try to arrive at a solution of the problem of interpretation with respect to the concept of force, by drawing on studies also critical of mainstream research on students’ conceptions. Avoiding the word ‘force’ In order to block associations with a physicist’s use of the word ‘force’, a useful initial strategy is to avoid it. Bliss & Ogborn (1993), for example, formulate a commonsense theory of motion in such terms as ‘support’ and ‘effort’. Some elements of this theory are: · If an object is supported it does not fall. If it is not supported it falls, until it is once more supported. Falling is a natural motion. One need not look for a cause for it to continue, only for a continued lack of support. · To support something needs strength or effort. If the strength of a support is not enough, it may break. If the strength is enough the support takes the weight of the thing it supports. DiSessa (1993) uses such terms as ‘agent’, ‘result’ and ‘effort’ in formulating his phenomenological primitives, e.g. Ohm’s p-prim: · An agent exerts some effort in order to achieve a result through some resistance. Increased effort begets increased result. Increased resistance begets reduced result. When I read what children say about familiar situations in which something moves, usually after it has been kicked, pushed, thrown, etc., I think I understand perfectly well what they are trying to say. When riding my bike I have to keep pushing to keep moving; the harder I throw something, the farther it gets, and so on. Put in somewhat more general terms, some basic intuitive rules can be formulated by which all of us operate (cf. Klaassen, 1995, section 2.2.3), for example:

450 · ·

THE CONCEPT OF FORCE AS A CONSTITUTIVE ELEMENT To keep things in motion, one has to keep making an effort, otherwise they will come to a stop. The motion of an object can cause something to happen. Faster motion begets more effect.

Dekkers (1996, section 2.7) presents a table, with on the left the most significant generalized students’ statements involving the word ‘force’ he culled from the literature; and on the right his attempts to understand those statements without using the word ‘force’. Here is a sample of his four-page-long table: If a body is not moving, there is no force Being at rest is a natural state of an object. acting on it. If a body is moving, there is a Absence of motion does not require an force acting on it in the direction of explanation. Motion does not spontaneously motion. start. Moving objects differ from stationary objects, which requires an explanation. Forces are present in static situations if an In some situations, the fact that nothing event will happen, is about to happen, or happens (yet) is out of the ordinary and requires is prevented from happening. an explanation. This is the case when it is believed that something will happen, is about to happen or prevented from happening. Properties of the current situation explain what will produce the later effect.

Without pretending to have now solved the problem of interpretation, the above attempts to avoid the word ‘force’ may already have cast doubt on the oftensupposed alternativeness of students’ conceptions. Making explicit the circumstances in which students hold the concept of force applicable Another useful strategy in solving the problem of interpretation is also intuitively clear. If we are to find out what someone’s reasons or criteria are for applying the concept of force, we can do no better than assume that whenever she believes her criteria are applicable, they are in fact applicable. That is, we must assume that she does not, as a rule, misapply her own criteria, however difficult she herself may find it to articulate them clearly. The above attempts to avoid the word ‘force’ suggest that in commonsense reasoning forces are involved in explanations of something that is ‘out of the ordinary’ or deviates from a ‘natural state’. An object ‘exerts a force’ if it causes a deviation, ‘has a force’ if it has the potentiality to do so, and ‘transfers a force’ if the caused deviation is such that another object causes a further deviation or has acquired the potentiality to do so. Dekkers lists examples of what students may consider as ‘out of the ordinary’: deformation, deflection, (reduction of) motion, stress. He summarizes these as ‘activity’, and formulates as rules governing the commonsense concept of force: ‘activity implies a force’ and ‘more activity implies more force’. DiSessa’s characterization of some p-prims can also be understood as indications of the kinds of deviation for which people may feel an explanatory need in terms of forces: ‘force as a mover’, ‘force as a deflector’, ‘force

K. KLAASSEN

451

as a spinner’. DiSessa and Dekkers also mention examples of what students may consider as sources of forces: animate agents, engines, collisions, supports, springs. Capturing the essence of the concept of force By combining some of the suggestions mentioned above, we may try to capture the essence of the commonsense concept of force in the slogan ‘force as a deviator from a natural state’. DiSessa plays with the thought that something like this might be a meta p-prim, but in the end he rejects it: ‘many kinds of change need explanation but […] there is no [fixed or] universal characterization of change [or of natural state, KK] that covers the circumstances in which people feel the need, or not, to look for deeper causation.’ So if ‘force as a deviator from a natural state’ is proposed as a solution to the problem of interpretation with respect to ‘force’, a problem of interpretation with respect to ‘natural state’ immediately surfaces. The latter problem is as difficult as the original one; in fact, they are the same. Does this imply that we have to give up the attempt to capture the essence of the concept of force; and must we rest content with an exemplary list of the kinds of circumstance that may be considered as ‘out of the ordinary’, and thus as standing in need of explanation? Precisely this is the ‘knowledge in pieces’ view that diSessa argues for: the explanatory needs that people feel ‘are distributed according to the kind of change; that is, they are embedded in the [phenomenological] primitives that connect to the particular circumstance.’ Dekkers is undecided. On the one hand, he admits that his analysis ‘does not yet provide an integrated, comprehensive theory for the common-sense concept of force’; on the other, he does ‘not even claim that all of the students’ views can be described in one comprehensive and coherent theory.’ I agree with diSessa that there is no single, all-purpose class containing all and only effects of forces, and the same goes for sources of forces. Nevertheless, I think diSessa’s pieces can find their place in a coherent whole. What still needs to be made explicit is the conceptual pressure that governs the rightness of fit between source and effect. ‘Force as a deviator from a state’ only makes explicit that the concepts of force and state are mutually dependent; a further element still is to be found which governs their mutual application. In my view the notion of generality will do. Part of the reason for holding that in a particular situation there is a deviation effectuated by a force, is that there is reason to believe that in relevantly similar situations relevantly similar deviations would occur. I suggest capturing the essence of the concept of force by the following two-tier explanatory strategy: · ·

a characterization of influence-free (force-free) states, checked by a characterization of plausible lawlike statements (force-laws) in which deviations from those states are correlated with properties of the configurations in which those deviations occur.

This two-tier strategy makes explicit the intimate connection between interaction theory (the collection of relevant force laws) and the notion of state (cf.

452

THE CONCEPT OF FORCE AS A CONSTITUTIVE ELEMENT

Friedman, 1983, section III.7). The strategy does not tell us what we ought to choose as forces, states, laws, etc. It only sets constraints on such choices. It offers an explanatory scheme into which the choices we make must fit. The scheme may need to be augmented with further basic elements, such as the semi-quantitative ‘the more force, the more deviation’ or a notion of laziness (inertia) to indicate that different bodies react differently to an applied force (the more inertia, the less deviation). DiSessa’s p-prims can also find their place here. 4. SPECIFIC APPLICATIONS OF THE CONCEPT OF FORCE AND THEIR INTERRELATIONS The general concept of force can still be applied in explanations in various specific ways. In this section I give examples of such specific applications and their interrelations. Examples of specific applications of the general concept of force The two-tier explanatory strategy indicated above can be seen at work in commonsense explanation of motion. In everyday life we take a strong interest in how to move objects from A to B, or to move ourselves from A to B by means of some object. Within this context, it makes good sense to consider as influence-free the way in which the objects would move without our interference (stand still or gradually come to a stop), given that at the same time we happen to know enough generalizations, however crude or vague they may be, in which relevant deviations (setting in motion, keeping in motion, braking) are satisfactorily correlated to kinds of action we can perform. Given another goal, e.g. hitting a target with a projectile, another type of motion can be considered as influence-free, as long as this is checked by the availability of sufficient rules of thumb to account for relevant deviations from that type of motion. Many of the intuitive rules concerning motion are (related to) crude laws between kinds of action and kinds of deviation from a particular type of motion. Thus, a ragbag of specific applications of the concept of force can be seen to function in commonsense explanation, all of which are ‘highly pragmatic not only in their conspicuous ties to action, but in their breezy disregard of the irrelevant or implausible’ (Davidson, 1995). The general concept of force can also be seen as structuring the Newtonian framework. It consists of (1) the specification of a kind of motion that is to count as a state (uniform rectilinear motion), and (2) interaction theory to account for all deviations from this type of motion in terms of force laws. Force laws, such as Newton’s law of gravitation, are general statements that specify the forces objects exert on each other as a function of their total configuration (cf. Jammer, 1957, chapter 12). One can continue to take uniform rectilinear motion as a state, and here the conceptual pressure of generality is felt, as long as one is able to find plausible force laws to account for the deviations from it.

K. KLAASSEN

453

Although Kepler himself has not systematically developed it, one can also specify a Keplerian style of explaining motion. It consists of (1) the specification of rest as the kind of motion that is to count as a state, and (2) interaction theory to account for deviations from rest. This is an application of the concept of force which differs from the Newtonian one. Keplerian net forces, just to name one difference, are of necessity always in the direction of motion. In order to account for planetary motion, Kepler imagined some kind of spokes emanating from the sun and pushing the planets along their orbits as the sun rotates about its axis (cf. Jammer, 1957, chapter 5). It is possible to make Kepler’s idea precise and to formulate more or less plausible Keplerian force laws, which lead to the same predictions of planetary motion as within the Newtonian framework on the basis of a gravitational influence directed to the sun. Relations between the various specific applications of the concept of force One aspect of the relation between the various specific applications of the concept of force is, of course, that they are specific applications of the same general concept. In all cases, the basic criteria for identifying forces are the same. In the circumstances under which the motion of an object deviates in some relevant way from the assumed influence-free motion, one will search for recurring configurations with which these deviations can be satisfactorily correlated. Such shared basic criteria still allow for different specific criteria, corresponding to the kind of motion considered as influence-free, in combination with why one needs an explanation and, given this purpose, with what one thinks can be explained by means of appropriate laws. In the same way, the shared basic explanatory strategy still allows for different specific explanations. Consider the question why one has to keep pedalling in order to keep speed on a bicycle. The specific commonsense explanation is straightforward. Keeping speed is a deviation from what, for the everyday purpose of moving oneself from A to B by means of some object, can plausibly be assumed as influence-free motion (stand still or gradually come to a stop), and according to the intuitive rules, a continuous effort by an agent can account for that kind of deviation. The specific explanation within Newtonian mechanics is more involved. Under normal terrestrial conditions, a moving bicycle comes to a stop. This kind of deviation from the Newtonian influence-free motion (uniform and rectilinear) can under normal terrestrial conditions be correlated with sources of opposing frictional forces. So to keep a body in motion one somehow has to make it the case that there is also a forward force. If this forward force balances the frictional forces, such that there is no net force, the result is uniform rectilinear motion. Whereas in common sense there is no need for uniform application of the concept of force, things stand differently in Newtonian mechanics, when it is pursued in a frame of mind in which we want to understand things ‘whether or not we can control them, and whether or not such knowledge will serve our mundane needs’ (Davidson, 1995). In that frame of mind, explanation of motion, though it may answer to various interests, in itself is not relative to any interests. Every

454

THE CONCEPT OF FORCE AS A CONSTITUTIVE ELEMENT

deviation from uniform rectilinear motion, whether it is of practical interest or not, ultimately has to be accounted for by means of exceptionless force laws. Such differences with respect to purpose, interest, and what is considered relevant, form another aspect of the relation between common sense and Newtonian mechanics. Due to rather disparate explanatory interests, there is no tension between Newtonian and commonsense explanation. Within one and the same theoretical mood, different specific applications of the concept of force can still be tried, as illustrated by the Keplerian and Newtonian schemes. Within each scheme, deviations from the assumed influence-free motion provide motives to construct a theory that succeeds in accounting for the deviations; but because there are no guarantees that one will be able to do so, a rivalry arises between the two schemes. Their relative merits must be evaluated in the light of a shared commitment to the usual epistemic virtues associated with their fundamental aspirations, such as those of strict empirical adequacy and unification (cf. Nagel, 1979, section 7.II, for further discussion of the status of laws of motion). Educational challenges In my view the analysis above contains useful suggestions concerning the kind of work that must be done to provide students with incentives and conceptual tools to bridge the gap from common sense to Newtonian mechanics (cf. Westra et al., 2003, for more details). Students will somehow have to be brought into the theoretical mood that suits a fundamental research type of practice. The general concept of force, if appropriately made explicit on the basis of commonsense explanations, can then serve as a general scheme to be further specified. The Keplerian specification with its associated assumption of rest as influence-free motion may especially appeal to students. If presented together with the Newtonian specification, students can try to construct, for both specific schemes, interaction theories that succeed in accounting for, say, planetary motion. What will be evident to them is that the force laws need to be empirically adequate in the sense that the predicted paths must match the actual paths as closely as possible. In a modelling process of fitting and adjusting parameters, they may thus, within both the Keplerian scheme and the Newtonian scheme, arrive at more or less adequate force laws. Such initial successes make it plausible to investigate whether for other cases than planetary motion it is also possible to find empirically adequate force laws within each of the two schemes. Thus, the schemes themselves can also come to be evaluated, and in an increasingly explicit way, in the light of the epistemic value of unification. In short, in a modelling process that both depends on and implements the general concept of force, and that is driven by epistemic values associated with a fundamental research type of practice, the Newtonian scheme can in the end be expected to fare better. 5. CLOSING REMARKS In this section I suggest a broader application of an analysis like the one just given with respect to the concept of force.

