Distributed Leadership: Different Perspectives

DISTRIBUTED LEADERSHIP

STUDIES IN EDUCATIONAL LEADERSHIP VOLUME 7 Series Editor Kenneth Leithwood, OISE, University of Toronto, Canada Editorial Board Christopher Day, University of Nottingham, United Kingdom Stephen Jacobson, Graduate School of Education, Buffalo, U.S.A. Bill Mulford, University of Tasmania, Hobart, Australia Peter Sleegers, University of Nijmegen, The Netherlands

SCOPE OF THE SERIES Leadership we know makes all the difference in success or failures of organizations. This series will bring together in a highly readable way the most recent insights in successful leadership. Emphasis will be placed on research focused on pre-collegiate educational organizations. Volumes should address issues related to leadership at all levels of the educational system and be written in a style accessible to scholars, educational practitioners and policy makers throughout the world. The volumes – monographs and edited volumes – should represent work from different parts in the world.

For further volumes: http://www.springer.com/series/6543

DISTRIBUTED LEADERSHIP Different Perspectives

Edited by

ALMA HARRIS Institute of Education, London

123

Editor Prof. Alma Harris Institute of Education University of London Bedford Way London WCIH OAL [email protected]

ISBN 978-1-4020-9736-2

e-ISBN 978-1-4020-9737-9

DOI 10.1007/978-1-4020-9737-9 Library of Congress Control Number: 2008942715 c Springer Science+Business Media B.V. 2009  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 on acid-free paper 9 8 7 6 5 4 3 2 1 springer.com

Acknowledgements

Books such as this are rarely possible without the intellectual endeavor and goodwill of colleagues in the field. Fortunately, I work in an international research community that is challenging but also supportive. I am particularly grateful to all of the contributors in this book for providing chapters of such quality and depth. I am also grateful to the series editor for encouraging me to publish this book and to the publishers for their professionalism and support. Special thanks go to Jim Spillane for his friendship and intellectual generosity. I have learned a great deal from him and continue to do so. In any career there are people who influence us and who provide opportunities that take us further than we could ever have anticipated. My own intellectual journey benefited from the early influence of two leading academics, David Hopkins and Christopher Day. I would like to thank them both. Alma Harris

v

Contents

Part I Setting the Scene Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Alma Harris

3

1 Distributed Leadership: What We Know . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Alma Harris Part II Empirical Perspectives 2 Investigating Connections Between Distributed Leadership and Instructional Change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Eric M. Camburn and S.W. Han 3 Taking a Distributed Perspective in Studying School Leadership and Management: The Challenge of Study Operations . . . . . . . . . . . . . . 47 James P. Spillane, Eric M. Camburn, James Pustejovsky, Amber Stitziel Pareja and Geoff Lewis 4 The Relationship Between Distributed Leadership and Teachers’ Academic Optimism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Blair Mascall, Kenneth Leithwood, Tiiu Strauss and Robin Sacks 5 Distributed Leadership in Schools: Does System Policy Make a Difference? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Philip Hallinger and Ronald H. Heck Part III Practical Perspective 6 Capacity Building Through Layered Leadership: Sustaining the Turnaround . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 Christopher Day vii

viii

Contents

7 The Relationship Between Distributed Leadership and Action Learning in Schools: A Case Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139 Stephen Dinham Part IV Conceptual Theoretical 8 The Role of Sensemaking and Trust in Developing Distributed Leadership . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 Karen Seashore Louis, David Mayrowetz, Mark Smiley and Joseph Murphy 9 Distributed Leadership: Democracy or Delivery? . . . . . . . . . . . . . . . . . . 181 Andy Hargreaves and Dean Fink Part V Future Perspectives 10 From Distributed to Hybrid Leadership Practice . . . . . . . . . . . . . . . . . . . 197 Peter Gronn 11 Fit for Purpose: An Educationally Relevant Account of Distributed Leadership . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 V.M.J. Robinson 12 Coda . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241 Alma Harris Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249

Contributors

Eric M. Camburn is an Assistant Professor in the department of Educational Leadership and Policy Analysis at the University of Wisconsin-Madison. Camburn’s current research centers around understanding efforts to improve instruction, including programmatic improvement efforts, the organizational factors that support such efforts, and the impact change efforts have on leadership practice and instruction. Recent publications include “Assessing the Validity of A Language Arts Instruction Log through Triangulation” coauthored with Carol Barnes and published in the Elementary School Journal, and “Distributed Leadership in Schools: The Case of Elementary Schools Adopting Comprehensive School Reform Models”, coauthored with Brian Rowan and James Taylor and published in/ /Educational Evaluation and Policy Analysis. Christopher Day is Professor of Education and Director of the Teacher and Leadership Research Centre (TLRC). Prior to this he worked as a teacher, lecturer and local education authority adviser. He is editor of “Teachers and Teaching: Theory and Practice” and co-editor of the Educational Action Research Journal. He has recently completed directing a four year national Government funded research project on Variations in Teachers’ Work, Lives and Effectiveness. He is currently directing a 12 country project on successful school principalship; a nine country European project on successful principalship in schools in challenging urban contexts; and national projects on school leadership and pupil outcomes; and effective classroom teaching. He is a founder member of the Primary Schools Learning Network. Steven Dinham is Research Director of the Teaching, Learning and Leadership program at the Australian Council for Educational Research, Melbourne, Australia. He is also Professorial Fellow at the University of Melbourne and Visiting Professorial Fellow at the University of Wollongong. He was formerly Professor of Education at both the University of New England and the University of Wollongong and Associate Professor at the University of Western Sydney. He has conducted a wide range of research projects in the areas of educational leadership and change, effective pedagogy/quality teaching, student achievement, postgraduate supervision,

ix

x

Contributors

professional teaching standards, teachers’ professional development, middle-level leaders in schools, and teacher satisfaction, motivation and health. Peter Gronn is Professor of Education, University of Cambridge. Previously, he was Professor in Public Service, Educational Leadership and Management at the University of Glasgow, prior to which he was Professor of Education (personal chair), Monash University. His research interests cover all aspects of leadership, including educational policy, distributed leadership, leadership formation and the development of leaders, leadership models and types, school improvement, and links between organisational culture, leadership and organisational learning. The major focus of his most recent research has been leadership coaching, and head teacher recruitment and retention, for which projects he has been principal investigator and has received funding from the Scottish government. Philip Hallinger holds a Chair in Leadership and Change at the Hong Kong Institute of Education. His research focuses on school leadership effects, leadership development and problem-based learning. He can be reached at [email protected]. His research publications can be accessed at philiphallinger.com Seong Won Han is a Ph.D. candidate in sociology, University of WisconsinMadison. Her research interests include social stratification, sociology of education, and quantitative methodology. Her areas of specialization are inequality in education, transitions from high school to postsecondary education, educational policy, school reform, and comparative and international education. Alma Harris is Pro-Director (Leadership) at the Institute of Education, London and Professor of Educational Leadership at the London Centre for Leadership in Learning. She has previously held academic posts at the University of Warwick, University of Nottingham and University of Bath. She is currently an Associate Director of the “Specialist Schools and Academies Trust” and she is the editor of “School Leadership and Management”. Her research work has focused upon organizational change and development. She is internationally known for her work on school improvement, focusing particularly on ways in which leadership can contribute to school development and change. Her writing has explored middle level leadership, teacher leadership and leadership in challenging circumstances. Her most recent work has focused on distributed leadership in schools. Ronald Heck holds the Dai Ho Chun Endowed Chair in Education at the U. of Hawaii-Manoa. His research focuses on the relationship between school leadership and school improvement. He has also written on research methodology. He can reached at rheck@hawaii Kenneth Leithwood is Professor of Educational Leadership and Policy at OISE/ University of Toronto. His research and writing concerns school leadership, educational policy and organizational change. He has published more than 80 referred

Contributors

xi

journal articles, and authored or edited more than three dozen books. For example, he is the senior editor of both the first and second International Handbooks on Educational Leadership and Administration (Kluwer Publishers, 1996, 2003). His most recent books include Distributed leadership: The state-of-the-science (2008) Leadership With Teachers’ Emotions In Mind (2008), Making Schools Smarter (3rd edition, 2006) and Teaching for Deep Understanding (2006). Professor Leithwood is the recent recipient of the University of Toronto’s Impact on Public Policy award and is a Fellow of the Royal Society of Canada. Blair Mascall is Associate Professor in the Department of Theory and Policy Studies at the Ontario Institute for Studies in Education, University of Toronto, Canada. His current research is divided between an empirical study to establish the outcomes of distributed leadership in schools and school districts in Canada, and a large-scale project to define the impact of leadership on student achievement in the United States. David Mayrowetz is an associate professor of educational policy studies at the University of Illinois, Chicago. His research agenda centres on how organizational and institutional forces shape the formation and implementation of educational policy. He is particularly interested in distributed leadership as a condition that could facilitate school reform and improvement. Articles authored and co-authored by Mayrowetz have appeared recently in Educational Policy, Educational Administration Quarterly, Journal of School Leadership, and Leadership, Policy in Schools. Viviane Robinson is a Professor of Education at The University of Auckland and Academic Director of the First-time Principals Programme – New Zealand’s national induction programme for school principals. Her research focuses on leadership and organisational learning. She has recently published in Educational Administration Quarterly, Leadership and Policy in Schools, Journal of Educational Administration and Educational Management and Leadership. Karen Seashore Louis is Rodney S. Wallace Professor in Educational Policy and Administration at the University of Minnesota. Her research focuses on school organization, leadership, and policy. She has served as a vice-president for the American Educational Research Association, and on the executive committee of the University Council for Educational Administration. Her most recent book is co-authored with Sharon Kruse (2009) Strong Cultures: A Principal’s Guide to Change. Thousand Oaks, CA: Corwin. Mark A. Smylie is Professor of Education in the Department of Educational Policy Studies at the University of Illinois at Chicago. His research focuses on education organizational change, teacher and administrative leadership, and urban school improvement.

xii

Contributors

James P. Spillane is the Spencer T. and Ann W. Olin Chair in Learning and Organizational Change at Northwestern University where he is a Professor of Human Development and Social Policy, Learning Sciences, and Management and Organizations. Spillane is a Faculty Fellow at Northwestern University’s Institute for Policy Research and is a senior research fellow with the Consortium for Policy Research in Education (CPRE). With funding from the National Science Foundation, Spencer Foundation, and Institute for Education Sciences, Spillane’s work explores the policy implementation process at the state, school district, school, and classroom levels, and school leadership and management. Tiiu Strauss is currently a project director working with Kenneth Leithwood in the Department of Theory and Policy Studies at the Ontario Institute for Studies in Education at the University of Toronto, Canada. She has published in the areas of leader problem solving and distributed leadership, and is involved in research projects related to leadership in turnaround schools.

Part I

Setting the Scene

Introduction Alma Harris

The field of school leadership is currently preoccupied with the idea of distributed leadership. Few ideas, it seems, have provoked as much attention, debate and controversy. Whatever your position on distributed leadership, and you cannot fail to have one, it is irrefutable that distributed leadership has become the leadership idea of the moment. Yet, it is an idea that can be traced back as far as the mid 20s and possibly earlier. So why the interest? Part of the answer can be found in a move away from theorizing and empirical enquiry focused on the single leader. This shift has undoubtedly been fuelled by structural changes, within schools and across school systems that have resulted in alternative models or forms of leadership practice. Evidence highlights how those occupying formal leadership positions are increasingly recognizing the limitations of existing structural arrangements to secure organizational growth and transformation (Fullan et al., 2007; Harris et al., 2008; Chapman et al., 2008). As a consequence, many heads and principals are actively restructuring, realigning and redesigning leadership practice in their school (Harris, 2008). While the terminology to describe such changes varies, the core principle is one of extending or sharing leadership practice. While scholars have long argued for the need to move beyond those at the top of organizations in order to examine leadership (Barnard, 1968; Katz and Kahn, 1966) until relatively recently, much of the school leadership literature has tended to focus upon the head or the principal. The growth of what Gronn (2003) has termed “greedy work” in schools has undoubtedly contributed to a renewed and pragmatic interest in sharing leadership responsibility. Teacher leadership, student leadership and community leadership, previously overshadowed because of the dominance of more traditional forms of leadership, are now receiving much greater attention (Wilmore, 2007). While many organizational theorists spent most of the twentieth century exploring the variations and consequences of hierarchies, interest has now moved to a

A. Harris (B) Institute of Education, 20 Bedford Way London, WCIH OAL

A. Harris (ed.), Distributed Leadership, Studies in Educational Leadership 7, C Springer Science+Business Media B.V. 2009 DOI 10.1007/978-1-4020-9737-9 1, 

3

4

A. Harris

consideration of peer based relations, interactions and co-leadership (Heckscher, 2007). Interest in collaborative systems, reflected in both the management literature and the academic social sciences literature, is best exemplified by the explosion of interest in complexity science and network theory. From a complexity perspective those in formal leadership positions “emphasize the management of independencies” and are primarily concerned with rich networks of relationships rather than “controls over process or outcomes” (Leithwood et al., 2009a: 7). Network theory is similarly concerned with interdependencies and the distribution of power across complex systems. As Wheatley notes (1998: 164) nothing exists independent of its relationships, whether looking at subatomic particles or human affairs. This position implies that maximizing interconnections and interactions among organizational members is more likely to result in positive growth and development. Organisational learning theory and theories of distributed cognition assume that existing capacities of individual members can be enhanced through social interaction and connecting sources. Distributed cognition views a system as a set of representations, and models the interchange of information between these representations. Hutchins (1995) emphasizes the interdependence of the individual and highlights how human activity is widely distributed across a complex system. His work also emphasizes the power of lateral agency, understood as the potential for change and learning to emerge in a horizontal as well as a vertical direction. It is possible for a team to organize its behavior in an appropriate sequence without there being a global script or plan anywhere in the system. Each member only needs to know what to do when certain conditions are produced in the environment. (Hutchings, 1963: 24)

Distributed cognition suggests that capacities are distributed throughout the social and material conditions of the organization and that they are fluid rather than fixed. The implication here is that making better use of existing capacities, including leadership, within in the organization is likely to result in some advantage. From this perspective, distributing leadership is more likely to have a positive impact on the organization if it is aligned to the contours of expertise and the provision of conditions that support social learning. Jim Spillane’s highly influential and groundbreaking work on distributed leadership theory draws heavily upon distributed cognition and social learning theory (Spillane et al., 2001). Distributed cognition is largely concerned with sources and patterns of influence that occur within organizations. Using this theoretical position, Spillane et al., (2004) suggests that a distributed perspective on leadership has two aspects: the leader plus aspect and the practice aspect. Drawing on distributed cognition theory, Spillane (2006: 19) argues that a distributed perspective necessitates understanding how aspects of the situation enable and constrain leadership practice and thereby contribute to defining it. Distributed leadership is a lens to understand leadership practice; it is a conceptual and analytical framework for studying leadership interaction. Others in the field take a different perspective on distributed leadership ranging across the normative, descriptive, predictive and discursive. Indeed, the popularity

Introduction

5

of distributed leadership has meant that interpretations of the term can easily slide between these various positions (Harris, 2007). The net result of this conceptual fluidity has meant that distributed leadership can be misleading because of the possibility of “meaning all things to all people” (Spillane, 2006: 102). The chameleon like quality of distributed leadership invites both misinterpretation and misunderstanding. One common misunderstanding is that distributed leadership is a convenient “catch all” descriptor for any form of shared, collaborative or extended leadership practice. This view is certainly quite prevalent, blurring further the meaning of the term. From the literature it is clear however that distributed leadership is much more than shared leadership practice. Distributed leadership focuses on the nature and form of leadership practice and the particular configuration of interactions between leaders, followers and the situation (Spillane, 2006: 14). Distributed leadership is concerned with the coperformance of leadership practice and the nature of the interactions that contribute to co-performance. Inevitably, patterns of distribution vary within; between and across schools with differential organizational outcomes or effects (Leithwood et al., 2009b; Spillane et al.,, 2001 Harris, 2008). To position distributed leadership as the antithesis of top-down, hierarchical leadership is also to misinterpret the term. While distributed leadership is certainly an alternative way of construing leadership practice and can be situated in relation to “top down” models of leadership, it is not the opposite. Distributed leadership, essentially involves both the vertical and lateral dimensions of leadership practice. Distributed leadership encompasses both formal and the informal forms of leadership practice within its framing, analysis and interpretation. It is primarily concerned with the co-performance of leadership and the reciprocal interdependencies that shape that leadership practice (Spillane, 2006: 58). As Lakomski (2009) has noted, despite such misunderstanding, there seems to be “unbridled enthusiasm” for distributed leadership and much critiquing, researching, writing, analyzing and advocating in its name. Some would argue that the idea is simply not worth pursuing. Others propose it is managerialisim in a new guise (Fitzgerald and Gunter, 2006). However looking at the literature, it is clear that unlike so many other “labels for leadership”, that tend dominate the school leadership field, distributed leadership not only brings a strong theoretical framing but also an increasingly compelling evidential base.

1 The Book This aim of this book is to bring together the latest thinking and research evidence on distributed leadership. Several of the contributions build upon work published in a special edition of the Journal of Educational Administration1 that explored different

1 Harris, A (guest editor) (2008) ‘Distributed Leadership through the looking glass’ Journal of Educational Administration Vol 46 No 2.

6

A. Harris

aspects of distributed leadership. Collectively the chapters represent the work of international scholars who offer divergent perspectives on distributed leadership. The book’s main purpose is to present the latest empirical evidence, conceptual development and theoretical analysis while also considering future directions for research in this area. The chapters in this book challenge as much as they confirm. They illuminate and illustrate some of the complexity, confusion and contradiction associated within distributed leadership. This is not a book seeking “cosy consensus” but rather is a text that draws together serious intellectual analysis and debate underscored with rigorous empirical evidence. Ken Leithwood and his colleagues (2009a: 281) suggested that without a more nuanced appreciation of the anatomy of distributed leadership it is not at all clear how one would have conceptualized and measured distributed leadership in order to assess its effects, whatever they might have been. The chapters that follow build upon the substantial intellectual foundation provided in Leithwood et al. (2009a) and place distributed leadership in even sharper focus. Through greater clarity about the contours, nature and operationalisation of distributed leadership, there is undoubtedly much greater scope to explore the issue of impact and effect. The book is divided into sections that consider the empirical, practical, conceptual, theoretical dimensions of distributed leadership. It is acknowledged that any such delineation is always problematic as many of the chapters deal with several of these dimensions. Consequently, it is acknowledged that the divisions are imperfect but they exist to provide a framework for the book. The first section of the book is concerned with the empirical dimension and encompasses a range of chapters that have data rich accounts of distributed leadership practice. The initial chapter by Harris provides a summary of what we know about the relationship between distributed leadership and organizational improvement. It considers a wide range of empirical evidence and highlights the growing number of studies, including those outlined in this book, that point towards a positive relationship between distributed leadership and certain organizational outcomes. The chapter argues that there is now a sufficient empirical foundation to design impact studies of distributed leadership. The chapter concludes that despite methodological challenges, investigation into the impact of distributed leadership on organizational outcomes is now needed. The next chapter by Camburn and Won Han focuses on the relationship between distributed leadership and instructional change. The chapter explores the outcomes of distributed leadership by drawing upon extensive evidence from an investigation into the America’s Choice CSR program. A core design feature of this program is the requirement to distribute leadership responsibilities to teacher leaders in schools, and this distribution of leadership in turn, is intended to act as a key lever for instructional change. The chapter outlines the impact of this program in 30 urban elementary schools. It investigates the association between the distribution of leadership to teachers and instructional change. This study concludes that distributing leadership to teachers can support instructional change. The next chapter by Spillane, Camburn, Amber Pareja and Lewis draw upon data from a large scale study to explore different operationalizations of a distributed

Introduction

7

perspective. The chapter’s central argument is that using a distributed framework to study leadership and management involves operationalizing core aspects of the framework. By taking a distributed perspective the authors suggest the need to develop study operations that allow researchers to validly describe and examine constructs implicit in a distributed leadership analytical framework. The chapter examines four operationalizations of the “leader plus” aspect of a distributed perspective by drawing upon data from two elementary schools. The chapter concludes by arguing that there is a need to give much more attention to study operations when investigating leadership and management practices when taking a distributed perspective. The chapter by Mascall, Leithwood and Strauss looks at the relationship between four patterns of leadership distribution and teachers’ academic optimism. The chapter argues that where teachers believe in the power of cooperation, and in the use of reflection and dialogue as the basis for decision-making, and when they hold realistic beliefs about the leadership capacities of their colleagues, they are more likely to perceive themselves to be engaged in more planned and aligned patterns of leadership distribution. The authors do not claim that academic optimism is either an antecedent or an outcome of planfully aligned forms of distributed leadership but their results suggest that there is more to be gained if distributed leadership in organized in a planful way. The final chapter in this section, by Hallinger and Heck explores the impact of system policies on the development of distributed school leadership and school improvement. The authors synthesize the results of a series of analyses of empirical data on distributed leadership and school improvement. These studies centered on the impact of new state policies that sought to create broader and deeper leadership capacity in schools as a vehicle for stimulating and sustaining school improvement. The quantitative analysis and results support a relationship between distributed leadership and school capacity for improvement. The results outlined in this chapter suggest that distributed leadership is an important co-effect of school improvement processes. Section three takes a practical or normative perspective on distributed leadership. This section incorporates two chapters which are richly grounded in data and focus primarily upon the lived reality of distributed leadership in various school contexts. The first chapter by Day focuses on a primary or elementary school principal in England. It follows a principal over a seven year period of leadership when she “turned around” a school which was under threat of closure. The chapter suggests that layered rather than distributed leadership is the key to sustained school improvement. The chapter considers the sequencing of leadership activities through the various phases of school development and recovery. It highlights how those in formal leadership positions orchestrate the conditions where informal leadership practices can develop and thrive. The second chapter in this section by Dinham is concerned with action learning and how distributed leadership is both a precondition and product of this process. The chapter reports on an evaluation of fifty Quality Teaching Action Learning (QTAL) projects coordinated by the New South Wales (Australia) Department of

8

A. Harris

Education and Training (NSW DET). These projects carried out as part of the Australian Government Quality Teaching Program (AGQTP) were designed to drive improvement in school education. The chapter provides an analysis of the findings and explores the importance of distributed leadership in the process of organizational learning. Section four takes a conceptual and theoretical perspective on distributed leadership. The initial chapter by Louis, Meyrowetz, Smyle and Murphy focuses on the development of distributed leadership and explores how various organizational factors might affect its configuration. The chapter draws upon in-depth empirical evidence collected in six high schools in the USA. Each of the schools participated in an initiative designed to promote distributed forms of leadership. Within this chapter the authors explore two crucial factors in the development of distributed leadership; sense making and trust. They examine how the process of understanding (making sense) of distributed leadership and the nature of existing relationships in the school shape the design, nature and the development of distributed leadership by school-level personnel. The chapter by Hargreaves and Fink offers a conceptual analysis of distributed leadership practice. It discusses distributed leadership in conjunction with the interrelated ideas of living systems, communities of practice and networks. The chapter explores how distributed leadership operates as a network of strong cells organised through cohesive diversity and emergent development rather than mechanical alignment and predictable delivery. The chapter also takes a critical stance on distributed leadership practice. It questions whether, in practice, these lateral leadership strategies are being used to extend democratic public and professional involvement in developing the goals and purposes of education or whether they are being primarily used as motivational devices to re-energize a dispirited profession into producing more effective and enthusiastic delivery of imposed government performance targets. The final section in the book considers future perspectives and directions for distributed leadership. The chapter by Gronn reflects upon and analyses the meaning of distributed leadership offering a critical perspective. The chapter considers future directions for distributed leadership and questions what if anything comes next? The chapter argues that it would be profitable for researchers to be thinking beyond it. The chapter suggests what this would entail, why this needs to occur and how it might be done. The final chapter by Robinson explores the potential of investigating the relationship between distributed leadership and student learning outcomes. It argues that such an exploration needs to discriminate between leadership practices in general and those particular leadership practices that develop more effective teaching and learning. The chapter argues that developing an account of distributed leadership that makes this kind of discrimination requires an integration of research on leadership with pedagogical theory and evidence. The chapter addresses the fundamental question of the shifts that might be required in order to discover and investigate the types of distributed leadership practice that have more and less impact on student outcomes.

Introduction

9

The main reason for compiling this book was to draw together the most recent thinking and evidence about distributed leadership. It is clear that while the knowledge base is increasing, there is still more conceptual ground clearing, theoretical sharpening and empirical investigation to be undertaken. While the idea of leadership distribution may be “old” the research activity focused on contemporary interpretations of the term are relatively new. In short, there is much more empirical work to be done. This is both good news and bad news. First the bad news; it is evident that there is still some way for the field to go to ensure that the research base on distributed leadership is extensive and substantial. The empirical evidence about distributed leadership and organizational change is encouraging rather than conclusive. This is still relatively un-chartered research territory. Now the good news; there is more intellectually rich work to be undertaken and interesting methodological challenges to be faced for those researchers who are serious about improving our understanding of the nature, impact and effect (negative or positive) of distributed leadership practice on organizational change. By undertaking such work there is the real possibility of making a significant and a considerable contribution to the field.

References Barnard, C. (1968). Functions of the Executive. Cambridge, MA, Harvard University Press, Cambridge, Massachusetts USA. Chapman, C. Ainscow, M. and Gunter H. (2008) New Models of Leadership: What Do We Know from the Literature? Nottingham, National College for School Leadership. Fitzgerald, T. and Gunter, H. (2006), “Teacher leadership? A new form of managerialism”, New Zealand Journal of Educational Leadership, Vol. 21, No. 2, pp. 44–57. Fullan, M. Hill, P. and Crevola, C. (2007). Breakthrough. Thousand Oaks, Corwin Press. Gronn, P. (2003) The New Work of Educational Leaders: Changing Leadership Practice in an Era of School Reform. London, Paul Chapman. Harris, A. (2007) Distributed Leadership: Conceptual Confusion and Empirical Reticence International Journal of Leadership in Education Vol. 10, No. 3, pp. 1–11. Harris, A (2008) Distributed Leadership: Developing Tomorrow’s Leaders, London Routledge Press. Harris, A. Allen, T. and Goodall, J. (2008) Capturing Transformation, London Specialis Schools and Academies Trust. Heckscher, C. (2007). The Collaborative Enterprise. London, Yale books. Hutchins, E. T. (1995) Cognition in the wild.. Cambridge, MA, MIT. Katz, D. and Kahn, R. L. (1966) The Social Psychology of Organizations. New York, Wiley. Lakomski, G. (2009) Functionally adequate but causally idle: whither distributed leadership, Journal of Educational Administration Vol. 46, No. 2, pp. 159–172. Leithwood, K. Mascall, B. and Strauss, T. (2009a) Distributed Leadership According to the Evidence, London, Routledge. Leithwood, K. Mascall, B. Strauss, T. Sacks, R. Memon, N. and Yashkina, A. (2009b) Distributing Leadership to Make Schools Smarter: Taking Ego out of the System in Leithwood, K. Mascall, B. and Strauss, T. Distributed Leadership According to the Evidence, London, Routledge. Spillane, J. Halverson, R. and Diamond, J. (2001) “Towards a Theory of Leadership Practice: A Distributed Perspective.” Northwestern University, Institute for Policy Research Working Article.

10

A. Harris

Spillane, J.P., Halverson, R. and Diamond, J.B. (2004). “Towards a theory of leadership practice: A distributed perspective.” Journal of Curriculum Studies, Vol. 36, No.1, pp.3–34. Spillane, J.P. (2006). Distributed leadership. San Francisco: Jossey-Bass. Wilmore, E. (2007) Teacher Leadership: Improving Teaching and learning from the Inside, New York, Corwin Press.

Chapter 1

Distributed Leadership: What We Know Alma Harris

Introduction Distributed leadership has caught the attention of researchers, policy-makers practitioners and educational reformers (Spillane, 2006; Harris, 2008; Leithwood et al., 2009a). It is the leadership idea of the moment, even though its genesis can be traced back to the field of organizational theory in the mid 1960s (Barnard, 1968). Critics argue that distributed leadership is nothing more than a “new orthodoxy” which reinforces managerialist principles (Fitzgerald and Gunter, 2007). Alternatively, others argue it offers a new way of thinking about leadership in schools and provides a powerful tool for transforming leadership practice (Spillane et al., 2001). Unfortunately, the leadership field is particularly susceptible to new theories or labels for leadership. The leadership industry is fuelled by new ways of defining and describing leadership practice. Many of these theories or labels appear in the leadership field without any empirical evidence or testing. They are championed, celebrated and exploited before being added to the stockpile of redundant leadership theories. Distributed leadership is undoubtedly “the new kid on the block” (Gronn, 2000) and is firmly centre stage in the discourse about leadership and organizational change. But what evidence do we have that distributed leadership is all that some claim or wish it to be? How do we know if it has more potential than any other form of leadership to secure organizational change and improvement? What evidence do we have about its impact on organizational change and development? This chapter reviews the evidence base and explores what is known about the relationship between distributed leadership and organizational change. It outlines the empirical evidence, highlights the main findings and identifies areas for future investigation.

A. Harris (B) Institute of Education, 20 Bedford Way, London, WCIH OAL Note: This chapter is based on Harris, A. (2008) Distributed Leadership: What We Know? Journal of Educational Administration 46 Issue: 2 Page:172–188.

A. Harris (ed.), Distributed Leadership, Studies in Educational Leadership 7, C Springer Science+Business Media B.V. 2009 DOI 10.1007/978-1-4020-9737-9 2, 

11

12

A. Harris

What We Know The contemporary evidence base about distributed leadership is still emerging but there is now increasing evidence about its relationship with organizational change (Leithwood et al., 2009a). A wide range of studies now exist that have explored how different patterns of distribution influence organizational outcomes and organizational change (Spillane and Camburn, 2006; Harris, 2008; Leithwood et al., 2009b; Mayrowetz et al., 2009; Camburn and Won Han; Hallinger and Heck in this book). Findings show that the configuration of leadership distribution is important and that certain patterns of distribution have a more positive effect than others upon organizational development and change (Leithwood et al., 2009a). The emerging evidence also shows that distributed leadership has a greater impact upon organizational development where certain structural and cultural barriers are removed (Harris, 2008). It highlights that the configuration of distribution matters in school development and change. These findings are further substantiated by the organizational development and improvement literature. Within this literature is the strongest indication yet that distributed leadership has the potential to positively influence organizational change (Iandoli and Zollo, 2008). Interest in the idea of distributed leadership has undoubtedly been fuelled by associations with certain organizational benefits (e.g., Manz and Sims Jr., 1993; Gronn, 2002; Burke, Fiore et al., 2003). Work by Graetz (2000) presents distributed leadership as a positive channel for change. He notes that “organisations most successful in managing the dynamics of loose – tight working relationships meld strong personalized leadership at the top with distributed leadership”. Similarly Gold et al. (2002) point towards the development of leadership capacity within the school and its distribution as a key lever of organizational success. In their review of successful school improvement efforts, Glickman et al. (2001) construct a composite list of the characteristics of the “improving school” (p.49) defined as a school that continues to improve student learning outcomes for all students over time. At the top of this list appear “varied sources of leadership”, including distributed leadership. The school improvement literature similarly highlights a positive relationship between distributed leadership and change in schools. This research base has consistently underlined the importance of teacher involvement in decision making processes and the contribution of strong collegial relationships to school improvement and change. Little (1990) suggests that collegial interaction lays the groundwork for developing shared ideas and for generating forms of leadership that promote improvement. Rosenholtz (1989) argues even more forcibly for teacher collegiality and collaboration as means of generating positive change in schools. Her research concludes that effective schools have tighter congruence between values, norms and behaviors of principals and teachers and that this is more likely to result in positive school performance. Distributed leadership also features predominantly in the literature relating to professional learning communities. This literature shows that professional learning communities make a significant difference to student achievement and that

1 Distributed Leadership

13

leadership within these communities is widely shared or distributed (Louis and Marks, 1998). Extending leadership responsibility beyond the principal is shown to be an important lever for developing effective professional learning communities in schools (Morrisey, 2000) A range of other studies (Portin, 1998; Blase and Blase, 1999; Hallinger, 2000; Hallinger and Heck in this book) reinforce a positive relationship between certain organizational outcomes and distributed forms of leadership practice. More recently, distributed leadership has been identified as a key influence in sustaining professional learning communities (Stoll and Louis, 2007). It might be expected therefore that if even just a few of the claims made in the literature about distributed were to materialize, then the impact on schools would be significant (Leithwood et al., 2009a). However the evidence about impact is limited. Even within the teacher leadership literature, which is extensive, there is relatively little evidence of a direct impact on student learning outcomes (York and Barr, 2004). Rather, the empirical weight supports the greatest impact on the professional growth of the teacher leaders themselves, in terms of higher self efficacy and morale (Murphy, 2005). There is some evidence which shows that informal leadership “dispersion” can negatively affect team outcomes by contributing to inefficiencies within the team. Therefore having fewer leaders rather than more would seem preferable. In their work, Heinicke and Bales (1953) found that having fewer informal leaders was positively related to task efficiency over time. Other writers have argued that having fewer informal leaders enhanced peoples’ feelings of being socially validated for their work (Festinger, Schacter et al., 1950). Distributed leadership has also been identified as one of six “obstacles” to effective team performance: the clear differentiation of role responsibilities and the assignment of those responsibilities to particular team members provides a measure of stability and predictability that is otherwise lacking on a team where role assignments are poorly defined. . . An individual’s security derives largely from his being able to count upon a stable social environment. (Melnick, 1982, p 3.)

The contemporary literature points to some of the practical difficulties associated with distributing leadership in schools. It shows that distributed leadership can result in conflicting priorities, targets and timescales. Boundary management issues and competing leadership styles can emerge (Storey, 2004: Timperley, 2009). The evidence shows that in practical terms, implementing distributed leadership implies crossing structural and cultural boundaries. While distributed leadership among teachers may be desirable, some caution needs to be sounded about the potential difficulties involved. Although formally appointed leaders do not automatically command respect and authority, teacher leaders may be particularly vulnerable to being openly disrespected and disregarded because they do not carry formal authority. On the other hand, nomination of teacher leaders by colleagues may not realize potential expertise within the group because colleagues may select their leaders using other criteria (Timperley, 2005). p 412).

14

A. Harris

In their work, Fitzgerald and Gunter (2007:6) question whether it is possible for distributed leadership to occur in a policy climate that affords authority and responsibility for leadership and management to those labeled according to an established hierarchy? The implication here is that existing school structures mediate against distributed leadership practice and that this type of informal influence and agency is not possible within the existing hierarchical structure of schooling. Fitzgerald and Gunter (2007) also suggest that teacher leadership merely cements authority and hierarchy whereby “leaders” monitor teachers and their work to ensure a set of pre-determined standards are met. The empirical evidence conversely shows that authority and hierarchy are more likely to be challenged by those who are teacher leaders (York-Barr and Duke, 2004). The evidence highlights that the role of “leader” is not one imposed upon teachers, it is usually selected. Similarly, there is little evidence to suggest that teacher leadership is used a mechanism to monitor performance (Lieberman, 1995; Murphy, 2005). As highlighted earlier, the literature does reveal some reservations about the relationship between teacher leadership and student learning outcomes. This relationship is worth particular attention and scrutiny.

Distributed Leadership and Student Learning Outcomes Positions on the relationship between distributed leadership and student learning outcomes vary. Some writers have argued that seeking to explore this relationship is a futile exercise. They suggest that the search for normative links between specific leadership distribution patterns and student achievement results is unlikely to yield clear guidelines for practice (Anderson et al., 2009:135). Others have argued that distributing leadership is only desirable if the quality of leadership activities contributes to “assisting teachers to provide more effective instruction to their students” (Timperley, 2009:220). Reflecting upon these positions, Robinson (2008) and Harris (2008) argue that any exploration of the relationship between distributed leadership and student learning outcomes is only desirable and possible, if the leadership literature is more closely connected to the pedagogical literature. It is suggested that linking studies of leadership practice much more closely to learning practice will bring us closer to addressing the question of whether and how different pattern or configurations of distributed leadership contribute to student learning outcomes. There are already a range of studies that have touched upon the relationship between distributed leadership and learning outcomes. There are two studies, in particular, that offer a useful starting point in highlighting what we know about distributed leadership and student learning outcomes. The first study undertaken by Leithwood and Jantzi (2000) in Canada and the second conducted by Silins and Mulford (2002) in Tasmania, both focus on this relationship as part of much broader empirical investigation. The findings from the Leithwood and Jantzi (2000) research suggest that distributing a larger proportion of leadership activity to teachers has a positive influence on teacher effectiveness and student engagement. They also note that teacher

1 Distributed Leadership

15

leadership has a significant effect on student engagement that far outweighs principal leadership effects after taking into account home family background. Silins and Mulford’s (2002) work has also provided cumulative confirmation of the key processes through which more distributed kinds of leadership influence student learning outcomes. Their work concluded that student outcomes are more likely to improve when leadership sources are distributed throughout the school community and when teachers are empowered in areas of importance to them. (p) Other, smaller scale studies, have also demonstrated a positive relationship between distributed leadership and student learning outcomes. A study of teacher leadership conducted in England found positive relationships between the degree of teachers’ involvement in decision making and student motivation and self efficacy (Harris and Muijs, 2004). This study explored the relationship between teacher involvement in decision making within the school and a range of student outcomes. These findings show a positive relationship between distributed leadership and student engagement. In addition, both teacher and student morale improved where teachers felt more included and involved in decision making within the school. The “Distributed Leadership Study” (Spillane et al., 2001) undoubtedly remains the largest contemporary study of distributed leadership practice in schools. This four year longitudinal study, funded by the National Science Foundation and the Spencer Foundation, was designed to make the “black box” of leadership practice more transparent through an in depth analysis of leadership practice. The central premise for the study is that distributed leadership is best understood as distributed practice, stretched over the school’s social and situational contexts. This research, which focused on 13 elementary schools in Chicago, found that the task of instructional improvement engaged multiple leaders and highlighted how understanding the interplay between different leaders is crucial to understanding leadership practice. This study suggests that the school rather than the individual leader is the most appropriate unit for thinking about the development of leadership expertise. It also noted that intervening to improve school leadership may not be most optimally achieved by focusing on the individual formal leader but may be secured through influencing the practices of multiple leaders. Work by Copland (2003) which focused on improvement in eighty six schools that were engaged in data-driven, whole school reform similarly found extensive staff involvement in the leadership of the schools. The study highlighted the importance of involvement at all levels in decision making. The research identified positive trends in performance resulting from the large scale move towards more participatory and distributed patterns of leadership. Other research that has looked explicitly at the relationship between leadership and student learning outcomes has also highlighted the importance of distributed leadership. Work by Day et al. (2007:17) notes: substantial leadership distribution was very important to a school’ success in improving pupil outcomes. The findings from this study also showed that distributed leadership was positively correlated to the conditions within the organisation, including staff morale, which in turn impacted positively upon student behaviour and student learning outcomes (Day et al. 2009).

16

A. Harris

The evidence from this project highlighted that leadership practices in the most effective schools were widely distributed and the relationship between vertical and horizontal leadership structures was porous and inter-changeable. It reinforced that the principals or heads largely determined the nature and pattern of leadership distribution in their schools. The patterns they chose were determined by three main influences: a) their personal view of leadership (e.g. need for control) b) their own stage of development as a leader c) their estimates of the readiness of their staff to take on greater leadership (Day et al., 2007:19). Other contemporary studies that have focused, in part, on the relationship between leadership and learning (Leithwood et al., 2009b; Mascall et al., Louis et al., in this book) have also highlighted the influence of distributed leadership practices. In their work, Mayrowetz et al., (2009) offer a theory of both how and why leadership distribution might be expected to have positive consequences for students. They argue that “redesigned work in schools can lead to motivated and better equipped educators with a broader view of their schools and that these educators in turn will perform leadership functions which will both deepen and lengthen reform efforts” (Mayrowetz, 2009:191). In their work Mascall et al., (in this book) explore the relationship between planfully aligned forms of distributed leadership and academic optimism. Their results offer reasonable support for efforts to ensure that leadership is distributed in schools in planful ways.

Patterns of Distribution Much of the recent empirical work concerning distributed leadership has focused on patterns of distribution (Leithwood et al., 2007, 2009a). Findings show that the patterns of leadership practice in a school affect organizational performance. They also reveal that the effects and impact of distributed leadership on organizational outcomes depends upon the pattern of leadership distribution. Leithwood et al. (2006) highlight two key features necessary for successful leadership distribution. First, leadership needs to be distributed to those who have, or can develop, the knowledge or expertise required to carry out the leadership tasks expected of them. Second, the initiatives of those to whom leadership is distributed need to be coordinated, preferably in some planned way (Leithwood, 2006). These two conditions for successful leadership distribution are the starting points for Locke’s (2003) “integrated model” of leadership. This model acknowledges both the reality and the virtues, in most organizations, of distributed leadership based on multiple forms of lateral (e.g., teacher to teacher) influence. Also acknowledged are the “inevitable” sources of vertical or hierarchical leadership in virtually any successful organization.

1 Distributed Leadership

17

Gronn (2002) distinguishes between two distinct forms of distributed leadership that help further clarify Locke’s model. Gronn labels these forms “additive” and “holistic”. Additive forms of distribution describe an uncoordinated pattern of leadership in which many different people may engage in leadership functions but without much, or any, effort to take account of the leadership efforts of others in their organization. Locke’s model suggests that such unplanned patterns of distributed leadership would do little to help the organization develop or grow. Holistic or person-plus leadership (Spillane, 2006) refers to consciouslymanaged and synergistic relationships among some, many, or all sources of leadership in the organization. These forms of distributed leadership assume that the sum of leaders’ work adds up to more than the parts. It is also assumed that there are high levels of interdependence among those providing leadership and that the influence attributed to their activities emerges from dynamic, multidirectional, social processes which, at their best, lead to learning for the individuals involved, as well as for their organizations (Pearce and Conger, 2003). Gronn (2002: 657) has suggested that concertive forms of distributed leadership may take three forms:

r r r

Spontaneous collaboration: From time to time groupings of individuals with differing skills and knowledge capacities, and from across different organizational levels, coalesce to pool their expertise and regularize their conduct for duration of the task, and then disband Intuitive working relations: This form of concertive distributed leadership emerges over time “. . .as two or more organizational members come to rely on one another and develop close working relations” and, as Gronn argues, “leadership is manifest in the shared role space encompassed by their relationship” Institutionalized practice: Citing committees and teams as their most obvious embodiment, Gronn describes such formalized structural as arising from design or through less systematic adaptation.

The extent and nature of coordination in the exercise of influence across members of the organization is a critical challenge from a holistic perspective. When role overlap occurs in a coordinated fashion there can be mutual reinforcement of influence and less likelihood of making errors in decisions. Some elaboration and refinement of Gronn’s (2002) holistic forms of distributed leadership have been proposed by Leithwood et al. (2006b):

r

Planful alignment – This configuration is comparable to Gronn’s “institutionalized practice”. The tasks or functions of those providing leadership have been given prior thoughtful consideration by organizational members. Agreements have been worked out among the sources of leadership (principals, heads of department and teachers etc) about which leadership practices or functions are best carried out by which source. Although alignment is generally considered a good thing for organizations, positive contributions of this configuration to productivity cannot be automatically assumed.

18

r

r

r

A. Harris

Spontaneous alignment – In this configuration, leadership tasks and functions are distributed with little or no planning, for example: the principal assumes she will be responsible for modeling values important to the school and everyone else makes the same assumption. Nevertheless, tacit and intuitive decisions about who should perform which leadership functions result in a fortuitous alignment of functions across leadership sources. There is no significant difference in the contribution to short-term organizational productivity of this “method” of alignment, as compared with planful alignment. However, the tacit nature of decisions this method entails seems likely to reduce the flexibility and adaptability of the organization’s responses to future leadership challenges. Spontaneity offers few guarantees of fortuitous alignment. Spontaneous misalignment – This configuration mirrors spontaneous alignment in the manner of leadership distribution, as well as its underlying values, beliefs and norms. However the outcome is different or less fortuitous – misalignment (which may vary from marginal to extensive). Both short- and long-term organizational productivity suffer from this form of (mis)alignment. However, organizational members are not opposed, in principle, to either planful or spontaneous alignment thus leaving open reasonable prospects for future productive alignment of one sort or another. Anarchic misalignment – This configuration is characterized by active rejection, on the part of some or many organizational leaders, of influence from others about what they should be doing in their own sphere of influence. As a result, those leaders’ units behave highly independently, competing with other units on such matters as organizational goals and access to resources. Active rejection of influence by others, however, stimulates considerable reflection about one’s own position on most matters of concern.

Findings by Leithwood (2006:2007) and Mascall et al. (in this book) suggest that planful patterns of alignment have the greatest potential for short-term organizational change. Furthermore, planful alignment seems more likely to contribute significantly than other patterns of distribution to long-term organizational productivity. The research found that both spontaneous misalignment and anarchic alignment were likely to have negative effects on both short and long-term organizational change and development.

Commentary The empirical evidence about distributed leadership and organizational development is encouraging but far for conclusive. We need to know much more about the barriers, unintended consequences and limitations of distributed leadership before offering any advice or prescription. We also need to know the limitations and pitfalls as well as the opportunities and potential of this model of leadership practice. The methodological challenges in addressing these questions are extensive (Timperley, 2005: 417). Yet, as Spillane et al. (2009; and in this book) have shown

1 Distributed Leadership

19

distributed leadership provides a powerful analytical lens to look at ways in which organizational development and learning take place. It offers a frame to help researchers in the leadership field build evocative cases that can be used to assist practitioners in thinking about their ongoing leadership practice. It also provides a basis for measuring and investigating a more comprehensive and complex set of leadership practices that go beyond the checklists of leadership characteristics, skills and strategies. Contemporary studies of distributed leadership are going beyond the conceptual and empirical descriptions of what distributed leadership is to look much more closely at issues of impact and outcome. Ironically, if we shun the opportunity to undertake research about impact because of the methodological challenges it is more, likely that we will be blinkered to the limitations of distributed leadership (Timperley, 2005). We undoubtedly need empirical studies that highlight both the inadequacies of distributed leadership practice, as well as the possibilities.

References Barnard, C. (1968). Functions of the Executive. Cambridge MA, Harvard University Press. Blase, J. R. and J. Blase (1999). “Implementation of shared governance for instructional improvement: Principals’ perspecitves.” Journal of Educational Administration 37(5): 476–500. Burke, C. S., S. M. Fiore, et al. (2003). The role of shared cognition in enabling shared leadership and team adaptability. Shared Leadership: Reframing the How and Whys of Leadership. C. J. Pearce and J. A. Conger. Thousand Oaks, CA, Sage: 103–122. Copland, M. A. (2003). “Leadership of inquiry: Building and sustaining capacity for school improvement.” Educational Evaluation and Policy Analysis 24(4): 375–475. Day, C., K. Leithwood, P. Sammons, A. Harris, and D. Hopkins (2007). Leadership and Student Outcomes, London, DCSF Interim Report. Day, C., K. Leithwood, P. Sammons, A. Harris, and D. Hopkins (forthcoming). Leadership and Student Outcomes, London, DCSF Final Report. Festinger, L., S. Schacter, et al. (1950). Social Pressure in Informal Groups: A Study of Human Factors in Housing. New York, Harper. Fitzgerald, T. and M. Gunter (2007). Teacher Leadership: A New Myth for Our Time? Chicago, AERA. Gold, A., J. Evans, P. Early, D. Halpin, and P. Collabone (2002). “Principled principals? Evidence from ten case studies of ’outstanding school teachers’.” Paper presented at the annual meeting of the American Educational Research Association, New Orleans, LA. Glickman, C. D., Gordon, S. P., and Ross-Gordon, J. M. (2001). Supervision and instructional leadership. MA: Allyn & Bacon., Needham Heights. Graetz, F. (2000). “Strategic change leadership.” Management Decisions 38(8): 550–562. Gronn, P. (2000). “Distributed properties: A new architecture for leadership.” Educational Management and Administration 28(3): 317–338. Gronn, P. (2002). Distributed leadership. Second International Handbook of Educational Leadership and Administration. K. Leithwood and P. Hallinger. Dordrecht, Kluwer Academic Publishers: 653–696. Hallinger, P. and P. Kantarmara (2000). “Educational change: Opening a window onto leadership as a cultural process.” School Leadership and Management 20(2): 189–205. Harris, A. and D. Muijs (2004). Improving Schools Through Teacher Leadership. London, Oxford University Press.

20

A. Harris

Harris, A. (2008). Distributed Leadership: What We Know? Journal of Educational Administration 46(2): 172–188 ISSN 9578234. Heinicke, C. M. and R. F. Bales (1953). “Developmental trends in the structure of small groups.” Sociometry 16: 7–38. Iandoli, L. and G. Zollo (2008). Organisational Cognition and Learning, New York, Idea Group Incorporated. Leithwood, K. and D. Jantzi (2000). The effects of different sources of leadership on student engagement in school. In K. Riley and K. Louis (Eds.), Leadership For Change and School Reform, London, Routledge.: 50–66. Leithwood, K. D., C. Sammons, P. Harris, A. and Hopkins, D. (2006a). Successful School Leadership What It Is and How It Influences Pupil Learning. London Report to the Department for Education and Skills. Leithwood, K., B. Mascall, T. Strauss, R. Sacks, N. Memon, and G. Yashkina (2006b). “Distributing Leadership to Make Schools Smarter.” Leadership and Policy. Leithwood, K., B. Mascall, and T. Strauss (2009a). Distributed Leadership According to the Evidence, London, Routledge. Leithwood, K., B. Mascall, T. Strauss, R. Sacks, N. Memon, and A. Yashkina (2009b). Distributing Leadership to Make Schools Smarter: Taking Ego out of the System. In K. Leithwood, B. Mascall, and T. Strauss (2009). Distributed Leadership According to the Evidence, London, Routledge. Lieberman, A. (1995). “Practices That Support Teacher Development: Transforming Conceptions of Professional Learning.” Phi Delta Kapan 6: 591–596. Little, J. W. (1990). The persistence of privacy: Autonomy and initiative in teachers’ professional relations. Teachers College Record 91(4): 509–536. Locke, E. A. (2003). Leadership: Starting at the Top in Pearce, C. J. and Conger, C. (eds) Shared Leadership: Reforming the Hows and Whys of Leadership, Thousand Oaks, CA Sage. Louis, K. S. and H. Marks (1998). “Does professional community affect the classroom? Teachers’ work and student work in restructuring schools.” American Educational Research Journal 33(4): 757–798. Manz, C. C. and H. P. Sims Jr. (1993). “Self-management as a substitute for leadership: A social learning perspective.” Academy of Management Review 5: 361–367. Mayorwetz, D. (2008). Making sense of Distributed Leadership: Exploring the Multiple Usages of the Concept in the Field Educational Administration Quartely 44: 424–432. Mitchell, C. and L. Sackney (2000). Profound Improvement, Building Capacity for a Learning Community. Lisse, Swets & Zeitlinger. Morrisey, M. (2000). “Professional learning communities: An ongoing exploration.” Austin, TX: Southwest Educational Development Laboratory. Murphy, J. (2005). Connecting Teacher Leadership and School Improvement. Thousand Oaks CA, Corwin Press. Portin, B. S. (1998). “Compounding roles: A study of Washington’s principals.” International Journal of Educational Research 29(4): 381–391. Rosenholtz, S. J. (1989). Teachers’ Workplace: The Social Organization of Schools. New York, Longman. Silins, H. and W. Mulford (2002). Leadership and school results. Second international handbook of educational leadership and administration. In K. Leithwood and P. Hallinger (Eds.). Dordrecht, The Netherlands, Kluwer: 561–612. Spillane, J., Halverson, R. and Diamond, J. (2001). “Towards a Theory of Leadership Practice: A Distributed Perspective” Northwestern University, Institute for Policy Research Working Article. Spillane, J. P. (2006). Distributed Leadership. San Francisco CA, Jossey-Bass. Spillane, J. P. and E. Camburn (2006). The practice of Leading and Managing: The Distribution of Responsibility for Leadership and management in the Schoolhouse. American Educational Research Association. San Francisco.

1 Distributed Leadership

21

Stoll, L. and Seashore Louis, K. (2007). Professional Learning Communities, New York, Open University Press. Storey, A. (2004). “The Problem of Distributed Leadership in Schools.” School Leadership and Management 24(3): 249–265. Timperley, H. (2005). “Distributed Leadership:Developing theory from practice.” Journal of Curriculum Studies 37(4): 395–420. York-Barr, J. and K. Duke (2004). What Do we Know about Teacher Leadership? Findings from Two Decades of Scholarship Review of Educational Research 74(3):255–316. Yukl, G. (2002). Leadership in Organisations. New Jersey, Prentice Hall.

Part II

Empirical Perspectives

Chapter 2

Investigating Connections Between Distributed Leadership and Instructional Change Eric M. Camburn and S.W. Han

Introduction To date, the literature on distributed leadership has mainly developed along two paths – conceptual writing about what distributed leadership is, and empirical studies describing whether and how leadership is distributed. At this stage of its development, this literature has not seriously addressed the potential consequences nor the benefits of distributed leadership. In summarizing a recent edited volume on the subject, Leithwood et al. (2008) acknowledge that the volume, and the larger body of research on distributed leadership has not yet assessed “the contribution of greater leadership distribution to the long list of desirable outcomes typically invoked by advocates” (p. 280). What benefits can be expected for schools in which leadership is distributed and how might distributed leadership help bring about such benefits? In this chapter, we take a small step towards addressing such questions by investigating the association between the distribution of leadership to teachers and instructional change in schools. There is growing evidence that comprehensive school reform (CSR) programs can support improved achievement by facilitating the adoption of new instructional strategies in schools (Borman et al., 2007). There is also emerging evidence that CSR programs can support student achievement and instructional change by reorganizing the ways in which schools manage and support instruction (Rowan et al., 2009). Research has further shown that distributed leadership is used by a number of these programs as a primary tool for reorganization (Camburn et al., 2003). This chapter reports an investigation of the America’s Choice CSR program. The design of this program is intended to distribute leadership responsibilities to teacher leaders in schools, and this distribution of leadership in turn, is intended to act as a key lever for instructional change. In this chapter, we examine how this intended design played out in about 30 urban elementary schools that implemented the America’s Choice design. We begin by describing persistent barriers to instructional change

E.M. Camburn (B) Department of Educational Leadership and Policy Analysis, University of Wisconsin-Madison, Madison, WI, USA

A. Harris (ed.), Distributed Leadership, Studies in Educational Leadership 7, C Springer Science+Business Media B.V. 2009 DOI 10.1007/978-1-4020-9737-9 3, 

25

26

E.M. Camburn and S.W. Han

and then present a conceptual argument for how distributing leadership to teachers in America’s Choice might serve as a mechanism for instructional change. We go on to describe our research strategy for testing this conceptual argument, and conclude with empirical results.

Persistent Barriers to Achieving Instructional Change The idea that classroom instruction is highly resistant to change has received consistent endorsement from scholars over the years (see for example Cuban, 1990; Fullan, 1991; Sarason, 1990). The list of potential barriers that are presumed to stand in the way of changing classroom instruction is familiar. Teachers are the ultimate brokers of change, and they may simply decide that change is not in their best interest, or they may make only surface changes to their practice that do not fundamentally alter the instructional experiences their students receive (Cohen, 1990; Firestone, 1989; Fullan, 1991; McLaughlin, 1987; Spillane et al., 2002). Even though many believe that teacher collaboration can serve as a springboard for instructional change, persistent teacher isolation and norms of privacy are believed to stand in the way of significant collegial work (Fullan, 1991; Little, 1990). Frequent shifts in national, state, and district priorities often mean that schools do not sustain a focus on a set of goals long enough for instructional changes to mature and take root (Cuban, 1990). Frequent policy shifts and multiple sources of instructional guidance can also create conflicting messages about expectations for teaching. Such incoherence in instructional guidance can serve as an impediment to change by muddying expectations for teachers and by diverting teachers’ attention away from desired targets of change (Cohen and Ball, 1999; Mazmanian and Sabatier, 1981; Newmann et al., 2001). Research suggests that placing leadership responsibility in the hands of teachers to act as instructional mentors and coaches to their peers might serve as an antidote to persistent barriers to instructional change. For example, a recent randomized trial found that teachers who received a coaching treatment had significantly more positive outcomes than control teachers, including higher levels of positive emotional climate in their classrooms, greater sensitivity to their students’ needs, and more effective classroom management (Raver et al., 2008). A meta-analysis of 112 studies conducted by Bennett (1987) found that when teacher training strategies included coaching (prolonged support in teachers’ classrooms from an expert teacher), the transfer of newly learned teaching strategies to the classroom was much more likely. But prior research also indicates that systematic efforts to distribute leadership among teachers, such as site-based management, career ladder initiatives, and early mentor teacher programs have often fallen short of their intended goal of enhancing the instructional leadership provided by schools. Explanations for the failure of these initiatives includes a lack of clearly defined expectations for leadership responsibilities, a lack of explicit or extended training for new leaders, and pervasive isolation in schools that severely limited social interactions through which teacher leadership is exercised (Hart, 1995; Lieberman, 1988; Little, 1995; Smylie et al., 2002; Smylie and Denny, 1990).

2 Investigating Connections Between Distributed Leadership and Instructional Change

27

Distributed Leadership and Instructional Change in CSR Programs Spurred in part by the ineffectiveness of school reforms of the 1970s and 1980s to overcome persistent barriers to instructional change, the New American Schools (NAS) initiative supported the development of “comprehensive” approaches to school improvement (Rowan et al., 2004). The comprehensive school reform (CSR) designs that emerged from this initiative reflected three features of the NAS theory of action (Berends et al., 2002).

r r r

The change process is centered around an explicit design that is a blueprint for change. Designs include both targets of change (features of schools that are to be restructured) and specific ideas about the people and processes that will be used to bring about the change (Rowan et al., 2004). The design should specifically and clearly communicate expectations about what school personnel need to do in order to implement the design. The model should provide sustained, ongoing assistance that directly supports school personnel’s implementation of the design.

Reflecting these design principles, CSRs have been shown to address persistent challenges of instructional change by strategically distributing leadership in a manner that supports teacher development. A study conducted by the first author and colleagues documented how this process worked in three CSR programs (Camburn et al., 2003). The study examined how responsibility for leadership functions was distributed among staff holding formal leadership roles in schools. The areas of leadership examined included instructional leadership, building management (scheduling, maintenance of physical plant, personnel), and boundary spanning functions (working with parents, community, school board, etc). The study further examined three sub-domains of instructional leadership including setting instructional goals, developing teachers’ instructional capacity, coordinating curriculum, and monitoring school improvement efforts. The study was based on an attempted census of staff holding formal leadership positions in 114 schools. Based on self-reports, leaders were categorized into four broad roles: principal, assistant principal, CSR Coach (CSR-facilitators and subject area coordinators), and “Other” Leader (non-CSR subject area facilitators, program coordinators, master teachers, and mentor teachers). Camburn et al. (2003) outlined a two step process by which leadership was distributed in schools implementing CSR programs. They called the first step “configuration” which refers to the creation of formal leadership positions in a school. The second step called “activation” refers to social processes that provide guidance about how the leadership functions associated with these formal roles should be enacted. This split between configuration and activation reflects Linton’s (1936) classic role theory in which he made a clear distinction between status, or one’s position within a social structure, and role, which are the behaviors expected of people who occupy a particular status. Camburn et al. (2003) viewed configuration as a process that defined formal leadership statuses and activation as a process that defined roles, that is, the expected performances for people holding particular leadership statuses.

28

E.M. Camburn and S.W. Han

The study by Camburn et al. (2003) found evidence of configuration and activation. Schools implementing the America’s Choice and Success for All CSR designs had more total leadership positions than a group of comparison schools, and had a higher percentage of staff assigned to the kinds of teacher leadership positions called for by the two designs. Schools implementing these two programs were also found to allocate more leaders per teacher, again, reflecting the intention of the two programs to add leadership positions to support teacher development. Schools implementing the Accelerated Schools Project allocated more leaders per teacher than comparison schools, but otherwise appeared very similar to comparison schools in the way in which leadership positions were configured. Camburn et al. (2003) also found evidence supporting the idea that CSR programs activated leadership by defining role expectations and training role incumbents. A review of the designs of the three CSR programs studied indicated that teacher leaders filling coach and facilitator roles added by the programs were to receive significant training in how to perform their instructional leadership roles. Statistical analyses revealed that the amount of professional development received by leaders was strongly and positively related to the performance of four instructional leadership functions. These analyses also illustrated a clear division of labor in the way in which leadership functions were distributed across role incumbents. While principals were more likely than CSR coaches to set instructional goals and monitor improvement initiatives, CSR coaches were more likely than principals to provide leadership that directly supported the improvement of teachers’ practice. This observed role specialization is precisely what one would predict given the primary responsibility of the coach roles as defined by the CSR programs.

Instructional Change in Schools Implementing the America’s Choice Design This chapter picks up where Camburn et al. (2003) left off. While that study established that CSR programs distribute leadership in a way that supports instructional change, this study goes a step further and investigates whether the configuration and activation of teacher leadership positions in a single CSR program supports the adoption of instructional practices advocated by that program. In this section we describe the program’s approach to literacy instruction and discuss how leadership responsibilities were distributed to teachers as part of the program design.

America’s Choice Approach to Literacy Instruction At the time data was collected for this study, the America’s Choice program outlined a set of research-based instruction routines for teacher-student interaction around complex literacy content. The routines had five distinctive components: (1) a significant emphasis on instruction in the writing process, (2) the use of genre and

2 Investigating Connections Between Distributed Leadership and Instructional Change

29

literary technique as a vehicle for understanding text structures, (3) routines through which students received responses to their writing from fellow students and teachers, (4) a focus on guided reading, and (5) the integration of reading and writing. A good deal of writing instruction in the America’s Choice program is encompassed by an activity called “Writers Workshop.” Work in Writer’s Workshop is divided into Mini-lessons and Work sessions. Mini-lessons are brief instructional segments (10–15 minutes) in which the focus of the day’s literacy work is introduced. In Mini-lessons, teachers also provide explicit instruction in writing skills and strategies including idea generation, genre-specific writing strategies, and the critique and revision of text. “Work sessions” comprise the remainder of the literacy period, and in these sessions students work on the focal literacy topic through independent writing and through conferences with teachers and peers. The America’s Choice design stipulates that students receive regular feedback on their writing from peers and teachers. One strategy for such feedback is for students to read drafts of their writing out loud and to receive feedback from their audience (either peers or teachers). Students are then expected to revise their writing based on this feedback. Peer critique is expected to take the form of “accountable talk” in which students use evaluative criteria outlined by America’s Choice to examine each other’s work. Students are also expected to apply these same criteria in critiquing their own writing. Students also receive instruction and feedback on their writing from teachers in student-teacher conferences. These interactions are intended to guide students and scaffold their progress during all phases of the writing process. The America’s Choice program’s approach to reading, called Reader’s Workshop, prominently features guided reading, an activity where students read aloud, where teachers diagnose students’ reading difficulties, and where students receive guidance and support from their teacher as they read. The program also stipulated that teachers should determine the reading level of books read by students so that this information could be used for diagnosis during guided reading. Whereas most proponents of guided reading describe it as a strategy to promote fluency, the America’s Choice program views guided reading as a strategy for promoting comprehension. At the time of this study, the activities America’s Choice used to implement guided reading paralleled the activities of Writer’s Workshop including the use of Mini-lessons, Work-sessions, teacher-student conferencing and peer feedback. One of the most distinctive features of the program’s approach to literacy is the integration of reading and writing instruction. The program’s focus on genre studies and literary techniques is one avenue through which integration occurs. By having students read and write in different genres, attend to and understand text structures and literary techniques in different genres, and work directly with different text structures through written composition and revision, writing instruction in America’s Choice does not merely help students with writing mechanics. Rather, literacy practices such as these are also intended to improve students’ comprehension abilities. In this sense, instruction in America’s Choice treats the reading and writing processes as integrated, as writing instruction is intended to help students better comprehend text.

30

E.M. Camburn and S.W. Han

The instructional activities advocated by America’s Choice can also be viewed as a vehicle for integrating writing and reading. Like the study of genre and literary techniques, activities where students read, listen to and respond in writing to different texts support the development of students’ comprehension abilities, while at the same time modeling quality writing. The children’s literature which is used for both writing and reading instruction also serves as an integrating force, as students were encouraged to recognize distinctive text structures in this literature and in their own writing, and to try to incorporate these structures into their writing. Considerable evidence supports the efficacy of instructional strategies making up the America’s Choice program’s approach to literacy. For example, evidence from NAEP demonstrates a positive association between instruction in the writing process and writing proficiency (Goldstein and Carr, 1996). Multiple studies have found positive effects of the National Writing Project, a program which advocates systematic teaching of the writing process (Fancsali and Silverstein, 2002; Hawkins and Marshall, 1981; Marshall, 1983; Pritchard and Marshall, 1994). Recent research also suggests the benefits of a focus on genre as a means of understanding text structure. A report by the RAND Reading Study Group (RSG) determined that “knowledge of text structure is an important factor in fostering comprehension” (Snow, 2002). Receiving feedback on one’s writing, both from teachers and students, has also been found to benefit students’ writing (Beach and Friedrich, 2006). Finally, a meta-analysis of 14 control group studies examining the effect of “repeated and guided repeated oral reading” strategies on reading achievement revealed an overall effect size of .41 (NICHD, 2000).

Distributed Leadership in America’s Choice Perhaps one of the most important design elements in America’s Choice is the creation of two leadership roles – Design Coach and Literacy Coordinator. Design Coaches provide broad leadership and support for all aspects of America’s Choice program implementation. This includes working with school administrators to plan program activities in the school, oversight of the development of curricula and student assignments that reflect program goals, and organizing and supporting the analysis of student performance data by school faculty. In addition to these broader program responsibilities, Design Coaches also provide direct support to teachers by providing instruction in how to analyze student work. The role of Literacy Coordinator is more specialized than that of Design Coach. Literacy Coordinators focus their efforts exclusively on direct assistance to teachers in implementing the early grades literacy curriculum. As part of this role, these teacher leaders model instructional strategies associated with the America’s Choice program, observe teachers’ use of the strategies, and follow up these observations with feedback and support. In addition, Literacy Coordinators also work together with teachers to analyze student work. According to the framework of Camburn et al. (2003), teacher leadership roles in America’s Choice are “activated” in a number of ways. Expectations for the two roles are expressed to role incumbents through formal program documents. Design

2 Investigating Connections Between Distributed Leadership and Instructional Change

31

Coaches and Literacy Coordinators also receive intensive training in all aspects of their roles. A major focus of this training is for these teacher leaders to become experts in the instructional approaches associated with the America’s Choice design. This professional development also prepares these leaders to effectively guide the work of faculty development through direct work with teachers and through the organization, oversight, and facilitation of program implementation. America’s Choice, like most CSR designs, can be viewed as providing a blueprint for instructional change. The America’s Choice design directs teachers to become exposed to AC instructional practices by working with teacher leaders (Design Coaches and Literacy Coordinators). This exposure comes through professional development conducted by the teacher leaders and through the observation of teacher leaders as they model AC practices. Teachers and teacher leaders also engage in ongoing collaborative work, such as the analysis of student work and the development of curricular units, that is designed to further support teachers’ adoption of America’s Choice practices. In addition to these sorts of peer collaboration, America’s Choice materials, including the standards and curriculum materials, also serve to encourage the adoption of literacy practices advocated by the America’s Choice design.

Research Approach We used data from a large-scale study of elementary school reform to examine the association between the distribution of leadership to teacher leaders and teachers’ adoption of instructional practices associated with America’s Choice. Data come from the Study of Instructional Improvement (SII), a large-scale study of the design, implementation, and instructional effectiveness of three of America’s largest and most widely-disseminated CSR models – the Accelerated Schools Project (ASP), America’s Choice (AC), and Success for All (SFA). The SII was conducted in approximately 115 schools located primarily in large, urban school districts. Of these schools, approximately 90 participated in one of the three CSR models. The remainder of the school sample was comprised of 26 comparison schools which were demographically-similar to the CSR schools and located in the same school districts. Since this study tests hypotheses about America’s Choice and not the other two CSR programs, our analyses are limited to the 31 America’s Choice schools and 26 comparison schools in the sample. Data used for this study come from teacher surveys administered in the spring of the 2001–2002 and 2003–2004 school years. Survey data for 981 teachers from the 2001–2002 school year and 1019 teachers from the 2003–2004 school year were analyzed.

Outcome Measures To investigate the potential consequences of distributed leadership, we analyzed three outcome measures. A central variable for this study is evidence of the extent to which teacher leaders served as resources for the adoption of new teaching practices. The theory of action implied by the America’s Choice design stipulates that

32

E.M. Camburn and S.W. Han

teacher leaders would provide just such a resource. But as Cohen et al. (2003) argue, resources such as these are inert, and do not assert an impact on school operations until they are actively used. In this case, the distribution of leadership responsibility to teacher leaders in America’s Choice schools will not actively support teacher development until classroom teachers tap into the new teacher leaders as resources. The America’s Choice design stipulates that teachers tap into this resource through regular interactions with Literacy Coordinators and Design Coaches. These interactions are intended to be squarely focused on the specifics of classroom teachers’ practice. With this logic of resource use in mind, we created a measure indicating the degree to which teachers worked with instructional leaders (e.g. coaches and facilitators) on their instruction. Teachers’ responses to four survey items were combined to form a scale. The items asked teachers how often they watched an instructional leader model instruction, how often an instructional leader observed them and provided diagnostic feedback, how often an instructional leader gave them feedback on their use of curriculum materials, and how often their work with instructional leaders focused on their students’ work. The resulting scale thus allows us to gauge the extent to which teachers in the sample were able to tap teacher leaders as resources for instructional change. The second outcome we examined is a measure of the extent to which teachers perceive clarity in expectations for instruction in their school. We hypothesize that one of the ways teacher leaders can support teachers’ adoption of new practices is to help them clarify what kinds of instructional strategies are valued, useful and appropriate in their school. As discussed earlier, research has identified a lack of clarity in instructional expectations as a persistent barrier to instructional change and the New American Schools initiative held that CSR designs should specifically and clearly communicate expectations. Our focus on the clarity of expectations for instruction also has a basis in recent learning theories which have come to be called a situative perspective. Situative theorists contend that learning is a social process that occurs in particular socio-cultural settings. Applied to teaching, this perspective argues that learning new instructional strategies is likely to be ineffective if new strategies are encountered in the abstract, divorced from the realities of one’s school and divorced from one’s classroom practice. In contrast, the situative perspective contends that lasting, meaningful learning for teachers (and subsequently instructional change) is more likely when learning experiences reflect the realities of teaching in one’s school (Putnam and Borko, 2000). Situative theorists argue that becoming attuned to constraints and affordances in one’s environment is a fundamental aspect of the learning process (Greeno and MMAP, 1998). Applied to the present study, this perspective suggests to us that teachers who perceive greater clarity in expectations for instruction may be better attuned to constraints and affordances for teaching in their school. Instructional leaders may thus support teachers’ adoption of new strategies by helping them become more attuned to these kinds of constraints and affordances. In order to investigate whether teachers’ work with instructional leaders was associated with the adoption of new teaching practices, we developed a third outcome

2 Investigating Connections Between Distributed Leadership and Instructional Change

33

Table 2.1 Survey items included in a measure of instructional practices advocated in the America’s Choice design Item description

AC practice measured

Teacher uses ability grouping

Reader’s Workshop – guided reading conducted with small groups with similar learning needs Reader’s and Writer’s Workshop – conferencing Writer’s workshop – students analyze author’s craft Writer’s workshop – students analyze author’s craft Writer’s workshop – students analyze author’s craft Reader’s Workshop – guided reading

Teacher individualizes instruction Analyzing or evaluating text Examining literary techniques Identifying author’s purpose Did think-aloud or explained strategy application Editing capitalization, punctuation, etc. Editing grammar, word use, etc. Revising writing through elaboration Revising writing through reorganization Writing with letter strings or words Writing individual sentences Writing connected sentences Writing two or more paragraphs

Writer’s workshop – explicit instruction in writing Writer’s workshop – explicit instruction in writing Writer’s workshop – explicit instruction in writing Writer’s workshop – explicit instruction in writing Writer’s workshop – explicit instruction in writing Writer’s workshop – explicit instruction in writing Writer’s workshop – explicit instruction in writing Writer’s workshop – explicit instruction in writing

measure of teachers’ use of literacy strategies called for by the America’s Choice program. The first step in constructing this measure was to identify instructional practices advocated by the America’s Choice design. This was accomplished by reviewing America’s Choice program documents and curriculum materials. We then identified items in the teacher survey that measured teachers’ use of strategies advocated by America’s Choice. The results of this process are summarized in Table 2.1. Recall that America’s Choice approach to literacy instruction was largely built around two general strategies – Reader’s Workshop and Writer’s Workshop. As Table 2.1 shows, our measure of instructional practice contains three items that are thought to indicate teachers’ use of practices consistent with Reader’s Workshop – the use of ability grouping (AC teachers were encouraged to group students of similar ability for guided reading), individualization of instruction, and the use of think-alouds (the latter two activities are indicative of conferencing advocated by AC). The remaining eleven items making up this measure are thought to indicate teachers’ use of practices that are consistent with Writer’s Workshop. Among the distinctive features of Writer’s Workshop tapped by these items are: the examination of literary techniques, the analysis and evaluation of text, and identifying the author’s purpose. As discussed earlier, America’s Choice advances these strategies as an effective means for students to work with and understand different text structures. Another distinctive feature of literacy instruction advocated by America’s choice that may also be captured in this measure is the integration of reading and writing as teachers will be higher on this scale to the extent that they emphasize both the reading and writing strategies measured with this set of survey items. We

34

E.M. Camburn and S.W. Han

note however, our ability to measure integration is limited because the items in this scale do not explicitly measure whether teachers are in fact are integrating the practices. All three outcome measures were scaled using a Rasch item response theory (IRT) model (Wright and Masters, 1982). This model allowed us to place teachers on comparable scales in 2001–2002 and 2003–2004, thus permitting us to gauge change in these three outcomes over time. All three measures had acceptable levels of reliability of .70 or greater.

Statistical Models Multilevel models were used to test propositions about the relationship between the distribution of leadership to teacher leaders and instructional change among schools implementing the America’s Choice program. In this section we describe the analytic models that were used and the process we undertook in choosing the models. The greatest analytical challenge we faced was to develop a valid way of assessing instructional change. Our first strategy in this regard was just discussed. To validly assess whether teachers in schools that implemented the America’s Choice program adopted practices advocated by the program, we developed an outcome measure that we believe reflected the instructional goals of the program. We also gain leverage on this issue from the quasi-experimental design of the Study of Instructional Improvement. As previously mentioned, the research design contains a group of comparison schools that are similar to the schools implementing CSR programs but that were not themselves implementing any of the three CSR programs. Since schools were not randomly assigned to the CSR programs we need to be concerned whether any differences we observe between America’s Choice and comparison schools existed prior to schools’ adoption of America’s Choice. The presence of any pre-existing differences between the two groups of schools would cast doubt on any observed “post-treatment” differences between the two groups. Propensity score stratification was used to statistically equate America’s Choice and comparison schools on 34 characteristics measured prior to schools’ adoption of America’s Choice. This technique essentially controls for any pre-existing differences between the two groups of schools. With these controls in place, we can be relatively confident that any observed differences in distributed leadership and instructional practice between comparison schools and America’s Choice schools did not exist prior to schools’ adoption of America’s Choice. By equating schools on “pre-treatment” differences, we thus feel reasonably comfortable concluding that any observed cross-sectional differences between America’s Choice and comparison schools are associated with program implementation. In 2001–2002, all of the America’s Choice schools had been implementing the program for at least two years. Given this timing and the propensity stratification equating, if we find that teachers in America’s Choice schools are more likely to use practices advocated by the program in 2001–2002, we believe it is reasonable to

2 Investigating Connections Between Distributed Leadership and Instructional Change

35

interpret this as a program effect. We further believe this interpretation would be equally reasonable for similar results obtained in later years. Since the SII research design is longitudinal, there is the possibility of going a step further and assessing whether individual teachers’ practices are changing over time. One possible approach to the problem is to use a statistical model to predict linear “growth” in teachers’ use of instructional practices advocated by America’s Choice. This strategy is commonly used to model change in individual student achievement over time, and the results are generally straightforward to interpret. At the outset of this study we had some doubts about whether a linear growth model would be a valid way to represent changes in teachers’ practice over time. Case study accounts of teacher change suggest that for some, change may not unfold in a linear fashion, but rather, move slowly in fits and starts (see Wasley, 1994 for example). If this is true for large numbers of teachers, operationalizing change as linear growth over time might not provide an optimal description of the phenomenon. With these cautions in mind we conducted exploratory growth analyses with a three level hierarchical linear model which nested teachers’ instruction scores for each of three years (level 1), within teachers (level 2), within schools (level 3) (see Raudenbush and Bryk, 2002 for technical details regarding growth models). The models indicated that on average, individual teachers’ rate of growth between 2001–2002 and 2003–2004 on the instructional practice outcome was essentially zero. Certainly some teachers’ use of practices advocated by America’s Choice increased, and some teachers use of these practices decreased during this period, but on average, most teachers stayed the same. In light of these results, we determined that models of individual teacher growth did not provide a useful representation of observed patterns in the data. We thus shifted our focus to an investigation of whether results consistent with an overall effect of America’s Choice on distributed leadership and instruction could be observed over time. We did this by examining schools at two time points – early in their adoption of the program, and later in the process, after schools had worked with the program for a number of years. In our view, examining results at these two time points provides an indication of whether any observed program effects persisted over time. We conducted these analyses using multilevel statistical models that provide more accurate estimates for data with a nested structure (Raudenbush and Bryk, 2002). For this study, the sample of teachers we studied are nested within the schools in which they work. Besides providing more accurate estimates for nested data, the multilevel models used for this study allowed us to simultaneously examine the effects of teacher and school level factors on outcome measures. For example, the models allow us to investigate, within a single analysis, how teachers’ background (e.g. the number of years they have taught) is related to their instructional practice, and how characteristics of teachers’ schools are related to school-level patterns in instructional practice. In total, six models were estimated – one apiece for each of three outcomes and for each of two years. Variables in the models were scaled so that estimates can be interpreted as standard effect sizes. With this kind of scaling the results indicate

36

E.M. Camburn and S.W. Han

the magnitude of relationships between independent variables and outcomes in a standard fashion thus permitting direct comparison of the effect sizes of different independent variables. In addition to controlling for pre-existing differences between America’s Choice and comparison schools as discussed earlier, models also controlled for differences in teacher background with respect to their gender, race, years teaching experience, content knowledge for teaching reading, and average achievement level of their students.

Results Our first goal was to establish whether teachers in America’s Choice schools were more likely than those in comparison schools to access teacher leaders in their schools as resources for instructional improvement. As previously mentioned, Camburn et al. (2003), who used data from the same study, established that schools implementing the America’s Choice design were indeed more likely than comparison schools to distribute leadership responsibilities for teacher development to teacher leaders. In this first analysis, we attempted to go a step further and establish whether teachers in America’s Choice schools were more likely to access the leadership resources that were distributed. To assess whether America’s Choice teachers were more likely to access leadership resources, we fit models predicting the outcome measure indicating the extent to which teachers worked with instructional leaders in their schools. Key results from this analysis are displayed in Fig. 2.1. We found that in 2001–2002, teachers in America’s Choice schools were about three times more likely than their counterparts in comparison schools to work with instructional leaders in their schools. Thus, not only were greater leadership resources allocated to teacher development in America’s Choice schools, but teachers were clearly taking advantage of these resources with greater frequency. It is important to note that this observed difference between America’s Choice teachers and comparison school teachers is up and above any differences between teachers associated with their teaching experience and knowledge. It is also important to note that in 2001–2002, half the America’s Choice schools in the sample were in their second year of program implementation, and half had been implementing the program for 3 or 4 years. Considering this fact, and that we have controlled for pre-existing differences between America’s Choice and comparison schools, we are relatively confident that this result is reflective of schools’ implementation of the America’s Choice program. We fit the same model using data from the 2003–2004 school year, and key results from that model are also displayed in Fig. 2.1. The results indicate that teachers in America’s Choice schools were slightly more likely to work with instructional leaders than teachers in comparison schools in 2003–2004, although the difference between these two groups was not statistically significant. The results presented in Fig. 2.1 do not indicate that teachers’ work with instructional leaders in America’s Choice schools decreased over time. Indeed, an examination of descriptive statistics

2 Investigating Connections Between Distributed Leadership and Instructional Change Fig. 2.1 The frequency of teachers’ work with instructional leaders in America’s Choice & comparison schools

37

0.4

Effect size

0.3

0.2

0.1

0 2001–02

2003–04

–0.1 America’s Choice

Comparison Schools

for this outcome measure indicates that teachers’ work with instructional leaders in America’s Choice schools was stable over time, while in comparison schools, teachers’ work with instructional leaders increased during this period. While the main purpose for including teacher characteristics such as race and years experience in the models was to control for differences in teacher background, the results for these variables are also of substantive interest. The results presented in Fig. 2.2 paint a portrait of the kinds of teachers that are more likely to work with 0.4 0.3

Effect size

0.2 0.1 0 –0.1 –0.2

Fig. 2.2 The frequency of teachers’ work with instructional leaders by teacher characteristics

2001–02

African American Content knowledge

2003–04 Years teaching experience

38

E.M. Camburn and S.W. Han

instructional leaders. We found that African American teachers spent significantly more time working with instructional leaders than teachers of other races both in 2001–2002 and 2003–2004 school years. Teachers with fewer years of experience spent more time working with instructional leaders in 2001–2002 school year, but the estimate for teacher experience was not statistically significant during the 2003– 2004 school year. The models also indicated that teachers with less content knowledge for teaching literacy spent more time working with instructional leaders. These latter two results perhaps indicate a strategic allocation of leadership resources in schools, with less experienced and less knowledgeable teachers receiving greater attention from instructional leaders. We next attempted to understand whether teachers in America’s Choice schools were more likely to use literacy instruction strategies advocated by the America’s Choice program than teachers in comparison schools. Figure 2.3 shows how teachers in these two groups differed on the instructional practice outcome in 2001–2002 and 2003–2004. The figure clearly illustrates that teachers in America’s Choice schools were much more likely than those in comparison schools to use the kinds of teaching practices called for in the America’s Choice design which included explicit instruction in writing, teacher conferencing, attention to literary techniques, and strategies associated with guided reading. The significant difference between America’s Choice and comparison school teachers on this outcome was observed in 2001–2002, when schools were fairly early in their implementation of America’s Choice, and in 2003–2004, when all America’s Choice schools had been working with the program for at least four years. Again, recalling the fact that these models control for pre-existing differences between America’s Choice and comparison schools, it seems reasonable to us to interpret these results as program effects on instructional practice. The fact that the instructional practices of teachers in America’s Choice schools differed significantly from those of teachers in comparison schools suggests to us that the program’s impact on instruction persisted over time. Indeed, this result is consistent with other research based on data from the Study of Instructional Improvement that has demonstrated strong program effects on instructional practices in schools implementing the America’s Choice design (Correnti & Rowan, 2007; Rowan et al., 2009). 0.4

Effect size

0.3 0.2 0.1 0

Fig. 2.3 Teachers’ use of literacy instruction practices advocated by America’s choice in America’s Choice and comparison schools

–0.1

2001–02 America's Choice

2003–04 Comparison Schools

2 Investigating Connections Between Distributed Leadership and Instructional Change

39

Beyond testing whether teachers in America’s Choice schools were more likely than teachers in comparison schools to use literacy instruction practices consistent with the America’s Choice design, this model also allowed us to examine the characteristics of teachers who used these practices more frequently. We were particularly interested in understanding whether teachers who worked more frequently with instructional leaders were more likely to use these kinds of instructional strategies. Since this analysis includes teachers in schools implementing America’s Choice and teachers in comparison schools, the model tells us whether teachers who work more often with instructional leaders are generally more likely to teach in a manner consistent with the America’s Choice program. Figure 2.4 displays results for teacher-level variables that were significantly associated with the instruction outcome measure. We found that African American teachers were much more likely to engage in practices such as the explicit teaching of the writing process, and having students examine literary techniques than teachers of other races. The models also predicted that teachers with greater teaching experience, and those with lower levels of content knowledge for teaching literacy were more likely to use these kinds of practices. Teachers’ work with instructional leaders was one of the strongest predictors of the instructional outcome measure, with standardized effect sizes for this variable exceeding .20 in 2001–2002 and 2003–2004.

0.4 0.3

Effect size

0.2 0.1 0 –0.1 –0.2

2001–02

2003–04

African American

Years teaching experience

Content knowledge

Work with inst. leaders

Fig. 2.4 Teachers’ use of literacy instruction practices advocated by America’s Choice by teacher characteristics

40

E.M. Camburn and S.W. Han

This preliminary evidence of the link between teachers’ interactions with instructional leaders and changed teaching practices is suggestive but far from conclusive. Because this analysis includes both America’s Choice and comparison school teachers, we cannot be sure whether this relationship holds in schools implementing America’s Choice. Additional analyses presented below probe this question in greater depth. Having established that teachers in schools that implemented America’s Choice were more likely to access instructional leaders as resources for instructional improvement, and that they were also more likely to use literacy instruction practices advocated by the program, we next sought to better understand whether there is a connection between these two results. In other words, we wondered whether there was evidence suggesting that America’s Choice teachers’ work with instructional leaders might have facilitated their use of literacy instruction practices advocated by the program. To address this question we conducted two additional analyses that were limited to the schools that implemented America’s Choice and the teachers in those schools. Limiting the analysis to these schools was intended to illuminate the internal dynamics of instructional change within schools that had adopted the America’s Choice design. As previously discussed, an important responsibility of Design Coaches and Literacy Coordinators is to provide guidance to teachers that clarifies the instructional expectations of the America’s Choice program. In a broader sense, situative learning theories posit that opportunities to learn new practices that are embedded in teachers’ day-to-day work may provide a similar kind of clarification. Specifically, these theories hold that when teachers’ learning opportunities are embedded in or at least connected to their daily work in classrooms, teachers are more likely to become aware of what constitutes effective, practical, or appropriate instruction within their schools. Teachers’ interactions with Design Coaches and Literacy Coordinators in America’s Choice are indeed embedded in and connected to teachers’ daily practice, and thus, we hypothesize that teachers’ work with these leaders will help them become more aware of what literacy instruction practices are effective and appropriate within their school. To test these ideas we fit a multilevel model predicting the outcome measuring the degree to which teachers perceived expectations for instruction in their schools as being clear. Our main goal with this model was to test the association between the clarity of instructional expectations and teachers’ work with instructional leaders. Key results from this model are displayed in Fig. 2.5. Within America’s Choice schools we found a strong association between teachers’ work with instructional leaders and having greater clarity about instructional expectations. Indeed, as Fig. 2.5 illustrates, working with instructional leaders was the strongest predictor variable in the model. This result remained consistent for both years examined, suggesting that Design Coaches and Literacy Coordinators may have continued to play a role in clarifying expectations for instruction even 4–6 years after schools began to implement the America’s Choice program. We also found that more experienced teachers were more likely to have greater clarity of instructional expectations than their less experienced counterparts, a result which struck us as intuitive. The model

2 Investigating Connections Between Distributed Leadership and Instructional Change

41

0.4 0.3

Effect size

0.2 0.1 0 –0.1 –0.2 –0.3

2001–02

Years of teaching experience

2003–04 Classroom achievement below natl. avg.

Work with instructional leaders

Fig. 2.5 Clarity of expectations for instruction within America’s Choice schools

also predicted that teachers whose students achievement was below the national average were significantly less likely to report clarity of instructional expectations. Like the result for teachers’ work with instructional leaders, the results for teachers’ years of experience and student achievement levels remained consistent over time. Finding a positive association between teachers’ work with instructional leaders and their perception of greater clarity in instructional expectations in America’s Choice schools adds another piece to the puzzle. The result suggests that instructional leaders in America’s Choice schools might facilitate teachers’ adoption of new practices by helping teachers better understand what constitutes effective and appropriate practice in their school. But at this stage, we have not yet established whether working with instructional leaders is positively associated with the use of literacy instruction practices advocated by America’s Choice in America’s Choice schools. In a final analysis we attempt to further complete the puzzle by subjecting this conjecture to an empirical test. Key results from the final model are presented in Fig. 2.6. We found solid evidence supporting our hypothesis of a positive relationship between teachers’ work with instructional leaders in America’s Choice schools and their use of literacy instruction practices advocated by the program. This significant result held across both years studied suggesting that Design Coaches and Literacy Coordinators continued to support teachers’ adoption of literacy practices consistent with the America’s Choice design even 4–6 years after the program was first implemented in schools. Evidence for the hypothesis that having greater clarity about instructional expectations was positively associated with the use of literacy instruction practices

42

E.M. Camburn and S.W. Han 0.4

Effect size

0.3 0.2 0.1 0 –0.1 –0.2

2001–02

2003–04

African American

Content knowledge

Work with inst. Leaders

Clarity of inst. expectations

Fig. 2.6 Use of literacy instruction practices advocated by America’s Choice within America’s Choice schools

advocated by America’s Choice schools was more equivocal. In 2001–2002, clarity of instructional expectations was positively associated with the instruction outcome measure, but the result was not statistically significant. In 2003–2004, the positive relationship between these two variables was statistically significant. The model also predicted that African American teachers were more likely than teachers of other races to use literacy instruction practices associated with America’s Choice, and that teachers with greater content knowledge for teaching literacy were less likely to use instructional practices advocated by America’s Choice. The latter result appears consistent with the earlier finding that teachers with lower levels of content knowledge tended to spend more time working with instructional leaders. It is possible that within America’s schools, such teachers were more likely to use literacy practices advocated by the America’s Choice program because they received greater attention from instructional leaders. We caution however, that this is merely a conjecture and our data do not support strong causal claims regarding this result.

Conclusions This study adds to the literature on distributed leadership by going beyond the conceptual and empirical descriptions of what distributed leadership is and what forms it takes that have thus far dominated the literature. When this study is considered alongside earlier research on distributed leadership and CSR programs, we believe a plausible case can be made that distributing leadership to teachers can support instructional change. Earlier research found substantial evidence that three CSR programs, including America’s Choice, distributed leadership responsibilities for

2 Investigating Connections Between Distributed Leadership and Instructional Change

43

faculty development to teacher leaders (Camburn et al., 2003). In this study, we documented that teachers in America’s Choice are more likely than those in comparison schools to access instructional leaders as resources for their development. Like other studies conducted as part of the Study of Instructional Improvement (Correnti and Rowan, 2007), we found evidence suggesting fairly widespread adoption of literacy instruction practices advocated by America’s Choice among teachers in schools that implemented America’s Choice, and this adoption appears to have persisted over a substantial period of time. We further found that teachers in America’s Choice schools were more likely to use these practices if they spent more time working with instructional leaders, a result that is consistent with the strategy of America’s Choice to support instructional change by distributing responsibility for teacher development to teacher leaders. Finally, we found mixed evidence consistent with the idea that teacher leaders may have facilitated teachers’ adoption of America’s Choice literacy practices in part by clarifying instructional expectations for teachers. There are at least two limitations of this study worth mentioning. One limitation is that our results are based on a very specific case, namely, the distribution of leadership to teachers within the context of a single comprehensive school reform program. There are likely many other forms of distributed leadership in other settings that might also support instructional change. Our relatively narrow focus likely limits the generalizability of these results. A second limitation of this study lies with our use of annual surveys to measure instruction and key independent variables. Camburn and Han (2005) have found measures of instruction from annual surveys to be systematically biased. Their research indicates that elementary school teachers may overstate the frequency with which they use literacy instruction practices by as much as 7 days per month. A second potential source of bias is what Shadish et al. (2002) call monomethod bias. This bias reflects the tendency for variables that come from the same instrument to be correlated with each other to some degree because those who provide the data (in this case teachers) tend to provide internally-consistent responses. For this reason, inferences based on multiple data sources are believed to be on a stronger footing than inferences based on a single data source. While we believe this study advances research in a constructive direction by beginning to examine the consequences of distributed leadership, much additional work is needed in this area. One of the most acute needs in our view is conceptual work about why and how distributed leadership might have consequences for important school outcomes. Without such conceptual development, empirical research in this area is likely to remain largely descriptive and is likely to unfold unsystematically. As other writers on the topic point out, there is also a strong need for empirical research on distributed leadership, especially research that examines the consequences of such leadership for schools. Descriptive case studies that attempt to draw connections between distributed leadership, intermediate outcomes, and student achievement could aid conceptual writing about these connections. Exploratory quantitative work might also help identify configurations of distributed leadership that hold promise for supporting teacher development and student performance. Making progress in these areas is important so that research can begin to inform

44

E.M. Camburn and S.W. Han

practitioners and policymakers about the potential benefits that might accrue when leadership is distributed.

References Beach, R., & Friedrich, T. (2006). Response to writing. In C. A. MacArthur & S. Graham & J. Fitzgerald (Eds.), Handbook of writing research (pp. 222–234). New York: Guilford Press. Bennett, B. B. (1987). The effectiveness of staff development training practices: A meta analysis. Unpublished Ph.D. dissertation, University of Oregon. Berends, M., Bodilly, S. J., & Kirby, S. N. (2002). Facing the challenges of whole-school reform: New American Schools after a decade. Santa Monica, CA: Rand Education. Borman, G. D., Slavin, R. E., Cheung, A. C. K., Chamberlain, A. M., Madden, N. A., & Chambers, B. (2007). Final reading outcomes of the national randomized field trial of success for all. American Educational Research Journal, 44(3), 701–731. Camburn, E., Rowan, B., & Taylor, J. E. (2003). Distributed leadership in schools: The case of elementary schools adopting comprehensive school reform models. Educational Evaluation and Policy Analysis, 25(4), 347–373. Camburn, E. M., & Han, S. (2005). Validating measures of instruction based on annual surveys, Annual Meeting of the American Educational Research Association. San Francisco, CA. Cohen, D. K. (1990). A revolution in one classroom: The case of Mrs. Oublier. Educational Evaluation and Policy Analysis, 12(3), 311–329. Cohen, D. K., & Ball, D. L. (1999). Instruction, capacity, and improvement (CPRE Research Report Series RR-443). Philadelphia, PA: Consortium for Policy Research in Education. Cohen, D. K., Raudenbush, S. W., & Ball, D. L. (2003). Resources, instruction, and research. Educational Evaluation and Policy Analysis, 25(2), 119–142. Correnti, R., & Rowan, B. (2007). Opening up the black box: Literacy instruction in schools participating in three comprehensive school reform programs. American Educational Research Journal, 44(2), 298–338. Cuban, L. (1990). Reforming again, again, and again. Educational Researcher, 19, 3–13. Fancsali, C., & Silverstein, S. (2002). National writing project final evaluation report. New York: Academy for Educational Development. Firestone, W. A. (1989). Using reform: Conceptualizing district initiative. Educational Evaluation and Policy Analysis, 11(2), 151–164. Fullan, M. (1991). The new meaning of educational change (2nd ed.). New York, NY: Teachers College Press Teachers College Columbia University. Goldstein, A. A., & Carr, P. (1996). NAEPFacts: Can students benefit from process writing? (Issue Brief NCES 96845). Washington, DC: National Center for Education Statistics. Greeno, J. G., & MMAP (1998) The situativity of knowing, learning, and research. American Psychologist. 53(1), pp. 5–26. Hart, A. W. (1995). Reconceiving school leadership: Emergent views. Elementary School Journal, 96(1), 9–28. Hawkins, M. L., & Marshall, J. C. (1981). Evaluating a writing program, grades 3–12 (ED290156). Ferguson, MO: Ferguson-Florissant School District. Leithwood, K. A., Mascall, B., & Strauss, T. (2008). What we have learned and where we go from here. In K. A. Leithwood & B. Mascall & T. Strauss (Eds.), Distributed leadership according to the evidence (pp. 269–281). New York, NY: Routledge. Lieberman, A. (1988). Building a professional culture in schools. New York: Teachers College Press. Linton, R. (1936). The study of man: An introduction (Student’s ed.). New York, London,: D. Appleton-Century company incorporated.

2 Investigating Connections Between Distributed Leadership and Instructional Change

45

Little, J. W. (1990). The persistence of privacy: Autonomy and initiative in teachers professional relations. Teachers College Record, 91(4), 509–536. Little, J. W. (1995). Contested ground: The basis of teacher leadership in 2 restructuring highschools. Elementary School Journal, 96(1), 47–63. Marshall, J. C. (1983). Student achievement in the holistic teaching of writing: Project impact and diffusion, Annual Meeting of the American Educational Research Association. Montreal, Canada. Mazmanian, D. A., & Sabatier, P. A. (1981). Effective policy implementation. Lexington, MA: Lexington Books. McLaughlin, M. W. (1987). Learning from experience: Lesson from policy implementation. Educational Evaluation and Policy Analysis, 9(2), 171–178. National Institute of Child Health and Human Development (NICHD). (2000). Report of the National Reading Panel: Teaching children to read: An evidence-based assessment of the scientific research literature on reading and its implications for reading instruction (NIH Publication No. 00-4754). Washington, DC: U.S. Government Printing Office. Newmann, F. M., Smith, B., Allensworth, E., & Bryk, A. S. (2001). Instructional program coherence: What it is and why it should guide school improvement policy. Educational Evaluation and Policy Analysis, 23(4), 297–321. Pritchard, R. J., & Marshall, J. C. (1994). Evaluation of a tiered model for staff-development in writing. Research in the Teaching of English, 28(3), 259–285. Putnam, R. T., & Borko, H. (2000). What do new views of knowledge and thinking have to say about research on teacher learning? Educational Researcher, 29(1), 4–15. Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods (2nd ed.). Thousand Oaks: Sage Publications. Raver, C. C., Jones, S. M., Li-Grining, C. P., Metzger, M., Champion, K. M., & Sardine, L. (2008). Improving preschool classroom processes: Preliminary findings from a randomized trial implemented in Head Start settings. Early Childhood Research Quarterly, 23(1), 10–26. Rowan, B., Camburn, E., & Barnes, C. (2004). Benefiting from comprehensive school reform: A review of research on CSR implementation. In C. T. Cross (Ed.), Putting the pieces together: Lessons from comprehensive school reform research (pp. 1–52). Washington, D.C.: National Clearinghouse for Comprehensive School Reform. Rowan, B., Correnti, R., Miller, R., & Camburn, E. (in press). School improvement by design: Lessons from a study of comprehensive school reform programs. In G. Sykes & B. Schneider (Eds.), Handbook of education policy research. Washington, DC: American Educational Research Association. Sarason, S. B. (1990). The predictable failure of educational reform: Can we change course before it’s too late? (1st ed.). San Francisco: Jossey-Bass. Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-experimental designs for generalized causal inference. Boston: Houghton Mifflin. Smylie, M. A., & Denny, J. W. (1990). Teacher leadership: Tensions and ambiguities in organizational perspective. Educational Administration Quarterly, 26(3), 235–259. Smylie, M. A., Conley, S., & Marks, H. M. (2002). Exploring new approaches to teacher leadership for school improvement. In J. Murphy (Ed.), The educational leadership challenge: Redefining leadership for the 21st century. 101st yearbook of the National Society for the Study of Education, Part1 (pp. 162–188). Chicago, IL: University of Chicago Press. Snow, C. E. (2002). Reading for understanding: Toward an R& D program in reading comprehension. Santa Monica, CA: Rand. Spillane, J. P., Reiser, B. J., & Reimer, T. (2002). Policy implementation and cognition: Reframing and refocusing implementation research. Review of Educational Research, 72(3), 387–431. Wasley, P.A. (1994). Stirring the chalkdust: Tales of teachers changing classroom practice. New York: Teachers College Press. Wright, B. D., & Masters, G. (1982). Rating Scale Analysis. Chicago: MESA Press.

Chapter 3

Taking a Distributed Perspective in Studying School Leadership and Management: The Challenge of Study Operations James P. Spillane, Eric M. Camburn, James Pustejovsky, Amber Stitziel Pareja and Geoff Lewis

Introduction Recent work suggests that viewing school leadership from a distributed perspective has the potential to provide useful insight into how management and leadership unfold in the daily lives of schools. Writing in the area of distributed leadership has identified numerous entities in the school across which leadership can be distributed, including people and aspects of the situation such as routines and tools (Harris, 2005; MacBeath et al., 2004; Spillane, 2006). While there have been advances in articulating conceptual frameworks for taking a distributed perspective on school leadership and management (Gronn, 2000; Spillane et al., 2004), the empirical research base in this area is less developed. With a few exceptions (Camburn et al., 2003; Leithwood et al., 2007; Spillane et al., 2007), most empirical work has involved small samples of schools. But as we argue in this chapter, before researchers begin to accumulate evidence on distributed leadership in schools, an important intermediate step needs to be taken: the operationalization of concepts, or in other words, the translation of theory into measurement. It is this intermediate step that is the primary focus of this paper. In this chapter, we examine the entailments of the distributed perspective for collecting and analyzing data. We then go a step further and examine the results obtained for different operationalizations of a distributed perspective, considering along the way the strengths and weaknesses of each operationalization. Within the larger domain of distributed leadership, we are concerned with the epistemological and methodological challenges involved in studying the distribution of leadership across people within the school – the leader-plus aspect of a distributed perspective (Spillane, 2006). Researchers who wish to study the leader-plus aspect of distributed leadership face two basic questions:

r

What aspects of leadership and management work are hypothesized to be distributed across people?

J.P. Spillane (B) School of Education and Social Policy Faculty Associate, Institute for Policy Research Northwestern University, Evanston, IL, USA

A. Harris (ed.), Distributed Leadership, Studies in Educational Leadership 7, C Springer Science+Business Media B.V. 2009 DOI 10.1007/978-1-4020-9737-9 4, 

47

48

r

J.P. Spillane et al.

Across which school actors do researchers hypothesize leadership and management work to be distributed? These questions surface a number of related methodological questions:

r r r

Who should provide evidence of distributed leadership – leaders, followers, or both? Among leaders, should researchers seek evidence on formal leaders, informal leaders, or both? Should evidence on distributed leadership come from self-reports, or from more “impartial” measurement strategies such as the reports of others (e.g., teachers) through network surveys? What tradeoffs do researchers make with each of these decisions? In other words, how do these various operationalizations of the leader plus aspect of a distributed framework affect the validity of data?

Ultimately, researchers’ answers to these questions will determine the kinds of inferences they will be able to make about distributed leadership. Our chapter is organized as follows. We first consider what it means to take a distributed perspective and briefly review the empirical evidence on school leadership and management from this perspective. We then provide a detailed account of the data used in this paper, addressing issues of validity where possible. Turning to results, we examine four operationalizations of the leader plus aspect of a distributed perspective by looking at data for two elementary schools. Based on this analysis, we identify two dimensions along which to examine the issues involved in these four operations of the leader plus aspect. We then consider how well our four operationalizations tap into variation between schools and between activity-types in the distribution of responsibility for leadership and management work. Our central argument is this: one challenge in using a distributed framework to study leadership and management involves operationalizing core aspects of the framework. In taking a distributed perspective we have to develop study operations that allow us to validly describe and examine constructs in our analytical framework. Study operations are important because they influence the validity of the inferences we can make based on the data we gather. For example, our operationalization of an aspect of a distributed framework (e.g., co-performance) can fail to adequately explicate that aspect, or can confound two or more aspects, and thereby pose a threat to construct validity. Further, relying on a single operationalization (i.e., monooperational bias) or relying on a single method to gather data on operationalizations of a construct (i.e., mono-method bias) can similarly threaten construct validity and call into question any possible inferences we can draw from our data.