K. KLAASSEN

455

Interpretation I have criticized studies on students’ conceptions because of their neglect of the problem of interpretation. Johnson & Gott (1996) have likewise given methodological criticism. They note that, contrary to the impression one might gain from the literature, finding out what a child thinks is not a straightforward task. It demands ‘more than simply asking a child a few ‘key’ questions and then categorizing verbal responses according to forms of words’. They challenge the findings of some well-known research studies in terms of some plain methodological principles, which every researcher, when explicitly asked, would support. Nevertheless, it takes some effort to observe those principles, and in much research into children’s thinking, they are insufficiently observed. Johnson & Gott also suggest alternative interpretations which they try to support by further research; they also stimulate a re-evaluation of existing data. Analyses like the one of Johnson & Gott and those presented earlier in this paper (Bliss & Ogborn, diSessa, Klaassen, Dekkers) suggest that the effect of a proper reinterpretation of studies into children’s ideas is rather sobering. The outcome simply is that most of children’s ideas are plain commonsense truths. This is part of the reason that it is difficult for me to maintain a genuine interest in studies into students’ conceptions, and that they are only addressed here in order to point to methodological issues. The other part is based on a further analysis of interpretation as provided by the philosopher Davidson. He concludes that ‘[i]nterpretation requires us to see the mental lives of others as enough like our own in point of overall coherence and correctness to allow us to assign reasons to their acts, intentions, beliefs and other attitudes – in other words, to understand them. It requires us to see other agents as more or less rational creatures inhabiting a shared world that they conceive much as we do’ (Davidson, 2000). If these interpretational maxims are correct, then their application to students’ speech and action cannot result in anything else than that what they say and do is by and large coherent, correct, and to the point. Constitutive elements of understanding the world The general concept of force with its associated explanatory scheme is a special case of the so-called cause-law thesis. If two particular events are related as cause and effect (a caused b), there is a law (lawlike generalization) to the effect that ‘all events similar to a will be followed by events similar to b’. We have reason to believe the singular causal statement only in so far as we have reason to believe there is such a law, and we may have good reason to believe there is such a law without knowing what the law is. Like the concept of force, the cause-law thesis offers an explanatory scheme. ‘[E]vents are changes that explain and require such explanations. This is not an empirical fact: nature doesn’t care what we call a change, so we decide what counts as a change on the basis of what we want to explain, and what we think available as an explanation. In deciding what counts as a change we also decide what generalizations to count as lawlike. [...] if you can’t explain it using one assumption of what counts as a change, adopt new categories

456

THE CONCEPT OF FORCE AS A CONSTITUTIVE ELEMENT

that allow a redefinition of change’ (Davidson, 1995). Ramberg suggests thinking of the cause-law thesis as a constitutive element of understanding the world. It expresses constraints on the application of such basic concepts as those of cause, change, state, kind, substance, and object. It thus forms a bond between common sense and the various sciences: no matter how distinct and diversely anchored in explanatory interests these various theoretical structures may be and remain, ‘the constitutive interest in generality, the inherent susceptibility to law of their terms, marks them as […] a part of the same general descriptive project’ (Ramberg, 1999). Holding that the cause-law thesis is a constitutive element of understanding the world implies that our thoughts do not simply picture the world. It does not imply, however, that the world is of our own making, as (social) constructivists often seem to hold. Nor does it imply, that how we think the world is has got little to do with how the world is, as sceptics in their various ways maintain. We do not, at least in any ordinary sense, choose the constitutive elements of understanding the world. They rather are ineluctable elements of what directs and explains our choices in understanding the world. So, I disagree with diSessa’s (1993) suggestion that there is no reason to suppose ‘that children cleanly and with high reliability encode the general principle that change needs a cause.’ Rather, we must so interpret children that by and large they come out as abiding by this principle. This involves finding out what, in the context at hand, they take as changes, states, and laws. I can interpret some empirical studies on ‘commonsense reasoning’ in this sense (Whitelock, 1991; Gutierrez & Ogborn, 1992). Other experimental work, however, such as an attempt to uncover basic dimensions of thought about the nature of entities (Mariani & Ogborn, 1991), rather seems to try to draw some constitutive elements out of students. This seems to me misguided. Although students’ understanding of the world also both implements and depends on the relevant constitutive elements, they simply are not the right kind of people to make those explicit. Obvious as the constitutive elements may sound once they are formulated, it takes the greatest minds to articulate them clearly and sharply. It is up to us as educationalists to explore whether and how they can be made educationally productive. REFERENCES Bliss, J. & Ogborn, J. (1993). A common-sense theory of motion: issues of theory and methodology examined through a pilot study. In: P.J. Black & A.M. Lucas (Eds.). Children’s informal ideas in science. London: Routledge. 120-133. Clement, J., Brown, D.E. & Zietsman, A. (1989). Not all preconceptions are misconceptions: finding ‘anchoring conceptions’ for grounding instruction on students’ intuitions. International journal of science education, (11). 554-565. Davidson, D. (1995). Laws and cause. Dialectica, 49. 263-279. Davidson, D. (2000). Objectivity and practical reason. In: E. Ullmann-Margalit (Ed.). Reasoning practically. Oxford: Oxford University Press. 17-26. Davidson, D. (2001). Inquiries into truth and interpretation (2nd edition). Oxford: Clarendon Press.

K. KLAASSEN

457

Dekkers, P.J.J.M. (1996). Making productive use of student conceptions in physics education: development of the concept of force through practical work. Amsterdam: VU Huisdrukkerij. DiSessa, A.A. (1993). Toward an epistemology of physics. Cognition and instruction, 10. 105-225. Duit, R. (1999). Conceptual change approaches in science education. In: W. Schnotz, S. Vosniadou & M. Carretero (Eds.). New perspectives on conceptual change. Amsterdam: Elsevier Science. 263-282. Friedman, M. (1983). Foundations of space-time theories: relativistic physics and philosophy of science. Princeton: Princeton University Press. Gunstone, R. & Watts, M. (1985). Force and motion. In: R. Driver, E. Guesne & A. Tiberghien (Eds.). Children’s ideas in science. Milton Keynes: Open University Press. 85-104. Gutierrez, R. & Ogborn, J. (1992). A causal framework for analysing alternative conceptions. International journal of science education, 14. 201-220. Halloun, I.A. & Hestenes, D. (1985). Common sense concepts about motion. American journal of physics, 53. 1056-1065. Jammer, M. (1957). Concepts of force: a study in the foundations of dynamics. Cambridge: Harvard University Press. Johnson, P. & Gott, R. (1996). Constructivism and evidence from children’s ideas. Science education, 80. 561-577. Klaassen, C.W.J.M. (1995). A problem-posing approach to teaching the topic of radioactivity. Utrecht: CD-ß Press. (http://www.library.uu.nl/digiarchief/dip/diss/01873016/inhoud.htm) Klaassen, C.W.J.M. & Lijnse, P.L. (1996). Interpreting students’ and teachers’ discourse: an underestimated problem? Journal of research in science teaching, 33. 115-134. Mariani, M.C. & Ogborn, J. (1991). Towards an ontology of common-sense reasoning. International journal of science education, 13. 69-85. McCloskey, M. (1983). Intuitive physics. Scientific American, 248. 122-130. Nagel, E. (1979). The structure of science: problems in the logic of scientific explanation. Indianapolis: Hacket Publishing Company. Ramberg, B. (1999). The significance of charity. In: L.E. Hahn (ed.). The philosophy of Donald Davidson. Chicago: Open Court. 601-618. Smith, J.P. III, diSessa, A.A., Roschelle, J. (1993). Misconceptions reconceived: a constructivist analysis of knowledge in transition. The journal of the learning sciences, 3. 115-163. Westra, A., Klaassen, K. & Lijnse, P. (2003). A new introduction for teaching-learning mechanics. Paper presented at ESERA 2003. Whitelock, D. (1991). Investigating a model of commonsense thinking about causes of motion with 7 to 16-year-old pupils. International journal of science education, 13. 321340.

CHANGES IN COLLEGE STUDENTS’ CONCEPTIONS OF CHEMICAL EQUILIBRIUM JOCELYN LOCAYLOCAY¹, ED VAN DEN BERG², MARCELITA MAGNO³ ¹University of San Carlos, Philippines, ²Vrije Universiteit, Netherlands ³UP National Institute of Science and Mathematics Education Development (UP-NISMED), Philippines,

ABSTRACT The purpose of this study was to describe the evolution of conceptions about chemical equilibrium based on observations/interactions with selected chemistry students using an instructional design which included constructivist strategies such as POEs, analogies, small group discussions, and journal writing. Two intact classes with a total of 75 students enrolled in a general chemistry course participated in the study. The conceptual evolution of six students was followed through the use of pre-tests, transcripts of audiotaped and videotaped group discussions, written answers to activity sheets, learning journals, interviews, and post-tests. The students started with concepts of complete reactions and progressed to developing concepts of reversibility but had difficulty with the dynamic nature of reversible reactions. The use of an analogy using double-sided disks helped in the students’ visualization of the microscopic processes taking place and in the properties of systems as they approach and when they reach equilibrium. However, students’ conceptions of complete reactions still competed with their conceptions of reversible reactions.

1. INTRODUCTION Chemical equilibrium is considered to be one of the most difficult topics in general chemistry. Several studies have investigated student difficulties in understanding the topic (e.g. Camacho & Good, 1989; Gussarsky & Gorodetsky, 1988; Hackling & Garnett, 1985; Bergquist & Heikkinen, 1990). The main alternative conceptions regarding chemical equilibrium have been summarized by Bucat & Fensham (1994), Huddle & Pillay (1996), and Van Driel and Gräber (2003). The most problematic concepts are the dynamic and reversible nature of chemical equilibrium, the use of Le Chatelier’s Principle, and the equilibrium constant. Prior to the teaching of chemical equilibrium, chemical reactions are presented as going to completion, meaning that all reactants are used up and converted into products. If the original amounts of reactants are not stoichiometrically equivalent, then one or more reactants might remain but then certainly one of the reactants is used up. This assumption is strongly reinforced by the kind of classroom work and homework problems given to the students. 459 K. Boersma et al. (eds.), Research and the Quality of Science Education, 459—470. © 2005 Springer. Printed in the Netherlands.

460

COLLEGE STUDENTS’ CONCEPTIONS OF CHEMICAL EQUILIBRIUM

The introduction of chemical equilibrium in the latter part of a two semester chemistry course exposes the student to the possibility of incompleteness and reversibility of chemical reactions. The students are confronted with the idea of two opposing chemical reactions occurring at the same time but for which no visible evidence is available. These concepts are at odds with well established conceptions that students have about chemical reactions. In order for conceptual change to take place, it is necessary for students to become dissatisfied with their present conceptions (Posner et al., 1982). In order to teach chemical equilibrium successfully, one must have knowledge of student preconceptions about the topic and what revisions are necessary. Furthermore, one has to understand the cognitive mechanisms responsible for the development of these personal theories and models, and why they are resistant to change (Glynn, Yeany & Britton, 1991). Many studies of misconceptions have measured pre- and post-study conceptions of students. To obtain optimal insight into the cognitive mechanisms of conceptual change, as well as the interaction of the teaching strategy with learning, a better approach is to monitor conceptual development throughout the learning process (White & Gunstone, 1992; Mazur, 1997; Berg, 2003). The purpose of this study was to describe the evolution of conceptions about chemical equilibrium based on observations/interactions with selected chemistry students using a well-tested instructional sequence. Specifically, the study aimed to answer the following research questions: 1. What conceptions do Philippine general chemistry students have before instruction about chemical reactions, regarding reversible reactions and the attainment of chemical equilibrium? 2. How do students’ conceptions of chemical equilibrium evolve during the different stages of teaching and as a result of diagnostic activities, POEs and other experiments, analogies, and small group discussion? 2. PHILIPPINE CONTEXT Unlike most other countries, the Philippines has 4 years rather than 6 years of secondary education. Subsequently, up to 40% of the age group attends some form of post-secondary education. This is one of the highest proportions among developing countries. Chemistry is taught in the third year of secondary education in daily lessons. Unfortunately 80% of the chemistry teachers are not qualified to teach chemistry but have been forced to teach the subject due to the lack of teachers with a chemistry background. As a result high school chemistry is taught as a set of definitions plus some tricks to be memorized to solve standard problems. Typical lessons consist of lecture/dictation followed by low level questions or fill-in-theblank exercises, as reported in observational studies (Somerset et al., 1999; Berg et al., 1998). Systematic development of concepts and reasoning is lacking. Therefore, this study, with its emphasis on concept development and reasoning with evidence, provided students with (what is for them) an unusual educational experience. Both

J. LOCAYLOCAY, E. VAN DEN BERG, M. MAGNO

461

in high school and university the sciences are taught in English, which is the second or third language of the students. 3. METHODOLOGY Prior to the study reported on here, the first author and a colleague experimented for several years in the teaching of chemical equilibrium. This resulted in a lesson sequence with laboratory activities, simulations, diagnostic activities, learning journals, and other components, most of which had been tried out and revised based on experience. Several elements of the sequence were adapted from van Driel (1990). A 35 item pre-post test went through several revisions; its validity with Philippine students was checked using interviews and assignments. The lesson sequence consisted of six lessons of two weekly 3 hour sessions each, for a total of 36 hours. The research was conducted in two General Chemistry II classes. One class consisted of 44 engineering students who took General Chemistry II as a compulsory course. The other class consisted of 31 students enrolled in either the B.S. Chemistry program or the Physics-Chemistry teacher education program. This second group was more selective and had chosen chemistry as their main subject. Both groups had taken General Chemistry I in the first semester. The typical age was 17-18 years. The data collected included pre-test and post-test results of all students in the two classes, written documents such as activity sheets, learning journals, and assignments. In order to describe students' conceptual development in detail, three students from each class were followed throughout the course through oral interviews, audiotapes and videotapes of small group and whole class discussions, and copies of all written work. The students were chosen to represent different cognitive levels: abstract, transitional, and concrete thinkers; the Test of Logical Thinking (Lawson, 1978) was used to establish this. Three engineering students were labeled X, Y, Z, while three Chemistry/Chemistry Education students were labeled A, B, and C. The classes were handled by one of the researchers (JRL) during the teaching of chemical kinetics and chemical equilibrium. 4. RESULTS AND DISCUSSION An understanding of chemical equilibrium requires that students understand reversible reactions and their dynamic nature. In the pre-test and the pre-interview, students had the pre-instructional conceptions that reactions go in one direction and that at least one of the reactants would completely react. The first lesson used experiments (van Driel et al., 1998) resulting in anomalous data. Experiment 1 involved the iron(III)-thiocyanate equilibrium and was intended to show the presence of unreacted iron(III) and thiocyanate ions and the incompleteness of the reaction. The use of anomalous data or discrepant events to create cognitive conflict causing students to be dissatisfied with their current conceptions and to adopt a target concept has been advocated by many educational researchers (Posner, Strike,

462

COLLEGE STUDENTS’ CONCEPTIONS OF CHEMICAL EQUILIBRIUM

Hewson & Gertzog, 1982; Hewson & Hewson, 1984; Champagne, Gunstone & Klopfer, 1985). Students A and B started to form concepts of reversible reactions: The reactants may have formed product and again formed back into reactants.1 (A, B) However, students C, X, Y, and Z still thought of reactions as complete and tried to explain the experiment to conform to their preconceptions. Chinn & Brewer (1993) reported findings that even when anomalous results are recognized, the student does not necessarily replace a prior idea with a better one. Students appear to use several strategies to discount the data and preserve their initial conceptions. The same results were obtained in this study. Student C tried to explain the results by assuming that the reaction is slow. Students X and Z explained the results by saying that at a certain point the reaction stops even if some reactants are still present. Student Y just ignored the observation. Maybe there is a certain point at which reactants stop their reaction. (X) There may be a time and factors when the reactants will not react even if one of the reactant is not completely consumed. (Z) There is a limiting reactant in the solution thus having an excess of ions that are not reacting. (Y) To introduce the reversibility of chemical reactions, the system of the tetrachlorocuprate(II) and the tetraamminecopper(II) complexes in water was used. [CuCl4]2- (aq) + 4NH3 (aq) '

[Cu(NH3)4]2+(aq) + 4Cl- (aq)

yellow

dark blue

After repeatedly adding drops of aqueous NH3 and HCl and observing the solution changing from yellow green to dark blue and vice versa, the students were asked to explain their observations in terms of the collision model of chemical reactions. The group discussions revealed that the students had formed conceptions of reversible reactions: We agree that this is a reversible reaction. But can the reaction be reversed, have we not added HCl to this solution? This is a reversible reaction. Reactants react to form the products and the product may 1

Quotations are verbatim, including language errors.