Conceptual and Empirical Anchors A distributed perspective is an analytical framework for investigating school leadership and management (Spillane, 2006; Gronn, 2003; Spillane et al., 2003). It involves two aspects: the leader-plus aspect and the practice aspect (Spillane, 2006;

3 The Challenge of Study Operations

49

Spillane and Diamond, 2007). The leader-plus aspect recognizes that leading and managing schools can involve multiple individuals. Moreover, school leadership and management potentially involves more than the work of individuals in formal leadership positions such as principals, assistant principals, and specialists; it can also involve individuals who are not formally designated leaders. The leadership practice aspect foregrounds the practice of leading and managing. A practice or “action perspective sees the reality of management as a matter of actions” (Eccles and Nohria, 1992, p. 13). Defining leadership and management as practice allows for the possibility that people without any formal leadership designations might take a part in that work (Heifetz, 1994). While people’s actions are important in studying practice, interactions are paramount in efforts to understand the practice from a distributed perspective. From a distributed perspective, studying the actions of individuals or aggregating their actions is insufficient; a distributed perspective frames practice as a product of the interactions of school leaders, followers, and aspects of their situation. We focus chiefly on the leader-plus aspect in this paper, exploring different operations of this aspect in research. Prior empirical work suggests that an exclusive focus on the school principal is indeed short-sighted (Harris, 2005; MacBeath et al., 2004; Leithwood et al., 2007; Spillane and Diamond, 2007). Defining leadership as a set of organizational functions rather than tying leadership to a particular administrative position, Heller and Firestone (1995) found in a study of eight elementary schools that multiple leaders, including school district personnel and external consultants, were taking responsibility for leadership. A recent study of more than one hundred U.S. elementary schools also found that responsibility for leadership functions was typically distributed across three to seven formally designated leadership positions per elementary school (Camburn et al., 2003). Camburn and colleagues surveyed formally designated leaders in each school to examine the distribution of responsibility for leadership functions. Such positions included principals, assistant principals, program coordinators or facilitators, subject area coordinators or facilitators, mentors, master teachers, or teacher consultants, and other “auxiliary” professional staff, such as family outreach workers. Individuals with no formal leadership designations also take responsibility for leadership activities. Studies that look beyond those in formally designated leadership positions show that teachers also perform key leadership functions and routines (Heller and Firestone, 1995; Spillane, 2006; Spillane and Diamond, 2007; Spillane et al., 2007).

Methods The research reported here is part of a larger evaluation of a professional development program intended to prepare principals to be outstanding instructional leaders within the context of standards-based accountability systems. The primary objective of the evaluation was to assess the effects of participation in the program on school principals’ practice and knowledge. While a component of the principal

50

J.P. Spillane et al.

development program exposed principals to a distributed perspective on school leadership and the distributed perspective framed our evaluation study, it was not the primary focus of the development program or our research.

Data Collection and Instruments The study was undertaken in a mid-sized urban school district in the Southeastern United States. Data collection involved 52 principals and 2,400 school personnel. For the purpose of this paper, we analyzed data from three different research instruments: an experience sampling method (ESM) log, a principal questionnaire (PQ), and a school staff questionnaire (SSQ). The first dataset contained responses from principals that were collected using the experience sampling methodology (ESM). The ESM log captures behavior as it occurs within a natural setting, mitigating the problems associated with measuring behavior using retrospective self-reports (Hektner et al., 2006). To complete the ESM log, principals were beeped at random intervals throughout their work day, alerting them to fill out a brief questionnaire programmed on a handheld computer (PDA). The principals were beeped fifteen times a day for six days during Spring 2005. Forty-two of the fifty-two participating principals provided multiple days of data. For these principals, the overall response rate to the beeps across the six-day sampling period was 66%. The second source of data was a web questionnaire (PQ) that was administered to principals. For the purpose of this paper, we focused on the question in the PQ that asked principals about formal leadership teams at their schools. Forty-nine of the principals in the sample completed the PQ (94%). Finally, we analyzed data collected using a questionnaire that was mailed to staff members in all 52 schools. The overall response rate for the SSQ was 87%; schoollevel response rates ranged from 73 to 100%. In this paper we focus on SSQ survey questions that asked staff about formal leadership roles and responsibilities and on social network questions that asked staff to identify who they turned to for advice about reading and mathematics. School staff indicated on the SSQ the specific leadership roles they fulfill in the school as well as the percentage of their time that is assigned to each role. These data provide us with an estimate of the number of formally designated leaders in each school along with an estimate of how much time they spend on management and leadership-specific responsibilities. On the SSQ, school staff also identified from whom they sought advice about mathematics and language arts.

Examining Multiple Operationalizations of the Leader-Plus Aspect The leader plus aspect of a distributed perspective can be operationalized in various ways, some of which also capture the leadership practice aspect of a distributed perspective. Rather than searching for the one best study operation of the leader

3 The Challenge of Study Operations

51

plus aspect, we, like others (see for example Campbell and Fiske, 1959 for perhaps the earliest discussion of the idea), contend that multiple operations are desirable in order to minimize threats to validity. And like others, we argue that understanding what different operations illuminate and capture is a critical step in efforts to generate more robust empirical knowledge about social constructs in general (Denzin, 1989, Camburn and Barnes, 2004), and about school leadership and management in particular. Below we use data from two of the 52 schools in our study to explore four different operationalizations of the leader-plus aspect. We look at how these operationalizations work and what insights about leadership and management can be gleaned from each of them. These four are not the only operationalizations possible for the leader plus aspect of a distributed perspective.

Four Operationalizations of the Distribution of Responsibility One of the most basic ways in which to understand how responsibilities are distributed across staff members is to examine an organization chart to see which staff members belong to which leadership and management committees or hold formally designated leadership/management positions. On the PQ, principals enumerated the membership of the school’s formal leadership team, thus providing a list that resembles a school’s organization chart for leadership and management. By examining how often different roles are included in school leadership teams (e.g., how many schools include the math coordinator on the leadership team), we gain a broad understanding of how often leadership and management is distributed to non-principal roles in general, and to particular roles in schools. Data from the SSQ allow us to identify formally designated leaders based on school staff self-reports, providing a second indicator of how responsibility for leadership and management is distributed. In particular, we examine the percentage of staff with formal leadership assignments, calculated as the total number of staff who indicated that they held a leadership/management role divided by the total number of staff that completed the SSQ. A third indicator of how responsibility for leadership and management is distributed in schools is obtained from the ESM log. This approach differs from both the PQ and the SSQ in that unit of study is exclusively the school principal’s workday. The ESM log actually provides us with two distinct indicators. First, school principals reported who had responsibility for the activities that they reported not leading. Second, when principals reported that they were co-leading, they identified the role of anyone who was co-performing with them. In this way, the ESM log enables us to tap into the leadership practice dimension of the distributed perspective by identifying instances of co-performance in the school principal’s workday. Principals are prompted by random beeps to respond to a survey that poses questions about their current activities. We can get an overall estimate of the percentage of time that principals spend leading alone and with co-leaders when we aggregate the reports (the beeps) across the six-day sampling period. In the analyses of ESM data below, reported percentages are based on the subset of beeps where

52

J.P. Spillane et al.

the respondent reported being engaged in a school-related activity. In our analyses, we calculate the mean percentages differently based on the area of interest. When comparing means (e.g., percentage of time leading alone for administration versus time leading alone for instruction and curriculum) we calculate percentages on all school-related beeps across principals and days. However, if we are interested in analyzing variance between days and/or principals, our percentage of time estimates are calculated for each principal and day and averaged across all principal/day combinations. The unit of study in an ESM log is the respondent’s practice. ESM log data therefore provide a relatively narrow picture of leadership and management practice in a school because only school principals completed it. Still, measuring his or her practice certainly captures an important part of all leadership and management practice in a school. However, having other formally designated leaders such as assistant principals and curriculum specialists complete the ESM log would provide a broader picture. A fourth view of the ways in which leadership is distributed across staff members in schools is obtained by examining the social networks through which teachers share advice about mathematics and language arts instruction. Our approach here differs from the other three approaches in two important respects. First, the social network approach is designed to focus on leadership (defined as a social influence relationships or interactions), whereas the other three approaches do not attempt to distinguish leadership from management activities. Second, in our study the social network questions focused on mathematics and language arts rather than leadership interactions writ large. Hence, by focusing on leadership (as distinct from leadership and management) for only two curricular domains these questions focused on a somewhat narrow slice of school leadership and management activity. We analyze data from social network questions that asked respondents to list those to whom they went to for advice about mathematics and language arts. Respondents are identified as being leaders in math and/or reading based on the reports of their peers, using a measure called in-degree centrality. In social network analysis, in-degree centrality is a measure of the number of ties directed to an actor from other actors. In an advice network, an actor’s in-degree indicates the number of people who approach that actor for advice. For purposes of identifying leaders, we make the assumption that any actor who provides advice to three or more others is a leader. Other researchers have proposed more complex methods of identifying leaders using social network data (Cole and Weiss, 2006), though how such calculations relate to leadership in a theoretical sense is rather unclear. For the illustrative purpose of this paper and without delving into various theoretical debates on leadership, we believe that our admittedly simplistic approach will suffice. By capturing the pattern of influence-relationships among staff, the socialnetwork approach also allows the analyst to move beyond the leader-plus perspective and focus on leadership practice. We demonstrate this approach by calculating the frequency with which influence relationships or interactions are focused

3 The Challenge of Study Operations

53

on formal leaders as a percentage of all influence relationships or interactions identified.

Contrasting Operations of the Leader Plus Aspect: Canton Elementary and Lowell Elementary We examine how leadership is distributed among school staff by looking systematically at what our different operationalizations turned up for two schools: Lowell Elementary and Canton Elementary. We selected these schools based on analysis of the ESM data, which suggested that the principals of these two schools differed in the degree to which they involved others in the work of leading and managing their schools. The two case study schools differ not only with respect to how responsibility for leadership and management is distributed, but also in terms of size and the student populations they served. Lowell Elementary has a staff of 72. Of the 885 students enrolled at Lowell, 28% receive free or reduced-price lunches and approximately one-third (31%) are African-American. In contrast, Canton Elementary’s staff of 37 serves only 415 students, half of whom receive free or reduced-price lunches and a little over half (56%) of whom are African-American. The two school principals have relatively similar characteristics and backgrounds. Ms. Kite has been an administrator for 13 years, 8 of which have been at Lowell Elementary. Ms. Bind has been an administrator for 14 years and principal at Canton Elementary for 11 years. Both middle-aged Caucasian women, Ms. Kite was a teacher for 12 years and Ms. Bind was a teacher for 18 years prior to entering school administration. Our various research instruments suggest that the work of leading and managing Lowell Elementary and Canton Elementary involves multiple people, in varying degrees, in addition to Ms. Kite and Ms. Bind. One of the basic challenges facing researchers taking a distributed perspective is to identify the staff among whom the responsibility for leading and managing is distributed – the leader plus aspect. In our study, we used four different operations to identify who these individuals are:

r r r r

Principals identified who was on the leadership team in the PQ instrument; Principals reported who lead an activity they participated in or with whom they co-lead when beeped on the ESM instrument; School staff identified themselves in the SSQ instrument as having a formally designated leadership position in the school; and School staff identified people who provided advice about mathematics and language arts in social network type questions (SSQ).

Overall, these four operationalizations suggest some convergence and divergence in identifying who takes responsibility for leadership and management work at Lowell Elementary and Canton Elementary (See Table 3.1). While full-time classroom teachers emerge as leaders in all four approaches at Lowell Elementary, they

54

J.P. Spillane et al.

Table 3.1 Nominated leaders across data sources for lowell elementary and canton elementary Assistant principal

Principal

Leadership team member (PQ) Identification as leader or co-Leader (ESM) Self-identification as designated leader (SSQ) Identified as math and/or reading leader (SSQ)

Formally designated leader

Classroom teacher

Lowell Canton

Lowell Canton

Lowell Canton

Lowell Canton

YES

YES

YES

YES

YES

YES

YES

YES

YES

YES

YES

NO

YES

YES

YES

YES

N/A

N/A

YES

YES

14

7

N/A

N/A

NO

NO

NO

YES

3

3

10

2

emerge in only three of the four approaches at Canton Elementary. The assistant principals emerge as players in three of the four approaches in both schools. Both Ms. Kite and Ms. Bind, the school principals, emerge in two of the three applicable approaches. At Lowell Elementary, the social network questions suggest that some ten regular classroom teachers play a leadership role. At Canton Elementary, two classroom teachers are identified as math and/or reading leaders according to the social network questions. We examine the four approaches below. Principal Report: “Leadership Team” membership (PQ Data). In the PQ instrument, both Ms. Kite and Ms. Bind identified who was on the leadership team at their respective schools. From Ms. Kite’s report in the PQ, we learn that the leadership team at Lowell Elementary consists of the principal, the assistant principal, a variety of specialists, regular classroom teachers, other staff members, and parents/community members. According to Ms. Bind, the leadership team at Canton Elementary consists of the principal, the assistant principal, a variety of specialists, various regular classroom teachers, and other staff members. These data suggest that more diverse players are involved in the work of leading and managing the school including parents/community members at Lowell Elementary compared to Canton Elementary. We acknowledge that an exclusive focus on the leadership team is limiting and the inclusion of other key organizational routines and committees such as the school improvement planning committee, Language Arts committee, and Mathematics committee may be desirable to provide a more nuanced account of how responsibility for leadership and management is distributed. Principal Report: Who is Leading or Co-Leading (ESM Data). We can get a sense of how responsibility for school leadership and management is distributed across staff at the two schools by exploring the extent to which the school principals report leading alone, co-leading with others, or not leading an activity when beeped at random during the workday. At Lowell Elementary, Ms. Kite has a tendency to share her leadership role with others, while at Canton Elementary, Ms. Bind

3 The Challenge of Study Operations

55

Table 3.2 Percentage of time principal was co-leading, not leading, or leading alone Principal participation

Lowell (n = 63)

Canton (n = 60)

Co-leading Not leading Leading alone

62 24 14

23 18 58

is inclined to lead alone. Ms. Kite, reported co-leading 62% of the time, a much larger proportion than Ms. Bind, who reported co-leading just 23% of the time (see Table 3.2). Further, Ms. Bind reported that she lead alone over half the time (58%), while Ms. Kite lead alone only 14% of the time. The two principals are somewhat similar in the amount of time they reported not leading, with Ms. Kite not leading 24% of the time and Ms. Bind not leading 18% of the time. Thus, through their reports of leading alone or co-leading the activities in which they participated, the two principals demonstrated divergent practice. We get a more detailed picture of how responsibility for school leadership and management is distributed across staff at the two schools by examining principals’ reports of who is leading an activity that they are not leading, and their reports of individuals with whom they co-lead an activity. At Lowell Elementary, Ms. Kite identified someone else as leading the activity for nearly a quarter of the activities in which she was involved over the six-day period. On those occasions where she reported that she was not leading the activity, the leaders she identified typically were other formally designated leaders including subject area specialists, the assistant principal, and teacher leaders, though informal leaders such as regular classroom teachers, parents, and students were also included (see Table 3.3). At Canton Elementary, Ms. Bind reported that someone else was leading just under one-fifth (18%) of the activities in which she was involved over the six-day period. During the activities in which Ms. Bind was participating but not leading, formally designated leaders such as subject area specialists and teacher leaders were more likely to be leading than were informal leaders. Indeed, classroom teachers, parents, and students were not reported to lead such activities at all. These data suggest that other formally designated leaders and informal leaders are important in understanding the work of leading and managing Lowell Elementary and Canton Elementary, although the extent to which such leaders led differs by school, with informal leaders figuring more prominently in Ms. Kite’s workday. Ms. Kite reported that for a little over four-fifths (81%) of those activities where she was leading, at least one other individual (and sometimes more than one) was co-performing the activity with her; Ms. Bind was more likely to lead alone and reported co-leading a little over one-quarter (29%) of the activities in which she was leading (See Table 3.4). While Ms. Kite was more likely to report co-performing an activity with another formally designated leader, she also reported co-performing activities with individuals with no formal leadership designations such as classroom teachers and even students and parents. Ms. Bind also led activities with both formal and informal leaders (See Table 3.4).

56

J.P. Spillane et al. Table 3.3 Percentage of time different people led when principal was not leading Leader

Lowell (n = 15)

Canton (n = 11)

Classroom teacher Subject area specialist Parents Other professional staff Other Assistant principal Teacher leader District staff Student Non-teaching staff Community member

47 47 40 33 33 27 20 13 7 0 0

0 9 0 9 36 0 18 9 0 18 0

Table 3.4 Percentage of time different people co-lead when principal was co-leading Co-leader

Lowell (n = 39)

Canton (n = 12)

Subject area specialist Other professional staff Classroom teacher Non-teaching staff Assistant principal Teacher leader District staff Student Parent Community members Other

21 21 18 18 15 15 10 3 3 0 0

0 0 14 36 0 14 7 14 0 7 7

The percentage of time that the two principals reported leading an activity when beeped varies depending on the type of activity. While Ms. Kite reported leading almost all administration activities (97%), she reported leading just under one-half (47%) of the instruction and curriculum activities. Ms. Bind, on the other hand, reported leading most administration activities (87%) and all of the curriculum and instruction activities (100%), although she only reported participating in curriculum and instruction activities during two of 60 beeps over the six-day period. These data on the school principal’s workday suggest that individuals other than the school principal may be even more important when it comes to managing and leading instruction and curriculum, especially at Lowell Elementary. Of course, the exclusive focus on the work of the school principal does not capture a variety of other leadership and management activities that may not involve the school principal. For example, some school principals may delegate certain leadership and management activities to other formally designated leaders such as an assistant principal; the ESM data would not capture these activities. Of the activities Ms. Kite reported leading, she worked alone for about one-fifth of them, and a co-leader was present for the other four-fifths. Ms. Bind, on the other hand, was much more likely to lead alone. She led alone almost three-fourths (71%) of the activities she reported leading, and reported co-leading for the other

3 The Challenge of Study Operations

57

fourth (29%). When co-performing an activity, Ms. Kite reported working with one other person 85% of the time, while 15% of the time there were two or more other people co-performing with her; Ms. Bind reported working with one other person 100% of the time when co-leading. Ms. Kite reported spending her time co-performing with a variety of people, including subject area specialists, other professional staff (e.g., guidance counselors, social workers), classroom teachers, and non-teaching staff among others. At Canton Elementary, Ms. Bind co-performed primarily with non-teaching staff, teacher leaders, classroom teachers, and students (See Table 3.4).1 When Ms. Kite was not leading an activity, a variety of people, including classroom teachers, subject area specialists, parents, other professional staff, others, and the assistant principal, were identified as leading. Ms. Bind identified others, teacher leaders, and non-teaching staff as the most prominent leaders when she was not leading an activity she was participating in (see Table 3.3). Staff Self-Report of Formally Designated Leadership Position (SSQ). Based on an analysis of the SSQ, Lowell Elementary has 15 and Canton Elementary has nine other individuals with formally designated leadership positions in addition to the school principal. This difference in the number of leaders is roughly proportional to school size, since Lowell has slightly more than twice as many students as Canton. Including the principals at both schools, a slightly larger proportion of the staff members hold formal leadership roles at Canton Elementary (29%) than at Lowell Elementary (23%). Table 3.5 shows the number of people who spent some portion of their time in each leadership position and the average percentage of time spent on each role.2 At both Lowell Elementary and Canton Elementary, these formal Table 3.5 Average percentage of time spent on each leadership role in SSQ Lowell Role Assistant principal Reading coordinator Math coordinator Other subject coordinator Special program coordinator School improvement coordinator School reform coach Mentor teacher Teacher consultant Other Any formal role

1

# of People

Canton % of Time

# of People

% of Time

1 1 2 7

100 38 69 27

2 0 0 2

56 n/a n/a 13

1

38

2

25

3

42

3

38

5 7 3 2 15

23 52 13 56 63

1 3 1 4 9

13 46 13 31 57

Note that the percentages in this table, and several other tables, will not total to 100% as respondents were able to select more than one category for several of the questions in the ESM. 2 Estimates for percentage of time spent on each leadership role were provided via a range (i.e., 0%, 1–25%, 26–50%, 51–75%, 76–99%, and 100%). When calculating the average percentage of time spent on each role, we used the midpoint of each of the ranges.

58

J.P. Spillane et al.

leadership designations include one full-time assistant principal who spends all of his or her time in the assistant-principal leadership position. At Lowell Elementary, the remaining 14 leaders include reform coaches, reading coordinators, math coordinators, other subject coordinators, mentor teachers, etc., who each spread their time across one or more roles. At Canton Elementary, the remaining eight leaders include mentors, other subject coordinators, school improvement coordinators, etc., who each perform one or more roles (see Table 3.5). Further, all of these individuals selected “regular full-time teaching appointment” when asked about employment status, while the two full-time assistant principals (one at Lowell Elementary and one at Canton Elementary) selected “administration” for this same question. School staff reports on leaders for mathematics and language arts (SSQ). In the SSQ we asked the staff at Lowell Elementary and Canton Elementary to identify those to whom they turn for advice about mathematics and reading instruction. Examining these data, we can identify both formal and informal leaders for two core school subjects and gauge how responsibility for leadership and management in these two school subjects is distributed. Using the measure of in-degree centrality, we can identify who provides leadership in mathematics and reading at Lowell Elementary and Canton Elementary. For illustrative purposes, we define a leader as any actor with an in-degree of at least three. Such a cut-off allows us to focus on those actors that are more influential in a given subject-area and to reduce sensitivity to random noise in the data. Using a fixed cut-off point also allows for simple comparisons across schools. Using this definition, we identify 13 mathematics leaders and 12 reading leaders at Lowell Elementary and three math leaders and five reading leaders at Canton Elementary. At Lowell Elementary, 10 of the math leaders are regular classroom teachers with no formal leadership designation, while only three are formally designated leaders: a mentor teacher who teaches first and second grades and two gifted education teachers who have leadership roles as whole school reform coordinators. The same mentor teacher is the only reading leader with a formally designated leadership position, while the remaining 11 reading leaders are regular classroom teachers with no formal leadership designation. Six of the informal leaders appear in both networks. At Canton Elementary, two of the three math leaders and three of the five reading leaders have formal leadership positions; the remaining math leader and two reading leaders are classroom teachers with no formal leadership designation (Table 3.6).

Table 3.6 Math and reading leaders: Number of staff with an in-degree of 3 or more Lowell

Canton

Role

Math

Reading

Math

Reading

Formal leader No formal leadership designation Total Total as a percentage of staff

3 10 13 18%

1 11 12 17%

2 1 3 8%

3 2 5 14%

3 The Challenge of Study Operations

59

Table 3.7 Percentage of advice-seeking directed towards formal leaders Lowell Role

Math

Principal Assistant principal Reading coordinator Math coordinator Other subject coordinator Special program coordinator School improvement coordinator School reform coach Mentor teacher Teacher consultant Other Any formal role

1 1 1 1 3 1 10 13 8 10 4 23

Canton Reading 1 2 1 1 5 0 3 2 9 2 2 13

Math

Reading

0 6 n/a n/a 9 6 9 3 21 6 6 33

0 8 n/a n/a 15 0 8 0 18 8 21 36

Using the social network approach, it becomes apparent that classroom teachers are involved in providing math and reading leadership at both schools. The social network questions identify more informal leaders at Lowell Elementary than at Canton Elementary. The difference in school size (Lowell Elementary is more than two times as large as Canton Elementary) is likely to be a contributing factor in the differing levels of informal leadership, since larger schools may require more reliance on informal leaders. At both Lowell and Canton, classroom teachers with no formal leadership role account for the majority of advice-relationships related to both mathematics and reading. At Lowell, 23% of mathematics advice relationships are directed to formal leaders, with the school-reform coach playing the single most prominent role (see Table 3.7). Only 13% of language arts advice relationships are directed to formal leaders at Lowell. Particularly striking in both networks is that formally-designated language arts and mathematics coordinators appear to play a very small role. At Canton, formal leaders account for approximately one third of the advice relationships in both mathematics and language arts. Mentor teachers are prominent formally-designated role in both networks at Canton.

Methodological and Epistemological Considerations As illustrated through the cases of Lowell Elementary and Canton Elementary, our four approaches to operationalizing the leader-plus aspect of the distributed perspective allow different ways of understanding how responsibility for leadership and management is distributed in schools, even when the same research instrument is used. Examining the four approaches, we can identify different epistemological assumptions. We can think about the four approaches along two dimensions (see Table 3.8). First, we can categorize the various approaches based on the data source, that is, who provides evidence about the distribution of responsibility for leadership

60

J.P. Spillane et al. Table 3.8 Epistemology and methodology Top-down Bottom-up

Designed organization

Lived organization

PQ leader team members SSQ – leadership designation

ESM SSQ – social network

and management work. Second, we can categorize the approaches by whether they generate data about the formal designed organization or the organization as lived (Meyer and Rowan, 1977). Both dimensions are critical in that they suggest different ways of coming to know and justifying knowledge claims about how responsibility for leadership and management is distributed in the school. The top-down/bottom-up dimension foregrounds who should provide evidence about the distribution of responsibility. The designed organization/lived organization dimension underscores that one can come to know how leadership is distributed in schools either through focusing on the formally designated leadership positions in the school (the designed organization) or through the day-to-day practice of leadership and management (the lived organization). While these two aspects of the organization are related, they are not mirror images of one another. The top-down approach relies on the reports of the school principals, Ms. Kite and Ms. Bind, and could be extended to include other formally designated leaders. From an epistemological perspective, examining how leadership is distributed by examining the work of the school principal is sensible given the principal’s position at the top of the organization. In this study, principals’ reports are of two sorts. In the PQ, principals report on the designed organization by listing members of the school leadership team. The designed organization, however, may not be an accurate representation of what actually happens in the day-to-day life of managing and leading the school. In the ESM, principals report on who actually performs, or co-performs with them, particular leadership and management activities. Each of the approaches shows that both formally designated and informal leaders have responsibility for school management and leadership at both schools. The ESM, however, go beyond the PQ in at least two respects. First, it goes beyond identifying those who have responsibility for leadership and management and identifies those individuals who actually do the work. Second, the ESM data provides a sense of the arrangements for distributing leadership and management work by tapping into the leadership practice aspect of the distributed framework. For example, we get a sense of the prominence of co-performance of leadership and management work; we can identify situations in which two or more actors coperform a leadership or management activity, albeit tied entirely to the principal’s practice. Using the ESM data, we can also gauge the prevalence of co-performance in the school principal’s work and examine how it differs by activity type. As we apply the method, ESM operationalizes the leader plus aspect and at least part of the leadership practice aspect. From an epistemological perspective, is sensible to include the perspective of all organizational members, especially those more often than not cast in the follower

3 The Challenge of Study Operations

61

role, such as regular classroom teachers. For leaders and managers to lead and manage, others must agree to be led and managed (Dahl, 1961; Cuban, 1988). Further, people in schools, even school principals and other formally designated leaders, can move in and out of leader and follower roles depending on the task or activity. Hence, we can learn about leadership and management by focusing on the perspectives of all organizational members so that we include the views of those who are chiefly in follower roles in addition to those of leaders. Items on the SSQ allow us to do this in two ways. First, the SSQ asks organizational members if they have a formally designated leadership position in the school. Second, the SSQ networks questions ask organizational members to identify people from whom they seek advice about mathematics and reading. The two SSQ items that were designed to incorporate the bottom-up perspective generate strikingly different accounts of how responsibility for leadership is distributed among people in the school. While the self-reports on formal leadership positions foreground the designed organization, the network questions focus on the organization as lived and allow for the emergence of both formally designated and informal leaders. At both Lowell Elementary and Canton Elementary, around one-fifth of the staff indicated that they held a formally designated leadership or management position, which is a very high percentage of the faculty. The social network questions suggest that the distribution of leadership, at least with respect to the sharing of advice about mathematics and reading instruction, is more evenly distributed between formally designated leaders and informal leaders (i.e., individuals with no formal leadership designation). Of the two principals and three assistant principals at the two schools, only one part-time assistant principal at Canton Elementary emerges as a leader from the network measures. Still, formally designated leaders remain important (three at Lowell Elementary and four at Canton Elementary) and are potentially some of the most important leaders in terms of the number of others who seek advice from them. Regardless of the way we operationalize the leader plus aspect, the data suggest that the work of managing and leading Lowell Elementary and Canton Elementary involves multiple actors, even according to the top down approach that privileges the school principal. Operationalizations that tap into the organization as lived suggest that when studying the distribution of responsibility for leadership and management, it is important to study school staff with no formal leadership designations. Even the ESM data, looking at leadership and management work from the top-down and only from the school principal’s workday, suggest that actors with no formal leadership designations are important to consider when examining how the work is distributed over people.

Variance Between Schools, Between Activity-Types, and Between School Types An important consideration is whether our study operations of the leader plus aspect manage to pick up variability in how leadership is distributed in schools. In this

62

J.P. Spillane et al.

section, we present a preliminary analysis focusing on variance between schools and between activity types in the distribution of responsibility for leadership and management work. Our main question is this: Do our various operationalizations of the distribution of responsibility for leadership and management enable us to identify differences between schools, between activity types, and between school types?

The Designed Organization The SSQ instrument provides self-reports of formally designated leadership positions. Overall, 30% (622 of 2,070 respondents) spend at least a portion of their time in a formally designated leadership role. Over 25% of these actors reported being in one or more full-time leadership positions. These percentages are fairly consistent across schools, as might be expected given that all of the schools are in the same district. A large proportion of the respondents who reported having a formal leadership designation also reported having a full-time classroom teaching assignment; over three-fourths (78.4%) of the formally designated leaders reported having a full-time teaching assignment. An average of 12.0 staff members per school (in addition to the school principal) reported having one or more formally designated leadership positions. Table 3.9 displays the total number of respondents (including those with a full-time teaching assignment) assigned to each role, the average number of people assigned to that role per school, and the average percentage of time spent on each role across schools. The percentage of staff holding formal leadership roles varied by school type. Approximately one-third of middle school and high school respondents reported having a formally-designated leadership role in their school but only slightly more than a quarter of the elementary respondents reported the same thing. Roughly one quarter of the leaders in all three types of schools reported holding a full time leadership position. Many staff holding formal leadership positions also had full time classroom teaching assignments (78% of elementary leaders, 77% of middle school Table 3.9 SSQ formal leadership designations Role

#

Avg # per school

Avg % of Time

Assistant principal Reading coordinator Math coordinator Other subject coordinator Special program coordinator School improvement coordinator School reform coach Mentor teacher Teacher consultant Other Any role

113 108 81 213 164 160 171 317 201 120 622

2.2 2.1 1.6 4.1 3.2 3.1 3.3 6.1 3.9 2.3 12.0

59 47 36 43 48 40 39 38 38 44 65

3 The Challenge of Study Operations

63

Table 3.10 Average number of formal leadership designations per school, by school type Role

Elementary school

Middle school

High school

Assistant principal Reading coordinator Math coordinator Other subject coordinator Special program coordinator School improvement coordinator School reform coach Mentor teacher Teacher consultant Other Any role

1.8 1.7 1.4 3.6 3.0 2.7 3.1 6.0 4.0 2.3 11.2

2.9 3.0 2.4 4.5 3.6 4.2 3.7 5.8 4.5 2.7 12.8

3.0 2.5 1.4 6.4 3.6 3.6 4.4 8.3 3.6 2.4 16.5

leaders, 77% of high school leaders). Not surprisingly, we found that middle and high schools had more leaders on average (12.8 and 16.5 leaders per school respectively) than elementary schools (11.2 leaders per school). Table 3.10 displays the average number of people (including those with a full-time teaching assignment) assigned to various leadership roles across schools according to school type. We suspect that these differences largely reflect the fact that high schools and some times middle schools tend to be larger, more complex organizations than elementary schools.

The Lived Organization In this section we explore variability among schools and activity types using data generated by the ESM log and the SSQ social network questions that focused on the lived organization. The ESM log captures the organization as lived from the perspective of the school principal, whereas the SSQ captures it from the perspective of all organizational members, regardless of whether they are formally designated leaders.

The Principal’s Work Day: Using ESM Data We begin by looking at variability in ESM measures by school and school type, then turn our attention to variability by activity type and the interaction between activity type and school type. We use box plots to depict the variability.3 On average, school principals lead 69% of the activities they engage in, and they lead alone for 52% of these activities. Figure 3.1 displays the distribution of 3 The “box” indicates the upper and lower quartiles of the distribution and the median is the dark horizontal line. The “whiskers” show the range of values that are within 1.5 times the box height. All of the values outside of the whiskers are considered outliers.

64

J.P. Spillane et al.

Fig. 3.1 Percentage of time leading and leading alone

the percentage of time principals spend leading an activity, and the distribution of time principals spend leading alone as a percentage of time spent leading. There is considerable variation across schools in the amount of time principals lead the activity in which they are engaged, ranging from 44% to 90%, excluding outliers (see Fig. 3.1). While some principals reported that someone else was leading over 50% of the activities they participated in over the six-day period, others reported that someone else was leading only 10% of the time. Focusing on those activities where the school principal reported leading, we see even more variation between schools in the amount of time principals lead alone as distinct from co-leading with someone else, ranging from 19% to 91%. Overall, the ESM log picks up considerable variability in the frequency with which staff other than the principal lead an activity, and the frequency with which principals co-lead activities with others. We found that principals in elementary, middle and high schools reported taking the lead on activities at very similar rates. While elementary school principals said they led 69% of the activities in which they engaged, middle school principals led 64% of the time, and high school principals led 66% of the time (Fig. 3.2). While the overall rate with which principals led activities was very similar across school types, there was greater variation among elementary and middle school principals than among high school principals. Middle school principals are considerably less likely than their elementary and high school counterparts to report leading an activity alone. Once again, the range of

3 The Challenge of Study Operations

65

Fig. 3.2 Percentage of time leading by school type

elementary principals’ reports of leading alone is much broader (7–58% excluding outliers) than the ranges of middle (14–43%) and high school (28–42%) principals As Fig. 3.3 illustrates, elementary principals vary tremendously in the frequency with which they lead alone, while differences between middle school principals are smaller. In comparison, the small number of high school principals in this sample exercise leadership alone at very similar rates. In general, the variation across principals is more pronounced when we consider the type of activity in which the principal participates. Table 3.11 reports the mean percentages for leading and leading alone by activity type. We find that principals lead the majority of administrative related activities by themselves. In contrast, principals lead just over half of the instruction and curriculum related activities. Table 3.11 reports only mean percentages by activity type. In Fig. 3.4, we examine variability between principals, focusing only on administrative activities (admin) and instruction and curriculum related activities (IC). There is more variability between principals in whether they are leading the activity for instruction and curriculum related activities than for administration related activities. There is also more variation between principals in whether they are leading alone or co-leading for instruction and curriculum related activities compared with administrative related activities.

66

J.P. Spillane et al.

Fig. 3.3 Percentage of time leading alone when leading, by school type

Table 3.12 reports the mean percentage of time that principals lead by activity type and school type. Reflecting overall patterns, 74% or more of principals in all three types of schools took the lead on administration activities, and at least 50% of the principals in each group took the lead on fostering relationships and curriculum and instruction. Principals in all three kinds of schools were considerably less likely to take the lead on professional growth activities. Within activities, we saw quite a bit of similarity between principals from different school levels, though there are a few exceptions. Elementary school principals were slightly more likely to lead administrative activities than middle and high school principals. Middle school principals were slightly more likely to take the lead in fostering relationships. Interestingly, high school principals were the most likely to take the lead on professional growth activities (40% reported doing so)

Table 3.11 Percentage of time leading and leading alone by activity Activity

% Leading

% Leading alone

Administration Fostering relationships Instruction & curriculum Professional growth

78 66 55 23

55 38 46 46

3 The Challenge of Study Operations

67

Fig. 3.4 Percentage of time leading activities

while middle school principals were the least likely (only 22% said they took the lead on such activities). Elementary school principals fell in the middle, with 30% of this group reporting that they led professional growth activities in which they were a participant. The percentage of time principals led instruction and curriculum activities was quite similar across elementary, middle, and high schools (51, 56, and 55% respectively). Among elementary and middle school leaders, there is a larger amount of variation between schools as to whether or not they are leading instruction and curriculum-related activities than there is for whether or not they are leading administrative activities. Among high school principals, the amount of variation is similar for both instruction and curriculum-related activities and administrative activities. Figure 3.5 illustrates the percent of time the principals reported leading administration- and curriculum-related activities according to school type. Table 3.12 Percentage of time leading by activity and school type Activity

Elementary school

Middle school

High school

Administration Fostering relationships Instruction & curriculum Professional growth

81 52 51 30

75 60 56 22

74 56 55 40

68

J.P. Spillane et al.

Fig. 3.5 Percentage of time leading by activity and school type

Excluding outliers, elementary leaders reported leading between 56 and 100% of the administrative activities and between 0 and 100% of the instruction and curriculum activities in which they participated, while middle school principals reported leading between 58 and 100% of the administrative activities and between 24 and 100% of the instruction and curriculum activities in which they participated. There is less variation among high school principals’ reports; they reported leading between 53 and 85% of the administration and between 50 and 75% of the instruction and curriculum activities in which they participated. There is more variability among elementary and middle school principals’ reports of leading alone compared to high school principals (see Fig. 3.6) (Table 3.13). Data generated by the ESM log also indicate variability in the staff members with whom principals engage in co-leadership. Overall, principals selected classroom teachers most frequently as co-leading an activity with them, followed by other professional staff, and teacher leaders (see Table 3.14). Principals are considerably more likely to co-lead with classroom teachers than with teacher leaders and assistant principals. The fact that principals spend more time co-leading with classroom teachers than teacher leaders is curious but may be driven by the total number of teachers relative to the number of teacher leaders and assistant principals.

3 The Challenge of Study Operations

69

Fig. 3.6 Percentage of time leading alone when leading, by activity and school type

Table 3.13 Percentage of time leading alone when leading by activity and school type Activity

Elementary school

Middle school

High school

Administration Fostering relationships Instruction & curriculum Professional growth

55 42 52 82

49 32 38 35

60 4 45 0

Table 3.14 Co-leader participation Co-leader

%

Classroom teacher Other professional staff Teacher leader Assistant principal Non-teaching staff Student Subject area specialist Parent Other District staff Community Members

29 24 24 21 16 16 10 7 7 4 2

70

J.P. Spillane et al.

On average, principals spend relatively little time co-leading with assistant principals on matters of instruction and curriculum. Principals at different types of schools (elementary, middle school, and high school) show slightly different patterns in the people with whom they co-lead. Elementary and middle school principals are most likely to report co-leading with teachers, while high school principals reported co-leading with teachers only 15% of the time (see Table 3.15). Assistant principals are the most common co-leaders for high school principals (41%); less common for middle school principals (27%); and even less common for elementary principals (16%). All three types of principals reported co-leading with teacher leaders 20–25% of the time. These differences very likely are related to school size with larger schools having more assistant principals than smaller schools. Examining the variability in the school principals’ co-leaders by activity, Fig. 3.7 displays the data for the four most frequently selected co-leaders. There is substantial variation between principals in the percentage of time they spent co-leading with teachers (either teacher leaders or classroom teachers) in activities related to instruction and curriculum and less variability for administration related activities. Again, the ESM log seems to pick up variation in how leadership is distributed over people by activity type. For example, comparing the distributions of these co-leaders to one another and comparing distributions within co-leaders across activity-types, we see that the range of variation in who co-leads differs by activity type. When broken down into school types, there is more variation among elementary, middle school, and high school principals in the percentage of time they spent co-leading with teachers on matters of curriculum and instruction than on administration (see Fig. 3.8). In regards to co-leading with assistant principals, there is variation in both curriculum and instruction and administration matters for elementary, middle school, and high school principals (see Fig. 3.9). Moreover, on average elementary principals are equally likely to work with teachers on administration and instruction and curriculum. This is not the case for middle and high schools, where principals appear to be significantly more likely to work with teachers on instruction and curriculum than on administration.

Table 3.15 Co-leader participation percentages by school type Co-leader

Elementary school

Middle school

High school

Classroom teacher Other professional staff Teacher leader Assistant principal Non-teaching staff Student Subject area specialist Parent Other District staff Community members

32 23 25 16 18 15 9 7 4 4 3

32 25 25 27 12 14 10 8 3 5 1

15 18 20 41 5 16 15 5 13 0 2

3 The Challenge of Study Operations

Admin

I&C

Teacher Leader

Admin

71

I&C

Classroom Teacher

Admin

I&C

Asst Principal

Admin

I&C

Other Prof Staff

Fig. 3.7 Percentage of time spent with co-leaders

In light of the above analysis, we conclude that the ESM log picks up variance by school, school-type, and activity-type in terms of whether the school principal is leading the activity or not, and given that the principal is leading, whether she or he is leading alone or co-leading with someone else.

Advice Givers: Using SSQ Network Data In this section we examine variability in the social network measures of leaders by activity type; in this case, mathematics and reading. For the purpose of this preliminary analysis, we define leaders as any staff member that provides advice to at least three of their colleagues. Using these criteria, we identify 181 math leaders and 200 reading leaders out of a total sample of 2492 people. Of the mathematics leaders, 45% have a formal leadership role, while 49% have no formally designated leadership role in the school (7% could not be classified due to survey non-response). Among reading leaders, 44% are formally designated as leaders while 48% are informal leaders (9% are unknown).

72

J.P. Spillane et al.

Fig. 3.8 Percentage of time spent with teachers as co-leaders by activity and school type

Fig. 3.9 Percentage of time spent with assistant principals as co-leaders by activity and school type

3 The Challenge of Study Operations

73

When divided into school type, we identify 132 elementary, 39 middle school, and 28 high school reading leaders. High school reading leaders are less likely to have a formal leadership role (36%) than are elementary or middle school reading leaders (45 and 49%, respectively), although the formal leadership role status of 18% of the high school reading leaders is unknown. We also identify 116 elementary, 32 middle school, and 31 high school math leaders. Here, elementary and high school math leaders are less likely to have a formal leadership role (43 and 42% respectively) than are middle school math leaders (56%). According to the social network approach, principals and vice-principals do not play a large role in leading mathematics and reading instruction. Only one principal emerges as a leader in mathematics, and three principals emerge as reading leaders. Assistant-principals are slightly more prominent; 9 were identified as math leaders and 10 were identified as reading leaders (see Table 3.16). More striking is that individuals with formal leadership designations in mathematics and language arts figure less prominently than we might have expected. While 37% of the reading coordinators emerge as leaders based on our analysis of the social network data, 63% did not emerge as leaders. The situation is similar for mathematics; over 50% of mathematics coordinators do not emerge as leaders based on our analysis of the social network data (see Table 3.16). Examining leadership roles by school type, we find that principals at the high school level are unlikely to play a large role in leading math or reading instruction according to the social network measures. Of the three principals identified as reading leaders, two are elementary and one is middle school. The only principal identified as a math leader is also elementary. No high school principals are identified as math and/or reading leaders using this approach. Elementary vice-principals are more likely to be identified as math leaders than are middle school and high school vice-principals, and they are also more likely to be identified as reading leaders than are high school vice-principals (see Tables 3.17 and 3.18).

Table 3.16 Percentage of formal leaders considered math and reading leaders according to network data Role

Math leaders

Reading leaders

Principal (n = 52) Assistant principal (n = 113) Reading coordinator (n = 108) Math coordinator (n = 81) Other subject coordinator (n = 213) Special program coordinator (n = 164) School improvement coordinator (n = 160) School reform coach (n = 171) Mentor teacher (n = 317) Teacher consultant (n = 201) Other leader (n = 120) Any formal role (n = 674) No formal leadership designation (n = 1, 448)

2 8 6 44 9 10 16 16 16 15 8 12 6

6 9 37 6 8 14 14 13 15 10 14 13 7

74

J.P. Spillane et al.

Table 3.17 Percentage of formal leaders considered reading leaders according to network data, by school type Role

Elementary school

Middle school

High school

Principal Assistant principal Reading coordinator Math coordinator Other subject coordinator Special program coordinator School improvement coordinator School reform coach Mentor teacher Teacher consultant Other leader Any formal role No formal leadership designation

7 11 39 10 11 14 19 17 20 14 16 17 7

9 9 39 4 8 20 11 10 11 8 13 13 7

0 4 35 0 2 7 10 6 5 0 11 8 5

As we expected, reading and mathematics coordinators do not play as prominent a role. Slightly less than two-fifths of the elementary (39%), middle school (39%) and high school (35%) reading coordinators emerge as informal reading leaders (see Table 3.17). This pattern is similar for elementary math coordinators, with only 36% emerging as informal math leaders, but between half and three-fourths of the middle school (50%) and high school (73%) math coordinators are identified as being informal math leaders (see Table 3.18). This could be due to a greater level of specialization by school subject in the middle and high schools. We need to point out two limitations with respect to operationalizing leadership using social network data. First, we define leadership for mathematics and language arts as a social influence interaction specifically related to advice-giving; therefore

Table 3.18 Percentage of formal leaders considered math leaders according to network data, by school type Role

Elementary school

Middle school

High school

Principal Assistant principal Reading coordinator Math coordinator Other subject coordinator Special program coordinator School improvement coordinator School reform coach Mentor teacher Teacher consultant Other leader Any formal role No formal leadership designation

3 15 12 36 14 10 20 17 17 15 12 15 7

0 0 3 50 2 10 17 20 20 16 3 13 5

0 4 0 73 6 10 7 9 12 17 5 10 6

3 The Challenge of Study Operations

75

Fig. 3.10 Percentage of school staff identified as math and reading leaders

our approach is unlikely to pick-up on forms of leadership that do not involve interactions. Second, as noted earlier, the social network questions in the SSQ focused exclusively on mathematics and reading and therefore are unlikely to pick up on social influence interactions that are subject matter neutral or generic (e.g., classroom management, student discipline). Hence, we urge caution in interpreting these findings as they are premised on a number of assumptions. On average, less than a tenth of the school’s respondents are identified as math (7%) and language arts (8%) leaders. However, the percentage of the school’s respondents who are identified as language arts and math leaders varies greatly between schools. The percentage of respondents who are mathematics leaders varies from 0 to 15% of the staff depending on the school, and the percentage of respondents who are identified as language arts leaders varies from 0 to 17% of the staff (see Fig. 3.10). The percentage of principals identified as mathematics leaders is similar across elementary, middle, and high schools (see Fig. 3.12). There is more variation in elementary schools and middle schools, 2–17% and 0–16% respectively (excluding outliers) than among high schools (0–12%) in the percentage of respondents who are identified as reading and/or mathematics leaders (see Fig. 3.11). Across schools, 36% of all mathematics advice relationships or interactions and 38% of all reading advice relationships or interactions are directed towards formal leaders in any role (see Table 3.19). One might expect that mathematics coordinators would play a large role in the mathematics networks, but they account for only 15% of all advice-relationships. Similarly, reading coordinators account for only 17% of advice relationships in the reading networks. The social network data

76

J.P. Spillane et al.

Fig. 3.11 Percentage of school staff identified as language arts leaders by school type

reveal considerable variation across schools in the degree to which formal leaders are involved in advice relationships. Across schools, formal leaders account for between 0 and 77% of math advice relationships, and 0% and 82% of reading advice relationships (see Fig. 3.13). Our data suggest fewer leaders when we use the social network measure rather than the self-report measure of a formal leadership position. This is to be expected as our social network measure focused narrowly on mathematics and reading rather than instruction writ large; moreover, the method did not include activities not directly tied to instruction (e.g., scheduling). Further, the social network questions attempted to zone in on leadership activities (defined as social influence interactions) rather than including both leadership and management activities. Hence fewer leaders might be expected. Comparing the two approaches, one advantage of the social network approach is that it identifies actors with no formal leadership designation as important actors in school leadership. Further, the social network data suggest that a formal leadership designation (e.g., mathematics coordinator, literacy coordinator) may not be a good indicator of who actually leads in the day-to-day life of the organization. As noted above, two-thirds of the reading coordinators are not identified as leaders for reading instruction by staff. Similarly, while 113

3 The Challenge of Study Operations

77

Fig. 3.12 Percentage of school staff identified as math leaders by school type

individuals identify themselves as assistant principals in the self-identification as formally designated leader SSQ question, only 8 and 9% of these individuals emerge as leaders for reading and mathematics respectively in the SSQ social network questions. Table 3.19 Percentage of advice relationships directed towards formal leaders Role

Math advice network

Reading advice network

Principal (n = 52) Assistant principal (n = 113) Reading coordinator (n = 108) Math coordinator (n = 81) Other subject coordinator (n = 213) Special program coordinator (n = 164) School improvement coordinator (n = 160) School reform coach (n = 171) Mentor teacher (n = 317) Teacher consultant (n = 201) Other leader (n = 120) Any formal role (n = 674)

2 5 3 15 8 8 11 12 20 12 5 36

2 4 17 3 7 9 9 11 19 11 7 38

78

J.P. Spillane et al.

Fig. 3.13 Percentage of advice relationships directed towards formal leaders

Discussion and Conclusion We have examined evidence of how responsibility for leading and managing schools is distributed across people along two dimensions. First, we examined how the data source – whether data came from principals at the top of the organization or from all organizational members (regardless of whether they were formally designated leaders or not) influences the conclusions one might draw about leadership distribution. Second, we considered how the target of data collection – be it the designed organization or the lived organization – impacts conclusions about distributed leadership. The various approaches show considerable agreement with respect to the individuals over whom leadership is distributed in schools. While acknowledging broad similarities among the various approaches, the different approaches also surface some divergence that has implications for thinking about the epistemological and methodological challenges in measuring leadership from a distributed perspective. Of the four approaches, the findings about the distribution of leadership generated by the social network questions on the SSQ differ most from the findings generated by the other instruments. This is in part a function of the fact that these questions focus on leadership for mathematics and reading rather than leadership and management for instruction writ large. Still, more than the content focus of the questions seems to be at play. For example, nearly two-thirds of the formally designated language arts leaders and one half of the formally designated mathematics leaders do not emerge as leaders based on our analysis of the social network data. Tapping into the lived organization, the social network questions suggest that an

3 The Challenge of Study Operations

79

exclusive focus on formally designated leaders may miss an important dimension of how the work of leading and managing schools is distributed over people. We acknowledge limitations inherent in using network data to identify leaders. This approach assumes that providing advice indicates the provision of leadership, an assumption about which we cannot be sure. This technique also relies on a single avenue for the exercise of leadership, namely, advice-giving relationships. While most leadership theories posit that leadership occurs through social influence interactions or relationships, leadership may also occur through non-interactional means such as modeling behavior. Approaches that target the organization as lived (ESM, SSQ) are important for tapping how leadership is distributed over actors with no formal leadership designation. Regardless of whether these approaches attempt to get at leadership from the top down (i.e., the principals’ on-the-spot reports regarding who is leading or co-leading) or from the bottom up, they appear to capture an important dimension of school leadership that may be missed by data collection approaches that focus exclusively on the organization as designed. We also examined whether the measures picked up variation between schools, school types, and activity types. Lamenting the lack of empirical work using a distributed framework, scholars have rushed to gather data to describe how leadership and management is distributed in schools, and in some cases to test whether there is a relationship between how leadership and management is distributed and student achievement. While these efforts are commendable, often a critical step in the process of moving from a broad analytic frame to empirical data collection has been glossed over and not subjected to critical reflection. In this paper, we have explored various ways of operationalizing a distributed perspective, showing how different study operations can result in different conclusions with respect to how leadership and management are distributed in schools. In doing so, we argue for much more attention to study operations when investigating leadership and management using a distributed perspective. Steps to generate an empirical knowledge base using a distributed perspective should strive to minimize threats to construct validity by carefully and critically appraise different ways of operationalizing aspects of the framework. Absent this, the validity of the inferences we can make from the empirical data will be compromised.

References Camburn, E. & Barnes, C. (2004). Assessing the validity of a language arts instruction log through triangulation. Elementary School Journal. 105, 49–74. Camburn, E., Rowan, B., & Taylor, J. (2003). Distributed Leadership in Schools: The Case of Elementary Schools Adopting Comprehensive School Reform Models. Educational Evaluation and Policy Analysis. 25(4), 347–373. Campbell, D. T., & Fiske, D. W. (1959). Convergent and discriminant validation by the multitraitmultimethod matrix. Psychological Bulletin. 56, 81–105. Cole, R. & Weiss, M. (2006, November). Follow the Leader: Identifying Organizational Leaders Using Network Data. Paper presented at Northwestern University’s Teacher Network Conference, Evanston, IL.

80

J.P. Spillane et al.

Cuban, L. (1988). The managerial imperative and the practice of leadership in schools. Albany: State University of New York Press. Dahl, R. A. (1961) Who Governs? Democracy and Power in an American City. New Haven, CT: Yale University Press. Denzin, N. K. (1989). The research act: A theoretical introduction to sociological methods (3d ed.). Englewood Cliffs, NJ: Prentice Hall. Eccles, R. G. & Nohria, N. (1992). Beyond the hype: Rediscovering the essence of management. Boston: Harvard Business School Press. Gronn, P. (2000). Distributed Properties: A New Architecture for Leadership. Educational Management & Administration. 28(3), 317–38. Gronn, P. (2003). The New Work of Educational Leaders: Changing leadership Practice in an era of School Reform. London: Paul Chapman. Harris, A. (2005). Distributed Leadership in Davies, B. (ed) (2005) The Essentials of School Leadership. London: Paul Chapman Press, pp. 160–172. Hektner, J. M., Schmidt, J. A., & Czikszentmihalyi, M. (2006). Experience Sampling Method: Measuring the Quality of Everyday Life. Thousand Oaks, CA: Sage Publications. Heifetz, R. A. (1994). Leadership Without Easy Answers. Cambridge, MA: Bellknap Press. Heller, M. F. & Firestone, W. A. (1995). Who’s in charge here? Sources of leadership for change in eight schools. Elementary School Journal. 96(1), 65–86. Leithwood, K., Mascall, B., Strauss, T., Sacks, R., Memon, N., & Yashkina, A. (2007). Distributing leadership to make schools smarter: Taking the ego out of the system [Electronic version]. Leadership and Policy in Schools. 6(1), 37–67. MacBeath, J. Oduro, G., & Waterhouse, J. (2004). Distributed Leadership. Nottingham, UK: National College of School Leadership. Meyer, J. W. & Rowan, B. (1977). Institutionalized Organizations: Formal Structure as Myth and Ceremony. American Journal of Sociology. 83(2): 340–363. Spillane, J. (2006). Distributed Leadership. San Francisco: Josey-Bass. Spillane, J. & Diamond, J. (2007). (Eds.) Distributed Leadership in Practice. New York: Teachers College Press. Spillane, J., Diamond, J., & Jita, L. (2003). Leading Instruction: The Distribution of Leadership for Instruction. Journal of Curriculum Studies. 35(5): 533–543. Spillane, J., Halverson, R., & Diamond, J. (2004). Towards a Theory of School Leadership Practice: Implications of a Distributed Perspective. Journal of Curriculum Studies. 36(1): 3–34. Spillane, J. Camburn, E., & Pareja, A. (2007). Taking a Distributed Perspective to the School Principal’s Work Day. Leadership and Policy in Schools, 6(1): 103–125.

Chapter 4

The Relationship Between Distributed Leadership and Teachers’ Academic Optimism Blair Mascall, Kenneth Leithwood, Tiiu Strauss and Robin Sacks

Introduction Current interest in distributed sources of leadership is pervasive among both researchers and practicing leaders (e.g., Harris, 2009; Hammersley-Fletcher and Brundrett, 2005; Storey, 2004). Nevertheless, systematic evidence is modest, at best, about the factors that influence the nature and extent of distributed leadership in schools, as well as the consequences of distributed patterns of leadership for schools and students. The study reported in this paper examined the relationships between four patterns of distributed leadership and a selected set of teacher beliefs likely influence teachers’ leadership distribution preferences. The study also examined the relationship between the four patterns of leadership distribution and teachers’ academic optimism. Much current leadership research aims to demonstrate the impact of leaders on schools and students (Leithwood et al., 2004; Silins and Mulford, 2002). But the direct effects of leadership on student achievement, the most defensible of the possible outcomes of interest, is weak (Hallinger and Heck, 1996). So the challenge is to identify the indirect path through which leadership influences students; this is a challenge to identify variables that leaders influence and which also influence students (Hallinger and Heck, 1996; Leithwood and Jantzi, 2000; Wahlstrom and Louis, in press). Among the wide range of possible variables mediating leaders influence on students, considerable research has focused on teachers’ beliefs and emotional states (Leithwood, 2006; Wahlstrom and Louis, in press). Work by Hoy and his colleagues (e.g., Hoy and Tarter, 1992; Hoy et al., 2006; Tschannen-Moran et al., 1998) has made an important contribution to the research on teachers’ emotional states and their effects on student achievement. The degree to which leadership is successful in improving the learning of students would appear to reflect, in part, the amount of influence leadership has on teachers’ motivations and related beliefs and feelings.

B. Mascall (B) Department of Theory and Policy Studies, Ontario Institute for Studies in Education, University of Toronto, Toronto, Canada

A. Harris (ed.), Distributed Leadership, Studies in Educational Leadership 7, C Springer Science+Business Media B.V. 2009 DOI 10.1007/978-1-4020-9737-9 5, 

81

82

B. Mascall et al.

Efforts to better understand how leadership influences schools and students is now being extended from leadership as it is exercised by individuals (such as principals) to distributed and collective leadership enactments. Most contemporary research on distributed leadership has focused on the work that leaders do, the practices which are distributed, and who takes on which practices. Recent efforts by Spillane et al. (2009) and Firestone and Martinez (2009) provide some insights into the sources of leadership in a distributed approach, and the practices of such leaders. This study, along with several other recent efforts (e.g., Mayrowetz et al., 2009), moves the agenda forward a step by asking about the causes and consequences of distributed leadership. In our recent consideration of the current work being done by researchers (Leithwood et al., 2009b), we were struck by the degree to which the work was normative. There appear to be wide ranging assumptions in the discussions about distributed leadership about how this approach is inherently “good.” Yet the research on distributed leadership is primarily descriptive, and provides only modest support for the concept. We know from our reading of the literature that leadership distribution is indeed a common phenomenon – indeed, it has been practiced in schools in some fashion for years. Leadership distribution is sensitive to its context: it is likely to vary according to the conditions found in the particular organization. The first phase of our current research concluded that leadership distribution not only exists in parallel with traditional individual leadership, but that the extent of leadership distribution is dependent on strong individual leadership from a formal leader.

Framework The framework for this study consisted of a set of teacher beliefs considered antecedent to a school’s patterns of leadership distribution, four distinct patterns of leadership distribution, and an outcome variable labelled “academic optimism.”

Antecedent Teacher Beliefs Considerable research has been done to highlight the role of teachers’ beliefs in shaping teacher practice. Leithwood et al. (2006) describe a succinct formula for the factors that define teacher performance: motivation, ability, and setting. While all three of these factors need to work together for effective teacher performance to occur, each of these factors has a significant contribution to make. In this paper, we focus on the beliefs that teachers hold, as part of their motivation to undertake their work. As Leithwood et al. (2006) conclude, understanding these factors has importance for defining what leaders do to contribute to these beliefs. Teacher beliefs can be divided in to those beliefs which teachers hold about their knowledge and abilities to undertake a task (capacity beliefs) and their beliefs about the ability of a setting to support their work (context beliefs). Capacity beliefs describe psychological states about how individuals believe they are capable of accomplishing their goals. In situations of school reform, teachers’ performance

4 The Relationship Between Distributed Leadership and Teachers’

83

is seen as being influenced by the confidence they have in their ability to implement the required changes (Heneman, 1998; Kelly and Protsik, 1997). Teacher beliefs about their ability to be successful are influenced to a great extent by the organizational context. In some situations, the conditions in schools have eroded teachers’ capacity beliefs (Leithwood et al., 1999). Such factors as the size of the school, the sense of community, or feedback from colleagues can have either a positive or a negative effect on teachers’ beliefs about their ability to succeed. Teachers’ beliefs about the context in which they work are often based on what they have seen in the past. Huberman (1988) reported that experience with mismanaged approaches to educational change can make it very difficult for a teacher to believe in the positive potential of other change initiatives. How a reform is introduced can greatly diminish teachers willingness to implement the reform. In the first phase of our research for this project, we interviewed teachers about the practice of leadership distribution. While much of the interview focused on what leaders did, what factors influenced these practices, and what the effects were (see Leithwood et al., 2009a, for a description of this earlier work), we found teachers talked at some lengths about the beliefs they held about such leadership. Four particular beliefs appeared in a number of interviews: beliefs about the leadership capacity of their colleagues, the decision-making practices in the school, the shared goals of the school, and the use of collaborative practices. We suspected that these beliefs tended to influence the pattern of leadership distribution which was practiced in the school. As a result we tested some of these beliefs in the study reported here.

Patterns of Leadership Distribution Current research in the field of leadership distribution has had two objectives. The first is to provide some empirical evidence about the practice and effects of distributed leadership. It is early days in this process, and there is as yet little strong evidence to demonstrate the power of this concept. The second approach is to bring some conceptual clarity to the concept. A number of current research projects have been concerned with attempts to define the nature of distributed leadership by describing the various patterns in the exercise of such leadership. Research by Gronn (2009), Harris (2009), Spillane et al. (2009), MacBeath (2009), Anderson et al. (2009), and Leithwood et al., (2009a) have all tried to define the various dimensions which might chart the patterns of distributed leadership. Leithwood et al. (2009b) describe the range of such dimensions: “These dimensions attempt to capture (a) differences in the range of organizational members to whom leadership is distributed, (b) the degree to which distributed forms of leadership are coordinated, (c) the extent of interdependence among those to whom leadership is distributed, (d) the extent to which power and authority accompany the distribution of leadership responsibilities, and (e) the stimulus for leadership distribution” (p. 272).

84

B. Mascall et al.

Gronn (2000) provided some early direction in our understanding of patterns of leadership distribution, identifying two contrasting patterns. The first, termed additive, describes how a number of different people may offer leadership in an organization, but without any attempt to coordinate this leadership, or for individuals to take account of what others are doing. Gronn’s second pattern was described as holistic, an approach to leadership which is deliberate, planned, and coordinated. In this approach, leaders work closely together, ensuring that the work of one leader builds on and links to the work of others. This coordinated approach ensures that all are working towards a common goal, and that all are learning as they go. Spillane et al. (2009) have picked up on this latter pattern, calling this leadership plus in their current work. Spillane and his colleagues have pushed this concept further by considering people in both formal and informal leadership roles. Both Spillane and Gronn see the importance of the relationships among the leaders: the degree of interdependence, coordination, inter-reliance, and mutual support become critical elements in the success of this approach to leadership distribution. Current research by Harris (2009) and MacBeath (2009) examines patterns of leadership distribution which vary according to the degree of power and authority exerted. Since these dimensions have been demonstrated to be of central importance in the study of other forms of leadership, this approach is expected to produce some useful results. This research raises questions about the way power is allocated, and the degree to which this allocation is coordinated or spontaneous. Our approach to the patterns of leadership distribution builds on Gronn’s holistic pattern by examining the degree to which approaches to leadership distribution are consciously aligned across the sources of leadership, and the degree to which the approach is either planned or spontaneous. This produces four different possible patterns: planful alignment, planful misalignment, spontaneous misalignment, and anarchic misalignment. Each of these is summarized briefly below. Planful alignment. In this pattern, the tasks or functions of those providing leadership have been given prior, planful thought by organizational members. The various sources of leadership consider which leadership practices or functions are best carried out by which source. This pattern is comparable to the holistic form which Gronn (2003) labels “institutionalized practice.” Although negative outcomes are possible, evidence from the first phase of our study suggests that this planned and aligned pattern of leadership distribution is likely to be associated with positive effects for the organization. The “pre-thinking,” reflective, or planful processes associated with this configuration seem likely to increase the chances of such effects in the long run. Spontaneous alignment. In this pattern, leadership tasks and functions are distributed with little or no planning. Despite this lack of deliberate planning, leadership functions appear to be aligned across leadership sources by chance, habit or for some other reason. This pattern is similar to Gronn’s (2003) “spontaneous collaboration.” While positive outcomes would be expected from this approach in the short-term, the lack of reflective feedback would make productivity over the long-term difficult to sustain. Short-term success, however, potentially reduces the

4 The Relationship Between Distributed Leadership and Teachers’

85

incentive members engaged in this form of leadership distribution have to move toward more planful and coordinated forms. Spontaneous misalignment. This pattern is similar to Spontaneous Alignment in the lack of planning for leadership distribution but chance, habit, and so on, in this case, result in misalignment and largely negative consequences for the organization. Lack of alignment makes it difficult to achieve even short-term successes. Under supportive conditions, it seems possible to shift the people associated with this form of leadership distribution to more planful forms. The lack of even short-term success in this pattern may provide an incentive to change. Anarchic misalignment. Associated with some secondary school departments in the first phase of our study, this pattern features substantial planning and alignment within a sub-unit (such as a department) but an oppositional or competitive disposition in relation to the organization as a whole. Movement toward forms of leadership distribution reflecting wider organizational goals is likely to hinge on the success others in the organization have in convincing those engaged in this pattern of the value of those wider goals.

Academic Optimism One of the challenges in any study of leadership is the identification of the outcomes that will be affected by such leadership. In the field of education, there has been considerable work in recent years to attempt to link leadership practices to educational outcomes (Leithwood et al., 2004). While the link between formal leadership and student success is at best tenuous, there is a body of work which helps to unpack the variables that mediate between leaders and students. A recent study by Hoy, Tarter and Woolfolk Hoy (2006) found large effects on student achievement by a variable they labeled “academic optimism.” This was an aggregate variable which included three teacher beliefs – trust, collective efficacy and academic emphasis. Hoy and his colleagues argued that optimism was “an appropriate overarching construct . . . because each construct contains a sense of the possible” (2006, p. 431). More recent work by Hoy et al. (2007) has tested and refined the earlier construct. We used an adapted version of Hoy et al.’s academic optimism as a measure of teacher beliefs for this study. The adaptation was to replace “academic emphasis” with “organizational citizenship behavior (OCB).” This replacement was not based on objections to the value of academic emphasis. Our reasoning was that successful change in schools depends on the willingness of teachers to engage in work with their colleagues outside of their own classrooms. Measures of OCB aim to capture that willingness. Organizational Citizenship Behavior. Organizational Citizenship Behaviour (OCB) describes the feelings of belonging (termed “citizenship”) in the organization, and the behaviours that promote this. Organ (1988) coined this phrase to indicate behaviours that were not required as part of the job, but were offered in order to help others in the organization. Bateman and Organ describe OCBs as practices that

86

B. Mascall et al.

“lubricate the social machinery of the organization” (1983, p.588). The willingness to undertake such altruistic deeds makes a significant contribution to the success of the organization. Applied to schools, DiPaola and Hoy (2005) found a significant relationship between the OCBs of staff in a high school and their students’ achievement on standardized tests. They describe the practices they observed related to teacher’s OCB: Teachers who voluntarily help their new colleagues and go out of their way to introduce themselves to others define organizational citizenship behaviors in schools. Teachers in such schools take it upon themselves to make innovative suggestions, to volunteer to sponsor extra-curricular activities, and to volunteer to serve on new committees. Moreover, teachers help students on their own time, stay after school to help if necessary, and resist the temptation to give students busy work. Organizational citizenship behavior in schools connotes a serious educational context in which teachers are rarely absent, make efficient use of their time while at school, work productively with their colleagues, and give high priority to professional activities over personal ones while in school. They use their talents and efforts to help both students and the school to achieve. (p. 36).

Our measure of OCB combines items from Podsakoff et al. (1990) 24-item scale with items from one sub-scale of Van Dyne and LePine’s (1998) survey. Podsakoff et al. measure the five dimensions of OCB identified by Organ (1988): altruism, courtesy, civic virtue, conscientiousness and sportsmanship. Van Dyne and Le Pine added a sixth dimension, “voice”, that describes a person’s capacity to make “innovative suggestions for change and to recommend modifications to standard procedures even when others disagree” (1998, p. 108). As yet, there is little evidence about how leaders influence the development of OCBs on the part of teachers. Trust. As with OCBs, there is now a small amount of evidence (in particular, Bryk and Schneider, 2002; Tschannen-Moran, 2004) demonstrating that the trust felt between and among teachers and administrators is both influenced by leader practices and an influence on student learning. This study measured both mutual trust among teachers and teachers’ trust in leaders. Evidence provided by Wahlstrom and Louis (in press) demonstrated that supportive principal behavior and faculty trust are significantly correlated, and that schools with higher levels of engaged teachers have higher levels of trust in colleagues. Directive (as opposed to supportive) principal behavior is negatively correlated with teachers’ trust in their principal, but has no impact on trust in their teacher colleagues. This implies that principals can build trust between themselves and their teachers, but they have little influence on the trust that teachers feel among themselves. Bryk and Schneider (2003) found that respect, personal regard, competence in core role responsibilities, and personal integrity, were associated with relational trust among teachers and leaders. Louis (2007) identified specific principal behaviors that had an impact on teacher-teacher trust, including effective communication, clear vision, consistency between words and actions, and competent management of school affairs. Trust was also built by reshaping the composition of the staff through hiring, and counseling out teachers who did not live up to the school’s mission and values. In buildings characterized by high trust, there was more collective decision making,

4 The Relationship Between Distributed Leadership and Teachers’

87

with a greater likelihood that reform initiatives were widespread and that there were demonstrated improvements in student learning (Louis, 2007). While discussions of trust have traditionally focused on the individual, more recent work has become concerned with interpersonal relations and organizational behaviour. As organizations attempt to promote more collaboration and cooperation, trust among employees and trust in leaders have become increasingly important variables in explaining key organizational outcomes (Tschannen-Moran, 2004). In organizations with high levels of trust, individuals are comfortable in seeking help from others and learning from their coworkers. Teacher Efficacy. Our study measured both individual and collective teacher efficacy. Bandura (1977) defined teacher efficacy as a component of self-efficacy; it is the confidence teachers have about their ability to accomplish a teaching task. According to Bandura (1995), four factors influence self efficacy, including mastery experiences (based on personal experience with the task); psychological and emotional states (an individual’s level of arousal); vicarious experiences (seeing others succeed or fail in a particular task); and social persuasion (formal or informal, from leaders or colleagues). Teacher efficacy has a demonstrable impact on student achievement. The earliest studies of teacher efficacy by the Rand organization (Armor et al., 1976) found that teacher efficacy explained a high proportion of the variation in reading achievement in minority students. Correlations between teacher efficacy and teacher practices are also very high. For example, Ross’s (1998) review of the consequences of teacher efficacy concluded that: Higher teacher efficacy is consistently associated with the use of teaching techniques that are more challenging and difficult, with teachers’ willingness to implement innovative programs, and with classroom management practices that promote student responsibility. High expectations of success enable teachers to set higher goals for themselves and others, take risks in experimenting, and learn new methods that contribute to higher student achievement. (p. 58)

While the concept of efficacy was initially focused on the individual, more recent work has defined this as a collective attribute, linking it to much of the literature on organizational constructs. Goddard et al. (2000) define collective efficacy as “the perceptions of teachers in a school that the effort of the faculty as a whole will have a positive effect on students” (p. XXX). They argue that collective efficacy will promote persistence in the school, and provide sanctions for those who do not persist. In a school with high collective efficacy, teachers accept responsibility for outcomes, and are not discouraged by temporary setbacks. Since evidence exists that teachers’ collective efficacy can be a stronger predictor of student achievement than students’ socioeconomic status (Goddard et al., 2000), there is a clear need to describe what school leaders do to support efficacy among their faculty. Principal leadership behavior and the development of teacher efficacy were studied in a small sample (N = 10) of middle schools (Hipp, 1996). Such leadership affected efficacy largely by addressing in-school problems within the principals’ control, such as discipline or shared decision-making.

88

B. Mascall et al.

Only one published study, to our knowledge, clearly links teacher efficacy with trust. Da Costa and Riordan (1996) explored the relationship between teachers’ sense of efficacy and the role of trust in teachers’ willingness to engage in workfocused relationships with colleagues. Teachers in high-trust work relationships who were also confident of their teaching abilities were willing to have conversations about instructional pedagogy in a variety of settings, including general team meetings. Conversely, when teachers perceived themselves to be lacking in efficacy they would not go beyond the bounded conversation for a pre- or post-observation conference with a colleague. Although these results are limited by the study’s small scale, they point to the need for further examination of trust and sense of efficacy. Starting with the patterns of leadership distribution, and their antecedents, identified in the first phase of our current research, we explored the relationship between these antecedents, the leadership distribution patterns and academic optimism. We hypothesized that: 1. Key beliefs predispose teachers’ choices of leadership distribution patterns in which to engage. 2. High degrees of academic optimism are associated with planned approaches to leadership distribution. 3. Low levels of academic optimism are associated with unplanned approaches to leadership distribution.

Methods Context Evidence for this paper was gathered as part of a three-year mixed-methods study of distributed leadership. The study was conducted in one large school district in Ontario, Canada, in which deliberate, sustained attempts had been made by district staff to expand the distribution of leadership in schools over more than a dozen years. Serving more than 100,000 students in about 150 elementary and 30 high schools, the district employed approximately 8800 teachers and 400 school administrators. The socio-economic status (SES) of families in this district varied widely, and the schools were located in urban, sub-urban, and rural settings. Overall, however, the families in the district had higher than average SES, and the majority of schools were in urban and suburban locations. Among other outcomes, the first phase of our study (conducted in eight case study schools) identified four patterns of leadership distribution (Leithwood et al., 2009a). One goal of second phase of this study – the focus of this paper – was to provide quantitative evidence about the relationship between these four different patterns of leadership distribution selected school variables linked to student achievement in prior research. Our end goal was to help move research on distributed leadership beyond its current, largely descriptive state to one which offers more insights about the improvement of leadership practice.

4 The Relationship Between Distributed Leadership and Teachers’

89

Sample The survey population for this study was all licensed, part-time and full-time teachers in the district (about 8800) and the study included this entire population. A total of 1640 teachers responded to one of two forms of an on-line survey. The low (18.6%) response rate was due to some combination of the on-line nature of the surveys and the unusually and inadvertently demanding nature of other extraclassroom tasks teachers were grappling during the period in which the surveys were being conducted (completing report cards using a centralized computer program that malfunctioned frequently). Three different strategies were used to increase response rates over a two month period in order to acquire the achieved sample. By most social science standards, this achieved sample is relatively large. For example, only 224 respondents are needed to represent a population of 8800 for a confidence interval of 0.99 and a margin of error of 0.93. With 1640 respondents, the margin of error drops to .032. But such calculations assume that the achieved sample is normally distributed, an assumption we cannot make in this case.

Measures Our research used a teacher survey to gather data on a number of constructs. The analysis reported in this chapter focuses on three groups of items, as described below. Responses to all survey items used a 6-point scale that ranged from “disagree strongly” to “agree strongly.” Beliefs influencing teachers’ choice of leadership distribution pattern. Based on what we heard in phase one interviews, we developed a series of statements that articulated the kinds of beliefs we expected to find among teachers. These statements concerned four topics of belief: the leadership capacity of their colleagues, the decision-making practices in the school, the shared goals of the school, and the use of collaborative practices. In each of these statements we asked teachers to evaluate the degree to which they felt that their colleagues in the school shared these beliefs. The statements then give a sense of what individual teachers saw as the belief set for teachers in the school.

r r r r

People in this school have unrealistic beliefs about the leadership capacities of their colleagues My colleagues and I generally believe that reflection and dialogue and essential for good decision making There is little commitment in this school to a common set of shared goals Competition is more productive than cooperation as an incentive for improving our practices

Patterns of leadership distribution. The items measuring patterns of leadership distribution ask respondents to identify the extent to which leadership distribution

90

B. Mascall et al.

is planned or spontaneous, aligned or anarchic. One item measured one pattern, as follows:

r r r r

We collectively plan who will provide leadership for each of our initiatives and how they will provide it. (planful alignment) The distribution of leadership tasks in this school is “spontaneous.” It is not planned and it usually works out well. (spontaneous alignment) The distribution of leadership tasks in this school is “spontaneous.” It is not planned and it often leads to conflict and confusion. (spontaneous misalignment) Leadership within individual divisions or departments coordinate their work carefully, but it is not done across the school as a whole. (anarchic misalignment)

Teachers’ views of their academic optimism. Table 4.1 lists the items used to measure all subscales of academic optimism. Variables encompassed by this aggregate construct were assessed using measures drawn from previous research. The 4 item scale for “mutual trust among teachers” was adapted from Bryk and Schneider (2002) and included items such as “teachers in this school really care about each other”. The internal reliability of this sub-scale was .82. Our measure of teacher trust in school leaders, a 7 item scale, was adapted from Podsakoff et al. (1990) and Table 4.1 Teacher ratings of the component variables included as part of academic optimism6 N

M

SD

Rel .

Mutual trust among teachers

• Teachers at this school really care about each other. • Teachers in this school trust each other. • It’s OK in this school to discuss feelings, worries and frustrations with other teachers. • Teachers in this school respect colleagues who take the lead in school improvement efforts.

1130

4.68

1.09

.88

Teacher trust in school leaders

• I feel quite confident the leaders at my school always try to treat me fairly. • Leaders at my school would not try to take advantage of teachers by deceiving them. • I feel a strong loyalty to our school leaders. • I would support the leaders at my school in almost any emergency. • I have a divided sense of loyalty toward my school leaders. (reverse scale) • It’s OK in this school to discuss feelings, worries and frustrations with school leaders. • Leaders in our school look out for the personal welfare of teachers in this school.

1609

4.51

1.05

4 The Relationship Between Distributed Leadership and Teachers’

91

Table 4.1 (continued) N

M

SD

Teacher self Efficacy

• If students aren’t disciplined at home, they aren’t likely to accept any discipline at school. • When I really try, I can get through to the most difficult or unmotivated students. • A teacher is very limited in what he/she can achieve because a student’s home environment is a large influence on his/her learning. • If parents would do more for their children, I could do more for my students. • If a student did not understand information from a previous lesson, I would know how to increase his/her understanding in the next lesson.

1068

4.31

.87

Teacher collective efficacy

• If a student doesn’t learn something the first time, teachers in this school will try another way. • Teachers in this school really believe every student can learn. • If a student doesn’t want to learn, most teachers here give up.∗ (reverse scale)

1397

4.20

.85

Rel . .42

• Teachers in my school need more training to know how to deal with the students who aren’t learning.∗ (reverse scale) • Teachers in my school don’t have the skills needed to produce meaningful student learning.∗ (reverse scale) • Most students come to school ready to learn. • Home life provides so many advantages students are bound to learn. • Students here just aren’t motivated to learn.∗ (reverse scale) OCB – altruism

• I have taken steps to try to help colleagues who have been absent. • I make a point to help colleagues who have heavy workloads. • I have helped orient a new teacher even though it is not required. • I am always ready to lend a helping hand to those around me.

1181

4.78

.984

.83

OCB – civic virtue

• I attend meetings that are not mandatory but are considered important. • I keep abreast of changes in the school. • I read and keep up with school announcements, memos and so on.

1052

5.00

.929

.73

92

B. Mascall et al. Table 4.1 (continued) N

M

SD

Rel . .93

OCB – voice

• I have made recommendations concerning issues that affect school staff. • I speak up and encourage others staff to get involved in school issues. • I have communicated my opinions about work issues to others in this school even if my opinion is different and others in the group disagree with me. • I keep well informed about issues where my opinion might be useful to this school. • I am involved in issues that affect the quality of work life here in this school. • I have spoken up at this school with ideas for new projects or changes in procedures.

1274

4.05

1.25

OCB – courtesy

• I have done or said something to prevent problems with other teachers and staff.

799

4.24

1.32

1603

4.48

Total OCB

.942

6

These data were collected using two forms of the teacher survey in order to keep the total amount of time required by a single respondent manageable (the two surveys included a total of xx and yy items). When items included in a scale were divided among the two instruments, results do not allow the calculation of a scale reliability.

Bryk and Schneider (2002): it includes items such as “I feel quite confident that my principal will always try to treat me fairly.” Previous research reported alpha coefficients in .92 range for this sub-scale. The five item “teacher self-efficacy” sub-scale was adapted from TschannenMoran et al. (1998) and includes items such as, “when I really try, I can get through to the most difficult or unmotivated students.” The reliability of the original scale is .77. The eight item scale for collective teacher efficacy was derived from Ross et al. (2004) and included items such as “teachers in this school really believe every student can learn.” The reliability of the original scale was .83. Our measure of organizational citizenship behaviour included 14 items from surveys by both Podsakoff et al. (1990) and Van Dyne and LePine (1998). Estimates of the internal consistency of the five OCB dimensions averaged across twelve samples reported by Podsakoff et al. were as follows: altruism (.88), courtesy (.87), conscientiousness (.85), sportsmanship (.88), and civic virtue (.84). Van Dyne and LePine report that, across six samples of data, the voice scale was found to have high internal consistency reliability and high test-retest reliability. These scale reliabilities from earlier research provide some confidence in the starting points for our own surveys but we used only the existing “voice” scale intact. As reported in Table 4.1, the ratings of OCB, trust, and efficacy in our research were moderate (4.20 for collective efficacy) to high (4.68 for teachers’ trust in teachers).

4 The Relationship Between Distributed Leadership and Teachers’

93

Data Analysis Individual respondents were the unit of analysis. We have become increasingly convinced that the individual rather than the school level is appropriate unit of analysis in the majority of the research carried out about school-level leadership. Our findings, across many previous studies, that there is more variation in the responses of teachers within than across schools to the leadership they experience, provide the justification for this claim. This claim also reflects the major premise of Leader Member Exchange Theory (Graen and Uhl-Bien, 1998) that leaders are likely to behave in different ways with (in the case of schools) their teacher colleagues depending on such perceived factors as their teachers’ levels of expertise, motivations to change, willingness to engage in school-wide decisions and interpersonal styles.

Results Overview of Results Table 4.2 reports means and standard deviations of teachers’ responses to the questions about the patterns of leadership distribution in their school. Table 4.2 Ratings of leadership distribution Planful alignment Anarchic misalignment Spontaneous alignment Spontaneous misalignment

Mean rating

SD

3.72 3.36 3.05 2.29

1.62 1.49 1.48 1.41

The mean ratings for patterns of leadership distribution ranged from a high of 3.72 (slightly agree) for planful alignment to a low of 2.29 (disagree) for spontaneous misalignment, with anarchic misalignment and spontaneous alignment in between. Standard deviations of responses are fairly high, suggesting that teachers varied widely in their perceptions of the patterns of leadership distribution in their schools.

Teacher Beliefs and Patterns of Leadership Distribution We were curious to see how teacher beliefs about teaching and learning correlated with the various patterns of leadership distribution. Table 4.3 presents the results of our analysis, and correlates those ratings to the values given for the patterns. We discovered that planful alignment was distinctly different from the other patterns in terms of the correlation with these beliefs. Planful alignment is the most likely pattern to emerge when the following beliefs are in place:

94

B. Mascall et al.

Table 4.3 Relationship between leadership distribution patterns and teachers’ underlying beliefs (Correlation coefficients)

People in this school have unrealistic beliefs about the leadership capacities of their colleagues My colleagues and I generally believe that reflection and dialogue and essential for good decision making There is little commitment in this school to a common set of shared goals Competition is more productive than cooperation as an incentive for improving our practices

Planful alignment

Anarchic alignment

Spontaneous alignment

Spontaneous misalignment

−.155(∗∗ )

.392(∗∗ )

.286(∗∗ )

.176(∗∗ )

−.171(∗∗ )

−.109(∗∗ )

−.055

−.291(∗∗ )

.394(∗∗ )

.278(∗∗ )

.044

−.076(∗ )

.327(∗∗ )

.191(∗∗ )

.052

.096(∗∗ )

∗∗ ∗

-

Correlation is significant at the 0.01 level (2-tailed). Correlation is significant at the 0.05 level (2-tailed).

teachers believe that there is a commitment to shared whole-organization goals; teachers believe in using cooperation rather than competition in the organization teachers believe in reflection and dialogue as basis for decision making teachers have well-rounded beliefs about the capacities of their leadership colleagues

Conversely, spontaneous forms and misaligned forms are less likely when these beliefs are present. This strong endorsement of planful alignment as the pattern of leadership distribution most associated with beliefs about a positive school culture leads us to conclude that teachers see a planned and aligned approach to leadership distribution as a positive part of an effective and supportive school.

Relationships Between Leadership and Academic Optimism Table 4.4 reports the correlations between patterns of leadership distribution and the sub-scale components of academic optimism as well as the aggregate measure. Planful alignment is moderately related to the aggregate academic optimism (.34). Among the component sub-scales, this pattern of distributed leadership is

4 The Relationship Between Distributed Leadership and Teachers’

95

Table 4.4 Correlations between patterns of distributed leadership and academic optimism Planful alignment

Anarchic misalignment

Spontaneous misalignment

Spontaneous alignment

Collective teacher efficacy

.114∗∗

−.127∗∗

−.170∗∗

−.028

Teacher self efficacy

.085∗∗

−.079∗

Trust in teachers

∗∗

.258

−.156∗∗

−.049

∗∗

−.286∗∗

.038

∗∗

∗∗

−.140

Trust in leaders

∗∗

.403

−.185

−.356

.002

Organizational citizenship behaviour

.123∗∗

−.040

−.027

−.084∗∗

Academic optimism

.339(∗∗ )

−.171(∗∗ )

−.301(∗∗ )

−.063(∗ )

∗∗ ∗

Correlation is significant at the 0.01 level (2-tailed). Correlation is significant at the 0.05 level (2-tailed).

most strongly related to trust in leaders (.40), while only weakly (but still significantly) related to the rest. Spontaneous misalignment has associations with academic optimism least similar to the relationships with planful alignment. Its relationship with aggregate academic optimism is negative (−.30) and this negative relationship seems likely accounted for by the corrosive effects of this form of leadership on trust, especially trust in leaders (−.36). The remaining two patterns of distributed leadership have weak but significant negative relationships with academic optimism. We explored this relationship between Academic Optimism and patterns of leadership distribution a little further. Respondents were divided into five groups or quintiles, based on their ratings of academic optimism. Ratings of leadership distribution were then mapped onto each of these quintiles. Figure 4.1 displays the results of this analysis. The horizontal axis in Fig. 4.1 presents the five quintiles of academic optimism, with the first quintile being a low level, and the fifth quintile being a high level of academic optimism. The vertical axis presents the ratings teachers gave each of our four patterns of leadership distribution. We used this analysis to explore how these views about the patterns of leadership distribution changed depending on the ratings for academic optimism. Planful Alignment is unique among the patterns in showing consistently increasing associations with academic optimism. Planful Alignment is rated lowest in the quintile of respondents with the lowest rating of academic optimism, and rated highest in the quintile of teachers with the highest ratings of academic optimism. All the other patterns of leadership distribution move more or less in the reverse direction, with higher ratings of the three other patterns of distributed leadership associated with lower ratings of academic optimism. In sum, the more academically optimistic are teachers, the more likely they are to report that leadership is distributed in their schools in a planfully aligned pattern.

96

B. Mascall et al.

Fig. 4.1 The relationship between patterns of leadership distribution and academic optimism

4.50

4.00

Mean

3.50

3.00

2.50

2.00

1.50 1.00

2.00 3.00 4.00 5.00 Acadmic Optimism (quintiles) Planful Alignment Spontaneous Alignment Spontaneous Misalignment Anarchic Misalignment

Conclusion This study explored the relationship between four different patterns of distributed leadership and a set of teacher beliefs which prior evidence has suggested help influence teachers’ preferences for leadership distribution, and academic optimism, a variable demonstrably consequential for student learning. The four patterns of distributed leadership varied along two dimensions – the planfulness of leadership distribution and the alignment of leadership with the purposes of the school organization. This conception of leadership distribution patterns reflects theory and evidence suggesting that more coordinated forms of leadership distribution make more productive contributions to organizational outcomes (e.g., Osborn and Hunt, 2007). Evidence from the survey responses of 1640 elementary and secondary teachers in one large district provided the evidence used to test our hypotheses about these relationships. Our hypothesis about key beliefs being associated with productive patterns of leadership distributions is supported by this evidence. When teachers believe in the power of cooperation, and in the use of reflection and dialogue as the basis for decision-making, and when they hold realistic beliefs about the leadership capacities of their colleagues, they are more likely to perceive themselves to be engaged in more planned and aligned patterns of leadership distribution.

4 The Relationship Between Distributed Leadership and Teachers’

97

Teachers’ academic optimism was also hypothesized to be most strongly and positively associated with a planfully aligned pattern of leadership distribution. The study has several methodological limitations. The survey response rate of about 19% is one of them. While there were clearly identified reasons for this, moderate response rate that had nothing to do with the survey itself, we cannot know how well our evidence represents the population of teachers in the study district – not to mention the much broader populations of teachers whose beliefs we aspire to understand. Nonetheless, the sample is relatively large and the primary limitation concerns the degree to which it is normally distributed in comparison with the population of teachers in the district. A second methodological limitation arises from the distribution of items measuring some of our variables across two forms of the survey. While this was done to reduce the demands on respondents, it prevented us from calculating the reliability of some of the scales and eliminated the possibility of testing the factor structure of our measures. The practical implications of these results depend in part on some speculations about the nature of the distributed leadership – academic optimism relationship. We have not claimed that academic optimism is either an antecedent or an outcome of planfully aligned forms of distributed leadership, although we were initially drawn to academic optimism as a potential consequence of distributed leadership. If that is actually the case, our results offer reasonably compelling support for efforts to ensure that leadership is distributed in schools in planful ways. Spontaneous patterns seem to have negative effects and our data also suggests that this may have something to do with trust in leaders. Planful forms of distribution make leaders decisions more transparent and less open to suspicion or concern, whereas spontaneous forms of distribution often may leave the motivations and decisions of leaders implicit. Academic optimism, however, may be an influence on the development of different leadership distribution patterns, an antecedent variable. When teachers hold high levels of such optimism, they may be much more disposed to engage intentionally with their colleagues in efforts to improve their schools. Such intentionality would foster planfulness. As these speculations about the directionality of the distributed leadership – academic optimism relationship suggest, further research aimed at clarifying the relationship would be very useful. That some set of teacher beliefs would have an influence on the nature of the distributed leadership which emerges in schools seems almost certain. But which beliefs? Is academic optimism and all of its components one of them? Some teacher beliefs are no doubt influenced by or a consequence of experiencing different forms of distributed leadership. Is academic optimism among them? What other consequences flow from coordinated or planful and uncoordinated or spontaneous patterns? Is improved student learning a likely consequence of planfully aligned patterns of distributed leadership? Further analysis of these data, and other future research will help to clarify the place of academic optimism in relationship to distributed leadership. Assuming further support for the value to organizations of planful alignment, it will also be important to know in much more detail what this pattern of leadership

98

B. Mascall et al.

distribution looks like in practice. As yet, our description of this pattern is best characterized as “thin.” Recent work by Spillane and Camburn (2009) examines the quite different ways in which principals and their colleagues co-perform leading activities, and how the sources of such leadership vary from activity to activity. This will be a productive area to explore in future work, as well. Normative approaches to leadership distribution will need to find a “third way” in which formal leaders and teachers work together to share leadership in a planned and aligned way, supporting each other in a trustful, collaborative and confident manner. Such an approach may lead to improving schools, and ultimately, to student success.

References Anderson, S., Moore, S., and Sun, J. (2009). Positioning the Principals in Patterns of School Leadership Distribution.. In Leithwood, K., Mascall, B., and Strauss, T. (Eds.), Distributed Leadership According to the Evidence. New York: Routledge. Armor, D., Conroy-Oseguera, P., Cox, M., King, N., McDonnell, L., Pascall, A., et al. (1976), Analysis of the school preferred reading programs in Los Angeles minority schools. (rep. No. R-2007-LAUSD), Santa Monica, CA: RAND (ERIC Document Reproduction Service No. 130 243). Bandura, A. (1977), Self-Efficacy: The Exercise of Control. New York, NY: W. H. Freeman and Sons. Bandura, A. (1993), Perceived self efficacy in cognitive development and functioning, Educational Psychologist, 28(2), 117–148. Bateman, J. and Organ, D. (1983), “Job satisfaction and the good soldier: The relationship between affect and employee “citizenship””, Academy of Management Journal, 26(4), 587–595. Bryk, A. S. and Schneider, B. (2002), Trust in Schools: A Core Resource for Improvement. New York, NY: Russell Sage Foundation. Bryk, A. S., and Schneider, B. (2003). Trust in schools: A core resource for school reform. Educational Leadership, 60(6), 40–44. Da Costa, J. L. and Riordan, G. P. (1996), Teacher efficacy and the capacity to trust, paper presented at the annual meeting of the American Educational Research Association, New York, NY. DiPaola, M. and Hoy, W. (2005), “Organizational citizenship of faculty and achievement of high school students”, High School Journal, 88(3), 35–44. Firestone, W. A. and Martinez, M. C. (2009), “Districts, teacher leaders, and distributed leadership: Changing instructional practice. In Leithwood, K., Mascall, B., and Strauss, T. (Eds.), Distributed Leadership According to the Evidence. New York: Routledge. Goddard, R. D., Hoy, W. K., and Woolfolk Hoy, A. (2000), “Collective teacher efficacy: Its meaning, measure and impact on student achievement”, American Educational Research Journal, 37(2), 479–507. Graen, G. B. and Uhl-Bien, M. (1998), “Relationship-based approach to leadership: Development of leader-member exchange (LMX) theory of leadership over 25 years: Applying a multi-level multi-domain perspective”. In Dansereau F. and Yammarino, F. (Eds.), Leadership: The Multilevel Approaches (Vol. 24, Part B), Stamford, CT: JAI, pp. 103–134. Gronn, P. (2000) Distributed Properties: A New Architecture for Leadership. Educational Management and Administration, 28(3), 317–338. Gronn, P. (2003), The New Work of Educational Leaders: Changing Leadership Practices in an Era of School Reform. London, UK: Paul Chapman. Gronn, P. (2009). Hybrid Leadership. In Leithwood, K., Mascall, B., and Strauss, T. (Eds.), Distributed Leadership According to the Evidence. New York: Routledge. Hallinger, P. and Heck, R. (1996), “The principal’s role in school effectiveness: An assessment of methodological progress, 1980–1995”. In Leithwood K, and Hallinger, P. (Eds.), Interna-

4 The Relationship Between Distributed Leadership and Teachers’

99

tional Handbook of Educational Leadership and Administration, Dordrecht. The Netherlands: Kluwer Academic Publishers, pp. 723–783. Hammersley-Fletcher, L. and Brundrett, M. (2005), “Leaders on leadership: The impressions of primary school headteachers and subject leaders”, School Leadership and Management, 25(1), 59–75. Harris, A. (2009), “Distributed knowledge and knowledge creation”. In Leithwood, K., Mascall, B. and Strauss, T. (Eds.), Distributed Leadership According to the Evidence. New York, NY: Routledge. Heneman, H. G. (1998). Assessment of the motivational reactions of teachers to a school-based performance award program. Journal of Personnel Evaluation in Education, 12(1), 43–59. Hipp, K. A. (1996). Teacher efficacy: Influence of principal leadership behaviour, paper presented at the annual meeting of the American Educational Research Association, New York, NY. Hoy, W. and Tarter, J. (1992), “Measuring the health of the school climate: A conceptual framework”, NAASP Bulletin, November, pp. 74–79. Hoy, W. K., Tarter, C. J. and Woolfolk Hoy, A. (2006), “Academic optimism of schools: A force for student achievement”, American Educational Research Journal, 43(3), 425–446. Hoy, A. W., Hoy, W., and Kurz, N. (2007). Teachers’ Academic Optimism: The development and test of a new construct. Teaching and Teacher Education, 24, 821–835. Huberman, M. (1988). Teacher careers and school improvement. Journal of Curriculum Studies, 20(2), 119–132. Kelley, C., and Protsik, J. (1997). Risk and reward: Perspectives on the implementation of Kentucky’s school-based performance award program. Educational Administration Quarterly, 33(4), 474–505. Leithwood, K. (2006). Teacher Working Conditions that Matter: Evidence for Change. Toronto, ON: Elementary Teachers’ Federation of Ontario. Leithwood, K. and Jantzi, D. (2000), “The effects of transformational leadership on organizational conditions and student engagement”, Journal of Educational Administration, 38(2), 112–129. Leithwood, K., Jantzi, D., & Steinbach, R. (1999). Changing leadership for changing times. Buckingham, UK: Open University Press. Leithwood, K., Jantzi, D., and Mascall, B. (2002). A Framework for Research on Large-scale Reform. Journal of Educational Change, 3(1), 7–33. Leithwood, K., Seashore Louis, K., Anderson, S., and Wahlstrom, K. (2004), How Leadership Influences Student Learning: A Review of Research for the Learning from Leadership Project. New York, NY: The Wallace Foundation. Leithwood, K., Mascall, B., and Strauss, T. (2009a). What we have learned and where we go from here. . In Leithwood, K., Mascall, B., and Strauss, T. (Eds.), Distributed Leadership According to the Evidence. New York: Routledge. Leithwood, K., Mascall, B., Strauss, T., Sacks, R., Memon, N. and Yashkina, A. (2009b), Distributing leadership to make schools smarter: Taking the ego out of the system. In Leithwood, K., Mascall, B., and Strauss, T. (Eds.), Distributed Leadership According to the Evidence. New York: Routledge. Louis, K. (2007), “Trust and improvement in schools”, Journal of Educational Change, 8(1), 1–24. MacBeath, J. (2009). Distributed Leadership: Paradigms, Policy and Paradox.. In Leithwood, K., Mascall, B., and Strauss, T. (Eds.), Distributed Leadership According to the Evidence. New York: Routledge. Mayrowetz, D., Murphy, J., Louis, K. S., and Smylie, M. (2007), Distributed leadership as work redesign: Retrofitting the job characteristics model. In Leithwood, K., Mascall, B., and Strauss, T. (Eds.), Distributed Leadership According to the Evidence. New York: Routledge. Organ, D. W. (1988). Organizational citizenship behavior, Lexington, MA: D. C. Heath. Podsakoff, P., MacKenzie, S., Moorman, R. and Fetter, R. (1990). “Transformational leader behaviors and their effects on followers’ trust in leader satisfaction and organizational citizenship behaviors.” Leadership Quarterly, 1(2), 107–142. Ross, J. A. (1998), “The antecedents and consequences of teacher efficacy.” Advances in Research in Teaching, 7, 49–73.

100

B. Mascall et al.

Ross, J. A., Hogaboam-Gray, A., and Gray, P. (2004), “Prior student achievement, collaborative school processes and collective teacher efficacy.” Leadership and Policy in Schools, 3(3), 163–188. Silins, H. and Mulford W. (2002), “Leadership and school results”. In Leithwood, K. and Hallinger, P. (Eds.), Second International Handbook of Educational Leadership. Dordrecht, the Netherlands: Kluwer Academic Press, pp. 561–612. Spillane, J. P., Camburn, E. M. and Stitziel Pareja, A. (2009). School Principals at work: A Distributed Perspective. In Leithwood, K., Mascall, B., and Strauss, T. (Eds.), Distributed Leadership According to the Evidence. New York: Routledge. Storey, J. (2004), “Changing theories of leadership and leadership development”. In Storey, J. (Ed.), Leadership in Organizations: Current Issues and Key Trends. London, UK: Routledge, pp. 11–38. Tschannen-Moran, M. (2004), Trust matters: Leadership for successful schools. San Francisco, CA: Jossey-Bass. Tschannen-Moran, M., Woolfolk Hoy, A. and Hoy, W. K. (1998), “Teacher efficacy: Its meaning and measure”, Review of Educational Research, 68(2), 202–248. Van Dyne, L. and LePine, J. A. (1998) “Helping and voice extra-role behaviors: Evidence of construct and predictive validity”, Academy of Management Journal, 41, 108–119. Wahlstrom, K. and Louis, K. (2008), “How teachers experience principal leadership: The roles of professional community, trust, efficacy and shared responsibility”. Educational Administration Quarterly, XLIV, 4, 458–495.

Chapter 5

Distributed Leadership in Schools: Does System Policy Make a Difference? Philip Hallinger and Ronald H. Heck

In a matter of a few short years, the idea of distributed leadership has evolved from a theoretical consideration of naturally-occurring social influence processes in school organization . . . to a mantra for reshaping leadership practice. More and more schools and school systems are attempting to develop distributed leadership. Increasingly, state education agencies and national education organizations are encouraging them to do it. (Louis et al., 2009)

The challenge of developing schools with the capacity for continuous improvement has led to a rapidly emerging focus on fostering leadership at all levels of the education system. Nowhere is this more evident than in the press towards distributing leadership among a broader set of key stakeholders, especially teachers, in schools (Barth, 1990; Gronn, 2002; Lambert, 2002, 2003; Spillane, 2006). Paradoxically, the latest thinking suggests that the drive to develop distributed leadership in schools neither diminishes nor comes at the expense of the principal’s responsibilities for leadership. Indeed, scholars and policymakers alike assert that principal leadership remains a key driver for change and source of support for building leadership capacity among others (e.g., Childs-Bowen et al., 2000; Gewirtz, 2003; Lambert, 2002, 2003; Murphy, 2009; Stricherz, 2001). As Mayrowitz and colleagues observe: “[P]rincipals occupy the critical space in the teacher leadership equation and center stage in the work redesign required to bring distributed leadership to life in schools.” This focus on the development of distributed, shared or collaborative leadership is especially evident in new policies and programs initiated at the school system and higher governmental levels (Barth, 1990, 2001; Clift et al., 1992; Fullan, 2006; Gronn, 2002; Harris, 2003; Lambert, 2002, 2003; Spillane, 2006; Spillane et al., 2004). In this chapter we explore the system-wide initiation of distributed leadership and its effects on school improvement. We examine these issues in the context of a state-level mandate in the USA to increase school accountability, enhance leadership capacity, and improve student learning. The specific policy context included the formation of school-community councils intended to engage a broad set of stakeholders in working with the principal to lead school-level improvements.

P. Hallinger (B) Hong Kong Institute of Education, Hong Kong, China

A. Harris (ed.), Distributed Leadership, Studies in Educational Leadership 7, C Springer Science+Business Media B.V. 2009 DOI 10.1007/978-1-4020-9737-9 6, 

101

102

P. Hallinger and R.H. Heck

Although distributed leadership has gained increasing prominence in discussions of school leadership during recent years, empirical data – especially data concerning its impact on school improvement – remain scarce. In this chapter we synthesize the results of a series of analyses of empirical data on distributed leadership and school improvement. The studies centered on the impact of new state policies that sought to create broader and deeper leadership capacity in schools as a vehicle for stimulating and sustaining school improvement. The policy initiatives underlying the research reported in this chapter reflect global trends, thereby making the study’s findings relevant to discussions of system policy, school leadership and school improvement globally.