J. LOCAYLOCAY, E. VAN DEN BERG, M. MAGNO

463

form back to the reactants but it depends on how you manipulate the reaction. (A) However, their idea of reversibility was such that they thought the reversible reaction only occurs when one of the products of the reaction is added. In order for them to consider the possibility of both the forward and the reverse reactions occurring simultaneously even without such additions, students were asked to add a reacting species until a color intermediate to the blue and the green color was obtained. The constancy in color of the solution despite the presence of both the reactants and the products could have been explained by assuming that both reactions take place at the same time and at the same rate. However, the students in their discussion did not arrive at this explanation. They reverted to their concept of complete reactions and reasoned that the reaction was not completed because there was not enough of the reactant. There is not enough concentration of HCl therefore there is no collision between ions in order for the reaction to occur again to form back the reactants to its original form. (X) Some of the reactants may not have formed products and will still be capable of reacting if one of the reactants is added. (A) Past research has shown that students have difficulty with the microscopic level and develop many misconceptions because of these difficulties (e.g. Garnett et al., 1995; Nakhleh, 1992). Lawson et al. (1993) suggested that students are better able to integrate concrete examples into existing knowledge than abstract ones and that analogies can enhance the intelligibility and plausibility of the explanations. The second lesson made use of an analogy, adapted from the simulations proposed by Huddle et al. (2000) and Harrison and Buckley (2000). The analogy used blue- and pink-sided disks to represent reactant and product molecules, respectively. The activity is supposed to show in concrete terms the dynamic nature of chemical equilibria and to explain the constant concentration of reactants and products at equilibrium. From their activity sheets, the six students had the correct concept that as the system approaches equilibrium, the concentration of the reactant decreases while the concentration of product increases, and that the rate of the forward reaction decreases while that of the reverse reaction increases. When the system attains equilibrium, the rates of the forward and reverse reactions are equal, the concentration of the products and the reactants remain constant but are not necessarily equal, and the ratio of the products to the reactants, [P]/[R], is constant. Class results shown in Table 1 indicate that a majority of the students in both classes acquired the scientific concepts regarding the characteristics of systems at equilibrium.

464

COLLEGE STUDENTS’ CONCEPTIONS OF CHEMICAL EQUILIBRIUM

Table 1: Post-test results and percent gain on questions about properties of systems as they approach and reach chemical equilibrium

As system approaches equilibrium [2NO (g) + Cl2 (g) ' 2NOCl (g) ] Concentrations of products increase Concentrations of reactants decrease NO (g) Cl2 (g) Rate of forward reaction decreases Rate of reverse reaction increases At equilibrium Reactions go on Concentrations of products and reactants remain constant Concentration of product is not equal to concentration of reactants [NOCl]2/ [NO]2[Cl2] = Kc Kc does not change with change in concentration of reactants

Post-test Chemistry

gain

Post-test Engineering

gain

100%

31%

82%

37%

97% 100% 68% 74%

25% 50% 15% 21%

59% 57% 45% 61%

-18% 12% 11% 29%

94%

53%

93%

43%

94%

38%

89%

14%

84% 97%

37% 16%

73% 91%

62% 30%

71%

52%

39%

19%

The entries in students' learning journals show how the analogies helped. The activity was very effective. The disks are the best way to show us in larger view what was happening in the microscopic level. (C9) The activity which helped me learn was using the disks. It was a helpful analogy of a reversible reaction. Through comparing the number of disks per time, I was able to infer about what would happen on the concentration of the reactants and products as it reaches equilibrium. (C2) The activity with the disks and the exercise helped the change. The total number of disks started with and the varying concentrations given in the exercise showed results which support that the reaction will always reach equilibrium at a similar constant and with the same product and reactant ratio. (E30) At this point, the students were already considering the dynamic nature of a chemical equilibrium and were attributing the constancy of the macroscopic properties to the fact that the two reactions are occurring at the same rate. However, in interviews after the lesson, it can be seen that the conceptions of students B, C, Y,

J. LOCAYLOCAY, E. VAN DEN BERG, M. MAGNO

465

and Z of complete reactions still competed with their conceptions of reversible reactions. T:

C: T: C: T: C:

If we have for example a reversible reaction 2 SO2 (g) + O2 (g) ' 2 SO3 (g) This is a reversible reaction. If we place in a one-liter container one mole of SO2 and one mole of O2, what will happen as I mix the two? It will react to form two moles of SO3. So one mole of SO2 will react to form two moles of SO3? One mole Then once that forms one mole of SO3? It will react back to form the reactants.

C's immediate answer of two moles of SO3 shows that she was looking at the coefficients in the balanced equation and thinking immediately of complete reactions. She had the conception of a reverse reaction occurring, but she was not yet sure as to when it occurs. She had the wrong conception that the concentration of the reactants decreases and then increases again as the reverse reaction occurs. T: Then, you let the reaction proceed. What will happen to the concentrations of the reactants as the reaction proceeds? C: It will become lesser, Ma’am. T: How about the concentration of SO3? C: It will be more concentrated as the reaction proceeds. T: Okay, these reactants become lesser and the products become more concentrated. Will it keep on decreasing? C: Yes. T: So, it will all be used up because you said it will keep on decreasing. C: No, Ma’am. T: So what will happen to it? C: It will just become lesser. As the product becomes more concentrated, so it will react back again. It will react again to form the reactants. So the reaction is reversible. T: So what happens to the concentration of the reactants? C: It will become more concentrated until it reached a point, the concentrations are both the same. When asked similar questions, student Z's immediate thought was to consider the reaction as complete. However, since students had already been exposed to reversible reactions and to the experiment that showed the presence of both reactants and products, student Z seemed confused as to how this came about. His idea was to

466

COLLEGE STUDENTS’ CONCEPTIONS OF CHEMICAL EQUILIBRIUM

think of reactions as oscillating.2 It seems that the completeness of all reactions is a difficult misconception to remedy. T: SO2 reacts with O2 to form SO3 in a ratio of 2:1:2. The reaction is reversible. I start with 2 moles of SO2 and 1 mole of O2. I give it some time. What will I find? Z: 2 moles SO3 T: Just 2 moles of SO3? Z: There is still SO2 and O2 gas. T: Where did I get my 2 moles SO3? Z: From the 2 moles of SO2 and the 1 mole of O2, so this is 2,2 then 1,1 and this is 2. T: Then where will I get the SO2 that you said is still present? Z: Because this is reversible Ma’am so there should be some unreacted. T: So if there are some of these unreacted, why will I get 2 moles of SO3? Z: Then this one here, this 2 moles of SO3 will dissociate and form SO2 and O2…it will already be in excess, I don’t know. It is again present. Then afterwards this will react to form 2 moles SO3. The lessons on manipulating equilibrium systems used experiments and exercises as the basis for group discussions to determine the effect of changes in concentration and temperature on a system at equilibrium. The discussions also examined the effect of a change on the basis of the equilibrium law, rates of reactions, and Le Chatelier’s Principle. All of the students developed the correct concept that an increase in the concentration of the reactants increases the rate of the forward reaction while an increase in the concentration of the products increases the rate of the reverse reaction. In their learning journals, students A, B and C wrote that if the concentration of the reactants is increased, this will shift the equilibrium forward, but the Kc value won’t change. They attributed this learning to the activity using the blue and pink disks. An example of an entry from student A is given below. A: The activity on the blue and pink disks indicates that an increase in number of discs increases the number of both discs at equilibrium while maintaining the same Kc value. Student Z had the misconception that when a substance is removed from the system, the reaction goes faster in order to replace the removed substance. The term “favored” may have caused the student to think that when a change favors a reaction, it means that the rate of that reaction always increases. The entry in his learning journal showed this.

2

The conception of oscillating reactions has been reported by Bergquist and Heikenen (1990).

J. LOCAYLOCAY, E. VAN DEN BERG, M. MAGNO Z

467

Removing a substance may also make the reaction faster to replace the removed substance but it will still reach to a point where the rates are equal.

Hackling and Garnett (1985) reported similar misconceptions. They found from their interviews that the students responded that the rate in one direction would be increased and the reverse direction decreased in order to explain the changes predicted by Le Chatelier’s Principle. This misconception can be addressed in the analogy by asking the students to look at what happens to the rates of forward and reverse reactions when a product is removed. From the number of blue disks and pink disks that are changing per minute students in this study could see the effect on the two rates. The effect of changes in temperature on systems at equilibrium was studied with the use of the NO2-N2O4 system. N2O4 (g) ' 2 NO2 (g) UH = +57.2 kJ mol-1 colorless reddish brown All 6 students were able to identify which direction is favored by an increase in temperature. They were able to relate the more intense color of the tube to the formation of more NO2 gas. They correctly inferred that the exothermic reverse reaction is favored by a decrease in temperature. However, they had difficulty with their explanation as to why the reaction is favored, as shown in the video transcript of a discussion between students A and B. A:

B: A: B: A:

B:

We have an increase in temperature, which increases the rates of both endothermic and exothermic reactions. How then can you explain the results of the experiment? You said that the exothermic reaction is favored by a decrease in temperature, however, we have agreed theoretically that endothermic and the exothermic reactions are favored by temperature increase, right? Because an increase in temperature increases the number of molecules with kinetic energy higher or at least equal to the activation energy barrier. Thereby aiding or increasing the rate of the endothermic and exothermic reaction. How are we to reconcile this concept, this theoretical concept with the results of the experiment? Let me ask, what will happen to the kf and kb values or the rate constants of the backward reaction if we will increase the temperature? It would also increase. It will also increase the… The rate constants of both forward and backward reactions. So, it will increase both the rate constants of the forward and backward reaction What is the effect of the increase of both, the rate constants of the forward and the rate constant of the backward reaction to the concentration of the reactants and products? More or less the concentration of the products and reactants will also increase because the rate constants have been increased.

468

COLLEGE STUDENTS’ CONCEPTIONS OF CHEMICAL EQUILIBRIUM

Student B did not take into account the fact that both reactions are taking place at the same time, thus the resulting concentrations will be affected by the change in rates of both reactions. Student A had developed the correct concept that both rates will increase but the increase will not be the same. The increase in rate of one reaction is greater than the increase in rate of the reverse reaction. Students X, Y, and Z also had difficulty explaining how the lowering of the temperature favors any reaction. They knew that when the temperature increased, the rates of both reactions would increase. They couldn’t understand how the reverse reaction was favored by a decrease in temperature. They couldn’t seem to picture an endothermic forward reaction occurring at the same rate as an exothermic reverse reaction. In an interview conducted after the lesson, Student Z was asked the following questions based on the reaction. A(g) + B (g) ' C (g) UH is positive T: What will happen to the rate of the forward reaction if you increase the temperature? Z: Faster T: What will happen to the rate of the backward reaction? Z: Slow, less.. seems slower. T: If I have it at 200o and I have it at 500o, what will be the rate of the forward reaction at 200o compared to the rate of the reaction at 500o? Z: The rate of the forward reaction at 500o will be greater than the forward reaction at 200o. T: How about the rate of the backward reaction at 200o compared to the rate of the backward reaction at 500o? Z: For the backward, the rate of the backward reaction at 200o is greater than the rate at 500o. T: Why is it greater? What determines the rate? Z: Concentration. T: One is concentration. What happens to the number of molecular collisions as you increase the temperature? Z: It will increase.T: Okay. So which will have more molecular collisions for the backward reaction, at 200 or 500? Z: At 200. T: Why? You said the number of molecular collisions will increase? Z: The backward reaction is exothermic. So if you decrease the temperature, the effect is to make the backward faster because it releases heat. Student Z had a different conception for effect of temperature on rates of an endothermic reaction taken separately as opposed to its effect on an endothermic reaction, which is part of a reversible reaction.

J. LOCAYLOCAY, E. VAN DEN BERG, M. MAGNO

469

5. CONCLUSIONS AND IMPLICATIONS OF THE STUDY A first conclusion is that the various student conceptions regarding chemical equilibrium of Philippine students are very similar to those of students elsewhere in the world (Hackling & Garnett, 1985; Bucat & Fensham, 1995; Huddle & Pillay, 1996). This holds both for preconceptions and for intermediate conceptions which develop as lessons proceed (van Driel et al., 1998a). Misconception problems with regard to this topic seem rather universal, just like alternative conceptions with regard to key concepts in physics (Thijs & Berg, 1995). Somehow, influences of language, culture, and educational history on the content of conceptions are small. The conceptual change methods promoted in high-income countries seemed to work out in the Philippines in the sense that they produced a rather similar mix of scientific and alternative conceptions as compared to studies elsewhere (e.g. van Driel et al., 1998a; van Driel et al., 1998b). Detailed preconceptions and intermediate conceptions have been described in this paper. Knowledge of these intermediate conceptions will help to improve the lesson sequence and the teaching and learning of equilibrium concepts, particularly if the information is used in real time during the learning process in the classroom (Mazur, 1997; Berg, 2003). Results of the study have shown that evolution of students’ conceptions show both progressions and regressions. To prevent regressions from occurring, multiple strategies must be utilized. The analogy used in this study helped students visualize the simultaneous occurrence of the forward and reverse reaction. However, students’ preconceptions of complete reactions made the conceptual change to simultaneous reversible reactions difficult. Strategies that allow improved visualization of the dynamic nature of chemical equilibrium are needed. REFERENCES Berg, E. van den, Alfafara, R., & Dalman, T. (1998). Case studies of science and mathematics teaching in the Philippines and lessons for teacher and school development. National Association of Research in Science Teaching. San Diego, U.S.A., April 19-22, 1998. Berg, E. van den (2003). Teaching, Learning, and Quick Feedback Methods. The Australian Science Teaching Journal, June 2003, 28-34. Bergquist, W. & Heikkinen, H. (1990). Students’ Ideas Regarding Chemical Equilibrium. Journal of Chemical Education, 67(12), 1000-1003. Bucat, R. & Fensham, P (1995). Teaching and Learning about Chemical Equilibrium. In R. Bucat and P. Fensham (Eds.) Selected Papers in Chemical Education Research. The Committee on Teaching of Chemistry of the International Union of Pure and Applied Chemistry. Camacho, M. & Good, R. (1989). Problem Solving and Chemical Equilibrium: Successful and Unsuccessful Performance. Journal of Research in Science Teaching, 27(2), 157-172. Champagne, A.B., Gunstone, R.F., & Klopfer, L.E. (1985). Effecting changes in cognitive structure among physics students. In: West, L., Pines, L. Cognitive structure and conceptual change. Orlando: Academic Press, 163-187. Chinn, C.A. & Brewer, W.F. (1993). The role of Anomalous Data in Knowledge Acquisition: A Theoretical Framework and Implications for Science Instruction. Review of Educational Research, 63, 1-49. Driel, J. van (1990). Betrokken bij Evenwicht (Involved with Equilibrium). Utrecht, CD-β Press.