Perspectives on School Leadership We define school improvement leadership as a process of influence by which leaders (i.e., school principals and others within the school) identify a direction for change, develop formal and informal strategies for action, and coordinate efforts towards improvements for students. For the purposes of this chapter, we use the terms collaborative, shared, and distributed leadership interchangeably to refer to leadership that is exercised by the principal along with other key staff members of the school. We undertook a series of analyses of data that described the effects of distributed leadership on efforts to upgrade the quality of the school’s learning environment, curriculum, instruction and student learning in schools. Consistent with current scholarship, we were particularly interested in understanding how leaders employed capacity-building strategies targeting school structure and culture in their efforts to improve learning (Clift et al., 1992; Copland, 2003; Fullan, 2006; Leithwood et al., 2004; Mulford, 2007; Robinson et al., 2008; Smylie et al., 2002; Stoll and Fink, 1996; Sweetland and Hoy, 2000; Walters et al., 2003; Witziers et al., 2003).

School Improvement Leadership Despite these broad-stroke conclusions concerning the central role of leadership in school improvement, there remain significant challenges in interpreting this literature (Louis et al., 1999). Indeed, we must note at the outset of this entry that despite the frequency of its use, there is no commonly accepted definition of the term “school improvement.” It has been defined variously as the successful implementation of a program, changes in teacher behavior, transformation of the school’s culture, an alteration of a school structure, or an increase in student learning or school effectiveness (Clark et al., 1984; Firestone and Corbett, 1988; Fullan, 1982; Heck and Hallinger, 1999; Leithwood, 1994; Louis et al., 1999). School improvement studies, taken as a whole, have been important in providing information about the process of improvement even if they have been less successful in documenting how improvement processes affect student outcomes (Hall and Hord, 1987; Reynolds et al., 2000).

5 Distributed Leadership in Schools

103

In addition to acknowledging the diversity of ends that define our understanding of school improvement, there is also disagreement over the means that educators may choose to facilitate improvement (Louis et al., 1999). School improvement processes have been described as complex, multi-dimensional and dependent on the relationship between the school, its community, and its cultural context (Hallinger, 1998; Hallinger and Kantamara, 2001; Hallinger and Leithwood, 1998; Meyer and Rowan, 1977; Sarason, 1982). Conceptualizations of school improvement, therefore, must go beyond adopting an innovation, changing a structure, copying the practices of successful schools, or implementing new management systems such as site-based management (Barth, 1986, 1990; Crandall et al., 1986; Cuban, 1988; Fullan, 1992; Sarason, 1982). School improvement efforts are, however, a study in contrasts. The past several decades have seen school improvement efforts organized around the adoption of a curricular program, the implementation of teacher development programs, planned school improvement and school development programs, development of schools as learning organizations, and whole school adoption of an organized set of teaching and learning practices. Despite this variety in approaches, a consistent theme emerging from research across this domain over a period of decades is the importance of leadership in facilitating improvement efforts (Berman and McLaughlin, 1978; Firestone and Corbett, 1988; Fullan, 1982, 2000, 2001; Fullan and Pomfret, 1977; Hall and Hord, 1987; Leighton, 1996; Leithwood, 1994; Leithwood et al., 2004; Leithwood and Montgomery, 1982). School leaders play a central role in initiating internal changes in schools, providing direction and support, and sustaining those changes over time by linking the internal and external environments of the school (Fullan, 2001; Meyer and Rowan, 1977). Indeed, based upon experience and research of the past five decades, it is possible to conclude that sustained school improvement in the absence of evidence of leadership is a rarity. With respect to research-based evidence, the positive contributions attributed to school leadership – specifically that of the principal – for school improvement derive from two general types of studies. The first type is cross-sectional studies of school effectiveness. The second type consists of studies of school change and school improvement projects that observed leadership as a key factor in successful school improvement projects, but which were not explicitly designed to test this as a causal variable. The most influential model for understand the instructional management role of the principal was proposed by Bossert and colleagues (1982) at the Far West Laboratory for Educational Research and Development 25 years ago (see Fig. 5.1). This model proposed four main levels of variables in seeking to understand the role of principals in school improvement: school context (including student composition), leadership, school processes, and student achievement. Moreover, although Bossert and colleagues acknowledged the complexity of modeling these relationships, the reader will note that the arrows are uni-directional. The Far West Lab model influenced the design of research on leadership for school effectiveness and improvement over the subsequent two decades in several respects. First, most of these studies focused exclusively on the principal as the

104

P. Hallinger and R.H. Heck Community and Institutional Context

Principal Leadership

School Organizational Processes

School Outcomes

Personal Characteristics

Fig. 5.1 Instructional management role of the principal (Bossert et al., 1982)

source of leadership. Second, many of the best examples of empirical investigation in the domain of school improvement leadership employed similar multi-dimensional models (e.g., Goldring and Pasternak, 1994; Hallinger et al., 1996; Heck et al., 1990; Leithwood and Jantzi, 1994, 2000; Wiley, 2001). Third, these studies tended to assume that leadership caused the effects on school outcomes. Finally, the quantitative studies tended to be cross-sectional surveys that described relationships among these variables at a single point in time (Hallinger and Heck, 1996a; Reynolds et al., 2000). As suggested above, this body of research has yielded important findings about the impact of school leadership on school improvement and school effectiveness (Bell et al., 2003; Hallinger and Heck, 1996a, b; Leithwood et al., 2004, 2006; Robinson, 2007; Witziers et al., 2003). Nonetheless, we would note that this approach to modeling the relationship of leadership to school improvement leaves unanswered important questions of causality. That is, this uni-directional model fails to model the complexity of relationships among the variables as they play out in real organizations. Moreover, cross-sectional surveys are unable to assess how school leaders actually facilitate efforts to improve student learning outcomes over time (Heck and Hallinger, 1999; Sleegers et al., 2002; Smylie and Hart, 1999).

Dynamic Models of School Improvement With these points in mind, we assert that school improvement represents a dynamic process in which the relationships among people, processes, and structures change over time (e.g., Jackson, 2000). These changes subsequently bring about changes in the state of the organization over time. Thus, the empirical study of school improvement requires the use of dynamic models that take into account the changing relationships among relevant variables (e.g., context, leadership, educational processes, and outcomes) over time. This conceptual requirement for research on school improvement has important implications for methodology. The scarcity of longitudinal data on school

5 Distributed Leadership in Schools

105

improvement, and until recently, analytic techniques with the power to examine longitudinal processes in multi-level data structures, have hindered the development of this field of research (Hallinger and Heck, 1996a, b; Heck and Hallinger, 2005). These conceptual and methodological requirements for the study of school improvement suggest the need for dynamic rather than static theories and models. Dynamic theories of organizational change seek to predict how changes in organizational structures (e.g., size, hierarchy, staffing) and social-cultural interactions (e.g., goals, organizational culture, decision-making structures, social networks) impact organizational outcomes over a period time (Langlois and Robertson, 1993; Nonaka and Toyama, 2002; Ogawa and Bossert, 1995; Williams and Podsakoff, 1989). The conceptual model depicted in Fig. 5.1 incorporates static and dynamic components of school improvement in one simultaneous model. Dynamic components (which change over time) are shaded. Static components, which represent organizational relations at a single point in time, are not shaded. Our proposed model conceptualizes school leadership as a distributed effort enacted by the principal and key teacher leaders. Distributed leadership takes place within a school context and drives a set of school conditions and educational processes aimed at improving learning outcomes for children. For the purposes of this study, these key educational processes include a sustained focus on academic improvement, stakeholder involvement in school decision-making, professional learning, student and faculty support, and open communication. The specific roles played by leaders include being catalysts for change, maintaining the improvement focus, facilitating the leadership of others, supporting instructional effectiveness, and providing tangible support for staff and students. In our proposed dynamic model, ending school status – which is defined in terms of educational processes and student academic achievement – results from changes in previous organizational conditions (i.e., educational processes and growth rates). Stated differently, where schools end in a given temporal sequence is in part a function of (1) where they begin in terms of organizational processes and achievement and (2) what they do to improve those initial conditions. In this chapter we seek to understand more about how leaders influence schools so as to impact the ending school status. Therefore, we expect relationships to exist between key ending conditions and variables that capture change over time. Dynamic models such as depicted in Fig. 5.2 take into account the importance of temporal sequences in key organizational relationships (e.g., leadership to school processes) as they vary between schools (Wilms, 1992). Thus, we call attention again to the difference between this model and the model proposed by Bossert and colleagues (1982) in Fig. 5.1. This model seeks to study not only relationships at specific points in time but the changes in relationships over time. The significance of this difference is especially important as we seek to understand the process of school improvement. We suggest that quantitative studies of school improvement must employ dynamic models with longitudinal data if we are to leverage a deeper understanding of the role that leadership plays in school improvement.

106

P. Hallinger and R.H. Heck

Fig. 5.2 Conceptual model of school improvement leadership and learning

Our Study As noted at the outset of the chapter, the impetus for this study stemmed from the implementation of a state-level policy that formalized the implementation of standards-based learning and school accountability for the state’s schools. The mandate indicated that primary accountability for school improvement continued to lie with school principals. At the same time, even as the policy proposed for principals to play a key leadership role as catalysts for change, it also formalized formal structures intended to create clear roles and responsibilities for planning and executing school improvement activities by a broader set of stakeholders (i.e., parents and teachers). State policies also fostered a normative expectation that principals would develop and validate the distribution rather than the centralization of leadership in the school. The leadership model embedded within the state policy is consistent with what we and the other authors in this volume would term collaborative, coordinating or distributed leadership. The findings that we report here derive from a series of analyses we conducted on distributed leadership within this state context. It is not the purpose of this chapter to report in-depth findings from this research. Rather it is our intention to provide a synthesis of the major findings and discuss their meaning for scholars, policymakers, and practitioners. Those who are interested in the in-depth research reports are referred elsewhere (Heck and Hallinger, 2008a, b). The research design used employed a post-hoc analysis of improvement in leadership, school processes and student learning outcomes.1 As suggested earlier, we framed the analyses in terms of dynamic models of school change and improvement and selected analytical methods that were suitable to this approach. We used multilevel growth modeling for studying the changing relationships among school

1 This chapter is not intended to convey an in-depth description of the research model and process. Instead the chapter synthesizes the main findings. Readers interested in detailed description of the research are referred to Heck and Hallinger, 2008a, b).

5 Distributed Leadership in Schools

107

context, leadership, school process and learning outcomes over a three year period of time. Data were collected from teachers and students in 200 elementary schools over a three-year period to test these relationships. The study drew a random sample (N = 13, 391) from a third-grade student cohort that was subsequently observed over a three-year period (i.e., 2004–2006). The data included:

r r r

Background data on student composition such as SES, ethnicity, English language learner etc. Data on school process inputs were collected through annual surveys from a sample of teachers at each of these schools over a three years period. The outcome data consisted of reading and math tests; notably, the data on each student’s annual results were linked to their teachers, thereby allowing for more sophisticated modeling of changes in relationships over time.

Data collected on school processes warrant additional description. These data were collected by soliciting teacher perceptions of key aspects of the school’s organization and operation that were believed to be associated with school improvement. These also represented key domains targeted by the state’s educational policies, including but not limited to distributed leadership. The school processes were as follows.

r r r r r r r r

Distributed leadership, Standards emphasis and implementation, Focused and sustained action on school improvement, Quality of student support, Professional capacity of the school, School communication, Stakeholder involvement, Student safety and well-being.

Consistent with our desire to test a dynamic model of school improvement, we also included variables measuring change in these school processes. We used a school survey to define a baseline (beginning) level of educational processes and distributed leadership for each school. The corresponding seven indicators of school process change within each school during the course of the study were developed by subtracting the percentage of agreement at year one from the percentage agreement at year three. Change in distributed leadership over the length of the study was developed in a similar manner. We proposed that increases in the level of distributed leadership and school processes should be positively associated with changes in student growth. The primary means of data analysis employed for the purposes of these studies was structural equation modeling. This approach allows us to analyze the static and dynamic portions of the model simultaneously. As suggested earlier in our discussion of dynamic modeling, the ability to describe and measure changes in relationships of variables over time is essential to the study of school improvement. SEM approaches have this capability.

108

P. Hallinger and R.H. Heck

While quantitative analysis proves useful insight into broad trends, we also appreciate the complementary benefits of more in-depth qualitative analysis. Therefore, we also draw upon follow-up case studies of a subset of 21 high-change elementary schools in the larger data base. These schools were identified based on making 20% or more growth in third grade reading proficiency levels against No Child Left Behind (NCLB) standards during the three-year period. This provided an opportunity to study in more detail what some schools did to improve school reading scores. We reasoned that schools which increased third grade reading scores probably were adopting strategies to increase reading outcomes throughout the school. We wondered whether the student cohort within this set of high-change schools would make exemplary growth in reading against other students in our study and whether stakeholders at these schools would have stronger perceptions about process changes taking place over time than stakeholders in more typical schools. We found considerable support for these propositions. First, students in these schools had a significantly higher reading growth rate of 13% per year more than students in average schools in the data base. Second, the mean level of process change in these schools was significantly higher (0.32 of a standard deviation) than the grand mean for process change in the full data base of schools.

Results This study sought to further our understanding of a range of issues related to the development and effects of distributed leadership on school improvement. We pose several questions around which we will frame our presentation of results: 1. Can state policy foster the development of school capacity for distributed leadership? 2. What are the effects of distributed leadership on school improvement processes? 3. What are the effects of distributed leadership on school learning outcomes? 4. What are the roles and effects of the principal in developing a broader and deeper capacity for distributed leadership within the school?

The Policy Context for Developing Distributed Leadership Our first question asks if state policy makes a difference in the development of school capacity for distributed leadership. The results of this research suggest that it does. Across the 200 elementary schools, there was a statistically significant increase in the perception of distributed leadership in the 200 elementary schools over the three year period. While this study did not employ an experimental design, the use of growth modeling of longitudinal data offers greater confidence than crosssectional surveys or case studies that systemic change was taking place. Ron – anything to add here?

5 Distributed Leadership in Schools

109

Impact of Distributed Leadership on School Improvement Processes The next question concerns the impact of distributed leadership on key school improvement processes. The reader will recall that our causal model proposed that changes in distributed leadership would be associated with changes in school capacity for improvement. Capacity for improvement was conceptually embedded in a set of core school improvement processes (i.e., Distributed leadership, Standards emphasis and implementation, Focused and sustained action on school improvement, Quality of student support, Professional capacity of the school, School communication, Stakeholder involvement, Student safety and well-being. The quantitative results confirmed a relationship between distributed leadership and school capacity for improvement. Distributed Leadership was significantly and strongly correlated with Ending School Educational Processes (r = 0.86) and Change in Leadership over time was significantly and moderately correlated with Change in the School Process factor (r = 0.53). These results provide support for the premise that distributed leadership is an important co-effect of school improvement processes (Heck and Hallinger, 2008b). That is, on average, as the capacity for distributed leadership increased in schools over the three-year period, so did their broader capacity to improve. This broad finding was reinforced by school portraits drawn from the qualitative data. Content analysis of school narratives found that 71% of the subset of highchange schools indicated that distributed leadership was a key factor in their focus on school improvement and their specific strategy to increase reading levels. As one school’s narrative suggested: School leadership has done the research to find the quality material to lead the curriculum reform to improve direct services to students. Material selected is supported by training in the delivery of the program. . ..The administrator has procured and provided high quality professional development for the faculty and staff. . ..There has been a continuous progression in understanding the purpose of and use of assessments. . ..Now we continue to work on more frequent and on-going assessments that affect our instruction during the lessons. (School #16, p. 46)

The roles of distributed leadership included examining student success, maintaining a focus on student learning, and creating staff ownership for results. This was emphasized in one of the narrative descriptions of change. Throughout all the school’s efforts in making student achievement gains in reading, the faculty, staff and community members have come together as a school-wide professional learning community. . ..The school leadership group. . .works together as a collaborative team to facilitate the school’s curricular/instructional programs, student support and school operations. . ..At the teacher meetings, the principal meets with the teachers to plan and develop grade level plans, discuss English Language Arts and Math and Science curriculum/instruction, analyze student work, review professional literature, and to hold student case reviews. . ..Family literacy has been the focus of parent workshops and classes. Workshops provide parents with training in supporting the growth and development of their child’s love of reading and writing. (School #6, pp. 16–17)

110

P. Hallinger and R.H. Heck

These and similar statements about the linkage between distributed leadership and improved instruction (e.g., alignment of curriculum with state benchmarks, articulation to provide instructional coherence, professional development in reading strategies and assessment) are corroborated by noting that almost 50% of these 21 schools increased 6% or more in distributed leadership over the 3-year period (against only 27% of the other schools in the full data set). We also compared the correlations between the nine process indicators for this subset of schools against the correlations within the larger data set (not tabled). Most importantly, the correlation between increased distributed leadership and increased implementation of the learning standards was much stronger in these schools (r = 0.62) compared with this relationship in the other schools (r = 0.20). This former correlation is about three times as strong as the correlation across other schools. Similarly, the correlation between increased implementation of standardsbased instruction and corresponding increases in student assessment was stronger (r = 0.58) in the subset of schools than in the full data base (r = 0.28). As a school narrative noted, Our professional development for teachers has largely focused on learning the. . .content and performance standards. We have worked hard to get a clear understanding of the appropriate student performance for each reading standard. Included in this understanding [are] the. . .assessment strategies, best practice instructional strategies and ways to provide intervention for students who struggle to meet the expectation (School #5, p 12).

The integration of quantitative and narrative data suggests in the small subset of schools leadership efforts were more strongly directed at instructional changes and corresponding improvements in assessment than the average schools in the data base. In sum, the development of capacity for distributed leadership did appear to positively impact the school’s broader capacity for improvement.

The Impact of Distributed Leadership on Student Learning Outcomes Research conducted over the past 25 years yields the conclusion that school leadership effects on student learning are largely indirect in nature. That is, leadership exercised by school principals operates through key organizational processes (Bossert et al., 1982; Hallinger and Heck, 1996a, b; Robinson, 2007; Witziers et al., 2003). Moreover, while these effects are rather small, they are both measureable, statistically significant and potentially important within the scheme of alterable school-level variables (Leithwood et al., 2004, 2006). While we have developed increasing levels of confidence in the above conclusion, similar evidence concerning the effects of distributed leadership remains scarce (e.g., Marks and Printy, 2003; Mulford and Silins, 2003). The present research seeks to add to this knowledge base. The findings reported above confirmed that distributed leadership was significantly linked to important school processes. These school improvement processes

5 Distributed Leadership in Schools

111

were, in turn, significantly correlated with ending levels of reading and math achievement levels. This finding of indirect effects of distributed leadership on important school improvement process variables and learning outcomes was confirmed by further analyses conducted in the dynamic portion of the model. That is, Changes in Distributed Leadership were significantly related to Changes in School Processes which were positively related to Growth in Student Reading and Math Achievement (Heck and Hallinger, 2008a, b). Thus, both sets of analyses confirm that Collaborative Leadership was directly related to important school improvement process variables and indirectly to math achievement outcomes. While the size of these indirect effects of leadership on learning outcomes may be considered small, the effects are potentially meaningful. This dual approach to analysis establishes greater credibility in the proposed linkages between collaborative leadership, important school-level variables and student learning outcomes.

The Roles and Effects of Principal Leadership The last question posed for the purposes of this chapter centered on the role and effects of the principal in school environments where there is increasing emphasis on the distribution of leadership. We should note that the nature of our data set limited the ability to address this question directly. We did not have a separate measure of the principal’s leadership. Therefore, we approach this issue indirectly by examining the effects of principal turnover in the quantitative and qualitative data sets (Heck and Hallinger, 2008a). Almost half the schools had the same principal over the 3-year period (against 26% of the other schools). As one school noted, “The stability of our administration and teaching staff was a critical factor helping to mitigate the enormous requirements and challenges imposed on the school system by both the federal and state Department of Education” (School #4, p. 9). Nearly two-thirds implemented a school-level structural change to support reading (e.g., setting aside time blocks during the day, looping younger students with the same teachers for two consecutive years). About 75% of the schools also mentioned considerable success with school-wide efforts to involve parents in their children’s reading progress. Some of the strategies used included holding school meetings and classes, having parents read in evenings with their children, having student logs for nightly reading signed by parents, and increasing home-school communication. These types of school-level changes require a certain level of stability in school-level leadership in order to be sustained over time (Hall and Hord, 2001). More broadly, we observed that principal stability (i.e., having the same principal over the three-year period) had a small but statistically significant effect on Ending School Processes. Having the same principal in the school was also positively correlated with stakeholder perceptions concerning the presence of distributed leadership.

112

P. Hallinger and R.H. Heck

Stated differently, stakeholder perceptions of distributed leadership in the school at the end of the three years were significantly higher in schools where the same principal had been present over the three years of the study. When analyzing the impact of principal stability in terms of changes in the school rather than at the end point alone, an interesting dynamic emerged. We found that while principal stability was not directly related to changes in school processes, there was an interesting relationship between the school context, principal stability and on changes in school processes. More specifically, in schools with more challenging contexts (i.e., greater percentages of low SES students, minority students, and students receiving English language services), principal stability had a significant relationship with positive changes in school processes. The same significant interaction effect was not observed across the sample as a whole. While this finding was somewhat unexpected, it is potentially important.

Conclusions This chapter explored the role of distributed leadership in school improvement. We employed three main approaches to data analysis to address these questions. First, we modeled the effects of the school context, distributed leadership, and key school improvement processes on reading and math achievement over three years.2 Second, we modeled the change in these key variables with corresponding changes in achievement year-by year over the same three-year period. Third, we analyzed qualitative data on a subset of high change schools that improved beyond the level of other schools in reading outcomes. The results support the view that distributed school leadership and a set of key educational processes are related to school improvement in several ways that are consistent with the proposed theoretical model. First we note that the Ending School Process factor was significantly related to Ending Math Achievement. Second, the Change in Process factor had significant effects on Student Growth in Math. With respect to the Change in Process factor, we noted that two components in particular – Professional Capacity and Sustained Focus on Improvement – contributed the most in terms of explaining student growth in math achievement. These findings suggest the identification of a rather robust set of policy-driven, school-level factors that are related to student achievement in meaningful ways. Up until the turn of the 21st century, most research on school leadership focused on the principal. In the subsequent years, scholars have made a logical case for distributing leadership more broadly in schools (Gronn, 2002; Harris, 2003; Spillane, 2006, Spillane et al., 2004). Researchers have also begun to conduct empirical research that draws upon conceptualizations of leadership as a distributed process 2 Although this report focused solely on the results for math outcomes, we noted that results for reading outcomes were remarkably similar for the model’s theoretical components. That is the results for reading achievement were very similar concerning the relationships among Distributed Leadership, key School Processes, and Reading Achievement.

5 Distributed Leadership in Schools

113

in schools (e. g., Marks and Printy, 2003; School of Education and Social Policy, 2004). In this study we sought to extend the knowledge base on collaborative or distributed leadership through an empirical study of a live policy initiative designed to foster more distributed leadership in schools. Our report has focused on understanding both the effects of distributed leadership on school improvement and the role played by principals in an environment where there is an explicit expectation that leadership responsibilities will be shared. Although our findings are bounded by limitations inherent in the techniques employed we offer several implications relevant to researchers, policymakers and practitioners. First, with respect to researchers in the domains of leadership effects and school improvement, our study demonstrates the utility of longitudinal studies. Indeed, the robustness of our results would no doubt be strengthened by several additional waves of data in our time series. Nonetheless, the ability to test static and dynamic models simultaneously represents a significant advance that provides a foundation for future research in this area. Notably, only the state education department’s willingness to cooperate with research made this longitudinal approach possible. We should also take note of how this research has altered our own view towards the study of leadership processes in schools. Publication of a series of influential reviews of research on school leadership in the early 1980s (Bossert et al., 1982; Bridges, 1982; Leithwood and Montgomery, 1982; Pitner, 1988), gave impetus to the more systematic empirical study of school leadership effects. Subsequent reviewers have suggested that progress has been made at identifying the and specifying the indirect nature of principal leadership effects (e.g., Hallinger and Heck, 1996a, b, 1998; Leithwood et al., 2004; Robinson, 2007; Witziers et al., 2003). Yet, nagging problems have remained. For example, while the vast majority of leadership studies in education focused on the principal, we must acknowledge that the reality of leading schools requires a broader set of leadership resources. It may be the case, that some of the “nagging problems” that have accompanied studies of school leadership effects arise from the fact that we have only been measuring an important but incomplete portion of the school’s leadership resources. Thus, future research would do well to assess the contribution of leadership contributed by the principal as well as by other key stakeholders. Second, with respect to policy, the research begins to validate the viability of a set of key educational processes that can be linked to school improvement. More specifically, the research supports the strategy advocated by Fullan (2006) that aims to build professional and leadership capacity in schools. This study adds to a small but growing body of empirical research that finds positive effects of collaborative or distributed leadership on school improvement processes and outcomes. While the finding on principal stability awaits verification through more explicit study and analysis, it should nonetheless be of interest to policymakers who manage the selection and assignment of principals to schools. Finally, with respect to practice, this research should give encouragement to superintendents, principals and teachers. Murphy asserts:

114

P. Hallinger and R.H. Heck

In some real sense, at the school level all change flows through the principal’s office . . . That is, principals occupy the critical space in the teacher leadership equation and center stage in the work redesign required to bring distributed leadership to life in schools. (forthcoming)

Our findings provide tentative empirical support. The implementation of policies designed to foster distributed leadership do not appear to lessen the importance of the principal’s own leadership role. The task of building professional capacity and distributed leadership requires principal support (Childs-Bowen et al., 2000; Copland, 2003; Murphy, forthcoming; Spillane, 2006). Principals and teachers both have important, though perhaps different, roles in leading school improvement. Although the nature of these differences needs to be investigated further, our results suggest that principal leadership remains a key success factor in school improvement, especially in contexts where the challenges are greatest.

References Barth, R. (1986). On sheep and goats and school reform. Phi Delta Kappan, 68(4), 293–296. Barth, R. (1990). Improving schools from within. San Francisco: Jossey-Bass. Barth, R. (2001, February). Teacher Leader. Phi Delta Kappan, 443–449. Bell, L., Bolam, R., and Cubillo, L. (2003). A systematic review of the impact of schoolheadteachers and principals on student outcomes. London: EPPI-Centre, Social Science Research Unit, Institute of Education. Berman, P. and Mclaughlin, M. (1978). Federal programs supporting educational change (Vol. 3). Santa Monica, CA: Rand Corporation. Bossert, S., Dwyer, D., Rowan, B., and Lee, G. (1982). The instructional management role of the principal. Educational Administration Quarterly, 18(3), 34–64. Bridges, E. (1982). Research on the school administrator: The state-of-the-art, 1967–1980. Educational Administration Quarterly, 18(3), 12–33. Childs-Bowen, D., Moller, G., and Scrivner, J. (2000, May). Principals: Leaders of leaders. NASSP Bulletin, 84(616), 27–34. Clark, D., Lotto, L., and Astuto, T. (1984). Effective schools and school improvement: A comparative analysis of two lines of inquiry. Educational Administration Quarterly. 20 (3), 41–68. Clift, R., Johnson, M., Holland, P., and Veal, M. (1992). Developing the potential for collaborative school leadership. American Educational Research Journal, 29(4), 877–908. Copland, M. A. (2003). Building the capacity to lead: Promoting and sustaining change in an inquiry-based model of school reform. Draft copy of a manuscript (pp. 1–34) that later appeared. In J. Murphy and A. Datnow (Eds.). Leadership for school reform: Lessons from comprehensive school reform designs (pp. 159–183). Thousand Oaks, CA: Corwin Press. Crandall, D., Eiseman, J., and Louis, K. S. (1986). Strategic planning issues that bear on the success of school improvement efforts. Educational Administration Quarterly, 22(3), 21–53. Cuban. L. (1988). The managerial imperative and the practice of leadership in schools. Albany, NY: SUNY Press. Firestone, W. A. and Corbett, H. D. (1988). Planned organizational change. In N. Boyan (Ed.), Handbook of Research on Educational Administration (pp. 321–340). New York: Longman. Fullan, M. (1982). The meaning of educational change. New York: Teachers College Press. Fullan, M. (2001). Leading in a culture of change. San Francisco, CA: Jossey-Bass Inc. Fullan, M. (2006). The development of transformational leaders for educational decentralization. Toronto, Canada: Michael Fullan. Fullan, M. and Pomfret, A. (1977). Research on curriculum and instruction implementation. Review of Educational Research, 47(1), 335–397.

5 Distributed Leadership in Schools

115

Gewertz, C. (2003, Jan. 8). N. Y. C. Chancellor Aims Bolster Instructional Leadership. Education Week, 22 (16), 7, 12. Goldring, E. B., and Pasternak, R. (1994). Principals’ coordinating strategies and school effectiveness. School Effectiveness & School Improvement, 5, 237–251. Gronn, P. (2002). Distributed leadership as a unit of analysis. Leadership Quarterly 13, 423–451. Hall, G. and Hord, S. (1987). Change in schools. Albany, NY: SUNY Press. Hall, G. and Hord, S. (2001). Implementing change: Patterns, principles, and potholes. Boston: Allyn & Bacon. Hallinger, P., Bickman, L., and Davis, K. (1996). School context, principal leadership and student achievement. Elementary School Journal, 96(5), 498–518. Hallinger, P. and Heck, R. H. (1996a). The principal’s role in school effectiveness: An assessment of methodological progress, 1980–1995 (723–783). In K. Leithwood, J. Chapman, D. Corson, P. Hallinger, and A. Hart (Eds.), International handbook of educational leadership and administration. Dordrecht, Netherlands: Klewer Academic Publishers. Hallinger, P., and Heck, R. H. (1996b).Reassessing the principal’s role in school effectiveness: A review of the empirical research, 1980–1995. Educational Administration Quarterly, 32(1), 5–44. Hallinger, P. (1998). Educational change in Southeast Asia: The challenge of creating learning systems. Journal of Educational Administration, 36(5), 492–509. Hallinger, P. and Kantamara, P. (2001). Learning to lead global changes across cultures: Designing a computer-based simulation for Thai school leaders. Journal of Educational Administration, 39 (3), 197–220. Harris, A. (2003) ‘Teacher leadership as distributed leadership: heresy, fantasy or possibility?’ School Leadership and Management, 23(3), 313–324. Heck, R. H. and Hallinger, P. (1999). Next generation methods for the study of leadership and school improvement. In J. Murphy and K. S. Louis (Eds.) Handbook of research on educational administration (2nd Edition, pp. 141–162). San Francisco: Jossey-Bass. Heck, R. H. and Hallinger, P. (2005). The study of educational leadership and management: Where does the field stand today? Educational Management, Administration & Leadership, 33(2), 229–244. Heck, R. H. and Hallinger, P. (2008a). Distributed leadership effects on school improvement: Testing a reciprocal effects model. Paper prepared for presentation at the annual meeting of the American Educational research Association, San Diego, CA. Heck, R. H. and Hallinger, P. (2008b). Testing a dynamic model of organizational leadership and school improvement. Paper prepared for presentation at the annual meeting of the American Educational research Association, San Diego, CA. Heck, R., Larson, T., and Marcoulides, G. (1990). Principal instructional leadership and school achievement: Validation of a causal model. Educational Administration Quarterly, 26, 94–125. Jackson, D. (2000). The school improvement journey: Perspectives on leadership. School Leadership & Management, 20(1), 61–78. Lambert, L. (2002). A framework for shared leadership. Educational Leadership, 59(8), 37–40. Lambert, L. (2003). Leadership capacity for lasting school improvement. Alexandria, VA: Association of Supervision and Curriculum Development. Langlois, R. and Robertson, P. L. (1993). Business organisation as a coordination problem: toward a dynamic theory of the boundaries of the firm, Business and Economic History, 22(1), 31–41. Leighton, M. S. (1996). The role of leadership in sustaining school reform: Voices from the field. Washington, DC: US Department of Education. Leithwood, K. (1994). Leadership for school restructuring. Educational Administration Quarterly, 30(4), 498–518. Leithwood, K., Day, C., Sammons, P., Harris, A., and Hopkins, D. (2006). Seven strong claims about successful school leadership. Nottingham, England: National College of School Leadership.

116

P. Hallinger and R.H. Heck

Leithwood, K. and Jantzi, D. (2000). The effects of transformational leadership on organizational conditions and student engagement with the school. Journal of Educational Administration, 38(2),112–129. Leithwood, K., Louis, K. S., Anderson, S. and Wahlsttom, K. (2004). Review of research: How leadership influences student learning. Wallace Foundation. Downloaded from http://www.wallacefoundation.org/NR/rdonlyres/E3BCCFA5-A88B-45D3-8E27B973732283C9/0/ReviewofResearchLearningFromLeadership.pdf on December 19, 2007. Leithwood, K., and Menzies, T. (1998). Forms and effects of school based management: A review. Educational Policy, 12 (3), 325–346. Leithwood, K., and Montgomery, D. (1982). The role of the elementary principal in program improvement. Review of Educational Research, 52(3), 309–339. Louis, K. S., Mayrowetz, D., Smylie, M., and Murphy, J. (2009). The role of sensemaking and trust in developing distributed leadership. In A Harris (Ed.), Distributed leadership: Different perspectives. Netherlands: Springer. Louis, K. S., Toole, J., and Hargreaves, A. (1999). Rethinking school improvement. In J. Murphy and K. Seashore Louis (Eds.) Handbook of research on educational administration (2nd edition). San Francisco: Jossey-Bass, 251–276. Marks, H., and Printy, S. (2003). Principal leadership and school performance: An integration of transformation and instructional leadership. Educational Administration Quarterly, 39(3), 370–397. Meyer, J. and Rowan, B. (1977). Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology 83 (2), 340–363. Mulford, B. and Silins, H. (2003). Leadership for organisational learning and improved student outcomes – What do we know? Cambridge Journal of Education, 33(2), 175–195. Mulford, B. (2007). Building social capital in professional learning communities: Importance, challenges and a way forward. In L. Stoll & K. Seashore Louis (Eds.). Professional learning communities: Divergence, depth and dilemmas (pp. 166–180). London: Open University Press and McGraw Hill. Murphy, J. (forthcoming). Administrative leadership. In D. Mayrowetz, M. Smylie, K.S. Louis, & J. Murphy (Eds.). No Child Left Behind Act of 2001. (2001). Public Law No. 107–110. Washington D.C. Nonaka, I. and Toyama, R. (2002). A firm as a dialectical being: Towards a dynamic theory of a firm. Industrial and Corporate Change, 11(5), 995–1009. Ogawa, R., and Bossert, S. (1995). Leadership as an organizational quality. Educational Administration Quarterly, 31(2), 224–243. Pitner, N. (1988). The study of administrator effects and effectiveness. In N. Boyan (Ed.), Handbook of research in educational administration. New York: Longman. Reynolds, D., Teddlie, C., Hopkins, D., and Stringfield, S. (2000). Linking school effectiveness and school improvement. In C. Teddlie and D. Reynolds (Eds.). The international handbook of school effectiveness research (pp. 206–231). London: Falmer Press. Robinson, V. (2007). School leadership and student outcomes: Identifying what works and why. Melbourne: Australian Council for Educational Leaders Monograph No. 41. Robinson, V., Lloyd, C., and Rowe, K. (2008). The Impact of leadership on student outcomes: An analysis of the differential effects of leadership types. Educational Administration Quarterly, 44(5) 635–674. Sarason, S. B. (1982). Culture of the school and the problem of change, 2nd edition. Boston, MA: Allyn & Bacon. School of Education and Social Policy. (2004). The distributed leadership study. Chicago: Northwestern University. Available at: http://dls.sesp.northwestern.edu Sleegers, P., Geijsel, F., and van den Berg, R. (2002). Conditions fostering educational change. In K. Leithwood and P. Hallinger (Eds.) Second international handbook of educational leadership and administration (pp. 75–102). Dordrecht: Netherlands: Kluwer Academic Publishers.

5 Distributed Leadership in Schools

117

Smylie, M., Conley, S., and Marks, H. (2002). Exploring new approaches to teacher leadership for school improvement. In J. Murphy (Ed.), The educational leadership challenge: Redefining leadership for the 21st century (pp. 162–188). Chicago: University of Chicago Press. Smylie, M., and Hart, A. (1999). School leadership for teacher learning and change: A human and social capital development perspective. In J. Murphy and K. S. Louis (Eds.), Handbook of research on educational administration (2nd edition, pp. 421–441). San Francisco: Jossey-Bass. Spillane, J. (2006). Distributed leadership, San Francisco: Jossey-Bass. Spillane, J., Halverson, R., and Diamond, J. (2004). Towards a theory of leadership practice: A distributed perspective. Journal of Curriculum Studies 36(1): 3–34. Stoll, L., and Fink, D. (1996). Changing our schools. London: Open University Press. Stricherz, M. (2001, Sept. 12). D.C. Principal’s training designed to boost instructional leadership. Education Week, 21(2), 13. Sweetland, S., and Hoy, W. (2000). School characteristics and educational outcomes: Towards an organisational model of student achievement in middle schools. Educational Administration Quarterly, 36 (5), 703–729. Walters, T., Marzano, R., and McNulty, B. (2003). Balanced leadership: What 30 years of research tells us about the effect of leadership on student achievement. Denver: Mid Continental Regional Educational Laboratory. Wiley, S. (2001). Contextual effects on student achievement: School leadership and professional community. Journal of Educational Change, 2(1), 1–33. Wilms, D. (1992). Monitoring school performance. London: Falmer. Williams, L., and Podsakoff, P. (1989). Longitudinal field methods for studying reciprocal relationships in organizational behavior research: Toward improved causal analysis. Research in Organizational Behavior, 11, 247–292. Witziers, B., Bosker, R., and Kruger, M. (2003). Educational leadership and student achievement: The elusive search for an association. Educational Administration Quarterly, 39, 398–425.

Part III

Practical Perspective

Chapter 6

Capacity Building Through Layered Leadership: Sustaining the Turnaround Christopher Day

Introduction Cultures do not change by mandate; they change by the specific replacements of existing norms, structures and processes by others; the process of cultural change depends fundamentally on modelling the new values and behaviour that you expect to displace the existing ones. (Elmore, cited in Fullan, 2006: 57a)

The literature on school improvement has from some years now been replete with good advice on change: it cannot be mandated (yet policy makers continue to do so); best practice must be modelled by school leaders (yet it often is not); existing, unacceptable, norms, structures and processes must be displaced and replaced by those which more accurately reflect the school values and moral and ethical purposes (difficult to achieve, especially in large schools); system change needs to be in the early phases of improvement; and, as the school moves forward on its improvement journey, they will be replaced by others. In recent years, also, there have been observations by those who are wise and knowledgeable of teacher and school improvement, observing that the huge financial investments by governments in “top-down” models of systemic change have given little return. They identify the need to, “reframe our entire reform strategy so that it focuses relentlessly and deeply on capacity building and accountability” (Fullan, 2006: 28). What is important this and other similar calls is that not only do they emphasis the need to bring together internal and external change agendas but that they recognise that change – real change that is sustained – involves everyone in the school, that it takes time and that, whilst there may be many leaders in a school, the principal is the key to bringing about and sustaining successful change. However, the successful principal knows that for change to work means that all – students, teachers, non-teaching staff and parents – must have a sense of participation in and ownership of the process and the product. Only then will commitment be gained; and without commitment, sustaining change will be difficult if not impossible.

C. Day (B) TLRC, School of Education, The Dearing Building, The University of Nottingham Jubilee Campus, Wollaton Road, Nottingham, NG8 1BB, UK

A. Harris (ed.), Distributed Leadership, Studies in Educational Leadership 7, C Springer Science+Business Media B.V. 2009 DOI 10.1007/978-1-4020-9737-9 7, 

121

122

C. Day

Whilst there are a number of conditions associated with turning around schools which are failing their students and their wider communities, for example the appointment of a new principal, the injection of additional resources or “re-badging” (Ansell, 2004), there are also a number of warnings. The new leader’s agenda may be too daunting; “quick-fix” changes may lead to temporary (and, therefore, false) recoveries which cannot be sustained; and these often exhaust the staff (and principal) rather than re-energizing them (Kanter, 2004; Hargreaves and Fink, 2006). This may be the case, especially when change processes do not acknowledge people’s emotions, (Hargreaves, 1998), vulnerabilities (Kelchtermans, 1996) or values (Sockett, 1993; Jackson et al., 1993); when they do not seek to renew their passion for teaching (Day, 2004); or when they fail to provide the training and development opportunities and ongoing support necessary for staff to learn new skills and acquire new knowledge – in other words, when they fail to build the individual and collective power of the whole staff to improve student well-being and achievement school wide, otherwise known as “capacity building”. Capacity building involves the use of strategies that increase the collective effectiveness of all levels of the system in developing and mobilizing knowledge, resources and motivation, all of which are needed to raise the bar and close the gap of student learning across the system. (Fullan et al., 2006: 88)

It is one thing diagnosing the problem and identifying strategies which are designed to enable successful, sustained change at different levels. It is quite another to achieve success in the complex worlds of schools and classrooms. Indeed, despite the plethora of good research and experience generated advice such as that listed above, if we observe studies of the effects of new public management agenda and the accountability driven systems that accompany this, it would appear that we know little about the performance of “turnaround” schools in the longer term i.e. whether change is sustained. A recent report by the National Audit Office in England, for example, noted: There is little evidence available about the performance of recovered schools in the longer term. (2006:14)

Nor, it appears, can we readily find the skills and talents necessary to effect the sustained improvement of schools which, in American terms, are “on probation”: Work motivation and commitment. . .[of staff]. . .to stay were strongly related to principal leadership, collegiality and perceived skills of colleagues. We found these skills and talents in short supply across the studied schools. . . (Milnthrop, 2004: 156)

This chapter provides an example of the work of one elementary school principal in her first principalship whose staff demonstrate such work motivation and commitment in abundance over the longer term. It explores how, over a seven year period, this principal turned around a school which was under threat of closure when she was appointed to the job. She is not “charismatic” or a “hero” in the traditional sense in which these terms are often used in the literature on school principalship. However, she does have a resolute sense of moral purpose, her focus is upon both the instrumental and the personal, she is herself a good classroom teacher and is brave enough to model this for others, and she possesses a range of intra and interpersonal qualities.

6 Capacity Building Through Layered Leadership

123

The Context Oasis Primary and Nursery School is located in an inner city area of extreme social and economic deprivation. 71% of the pupils live in the most deprived 1 and 96% in the most deprived 5% of areas nationally. Household income is in the lowest 10% nationally and 80% of pupils are eligible for free school meals. The crime and disorder statistics are in the worst 3% nationally. Armed police patrol the streets because of the high number of shootings and there is a significant drug problem. In the past the school has had problems with hypodermic needles being buried “needle” up in the school grounds. A number of pupils aged 7–11 are involved in drug dealing for street gangs at night. Pupils tell the school that many spend their school holidays in the City Centre. Few parents have a history of accessing further education. There are 169 pupils in this smaller than average school. Pupil mobility is high, with 42% not completed the full four year education programme in 2005. 40% of pupils have social services involvement and 35% are classified as having special educational needs (SEN). The school has an increasing number of pupils whose parents are asylum seekers and many of these are not fluent English speakers. There are 13 different languages spoken and the proportion of pupils from minority ethnic backgrounds is almost three times higher than the national average. 80% of pupils come from one-parent families and the same percentage are eligible for free school meals. Attainment on entry is well below average. There are 7 teachers and 12 non-teaching staff. A key characteristic of the school is the quality of the physical environment. Whereas externally it appears to be “just another”, 1960s, concrete and glass, flat roofed building set in a small green field, on entry it is transformed into a fully carpeted series of areas with light pinewood doors, calico blinds on windows, art work and other stimuli on walls, a staff room which would grace any good class apartment, and internal/outdoor decking areas for children and staff. In other words, it is an environment which is in direct contrast to those from which the pupils come. ICT in the forms of a dedicated area, laptops and video screens is an integral part of the purposeful learning environments which constitute the school. Active rather than passive learning is encouraged and there is an emphasis upon values education together with “educating for excellence” and upon extending pupils’ horizons by promoting world citizenship, for example through a school-led Comenius project with schools in Poland and Italy. Yet only seven years previously, the school was on the verge of being closed, placed in “special measures” by external inspectors; staff morale was low, classroom resources scarce, and pupil achievement and self esteem rock bottom.

The Principal The principal (Diana) has been described as having achieved, “remarkable success” . . .and her school is described as being, “incomparably better for the years you have been there” (Urban Leaders Assessment Centre, 2005). Previously she worked in

124

C. Day

inner city schools in London and her present city before taking up the headship of this, “derelict, unpromising and challenging” school. In order to reach the level of success that has been achieved, she had to overcome the “embedded pessimism” of staff, quickly deal with two staff whose teaching was consistently poor and transform the ethos and environment. A key strength of Diana’s leadership is the generation of a shared vision for the education of the pupils and this is accompanied by close working relationships with parents, governors, external agencies. At Oasis, one tangible result of the new ways forward has been the rise in Key Stage 1 and 2 national test results of 7 and 11 year old students in Literacy, Numeracy and Science. Tables 6.1 and 6.2 show clearly what a difference this head and her staff made. Over the period of her leadership results improved in KS1 Mathematics from 24 to 95%; and in KS2 from 7 to 81%, in KS1 Science from 41 to 89% and in KS2 from 11 to 94%; and in KS1 Literacy from 31% (writing) and 28% (reading) to 79 and 84% respectively and in KS2 from 30 to 81%. It is important to note that whilst the persistence of external challenges remained constant, the nature of internal challenges – over which Diana had more direct

Table 6.1 Key stage 1 SAT results 1998 1999 2000 2001 2002 2003 2004 2005 2006 Maths % pupils reaching level 2 in Maths % pupils reaching level 3 in Maths Overall % of pupils reaching level 2+ in Maths Writing % of pupils reaching level 2 in Writing % of pupils reaching level 3 in Writing Overall % of pupils reaching level 2+ in Writing Reading % of pupils reaching level 2 in Reading % of pupils reaching level 3 in Reading Overall % of pupils reaching level 2+ in Reading Science teacher assessment % of pupils reaching level 2 in Science % of pupils reaching level 3 in Science Overall % of pupils reaching level 2+ in Science

24 0 24

46 19 65

39 33 72

77 14 91

74 22 96

79 21 100

79 11 90

85 4 89

95 0 95

31

46

77

76

74

69

74

58

79

0

0

6

10

4

11

0

0

0

31

46

83

86

78

80

74

58

79

21

48

29

62

61

64

63

80

63

7

13

38

14

4

16

11

12

21

28

61

67

76

65

80

74

92

84

41

51

50

81

56

68

79

73

84

0

7

17

5

35

32

11

19

5

41

58

67

86

91

100

90

92

89

6 Capacity Building Through Layered Leadership

125

Table 6.2 Key stage 2 SAT results 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Maths % pupils reaching level 4 in Maths % pupils reaching level 5 in Maths Overall % of pupils reaching level 4+ in Maths

7

33

63

70

61

74

43

63

68

50

0

0

8

15

17

0

14

25

16

31

7

33

71

85

78

74

57

88

84

81

English % of pupils reaching level 4 in English % of pupils reaching level 5 in English Overall % of pupils reaching level 4+ in English

30

44

46

63

61

45

35

67

64

18

0

0

4

11

17

19

29

0

12

63

30

44

50

74

78

64

64

67

76

81

Science % of pupils reaching level 4 in Science % of pupils reaching level 5 in Science Overall % of pupils reaching level 4+ in Science

11

44

46

63

61

45

35

67

32

44

0

0

71

56

22

13

36

25

68

50∗

11

44

96

96

100

90

86

92

100

94

influence – did not. Diana’s response was to manage both in tandem over a number of development phases: I had to go into everything to make sure that things are moving forward. . . You have to be realistic about the pace of improvement. . .Initially, it was a constant ‘hands on’ approach. I did not leave the school for a year. I was wrapped up in the school, couldn’t think beyond it.

Four Phases of Development Development Phase 1 (1999–2000) “Coming out of Special Measures” When Diana was appointed, the school was in “Special Measures”. This is a category of major weakness as judged by external inspection by a team of independent inspectors employed by government in England as part of its system for monitoring school standards. Schools in this category are given two years to improve or face closure. In a sense, then, Diana’s phase one improvement agenda was set. She had at the very least to i) raise the standards of teaching from “unsatisfactory” to “good or very good”; ii) reduce student absenteeism and lateness to lessons; iii) raise the test results levels of students who experienced poor reading and writing

126

C. Day

home cultures. In doing so, she had to address low aspirations and expectations of achievement held by staff, parents and students themselves. Moreover, she had to do so in an area in which, “the high crime rate and gang culture has an effect on some pupils’ performance as they become emotionally traumatized after witnessing or being involved in crime”, and in a school which provided an impoverished setting for teaching and learning. On her first day in the school she set her expectations: I talked a lot about my beliefs and expectations. . .I acknowledged that I had just come out of the classroom. . .So I knew what it was like, and that there are days when some people work better at different times and that there would be flexibility as long as the job’s done. . .I talked about working habits. . .and I brought some examples of the standards of work that I would expecting. . .I talked about the ethos of displays as part of the classroom learning environment. . . (HT)

Five First Steps i) Enriching the teaching and learning environment Diana redecorated the school building, including reshaping the interior so that it became a welcoming, rich environment (and provided a direct contrast to the external worlds of the students) with quality resources in classrooms. She prioritized the environment of the school – the feel good factor. (T1)

She transformed the school into a learning “oasis”, with brightly coloured furniture, carpeted corridors, walls adorned with photographs and children’s’ work clearly mounted, inner courtyards and a staffroom which looked more like a well appointed business lounge. As Corrie Giles (2007) observes “in schools where order had or could easily break down, effective management of the learning environment was a crucial first step in winning the support of teachers and parents” (p. 36) ii) Providing security The school site was secured, used hypodermics littering the field became a forgotten nightmare. Investment in a child friendly playing area was made (with consultation with staff and students) and the entrance into the school itself was both secured and made more friendly for parents and other visitors through the provision of a chaired, carpeted area. Diana built close, regular meetings with the police so that the knowledge gained would help understand the behaviour of certain students. He was in trouble with the police (because) he threw fireworks in a window.

iii) Establishing a Student Behaviour Policy and Improving Attendance Prior to Diana’s arrival, according to parents, children spent more time out of their classrooms than in. She introduced a positive behaviour system, insisting upon standards of talking by teachers to children, classroom organisation (particularly routines for the beginning and ending of lessons), and behaviour in corridors, playground and classrooms. There’s so much in terms of reward. . .we’ve got the ‘smiley faces’, the marble tree, parents are brought in for discussion if we’ve got a problem with their child. They are

6 Capacity Building Through Layered Leadership

127

also phoned and brought in if their child has done extremely well, or we will contact the parents to say there child has had a good day. We push the positive. (T) The children were involved in the rewards for the behaviour system. We were involved also, and it worked because people feel important because they have their say. (T)

iv) Community Acceptance It was really important that [head] fitted. . .she came under scrutiny to start with. . .but good relationships with the community developed because she listened. . .some of them would just turn up without an appointment, and expect to be seen. . .and she will always make time. And we set up a ‘Harmony Club’. They just talk to [head], tell her about anything that’s happening in the area, like how many drug raids. . .and they feel like they can talk to [head] and they come in and help out in the school. . .And, also, when we had Harvest Festival, we’ve got quite a nice relationship with the elderly in the bungalows. Most of them are disabled, and the children take them parcels. . .we’ve got good relationships with the community, police and the road safety people. . . (A)

Diana invested considerable time in talking with and – wherever possible – visiting parents in order to increase interest in attendance. However: No matter what you do, pupil attendance remains a constant challenge.

Many parents did not value education due to their own negative school experiences. Many had poor parenting skills and few had accessed formal education beyond school. v) Improving teaching and learning in classrooms with staff and students: leading by example Because the standards of work in the classrooms had been judged unsatisfactory, Diana had made this a priority: I discussed expectations with children, parents and staff in the first days of headship. . .I talked a lot about my beliefs and expectations. The children are educated for excellence. . .there’s no reason why any child at all can’t achieve, and I think that’s what she’s striving for, that’s really important. . .She’s firm when she has to be firm. . .If something’s not right, or something’s wrong, then you will be told (but) there’s no bullying involved. . .you’re told in a way that you’re offered support as well. (A)

Early on, she worked with staff to created detailed job descriptions so that everyone was clear about their roles and responsibilities. More importantly, she went into classrooms to observe and feed back upon teaching and to work with students, herself being observed by the teachers, “to show staff what they can do and what’s expected”. She also modelled high standards of the display for students’ work, working with the students to, “show the staff that they can do it”, and that, “I expect them to do the same”. She identified, also, areas for the development of individual staff: Everybody’s got strengths and weaknesses, and sometimes the strengths go unnoticed. But [head] seems to pick up on that. . .I was able to do a course about children with

128

C. Day autism. I didn’t think I could do that, but [head] said, ‘Would you be interested?’ It’s not a bullying tactic. It’s trying to get people motivated in the nicest possible way, without telling them what to do. (A)

These five steps were taken as a school and accompanied by sustained discussion and consultations with key stakeholders – staff, parents, community groups and the local authority-through a planned programme of formal meetings and many informal, one-to-one conversations, as the groundwork for further development was being laid. Everybody feels they have a part to play, no matter how small. I think the general feeling is happiness. And people feel valued. . .[head] is quick to pick up on things that aren’t going well and they will be sorted.

It would be a mistake, however, to conclude that there were no “bumpy moments” in these early stages of the journey to success. Two teaching staff were encouraged to resign, other teaching and non-teaching staff who had lost their motivation under the previous regimes had to be won over. Focussing on continuing professional development for all staff, principally through organising a range of in-house training to improve the quality of teaching and keep knowledge, skills and practices up-to-date was, and remains, a key developmental strategy. But it was the combination of external imperatives to improve together with modelling high standards of teaching and learning and creating the internal conditions (security, behaviour and the physical environment) for students and staff to do their best which were the keys to early success.

Phase 2 Taking Ownership: An Inclusive Agenda (2000–2002) With the school successful in being released from the “special measures” category in 2000, initiatives which had been began in Phase 1 began to yield dividends. The school achieved “Investors in People” (IIP) status in 2001 – a recognition of the investment Diana had made in the continuing development of all staff. It is easy to forget, also, that year on year improvement of the national test results vindicated Diana’s strategies and acted as rewards to staff and students for their work, contributing incrementally to increased motivation, commitment and sense of recognition of achievement. Two of the teachers for whom teaching at the school previously had changed significantly – an inexperienced, first year teacher who had been failing, now was succeeding; and the most experienced teacher in the school who, prior to Diana’s appointment, had been taking extended periods of stress-related leave, now was taking an active senior leadership role in the school, with renewed motivation and commitment. Prior to [head] I had two or three periods of three months off with stress, and I think I haven’t had a day off in the last two years. I will come in because I want to, whereas I’d have looked for any excuse not to before. (T)

Sustained change and improvement rarely occur as the result of a single action or combination of actions at any given time. Rather, success occurs over time as

6 Capacity Building Through Layered Leadership

129

a consequence of incremental, sustained changes as a result of the planning and implementation of a range of strategies in a number of key areas identified by the headteacher as being crucial to success. Phase 1 was primarily about achieving short term success in such key areas as expectations, attendance, behaviour, teaching and learning standards, and the physical environment. Creating the conditions for longer term success required, also, however, for a number of capacity building initiatives to be seeded during the first period. Among these were “performance management”, lesson observation, distributed leadership, training staff to collect and analyse classroom level data about student progress and relate this to their teaching and learning decisions whilst continuing to provide personal and emotional support to staff, parents and students. These next five steps together built staff’s capacity to sustain improvement within a sense of agreed common purpose. Prior to this, however, she involved the whole school community. i) Vision and Values: developing the school’s mission Diana wanted to ensure that the school’s mission and purposes were developed collectively so that they would be reflected consistently in all its work. Hallmarks of her leadership in this phase were inclusivity and participation in decision-making – strategies which contributed to promote her vision for the school through others. The start of phase 2 was an INSET day with the governors, all the staff, mid-day supervisors, cleaning staff, site manager, on visions and values. Coming up with an agreed set of values. . .this phase was crucial because we knew we would be inspected in two years time but we had got the freedom to move on. (HT)

ii) Distributing leadership This phase began with two key actions affecting the leadership and management of the school. First, a new deputy headteacher was appointed to work alongside her and to teach Literacy to Year 6 (11 year old) pupils. Second, leadership of various strands of the school’s work was distributed through the establishment of a new leadership team, entrusting staff with greater responsibilities. This included a member of staff with responsibility for overseeing inclusion, ensuring that all students have equal access to the curriculum. Distributed leadership, especially at this point in the school’s history, provided two messages. First, it communicated a belief that many people, rather than one, are able to take responsibility for leading change. Second, it demonstrated the confidence and trust of the headteacher. She is a very astute people manager. She knows how to inspire people and make them believe in themselves. . .very good at focussing on somebody’s strengths. . .empowering people. . .She’s always there. If things get too much, she’ll sit down with you and work it out. . .prioritize, make it manageable. . . (DH)

iii) Inclusivity: integrating students from different social and cultural environments Diana’s policy of including students as well as other key stakeholders in discussing ways forward for the school continued and the strong ethos of inclusion

130

C. Day

meant that equality of opportunity was central to all aspects of the curriculum. Special attention was given to the inclusion of students with emotional and behavioural difficulties. Incidents of racism or bullying were dealt with fairly and consistently. The multi-cultural population provided the school, in Diana’s view, with an invaluable resource from which to help students learn about each others’ cultures; and before students ate lunch, they thanked the kitchen staff for their food in the different mother tongue languages. There’s a lot of care around them, if you listen to teachers trying to sort out a pair of glasses, to get the parents to get a pair of glasses. . .they know that if the child doesn’t have a pair of glasses they’re not going to be able to see them and do their work. . .We can’t be the social worker, but we can be instrumental in getting other agencies involved when we feel we need to. . .you know, when they’re having a hard time at home. There’s a girl who though she had a fantastic birthday because she’d had two bags of crisps in bed, with a friend. That was her birthday. . .some our children’s parents are in prison so they don’t see them from one month to the next. . .So you do special little things for her in the classroom, you get the whole class to do something, to make sure she feels special. . .It’s those little touches that make a difference to a child being happy. . . (DH)

iv) Performance Management and Continuing Professional Development All staff had developed detailed job descriptions in the early stages of Diana’s headship. Government legislation ensures that all staff are interviewed (appraised) on an annual basis in order to set and evaluate targets for work related to whole school, classroom and individual needs. Diana ensured not only that the system of performance management was linked to school priorities, but that staff’s individual concerns and needs were supported through planned continuing professional development (CPD). We’ve tried to link into the bigger picture and, beneath that, there’s an objective for individual staff as well. . .I identify [from the interview] what individuals need and make a list of things that are not just objectives, but particular things that teachers need. (Headteacher)

CPD was both inward and outward looking in its focus. For example, Diana created a strategic plan of focussed staff meetings, some of which involved staff forming working groups to conduct action research on various aspects of learning. Classroom observations were used to identify both individual and whole school teaching and learning issues which were then supported by training opportunities. ICT expertise was developed – with a member of staff responsible for its leadership throughout the school – to meet the learning needs of all students. External sources were also used to support and raise staff expectations for what it was possible to achieve. The school became a member of the Local Authority [school district] Education Improvement Partnership. Through this a range of extended activities for students were able to be offered before and after school. School governors dedicated a considerable amount of the school budget to helping children achieve their full potential through the provision of teaching assistants to assist teachers in the classroom. Again, Diana led by example,

6 Capacity Building Through Layered Leadership

131

becoming involved in out-of-school education projects (e.g. as a Primary Strategy Consultant leader working with other city headteachers and their leadership teams to raise standards of literacy and numeracy). These helped her, “keep ahead of the game”, review her own vision and sense of purpose to the benefit, also, of her own school. [Head] is very good at knowing about external initiatives, for example the launch of the ‘Excellence and Enjoyment’ policy by DfES. It was in here straight away. We were dissecting it. A member of staff read a chapter a week and fed back to the staff meeting on what it would mean. . .We all took responsibility for different chapters. . .[head] is very outward looking. . .She feels a great sense of responsibility to the children and that if new, innovative things are coming up that can improve their lives and their experiences in school, then we go for it. . . (T)

v) Persisting priority on teaching and learning: becoming a thinking school Diana continued to observe and feedback on classroom work, and began now to involve other staff [with a view to fostering teaching which encouraged students to take responsibility for their own learning through a personalised approach]. Teachers watch each other and analyse. . ..and, slowly, staff attitudes have changed from being self-engrossed and defensive to a professional atmosphere in which teachers are comfortable with observing and supporting each other. They are going to each other more, and we are becoming a thinking school. (Headteacher)

This work with colleagues in classrooms was not only used as a means of “purging poor teaching” but also showing staff, “how students could be empowered to take greater responsibility for their learning”. What had initially met with resistance from some staff was now beginning to become second nature to them as they experienced the benefits. Diana was to develop the use of classroom data much further in the next phase of the school’s journey to success.

Phase 3 Going Deeper and Wider: Sustaining the Momentum (2002–2005) As Oasis continued to improve its physical environment, test results and teaching and learning, there might be some expectation that it could sit back on its laurels a little. However, this was not the case. Diana – and by now all the staff – were committed to sustaining and developing further the strategies for improvement. I don’t think we’ve ever coasted here. We achieve one thing and then we seem to go on to the next. (Deputy Head)

The school went both deeper and wider in its efforts to continue to improve. Induction packages were introduced for new staff, curricular targets were set with teachers for particular children as teaching and learning became more personalised: As a staff, everybody brought their records (of students) and Diana and I had a look at which children in each class were not achieving. And then we took out the special needs

132

C. Day

children, the EAL (English as an additional language) children, and then looked at the ones who were left and discussed with the teachers what they were doing with these particular children. . .That’s made teachers look and think, ‘What am I doing for those children?’. (Deputy Head)

As a result of the success of foundational strategies used in Phase 1 and 2, confidence and self esteem had grown and staff were ready to engage in more detailed analysis of their work. Every child has an individual educational plan. . .personalised. The teachers know which learning style the children prefer e.g. kinaesthetic. . .homework is personalised with the parents. . . (T)

Diana and her management team prioritised issues related to school culture, leadership development, recruitment and retention and workforce reform (a government initiative design to enable all staff to spend the equivalent of half a day each week out of the classroom). Additionally, the “Investor in People” status gained in the previous phase continued with significant investment in the ongoing development of staff. Perhaps the most significant development in this phase, however, was the increasing use of classroom data to inform decision making. A further five steps of the journey may be identified: restructuring the leadership; involving the community; assessment for (personalised) learning; staff development; and broadening horizons. Again, it is important to note that these were not new nor were they suddenly introduced. Rather, they were the development of journeys which had been begun in previous phases. i) Restructuring the leadership Two distinctive teams were established – a senior leadership team and a management team. The former focused upon the strategic development of teaching and learning throughout the school; whereas the latter was mainly concerned with the successful implementation and evaluation of procedures and systems. ii) Involving the Community Since the beginning, Diana had worked to engage parents and other members of the community, in supporting the school. The formal presence of this is a “Harmony” group of ex parents, “pillars” of the community who meet in the school and seek to raise funds. In addition, to this, the school liaises with a wide range of community groups and local government (district) committees responsible for implementing government policy on extending the provision of integrated services. The plan is that a “Children’s Centre” will be built on the school site in 2008 and this will service the life needs of parents and pupils, drawing more adults into engagement with the school. In addition, the school is the site for parenting classes, liaises with the Further Education College, and involves parents in classroom activities. iii) Assessment for (personalised) learning From the beginning, Diana’s vision had been to encourage pupils to take more ownership of their own learning, but in order to do so, the conditions in which

6 Capacity Building Through Layered Leadership

133

they learned and were taught had to provide psychological and social security and they had to begin to believe in their own abilities to achieve. We do a lot of role play. . .it might be something on being offered drugs. . .so it’s getting them to think about consequences of taking different actions. . .in class and assemblies. (HT)

Earlier phases had, as we have seen, established these conditions, through behaviour policies, raising expectations, increasing resource availability and accessibility and raising standard of teaching and learning. It was a combination of success in these, together with the new government policies which supported personalised learning, which now created opportunity in Oasis for students to take more responsibilities. Involving students in evaluating their own progress and learning needs through formative assessment methods had been a first step and now: We have pupil discussions about progress, use data from assessments to see whether children are progressing, whether they’re achieving. If children are underachieving, not at the level they should be, we note that and make individual plans. (HT)

iv) Placing staff development and well-being at the centre of school improvement She sees our needs, doesn’t overload us. (Classroom Teaching Assistant)

Having created and nurtured over time a common sense of purpose and matched teaching and learning relationships and practices to this, continuing professional development and well-being remained key drivers for Diana’s strategies for facilitating and stimulating change and improvement. She’s very good at pacing anything new that comes into school so that staff don’t feel overwhelmed. . .so she’s very good at prioritizing. . .and everything is brought in for discussion. . .and I think that’s the key. So we all feel as though we matter, that we all count and our opinions count. . .so we’re carried along and not pulled. . .we talk it through until we’re happy with what’s going on or we agree to give it a try. And then, once the school was running smoothly, initiatives started to come in, staff training and so on. . .Now we can get involved in all kinds of exciting initiatives: thinking hats, university links. I get the university people coming in doing science experiments; take the children to the university to get them to aspire. We’re in a position where we just want to keep moving forward and we’re assessing where we are and how we can improve and it’s exciting. (T)

There was a well planned programme of meetings to ensure regular communication and staff development relevant to the needs of school and individual: Diana is very good at focusing on people’s strengths. . .using those to inspire and promote. . .(John) is in IT. . .so we looked at that with (John) and gave him time to develop it. And I think Diana is very good at doing that, at empowering people and getting them to develop their strengths. (Deputy Head)

v) Broadening Horizons I think they enjoy school. We widen their worlds with what we provide and what we do, compared to the context where some of these children are living. . .We’ve got lots

134

C. Day of projects on the social and personal side. Some of the boys are out on the streets at night. (T)

Children from highly disadvantaged communities are often deprived of experiences from outside their immediate environment, and as part of the whole vision for the school was to extend their view of what might be possible for them by looking outside their normal environment. As part of this, the NQT (newly qualified teacher) with whom Diana had worked in the early years of her headship had bid successfully for an externally funded European project in which partner schools from five countries worked together on ways of recycling and saving the environment. Video-conferencing links and visits bought experiences of new and different worlds to the students, which, together with a range of other visits to locations outside the school, had the effect of extending students’ experiences beyond their classrooms and normal life environments.

Phase 4 Excellence and Creativity: Everyone a Leader (2005–present) The agendas for excellence and creativity in this phase are important because they represent a continuing optimism for what the students and teachers can achieve. They are not the final steps in an improvement “cycle”, for those already begun and developed in earlier phases will continue. The improvement journey is not one which will end. i) Creative Partnerships and Creativity But now it’s more of a focus on the creative. . .we’ve got drama and we are just linking the creative partnerships to get them involved in the school. (Deputy Head)

“Creative partnerships”, a government sponsored scheme which brings artists into schools to work with students over an extended period, is being used by the school to complement its policy of exposing students to a variety of positive learning opportunities in drama, music and sport. Yet “creativity” is defined by Diana as also opportunities to celebrate different festivals and aspects of the students’ cultures: We had a Divali concert and loads of parents came. (Deputy Head)

It is defined also through work in which she seeks to develop all staff as leaders. In these ways, Diana is seeking to develop, in adults and students alike, the ability to look at things in different ways, to take risks and to combine the use of the cognitive and the emotional in intellectually imaginative ways. ii) Excellence through self-evaluation The new challenges for the school are the same as the old – to continue the upward trend in achievement – but the context is different now. Instead of needing

6 Capacity Building Through Layered Leadership

135

to move the school from “special measures” to a standard which is acceptable, the challenge now is to move the school from “good” to “better” or “excellent”. Each new intake of students presents the same challenge, but now the school is better positioned to meet these. The children want to achieve now. You can feel it. I can feel it with my class. A few years ago it was quite a struggle. You were up against quite a lot of negativity from the children. . .not despair, but disinterest. And now, because their success is celebrated, they want more of it. . .they want to do well for themselves, and can see a means of doing it. (T)

There is a well established positive approach to inclusion and integration; an atmosphere of mutual appreciation; a secure, non-threatening environment for teaching and learning; a committed, purposeful staff; specific support for students with special educational needs; targeted support for students who are not yet achieving their potential; shared leadership and management responsibility among the staff; a vibrant ICT environment; staff development which is embedded in the needs of the school and individuals; working partnerships with external organisations; a culture of care and achievement. All of these have been achieved through the articulation, communication and implementation of shared moral purpose, an ethic of care and equality of opportunity, an insistence that all students are entitled to a first class education from dedicated, knowledgeable and skilled staff, and a belief that all students can achieve and demonstrate their achievement to themselves and the world outside the school.

Discussion: Layered Leadership Attend initially to three basics. . .literacy, numeracy, and well-being of students (sometimes called emotional intelligence, character education, safe schools). These are the three legs of the improvement school. (Fullan, 2006: 46)

At one level, what Michael Fullan recommends in discussing turnaround success is what this school principal seems to have done. Yet this account of her work over seven years demonstrates that creating conditions, achieving and sustaining success is far more complex than the image of a three legged stool suggests. Whilst Kanter’s (2004) three “cornerstones” solutions of accountability (in which people seek feedback and self-improvement); collaboration (where interpersonal bonds are strong and where there is a sense of belonging); and initiative (in which expectations of success produce the energy to put in extra effort and keep going under pressure) apply, they also provide an oversimplified image of conditions for successful change. The process of change is not linear, nor does it involve the application of discrete strategies, however potentially powerful their individual or collective impact might be. Rather, as we have seen, it is likely that a layered approach with more emphasis being placed on certain strategies whilst others are “seeded” during particular development phases of the school and those who work there, will be

136

C. Day

more effective. As “milestones” are reached, so further progress, with more emphasis on different strategies, whilst continuing with those already in place, can be made. I think what drives me is thinking that we give these children an opportunity to make a choice. . .to have a good education. They’re not going to have an equal chance but they’ve got as good a chance as we can give them. . .the academic side is important but so is the social. They have to have good social and interpersonal skills if they are to survive and hold their own. (HT)

What we learn from Diana’s story is that the challenges for schools located in areas of high deprivation and crime are persistent. Schools themselves cannot be responsible for solving these. What we also learn, however, is that school principals and their staff can: i) affect the life chances of their students through providing environments and teaching which raise expectations for achievement; ii) provide opportunities for students to fashion a broader view of future life and work prospects (some of the Oasis students are, for example, aspiring to become doctors, politicians and lawyers); and iii) instil in all students the willingness, will and self confidence to succeed. It is worth remembering that where the school is now on its journey is some distance from where it began, and that it has, by and large reached this point is in no small measure due to the values, qualities and skills of its headteacher. Today a visitor to the school could be forgiven for appreciating its bright, almost pristine environment. Classrooms are purposeful, the welcome for visitors is genuine, the warmth of relationships is evident and the sense of achievement by staff, students and all those connected with the school palpable. The students seem to be like those in any school in a thriving area of town or city. It is difficult to imagine that the homes from which they come and to which they return at the end of each day, the areas in which they play and the older children and adults with whom they spend a good deal of their time are likely to be in direct contrast to their school experiences. Many continue to experience emotionally dysfunctional home relationships, some are involved in crime and witness drug taking and its effects. The work of the school in striving for excellence must always be set in these contexts. Schools and leaders of schools in locations which serve highly disadvantaged communities such as this can never “coast” if they are to succeed. Bullet pointed lists of strategies which may help to achieve success can never fully represent how leaders achieve and sustain success. Acknowledgement that success is relative (to the history and present context of the school and its environment), that successful leaders are driven by intentions which are associated with “beliefs”, “knowledge” and “skills” and their application in specific contexts at particular times and with particular individuals and groups, and that successful improvement processes are both multi-layered and time related, may provide more useful assistance to others who wish to set out on their own leadership journeys to success.

6 Capacity Building Through Layered Leadership

137

References Ansell, D. (2004) Improving Schools Facing Challenging Circumstances, Nottingham, England: National College for School Leadership. Day, C. (2004) A Passion for Teaching, London: Routledge Falmer. Elmore, R. The Hollow Core of Leadership Practice in Education (unpublished paper), Harvard University Graduate School of Education, cited in Fullan, M. (2006), p. 57. Fullan, M. (2006) Turnaround Leadership, San Francisco: Jossey-Bass. Fullan, M., Hill, P., Cr´evola, C. (2006) Breakthrough, CA: Corwin Press. Giles, C. (2007) Building Capacity in Challenging US Schools: An Exploration of Successful Leadership Practice in Relation to Organizational Learning. International Studies in Educational Administration, Vol. 35, No. 3, 2007, pp. 30–8. Hargreaves, A. (1998) The Emotional Practice of Teaching, Teaching and Teacher Education, Vol. 14, No. 8, pp. 835–854. Hargreaves, A. and Fink, D. (2006) Sustainable Leadership, San Francisco: Jossey-Bass. Jackson, P. W., Boostrom, R. E. and Hansen, D. T. (1993) The Moral Life of Schools, San Francisco: Jossey-Bass. Kanter, R. M. (2004) Confidence: How Winning and Losing Streaks Begin and End, New York: Crown Business. Kelchtermans, G. (1996) Teacher Vulnerability: Understanding its Moral and Political Roots, Cambridge Journal of Education, Vol. 26, No. 3, pp. 307–324. Milnthrop, H. (2004) Schools on Probation, New York: Teachers College Press. National Audit Office (2006) Improving Poorly Reforming Schools in England, London: Department for Education and Skills. Sockett, H. (1993) The Moral Base for Teacher Professionalism, Columbia University: Teachers College Press.

Chapter 7

The Relationship Between Distributed Leadership and Action Learning in Schools: A Case Study Stephen Dinham

Introduction Teachers’ professional learning has been shown to be fundamental to successful teaching, student achievement and successful schools (Hattie, 2003, 2007; Dinham, 2007a, 2008b). In recent times there has been a movement away from traditional approaches to teachers’ professional learning towards more decentralised, contextualised forms of learning. (Dinham, 2007a). This chapter is mainly concerned with the method of action learning, and how distributed leadership is both a precondition and product of this process. In doing so, this chapter reports on an evaluation of 50 Quality Teaching Action Learning (QTAL) projects coordinated by the New South Wales (Australia) Department of Education and Training (NSW DET). Projects were funded and carried out as part of the Australian Government Quality Teaching Program (AGQTP), a national initiative designed to drive improvement in school education across all Australian sectors.1 The evaluation brief from the NSW DET was to investigate the conditions influencing teachers’ implementation of an inquiry-based approach to action learning. The evaluation encompassed 50 individual projects involving 82 NSW public (state) primary and secondary schools that had successfully tendered for grants to investigate school-based and school-driven action learning using the framework provided by the NSW model of pedagogy (NSW DET, 2003). Within the overarching QTAL project, each school or group of schools pursued an individual project (e.g., gifted and talented programs, literacy, quality teaching in science, primary to secondary transitions, etc.). The common approach taken by schools was to use the funding provided to release small teams from some of their teaching duties to work together on an

S. Dinham (B) Australian Council for Educational Research, Melbourne, Australia 1

http://www.qualityteaching.deewr.gov.au/agqtp/

A. Harris (ed.), Distributed Leadership, Studies in Educational Leadership 7, C Springer Science+Business Media B.V. 2009 DOI 10.1007/978-1-4020-9737-9 8, 

139

140

S. Dinham

approved Quality Teaching Action Learning project with the assistance of a designated university academic partner, “expert” in the area of the project. Teams were usually volunteers and comprised a mixture of classroom teachers and those in formal leadership positions. Principals were not usually part of the teams, although they played important roles in developing and supporting the projects. The evaluation (Aubusson, et al., 2005) found that the QTAL projects undertaken by school teams as part of the AGQTP were successful both in promoting and utilising action learning and in achieving individual project aims. Being part of such teams led to the professional growth of those involved and this manifested in increased individual and collective leadership capacity, activity, and influence in the school and sometimes beyond (Aubusson et al., 2007; Brady et al., 2006). The Role of the NSW Model of Pedagogy The document Quality Teaching in NSW Public Schools, incorporating the NSW model of pedagogy (NSW DET, 2003), provided an important rubric for action learning around improving pedagogy and for the evaluation of the project reported in this chapter. It has been found that such rubrics, coupled with associated concepts and terminology, provide an important vehicle and frame for professional discussions, planning, action and evaluation around pedagogy (Dinham and Rowe, 2007; Dinham, 2008b). The NSW model has been designed to be used by principals, school executive, and teachers “to lead and focus the work of the school community on improving teaching practice and hence student learning outcomes” (NSW DET, 2003: p. 3). The model includes three dimensions of pedagogy (p. 5):

r r r

pedagogy that is fundamentally based on promoting high levels of intellectual quality; pedagogy that is soundly based on promoting a quality learning environment; pedagogy that develops and makes explicit to students the significance of their work.

“Intellectual quality” includes the elements of deep knowledge, deep understanding, problematic knowledge, higher-order thinking, metalanguage, and substantive communication. “Quality learning environment” includes explicit quality criteria, engagement, high expectations, social support, students’ self-regulation, and student direction. “Significance” includes background knowledge, cultural knowledge, knowledge integration, inclusivity, connectedness and narrative (p. 9). Since the introduction of the NSW model of pedagogy in 2003, anecdotal evidence suggests that while many schools have fully engaged with and used the model to rethink and revitalize teaching and learning, other schools have largely ignored it. Some teachers have welcomed the focus on pedagogy after years of more extraneous imposed management and accountability mindsets, paradigms and policies, while other teachers have disparaged the model as just another fad or imposition from out of touch theorists and system bureaucrats.

7 The Relationship Between Distributed Leadership and Action Learning in Schools

141

Distributed Leadership There has been a subtle shift in conceptions of educational leadership in recent times. An earlier focus on educational administration and later management has turned more to leadership for enhanced teaching and learning. There has also been concern with an over-emphasis on the supposed attributes of the charismatic, heroic, “super leader,” and the finding that such leaders can be “negatively associated with leadership sustainability” has called into question the wisdom of seeking out and appointing such leaders (Fullan, 2005: pp. 30–31). Additionally, an earlier focus on formal leadership – especially the principal – has broadened to consider the influence of other school leaders and teachers, i.e., distributed (or distributive) leadership (Harris, 2004: p. 1; Dinham, 2007b). Although the concept of distributed leadership can be traced back to social psychology in the 1950s, it is only in the last decade or so that the concept has received widespread prominence and attention (Gronn, 2002: p. 653), although problems of loose definition remain (Harris, 2008: p. 173–175). These changes in how educational leadership is conceived and enacted reflect a number of realities: that teaching and learning should be the prime focus of the school; that principals cannot bear all the burden of school leadership due to increasing pressures and demands being placed upon themselves and schools; that others in formal leadership positions in schools are also under pressure (Dinham and Scott, 2002; Dinham et al., 2000), and that the contribution to educational outcomes of distributed leadership has tended to be overlooked or undervalued (Spillane et al., 2001; Gronn, 2002: p. 654). There is also the issue of leadership succession, especially when leaders who have attempted to keep leadership power largely to themselves depart (Lambert, 1998: p. 10; Hargreaves and Fink, 2004: p. 8; Dinham, 2007c). Overall, there is growing recognition that there is much unreleased and unrealized leadership potential and capacity for improvement residing in educational organizations (Crowther et al., 2002: pp. 3–16; York-Barr and Duke, 2004) and that distributed leadership is important in growing this organisational and individual capacity. Gronn (2002: pp. 654–660) has considered the multiple meanings of distributed leadership, which fundamentally fall into two groups, the first seeing distributed leadership as essentially additive (more leaders, spread leadership) and the second more holistic, including all forms of collaboration and participation. Rather than spreading existing leadership across more people, an holistic view of distributed leadership is concerned more with the synergies that can occur when people come together to work, plan, learn, and act, thus generating further leadership capacity within the individual and the organization. Distributed leadership, including teacher leadership (see also delegated leadership, democratic leadership, shared leadership, dispersed leadership; Bennett et al., 2003: p. 4) is now a major aspect of and influence upon constructs of educational leadership (Duignan and Bezzina, 2006), although as Harris has noted (2005: p. 170), as well as enthusiasm for the perceived benefits of the concept,

142

S. Dinham

“we urgently need contemporary, fine-grained studies of distributed leadership practice . . . without the associated empirical base it is in danger of becoming yet another abstract leadership theory.” York-Barr and Duke (2004: p. 292) concur: “there is little empirical evidence to support [teacher leadership’s] effects”. However York-Barr and Duke are optimistic about the potential for educational improvement through teacher leadership “despite being thwarted by centuries-old structures and conditions of schools that resist change”.

Leadership and Student Achievement The ultimate challenge for all educational leaders is to make things happen and improve in the classroom to facilitate learning. While there is now little doubt of the influence of the individual teacher on student achievement (Hattie, 2003, 2007; Rowe, 2003), leadership has been shown to influence what happens in the classroom through a variety of ways (see Mulford, 2008), with its influence perhaps being underestimated due to difficulties associated with measurement (Dinham, 2007d). A study of 38 government secondary schools in New South Wales where “exceptional” educational outcomes were thought to be occurring (Dinham, 2005) found that leadership (principal, other executive and teacher leadership), influenced student outcomes through:

r r r r r r r

A central focus on students and their learning; Teacher learning, responsibility and trust; External awareness and engagement; A bias towards innovation and action; Student support, common purpose and collaboration; Personal qualities and relationships; Vision, expectations and a culture of success.

In the above study, leadership – both principal and distributed – created the climate and conditions where teachers could teach and students could learn. Further, those in formal leadership positions, particularly principals, exercised leadership that resulted in others being encouraged and supported to develop and exercise their own leadership. Trust, sharing of power, giving people discretionary space, collegiality, and mutual respect were important elements in this process. Distributed leadership is particularly important in larger schools that tend to be fragmented or broken into “silos”. Size, complexity, workload and fragmentation tend to militate against the effectiveness and reach of a central, “hands-on” leader. In the secondary schools achieving exceptional outcomes, it was found that leadership capacity was developed and exercised by teams and functional groupings (i.e., faculties, other teams and informal groups) through whole-school programs and initiatives. Effective leaders were found to have the capacity to identify, develop and release the leadership capacity of others, for the benefit of all (Dinham, 2005, 2008b).

7 The Relationship Between Distributed Leadership and Action Learning in Schools

143

In reviewing the literature on distributed leadership and teacher leadership, Harris (2004: pp. 6–7) identifies common messages about ways in which teacher leadership and distributed leadership are enhanced and supported:

r r r r r

time needs to be set aside for professional development and collaborative work between teachers . . . teacher leaders need opportunities for continuous professional development in order to develop their role . . . The success or otherwise of teacher leadership within a school is heavily influenced by interpersonal factors and relationships with other teachers and the school management team . . . The ability of teacher leaders to influence colleagues and to develop productive relations with school management, who may in some cases feel threatened by teacher taking on leadership is therefore important . . . Overcoming these difficulties will require a combination of strong interpersonal skills on the part of the teacher leader and changes to the school culture that encourage change and leadership from teachers’.

These “messages” resonate strongly with the findings of the Quality Teaching Action Learning evaluation reported in this chapter.

The Study A research team from the University of Technology Sydney (Peter Aubusson and Laurie Brady) and the University of Wollongong (Stephen Dinham), with support from the New South Wales Department of Education and Training (NSW DET) staff, conducted an evaluation of the Quality Teaching Action Learning (QTAL) project in NSW. Schools had been invited to apply for Australian Government Quality Teaching Program (AGQTP) funding and 50 QTAL projects involving 82 schools were successful in having their projects – which had to meet both AGQTP and DET guidelines – approved.

Method The method used in the evaluation progressed through six phases: Phase 1. Planning and Design. August 2004–September 2004. This phase involved liaison between the evaluators and DET personnel; design of methodology (questionnaires, focus groups, mini-journals); and the recruitment and training of research assistants. Phase 2. Preliminary Research. September 2004–December 2004. This phase involved the collection of demographic, personal, and professional data from all participating schools; analysis of the 50 successful school

144

S. Dinham

applications representing 82 schools; selection of nine case study schools; and analysis of the first progress reports from all participating schools. Phase 3. Initial Case Study Research. January 2005–March 2005. This phase involved the first on-site collection of data from the nine case schools; the initial collection of mini-journals from the case schools; and analysis of school policies, meeting minutes, and resources relating to school projects. Phase 4. Mid Term Review. April 2005–May 2005. This phase involved evaluator sharing of aggregated data; analysis of second progress reports from 50 schools; and the collection of data at the DET midprogress sharing conference where representatives of all 50 QTAL project teams came together. Phase 5. Final Case Study Research. June 2005–July 2005. This phase involved the second collection of data from the nine case study schools; the second collection of mini-journals; the preparation of the case studies and evaluator synthesis of common insights; the analysis of final school progress reports; and further examination of relevant school documents. Phase 6. Validation and Analysis. August 2005–October 2005. This phase involved the content analysis of the action learning project reports and the writing and submission of the report to the NSW DET.

Findings of the Evaluation Findings from the various data sources were consistent, with the QTAL project found to be very successful overall. It needs to be noted that this was one of the most successful programs with which the evaluators have been associated. Not all evaluations report such positive findings. Broad findings of the evaluation are summarized below. The final report (Aubusson, et al., 2005) contains full details on methodology, findings and recommendations for the QTAL project.

Broad Findings from the Evaluation 1. 2. 3. 4. 5. 6. 7.

Successful projects were built upon a genuine, recognized need in the school(s). Successful projects had clear, agreed, achievable, and suitable goals. Support from the principal (and other leaders) was essential in project success. A credible, suitable leader for the project was also vital. Successful projects were characterized by effective teams and team building. Schools found it difficult to start and to build momentum. It was important to maintain communication with all school staff about the school’s project. 8. Academic (university) partners provided valuable conceptual and theoretical background and assisted with framing, implementing, and evaluating projects.

7 The Relationship Between Distributed Leadership and Action Learning in Schools

145

9. Teacher release time was a major factor in project success. 10. Schools found the NSW model of pedagogy (Quality Teaching) a useful rubric. 11. The most successful schools considered long term sustainability of the projects from the start. 12. Overall there were strong indications that projects were successful, but evidence of student outcomes was inevitably lacking given the time frame. 13. There was increased although still limited sharing of the successes of schoolbased initiatives with other schools. 14. Schools and individuals valued and benefited from the sharing conferences which brought project teams and facilitators together. 15. Distributed leadership was both a major factor in the success and a significant outcome of teachers’ action learning. The focus of this discussion now turns to the roles of teacher action learning and distributed leadership in the success of the projects.

Distributed Leadership and the Study Findings Overview Successful projects were characterized by effective teams and team building. . . . Distributed leadership was both a factor in the success and an outcome of action learning (Aubusson, et al., 2005).

The opportunity to work in teams and how teams were empowered and supported provided the opportunity for the development of distributed leadership, a key aspect of project success. It was evident that team members grew in expertise, confidence, and influence during the projects. Both individual and school leadership capacity were enhanced as a result of project participation. Comments from the nine case study project visit reports included: Leadership is more distributed with teachers taking more responsibility for their professional learning and increasing their contribution across the school – “leadership is more spread now, more pedagogic thought . . . more receptive” [Principal] (Red Gum Primary).2

Project Teams: Formation and Leadership The usual scenario in the 82 schools taking part in the 50 projects was for the school(s) to have previously identified an area of need, and to have completed some prior development work on this. The AGQTP and the QTAL project provided the means to address this need in a more systematic, in-depth way. A number of schools described the timing of the QTAL project as “fortuitous . . . it came at the right time.” 2

All school names are fictitious.

146

S. Dinham

Typically, it was the Principal, with a few other staff, who developed the proposal for funding. Once funding was granted, principals handed over project direction to a project leader, with the Principal acting as an advisor for the duration of the project. The case study report for Iron Bark High noted: The Principal of the secondary school said she was involved in all stages although [the project leader/deputy principal] was the driver . . . Distributed leadership was enhanced through the project, which had “spread leadership across faculties . . . staff are taking on leadership roles.”

Sometimes the project leader was a member of the school executive team and at other times a classroom teacher. The latter was more common in primary schools. A small team worked with the project leader and team members were usually volunteers with prior experience and/or interest in the substance of the project. Most project teams comprised both teachers and school executive (promoted teachers), but this does not appear to have resulted in problems of inequity. For example, the experience at Finch Primary School was that: Initially some members of the group were fearful of the workload and were concerned that the executive members of the group might act as “supervisors.” Even though the eight members included four members of the school executive, the group did not have a “supervisory” feel. All group members found the whole experience non-threatening.

The case study report for another school noted: The support from school leaders for the project, especially the Principal and project coordinator was seen as essential. The project leader was described as: “constantly actively involved” and “a big lynch pin but knew how to distribute leadership” (Iron Bark High).

Principals had significant influence over the composition of project teams without directing teachers to take part, although in several cases principals confided how they induced potentially negative or obstructive teachers (“blockers”) to be part of the teams to neutralise their influence and get them “on-side”(see Dinham, 2008a). The case study report for Wollemi Primary School noted: Some teachers were invited onto the team to provide an opportunity for building leadership expertise rather than because of a special commitment to the project or perceived leadership qualities. In this way, it was hoped [by the co-leaders] that the QTAL project could contribute to building long-term leadership capacity of the school. In this distributed leadership model, each member of the QTAL leadership team would plan the project’s progress determining what actions to take, what evidence to collect and analysing this evidence to determine further actions.

Project teams spent time prior to and in early stages of the project meeting and planning to formulate the goals for their individual project. These conversations were important in framing and directing projects, although in a minority of cases, goals proved overly ambitious and needed to be scaled down due to the limited time frame. However, things did not always run smoothly, as the above report from Wollemi Primary noted: Members of the QTAL leadership team confided that at least two members of the initial project team were reluctant members and did not develop the enthusiasm or leadership qualities needed to promote and lead the project within their stage [grade/year] groups.

7 The Relationship Between Distributed Leadership and Action Learning in Schools

147

However, both left the school during the project and their replacements in the [project] leadership group proved more productive.

Typically, the project leader was partly released from teaching responsibilities during the project to work with other staff and the university academic partner and to attend planning and sharing conferences associated with the QTAL project. Project leaders assumed a higher profile within, and in some cases outside their schools, than previously. They worked with members of the project team drawn from across the school, and in some cases with teachers engaged with the project from other schools. The case study report for Bilby Primary noted the importance of committed leadership from the project leader: The ICT [Information Communication Technology] teacher “knew where the school needed to go; she was really committed to it”. She was described by a team member as “our guiding light.”

In citing conditions for the success of the project at Wollemi Primary, the evaluation team found as major factors: [The] Established strong leadership team who developed their expertise in leadership, Action Learning, the NSW model of pedagogy and mathematics teaching. . . Leadership of the two assistant principals who had experienced similar projects, were confident, well respected and “had clout” with both staff and the executive. . . [and] Strong sense of commitment, shared responsibility and mutual support initially between the two executive leaders, which later developed more widely among the majority of the leadership group.

In interviews at the case study schools, principals recounted how they had selected project leaders both on the basis of their leadership skills, and on their potential for leadership. Over all, it was apparent that project leaders had grown into the role, gaining leadership skills, experience and confidence. It was also found that members of project teams grew in their leadership capacity during the course of projects, particularly those not in formal leadership positions. Clearly, the project leaders, with support from their principals, led their teams well, with collaboration and teamwork being essential factors in the success of the projects and in connecting the projects with their colleagues in the rest of the school.

Collaboration and Professional Dialogue Increased collaboration and communication among teachers was reported as an outcome of QTAL projects by the majority of teams (30, 60%). This trend was evident from reports at all stages. Many teams (28, 56%) reported the value of shared professional dialogue regarding teaching, often noticeable in faculty rooms, as a replacement for discussion about lesson content, student behaviour or social matters (sport, recreation, current events). Comments were often enthusiastic and illustrated the positive nature of the dialogue, for example:

148

S. Dinham

What worked was the real teamwork and collaboration within the executive and between staff members that has generated professional discussion and the ability to try new ways of doing things’. (Kangaroo High School) Faculties were seen to be talking more and working more closely together: “Staff resistant to change are now getting up and sharing.” A dialogue about teaching and learning has developed and people from different faculties are now talking and sharing, whereas they were “their own cells in the past.” There is more understanding of secondary strategies in stage 3 [primary grades 5–6] and of primary strategies in stage 4 [secondary grades 7–8]. (Iron Bark High) [Quotes from Principal]

The case study for Quoll High School reported that an outcome of successful team building was: Extensive teacher learning and teacher growth in risk taking and in confidence. Teachers who felt very hesitant about ICT in the classroom have developed new programs, which they are trialling, incorporating ICT and QT principles. . . . These teachers have learnt new skills with the technology, and are using a greater range of resources.

Team building and distributed leadership provided a critical mass and force for change. The case study report for Iron Bark High noted: The view was that there is “a critical mass now, momentum.” A teacher stated it was “a highlight of my career . . . so positive . . . I have learnt so much.” Teachers were “enthusiastic, everyone likes it because it worked . . . agreed to do it, really enjoyed it, understood it, feel confident, even people teaching for years . . . feedback, reaffirmation, reassurance . . . re-enthused some teachers.”

Empowerment, Learning and Growth An important aspect of the projects was their empowering nature. Teachers were given time, space, guidance and resources to engage in action learning. Rather than being imposed from above, projects grew from within and staff developed professionally through the success of the projects. A number of teams (10, 20%) noted an increase in teacher confidence in teaching a new content area, where this was the project focus. For example the Egret Primary school team reported: Staff have become more aware and have a greater understanding of science and technology and recent documentation. Teachers are more confident and willing to teach science and technology and the collaborative planning of units had increased . . . Staff generally enjoy teaching science and technology, as compared to not enjoying it earlier.

The case study report for Cedar High noted: The project has been a very effective professional learning activity for those teachers involved. It has “renewed a lot of personal interest.” It has “empowered the school and teachers . . . provided resources,” and “provided time and a framework for reflection on teaching and collecting data.” . . . Teachers are more confident and assertive in their professional learning. They “are increasingly using the language” of QT [quotes from Principal and teaching staff].

7 The Relationship Between Distributed Leadership and Action Learning in Schools

149

Selection as project leaders enabled these people to develop and demonstrate their leadership expertise, so much so that some were noticed for the first time and subsequently offered other leadership opportunities. The case study report for Red Gum Primary noted: A number of those interviewed commented how the project leader had grown in confidence and leadership capacity during the process. . . . The leadership, drive and enthusiasm of the project leader before, during and after the project and her availability to staff was seen as essential – “Without [her], it was not a viable option . . . one person to drive was a major factor . . . needed to keep pushing in early stages”.

Subsequently, the project leader from Red Gum Primary was appointed to a quality teaching consultant position in a DET branch office, something which “would not have happened without the [AGQTP] project” [Principal]. Academic (university) partners played an important role in the professional learning and growth that occurred. In some cases academic partners had an existing relationship with the school but in most instances they were appointed to a school or group of schools following project approval. There was an attempt to match the interests and expertise of the academic partner to the individual projects. By common consensus academic partners performed a valuable role in refining and scoping projects and in project implementation. School staff tended to lack background knowledge, research and evaluation skills and academic partners were particularly helpful in these areas. School staff in the more remote parts of the state understandably expressed the view that they would have liked more face to face contact with their academic partner. The NSW model of pedagogy introduced in 2003 was found to have been a useful framework for teachers to reflect on and improve teaching and learning. The fact that it is common to all public schools in the state from K-12 has meant that teachers have a framework and language to discuss pedagogy. All schools involved with the QTAL projects reported a heightened awareness of pedagogy and increased professional learning and discussion flowing from the use of the model and its (mandatory) application to the projects.

Time, Space, Control, and Community A major factor in the success of the school projects was the funding which released team members from some of their teaching. This enabled team members to meet with staff from other faculties and schools, to attend planning meetings and to present and share at conferences. One of the main outcomes made possible by the partial release from face-toface teaching was the building of community – within teams, within schools, and among schools in cluster (multi-school) projects. Common consensus was that such professional interaction and learning is very difficult in the day-to-day operation of

150

S. Dinham

a school where managing behaviours and events overrides deeper consideration of more significant issues. Some comments from case study reports included: Action learning thrives in a high school setting through team teaching. The collaborative nature of action learning is lost when teaching is independent and only reflections, rather than experiences, are shared (Dragon High). The collaborative approach was embraced enthusiastically by all, and it proved to be the catalyst for many other aspects of the project – such as peer mentoring, group planning sessions and collaborative classroom observations. Collaboration seems to have built a sense of team spirit at Cicada and this in turn led teachers to bond in a way they would not have experienced otherwise. The ability of staff from various schools to have time to meet, reflect and carry out stage based planning together has been one of the highlights of this project (Cicada Primary).

In accounting for the success of the action learning project at Banksia (Special School), the evaluation team noted the significance of: Dispersed leadership with choice and control given to teachers. Each team determined “its own direction” and responsibility for it. They were enthusiastic about their projects.

These comments illustrate the enjoyment that teachers derived from their ability to plan and work together. This coming together fostered new ideas and created a supportive atmosphere that encouraged the risk taking and shared learning that was evident in Quality Teaching Action Learning projects. Thus it can be argued that the collaboration that characterises action learning contributes, in a fundamental way, to its effectiveness as a means of individual and group professional learning and the development of learning communities. The story of one project can be used to illustrate the often complex journey toward effective community. Toad Primary School is an example of a school that ultimately achieved a great deal, but staff felt that obstacles had been encountered which had to be overcome. One problem had been the breadth of the initial project aims; another was that some teachers did not want to be included in the project. This team found the timelines difficult to adhere to, and often had the feeling that they were struggling. It was only when they reviewed the project that “the evidence revealed just how far we have come in terms of quality teaching and how our practice has improved as a result of this project” [project team member]. The way that the Quality Teaching Action Learning projects were instigated and implemented allowed the teachers involved to take responsibility for their own professional learning. This was regarded as a strength of the overall QTAL project by a number of teams. For example the Seagull Primary School case study report concluded that: Action learning proved to be a successful mode of delivery for teacher professional development as it allowed for individual teacher needs, and was driven by the individuals involved. It enabled teachers to be actively involved in their own learning and it wasn’t something done to them but rather something they had ownership of and could control.

The team at Blue Wren Primary School suggested that this ownership is essential as:

7 The Relationship Between Distributed Leadership and Action Learning in Schools

151

Schools change when individuals change and improve their professional practice. The combined use of the quality teaching lesson plan with the observation guide and follow up discussions and personal reflection had an impact on changing individual teaching practices. School change is a slow and incremental process; action learning is an effective agent for change since those involved in the research have ownership of their professional development.

The sharing conferences where representatives of the 50 QTAL project teams came together from across the state provided a vehicle for sharing, affirmation, and further leadership development. Project leaders and team members recounted with some pride how they had made presentations at these conferences, something they had not experienced before. The sharing of a project at a QTAL conference through a team presentation was described as “outstanding” by staff at one school (Iron Bark High) while the case study report for Peppermint Grove Senior College noted: The two NSW DET run conferences provided a forum for hearing what other schools were doing, and provided the opportunity to showcase their own achievements. One teacher, with a certain concealed glee, reported on how the team “gave our workshops as though the principals [in the audience] were class members.”

Beginning, Building Momentum, Communication, and Sustainability Some team members had prior experience with action research and action learning but most did not. Framing projects, agreeing on goals and determining strategies were important issues that took some time. There was also trepidation about beginning the projects. The time taken to reach the “take-off point” varied from a week to a month or more. One team member noted that “if we had planned everything, we wouldn’t have started,” the implication being that it is preferable to “break the ice” than to wait until all details are determined. It became apparent from the evaluation data how important it is to maintain communication with other staff. The literature on educational and organisational change stresses the need to clearly communicate to those not directly involved about a change project and its progress in order to build support and overcome “the culture of resistance” (Evans, 1996: pp. 40–51). The fact that in all projects staff were released from teaching duties at certain times provoked some negative reactions from other staff and in a few cases, parents, who did not understand why teachers couldn’t engage in such learning and planning “after hours”. Team members also expressed disquiet and even guilt over the disruption to their classes caused by the employment of replacement teachers. Sharing small successes and the progress of the projects was important both in keeping teams energized and in overcoming negativity from those not directly involved. This is essential if the projects are to be sustainable and to spread across the school. Because of the scope of the projects, – in effect, half a school year – the majority of principals and teams had identified ways to continue the work of project

152

S. Dinham

teams beyond this time frame. Some principals had arranged to use school funds to continue to support teams and most schools had planned to spread their project across more teachers and faculties to build on the success of the projects. In this way, projects were seen more as means than ends to address areas of need and concern. With success and progress, sustainability and further development became important issues. The case study report for Bilby Primary noted: The Principal proposes that the current project team will continue as the ICT team, and that a new Quality Teaching team will be established. She further proposes that someone from the project team will also be a member of the new QT team, and will guide its progress in selecting and developing a new project. The Principal has recently completed a training course in INTEL (integrating ICT into all school learning areas), and will share her newly acquired knowledge with the whole school staff (including the project team).

Further Discussion and Final Comments Research data derived from teachers, school reports and journals, academic partners and the researchers’ site visits, demonstrated that the Quality Teaching Action Learning projects stimulated and enhanced teacher professional learning in the schools concerned. The use of teams of interested and committed teachers was fundamental to this process. Team members were encouraged, empowered, and grew in the course of the action learning projects. Important factors in the operation of teams and their projects included the time, focus and support for professional learning, the teamwork and collaboration of team members, and the work of team leaders. The willingness of principals to share power and responsibility and to identify, respect and foster the leadership capacity of others was also crucial. The QTAL projects were effective in facilitating teachers’ action learning, but were also effective in clarifying, valuing and affirming what teachers and schools were already doing. Thus, the projects operated in a challenge, rather than a deficit context, improving and building, rather than “fixing”. Likewise, the use of the NSW model of pedagogy was seen to validate and affirm what “good teachers do,” whilst providing a useful framework for reflection and action. On a cautionary note, it is fair to say that teams were more adept and successful in promoting professional learning than in the research aspects of the projects. There was some uncertainty over the tools and data needed to track changes in student outcomes over the projects and the longer term. Most school projects had originally included the strategy of peer observation of teaching, and most of these schools had postponed this. It was clear that there are still feelings of risk, fear, and exposure associated with being observed teaching, which has connotations of judgment rather than development for some teachers. However, on a positive note, the teamwork and professional learning arising from the QTAL projects provided a foundation whereby teachers were now feeling less threatened and more comfortable about such professional sharing which was breaking down the isolation of the classroom.

7 The Relationship Between Distributed Leadership and Action Learning in Schools

153

The evaluation team, while being convinced of the overall effectiveness of the QTAL projects in achieving their aims, suggested caution in making such innovations mandatory. The voluntary nature of involvement and the fact that projects grew from needs already identified within the schools appeared important conditions for project effectiveness. While the evaluation team was not directly focused on distributed leadership as either a precondition or product of the action learning projects, it was apparent how important distributed leadership was to action learning and project success. Leadership cannot easily develop in a vacuum, and the action learning projects provided the vehicle to build on and further develop leadership capacity in the schools concerned. The time frame for the QTAL projects was relatively brief, yet there was sufficient evidence to suggest that distributed leadership has the capacity, when aligned with teacher learning, to help foster that elusive phenomenon of the learning community. Because of the timing of the evaluation it was too early to obtain hard data on improved student achievement attributable to the projects. However, there was no doubt that significant teacher learning had occurred and that new approaches to pedagogy had been put into place. Early indicators were that students were responding positively to new programs and more student-centred approaches to learning. Both teachers and students were more enthusiastic, and school progress report, interview, journal and observational data indicated enhanced teacher reflection, cooperation, and understanding. Teachers were moving outside their comfort zones with greater preparedness to take risks and adoption of new approaches. Productive self-criticism had increased. Acknowledgments This chapter draws upon both the AGQTP project and Dinham, S.; Aubusson, P. & Brady, L. (2008). “Distributed Leadership as a Factor in and Outcome of Teacher Action Learning”, International Electronic Journal for Leadership in Learning, 12(4), available at: http://www.ucalgary.ca/∼iejll/volume12/dinham.htm. Permission from Peter Aubusson and Laurie Brady to quote from the various publications arising from the project is gratefully acknowledged.