470

COLLEGE STUDENTS’ CONCEPTIONS OF CHEMICAL EQUILIBRIUM

Driel, J. van, Vos, W. de, & Verloop, N. (1998a). Introducing Dynamic Equilibrium as an Explanatory Model. Journal of Chemical Education, 76(4), 559-561. Driel, J. van, Vos, W. de, & Verloop, N. (1998b). Developing Secondary Students’ Conceptions of Chemical Reactions: The Introduction of Chemical Equilibrium. International Journal of Science Education, 20(4) , 379-392. Driel, J. van, Gräber, W. (2003). The teaching and learning of chemical equilibrium. In: J. Gilbert et al Chemical Education: Towards Research-based Practice. Dordrecht (Netherlands): Kluwer. Garnett, P.J., Garnett, P.J. & Hackling, M.(1995). Students’ Alternative Conceptions in Chemistry: A Review of Research and Implications for Teaching and Learning. Studies in Science Education, 25, 69-95. Glynn, S.M., Yeany, R.H. & Britton, B. (1991). A Constructive View of Learning Science. In S.M. Glynn, R.H. Yeany and B. Britton (Eds.), The Psychology of Learning. Hillsdale, New Jersey: Lawrence Erlbaum Associates. Gussarsky, E. Gorodetsky, M. (1988). On the chemical equilibrium concept: Constrained word associations and conceptions. Journal of Research in Science Teaching, 25(5), 319-333. Hackling, M.W., & Garnett, P.J. (1985). Misconceptions of chemical equilibrium. European Journal of Science Education, 7, 205-214. Harrison, J.A. & Buckley, P. D. (2000). Simulating Dynamic Equilibria: A Class Experiment. Journal of Chemical Education, 77(8), 1013-1014. Hewson, P.W., & Hewson, M.G. (1984). The role of conceptual conflict in conceptual change and the design of science instruction. Instructional Science, 13, 1-13. Huddle, P. A. & Pillay, A. E. (1996). An In-depth Study of Misconceptions in Stoichiometry and Chemical Equilibrium at a South African University, Journal Of Research in Science Teaching, 33(1), 65-78. Huddle, P.A., White, M. & Rogers, F. (2000). Simulations for Teaching Chemical Equilibrium. Journal of Chemical Education, 77(7), 920-926. Lawson, A.E. (1978). The Development and Validation of a Classroom Test of Formal Reasoning, Journal of Research in Science Teaching, 15(1), 11-24. Lawson, A.E., Baker, W., DiDonato, L., Verdi, M. & Johnson, M. (1993). The Role of Hypotheticodeductive Reasoning and Physical Analogues of Molecular Interaction in Conceptual Change. Journal of Research in Science Teaching, 30(9), 1073-1085. Mazur, E. (1997). Peer Teaching. New York, Wiley. Nakhleh, M.B. (1992). Why Some Students Don’t Learn Chemistry ? Chemical Misconceptions. Journal Of Chemical Education, 69(3), 191-196. Posner, G.J., Strike, K.A., Hewson, P.W. & Gertzog, W.A. (1982). Accommodation of a Scientific Conception: Towards a Theory of Conceptual Change. Science Education, 66(2), 211-227. Somerset, A., Alfafara, R. & Dalman, T. (1999). Effective and ineffective pedagogy compared: Science and mathematics teacher needs assessment study, Part II. Cebu City (Philippines): University of San Carlos Science and Mathematics Education Institute. Thijs, G.D., & Berg, E. van den (1995). Cultural factors in the origin and remediation of alternative conceptions in physics. Science and Education, 4 , 317-347. White, R.T., & R. Gunstone. (1992). Probing Understanding. London: Falmer Press

PARALLEL CONCEPTIONS IN THE DOMAIN OF FORCE AND MOTION

SUSANN HARTMANN, HANS NIEDDERER University of Bremen, Germany

ABSTRACT The basic assumption, for which we try to provide evidence in this paper, is that students always use multiple explanations before and after teaching. Other studies also give evidence of competing conceptions used in one content area, yet often a variation of context is seen as the cause of multiplicity. The study presented here focuses on individual answers within one context. A total of 47 students from grade 7 up to university level participated in interviews which dealt with three qualitative tasks in the domain of force and motion. As the interview technique was based on waiting and asking questions of specification without giving additional information, the context is assumed to be stable when dealing with one task. Data interpretation focused on 27 students from four different schools (age 16), who were interviewed before and after they attended a class in mechanics. Results show that most answers, even with respect to one task, reveal multiple explanations.

1. THEORETICAL BACKGROUND Students’ ideas about physical phenomena have been studied and described for many years. While some researchers focus on the difference between everyday life and scientific thinking (Duit et al., 1981; Reif & Larkin, 1991), others emphasize that both perspectives have much in common (Westra, 2003). The question of coherence and stability was raised in connection to this discussion. How unstable and fragmented is thinking in everyday life? Much research has been undertaken to clarify this question but no agreement has been reached so far. Viennot’s statement gives a short insight into this debate: “...if common knowledge is indeed made up of bits and pieces, they are rather large; zones of coherence have, in fact, emerged from what first seemed a muddle of unrelated errors.” (Viennot, 2001, p. 11). If we think of students’ knowledge as scattered in bits and pieces (diSessa, 1993), multiplicity may seem natural. In many empirical studies multiple conceptions were found to be related to context variations (Gunstone & Watts, 1985; Gerdes & Schecker, 1999). But is it possible to find competing conceptions in the answer of one person to one task if we look at conceptions which have the magnitude of explanations? An explanation is more than a small piece of thought as it combines several elements. Multiplicity in this paper is defined as the ability to form more than one explanation to a given task; 471 K. Boersma et al. (eds.), Research and the Quality of Science Education, 471—481. © 2005 Springer. Printed in the Netherlands.

472

PARALLEL CONCEPTIONS IN THE DOMAIN OF FORCE AND MOTION

in this paper the terms multiple explanations and parallel conceptions are used interchangeably. To us an explanation can be viewed as a cluster of elements that are linked to each other. Parallel conceptions have also been described as multiple cognitive layers in students’ understanding (Petri & Niedderer, 2003). In other studies parallel conceptions were also found. Tytler (1998) describes the answer of a student called “Noel”: “In generating all these explanations, he is drawing on a cocktail of conceptions to which he accords varying allegiance. (...) Noel is not only generating different conceptions for different tasks, but is doing so within tasks” (p. 911). Maloney and Siegler (1993) state: “For years after encountering physics concepts, students may possess not a single coherent understanding but rather a variety of alternative understandings that coexist and compete with one another” (p. 283). Taber (2000) concludes from his observations “...an individual learner can simultaneously hold in cognitive structure several alternative stable and coherent explanatory schemes that are applied to the same concept area” (p. 399). We know little from these studies about context variations as a possible cause for multiplicity. It is therefore difficult to tell whether parallel conceptions in one concept area, or in respect to one task, were due to shifts in context. Contexts may differ in place and surrounding (Roth, 1996), in time (length of an interview), in interactions, and thematically (task). In designing our study we kept all of these 4 parameters as constant as possible. We are aware that slight variations, especially those due to interactions, are inevitable. Aim Our main research aim was to find out whether or not students’ answers to questions about force and motion revealed the existence of multiple explanations, even if the context did not change. As our interview technique carefully avoided giving additional information, but relied on waiting and asking questions of specification, we have assumed that the context is stable when dealing with one task. Relevance If students usually formulate more than one explanation with respect to one task in a given context when they are given sufficient time, there is an urgent need to consider multiple points of view in a classroom situation. Teaching strategies that allow students to consider competing answers and to develop criteria for judging their appropriateness are then more valuable than other strategies. With respect to multiplicity, learning means to develop the abilities to distinguish different views and to find the strengths and weaknesses of each.

S. HARTMANN, H. NIEDDERER

473

2. METHODS AND SAMPLE Design and Data Semi-structured interviews were used in order to find out whether students' answers to questions about force and motion can be interpreted in terms of multiple explanations. Twenty seven students in four upper secondary schools (year 11) participated before and after taking a course in Newtonian mechanics. Additionally, the students filled out a questionnaire based on the Force Concept Inventory (FCI) both before and after the course. Four months passed between the interviews. After a further four months, six of the original 27 students were interviewed a third time. Data were also collected about the content of the classes and frequently used teaching methods. All students had chosen Physics as their main subject, and all classes dealt with mechanics (there were 50 to 80 hours of teaching in the four month interval between interviews). Also, ten students at university level and ten students in grade 7 were interviewed.

Grade 7

Table 1. Overview of data and time schedule Interviews with 10 students

March 2002 Grade 11, before teaching Interviews and tests with 27 students August (BT) 2001 Grade 11, after teaching Interviews and tests with the same 27 January (AT) students 2002 Grade 11, long after Follow-up interviews with 6 of the May 2003 teaching (LAT) original 27 students University Interviews with 10 students June 2001 Interview tasks Each task was distributed together with a graphical representation. Three tasks were used in all interviews: 1. A plastic ball being tossed up and falling down. 2. A boy tobogganing down a slope and then gliding horizontally. 3. A body moving (without friction) on a linear air track with equal balancing forces. For each task the same questions were asked: How does the object move? Why does it move like this? Interview Method The basic idea of the interview technique was to prompt students’ thinking and statements without giving additional information. This was in order not to vary the context. The interview strategy was based on: • Additional time (“wait time”) During the interviews the students were given as much time as possible to think

474

• • •

PARALLEL CONCEPTIONS IN THE DOMAIN OF FORCE AND MOTION on their own. This is seen as giving encouragement without providing additional information. Questions of specification For example, “Can you explain this in more detail?” To draw forces as vectors Each time the students talked about forces they were asked to indicate the forces they mentioned on the prepared pictures of the experiment. Probing acceptance (Jung, 1992; Wodzinski, 1996) In an additional part of the interview, students were confronted with 3 different explanations which were given as drawings and written explanations.

Each interview consisted of 3 tasks and lasted 45 minutes. Categories Table 2. Overview of categories Type name / values example 1 explanatory element “The ball moves upwards because it received a force upwards.” 1a A = non-scientific P = scientific explanatory element 2 explanation

2a - very clearly distinguishable - clearly distinguishable - distinguishable

The explanatory element above was labelled A, as from a scientific viewpoint forces cannot be transferred. “The ball moves upwards because of the force it got from my hand. While moving upwards it loses force and as soon as there is no force left, gravitation can act on the ball and it is because of gravitation that the ball comes down again.” For criteria, see section on Differentiating Explanations.

used for general capture of content figure 2

general capture of multiplicity

figure 1

In order to interpret the data without needing full transcriptions of all interviews, categories of responses were developed. In a first step, categories were used for a general capture of content and the number of explanatory elements expressed by students in different tasks. These categories were developed in a qualitative approach using 10 interview transcripts, together with background knowledge about students’ conceptions from the literature. The intercoder-reliability of these categories was tested. When two different persons categorized 24% of all interviews, 88% of all categories given were identical.

S. HARTMANN, H. NIEDDERER

475

In a second step each category was classified as A or P; P was used for “physics element of explanation”, A for “alternative element of explanation”. In a third step, it was decided whether several explanatory elements formed one explanation or more than one. If a student's answer had more than one explanation, we classified this as parallel conceptions or multiple explanations and defined how different the explanations were (see also paragraphs Differentiating Explanations and Example). An overview of the categories is shown in the table above. Differentiating Explanations When deciding whether or not an answer shows multiplicity, criteria have to be found to group the explanatory elements into one, or more than one, explanation. The most obvious criterion is contradiction: if two explanatory elements contradict each other, they belong to different explanations. How do we decide whether contradiction exists? There are two possibilities, a normative or an ideographic point of view (Driver et al., 1996, p. 58). One takes the position of a physicist while the other tries to figure out if the students themselves distinguish their explanations. Both perspectives differ in result. A normative approach might see students’ understanding of physics as related to two worlds, the world of everyday life and the world of physics. From the students’ own point of view all elements might belong together regardless of a physicist’s view. Our data interpretation distinguishes between the students’ point of view and a physicist’s point of view, though both interpretations are but our own constructions aiming at taking different standpoints. From both perspectives we looked at contradictions, parallelism, negations, and the use of key expressions like force, energy, momentum. If the distinction between two explanations was similar from both perspectives, we considered the explanations to be “very clearly distinguishable”. If two explanations clearly contradicted each other but only from one perspective, we thought them to be “clearly distinguishable”. If only some elements were found to be distinguishable and only from a physicist’s point of view, we labelled them “distinguishable”. Example On average, 6 explanatory elements were found in individual answers to the first interview task. The following example shows two elements of explanation that contradict each other. A student, Lajosch, explains the upwards movement of a tossed up ball (task 1) after the ball left the hand. One explanatory element of the first explanation: Lajosch: “Well in relation, of course, on the way upwards my force is stronger [than gravitation], else the ball would not move up, the ball would fall down.” (after 8 minutes) One explanatory element of the second explanation:

476

PARALLEL CONCEPTIONS IN THE DOMAIN OF FORCE AND MOTION Lajosch: “As soon as the ball leaves my hand it slows down. Therefore gravitation has to be strongest even while the ball moves up.” (after 12 minutes)

Lajosch’s explanations show a typical and well known conception which is often referred to as an impetus conception. Both explanations share the assumption that: “The ball moves upwards because it has received a force”. Yet the elements of explanation also contradict each other. While in the first element the conception “the strongest force determines the direction of motion” plays a prominent role, this conception is negated in the second element, where gravitation is assumed to be the strongest force. Therefore, the direction of movement is not in the same direction as the strongest force. When Lajosch draws the forces as vectors, the difference is visible. In his first picture, the resultant force points upwards, while in the second picture the resultant force points downwards. From a physicist’s point of view, both explanations are not correct, as forces are never given to objects. Still, the second explanation is nearer to a Newtonian view, as a relation between force and acceleration is postulated. From the student’s own perspective both explanations are also different, and for a while Lajosch tries to argue for one explanation being more convincing than the other. In the end Lajosch sighs and states: “I still don’t know which of my two models I found more convincing”. (after 17 minutes)

80%

% of students

23% 36%

60%

37%

28%

40%

20%

7%

30%

37% 32%

27%

9%

12%

11%

11 BT

11 AT

11 LAT

very clearly dist. clearly dist. distinguishable

22% 20%

7%

0% Grade 7

University

Figure 1. Percentage of cases showing more than one explanation N (grade 7) = 3x10, N (11 BT, AT) = 3x27, N (11 LAT) = 3x6, N (university) = 3x10

S. HARTMANN, H. NIEDDERER

477

Because of Lajosch’s own distinction of his two explanations (ideographic perspective) and because one explanation is nearer to a Newtonian view (normative perspective), his two explanations are labeled “very clearly distinguishable”. 3. RESULTS How often is more than one explanation found in the answer of one student to one task? In the following chart BT stands for “before teaching”, AT for “after teaching” and LAT for “long after teaching”. For definition of “distinguishable”, “clearly distinguishable” and “very clearly distinguishable” see paragraphs above. In order to find the number of cases that are the basis for this chart, the number of interviewed students has to be multiplied with the number of interview tasks, as each respondent could answer each task by formulating one or more explanations. Also note that students in grade 7 had not yet attended any physics classes, and students at university level had not dealt with physics for at least five years. Except for grade 7 and grade 11 (LAT), more than one explanation is found in over 70% of all answers. If a respondent explained the movement of an object and the interviewer waited and showed interest, the chances of hearing a second explanation were high. Third explanations were only rarely found. Multiple explanations show students’ ability to argue in more than one way. As the results show, multiple explanations are not caused by context variations alone, as they are also found within one context. If we imagine a grade 11 teacher, s/he can be fairly sure that more than half of the students are able to explain the movement of an object in two different ways. In grade 7 nearly every second child has the ability to form two different explanations, but the explanations are not as clearly distinguishable as in grade 11. University students who study different subjects but not physics, still have the ability to think of more than one way to explain a movement. Explanations in this group are often called “clearly distinguishable” which means that two explanations are either distinguishable from the point of view of a physics teacher (normative perspective) or from the point of view of the student him- or herself (ideographic perspective). Looking at the changes students made while answering the multiple choice test after attending the class also adds to the aspect of multiplicity. The written questionnaire was based on the FCI (Force Concept Inventory), a multiple choice test with qualitative questions about Force and Motion given to 40 students in grade 11, both before and after teaching. After teaching, 59 “cross outs” (changes) occurred. Each question has five possible answers, of which one is Newtonian while the others represent alternative conceptions. When students changed (crossed out) an answer, because on second thought another answer appeared more plausible, this indicates parallel conceptions. An analysis of crossed out answers shows that 68% of all “cross outs” after teaching were the answers chosen before teaching, 26% of the “cross outs” were Newtonian answers, and 6% were those answers the student

478

PARALLEL CONCEPTIONS IN THE DOMAIN OF FORCE AND MOTION

had already crossed out before taking the class. Only the Newtonian “cross outs” suggest novelty while all other “cross outs” suggest stability and change at the same time: “stability” because the answers chosen before teaching are again considered after teaching and “change” because these answers are no longer found to be the most plausible. It is remarkable that no other “cross outs” occurred. After four months and with 19 questions each with five possible answers, it is unlikely that students remembered what they had chosen before teaching. Even though a crossed out answer does not reveal the reason or argumentation behind the choice, these “cross outs” do indicate multiple explanations. Change and stability were also noted in the interview data. On average, about half of the explanatory elements were identical in the interviews before and after teaching. How many of the used elements are scientific ones? The next chart shows the number of scientific elements in relation to all elements of explanation.

80% 70%

70%

60%

47%

48%

41%

40%

32% 20%

30% LAT

12% AT

0% BT

test interview acceptance

Figure 2. Percentage of scientific elements in explanations Figure 2 shows the percentage of scientific elements for the multiple choice test, the oral interview, and the probing-acceptance part of the oral interview. Only the three interview questions were considered in this comparison. It is obvious that we misjudge students’ knowledge if we look at the multiple choice test results only. Students who did not choose the scientific answer in the multiple choice test often gave explanations expressing both scientific and nonscientific explanatory elements in an interview situation. The difference was especially high in grade 11 before teaching. Students in an interview situation often argued in more than one way by using alternative and scientific elements of

S. HARTMANN, H. NIEDDERER

479

explanation. Although they expressed more scientific elements in interviews after teaching, their ability to recognize the physics argumentation in the test improved even more. As expected, students used more scientific elements in their explanations after teaching. In the interviews the number of scientific elements increased from 32% before teaching to 47% after teaching. Four months after the end of teaching the answers show less multiplicity (see fig. 1), yet the percentage of scientific elements of explanation increased to 70%. The six students interviewed a third time had shown average results in the first interview. So, there seems to be a learning process related to the basic understanding of force and motion even after the teaching related to that content stopped. This goes together with findings of Häußler (1983) who found a similar increase in basic understanding from after teaching to long after teaching. It seems that learning at this stage means reducing multiplicity to what is assumed to be the scientifically correct answer. Still, non-scientific elements are also used to explain the movements of tossed balls, gliding sledges, and frictionless movement on air tracks. Students often used both, scientific and non-scientific elements within one explanation. This was also pointed out by Wittmann (2002): “When studying student understanding of a physics content area, an individual student response is often interpreted in terms of a single model or misconception. (...) Results described above show that such a description is not necessarily productive, nor is it complete. Instead, the results show that students use reasoning elements both inappropriately and appropriately at the same time.” (p. 115) 4. CONCLUSIONS We may summarize the main result as follows: Students’ individual answers with respect to one task (one context) show multiple explanations. Therefore, we should no longer believe that we can describe the knowledge of a person sufficiently by just looking at his or her spontaneous answer. But what shall we do with this knowledge of multiplicity? It is often assumed that if someone argues in more than one way, his or her answers cannot make sense. This is due to the belief that there is only one right answer to each question. To argue in more than one way is therefore interpreted as a weakness. Shall we sympathize with this opinion and encourage students to reduce multiplicity to one scientific answer? For people with a desire for orientation this might be a good way (Pantaleo, 1997). Yet, looking for one answer only, combined with the assumption that the teacher “has it”, may lead to reproduction instead of thinking. Deeper knowledge or “in-sight” is gained if students learn to distinguish between different explanations and to develop criteria to name their explanations' strengths and weaknesses. Further consequences related to learning are presented by Hartmann (2004). No one would deny that we should encourage students’ own reasoning

480

PARALLEL CONCEPTIONS IN THE DOMAIN OF FORCE AND MOTION

abilities instead of simply training them to memorize scientific concepts; taking multiplicity seriously is one way to do this. REFERENCES DiSessa, A. A. (1993). Towards an Epistemology of Physics. Cognition and Instruction, 10 (2 & 3), 105-226. Driver, R., Leach, J., Millar, R. & Scott, P. (1996). Young People’s Images of Science. Milton Keynes: Open University Press. Duit, R., Jung, W. & Pfundt, H. (1981). Alltagsvorstellungen und naturwissenschaftlicher Unterricht. Köln: Aulis. Gerdes, J. & Schecker, H. (1999). Der Force Concept Inventory. Der Mathematische und Naturwissenschaftliche Unterricht, 5/99, 283-288. Gunstone, R. & Watts, M. (1985). Force and Motion. In R. Driver, E. Guesne & A. Tiberghien (Eds.), Children’s Ideas in Science. Philadelphia: Open University Press. Hartmann, S. (2004): Erklärungsvielfalt. Doctoral Dissertation, University of Bremen. Häußler, P. (1983). Wie sich physikalisches Wissen im Gedächtnis der Lernenden verändert. Lernzielorientierter Unterricht, 4, 83, 27-37. Jung, W. (1992). Probing acceptance: A technique for investigating learning difficulties. In R. Duit, F. Goldberg & H. Niedderer (Eds.): Research in Physics Learning - Theoretical Issues and Empirical Studies. Kiel: IPN, 278-2295. Maloney, D. P. & Siegler, R. S. (1993). Conceptual competition in physics learning. International Journal of Science Education, 15, 283-295. Niedderer, H. & Goldberg, F. (1995). Lernprozesse beim elektrischen Stromkreis. Zeitschrift für Didaktik der Naturwissenschaften, 1 (1), 73-86. Niedderer, H. (2001). Physics Learning as Cognitive Development. In R. H. Evans, A. M. Andersen & H. Sørensen (Eds.): Bridging Research Methodology and Research Aims. Student and Faculty Contributions from the 5th ESERA Summerschool in Gilleleje, Danmark. The Danish University of Education. Page 397 – 414. (ISBN: 87-7701-875-3). Pantaleo, G. (1997). Explorations in Orienting vs. Multiple Perspectives. Lengerich: Pabst. Petri, J. & Niedderer, H. (1998). A Learning Pathway in High-School Level Quantum Atomic Physics. International Journal of Science Education 9, 1075-1088. Petri, J. & Niedderer, H. (2003). Atomic Physics in Upper Secondary School: Layers of Conceptions in Individual Cognitive Structure. In D. Psillos, P. Kariotoglou, V. Tselfes, E. Hatzikraniotis, G. Fassoulopoulos & M. Kallery (Eds.). Science Education in the Knowledge-Based Society, Kluwer Academic Publishers, 137-144. Reif, F. & Larkin, J. H. (1991). Cognition in Scientific and Everyday Domains: Comparison and Learning Implications. Journal of Research in Science Teaching 28 (9), 733-760. Roth, M. (1996). Situated Cognition. In R. Duit & C. von Rhöneck (Eds.), Lernen in den Naturwissenschaften. Kiel: IPN. Schnotz, W., Vosniadou, S. & Carretero, M. (Eds.), (1999). New Perspectives on Conceptual Change. Amsterdam: Pergamon. Taber, K. S. (2000). Multiple frameworks?: Evidence of manifold conceptions in individual cognitive structure. International Journal of Science Education, 22, 399-417. Tytler, R. (1998). The nature of students’ informal science conceptions. International Journal of Science Education, 20, 901-927. Viennot, L. (2001). Reasoning in Physics – The part of common sense. Dordrecht: Kluwer.

S. HARTMANN, H. NIEDDERER

481

Westra, A. (2003). A new approach to teaching/learning mechanics In Dusan Krnel (Ed.). Proceedings of the Sixth ESERA (European Science Education Research Association) Summerschool. CD, University of Ljubljana, Faculty of Education, 258-262. Wittmann, M. C. (2002). The object coordination class applied to wave pulses – analysing student reasoning in wave physics. International Journal of Science Education, 24, 97118. Wodzinski, R. (1996). Untersuchungen von Lernprozessen beim Lernen Newtonscher Dynamik im Anfangsunterricht. Münster: Lit

A CROSS-SECTIONAL STUDY OF THE UNDERSTANDING OF THE RELATIONSHIPS BETWEEN CONCENTRATION AND REACTION RATE AMONG TURKISH SECONDARY AND UNDERGRADUATE STUDENTS

GULTEKIN CAKMAKCI, JIM DONNELLY, JOHN LEACH The University of Leeds, UK

ABSTRACT This research describes a cross-sectional study, which will give insights into the development of students’ understanding of chemical kinetics (at key points, in relation to relevant teaching) from secondary to university level, in Turkey. The study is based mainly on the written responses given by school and undergraduate students to a series of written tasks involving concepts and phenomena in chemical kinetics. A small-scale interview study was also carried out with a number of students to obtain further information regarding students’ ideas about chemical kinetics and to check for appropriate interpretation of the written responses. In this paper, our focus is on the students’ understanding of the relationships between the concentrations of reactants/products and reaction rate. Analysis of students’ responses on written probes and in interviews indicates that, after instruction, many students use conceptions not consistent with scientific perspectives, and have conceptual difficulties in understanding the relationships between concentration and reaction rate. Furthermore, the results show that students did not frequently use “particulate” and “mathematical” modelling, and in most cases such modelling was not used as intended by the curriculum. The results indicate a need to review curricula, and instructional practices, in the light of the students’ difficulties in understanding chemical kinetics.

1. INTRODUCTION Chemical reaction rates and the factors that affect them constitute an important area of the chemistry curriculum (Cachapuz & Maskill, 1987; Ragsdale et al., 1998). Given the importance of chemical kinetics and the diverse nature of the concepts and relationships that comprise it, it is surprising that “the development of students’ understandings of chemical kinetics in relation to teaching” has not been the focus of educational research over the years (Justi & Gilbert, 1999; Cakmakci, 2004). There are very little data available on how understanding of chemical kinetics progresses as students move through the curriculum. It is intended that this study will provide empirical evidence about students’ understanding of chemical kinetics. Since students experience difficulties in understanding in chemical kinetics both at 483 K. Boersma et al. (eds.), Research and the Quality of Science Education, 483—497. © 2005 Springer. Printed in the Netherlands.

484

THE RELATIONSHIPS BETWEEN CONCENTRATION AND REACTION RATE

school (De Vos & Verdonk, 1986; Justi, 2002) and university level (Lynch, 1997), further research is required in order to give insights into the ways in which students conceptualise chemical kinetics at school and university level. Longitudinal and cross-sectional studies have been undertaken to identify how understanding of specific ideas progresses as students move through the curriculum. In order to evaluate the success of a curriculum at achieving its objectives, it is necessary to look at some aspects of students’ learning. Several longitudinal and cross-sectional studies have been undertaken on students’ domain-specific reasoning in other areas of science, and the data from these studies have provided important information when decisions are made in planning and sequencing the curriculum (Abraham et al., 1994; Driver et al., 1994). This study is aimed at identifying and characterising the development of students’ understanding of the relationships between the concentrations of reactants/products and reaction rate, at key points, in relation to relevant teaching at secondary and university level, in the Turkish educational system. It is intended that findings of the study can be used to inform teaching interventions by highlighting possible mismatches between the objectives of the curriculum and students’ level of understanding in chemical kinetics at school and university level. 2. METHODS This study is part of a continuing research project (Cakmakci, 2003a). It is based mainly on the written responses given by 191 school and undergraduate students to 11 open-ended diagnostic questions (termed ‘probes’) involving concepts and phenomena in chemical kinetics. The sample includes 108 secondary school students (Grade 10, ages 15-16) in three classes from two different schools, 48 firstyear (age 17+) and 35 third-year (age 19+) pre-service chemistry teachers. Data were collected 5-6 weeks after secondary school students had been taught chemical kinetics; data were collected immediately after teaching for university students. The same diagnostic tests were conducted on secondary school and undergraduate students. By using the same probes, we would be able to compare the development of student reasoning as a result of instruction at school and university. A sub-sample of the students (10 SS, 7 UF and 7 UT)1 was also interviewed in order to obtain further information regarding their ideas about chemical kinetics and to check for appropriate interpretation of the written responses. This sub-sample was chosen to represent diversity in responses to the written probes. The probes used in interviews were the same as those used in the diagnostic tests. The interview, therefore, could improve the credibility and validity of the data and findings. In order to identify the intended development of the subject of chemical kinetics within the school and university courses, the science curriculum, chemistry textbooks, and students’ notes were analysed, based on a conceptual analysis of the domain. 1

‘SS’ refers to secondary school students; ‘UF’ and ‘UT’ to university first year and university third year students, respectively

G. CAKMAKCI, J. DONNELLY, J. LEACH

485

Students’ ideas about the relationships between the concentrations or pressures of reactants/products and reaction rate were elicited throughout their responses on four probes. This paper focuses on students’ responses on two of these four probes (see the Appendix). The probes targeted both students’ scientific knowledge about the topic and how they use this knowledge both in the context of the school science and in everyday experience. Since the study aims to allow subjects to respond to the probes in their own ways, using their terminology that they thought appropriate, both conceptually and phenomenologically framed probes (Driver & Erickson, 1983) were designed, in an open-ended format. The intention was that, as a result of this, the subsequent analysis would be ideographic, exploring the content of the students’ reasoning in their own terms and language (Driver & Easley, 1978). The aim of conceptually framed probes was to find out how well particular ideas that have been taught have been understood by students. The reason for using phenomenologically framed probes was to explore which knowledge students use when given minimal contextual support within the probes and to find out how appropriately they use their knowledge in the context of everyday experience. These probes were mainly designed to assess students’ ability to apply science concepts in a given situation (using Driver and Erickson’s [1983] term, students’ theories-inaction). All school students who participated in this study majored in mathematics and science. In Turkey, formal chemistry courses, which take three years, start with secondary education (Grades 9-11, ages 14-17). School students follow a common science curriculum which is developed and approved by the Ministry of National Education. However, there is no national or centralised curriculum for universities. The concept of chemical kinetics is first taught to students in Grade 10 (ages 15-16). Students receive both classroom and laboratory instruction on chemical kinetics. However, it has been claimed that teachers in Turkish secondary schools usually prefer to teach chemistry with non-laboratory based instruction (Nakiboglu, 2003). Chemical kinetics is also taught in the first year in general chemistry course and in the third year in physical chemistry course in a five-year pre-service chemistry teacher-training course. Furthermore, the first year pre-service chemistry teachers do experiments on chemical kinetics in the general chemistry laboratory course. 3. RESULTS Conceptual Analysis of Chemical Kinetics A distinction has been made between the world of objects/events and the world of laws/theories/models (Logan, 1984; Tiberghien, 2000). Logan (1984) states that chemical kinetics has an unusually complex structure in that it is composed of two distinct but complementary lines of development: the “empirical” and “the “theoretical”. The relationship between chemical phenomena and theories/models is shown in Figure 1. Conceptual analysis of the domain suggested that the rate of chemical reactions can be explained by a qualitative approach (Particulate

486

THE RELATIONSHIPS BETWEEN CONCENTRATION AND REACTION RATE

Modelling), and it may also be quantified in terms of a quantitative approach (Mathematical Modelling). It might be concluded that chemical kinetics is concerned with the rates of chemical reactions and the factors affecting rates of chemical reactions that are explained by particulate modelling and which may be represented/quantified in terms of mathematical modelling. At university level, the main emphasis of chemical kinetics courses seems to focus on the quantitative aspect of the subject. The rate of chemical reactions (Chemical kinetics)