References Aubusson, P., Brady, L., and Dinham, S. (2005). Action Learning: What Works? A research report prepared for the New South Wales Department of Education and Training. Sydney: University of Technology Sydney. Aubusson, P., Steele, F., Dinham, S., & Brady, L. (2007). ‘Action learning in teacher learning community formation: informative or transformative?’, Teacher Development, 11(2), pp. 133–148. Bennett, N., Wise, C., Woods, P., & Harvey, J.A. (2003). Distributed Leadership. Nottingham: National College of School Leadership. Brady, L., Aubusson, P., & Dinham, S. (2006). ‘Action Learning For School Improvement’, Educational Practice and Theory, 28(2), pp. 27–39. Crowther, F., Kaagan, S., Ferguson, M., & Hann, L. (2002). Developing Teacher Leaders. Thousand Oaks, California: Corwin. Dinham, S. (2005). ‘Principal Leadership for Outstanding Educational Outcomes’, Journal of Educational Administration. 43(4), pp. 338–356.

154

S. Dinham

Dinham, S. (2007a). ‘The dynamics of creating and sustaining learning communities’, Unicorn Online Refereed Article No. 43, Australian College of Educators. Dinham, S. (2007b). ‘The Waves of Leadership’, The Australian Educational Leader, 29(3), pp. 20–21, 27. Dinham, S. (2007c). ‘Authoritative Leadership, Action Learning and Student Accomplishment’, Conference Proceedings, Australian Council for Educational Research, 2007 Research Conference, pp. 33–39. Dinham, S. (2007d). ‘How Schools Get Moving and Keep Improving: Leadership for Teacher Learning, Student Success and School Renewal’, Australian Journal of Education, 51(3), pp. 263–275. Dinham, S. (2008a). ‘Counting the Numbers in Educational Change’, The Australian Educational Leader, 30(1), pp. 56–57. Dinham, S. (2008b). How to Get Your School Moving and Improving: An evidence-based approach. Melbourne: ACER Press. Dinham, S., Brennan, K., Collier, J., Deece, A., & Mulford, D. (2000). ‘The Secondary Head of Department: Key Link in the Quality Teaching and Learning Chain’, Quality Teaching Series, No 2, Australian College of Education, pp. 1–35. Dinham, S. & Rowe, K. (2007). Teaching and Learning in Middle Schooling A review of the literature – A Report to the New Zealand Ministry of Education. Melbourne: ACER. Dinham, S. & Scott, C. (2002). ‘Pressure points: School executive and educational change’, Journal of Educational Enquiry, 3(2), pp. 35–52. Duignan, P. & Bezzina, P. (2006, February). ‘Building Leadership Capacity for Shared Leadership in Schools – Teachers as Leaders of Educational Change’, Australian Centre for Educational Leadership International Conference, University of Wollongong. Evans, R. (1996). The Human Side of School Change. San Francisco: Jossey-Bass. Fullan, M. (2005). Leadership and Sustainability. Thousand Oaks, California: Corwin. Gronn, P. (2002). ‘Distributed Leadership’, in K. Leithwood & P. Hallinger, (Eds.), Second International Handbook of Educational Leadership and Administration. Dordrecht: Kluwer. Hargreaves, A. & Fink, D. (2004). ‘The Seven Principles of Sustainable Leadership’, Educational Leadership, 61(7), pp. 8–13. Harris, A. (2004). ‘Teacher Leadership and Distributed Leadership: An Exploration of the Literature’, Leading and Managing, 10(2), pp. 1–9. Harris, A. (2005). ‘Distributed Leadership’, in B. Davies (Ed.), The Essentials of School Leadership. London: Paul Chapman. Harris, A. (2008). ‘Distributed Leadership: According to the evidence’, Journal of Educational Administration, 46(2), pp. 172–188. Hattie, J. (2003). Teachers make a difference: What is the research difference?, available at: http://www.acer.edu.au/workshops/documents/Teachers Make a Difference Hattie.pdf. Hattie, J. (2007). ‘Developing Potentials for Learning: Evidence, assessment, and progress’, EARLI Biennial Conference, Budapest, Hungary, available at: http://www.education. auckland.ac.nz/uoa/education/staff/j.hattie/presentations.cfm Lambert, L. (1998). Building Leadership Capacity in Schools. Alexandria, Va.: Association for Supervision and Curriculum Development. Mulford, B. (2008). The Leadership Challenge Improving Learning in Schools. Melbourne: ACER. Mumford, A. (1995). ‘Learning in Action’, Industrial and Commercial Training, 27(8), pp. 36–40. NSW Department of Education and Training. (2003). Quality teaching in NSW public schools: Discussion paper. Sydney: Professional Support and Curriculum Directorate, NSW DET. Rowe, K.J. (2003). The importance of teacher quality as a key determinant of students’ experiences and outcomes of schooling, discussion paper prepared for the Interim Committee of the NSW Institute of Teachers, available at: http://www.nswteachers.nsw.edu.au/library/Rowe.html. Spillane, J., Halverson, R., & Diamond, J. (2001). ‘Investigating School Leadership Practice: A Distributed Perspective’, Educational Researcher, 30(3), pp. 23–28. York-Barr, J. & Duke, K. (2004). ‘What do we Know about Teacher Leadership? Findings from two Decades of Scholarship’, Review of Educational Research, 74(3), pp. 255–316.

Part IV

Conceptual Theoretical

Chapter 8

The Role of Sensemaking and Trust in Developing Distributed Leadership Karen Seashore Louis, David Mayrowetz, Mark Smiley and Joseph Murphy

Introduction Of all the “big” ideas now on the landscape of educational leadership, few are more prominent than “distributed leadership.” In a matter of a few short years, the idea of distributed leadership has evolved from a theoretical consideration of naturally-occurring social influence processes in school organization (e.g., Gronn, 2000; Spillane et al., 2001) to a mantra for reshaping leadership practice. More and more schools and school systems are attempting to develop distributed leadership. Increasingly, state education agencies and national education organizations are encouraging them to do so. Among the best known of these efforts in the United States has been the State Action Education Leadership Projects (SAELP), funded and promoted by the Wallace Foundation, the Education Commission of the States, and the Council of Chief State School Officers. At the time that this chapter was being prepared, many of the states that received SAELP grants were actively promoting the development of distributed or teacher leadership as part of these projects. To date, most empirical work on distributed leadership has sought to describe what it might look like in practice. A small but growing number of studies have explored outcomes. Our attention has focused on the development of distributed leadership and how various organizational factors might shape its course. For the past three years, we have been studying distributed leadership development in six secondary schools, each of which participated in an initiative to promote distributed leadership. In this chapter we explore two crucial factors in the development of distributed leadership – sensemaking and trust. We examine how the process of understanding (making sense) of distributed leadership and the nature of existing relationships in the school shape (a) the design of opportunities for distributed leadership; (b) how distributed leadership is enacted; and (c) the development of distributed leadership by school-level personnel. Our analysis presents a theoretical

K.S. Louis (B) Department of Educational Policy and Administration, University of Minnesota Minneapolis, MN, USA

A. Harris (ed.), Distributed Leadership, Studies in Educational Leadership 7, C Springer Science+Business Media B.V. 2009 DOI 10.1007/978-1-4020-9737-9 9, 

157

158

K.S. Louis et al.

argument with supporting evidence of how these two factors in distributed leadership development.

Distributed Leadership: A Job Redesign Perspective Our definition of distributed leadership follows from Firestone and his colleagues’ work on leadership functions (Firestone, 1989; Heller and Firestone, 1995; Mayrowetz and Weinstein, 1999) and from later work by Spillane and Scribner and their respective colleagues that views distributed leadership as leadership activity spread over leaders, followers, and a school’s context (Scribner et al., 2007; Spillane et al., 2001). We see distributed leadership as the sharing and spreading of leadership work across individuals and roles throughout the school organization. As Scribner et al. argue, the success of distributed leadership depends not only on individuals performing different leadership functions effectively; it also depends on new patterns of interaction and influence among people doing this work. Our work is organized around the idea that initiatives to develop distributed leadership are forms of work redesign (Mayrowetz and Smylie, 2004). As part of the organizational of leadership work, distributed leadership calls on teachers to conceive of their roles differently and to assume different responsibilities, mostly beyond the classroom and often for purposes of school-level improvement. As teachers’ work becomes redefined so too does administrators’ work, not only with regard to distributing particular leadership functions but also supporting redefined teacher work and creating conditions conducive to its success (Murphy, 2005). Our model of distributed leadership as work redesign proceeds from the assumption that a major factor in organizational improvement is the “person-job relationship” (Hackman and Oldham, 1980). In an earlier analysis we extended the original JCM to better align it with the development of distributed leadership in school organizational context (Mayrowetz et al., 2007). Our model emphasizes the centrality of sense-making in creating the motivation to work performance, and points to trust as an antecedents to and moderator of both sensemaking and the way in which leadership distribution is enacted, as well as featuring the importance of preexisting leadership patterns. Finally, how distributed leadership work is performed will relate to outcomes achieved. A highly simplified version of our model that highlights the assumptions and relationship to be explored in this chapter is shown in Fig. 8.1.

Sensemaking and Distributed Leadership There are a variety of perspectives on sensemaking and change in the educational literature, but one finding is clear: When teachers or administrators are confronted with a new set of practices (such as those implied by distributed leadership initiatives) their interpretations of it will determine whether they engage in significant change,

8 The Role of Sensemaking and Trust in Developing Distributed Leadership

(pre-existing) JOB CHARACTERISICS

• • •

Motivation Learning Sensemaking

Leadership Distribution Patterns

159

Org. Performance

Trust .

Fig. 8.1 Simplified job characteristics model

incremental change, or resistance. New programs and practices may be ignored or resisted, usually without serious consequences (Gold, 2002; Louis and Dentler, 1988). While many studies focus on individual responses to disruptions or demands for change (Zembylas, 2003), or the role of context and culture as conditions mediating individual change (Angelides and Ainscow, 2000; Gioia and Thomas, 1996), our focus is on a recent direction in the sensemaking literature: the development of collective interpretations of demands for change in school leadership. This approach assumes that significant disruptions to current patterns of behavior or expectations will induce the group to reassess what they are doing and how they are doing it. Once they have considered what is going on, and how it can be handled, they may choose to do nothing, or they may, with more or less coherent deliberation, change the way in which the organization is operating. This approach is consistent with the prevalent assumption that if the goal is to alter traditional patterns of responsibility and leadership in the school then groups are the most effective unit of change. In particular, it is hard to imagine how one would realize distributed leadership, which calls for more people to do more different kinds of work, without engaging in collective deliberation. Sensemaking is the process by which individuals and groups evolve shared understandings of their setting (Boje, 1991; Coburn, 2001; Kezar and Eckel, 2002; Weick and Roberts, 1993), in this case their understanding of who is a leader, and what being a leader means. We treat sensemaking not as an event, but is an on-going process. Individuals pay attention when something in their surroundings does not fit with their usual routines, and use their experience to find patterns that help to explain new situations. Similarly, collective sensemaking occasionally occurs as part of a deliberate activity (like strategic planning), but more often emerges from informal communication that leads to common actions or agreed upon activities (Coburn, 2001; Weick and Roberts, 1993). For example, in education, the nature of professional communities and dialogue has emerged as a powerful factor determining collective understanding of new initiatives (Coburn, 2001; Honig and Hatch, 2004; Spillane et al., 2002), as well as organizational learning, or the creation of coherent and shared explanations for “how we do things around here” (Bryk et al., 1999). Teachers work

160

K.S. Louis et al.

together, and learn from each other’s conversations, which leads them to interpret changes in their setting and practice in a consistent and collective manner even when reflection also occurs individually (Coburn, 2001; Craig, 1995; Louis et al., 2005). Making sense of any change effort such as distributed leadership is affected by the public discussions surrounding the construction of any effort to make this change, and by the initiative’s alignment with existing conditions in the school (Firestone et al., 1998; Spillane, Reiser et al., 2002). But statements about new expectations do not by themselves construct knowledge for teachers and administrators. When the focus is only on changing leadership structures and the teacher’s work roles, teachers may not see any connection to their main task, which is supporting student learning. If distributed leadership is framed, on the other hand, as an opportunity for teachers to change school and classroom conditions so that they can carry out their main job more effectively, they are more likely to see it as central to their work rather than an “add on,” rather like lunch duty or hall monitoring. Sensemaking is a form of social processing but not necessarily deep or reflective processing. Casual conversations and narratives can reflect superficial behavior expectations rather than addressing core assumptions about how the school should function (Craig, 1995). In order to create more fundamental change, both time and deeper challenges to embedded assumptions are needed (Huy, 1999; Kezar and Eckel, 2002). As Hofstede demonstrates, assumptions about leadership are deeply embedded in both organizational and national cultures (Hofstede, 1991). Because they are fundamental to our assumptions about how work settings operate, any significant change to patterns of leadership (beyond minor “leadership style” differences between individuals) is likely to be disruptive and possibly controversial. The paradox of distributing leadership is that it may require a significant “push” from the top of the organization (the principal) in order for more initiative to be taken up by other school professionals or even students and parents. The role of principals in creating the conditions for both distributed leadership and learning how to enact distributed leadership are rather obvious. First, they play a central role in determining the opportunities for sensemaking and organizational learning because they have a role to play in determining structures (time to meet and talk) and the allocation of other resources to any change activity (Marks et al., 2002; Spillane et al., 2002). Second, because they have traditionally been regarded as the “head” of the school and the person with the greatest legitimate influence over school operations, their behavior will determine the degree to which teachers trust that taking on new leadership roles will be rewarding and have long-term benefits to themselves and others.

Trust and Distributed Leadership Our perspective starts from the assumption that any collective action in schools will be affected by the level of trust among its members. In particular, our model

8 The Role of Sensemaking and Trust in Developing Distributed Leadership

161

assumes that trust will affect the way in which people make sense of any effort to change leadership patterns, and will also have a direct effect on any enactment of leadership (whether centralized or distributed). Despite growing consensus on the importance of trust in organizations, a widely agreed-upon definition of trust remains elusive. Most definitions share the view that trust is a psychological state of “perceived vulnerability or risk that is derived from an individuals’ uncertainty regarding the motives, intentions, and prospective actions of others on whom they depend” (Kramer, 1999, p. 571). Trust is an expectation that another party will not act opportunistically, will be honest, and will make a good faith effort in accordance with previous commitments. A number of scholars distinguish among different sources of trust including, for example, transactional obligations, sustained interactions, and moral authority (Bryk and Schneider, 2002; Louis, 2007). Kramer (1999) argues, however, that it is difficult to easily make such attributions. Trust can be influenced by predispositions, interaction histories, and reputation, as well as social categories (e.g., minister, doctor, attorney). It can also emanate from norms and formal rules of conduct that define and govern the contexts in which persons interact. In other words, trust is generally considered to be a dynamic phenomenon operating systemically within and across individual, group, and organizational levels (Tschannen-Moran and Hoy, 2000). It is usually developed provisionally and from that point it can strengthen, weaken, be breached, lost, and restored. Trust can have history and become an embedded property of organizations (Louis, 2007). It can apply to certain aspects of a personal or organizational relationship and not to others. Levels of trust can vary within the same organization even with regard to the same object, say the principal or the central office (Goddard et al., 2001). Trust has a variety of benefits that could have a significant impact on the developing of distributed leadership initiates, ranging from reducing the perceived need for bureaucratic controls to increasing the flow of help and assistance (Bryk and Schneider, 2002; Kramer and Cook, 2004; Tschannen-Moran, 2004). This may be particularly important in professional settings, like schools: According to Tschannen-Moran and Hoy (2000), people in high trust situations are more likely to disclose “accurate, relevant, and complete data about problems. They will also be more willing to share their thoughts, feelings, and ideas.” (p. 581) Trust allows for open exchange of information and creates a context in which problems can be disclosed and addressed before they are compounded. In other words, trust promotes cooperation which are at the core of efforts to make significant changes in the leadership patterns in schools. Trust can also foster change by itself, by establishing a sense of “safety” that in turn can support communication and critique, examination of taken-for-granted assumptions, and risk-taking when there is a need to change (Bryk and Schneider, 2002). Distributed leadership is likely to be seen as risky by teachers who have confined their work to their classrooms, and trust may thus promote a willingness to try out a new (and public) role. Trust makes collective action more feasible (Uphoff, 2000), and can also promote organizational citizenship, that is, making contributions

162

K.S. Louis et al.

beyond formal job requirements without expectation of recognition or compensation (Tschannen-Moran and Hoy, 2000). Particularly important for this study is the finding trust enhances support for, cooperation with, and desire for interaction with superiors (Tschannen-Moran and Hoy, 2000), and enhances confidence in the completeness and accuracy of information coming from superiors (Tschannen-Moran and Hoy, 2000). Such trust in leadership relates positively to work attitudes (e.g., job satisfaction), productivity, and performance (Dirks, 2000; Dirks and Ferrin, 2002). The absence of trust often produces the opposite results (Tschannen-Moran, 2004).

Sensemaking and Trust as Tools for Investigating Distributed Leadership Over a decade ago, but well into a decade-long effort to pour state and federal funds into “restructuring” efforts, many authors began to raise serious question: Why is it so hard to create deep change in the institutional characteristics of American education? (Grossman and Stodolsky, 1995; Tyack and Tobin, 1994). Among the taken for granted institutional characteristics of teachers’ work that emerged in the last half of the 20th century are, for example, the increasing specialization of teachers who work with small groups of students, the increase of planning time in fragmented, short blocks within the school day, and the “7 period day” (Miles, 1995). Others, however, pointed out that the problems of change cannot be fully accounted for by institutional rigidities, but are also affected by how individuals and groups experience reform (Hargreaves, 2002; Judith W. Little, 1996). Our approach assumes that both trust and sensemaking are conditions that affect reform. Current calls for distributed leadership, including substantial funding provided by national foundations, suggest that it is an effort to disrupt the “grammar of schooling” in high schools. Sensemaking is a crucial mechanism that sits at the heart of developing and exercising distributed leadership, but it occurs in a context that has been resistant to fundamental change, in part because of a significant divide between the perceived interests of teachers and administrators (Copeland, 2003; Elstad, 2008; Louis, 2007). The role of trust and sensemaking are best exposed as disruptions hit the school as a social system and its members respond to that disruption. We limit our discussion to those disruptions that are related to distributed leadership. Elements of the social system refer to several of our antecedent/moderators like trust, micro politics, organizational culture/mental models, and possibly structure. Based on the framework presented above, we propose two simple descriptive questions for investigation: 1. What school conditions cause teachers and administrators to ignore or try to make sense of current efforts to change leadership patterns in high schools? 2. Does the current call for distributed leadership constitute a strong or weak potential disruption to the current “grammar of schooling” in secondary settings?

8 The Role of Sensemaking and Trust in Developing Distributed Leadership

163

Our Study of Distributed Leadership Development Our study involved three-year longitudinal comparative case studies of six secondary schools in two mid-Atlantic states. Fieldwork began in 2004–2005, the year these schools began their distributed leadership initiatives, and continued through the end of the 2006–2007 school year. These cases were selected from a larger group of schools that were known to be experimenting with distributed leadership.1 Two were middle schools and four were high schools. They are located in urban, suburban, and rural communities, with enrollments ranging from just more than 500 students to approximately 1,500 students. Their racial and ethnic compositions ranged from predominantly African American to racially mixed to predominantly white. Proportions of low income students ranged from about one-fifth to almost two-thirds. Finally, levels of academic achievement ranged from relatively high, with more than three-quarters of students meeting of exceeding goals on state achievement tests to relatively low with only one-fifth to one-half meeting or exceeding goals. Each spring each school was visited for three-to-five days by one of the authors. Approximately 15 key informants at each school and central office were interviewed each visit. These informants included principals and assistant principals, teachers who participated in the distributed leadership initiatives and teachers who did not, central office administrators, and teacher union representatives. Interviews focused on the distributed leadership initiative, the development of leadership practice in the school, the relationship of the distributed leadership initiative to other improvement efforts, and factors that might influence any or all of it. Whenever possible, we observed distributed leadership team meetings and school personnel “at leadership work.” Mid-year interviews of each principal were also conducted. Each year we collected documents concerning these initiatives, the schools, and their districts. We used a common protocol to ensure consistency in data collection. As an internal “check,” we exchanged data and vetted our observations with one another at regular meetings. In addition, one of the authors had regular contact with representatives of each school’s distributed leadership team. His familiarity with all of the schools served as another “check” on the general accuracy of our fieldwork. We conducted thematic analyses of our data to search for evidence of ways that sensemaking and trust might relate to distributed leadership development in these six schools. We were guided by our model and by the literature that suggests how trust might function as an element of that model. We combed our cases for “stories” pertaining to the design of distributed leadership work, its performance, and perceptions of that work.

1 The schools were located in two states funded by the Wallace Foundation’s State Action for Educational Leadership Program (SAELP). Both states encouraged schools that were interested in distributed leadership to become involved.

164

K.S. Louis et al.

Our findings are presented in the form of two contrasting focal cases. Overton High School is our “positive” case, which will be used to illustrate the process of sensemaking around distributed leadership over a three year period, and Middle Fork High School our “negative” case, where we show the limits that low trust places on the sensemaking process.2 These cases were chosen to illustrate the major contours of the relationship between sensemaking, trust and distributed leadership development suggested by the literature and by evidence from our six-school sample. Space limitations prevent us from discussing all our study sites and curtail examination of all dimensions of the relationship between trust and distributed leadership development in the cases.

Distributed Leadership Development at Overton Overton is one of three non-selective vocational high schools located in a specialized vocational district. Vocational education enrollments and offerings have declined over the past fifty years, and the introduction of new accountability legislation appears to have increased this trend in U.S. comprehensive schools (Benavot, 1983; Thomas, 2004). This is not true for Overton or its sister schools, which are oversubscribed as schools of choice for the metropolitan area that they serve. Although a “successful” by most standards, Overton faced several challenges at the beginning of our study. Like many vocational programs during the 80s and 90s, Overton’s curriculum focused on “terminal” students who had limited aspirations for post-secondary education, and one long-term faculty member described it as “a last chance school.” In the last six or seven years, however, efforts to close programs with limited employment/career options and to open new programs focused on skilled options with potential for career advancement have begun to revitalize the school. A more persistent issue is lack of coordination and occasional friction between the vocational and “regular” teachers, who are located in different parts of the building and rarely collaborate (Little, 1995). As the principal noted in 2004, there are some teachers in both groups who had never been in the “other” part of the building. Overton became involved in distributed leadership when it wrote a proposal to the state to receive approximately $25,000 a year as part of the state’s leadership development initiatives, funded by the Wallace Foundation. The initial proposal emphasized the existing commitment to developing professional learning communities (PLCs), and the intersection of distributed leadership (DL) with this effort. Since the funding of the proposal, these interlocking themes have persisted – at least in the minds of administrators. While most teachers are eager to talk about their work, they are less likely to label changes as PLC/DL related. 2 Pseudonyms are used instead of the real names of our study schools. We promised each school and each person who participated in this research as much anonymity as possible. Any association between a pseudonym and the name of an actual school in the states where we conducted our study is purely coincidental.

8 The Role of Sensemaking and Trust in Developing Distributed Leadership

165

Changing Job Characteristics Teachers. Obvious changes to teachers’ work began to occur prior to the involvement in the distributed leadership initiative when the principal began a gentle but clear effort to close programs that were not serving students well, and to open new ones.3 Newly hired vocational coordinators were recruited from industry (often without previous teaching experience), and brought with them both a sense of commitment to teaching students who were “like us when we were in high school” and an entrepreneurial spirit. They were given a free hand to develop new programs, and were often provided with supplemental funding to reequip facilities. A teacher who had completely rejuvenated her program described the process: The teacher that was there prior to me had been here for a long time. And the first thing I did was bring the dumpster around the back, and I thought that anything that’s not relevant to today’s work environment is going to the dumpster. . . (the principal) supported me all the way.

Her experience was not unique. . ..As one (math) teacher said: There is just so many cells pulsing that connect (to our improvement work). And (the principal) says, I just find good people and I get out of their way.

At roughly the same time, a small group of teachers became concerned about the behavior of freshman students (who accounted for the majority of disciplinary problems in the school), and they soon volunteered to develop a 9th grade academy that would provide a link between career counseling/choice of vocational specialties, support in academic learning skills, and a strong advisory system. The 9th grade academy provided an important bridge between the academic teachers (who were most affected by behavior and low academic performance), and the career teachers (who wanted students to make more informed choices), and was universally popular. Overall, a small number of teachers were involved in entrepreneurial program development, but a larger number participated in a re-invented governance structure and task forces. Although the traditional administrative leadership team remained intact, a larger “Steering Committee” composed of the 4 vice principals, all department/program chairs and a number of other key faculty members met weekly with the principal to brainstorm and think strategically about the school’s future. A teacher pointed to the change: It used to just be a department chair meeting, and all of the chairs would sit around, they would say, do you have any problems, do you have any problems, . . ..(now) specific committees handle problems. . .and basically we’re supposed to be part of the visioning for the future of the school. . ..we get assignments to read and then we’ll talk about it. And if it’s important enough, it’ll go back, take it back to the departments, and talk about it. . ... (the principal’s) the engineer. He’s sitting on the engine, and he’s moving the train. But he doesn’t pretend to know it all.

3 We refer to this as “gentle” because program directors were either counseled out or chose to retire.

166

K.S. Louis et al.

In fact, there were so many committees that a new principal, who arrived at the school two years after our research began, cut a number of them. However, he expanded Steering Committee to include additional teachers who served as volunteer coordinators for the school’s newly identified PLCS. According to the principal, this expansion served to put teacher leadership issues at the forefront of the school’s “visioning committee.” The typical teacher who is not on the Steering Committee appears to be involved in between 3 and 5 other non-mandatory leadership assignments, ranging from membership on an ad hoc or standing committee, to coaching events such as Skills USA. Committee work is viewed usually viewed as problem solving – action, not talking. Administrators. The effort to change the school began with the appointment of the current superintendent in the late 1990s from an associate superintendent position in the district; he then appointed the principal who was leading the school during the first two years of our study from his former position as vice-principal. Because both the superintendent and principal were predisposed to share leadership with others, the main challenge for the administrative team was to prepare the novice vice principals to carry the message when they moved up.4 This meant giving each of them opportunities to exercise leadership of different kinds, and the allocation of responsibilities among them included some direct supervision of departments, some responsibility for student issues, and some supervision of non-teaching staff. All of the vice principals were involved in the PLC/DL initiative, helping to coordinate all-staff retreats and work with committees. A new Vice Principal commented about the difference between her previous school and Overton: (My former principal’s) job description was his job description and he did that. And mine was mine and I did that. And Bill was over in guidance, and he did his thing. In this team there’s a flow . . . To a certain degree what you supervise is what you supervise – But they’re not afraid necessarily to cross boundaries.

School Culture The most obvious characteristic of Overton’s culture was the high level of trust and respect among all parties. Teachers, in particular, believed that the administrators were committed to making sure they had the resources they needed to make the school better. The following reflects the statements of many: . . .if you really can show that there is a need, not because of the state test, but because that’s what the kids need, then they’re going to find a way. And if we can’t find a way, (the principal) will say, “we’ll go someplace else and we’ll help you get what you need.” And they don’t pretend to be experts in it. . .they say, you’re the expert.

4 During our study a new principal, who had been a vice principal at another school, was appointed, one of Overton’s vice principals was appointed as a principal at another school, and two other vice-principals rotated to other schools.

8 The Role of Sensemaking and Trust in Developing Distributed Leadership

167

The school’s teachers have also gone from thinking of their students as “last chance” material to emphasizing their potential – and their own responsibility for caring. One vocational teacher, who had been at the school for nearly five years, noted that the school emphasizes personal relationships: I know those kids inside and out. . .We’re not just a teacher. . .We have a connection, and I think that’s what I love most about my job.

The lack of serious behind the scenes micro-politics was, according to most people, largely due to the fact that almost all teachers believed that they could influence the school’s decisions and directions. According to one administrator: They know that they will be heard, and they know that we will, listen to them with an open mind and take heed of what they’re saying; we’re willing to share. It may be that where I was (working before), people just didn’t complain because they didn’t think anything would happen if they did or not.

A new teacher made the same point, noting that he had expressed concerns about the performance of a teachers’ aide in his class, and that the principal allowed him to make the decision about whether to let her go: And what I really liked about (the principal’ reaction), and again, I respect him so highly for this, he empowered me to make that decision. He said, here’s the deal. Keep her, don’t keep her. But you have to live with the decision.

School Structure Creating professional communities and distributing leadership in a school that has twice as many departments as most schools (due to the many vocational areas), and many specialty career programs staffed by one or two people, is a daunting task. The scheduling issues are complicated by the need for more advanced students to spend blocks of time in their vocational specialty, and for the eligible senior students to be off-campus for much of the day doing co-op work. The availability of funds for all-staff retreats has helped, to some extent, to compensate for the nearly impossible task of creating common preparatory periods during the day, and the equally difficult job of scheduling after-school meetings (which teachers would voluntarily attend) when the vocational teachers often used that time to contact potential employers or work advisors. In spite of much effort to create committees and events in which vocational and “regular” teachers could work together, the integration of the two halves of the school remains incomplete, and leadership positions, other than those with clear administrative responsibilities, rarely transfer across the boundaries. As a vice principal noted on our last visit, the school has many teacher leaders, but has not cohesively focused its energies on improving student achievement. The new principal was less concerned about concentrating on student achievement scores, but noted that there was no way around it – the school was still a hierarchy, in which some decisions could only be made by administrators. His goal is to reassure teachers that this “fact” doesn’t

168

K.S. Louis et al.

diminish the need to have many more people stirring the pot and creating new activities. A key structural feature of Overton that cannot be underestimated is turnover, whose effect is largely positive. During a five year period between 2003 and 2008, nearly half of the faculty retired or left. While the school’s “last chance” reputation still dominated among staff and the public at the beginning of this period, the principal and the department chairs (who were heavily involved in the hiring process) nevertheless had the opportunity to restaff the school. They chose, where possible, younger teachers who had some experience and wanted to come to the district and to Overton because of the reputation is was acquiring as “a good place to work.” They also actively recruited novice (but not necessarily young) teachers to the new vocational areas – often people who were initially uncertified, but had a strong passion for promoting a new style of vocational education. It is hard to underestimate the level of energy that many of these new recruits brought with them, in part because they felt that they had joined a team with a mission to improve the life chances of students like those who chose Overton rather than the local comprehensive high school. Because so many of the teachers are new, the PLC/DL initiative is not a disruption, but just the way things are at the school.

Trust, Sensemaking and DL Outcomes in Overton The distributed leadership initiative landed on rich soil at Overton, and local conditions increased the rapidity and depth of its penetration into the routine practices of the school. There was little change over the three year period in which Overton was studied, in spite of the continuous flow of new teachers, the retirement of the much-loved principal, his replacement with a much younger person from another school in the district (teachers were involved in the interviews and concurred with the choice), and the departure of three of the original vice principals. This level of personnel change might have overwhelmed an initiative that was less well integrated into the school, but DL was never very distinct because it was embedded as a feature of the school’s routines before it was formally inaugurated. A great deal of institutionalization was due to the leadership style of the former principal, who was viewed by all as someone who deliberately gave power to others, and who preferred a stimulating conversation about what should be done than making a decision himself. The influx of new faculty, and the openness of the school to program innovation from teachers, created a spirit of entrepreneurship (“It’s a good idea – why don’t we try it?”) that reflected an approach to innovation that encouraged people to join. Strong support from the Superintendent was also noted by a variety of respondents. The principal – and the superintendent – were viewed as articulate supporters of the perspective that resources should follow priorities: The first priority was students and their learning, and the second was teachers and their creativity. The

8 The Role of Sensemaking and Trust in Developing Distributed Leadership

169

new principal, who was in place starting in the fall of 2007, stated very strongly that he supported the previous principal’s approach and planned to implement it as consistently as he could, Teachers consistently expressed the sense that they were valued, and that they had as many opportunities to exercise influence as they wanted. Teachers thus became additional story-tellers supporting the important role of teachers as leaders in the school. We heard about the heroism of the group that developed the 9th grade academy not only from them, but from others; we learned about the renewal of the science department’s curriculum not from the chair of science, but from a math teacher. The importance of participatory development and planning was underlined by the new principal, who argued that the somewhat unwieldy expansion of the Steering Committee to include more teacher leaders was important for symbolic reasons, as well as giving an opportunity for more people to contribute to discussions about the school’s future. The key issue here is how teacher leadership was defined: for most it meant entrepreneurship and individual initiative in program development rather than involvement in larger collective work. School-wide PLC committees were established after more than two years of planning, but are less well understood by most as a vehicle for exercising influence. The difficulty that many teachers (and administrators) in the school had in identifying distributed leadership as a distinctive initiative exemplifies the degree to which it was consistent with the directions in which the school was already headed. Although a name was given to it in order for the school to qualify for state funding that helped to support faculty retreats, distributed leadership was inseparable in most people’s minds from the effort to develop more intense conversations among faculty around solving problems that were identified by faculty – and everyone was able to identify PLCs as a feature of the school’s work. Teachers were more likely to spontaneously talk about the support that they had for stretching, taking on new challenges, and being creative than they were in identifying themselves as leaders. As the chair of the math department pointed out, she often had to tell teachers when they were being leaders because they saw what they were doing, whether chairing a committee or mentoring a new staff member, as simply part of their job. Of course Overton is not a perfect participatory democracy. As the principal points out, administrators still have decisions to make, and teachers still spend most of their time in classrooms. There are still curmudgeons and those who want to be left alone to teach. The depth of formal and informal leadership participation among the faculty is, however, unusual for a large high school, and the spirit of collective responsibility for creating new programs and getting the work done suggests a fuzziness of boundaries between the roles of teacher and administrator that corresponds to the similar flow and flexibility within the administrative team. While many factors have contributed to Overton’s ability to make sense of distributed leadership while experimenting with the form that it should, the importance of cultivating trusting (and caring) relationships between all parties underlies the ability of all parties to move forward.

170

K.S. Louis et al.

Distributed Leadership Development at Middle Forks One of a few of high schools in a mid-size urban school district, Middle Fork High School had the reputation of being the worst. This school of about 1,100 students is located in a section of the city that has experienced significant white flight. The now predominantly African-American neighborhood surrounding the school wrestles with growing poverty and crime. Student enrollment has steadily fallen nearly two-thirds since the early 1970s. In 2004–2005, student enrollment was about 80% African American, 14% Latino, and 6% white. Almost two-thirds of students were eligible for free or reduced-price lunch, and student mobility was also high. While almost half of students met or exceeded grade-level goals on state achievement tests in reading and language arts, only 20% met or exceeded goals in mathematics. Gang activity from outside the school had predictably spilled in. In consultation with a newly appointed principal, the district administration selected Middle Fork to participate in the state’s distributed leadership initiative. Both district and school-level administrators saw distributed leadership as a way to identify teachers and others in the school who would “step up” and ease the burden on the school’s administrators. The results of an internal faculty survey conducted that year also reflected a desire among a number of teachers to take a more active role in school leadership. The hope that distributed leadership could help galvanize reform in Middle Forks proved, however, to be elusive.

Changing Job Characteristics Administrators’ work. We began the Overton story by talking about changes in teachers’ work, but understanding the Middle Forks story requires starting with administrators. Like many urban schools, the history of administrative leadership at Middle Fork has been unstable and unsettling. The school had twelve principals in the past 25 years. The period leading up to the distributed leadership initiative was especially tumultuous. In the middle of the 2001–2002 school year, the recently appointed superintendent fired Middle Fork’s principal and vice principals and replaced them with a hand-picked administrative team. This seven-member team was composed of subject matter specialists from the district’s central office and administrators from other high schools in the district. The team met in secret for several weeks (in an unmarked room in the district office) to strategize what to do in its first months at Fox River. When the superintendent came to the school to announce the “take over” and introduce the team, he berated Middle Fork’s teachers for failing their students and he promised change. One teacher recalled that the superintendent seemed “very threatening.” “He would get us,” she continued. “He would be sure to get us. So that’s what we expected to happen.” Middle Fork’s new administrative team continued working in relative isolation from the faculty during its first few months at the school, in “a foxhole” as one member put it. The team felt a strong collective mission to clean up lawless

8 The Role of Sensemaking and Trust in Developing Distributed Leadership

171

hallways, in its words, the “wild west city.” The team administered a “tough love” style of student discipline, requiring that students enter the school through metal detectors and later instituting a strict uniform policy. Administrators took primary responsibility for enforcing the new rules, calling on only a few teachers and guidance counselors to assist. They believed that most teachers had given up on students, had stopped enforcing rules, and had excused poor academic performance by blaming students and their families rather than taking responsibility for their own failures. Teachers work. The administrative team began to “lay down the law” to the faculty. Many teachers were insulted by what they perceived to be an arbitrary “crackdown.” One teacher recalled: When they came in, it wasn’t nice at all. You were told, “This is what you’re going to do and that’s it.” You’ll be written up for this and written up for that, and it’s like my mom’s at home. . ..

In response, many teachers withdrew into their classrooms and a number of them stopped volunteering for extra duties and working with after-school programs. A very small number of staff, mostly those released from other duties to do so, continued to work sporadically on formal committees ancillary to the core operation of the school. A few teachers, support staff, and parents were elected by peers to a School Leadership Committee (SLC), required of the school under a state accountability program and constituted to advocate on behalf of students. But overall, the faculty took solace in the school’s history of administrative turnover and in the belief that “this too shall pass.”

School Culture By the fall in which our study began, tensions between faculty and administration remained high. The pre-existing distrust was exacerbated by new actions by the administrative team, which continued to believe that too many teachers lacked respect for students and failed to support them adequately. The team turned its attention more and more toward students and, according to some teachers, went to excessive lengths to support them. Administrators listened to students’ complaints about individual teachers and with increasing regularity questioned teachers about what they might have done to instigate the problems. Teachers began to doubt that they could rely on administrators’ support against what they considered to be an unruly student body. One teacher remarked that administrators are “more pro for the student than for the staff.” During the summer before the 2004–2005 school year, the superintendent made the administrative team’s appointment permanent. He named the head of the team Middle Fork’s principal and the other team members vice principals. The faculty interpreted this action as a sign that things were not going to change. Like the teachers, the principal and vice principals were under a great deal of stress. Their psychological and physical health began to suffer. Two of the original members of the administrative team retired rather than stay on at Middle Fork.

172

K.S. Louis et al.

Soon after the start of the 2004–2005 school year, another vice principal began a year-long medical leave. That school year began with no sign that tensions would subside. Predictably, few teachers embraced the idea of the distributed leadership initiative. Most were skeptical and reticent to join in. According to one of the vice principals, “They’re gun shy. I think they’ve been beat so many times that they’re afraid to come out of that foxhole because they’re going to get their heads blown off.”

School Structure The story of structural changes related to distributed leadership in Middle Forks is a simple one: No real change. Not surprisingly, it took some time for Middle Fork to establish a distributed leadership team, and its initial efforts to change the decision making structure were undermined by a variety of additional factors. At first, the team consisted of the principal, a vice principal, a guidance counselor, and a teacher. The counselor and teacher were told by administrators that their participation was mandatory. The team was supposed to attend monthly off-site training sessions on collaborative leadership development with teams from other schools in the area. Actively involved in many other projects and frequently called to meetings at the central office, the principal rarely attended the sessions. The vice principal who took a medical leave and was replaced on the team by the newest member of the administration. By January 2005, the sole teacher member quit the team because she believed that distributed leadership would not “ever happen” at Middle Fork. The principal was furious. The new vice principal and counselor urged the principal to appoint additional teachers to the team. The principal selected five teachers – two veterans and three teachers new to the school – and for the remainder of the school year Middle Fork had a larger group attending the training sessions. Still, according to one team member, the principal attended “maybe one session.” By the end of the 2004–2005 school year the distributed leadership team had done little except to attend the training sessions sponsored by the state. The full team returned intact for the 2005–2006 school year. The vice principal, who returned from medical leave, was told by the principal to join the team. The team began attending training sessions again and like most of the previous year without the principal. Gradually, the team became more engaged and more vocal.

Continuing the Saga of Distrust and Distributed Leadership At one training session late in the fall Middle Fork’s team was asked to discuss issues at the school while team members from other schools would observe, take notes, and give feedback. During the exercise, a Middle Fork teacher recalled an incident where several students had pushed and physically injured a teacher, who later claimed that the school administration did not make a strong case for stu-

8 The Role of Sensemaking and Trust in Developing Distributed Leadership

173

dent expulsion. Other team members joined in, expressing concerns about the poor relationship between the school’s faculty and administration. As word about this incident traveled back to Middle Fork, the school administration were incensed, and even some teachers were angry, feeling that the airing would make the school’s bad reputation worse. Through one of the vice principals on the team, the principal reprimanded the teacher who had told the story at the training session and also (for the first time) began to accompany the team to the state training sessions. According to one team member, after the principal starting attending the meetings regularly “everyone else was mute.” Other opportunities to expand teacher involvement were, on the other hand, overlooked. District officials presented a proposal to restructure the school into smaller learning communities. Teacher members of the first distributed leadership team felt some excitement and were eager to engage the task. But Middle Fork’s administration decided that it would be unwise for the school to embark on this initiative when it was already engaged in other improvement activities that included a “paperless” office and teacher professional development with another outside consultant. The cycle of cynicism and decreasing involvement continued. As the end of the second year approached, Middle Fork’s administrators were struggling to figure out how they were “best going to use” the staff on the leadership team. In a typical distributed leadership team meeting, a first-year technology coordinator was tapped by the administration to present the plan and lead a discussion about how the school could meet a district professional development requirement, during which less than half of the team members in attendance said a word. After the meeting, a few teachers confided that felt they were being asked to “rubber stamp” the administration’s work. A few administrators saw it differently, considering the plan “only a draft,” but, the principal was sending a different message, that she wanted the leadership team “to be the cheerleaders, to go out and sell it. . ., to take ownership of it.” Eventually, the plan was approved and then shelved. Moreover, the revolving door of administrative leadership was turning again. During the first months of the 2006–2007 school year, one of Middle Fork’s vice principals became the principal of another school in the district, and the principal was called to the central office to assume the position of special assistant to the superintendent to lead a district-wide high school reform initiative. One of the vice principals, a man with a long history in the district, the school, and community, was appointed to succeed her. More gregarious and less controlling than his predecessor, he began to make earnest attempts to involve teachers, by inviting them to help interview candidates for a vice-principal’s position, and developing a plan to involve teachers in planning for small learning communities. A vice-principal in charge of supervising the faculty planning committee grew impatient and thought some were trying to organize a “coup” and develop school policy themselves. She distributed copies of what she expected to be final list of teacher committees and walked out of the meeting. Although the principal tried to patch up this debacle, in the end only a few committees were directly related to school improvement (e.g., teacher

174

K.S. Louis et al.

professional development) rather than traditional teacher roles such as the planning the prom and mentoring students.

Trust, Sensemaking and DL Outcomes in Middle Forks Perceptions of the distributed initiative at Middle Fork were colored from the start by the larger tensions between the school administration and the faculty. There is little evidence to suggest that the initiative was perceived by many teachers or the administration as a serious opportunity to develop leadership at the school or to develop the school’s capacity for improvement and future performance. From the beginning school and districts administrators saw some benefit in a relatively narrow and self-serving way – as a means to shift some of its workload to others, as an opportunity to scout for and groom “future administrative talent,” and an opportunity to cultivate a base of support for their ideas and initiatives. Many teachers saw the initiative as a way for the administration to pass off work, to have them “step up and share some of the leadership responsibilities.” The narrow range of opportunities for meaningful leadership work presented to the cohorts turned out to be less than met the eye as the administration curtailed, rejected, or abandoned them, leaving teachers with a sense that stepping up was a responsibility with no rewards. They saw the administration directly or indirectly controlling the entire process, and what little feedback the team received was largely negative. For most teachers in Middle Forks, distributed leadership initiatives reinforced their belief that they should not trust either the goodwill or the support that they would get from administrators. By the end of the third year, even committed teachers on the TLC were struggling to find a way to be more than “a bridge” or conduit to take a few ideas from the administration to the faculty and staff. They wanted to contribute their own thoughts and help shape school initiatives too, and a few teachers inside and outside the TLC saw in the new principal a glimmer of hope.

Discussion and Conclusion Our descriptive analysis of Overton and Middle Forks High School has ranged broadly, but covered the major elements in the model that we presented in Fig. 8.1. The two cases were selected primarily because they represent a useful comparison of the way in which initial conditions and subsequent behaviors, both on the part of teachers and administrators, condition the outcomes of a formal initiative to try and create greater teachers leadership and involvement in secondary schools. We summarize the differences between the two schools using the key elements of our model – job characteristics, sensemaking process, trust, and evidence of distributed leadership – in Table 8.1. Drawing from the literature, our model, and the two cases, we can begin to answer the two questions that we posed at the beginning of the chapter.

8 The Role of Sensemaking and Trust in Developing Distributed Leadership

175

Table 8.1 Sensemaking, trust, and distributed leadership in two schools Overton high school

Middle forks high school

Pre-existing job characteristics

Entrepreneurial faculty behavior; high Individual and collective sense of failure; weak teacher autonomy and professionalism; involvement outside the supportive leadership from classroom; leadership turnover school/district; high level of leads to gap between teachers resources; weak internal and administrators; no clear communication stimulated recent teacher leaders; limited emphasis on PLCS; well-regarded professionalism/professional “old-timer” teacher leaders development.

Sensemaking process

Principal viewed as detached from Principal promotes DL and DL initiative; successive sensemaking (tolerates ambiguity); interruptions of efforts to active committee work provides develop teacher leaders deepen insight into connection between administrator-teacher gap; DL individual professionalism, and DL; seen by teachers as separate gradual emergence of new teacher from other school improvement leaders; merging of PLC and DL work; negative stories about initiatives; teacher role in undermining teachers circulate hiring/socializing new principal within the faculty; DL seen as deepens sense of influence; DL an effort to shift work from (informally) seen as part of efforts administrators to teachers; some to improve the school’s programs teachers maintain optimism and performance. when new principal is hired.

Trust

High levels of initial distrust are Some initial mistrust between confirmed by principal vocational and “regular” teachers indifference to DL initiative; a diminishes with joint committee few episodes confirm work and departure of willingness of administrators to “curmudgeon” faculty; trust in limit teacher involvement; administrators persists through teachers rarely speak their leadership turnover; trust in minds in public; teachers report administrative support explicitly “punishment” incidents; supports risk-taking on the part of teachers individually fearful; teachers; teachers “rewarded” by teacher-teacher communications peers for initiative; trust becoming appear truncated. an embedded cultural characteristics

Evidence of distributed leadership

Increasing number of teacher leaders emerge over three years; time-consuming committee structure involves almost all teachers; DL survives principal turnover.

Very limited collaborative effort with only a few tenured teachers participating; non-tenured teachers are not involved; “distribution” limited to principal choices

1. What school conditions cause teachers and administrators to ignore or try to make sense of current efforts to change leadership patterns in high schools? First, we see that trust helps to determine who will perform distributed leadership work (both delegating and “stepping up”) and the characteristics of the leadership work that is designed (meaningful, significant, with implications for student learning). If trust is strong, designers of DL activities may be more likely to:

176

K.S. Louis et al.

r r r r

develop distributed leadership work that extends to a wider variety of tasks on a wider range of significant subjects than if trust is weak or distrust reigns. design work that is more challenging and “risky” and that may also be more meaningful, significant to the organization, and more readily identified with. introduce fewer bureaucratic controls and extend greater autonomy to those who assume distributed leadership work. more likely to incorporate processes of collaboration and open, bi-directional communication, joint problem solving, full exchange of information, honest assessment, and unbridled feedback.

These observations are consistent with the literature, but also help us to understand the differences that were apparent in Overton and Middle Forks from the beginning. Second, our model suggests that it is not only the presence of particular design characteristics that matters to the performance of distributed leadership; it is also how participants make sense of these design characteristics. How they are initially perceived and how they are modified over the longer haul are both important. For example, an initiative that provides for wide task variety was seen and experienced much more narrowly in Overton, where teachers were highly entrepreneurial in their own classrooms and departments, but saw little connection between these opportunities and wider leadership. Work that calls for substantial autonomy may also be seen and experienced as much less. Middle Forks teachers’ weak response to the “opportunity” to design a new committee structure is a clear example. In each of these examples, perception and experience may further moderate the influence of design. High levels of trust are likely to promote or positively amplify how people make sense of the design characteristics of distributed leadership. On the other hand, low levels of trust or distrust are likely to suppress or negatively affect perceptions. A related assumption is that the performance and outcomes of distributed leadership work will be moderated by the context in which that work is developed and performed. A committee with a certain charge in one context is not the same as a committee with the same charge in another. As teachers and administrators on the committee relate to each other around “joint work” their perceptions will change, but will consistently be affected by trust. In a high trust setting like Overton, administrative gaffes are noted, but quickly forgotten; in Middle Forks similar incidents become part of the stories that are told from year to year. The role of key administrators is particularly important in determining how teaches make sense of a distributed leadership initiative. When initial tasks that require joint teacher effort are meaningful and connected with the core classroom work of teachers (developing a 9th grade academy in Overton), the result is different from one that is less meaningful (reviewing a district-mandated professional development document in Middle Forks). Interestingly, relational trust was built initially built in Overton because administrators tended to stay out of teachers’ classrooms and gave them a

8 The Role of Sensemaking and Trust in Developing Distributed Leadership

177

high degree of autonomy with program development; similar lack of classroom oversight in Middle Forks was not viewed as supportive because it was not accompanied by encouragement for risk and change. In other words, the same general behavior (limited direct instructional leadership from administrators) is interpreted differently in the two schools, with different effects on teachers’ willingness to take responsibility for program change and other assorted leadership tasks. In terms of sensemaking what is particularly striking about the difference between Middle Forks and Overton is the degree to which teachers’ stories end up determining the sensemaking process and its outcomes. While it would be a stretch to say that teachers had become the primary sensegivers in Overton, with an almost entirely new administrative team, they were the keepers of the stories about the school’s transition from traditional to teacher-powered. In Middle Forks, on the other hand, there was no coherent story that incorporated distributed leadership. It remained a concept rather than any set of practices that could be described. The only events that stand out in Middle Forks stories are instances of failure, where teachers thought they were supposed to “step up” but found that they were marginalized in making real decisions. Collective sensemaking around distributed leadership occurred in a context favorable to its development Overton. In turn, the sensemaking was seen as only a minor disruption, over one-quarter of the faculty engaged the disruption deeply but all had a chance to process it. The end result was that many became sensegivers and kept the momentum for teacher leadership going during significant personnel turnover among faculty and administration. This process impacted the organizational conditions again. The cultural mores that existed in the school rewarding teacher initiative were strengthened and through the committee structures, those mores began to shift to favor teacher activity that was collective and crossed disciplinary boundaries and even what is traditionally the administrative sphere. 2. Does the current call for distributed leadership constitute a strong or weak potential disruption to the current “grammar of schooling” in secondary settings? Based on only two cases, the answer is “possibly” and “under some conditions.” No one visiting Overton would argue that it represents a dramatically different vision of high school education, but there is evidence that it has been successful in meeting some key challenges: Providing “average” students with workforce skills while preparing them for entry into the post-secondary education system, and creating an equitable environment in which the “gaps” in achievement and graduation rates are limited. The usual expectations, which include a sulky faculty moored in disciplinary silos has also been challenged, largely through the DL/PLC initiatives. The same initiative (with a similar level of external resources) had little impact in Middle Forks. In some ways, Overton and Middle Forks present an interesting contrast because in both cases there were efforts to blend a DL initiative with another innovation that could have become a vehicle for changing the grammar of

178

K.S. Louis et al.

schooling and teacher leadership (Professional Learning Communities in Overton and Small Learning Communities in Middle Forks) In Overton, the distributed leadership was neatly folded into existing initiatives (the PLCs) and fairly well exploited for the development of teacher leaders. In Middle Forks, in contrast, the two initiatives remained separate in most people’s minds. This contrast illustrates the way in which sensemaking is responsive to the presence or absence of the ability to weave initiatives together to create a wider story of change than exists of they remain separate. But while we did not organize our cases to focus on leadership, the Middle Forks and Overton stories again draw attention to the importance of superintendents and principals in creating opportunities for real change. Telling that leadership story requires moving beyond distributed leadership to the role, beliefs and behaviors of key formal leaders and, in particular, to their ability to help steer a flexible design that can, over a period of years, create systematic opportunities for engagement and risk taking among secondary school teachers. That story will be told in a later analysis.