The world of objects/events

is explained by

may be represented /

The world of laws/theories/models Particulate Modelling

has relationship

Mathematical Modelling

Figure 1. The relationship between chemical phenomena and theories/models Analysis of the De-scaler probe The De-scaler probe is set in an everyday context and asks how changes of concentrations of reactants affect reaction rate (see the Appendix). A coding scheme was developed by reviewing students’ responses in interviews and to written probes, and by identifying common ideas and ways of explanation. The coding schemes were developed from students’ responses rather than being based, for example, on the normative science. Five main categories of responses were identified and used in the reporting of results. The first category, termed “macroscopic modelling”, tends to be descriptive in nature (knowledge of what happens, interpreting the phenomenon in terms of what might be perceived (i.e. at a macroscopic level) but not referring to unseen entities and processes – such as molecules, collision/interaction of particles/molecules/ions, or mechanisms to account for the phenomenon. For example, the answer of the respondent who said “concentrated acid cleans more quickly limestone in the kettle, because it is stronger” was placed in this category. The category termed “particulate modelling” incorporates those responses in which students use corpuscular models such as collision/interaction of particles/ions/atoms/molecules, or use the principles of the collision model in their

G. CAKMAKCI, J. DONNELLY, J. LEACH

487

reasoning. The emphasis is on the microscopic processes taking place during the reaction and the relationship of such processes to the macroscopic behaviour of the reaction. For example, when a respondent says that “concentrated acid [solution] has got more acid molecules; therefore more acid molecules will react with limestone molecules in a period of time”, the respondent’s explanation goes beyond the phenomenon by drawing on theoretical entities and processes that are not observable in the phenomenon itself. The category termed “mathematical modelling” is allocated to cases where justification involves algebra, diagrams, or mathematical formulae. For example, as one of the students states that “Reaction rate = k.[A].[B], therefore the more concentrated reactants are, the faster a reaction happens”, she justifies her answer based on the rate equation. Whilst macroscopic modelling makes few claims about the nature of systems and has little predictive and explanatory power, particulate and mathematical modelling allow more to be explained, quantified, and predicted. In some cases, students used particulate and mathematical modelling in tandem. These students invoked macroscopic-particulatemathematical relationships (appropriately or inappropriately) when accounting for the phenomenon. These responses are placed in “P&M Modelling” category. The category “Uncodeable” is allocated to incomprehensible responses or cases where there is no response given in any part of the probe. Results of the De-scaler probe Drawing upon the analysis of the ‘De-scaler’ probe, it is found that there is a statistically significant difference between school and undergraduate students’ reasoning (X2, p<0.0001) in that the reasoning based on macroscopic properties gradually decreases from school to university level (see Figure 2). Despite being directed to consider the phenomenon “in terms of particles”, school students tend to use explanations based upon macroscopic properties: only a small number of the students use some form of theoretical model or causal mechanism to account for the phenomenon. Terms like “dissolving/solubility” and “react/reaction” were not differentiated by many school students, and in some cases these terms were used interchangeably. It seems that regarding terminology where everyday conceptions are interlaced with the scientific concepts, expressions such as “concentrated acid cleans / removes / dissolves / eats / destroys / burns / irritates more quickly limestone / limestone particles in the kettle” are quite common amongst school students, but are practically absent by the end of secondary education. In some instances, the objects of the macroscopic world (e.g. strong acid reacts more quickly with limestone) are treated as if they were particles, atoms, or molecules (the objects of the sub-microscopic world). It was noticeable that university students were more likely to use reasoning based on sub-microscopic or mathematical levels. However, around half (45.8%) of the first-year undergraduates did not refer to any ideas concerning the dynamic nature of particles. Undergraduates’ (particularly university third-year students) responses were richer in the terminology and the range of justifications provided, in that they used the principles of collision or transition-state theory/model more

488

THE RELATIONSHIPS BETWEEN CONCENTRATION AND REACTION RATE

appropriately, frequently, and confidently in their reasoning. Particulate modelling becomes increasingly popular as undergraduates move through the curriculum. A few students, mainly undergraduates, relied on mathematical formulae (e.g. the rate equation) and a functional vocabulary (e.g. reaction rate is directly proportional to the concentrations of reactants, thus increasing the concentration of reactants increases the reaction rate) to describe the phenomenon. 6.3% of the first year and 8.3% of the third year undergraduates explained the macroscopic phenomenon using both particulate and mathematical modelling (P&M Modelling). Many of those students were able to move between different levels of modelling and integrated one to the other to express their understanding of the chemical phenomenon.

80 70 60 % Percentage

50 40 30 20 10 0

Grade 10

Uni 1

Macroscopic Modelling

64.8

45.8

5.7

Particulate Modelling

26.9

31.3

68.6

Mathematical Modelling

2.8

12.5

17.1

0

6.3

8.6

P&M Modelling

Uncodeable Total %

5.5 100

Figure 2.

4.1 100

Uni 3

0.0 100

Responses to the ‘De-scaler’ probe

Analysis of the Reaction rate-Time probe The purpose of the probe is to find out to what extent students know and understand the relationships between concentration of reactants and reaction rate. The probe presented students with data; they have to assess the data and find out how the reaction rate changes with time (see the Appendix). Written responses were categorised by comparing the similarities and differences in the answers of all respondents. Students’ responses were examined with a commitment to reflecting

G. CAKMAKCI, J. DONNELLY, J. LEACH

489

each student’s conceptions, rather than evaluating a particular response in terms of a set of normative concepts. Seven main categories of responses were identified and used in the reporting of results. Six mutually exclusive models were identified from students’ responses to the “Reaction rate-Time” probe. They are mutually exclusive in the sense that no written response has been coded as more than one model. The models represent different ways of thinking about the relationships between reaction rate and time / concentration. The models and the data are illustrated in Figure 3. The models, described below, are elucidated by quotes from the written responses and the transcripts.2 60 50 40 30 20 10 0

Grade 10

Uni 1

Uni 3

ICM IM

7 2 19 8 23

0 10 6 25 13

0 9 6 51 3

SM

33

44

26

DM ICDM CM

Uncodeable Total %

8 100

2 100

5 100

Figure 3. The models used in the ‘Reaction rate-Time’ probe

Results of the Reaction rate-Time probe The Scientific Model (SM): The rate of reaction is described as being dynamic in nature. The relationship between the concentrations of reactants/products and reaction rate is described in terms of collision model or in terms of mathematical 2

In the text, the quotations taken from students’ written responses and the transcripts were identified with the codes such as [Sa-D-13] or [UF-I-15]. In this code, first two letters ‘Sa’, ‘Sb’ and ‘Sc’ strand for secondary school students and ‘UF’ and ‘UT’ strand for university first year and university third year students, respectively. The third letter ‘D’ and ‘I’ indicate the Diagnostic tests responses and Interview responses, respectively, and the number shows the student’s number. The comments between brackets […] in the excerpts aim to make the excerpts easier to understand. They are not the words of students.

490

THE RELATIONSHIPS BETWEEN CONCENTRATION AND REACTION RATE

modelling (see quotation below). This category includes responses such as: “the reaction rate (Rxn) would decrease, because the higher the concentration of molecules, the greater the number of collisions in unit time and hence the faster the reaction. As reactants are consumed, their concentrations drop, collisions occur less frequently, and the reaction rate decreases” or “reaction rate is directly proportional to the concentrations of reactants, therefore the reaction rate would decrease”.

Reaction Rate

The reaction rate depends on the concentrations of reactants, and as a result, the graph [referring to the concentration of reactant-time graph] shows that the gradient at any point along with the line will decrease. As the gradient is equal to reaction rate, as rate=d[C]/dt, this shows that reaction rate decreases as the reaction progresses. [UF-I-05] Time

It is surprising that of the 35 third year university students (UT), only 9 (26%) used the scientific model, and that the first year university students (UF) had the highest percentage of using the scientific model. The Increasing Model (IM): According to this model the reaction is conceived to start slowly and occur faster thereafter. The students may use macroscopic or particulate or mathematical modelling for their justifications.

Rxn

At the beginning of the reaction, reactant molecules are far away from each other; therefore the reaction rate is zero at the beginning. During time interaction of molecules increases and as a result the reaction rate increases. [Sc-I-04] Time

The concentration of A is decreasing, which means the rate of the reaction would increase. [Sc-D-12]

The second excerpt shows that the subject may have confused the rate of reaction and amount of product. This model is particularly common among school students (SS). 23% of the SS, 13% of the UF, and 3% of the UT students used the Increasing Model. The Increasing-Constant Model (ICM): The reaction rate is zero at the beginning, then gradually increases up to a maximum value and remains constant at this value. This model has two sub-categories. One, ICMa, includes students who confuse reaction rate with amount of product.

G. CAKMAKCI, J. DONNELLY, J. LEACH

491

A [reactant] is decreasing, that means product is formed, As A and B [reactants] are used up, the formation of C [product] increases and accordingly the reaction rate increases until all A and B are consumed where the reaction rate is constant. [Sa-D-14]

This subject seems to confuse the reaction rate and amount of product, as she says reaction rate is constant at the end of the reaction. She might have thought that if the reaction rate is constant, no product is formed. The other sub-category, ICMb, involves responses in which students’ verbal justifications and reasoning are appropriate from a chemist’s perspective, but the knowledge is transferred into a symbolic/graphical form that is different from a chemist’s perspective (i.e. see the excerpt given below). This model (ICMb) is common amongst undergraduates.

Rxn

The reaction rate decreases because consumption of [A] is greater at the beginning thereafter it reduces. Since reaction rate is proportional to the concentrations of products, we can say that the reaction rate decreases during reaction time. [UT-D-16] Time

As quoted above and in many other cases, although the subject’s justifications and reasoning were correct, representation of that idea on a graph was different from a formal chemist’s perspective. The reason might be that the student has conceptual difficulties in understanding that reaction rate is maximum at the beginning and is zero at the end of the reaction or the student simply confuses it with the concentration of product vs. time graph. 8% of the SS, 25% of the UF, and perhaps surprisingly, around half of the UT students used this model (ICM). The Constant Model (CM): This category refers to respondents who state that the reaction rate is constant as the reaction progresses. This model is common among school students, many of whom used “if… then” reasoning. “If the temperature or concentration is changed, then reaction rate will change, otherwise reaction rate is constant during a reaction”. Many of the students assume that as long

492

THE RELATIONSHIPS BETWEEN CONCENTRATION AND REACTION RATE

as certain factors (e.g. temperature, concentration, catalysts) are not altered, reaction rate remains constant during a reaction.

Rxn

[A] and [B] decreases and [C] increases with time. The rate of the reaction would not change, because temperature has not been changed. [Sb-D-02]

Time

The reaction rate is constant as time passes, because reaction rate does not depend on time. Time does not change reaction rate. [Sb-D-13]

The Increasing-Constant-Decreasing Model (ICDM): The reaction is seen to start slowly and its rate increases up to a maximum value; at this level it decreases gradually to zero when the limiting reactant is consumed.

Rxn

The rate of the reaction increases at the beginning of the reaction. When reactants are used up the reaction rate drops and at the end of the reaction, the rate is zero. [UF-D-36]

Time

The Decreasing Model (DM): This model is mainly based on the idea that “a slow reaction takes a long time and a fast reaction takes a short time” or “reaction rate is inversely proportional to time”. The subjects have a general picture of how different reactions tend to occur, however they cannot/did not explain how reaction rate changes as a reaction progresses. While the relationship between reaction rate and time was graphically illustrated in a scientific way, the students’ interpretations and rationale were different from the established chemical perspective. In the transcript below, R and S stand for the researcher and the student respectively.

Rxn

G. CAKMAKCI, J. DONNELLY, J. LEACH

Time

493

S- Reaction rate is inversely proportional to time [mentioning on the graph]. A slow reaction takes a long time and that a fast reaction takes a short time. R- So, for this reaction what would you say about the reaction rate? How would it change during reaction time? S- If this reaction is fast, it takes less time, and if it is slow it takes more time to occur. R- How would the reaction rate change during this reaction?

The discussion did not go beyond this point, the subject could not explain how the reaction rate changed during the reaction, rather he was talking about the situation for different reactions. Uncodeable (U): This category is used for responses not represented by any of the above categories, or in cases where there is no response given in any part of the probe.

An overview of students’ responses to the Reaction rate-Time probe Six models were deduced from students’ responses to the ‘Reaction rate-Time’ probe. However, it has been found that context has a significant influence on students’ usage of these models (for details see Cakmakci, 2003a); therefore the results should be regarded as tentative. The results suggest that many students at both secondary and university levels use conceptions not consistent with scientific perspectives and have conceptual difficulties in explaining how reaction rate changes as a reaction progresses. The ideas that “reaction rate increases as the reaction progresses” (IM), “reaction rate is constant” (CM), or “reaction rate increases up to a maximum value, and remains constant at this value” (ICM) were quite common among both school and undergraduate students. The students who gave the explanation that the rate of reaction was constant (CM) did not anticipate that the reaction had a different rate at different stages of the reaction and that the reaction rate was dynamic, but rather believed that the reaction rate was static and had a constant quantity. Many students had difficulties in understanding that the reaction had the highest rate at the beginning of the reaction and the lowest rate at the end: rather, they thought the opposite. The reaction was thought to start slowly. Only 33% of secondary school, 44% of the university first year and 26% of university third year students used the ‘scientific model’ (SM). It is worth underlining that whilst many undergraduates gave appropriate interpretations and rationales concerning how reaction rate changes while the reaction progresses (particulate and/or mathematical modelling), they had difficulties in representing that knowledge symbolically, e.g. by representing it on a graph- (mathematical modelling). Research in other areas of science also shows that students have difficulties in making transformations within and across different representational

494

THE RELATIONSHIPS BETWEEN CONCENTRATION AND REACTION RATE

forms, e.g. transforming a chemical equation into a corresponding graph (Kozma, 2003). 4. CONCLUSIONS AND IMPLICATIONS In this paper we have discussed students’ ideas about chemical kinetics after they have been taught the topic. The purpose of the study was to explore the development of students’ understanding of chemical kinetics in relation to relevant teaching at school and university level. It should be noted that the study is cross-sectional, which allows for investigation of students’ reasoning in response to particular tasks at different educational levels. The study does not describe pathways in the development of individual students’ reasoning. Doing this would require longitudinal studies. The results of the De-scaler and Reaction rate-Time probes suggest that the students following the curriculum made minimal progress from secondary through university level. Many students had difficulties in providing appropriate explanation for the phenomenon/event. (By “appropriate” we mean involving reference to particulate and/or mathematical modelling, both of which are introduced to these students in the curriculum.) Many of them, particularly secondary school students, drew upon macroscopic modelling when explaining the phenomenon. Whilst macroscopic modelling says little about the nature of reaction rates and is essentially descriptive, particulate and mathematical modelling allow more to be explained, quantified, and predicted. However, particulate and mathematical modelling were not used frequently by the students, and in most cases not effectively and appropriately in accordance with the approach used in the curriculum. The students had conceptual difficulties in understanding how reaction rate changes during a reaction. One of the major difficulties the students experienced was that while they provided appropriate explanation for the relationship between reaction rate and time in written or oral form, they failed to construct a symbolic representation for this relationship. Since the students could not appropriately transform ideas within and across different types of modelling/representations, the curriculum should guide students to use multiple, linked representations in the context of collaborative activities (Cakmakci, 2003b; Kozma, 2003; Wu, 2003). Our claim is that all these three levels of modelling and their relationships with each other are necessary for achieving a full scientific understanding of chemical kinetics. Supporting and mediating students’ conceptualisations of the interrelationships between these three levels of modelling have potential for improving their conceptual understandings of chemical kinetics. As social and discursive interaction constitutes the process of meaning making (Mortimer & Scott, 2003), the teacher’s role should be supporting and mediating students’ conceptualisation of the relationships between macroscopic, particulate, and mathematical modelling. As Johnstone (1982) expressed it, “trained chemists jump freely from level to level in a series of mental gymnastics” (p. 377). The ability to pass confidently between these models should be an important goal for pre-service chemistry teachers to ensure that

G. CAKMAKCI, J. DONNELLY, J. LEACH

495

they will not pass conceptual difficulties and scientifically incorrect ideas on to their students. An analysis of the scientific knowledge presented in the curriculum indicates that although the nature of these three models is generally presented clearly, the links between them are not explicitly specified. In most cases, theoretical models are disconnected from the physical/chemical phenomena and practices. The results suggest that it would be helpful if the relationship between these three levels of modelling was clearly expressed in the curriculum, and students were given explicit teaching and support in moving among them. ACKNOWLEDGEMENT This project is funded by the Turkish Ministry of National Education. REFERENCES Abraham, M.R., Williamson, V.M. & Westbrook, S.L. (1994). A cross-age study of the understanding of five chemistry concepts. Journal of Research in Science Teaching, 31(2), 147-165. Cachapuz, A.F.C. & Maskill, R. (1987). Detecting changes with learning in the organization of knowledge: use of word association tests to follow the learning of collision theory. International Journal of Science Education, 9(4), 491-504. Cakmakci, G. (2003a). A cross-sectional study of the understanding of chemical kinetics among Turkish secondary and undergraduate students. Unpublished PhD proposal, The University of Leeds, Leeds, UK. Cakmakci, G. (2003b). Upper secondary school and undergraduate students' understandings of the concept of catalysis. Paper presented at the Variety in Chemistry Education, Dublin, Ireland, 31st August-2nd September 2003. Cakmakci, G. (2004). Science Education In Turkey: A Bibliography on Teaching and Learning Science. Retrieved April 2004, from the World Wide Web: http://www.geocities.com/ScienceEducationinTurkey/ De Vos, W. & Verdonk, A.H. (1986). A new road to reaction: Part 3. Teaching the heat effect of reactions. Journal of Chemical Education, 63(11), 972-974. Driver, R. & Easley, J. (1978). Pupils and paradigm: A review of literature related to concept development in adolescent science students. Studies in Science Education, 5, 61-84. Driver, R. & Erickson, G. (1983). Theories-in-action: Some theoretical and empirical issues in the study of students' conceptual frameworks in science. Studies in Science Education, 10(2), 37-60. Driver, R., Leach, J., Scott, P. & Wood-Robinson, C. (1994). Young people's understanding of science concepts: implications of cross-age studies for curriculum planning. Studies in Science Education, 24, 75-100. Johnstone, A.H. (1982). Macro-and microchemistry. School Science Review, 64, 377-379. Justi, R. (2002). Teaching and learning chemical kinetics. In J. K. Gilbert, O. De Jong, R. Justi, D. Treagust & J. H. Van Driel (Eds.), Chemical Education: Towards Researchbased Practice (pp. 293-315). Dordrecht: Kluwer Academic Publishers. Justi, R. & Gilbert, J. (1999). A Cause of A historical Science Teaching: Use of Hybrid Models. Science Education, 83(2), 163-178.

496

THE RELATIONSHIPS BETWEEN CONCENTRATION AND REACTION RATE

Kozma, R. (2003). The material features of multiple representations and their cognitive and social affordances for science understanding. Learning and Instruction, 13, 205-226. Logan, S.R. (1984). Introductory reaction kinetics – an unacknowledged difficulty. Education in Chemistry, 21, 20-22. Lynch, M.D. (1997). The effect of cognitive style, method of instruction, and visual ability on learning chemical kinetics. Unpublished PhD thesis, Iowa State University, Iowa. Mortimer, E.F. & Scott, P.H. (2003). Meaning making in secondary science classrooms. Maidenhead: Open University Press. Nakiboglu, C. (2003). Instructional misconceptions of Turkish prospective chemistry teachers about atomic orbitals and hybridization. Chemistry Education: Research and Practice, 4(2), 171-188. Ragsdale, R.O., Vanderhooft, J.C. & Zipp, A.P. (1998). Small-scale kinetic study of catalyzed decomposition of hydrogen peroxide. Journal of Chemical Education, 75(2), 215-217. Tiberghien, A. (2000). Designing teaching situation in the secondary school. In R. Millar, J. Leach & J. Osborne (Eds.), Improving Science Education: the contribution of research (pp. 27-47). Buckingham: Open University Press. Wu, H.K. (2003). Linking the microscopic view of chemistry to real-life experience: Intertextuality in a high-school science classroom. Science Education, 87, 868-891.

APPENDIX “De-scaler” Probe: Serap’s mother usually used Tudor brand kettle de-scaler, which contains a 3% solution of acid. However this time her husband went shopping, and he bought a different kettle de-scaler called Apex, containing a 5% solution of acid. When Serap’s mother used the new kettle de-scaler for removing limestone collected in the kettle she realised that the new de-scaler Apex removed the limestone faster than Tudor did. Why did it take less time to remove limestone in the kettle with concentrated kettle de-scaler (Apex)? Explain your answer as fully as you can in terms of particles.

G. CAKMAKCI, J. DONNELLY, J. LEACH

497

NAME INDEX Abd-El-Khalick, F. 260, 261, 267, 275, 291, 369 Abraham, M.R. 21, 172, 484 Adams, J.E 170 Adey, P. 130, 134 Agresti, A. 239 Aikenhead, G.S. 108, 113, 182, 420 Albe, V. 182 Alexopoulou, E. 382 Alfafara, R. 469, 470 Al-Hasan Ibn Al-Haitham 260 Alpe, Y. 181 Alverman, D.E. 370 Ametller, J. 209 Andersen, C. 180, 430, 480 Anderson, B.R. 350 Anderson, C. 259 Andersson, B. 107, 200, 224, 227, 337 Angell, C. 31, 32 Archimedes 260 Arnold, M. 198, 200, 297, 307 Artigue, M. 199, 200, 203 Asoko, H.M. 200 Atkin, J.M. 57, 61 Atkinson, P. 108 Aufschnaiter, S. 200 Austin, J.L. 386 Aznar Cuadrado, V. 422 Bach, F. 200, 223, 224, 227 Bachelard, G. 371 Bailey, S. 259, 260 Baird, J. 222 Baker, W.P. 297 Bakewell, K. 270 Bakhtin, M.M. 395, 396, 397, 405 Bano, Y. 435 Baquedano-Lopez, P. 114 Barufaldi, J.-P. 199 Baxter, G.P. 322 Beach, D. 108 Beaton, A.E. 34, 36, 48 Beers, J. 13

Beeth, M. 91 Beijaard, D. 119, 126 Bell, B. 130, 197 Bell, R. 247, 260, 261, 275 Bennett, S.N. 168 Bent, H. 62 Ben-Zvi, R. 337 Berg, E. 460, 469 Bergquist, W. 459, 460 Bergqvist, K. 108 Berliner, D.C. Beynon, J. 232 Bianchi, L.J. 135 Billig, M. 395 Bisanz, G.L. 281 Bisanz, J. 281 Bishop, K. 131 Black, P.J. 57, 61, 62, 63, 224 Blair, L.M. 256 Bliss, J. 449, 455 Blumenfeld, P. 312 Bodzin, A.M. 150 Boekaerts, M. 19 Boersma, K.Th. 42 Bol, L. 231 Boohan, R. 200 Bos, K. Tj. 42, 45, 47, 51 Boulter, C.J. 198, 325, 332 Bourdieu, P. 8,10 Bransford, J.D. 218, 221 Braund, M. 140 Brewer, W.F. 462 Brickhouse, N.W. 291 Bridges, D. 78 Britton, B. 460 Britton, E. 38, 53 Broekkamp, H. 26 Brown, A.L. 221 Brown, C. 115 Brown, D.E. 209, 297 Brown, J.S. 108, 114, 419 Brown, S. 284, 291, 292 Bryant, N. 10 Bucat, R. 459, 469

499

500 Buck, P. 338 Buckley, P.D. 463 Budzinsky, F. 260 Buffler, A. 242 Bulte, A.M.W. 73, 75, 172 Bybee, R.W. 150, 231, 232 Bye, A. 276 Cachapuz, A.F.C. 483 Cakmakci, G. 483, 484, 493, 494 Caldwell, K. 275 Calley, J. 310 Camacho, M. 459 Campbell, B. 75, 107, 169, 232 Campbell, J.A. 58 Carambo, C. 13 Carlgren, I. 114 Carlo, G. 186, 191 Carmeli, M. 150, 153 Carnot, S. 299 Carre, C.G. 199 Carretero, M. 457, 480 Carter, G. 223 Cassileth, B.R. 280 Castello, M. 232 Chambers, F.W. 189 Chambers, F. 271 Champagne, A.B. 462 Chang, C.-Y. 199 Chauvet, F. 198 Chen, C. 416 Chen, K. 428 Chinn, C.A. 462 Chittleborough, G. 351 Chomat, A. 198, 201, 202 Clark, C. 119 Clement, J. 209, 297, 326, 449 Cobb, P. 15, 225 Cobern, W. 108, 113 Cochran, K. 119, 120, 361 Cocking, R.R. 221 Cohen, D. 391 Cohen, L. 333, 330, 327 Cole, S. 248 Coll, R.K. 146 Coleman D. 127 Collins, A. 114, 419 Collins, S. 378

INDEX Comber, C. 167 Confrey, J. 25, 229 Cook, P. 119 Corbin, J. 251 Cowie, H. 382 Craft, A. 129, 130, 137 Crawford, B.A. 309, 310, 312, 313 Crawford, T. 108 Crebbin, C. 130, 135 Creswell, J.W. 314 Cronbach, L.J. 251, 274, 276 Cullin, M.J. 309, 310, 312, 313 Curtis, R.V. 354 Dagher, Z.R. 291, 354 Dalman, T. 469, 470 Davidson, D. 449, 452, 453, 455, 456 Davies, D. 134 Davis, C. 10 De Boer, G. 115 De Jong, O. 120, 126, 310, 311, 355 De Pomeroi, D. 185, 186, 187, 190 De Vos, W. 67, 170, 172, 484 Dekkers, P. 200, 450, 451, 455 Delamont, S. 108 Demuth, R. 180 Denley, P. 131 DeRuiter, J. 361 Désautels, J. 181, 407 Dewey, I. 197, 236, 248, 249, 255, 381 DiDonato, L. 470 Dieudonné, Prof. 60, 61 Dijkstra, Jr. Din Yan Yip 122 diSessa, A. 451, 455, 456, 471 Dreyfus, H.L. 291 Dreyfus, S.E. 291 Driel, J.H. 119, 126, 169, 170, 223, 310, 311, 316, 355, 459, 461, 469 Driver, R.H. 108, 197, 210, 225, 270, 271, 272, 370, 381, 382, 407, 420, 475, 484, 485 Drucker, S. 428 Druger, E. 293 Druker, S. 416 Duguid, P. 114

INDEX Duit, R. 5, 123, 198, 200, 209, 337, 353, 354, 362, 395, 435, 447, 471 Dumas-Carré, A. 199 Dupin, J.J. 354 Duschl, R.A. 368 Easley, J. 485 Edison, T.A. 260 Eflin, J.T. 280 Eijkelhof, H.M.C. 75, 437 Elkis, I. 153 Elmesky, R. 13 Engeström, Y. 5, 13 Ensign, J. 114 Eraut, M. 134 Erduran, S. 367, 368, 370, 381, 384, 391, 392 Ergazaki, M. 409 Erickson, G. 197, 485 Eylon, B-S. 231, 232 Feldman, A. 141 Fensham, P. 459, 469 Ferguson-Hessler, M.G.M. 26 Finegold, M. 310, 311, 316 Fink, D. 129 Fischler, H. 337, 338 Fisher, D. 157 Fleiss, J.L. 239 Freeman, J. 179 Friedman, M. 160, 452 Frost, R. 249 Fullan, M. 129, 130 Furth, H. 224 Gabel, D.L. 353 Gagné, B. 188 Galhoun, E. 121, 125 Gallagher, S.A. 158 Gallard, A.J. 150 Galton, M.W. 167 Gao, L. 120 García-Milá, M. 230, 430 Garden, R.A. 38 Garnett, P.J. 146, 459, 463, 467, 469 Gauthier, R. 188 Gayford, C. 78

501

Gee, J. 382 Genseberger, R. 75 Gentner, D.R. 297, 298, 299, 301, 306, 354 Gerdes, J. 471 Gertzog, W.A. 462 Gess-Newstome, J. 119, 120, 247, 256 Giachardi, D. 107 Gibbs, A. 259 Gick, M.L. 297 Giddens, A. 393 Giere, R.N. 408 Gil, J. 379 Gilbert, J.K. 130, 198, 310, 311, 314, 325, 326, 327, 328, 329, 330, 382, 483 Gillett, G. 395 Gipe, J.P. 143 Girardet, H. 409, 412 Giuffre, M. 270 Glaser, B.G. 109 Glaser, E.M. 271 Glennan, S. 280 Glynn, S.M. 297, 354, 460 Gnoffo, G. 293 Goldberg, F. 200 Gonzales, E.J. 37, 38 Good, R. 459 Goodrum, D. 157 Gore, J.M. 143 Gorodetsky, M. 459 Gräber, P.J. 127, 355, 459 Graber, W. 364 Grace, M.M. 78, 181 Graesel, C. 180 Gregory, K.D. 38 Griffith, A. 337 Griffiths, M. 91, 104 Gropengieber, H. 127, 205 Grosslight, L. 311, 312, 313, 314, 317, 320, 321, 328 Guerra, M.R. 382 Gunstone, R. 16, 120, 448, 460, 462, 471 Guskey, T.R. 129, 130, 377 Gussarsky, E. 459 Gutiérrez, K.D. 114, 456

502 Hackling, M. 157, 459, 467, 469 Haeberlen, S. 298, 306 Hagman, M. 223, 224 Hall, G.E. 91, 131, 137 Halliday, M.A.K. 395, 437 Halloun, I.A. 448 Hanks, W.F. 7 Hardell, L. 186, 187, 191 Hargreaves, A. 91, 130, 159 Harland, J. 131, 132, 133, 134, 135, 136, 137, 139 Harlen, W. 63, 107, 157 Harmon, M. 31 Harré, R. 395 Harrison, A. 337 Harrison, A.G. 148, 310, 311, 326, 353, 354, 355 Harrison, J.A. 463 Hart, C. 437 Hartmann, S. 479 Hartree, D. 63 Häuβler, P. 480 Heikinnen, H. 459, 466 Hein, G.H. 4 Henderson, J. 281 Herrenkohl, L.R. 382 Herrmann, F. 299 Hestenes, D. 448 Hewison, A. 269 Hewson, M.G. 470 Hewson, P.W. 119, 121, 142, 462 Hiebert, J. 222 Hind, A. 183, 284, 285 Hoban, G. 91 Hodson, D. 310 Hoffman, J.L. 148 Hofstein, A. 142, 150, 153 Hofstein, J.L. Hogan, K. 270, 382, 429 Holly, P. 142 Holowchak, M. 416 Holroyd, C. 157 Holyoak, K.J. 297, 298, 299 Hooymayers, H.P. 25 Hopkins, D. 121, 125 Hord, S.M. 91 Horton, R. 378

INDEX Houang, R.T. 53 Hoyle, C. 38 Huberman, M. 129, 377 Hubermann, A.M. 314 Huckle, J. 79 Huddle, P.A. 459, 463, 469 Hynd, C.E. 370 Ilénkov, E. 6 Ilya, M. 379 Ingvarson, L. 91, 130 Jackson, S.L. 313 Jammer, M. 452, 453 Janesick, V.J. 355 Jarman, R. 354 Jarvis, T. 130, 134, 135, 159, 160, 163, 164 Jay, E. 311 Jenkins, E.W. 27, 32, 181, 182 Jewitt, C. 108 Jimenez-Alexandre, M.P. 407, 409 Johnson, M. 279 Johnson, P. 455 Johnstone, A.H. 494 Johsua, S. 354 Joling, E. 16 Jones, L. 119, 120 Jones, M.G. 223 Jordan, B. 8 Joslin, P. 354 Joyce, B. 91, 121, 125, 134, 137, 153, 161, 377 Jung, W. 474 Jungck, J. 310 Justi, R.S. 310, 311, 314, 326, 327, 328, 329, 330, 338, 355, 483, 484 Kaminski, W. 198 Kaper, W. 297 Kariotoglou, P. 119, 125, 126, 198, 199, 200 Kattmann, U. 122, 198 Kaunda, L. 242 Kellaghan, T. 42 Kelley, A. 221, 229

INDEX

503

Kelly, D.L. 37 Kelly, G. 115, 416 Kemmis, S. 141 Kemp, A. 127 Kent, D.T. 122 Keogh, B. 369 Kepler, J. 453 Kerby, H. 119 Kerr, D. 79 Kersten, J. 270 Khishfe, R. 293 Kind, P.M. 249 Kinder, K. 131, 132, 133, 134, 135, 136, 137, 139 King, R. 361 Kircher, E. 347 Kjaernsli, M. 30, 31, 32 Klaassen, C.W.J.M. 72, 223, 449, 455 Klaassen, K. 449, 455 Klopfer, L.E. 462 Knain, E. 108 Koballa, T. 120 Kolstø, S.D. 182, 248, 272 Komorek, M. 198 Korpan, C.A. 271, 274 Korthagen, F.A.J. 120, 143, 170 Kortland, J. 20, 181 Kortland, K. 437 Koumaras, P. 198, 199, 435 Kozma, R. 494 Kraijick, J.S. 76 Kress, G. 395 Kruger, C. 60, 61, 122, 158, 161, 223 Kuhn, D. 371, 382, 395, 420 Kuhn, T.S. 4, 12 Kuiper, W. 42, 45, 46, 47, 48, 51 Kurtz, K.J. 297, 298

Leach, J. 15, 181, 199, 210, 212, 218, 247, 248, 284, 285, 286, 291, 381, 403, 404 LeCompte, M. 326, 327 Lederman, J.S. 283, 291 Lederman, N.G. 119, 247, 248, 256, 260, 261, 275, 369 Legardez, A. 181 Lehrer, R. 25, 229 Lemberg, J. 127 Lemeignan, G. 197 Lemke, J.L. 67, 72, 108, 224, 382, 395, 414 Leont’ev, A.N. 5 Lessard, N. 188 Letts, W.J. 291 Levinson, R. 77, 78, 182 Lewin, K. 142 Lewis, I. 141 Lewis, J. 181 Lichtfeldt, M. 350 Lie, S. 27, 31, 32, 35 Lijnse, P.I. 15, 20, 68, 75, 122, 171, 198, 221, 222, 223, 224, 449 Lindauer, I. 354 Linn, M.C. 231, 232 Lloyd, J. 130 Logan, S.R. 485 López, A. 107, 112, 114 Lotman, Yu. M. 397 Loucks, S. 129, 131, 137 Loucks-Horsley, S. 142, 169 Loughran, J. 91 Love, N. 142 Lubben, F. 107 Lucas, K.B. 283 Lunetta, V.N. 314 Lynch, M.D. 484

Lai, H. 185, 186, 187, 190 Larkin, J.H. 471 Larochelle, M. 407 Latour, B. 248 Lave, J. 108, 114, 284 Lawson, A.E. 259, 297, 407, 408, 413, 461, 463 Lazonby, J. 76, 179

MacDermott, L. 63 Malin, A. 35 Maloney, D.P. 472 Mamlok, R. 150 Mamlok-Naaman, R. 153 Maniala, T.L. 351 Manion, L. 327 Marentic Pozarnik, B. 120 Mariani, M.C. 456

504 Marks, A. 140 Martin, I. 116, 406 Martin, J.R. 395 Martin, M.O. 34–36, 48 Marx, R. 169, 312 Maskill, R. 483 Mason, L. 382, 407, 409 Matthews, M. 104, 209, 219, 283, 292 Maxwell, J.C. 60, 285, 286, 287, 289, 290 Mayer, R.E. 326 Mazur, E. 460, 469 McCloskey, M. 448 McComas, W. 259, 292 McGillicuddy, K. 116, 406 McIntyre, D. 284, 291, 292 McKeon, F. 160 McKittrick, B. 445 McKnight, C. 45 McKnight, Y. 13 McMahon, K. 134 Mctaggart, R. 141 Means, M.L. 382 Mehan, H. 401 Meheuet, M. 127 Merriam, S.B. 251, 328 Mestad, I. 249 Meyer, D. 270 Miao, C.-H. 297, 298 Mikelskis-Seifert, S. 337, 338, 344 Miles, M.B. 314 Millar, R. 24, 25, 181, 198, 200, 271, 272, 291, 297, 307, 381 Minstrell, J. 199, 381 Mintrop, H. 120 Mirham, G. 328 Molander, B.O. 30, 31 Monereo, C. 232 Monk, M. 384 Morge, L. 204 Morine-Dercshimer, G. 122 Morrison, K. 327, 328 Mortimer, E.F. 200, 210, 211, 395, 396, 397, 402, 403, 404, 419, 420, 494 Mulhall, P. 435, 436, 444 Mullis, I.V.S. 28, 49

INDEX Munn, P. 141 Nagel, E. 454 Nakhleh, M.B. 463 Nakiboglu, C. 485 Nakos, S. 94, 104 Navon, O. 153 Naylor, S. 369 Nemet, F. 375 Nentwig, P. 169 Nersessian, N.J. 326 Neuman, Y. 379 Newman, S.E. 419 Newton, I. 260, 261, 263, 265 Newton, P. 260, 261, 264, 265, 370, 382, 420, 458 Niaz, M. 146 Nicholson, P. 285, 286, 287 Nicol, C. 53 Niedderer, H. 200, 472 Nikolopoulou, K. 199 Nolen, S.B. 378 Norman, D.A. 326 Norris, S.P. 271 Nortfield, J. 127 Northfield, J. 104, 121, 222 Novick, S. 197 Ntombela, G.M. 255 Nussbaum, J. 197 O’Hanlon, C. 141 O’Malley, M. 261 Obaya, O. 142 Ogborn, J. 57, 64, 108, 224, 395, 449, 455, 456 Olander, C. 223, 224 Oliver, J.S. 157 Orpwood, G. 27 Osborne, J. 15, 19, 157, 164, 181, 368, 370, 378, 381–384, 391, 392, 420 Osborne, R.J. 326 Ostermeier, C. 104 Ozkaya, A.R. 146 Paige, K. 91 Palacio, D. 127, 167 Pantaleo, G. 479 Papamichael, Y. 197, 199

INDEX Parchmann, I. 180 Paulsen, A.C. 247, 248 Pelgrum, W.J. 34, 36 Pell, A. 130, 134, 159, 160, 163, 164 Pereiro-Muñoz, C. 420, 422 Peterson, P. 119 Peters-Sips, M. 45, 49 Petri, J. 200, 427 Pettersen, S. 270, 271, 273, 276, 279, 280 Peuckert, J. 338 Pfundt, H. 209 Phipps, R. 140 Pillay, A.E. 459, 469 Pilot, A. 67, 170, 172 Pine, J. 322 Plomp, T. 34, 36 Pontecorvo, C. 416 Pordhan, H. 435 Posner, G.J. 460, 461 Preece, A. 186, 187, 190 Preissle, J. 326, 327 Prenzel, M. 104 Preston, K. 337 Prideaux, N. 188 Prigogine, I. 6 Prins, G.J. 75 Psillos, D. 15, 122, 125, 126, 196, 198, 200, 221, 229 Qian, G. 370 Ragsdale, R.O. 483 Rainson, S. 198, 209, 223, 435 Raizen, S. 45 Ralle, B. 153 Ramberg, B. 456 Ramsden, J. 76, 179 Ratcliffe, M. 78, 84, 181 Ravanis, K. 197, 199 Ravetz, J. 377 Reddy, M. 368 Redish, J. 63 Reif, F. 471 Reigeluth, C.M. 354 Reisch, G. 280 Reiser, B. 312 Rennie, L. 157

505

Repacholi, M. 185-187, 190, 191 Resnick, L. 409, 437 Richards, J.C. 143 Rivet, A. 70, 75 Robardet, G. 197 Roberts, D.A. 67, 285, 286, 288, 289 Robitaille, D.F. 43 Rodrigues, S. 130, 135 Rodriguez, A.B. 416 Rogers, F. 470 Rogoff, B. 378 Roschelle, J. 457 Roskell, C. 269 Roth, W.-M. 3, 5, 9, 10, 108, 114, 283, 395, 409, 427 Roychoudhury, A. 395 Rua, M.J. 223 Rudduck, J. 382 Rushworth, P. 229 Russell, T.L. 393 Rutherford, F.J. 58 Rutherford, M. 325 Ryder, J. 181, 248, 284, 285, 291 Saari, H. 406 Sadier, T.D. 271 Sadler, T.D. 181 Säljö, R. 108, 114 Salmon, M. 416 Sampson, W. 276 Samuel, K.V. 353 Schauble, L. 25, 229 Schecker, H. 471 Scherz, Z. 243 Schmidt, D. 198 Schmidt, W.H. 29, 43, 45 Schneider, R. 76 Schnotz, W. 480 Schon, D.A. 142 Schuurmans, L. 53 Schwab, J.J. 67, 381 Schwartz, A.T. 75 Schwartz, R. 261 Schwarz, C. 323 Schwedes, H. 198, 298, 306 Scott, P.H. 15, 17, 108, 199, 210, 211, 212, 218, 286, 381, 383, 395, 396, 402, 403, 404, 419, 420, 494

506 Seifert, S. 337 Semmelweis, I. 30, 39, 40 Séré, M.G. 291 Sewell, W.H. 5, 8 Shamos, M.H. 232 Shavelson, R.J. 322 She, H. 157 Shen, S. 38 Shipman, H.L. 291 Showers, B. 91, 134, 137, 153, 161, 377 Shulman, L.S. 291, 292, 325 Siegler, R.S. 472 Silberstein, J. 350 Simmonneaux, L. 415 Simmons, E.R. 314 Simon, S. 112, 157, 164, 367, 370, 378, 384, 391, 392 Simpson, R.D. 157 Singer, J. 285-287 Slattery, M. 127 Smit, J.J. 310, 311, 316 Smith, C. 322, 334 Smith, D.C. 127 Smith, J.A. 363 Smith, J.P. III 457 Smith, T.A. 37, 38 Solberg, J. 270, 271, 274 Solomon, J. 78, 84, 369 Soloway, E. 310, 312, 313 Somerset, A. 460 Songer, N.B. 243 Sormunen, K. 406 Spektor-Levy, O. 243 Spyrtou, A. 125, 126 Squires, A. 229 Stavrou, D. 205 Steel, P. 140 Stengers, I. 6 Stigler, J.W. 222 Stiles, K. 142 Stoddart, T. 120-122 Stofflett, R. 128 Stokke, K.H. 37 Stolk, M.J. 170 Stoll, L. 129 Strage, A. 231 Stratford, S. 310

INDEX Strauss, A.L. 109, 251 Strike, K.A. 461 Suchman, L.A. 8 Summers, M. 122, 158, 161, 223 Suppe, F. 382 Sutton, C. 395 Szybek, P. 108, 113 Tabachnick B.R. 127 Taber, K.S. 472 Taconis, R. 19 Takahashi, T. 297 Tantoco, C. 293 Tattersal, J. 186, 187, 190 Taylor, A. 38 Taylor, T. 146 Tejeda, C. 114 ten Voorde, H.H. 25 Terlouw, C. 175 Thagard, P. 354 Thiele, R.B. 354 Thijs, G.D. 469 Thomsen, P. 58 Tiberghien, A. 198, 209, 223, 485 Tippins, D.J. 150 Tobin, K. 9, 10, 12, 150 Toulmin, S. 371, 392, 420, 422 Treagust, D.F. 146, 148, 209, 310, 326, 337, 353, 354, 362, 395 Trnobranski, P. 270, 273 Trowbridge, L.W. 150 Tsatsarelis, C. 405 Tsoumpelis, L. 197 Turmo, A. 35 Turner, S. 77, 78, 182 Twyman, M. 328 Tytler, R. 91, 94, 472 Uexküll, J. von 5 Unger, C. 311 Urmson, J.O. 392 Välijärvi, J. 35 Valverde, G.A. 53 Van Berkel, B. 170 Van den Akker, J.J.H. 53 Van den Berg, H. 53, 54

INDEX Van Driel, J.H. 119, 126, 169, 170, 223, 310, 311, 316, 355, 459, 461, 469 van Genderen, D. 25 van Lierop, A. 25 Van Oers, B. 73 van Soest, W. 25 Van Zee, E.H. 142, 381 Vanderhooft, J.C. 496 Veal, W. 355 Venville, G. 353 Verdi, M. 470 Verdonk, A.H. 170, 484 Verloop, N. 119, 126, 223, 310, 311, 316 Viana, A.P.P. 330 Viennot, L. 63, 198, 209, 218, 223, 435, 476 Viiri, J. 403 Vollebregt, M.J. 26, 351 Von Aufschnaiter, C. 376 Vos, F.P. 42-48 Vosniadou, S. 457, 480 Voss, J.F. 382 Vygotsky, L.S. 395 Waddington, D. 76, 179 Walberg, H.J. 150 Waldrip, B. 91, 104 Wallin, A. 223, 224 Walsh, A. 165 Warnock, M. 78 Wathen, S.H. 416 Watkins, D.A. 120 Watson, G. 271 Watson, N. 129 Watson, R. 259, 260 Watts, D.M. 325 Watts, M. 165, 448, 471 Webb, C. 91 Webb, P. 158 Weil-Barais, A. 197, 199 Wells, G. 368 Welzel, M. 200 Wenger, E. 284

Wertsch, J.V. 397, 419 West, R. 61 Westbroek, H.B. 171 Westbrook, S.L. 495 Westbury, I. 42, 47 Westra, A. 454, 471 White, B. 323 White, M. 470 Whitehouse, M. 57 Whitelock, D. 456 Wickman, P.O. 104 Wijnstra, J.M. 42 Wilbers, J. 127, 205 Wildman, S. 269 Wiley, D.E. 53 Wiliam, D. 224 Williamson, V.M. 495 Wisnudel-Spitulnik, M. 310 Wittmann, M.C. 479 Wodzinski, R. 474 Wolcott, H.F. 108, 109 Wolfe, R.G. 38, 53 Wood-Robinson, C. 220, 229, 495 Wragg, E.C. 160 Wu, H.K. 494 Wubbels, T. 120 Wyndhamn, J. 114 Wynne, B. 63, 181 Yeany, R.H. 460 Zabulionis, A. 31 Zeichner, K.M. 143 Zeidler, D.L. 271 Zeitoun, H.H. 308 Zeitz, C. 416 Zietsman, A. 456 Ziman, J. 248 Zimmermann, A. 10 Zipp, A.P. 496 Zogza, V. 409 Zohar, A. 375 Zoller, U. 181 Zuber-Skerritt, O. 141 Zwarts, M. 45

507