References Angelides, P., & Ainscow, M. (2000). Making sense of the role of culture in school improvement. School Effectiveness and School Improvement, 11(2), 145–163. Benavot, A. (1983). The rise and decline of vocational education. Sociology of Education, 56(2), 63–76. Boje, D. M. (1991). The storytelling organization: A study of story performance in an office- supply firm. Administrative Science Quarterly, 36(1), 106–126. Bryk, A., Camburn, E., & Louis, K. S. (1999). Promoting school improvement through professional communities: An analysis of Chicago elementary schools. Educational Administration Quarterly, 35(4), 707–750. Bryk, A. S., & Schneider, B. (2002). Trust in schools: A core resource for improvement. New York: Russell Sage Foundation. Coburn, C. E. (2001). Collective sensemaking about reading: How teachers mediate reading policy in their professional communities. Educational Evaluation and Policy Analysis, 23(2), 145–170. Copeland, M. (2003). Leadership of inquiry: Buidling and sustaining capacity for school improvement. Educational Evaluation and Policy Analysis, 25(4), 375–395. Craig, C. J. (1995). Knowledge communities: A way of making sense of how beginning teachers come to know in their professional knowledge contexts Curriculum Inquiry, 25(2), 151–175. Dirks, K. T. (2000). Trust in leadership and team performance: Evidence from NCAA basketball. Journal of Applied Psychology, 85(6), 1004–1012. Dirks, K. T., & Ferrin, D. L. (2002). Trust in leadership: Meta-analytic findings and implications for organizational research. . Journal of Organizational Psychology, 87(4), 611–628. Elstad, E. (2008). Towards a theory of mutual dependency between school administrators and teachers. Educational Management Administration & Leadership, (36), 3, 393–414. Firestone, W. (1989). Using reform: Conceptualizing district initiative. Educational Evaluation and Policy Analysis, 11(2), 151–165. Firestone, W., Meyrowitz, D., & Fairman, J. (1998). Performance-based assessment and instructional change: The effects of testing in Maine and Maryland. Educational Evaluation and Policy Analysis, 20(2), 95–113.

8 The Role of Sensemaking and Trust in Developing Distributed Leadership

179

Gioia, D. A., & Thomas, J. B. (1996). Identity, image, and issue interpretation: Sensemaking during strategic change in academia. Administrative Science Quarterly, 41(3), 370–403. Goddard, R., Tschannen-Moran, M., & Hoy, W. (2001). A multi-level examination of the distribution and effects of teacher trust in students and parents in urban elementary schools. Elementary School Journal, 101(1), 3–19. Gold, B. (2002). Social Construction of Urban Education: New Jersey Whole School Reform and Teachers’ Understanding of Social Class and Race: Pace University. Gronn, P. (2000). Distributed properties: A new architecture for leadership. Educational Management and Administration, 28(363–383). Grossman, P., & Stodolsky, S. (1995). Content as context: The role of school subjects in secondary school teaching. Educational Researcher, 24(8), 5–23. Hackman, J. R., & Oldham, G. R. (1980). Work redesign. Reading, MA: Addison-Wesley. Hargreaves, A. (2002). The emotional geographies of teaching. Teachers College Record, 103(6), 1056–1080. Heller, M., & Firestone, W. (1995). Who’s in charge here? Sources of leadership for change in eight schools. Elementary School Journal, 96(1), 65–86. Hofstede, G. (1991). Culture and organizations: Software of the mind. London: McGraw-Hill, Inc. London: McGraw Hill. Honig, M., & Hatch, T. (2004). Crafting coherence: How schools strategically manage multiple, conflicting demands. Educational Researcher, 33(8), 16–30. Huy, Q. N. (1999). Emotional capability, emotional intelligence, and radical change Academy of Management Review, 24(2), 25–345. Kezar, A., & Eckel, P. (2002). Examining the Institutional Transformation Process: The Importance of Sensemaking, Interrelated Strategies, and Balance. Research in Higher Education, 43(3), 295–328. Kramer, R., & Cook, K. S. (2004). Trust and distrust in organizations: Dilemmas and approaches. New York: Russell Sage Foundation. Kramer, R. M. (1999). Trust and distrust in organizations: Emerging perspectives, enduring questions. In Annual Review of Psychology (Vol. 50, pp. 569–598). Palo Alto: Annual Reviews Inc. Little, J. W. (1995). Contested ground: The basis of teacher leadership in two restructuring high schools. Elementary School Journal, 96(1), 47–63. Little, J. W. (1996). The emotional contours and career trajectories of (disappointed) reform enthusiasts. Cambridge Journal of Education, 26(3), 345–359. Louis, K. S. (2007). Trust and improvement in schools. Journal of Educational Change, 8(a), 1–24. Louis, K. S. (2007). Trust and improvement in schools. Journal of School Change, 8(1), 1–24. Louis, K. S., & Dentler, R. (1988). Knowledge use and school improvement. Curriculum Inquiry, 18(1), 32–62. Louis, K. S., Febey, K., & Schroeder, R. (2005). State-mandated accountability in high schools: Teachers’ interpretations of a new era. Educational Evaluation and Policy Analysis, 27(2), 177–204. Marks, H., Louis, K. S., & Printy, S. (2002). The capacity for organizational learning: Implications for pedagogy and student achievement. In K. Leithwood (Ed.), Organizational learning and school improvement. Greenwich, CT: JAI. Mayrowetz, D., Murphy, J., Louis, K. S., & Smylie, M. (2007). Distributed leadership as work redesign: Retrofitting the jobs characteristics model. Leadership and Policy in Schools, 6(1), 69–101. Mayrowetz, D., & Smylie, M. (2004). Work redesign that works for teachers. In M. A. Smylie & D. Miretzky (Eds.), Teacher workforce development: The one hundred and third yearbook of the National Society for the Study of Education, Part I (pp. 274–302). Chicago: National Society for the Study of Education. Mayrowetz, D., & Weinstein, C. (1999). Sources of leadership for inclusive education: Creating schools for all children. Educational Administration Quarterly, 35(4), 423–449.

180

K.S. Louis et al.

Miles, K. H. (1995). Freeing resources for improving schools: A case study of teacher allocation in Boston Public Schools. Educational Evaluation and Policy Analysis, 17(4), 476–493. Murphy, J. (2005). Connecting teacher leadership and school improvement. Thousand Oaks, CA: Corwin. Scribner, J. P., Sawyer, R. K., Watson, S. T., & Myers, V. L. (2007). Teacher teams and distributed leadership: A study of group discourse and collaboration. Educational Administration Quarterly, 43(1), 67–100. Spillane, J., Diamond, J. B., & Burch, P. (2002). Managing in the middle: School leaders and the enactment of accountability policy. Educational Policy, 16(5), 731–762. Spillane, J., Halvorson, R., & Diamond, J. (2001). Investigating school leadership practice: A distributed perspective. Educational Researcher, 30(3), 23–28. Spillane, J., Reiser, B. J., & Reimer, T. (2002). Policy Implementation and Cognition: Reframing and Refocusing Implementation Research. Review of Educational Research, 72(3), 387–431. Thomas, D. W. (2004). Perceived challenges of high-stakes assessments to high school career and technology education programs in Maryland.College Park: University of Maryland. Tschannen-Moran, M. (2004). Trust matters: Leadership for successful schools. San Francisco: Jossey Bass. Tschannen-Moran, M., & Hoy, W. K. (2000). A multidisciplinary analysis of the nature, meaning, and measurement of trust. Review of Educational Research, 70(4), 547–593. Tyack, D., & Tobin, W. (1994). The grammar of schooling: Why has it been so hard to change? American Educational Research Journal, 31(3), 453–479. Uphoff, N. (2000). Understanding social capital: Learning from the analysis and experience of participation. . In P. Dasgupta & I. Serageldin (Eds.), Social capital: A multifaceted perspective (pp. 215–259). Washington, D.C.: The World Bank. Weick, K. E., & Roberts, K. H. (1993). Collective Mind in Organizations: Heedful Interrelating on Flight Decks. Administrative Science Quarterly, 38(3), 357–381. Zembylas, M. (2003). Interrogating “Teacher Identity”: Emotion, Resistance, and Self-Formation. Educational Theory, 53(1), 107–127.

Chapter 9

Distributed Leadership: Democracy or Delivery? Andy Hargreaves and Dean Fink

Introduction In the face of mounting evidence that top-down, micro-managed educational change models have failed to enhance student achievement over time, alternative models of lateral and distributed leadership, cross–school networks and professional learning communities are now being promoted as ways to harness the energy, motivation and professional learning of teachers and school leaders to secure sustainable innovation and improvement. In contrast to technocratic emphases imported from the corporate world on performance targets, line management, and delivery systems, emerging models of distributed leadership, networks and communities of practice regard organizations more as “living systems” or complex, evolutionary, “networks” that are much less amenable to top-down regulation. This chapter discusses the nature and benefits of lateral approaches to educational change, especially in the form of distributed leadership, that treat schools, localities, states, or nations, as “living systems “ interconnected by mutual influence. Through a conceptual discussion of the interrelated ideas of living systems, communities of practice and networks, the chapter underlines how, within this conception, distributed leadership operates as a network of strong cells organized through cohesive diversity and emergent development rather than mechanical alignment and predictable delivery. However, more deeply and more critically, the chapter also investigates whether, in practice, these lateral strategies are being used to extend democratic public and professional involvement in developing the goals and purposes of education or whether they are being primarily used as motivational devices to re-energize a dispirited profession into producing more effective and enthusiastic delivery of imposed government performance targets? Research examples from England, North America, and Finland are used to underscore our argument.

D. Fink (B) Director of Dean Fink Consulting Associates e-mail: [email protected]

A. Harris (ed.), Distributed Leadership, Studies in Educational Leadership 7, C Springer Science+Business Media B.V. 2009 DOI 10.1007/978-1-4020-9737-9 10, 

181

182

A. Hargreaves and D. Fink

Three Concepts Sustainable improvement in organizations, like sustainable improvement in bodily health or protected ecosystems, does not occur through singular strategies, emphasizing only one crop or health solution. Rather, it is the interaction of these elements in complex and holistic systems that move organizations and environments forward. Three interrelated concepts articulate the importance of this argument: living systems, communities of practice and networks.

Living Systems All living systems, natural and human, possess two qualities. First, they foster creativity, imagination, and innovation. Learning and growth are inherent in all living systems and the appearance of a qualitatively new order of things emerges with the creation of meaningful novelty in the environment. This novelty may be an insightful remark or the learning and breakthrough that come from interdisciplinary or cross-cultural contact, or the development of new government policy. Such living systems can be spontaneous or emerge by design. Second, living systems are self-organizing networks of communication. Schools, districts, and indeed nations are organized into multiple communities of practice such as civil rights or environmental movements that can interconnect to move society forward, or that can conversely join together to inhibit change or block new directions. Like a web of interconnected communities, each network has an essential skeletal structure of rules and regulations that frames relationships among people and tasks, distributes political power, and guides daily practice. It is these arrangements that eventually appear in or disappear from classroom seating plans, policy documents, organizational charts, written contracts, and budgets. These are the structures, forms, and functions designed by policy makers, leaders, and teachers to provide stability, order and direction in organizations such as schools. This ability to design is a solely human function. In nature, most change occurs through emergence, evolution, and the survival of the fittest. Not all evolution is universally benign, however, as farmers whose fields have been devastated by locusts will testify. Human design therefore seeks to protect society from the destructive forces that also afflict nature by providing purpose, meaning, structure, and cohesion. Taken too far, however, human design can overwhelm and stifle emergence within organizational and natural ecosystems as global warming, holocausts, technologies of torture and insensitive bureaucracies, all illustrate. It is the informal interconnections and interrelationships among people that cut across and intersect with an organization’s formal structures, “the fluid and fluctuating networks of communication” that give web-like organizations and communities their “aliveness.” The aliveness of an organization – its flexibility, creative potential, and learning capability – resides in its informal “communities of practice”. The formal parts of an organization may

9 Distributed Leadership

183

be “alive” to varying degrees depending on how closely they are in touch with their informal networks (Capra, 2002, p. 111).

Everything and everyone within a web is connected in some way. When one touches a spider’s web for example, the sensation reverberates throughout the entire web. Human organizations are similar. An action in one part of the web affects all other parts to a greater or lesser extent. In human systems that are designed to support emergence and growth through mutual learning, such as strong professional learning communities or interdisciplinary teaching environments at their best, the influence of enthusiastic and effective teachers in some classrooms can easily spread to others. But in technocratic structures of command and control where teachers, departments, and schools are separated from or even pitted against one another, living systems exert their effect anyway – just less productively. Effectiveness begets envy, innovation invites cynicism, control provokes resistance and communities become poisoned by gossip and cliques. In both cases, leadership is not just something that formal leaders hang on to or dole out, it is already distributed across the organization (Spillane, 2006) as a healthy or sick living system, with good or bad effects (Hargreaves and Fink, 2006). The point, therefore, is not merely to support or celebrate living systems as an innovation or an idea, but to design them in ways that make constructive rather than destructive emergence more likely. One particular and prominent form of such design is that of communities of practice.

Communities of Practice Communities of practice are everywhere in our daily lives and particularly in the organizations in which we work. They exist within families, service organizations, recreational activities and places of worship. They are present wherever people voluntarily come together for mutual engagement and develop a shared repertoire of how they do things together. Communities of practice are not necessarily synonymous with a department or a team or any similar organizational units even though such formal structures can evolve into them. Etienne Wenger (1998) explains that: developing a practice requires the formation of a community whose members can engage with one another and thus acknowledge each other as participants. As a consequence, practice entails the negotiation of ways of being a person in that context. . .. the formation of a community of practice is also the negotiation of identities (p. 149). Schools have many “communities of practice”. There are the teachers of grade four, who voluntarily collaborate on the children’s learning program, or the team of teachers who work together to support particular students with special educational

184

A. Hargreaves and D. Fink

needs. A community of practice might include the group who meet annually to prepare for the school’s graduation ceremonies or the staff’s annual retreat. It might include the union representatives who present a common front on workload issues, or it can encompass the professional learning communities using data and other evidence to pinpoint how to identify and assist the students in school who are at greater risk. Some communities of practice contribute significantly to students’ learning while others get in the way. There are no membership cards in these communities. Some don’t even have names, but everyone who is part of a community of practice knows who its members are. Some individuals belong to a number of communities and easily traverse these boundaries. These webs are not without structure, but they are structured according to a different logic than traditional bureaucracies and hierarchies. In communities, networks or webs, hierarchical control gives way to shared collaboration. There is no center or apex, but a multiplicity of connections and threads that link the organization’s various communities within and beyond its own boundaries. It is these that leaders must try to understand and influence if they are to establish and achieve organizational goals. As Henry Mintzberg (2004) explains: Management has to be everywhere. It has to flow with the activity, which itself cannot be predicted or formalized. . ... Management also has to be potentially everyone. In a network, responsibility for making decisions and developing strategic initiatives has to be distributed, so that responsibility can flow to whoever is best able to deal with the issue at hand (p. 141).

Mintzberg contends that “bosses and subordinates running up and down the hierarchy have to give way to the shifting back and forth between ‘colleagues’ on the inside and ‘partners’ on the outside”. Webs need designated leaders to connect and contribute, not command and control. “And that means that managers have to get inside those networks. Not be parachuted in, without knowledge, yet intent on leading the team. No, they must be deeply involved, to earn any leadership they can provide” (Mintzberg, p. 141). He insists that leadership within the organizational logic of a web is. Not about taking clever decisions and making bigger deals, least of all for personal gains. It is about energizing other people to make better decisions and do better things. . .. It is about releasing the positive energy that exists naturally within people. Effective leadership inspires more than empowers; it connects more than it controls; it demonstrates more than decides. It does all this by engaging – itself above all and consequently others (Mintzberg, p. 143).

These shifting communities of practice, therefore, do not require heroic or hierarchical leaders, but leaders who can help design a culture in which leadership is distributed in an emergent and benevolent way – so the community engages in robust dialogue, in an evidence-informed and experience-grounded manner, about the best means to promote the goals of deep and broad student learning for all.

9 Distributed Leadership

185

Networks Living systems are self- organizing networks of communication. As Fritoff Capra has observed,” wherever we see life, we see networks” (2002, p. 9). Governments and educators are beginning to work together to create systems that retain a common focus on and sense of urgency and consistency about learning and achievement for all on the one hand, while respecting and encouraging the diversity and flexibility that makes systems strong, resilient and sustainable on the other. One of the most favored solutions for creating this kind of synergy across schools and throughout entire systems is professional networks that distribute leadership across traditional boundaries to release the potential of those within the network. In the words of David Jackson (2003) Networks provide support structures for innovative schools – enabling in the dissemination of “good practice” but, more importantly, facilitative of the sharing of “good process”. They offer the potential to overcome the endemic isolation of schools, and to challenge traditional hierarchical system structures through lateral leadership and learning norms (p. 21).

Professional learning networks for teachers are based on the belief as well as some evidence that “teachers learn best by sharing ideas, planning collaboratively, critiquing each other’s idea and experiences and reducing the isolation encountered in most schools” (Veuglers and O’Hair, 2005, p. 2). Networks increase professional interaction and learning across schools, and for those who participate in them, they generate excitement about teaching and learning. At their best professional networks enable and encourage schools to share and transfer existing knowledge that can help children learn better; they stimulate the professional fulfillment and motivation that comes from learning and interacting with colleagues; they capitalize on positive diversity across teachers and schools who serve different kinds of students, or who vary in how they respond to them; and they provide teachers and others with opportunities for lateral leadership of people, programs and problem-solving beyond the immediate school setting (Jackson, 2004). In addition, networked learning communities, or communities of practice (Wenger, 1998) provide opportunities to draw on and develop evidence-informed, research-derived practices; (Veuglers and O’Hair, 2005) to promote and spread innovation across interested schools (Hadfield et al., 2002); to give teachers more voice in professional and school-based decision-making (Veuglers and O’Hair, 2005), and to help personalize every school as a learning community, where they can adopt emergent solutions for their own needs (Association of California School Administrators, 2001). Last, but not least, professional networks are flexible and resilient in the face of crisis or misdirected system initiatives that turn out to be unsuccessful – allowing new learning and fresh solutions to emerge and fill the gap that the false starts and failures have left behind (Giles, 2005). Network are not always naturally good. If their purpose is unclear, participation can quickly evaporate; talk may not always proceed or translate into action; poor

186

A. Hargreaves and D. Fink

ideas and harmful practices can be circulated and exchanged as easily as good ones; and they can be designed in ways that appeal only to hyperactive enthusiasts. Communities of practice can similarly serve noble causes, or nefarious ones. It is therefore important not only to understand or even advocate for educational organizations as living systems, communities of practice or networked learning communities, but also to discern under what conditions they are more or less effective, and more or less supportive of ethically and democratically defensible ends.

Two Examples Living systems can be articulated and advanced through human design when they closely incorporate principles of natural systems, so as to encourage productive growth and cross-pollination within a broad yet commonly agreed set of purposes. Living systems can also be appropriated or undermined by surrounding or constricting agencies of hierarchical political control. Two key examples – from Finland and England – demonstrate the difference between the two and raise essential issues for distributed leadership and educational reform that we will then explore in our final analysis.

Finland In January 2007, with colleagues Beatriz Pont and Gabor Halasz (2007), one of us (Hargreaves et al., 2007) visited Finland for OECD to examine the relationship between leadership and school improvement and report on what we observed. Finland provides a mature and comprehensive example of a designed “living system” that has energized dramatic change through creativity, imagination and self- organizing networks. From being a rural backwater economy, Finland has transformed itself into a high performing economic powerhouse. In the few short years of the twenty-first century, Finland has already been ranked as the world’s most competitive economy by the World Economic Forum (Porter et al., 2004). Educationally, in OECD’s 2003 PISA results (OECD, 2004), Finland’s 15 year olds ranked top in reading, mathematics and science, while in equity terms the country displayed the lowest variance between schools – just one tenth of the OECD average (OECD, 2004). At the core of Finland’s remarkable transformation is the nation’s long-standing and also recently reinvented struggle to develop and be guided by a distinctive moral purpose that binds its people together. At the core of this country’s success and sustainability is its capacity to reconcile, harmonize, and integrate those elements that have divided other developed economies and societies – a prosperous, highperforming knowledge economy and a socially just society. It has also done this in a way that connects the country’s sense of its creative craft-like history to the struggle for its future technological and creative knowledge-based destiny. So for instance, while the Finns invest heavily in scientific and technological development, they also have the highest number of musical composers per capita of any developed country,

9 Distributed Leadership

187

and ensure that all young people engage in creative and performing arts to the end of their secondary education. While the knowledge economy has weakened the welfare state in many other societies, in Finland, a strong welfare state is a vital part of the national narrative that supports and sustains a strong economy. Yet this has been achieved not through centralized command and control systems of regulation. Rather, after an international banking crisis, high unemployment and the collapse of Finland’s Russian market along with the Berlin Wall, economic regeneration was fueled by educational decentralization. At the centre of this successful integration of purpose and decentralization of effort is Finland’s educational system (Aho et al., 2006) that provides education free of charge as a universal right all the way through school and higher education – including all necessary resources, equipment, musical instruments and free school meals for everyone. Through consultation and discussion, the National Board develops guidelines that provide the support and strategic thinking. With the support of educational research, the National Board provides a “steering system” for educational policies in an evidence-based way, through funding, legislation, evaluation, and curriculum content. Within this generally understood social vision, the state steers but does not prescribe in detail the national curriculum, with trusted teams of highly qualified teachers writing much of the curriculum together at the level of the municipality, in ways that adjust to the students they know best. The government sets the parameters that are designed to allow complexity and responsiveness to evolve without this degenerating into chaos. Finland – the world’s leader in measured student performance – places no emphasis on nor does it give any particular place to individual testing or measurementdriven high-stakes accountability. It does not improve performance by perseverating on it. Learning rather than measured performance defines the focus and the form of systemic distributed leadership. Since public education is seen as vital to the country’s growth and identity and given the shared high status for educators who are seen as central to this generational mission, the teaching profession is made up of highly qualified candidates. Learning and teaching are valued throughout schools and society, learning starts early but is unhurried and untested, and learning is broad and lifelong rather than concentrated on test preparation. Teacher quality and performance are addressed at the point of entry through working conditions, missions, status, autonomy, and reward. Improvement of schools that employ these highly capable and trusted professionals is achieved by processes of self-evaluation within learning organizations that are allocated national and local government resources so they can solve problems for themselves. System leadership, in this sense, is leadership for learning, leadership by learning, and leadership as learning – not leadership for performance and testing. At the heart of the human relationships that comprise Finland’s educational system and society is a strong and positive culture of trust, cooperation and responsibility. Principals and teachers are trusted, to a degree, because of their high qualifications, expertise and widespread commitment and responsibility. If principals

188

A. Hargreaves and D. Fink

become sick or ineffective, the community of teachers takes over because it sees that the school belongs to all of them. Yet trust is not blind or indifferent. Rather, active trust is built through deliberately created structures and initiatives. This is evident in:

r

r r r

Networks: National projects always have “very strong and big networks” for cooperating with national authorities, in forums where people “learn and work from each other”. Municipalities stress the importance of all teachers participating in local and school- based processes as well as in curriculum development, not to implement government strategies and initiatives but to spread ideas, transformatively and interactively through non- linear processes of learning and experimentation. Shared Targets: which are produced at the local level through action plans rather than imposed by political or administrative means. Self-Evaluation: as the key to continuous improvement as compared to imposed inspection or test-based accountability that ranks schools competitively on the basis of their test scores. Local principal cooperation where even at the secondary level principals across schools share financial and other resources when needed, and feel genuinely responsible together for all the children and young people in their town and city – for their community’s future – rather than competing only for the advantage of children in their particular schools.

Through these relationships of responsibility, cooperation and trust, Finland exhibits a pattern of systemic leadership in strong cultures of lateral and vertical teamwork, networking, participation, target-setting and self- evaluation. Hierarchies are not feared, and interventions (as compared to cooperative problem solving) are virtually unknown. Gaps are narrow and performance is strong – not as a result of precise external interventions that concentrate on these things but as a consequence of creating the living systems where highly qualified and responsible professionals produce these results for themselves. For observers of education external to Finland, it is important to understand the connectedness of the system’s various components and the underlying philosophy that drives it forward. It is too easy to “cherry-pick” those aspects that fit existing systems elsewhere and claim that this is the Finnish model. For example, arguing that, like Finnish teachers, all teachers must have Masters degrees if quality is to be improved (Barber and Mourshed, 2007), misses the point of Finland as a living system, where high quality teachers are drawn by the inspired societal mission, the profession’s public regard, and the supportive conditions of work, as much as they are created by particular levels of qualification.

England A second example of an effort to employ principles of living systems but within the context of more hierarchical and technocratic political systems, is a large project developed by the Specialist Schools and Academies Trust in England. This

9 Distributed Leadership

189

organization stimulates, coordinates and to some extent regulates the re-designation and sometimes re-construction of over 90% of English secondary schools so that each has a themed identity (such as languages, the arts, or sports). In its project, Raising Achievement/Transforming Learning (RATL), the SSAT devised an improvement model that promoted schools learning from other schools as an explicit, and even emphatic theme in its theory-of-action (Hatch and White, 2002) by:

r r r r r

emphasizing the principles of bringing about improvement by schools, with schools in peer driven networks of lateral pressure and support where the peer factor replaces the fear factor as the key driver in raising standards. combining outside-in knowledge of experts at conferences, with inside-out knowledge of successful and experienced practitioners working openly and inclusively with less successful colleagues in transparent processes (not just outcomes) of assistance and support. making mentor schools available to lower performing peers in cultures of strong expectation for improvement within transparent lateral systems, but not mandating these mentor relationships in general or in any particular case. supplying modest amounts of additional resourcing to facilitate these improvements and interactions. providing clear, practical menus of short- term, medium- term, and long- term strategies for improvement and transformation with proven success among experienced administrators.

Through three successive annual cohorts of phased-in participation that encompassed approximately 300 schools in total, RATL’s project leaders invited schools to join what was initially called an underperforming schools network, on the basis of the schools having experienced a measurable dip in achievement results over one or two years. An evaluation of this network conducted by one of us (Hargreaves and Shirley, 2007) points to the success and promise of this professionally peer-driven strategy of the strong helping the weak in cultures of committed and transparent improvement. Two thirds of the schools improved at double the rate of the national secondary school average within two years. The evaluation also shows that these laterally-driven improvement processes do not remain unaffected by the legacies of standardization that continue to linger in the surrounding policy environment – in the unrelenting emphasis on standardized test scores in basic subjects; in short term cycles of funding for projects such as this; in hit and run inspections that over-rely on printed achievement data rather than direct observation to assess the satisfactoriness of schools, and on unending waves of short term government initiatives. Lateral network activity with cross-school and within-school distributed leadership focuses disproportionately on adapting short term improvement to deliver existing learning rather than on long term transformations towards creating different and better teaching and learning. In RATL, schools are energized and enthused by the process of exchanging strategies such as e-mentoring, examination review skills, using former students to mentor existing ones, and nourishing children with bananas and water especially on

190

A. Hargreaves and D. Fink

test days, that have an immediate impact on tested achievement and school results. Instead of complying with the short term demands of outsiders, educators become addicted to developing and exchanging immediate improvements of their own – far more attractive than challenging and confronting their own and each other’s, fundamental approaches to teaching and learning. In effect, therefore, there are signs within this initiative that lateral professional energy and motivation is being harnessed and even hijacked to secure more efficient and enthusiastic delivery of unquestioned government purposes and targets in conventionally tested learning, rather than contributing to a redefinition of those educational goals and purposes in a more visionary, inclusive and transformational way appropriate to life and learning in the 21st century. Deep reflection and long term sustainable transformation are joyfully jettisoned to embrace short-term success in easily tested achievement.

One Choice? In the age of standardization, along with accompanying market competition between schools, top-down pressure to meet prescribed performance standards in England or Adequately Yearly Progress in the United States has been accompanied by absence of support in some instances, in the shape of poor funding and punitive intervention; or increased support in cases like England’s National Literary and Numeracy Strategy with its supply of strategies additional materials and extensive training and coaching. But while large scale reform and a coordinated top-down response to raising standards may have solved a crisis of public legitimation in education where national or statewide numerical standards have been viewed as a political necessity, another crisis of professional motivation loomed in its wake, creating difficulties of attracting and retaining high quality teachers and leaders within the profession (See Fig. 9.1). In an emerging era of post- standardization where many of the old elements of standardization still persist alongside a range of more motivational initiatives, lateral

Top-Down Government Goals

Performance Targets

Market Competition

Parent Choice

Resources

Fig. 9.1

Materials Support Bottom-up

Training

9 Distributed Leadership

191

Fig. 9.2

Top-Down Government Goals

Performance Targets

Lateral Learning Peer Pressure and Support

Public Engagement Integrated Services

Resources

Materials

Training

Support Bottom-up

strategies of involving parents and communities in their children’s education and of peer-driven, network-based improvement initiatives across schools have increased professional engagement and even shown evidence of further gains in conventional student achievement. However, this development of lateral capacity and greater distributed leadership (Fullan, 2007) is largely being placed in the service of meeting externally set performance targets that are still decreed non-democratically, and in a way that is politically arbitrary work and professionally exclusionary (Fig. 9.2). In the words David Hartley (2007) in his critical essay on the political appropriation of distributed leadership. The emergence of distributed leadership is very much a sign of the times: it resonates with contemporary culture, with all its loose affiliations and ephemeralities and it is yet another sign of how the public sector purports to legitimate policies by appeals to the new organizational forms within the private sector. . .. But what it is to be distributed remains very much within the strategic parameters and targets set by government. It is the teachers, not the strategy, which are available for distribution. Hierarchical forms of accountability remain (p. 211)

The hardest questions about distributed leadership are moral and democratic ones. What kind of distributed leadership do we want, and what educational and social purposes will it serve? Are such forms of leadership merely more subtle and clever ways to deliver standardized packages of government reforms and performance targets in easily measurable areas like literacy that have more to do with expedient politics than with sustainable educational change? Or, like Finland, can distributed leadership be a key principle in a coherent and inclusive democratic consensus that joins the entire community in the pursuit of a compelling social vision? The challenge posed by this question will be found not only in the nature of distributed leadership itself, but also in the courage and commitment of supposedly democratic nations to define their visions inclusively and inspirationally. Such visions must move beyond arithmetical achievement gaps or vacuous calls for world-class standards, to embrace the most basic questions of economic creativity and sustainability, ecological survival, and fundamental human rights that face the

192

A. Hargreaves and D. Fink

generations of today and tomorrow. At that moment, distributed leadership can then transcend from being a motivational tool of management to an inspirational and living principle of democratic citizenship. As we have seen in the case of Finland, these living principles are not merely ideals but actualities. Civil rights in the United States, women’s liberation, the end of apartheid in South Africa and the quiet revolution in attitudes to the environment, all began with people, not policies – often against the original opposition and entrenched resistance to change of governments. Each was a “living system” driven forward by emergent forms of distributed leadership, organized in cohesive networks and diverse communities of practice in pursuit of socially just moral purposes and shaped by designed formal structures that coordinated without inhibiting creativity and passion. In all these cases, distributed leadership has been about democracy, not delivery. If we are, once more, to give distributed leadership this bigger chance, it is time for the government’s delivery van to return to its depot.

References Aho, E., Pitkanen, K. and Sahlberg, P. (2006). Policy Development and Reform Principles of Basic and Secondary Education in Finland Since 1968. Washington, DC: World Bank. Association of California School Administrators. (2001). Recruitment and Retention of School Leaders: A Critical State Need. Sacramento: ACSA Task Force on Administrator Shortage. Barber, M. and Mourshed, M. (2007). How the World’s Best School Systems Come out On Top. London: McKinsey Company. Capra, F. (2002). The Hidden Connections: A Science for Sustainable Living. New York: Anchor Books. Fullan, M. (2007). Turnaround Leadership. San Francisco CA: Jossey Bass. Giles, C. (2005). ‘Change over time: Resistence, resilience, and sustainable educational reform’, Unpublished Manuscript, Buffalo: University of Buffalo, State University of New York. Hadfield, M., Kellow, M. and, C. (2002) Schools as Learning Communities: Building Capacity through Network Learning. In Education Vol 30 No.3, pp. 3–13. Hargreaves, A. and Fink D. (2006), Sustainable Leadership. San Francisco: Jossey Bass. Hargreaves, A. and Shirley, D. (2007). The Long and Short of Raising Achievement: Final Report of the Evaluation of the “Raising Achievement Transformation Learning”, project of the U.K. Specialist Schools and Academies Trust. Chestnut Hill MA: Boston College. Hargreaves, A, Hal´asz, G and Pont, B. (2007). Finland a Systemic Approach to School Leadership, A Case Study Report for the OECD Activity Improving School Leadership. Hartley, D. (2007) ‘The emergence of distributed leadership in education; Why Now?’ British Journal of Educational Studies, 55(2), 202–214. Hatch, T and White, N. (2002). ‘The RAW materials of reform: Rethinking the Knowledge of school improvement,’ International Journal of Educational Change. 3 (2), pp. 117–134. Jackson, D. (2004). Networked Learning Communities: Characteristics of networked learning – What we are learning? Paper presented at the annual conference of the International Congress of School Effectiveness and School Improvement. Rotterdam, Netherlands. Mintzberg, H (2004). Managers Not MBAs: A Hard Look at the Soft Practice of Managing and Management Development. San Francisco CA: Berrett-Koehler Publishers. OECD (2004), Learning for Tomorrow’s World – First Results from PISA 2003, OECD, Paris.

9 Distributed Leadership

193

Porter, M., Schwab, K., Sala-i-Martin, X. and Lopez-Claros, A. (eds.) (2004), The Global Competitiveness Report, Oxford University Press, New York. Spillane, J. (2006). Distributed Leadership. San Francisco CA: Jossey-Bass. Veuglers, W. and O’Hair, J. M. (eds.) (2005). School-University Networks and Educational Change. Maidenhead, UK: Open University Press/McGraw-Hill. Wenger, E. (1998). Communities of Practice. Cambridge MASS: Cambridge University Press.

Part V

Future Perspectives

Chapter 10

From Distributed to Hybrid Leadership Practice Peter Gronn

Introduction Ideas come and go. Some retain their usefulness, while some fall by the wayside. Still others survive with their original integrity intact, while others undergo major surgery or revision. To this point in its career history, the signs for distributed leadership are optimistic. If one casts a roving eye back across a decade or so, between the point of its arrival and its more recent uptake, then this particular view of leadership appears to have weathered an initial stage of conceptual exploration and is now well into a phase of empirical investigation. Moreover, some sense of its impact (and the difference, if any, that it makes) is becoming clearer. In short, distributed leadership displays a number of the hallmarks of survival. Having said that, and without wanting to adopt a glass half-empty mentality, the purpose of this chapter is to give voice to some caveats and concerns. These include: first, a need to revisit the provenance of distributed leadership and to set the genealogical record straight; second, to rethink the meaning of the term; and, third, to inquire about future directions and what comes next. Taken at face value, these first two concerns have a rather deceptive “housekeeping” quality about them. That is, answers to the questions appear to serve the interests of conceptual tidiness. In fact, the clarity they introduce has significant implications for how researchers and commentators understand leadership more generally. In relation to the third concern, somewhat paradoxically, I am suggesting that, just as distributed leadership appears to have come into its own, it would be profitable to begin thinking beyond it. Throughout the chapter (but especially in the final section) I suggest what this would entail, why this needs to occur and how it might be done. To expand on each of these points, it can be said, for example, in relation to the first that, notwithstanding the current level of interest in distributed leadership, the claim that the leadership of an organization observes a distributed pattern is

P. Gronn (B) University of Cambridge, Cambridge, England This chapter draws extensively on the author’s previously published article, “The future of distributed leadership”, Journal of Educational Administration, vol 46, no. 2 (2008): 141–158.

A. Harris (ed.), Distributed Leadership, Studies in Educational Leadership 7, C Springer Science+Business Media B.V. 2009 DOI 10.1007/978-1-4020-9737-9 11, 

197

198

P. Gronn

basically unremarkable. In one sense, when scholars in the field accord legitimacy to “distribution of leadership” they really do little more than play catch up with antecedent developments in the closely related conceptual domains of power and influence, both of which form part of Lukes’ (2005, p. 35) family of cognate modes of “significant affecting”1 . Among sociologists and political scientists, for example, such expressions as “distribution of power” and “distribution of influence” are wellrehearsed analytic categories (Gronn, 2003, pp. 60–62). Moreover, if it can be shown (as I endeavour to do in the main body of the chapter) that there is continuity between current understandings of distributed leadership and some early management research and theory, then that puts the idea in a different light. Evidence of continuity implies that, rather than a pre-occupation with the timeliness of distributed leadership’s emergence, for example, what probably has to explained is the upsurge of heroically informed understandings of leadership in the two decades or so that preceded its arrival in the literature and which effectively hijacked the field during that period. It was largely as a critical response to the Trojan horse of heroism— central to the visionary and charismatically informed dynamism of the so-called “new” leadership of the 1980s (Bryman, 1996)—that distributed leadership won scholarly recognition. For this reason, the heroic leader impulse (which, of course, resonated closely with antiquated archaic leader archetypes of greatness and glory) may be the aberrant development, rather than distributed leadership. In support of this claim, there is much that is currently being asserted to justify the significance of distributed leadership as an idea which already had been anticipated by a handful of pioneer theorists. As I indicate below, one of the few differences between now and when these scholars published (i.e., mainly the 1950s and 1960s), apart from the mostly functional approach that they adopted to distribution, is that this earlier generation chose not to elaborate on or pursue empirically the possibilities which the idea opened up. Given the close conceptual links between distributed leadership and such longstanding organisational phenomena as power, influence, co-ordination, collective decision-making and delegated authority, a key question for proponents of distributed leadership is what this idea adds by way of value. On the credit side of ledger, distributed leadership has helped to expose limitations inherent in leadership understood individually and has tempered its rather inflated view of human agency. Moreover, thanks also to distributed leadership, the field seems much more willing to accommodate a plurality of analyses that may be positioned somewhere on a continuum from concentrated to dispersed leadership. This assertion is true of researchers in the field and also of the declarations of public agencies (e.g., HMIE, 2007). While these are no mean achievements, they raise the issue of where to from here? As Hartley (2007) has recently pointed out, a number of scholars whose work has been closely associated with distributed leadership (including myself) have been promoting it as an idea whose time has come. One thing that might be timely about distributed leadership is the need for a shift in direction. Having endured for much

1

Although, interestingly, Lukes does not include leadership in his family.

10 From Distributed to Hybrid Leadership Practice

199

of the 1980s and 1990s a convergence of the field around a near monolith of orthodox individualism, which has finally begun to erode, little point would seem to be served by substituting it with an alternative form of convergent hegemony, in this case a distributed one. As I have recently argued elsewhere (Gronn, 2008), one of the weaknesses of both focused and distributed views leadership, when they are championed singly or co-exist as polarized alternatives, is that they may do less than full justice to patterns of diverse leadership practice increasingly manifest in schools. Sources of influence (the attribute accepted by most scholars as the key to leadership), can for a time be concentrated and at others dispersed, and agents of influence may sometimes be individuals and at others collectivities. A similar point has been made about leadership in the higher education sector. Bolden et al. (2008, p. 39), for example, note that while all 152 of their interview informants in 12 UK HIEs affirmed that leadership “was in some way distributed within the university”, they also expressed “a clear desire for strong and inspiring leadership from individuals in key roles”. My argument is that the interests of leadership research would be best served by a rubric which accommodates this and other similar possibilities. That rubric is hybridity. This chapter, then, takes the form of a report card or stock taking on distributed leadership. In order to progress my argument about the future of leadership, I begin by discussing the issue of the timeliness of distributed leadership. This is followed by a review of the work of early generation leadership theorists. Next, to substantiate the claim that the preferred analytical focus of researchers is hybridity (i.e., varying combinations and degrees of both concentrated and distributed leadership, the balance and form of which may oscillate over time) rather than particular types of leadership, in the succeeding section of the discussion I review a selection of recent distributed leadership research along with evidence from some earlier sources. Finally, in the discussion section, I consider the relationship between leadership and power, whether distributed leadership is a synonym for, or an alternative to, democratic leadership and I develop further the argument about hybridity, diversity and variation when mapping practice.

The Distribution of Just About Everything? The first possibility worth considering is whether or not the current absorption in distributed leadership may be a proxy for something else. That is, instead of distributed leadership, it may be the idea of “distribution” whose time has come. In the early 2000s, there are a number of distributed modes and forms. In times past, one of the most frequently encountered usages of “distribution” (at least for students of social and political systems) was the socialist objective of Western labour and social democratic parties. In Australia, for example, the wording of the objective said something like: “state ownership of the means of production, distribution and exchange”. As mentioned in the introduction, “distribution of power” was also not uncommon. Caplow (1956, p. 489) referred to “asymmetric distribution” of power

200

P. Gronn

and, in the early 1960s, in his pioneering study of power in New Haven, Who Governs?, Robert Dahl (1961) entitled the second part of the book (an 80-page section devoted to leadership) “the distribution of influence”. At about the same time, Crozier (1964, p. 19) employed “the distribution of the work load”. Current examples of distributed phenomena include: distributed information systems, distributed knowledge, distributed cognition, distributed decision-making, distributed work, distributed learning systems and even, according to a report in the Guardian in early 2007, distributed denial of service (the calculated crashing of computer network servers by flooding them with traffic)—an alleged Russian retaliatory measure inflicted in response to the Estonian government’s removal of a wartime statue. This recent prominence of distribution parallels a reappraisal of a range of ways of thinking across the social sciences, including the field of education. Space limitations do not permit a thorough elaboration of the point, but in some instances this reappraisal is little more than an attempt to reconstitute phenomena interpretively or discursively (i.e., by invoking a “lens” through which to represent aspects of reality), although in other instances a much more deep-seated ontological reworking of constitutive elements is occurring. The examples I have in mind include networks, which previously were principally a staple in the diet of anthropology but which, via the ideas of networked forms and communities of practice, now provide disaggregated understandings of organizations and organizing processes. Other examples include the notions of boundarylessness and portfolio careers in career theory, diffusion and flows in discussions of globalization, flexibility in the workplace, structuration and structuring in the constitution of social reality, the use of “liquid” as an alternative to characterizations of the contemporary epoch as post-modern or high modern, and a growing sense of scholarly resistance (difficult to substantiate, admittedly, but nonetheless evident) to hard and fast conceptual distinctions, and to the classifications of phenomena based on dualisms, dichotomies and bifurcation. These developments are indicative of a trend in thinking away from fixed and invariant forms; away from entities, tightly drawn lines, edges, borders and boundaries; and (in organisation studies) away from the kinds of command and control, line and staff management systems associated with a Weberian bureaucratic paradigm. The emphasis on distribution is consistent with these developments which, while they do not necessarily represent a triumph of process over structure, nor necessarily a re-assertion of a phenomenological view of being and becoming, express a preference for looseness and open-endedness to accommodate a sense of reality as fluid and continually emerging. There may be a downside to these developments. In the sphere of school education, now that distributed leadership is well entrenched in the linguistic furniture, there is a somewhat promiscuous inclination to think of virtually every initiative on the part of teachers and administrators as leadership. Equally, distributed, distributive or shared leadership tend to be extolled as solutions to problems associated with improving students’ learning. These habits of mind are certainly discernible in conversations with school and system personnel, and they flavour the reports and discussion papers of public sector agencies. The adoption of self-managed

10 From Distributed to Hybrid Leadership Practice

201

or site-managed schooling illustrates these points. Initially in the school reform movement in Australia, for example, it was customary to think of principals as the officially endorsed drivers of reform. After a while, official thinking shifted to include senior level teams to supplement and support the leadership work of principals. Professional development programs in teaming followed shortly afterwards. Implicit in this unfolding set of developments was an acknowledgement (rarely officially admitted to in these terms) that there were obvious limits to the ability of principals to deal with the complexity entailed in implementing the (political) reform agenda. Fast-forwarding about a decade later and focusing on England, the extent of the subsequent fusion of these individual and collective efforts is such that when Reid et al. (2004, p. 252) investigated the leading of learning in a secondary school, 76% of a staff of 68 (or 52 teachers) either were formally labelled, or thought of by their colleagues, as leaders. In other words, there were only 16 teachers were “without a significant leadership role, or roles”. Essentially, then, the language of leadership has been morphing from a gung-ho, topdown view of change management into a rubric that incorporates a lateral perspective, such that distributed leadership has become the “favoured strategy” (Hatcher, 2005, p. 254) for securing English teachers’ commitment to official management agendas. This trajectory in the direction of official legitimation of distribution highlights the virtue of a historical appreciation of developments. Despite the sudden discovery and love affair with distribution, this idea has not emerged from nowhere. Indeed, perhaps it has always been part of the work of leaders and managers. If so, then the question is: Why might researchers and others have chosen to ignore or gloss over this fact? Organizations whose operations have been geographically dispersed provide an obvious illustration of the historically enduring force of distributed work patterns. Re-winding to the 17th and 18th centuries, for example, in their analysis of the lines of communication and control between England and Canada in the Hudson’s Bay Company, O’Leary et al. (2002, p. 51) document the tensions involved in managing vastly dispersed operations. These tensions arose out of the need to achieve a “workable balance between central control and local discretion”. Moving forward to the more recent past, remarkably similar difficulties with managing distributed responsibilities were experienced in Australia in the 1950s when a headmaster and a campus head combined their efforts to establish an educationally innovative branch school in a geographically remote location (Gronn, 1999).

Back to the Future? A historical perspective brings to light other lessons. Invariably, in this regard, conceptual genealogies are enlightening. Not only do they set the record straight about antecedents, but they also reinforce the importance of continuities of understanding and practice. Distributed leadership itself is a case in point. While the results of the scholarly archeological dig reported below are not exhaustive—because examples

202

P. Gronn

of early generation theorists who dabbled in distributed leadership keep on coming to light—they indicate that the heritage of thinking which provides the foundation for this currently in vogue idea is considerably more extensive than was initially believed. Having acknowledged the potential of distributed leadership, it is unclear why the early writers whose work is catalogued in this section went no further. One explanation (admittedly speculative) for their refusal or inability to develop the idea may have been the hegemony of methodological individualism in the social psychology and organisation theory of their day. (I return to this point when I discuss holism and emergence in the next section.) The following writers are listed in approximate order of date of publication, with the recency of dates in some cases resulting from newly published editions.

Benne and Sheats In all that I have written on distributed leadership, I have suggested that C.A. Gibb was the progenitor of the idea. This claim remains valid, although with two minor qualifications. First, Gibb’s was not a lone voice. Second, while he was the first to use the words “distributed leadership”, to correct the record there were two other writers (Benne and Sheats, 1948) who had previously hinted at this possibility. Benne and Sheats (1948, p. 41) renounced sharp distinctions between leader and member functions, and proposed instead that: Groups may operate with various degrees of diffusion of “leadership” functions among group members or of concentration of such functions in one member or a few members. Ideally, of course, the concept of leadership emphasized here is that of a multilaterally shared responsibility.

This unelaborated observation about diffusion gave these authors a platform from which to argue for group training, for which purpose they classified 19 possible group member task and maintenance roles.

Gibb Gibb’s initial articulation of distributed leadership appeared in his chapter on leadership for the Handbook of Social Psychology (Gibb, 1954), which was slightly revised subsequently for the second edition of the Handbook (Gibb, 1968a). By this time Gibb had become something of a leadership agnostic, as is indicated by his claim that, while it might be vital to general discourse, leadership “had largely lost its value for the social sciences” (Gibb, 1968b, p. 91). A decade earlier, against the tide of opinion of his day, Gibb (1958, p. 103, original emphasis) doubted that leadership was the monopoly of one individual: There is still a tendency among psychologists and sociologists to think of every group as having a leader. . .however. . .unequivocal unipersonal leadership rarely, if ever, occurs.

10 From Distributed to Hybrid Leadership Practice

203

More than 50 years ago, the word “still” was a good indicator even then of the hegemony of focused individualist assumptions. The reason for Gibb’s skepticism was that, in the flux of group processes, leadership tended to “pass from one individual to another as the situation changes” (Gibb, 1954, p. 902). Here, Gibb pre-figured an emergent view of leading that he developed further in the second edition of the Handbook, where he also dismissed the bifurcated and mutually exclusive categories of leader and followers. He regarded this distinction as unhelpful because, in reality “each of these is but a transient status”, so that “leaders and followers frequently exchange roles and observation has shown that the most active followers often initiate acts of leading” (Gibb, 1968a, p. 252). This discomfort with solo leaders was evident in Gibb’s rejection of motivational drive as the overriding factor with which to account for the success of larger-thanlife figures of history (e.g., Napoleon). An emphasis on drive ignored how leaders were frequently captive of longstanding traditions and constrained to act in conformity with them. In fact, leaders were often reactive, and “forced to follow the behaviour of those who [in experimental pre-test situations] had followed” and who had simply “led the group in the direction it would have taken had he not been there” (Gibb, 1954, p. 898). Moreover, leaders were also captive of the prior expectations held of successors. In essence, then, the leader-follower dichotomy was untenable for Gibb, given that leadership “inevitably embodies many of the qualities of the followers, and the relation between the two may often be so close that it is difficult to determine who influences whom and to what extent” (Gibb, 1968a, p. 271). Indeed, so difficult was the measurement of influence flows in groups that Gibb suggested official designations of leaders be ignored. As an alternative, he endorsed an approach to group fieldwork in which “no a priori assumptions are made as to the distribution of those functions among members” (Gibb, 1968a, p. 272). Pushing the boundaries a little further, Gibb (1954, p. 884; 1968a, p. 215) then claimed that leadership was “probably best conceived as a group quality, as a set of functions which must be carried out by the group”, functions which could either be focused, on the one hand, or distributed, on the other, as there was no “force in the nature of the leadership relation itself” that was “making for ‘focused’ rather than ‘distributed’ leadership” (Gibb, 1968b, p. 94). For Gibb, then, groups had one or more leaders who were identifiable by the frequency of their acts of leading, except that there were two difficulties. First, if leaders were differentiated from followers according to a hierarchy of influence in groups, then observers had no option but to determine leader status “by drawing an arbitrary line on a frequency continuum” (Gibb, 1958, p. 103). Second, there remained the problem of reciprocity: a strategy of distinguishing leaders as group members who influenced their fellows more than they were influenced by them, made no allowance for the cumulative influence of all the individuals on one another. As Follett (1973 [1927], p. 159) pointed out, reciprocal influence means more than “all the ways in which A influences B, and all the ways in which B influences A”. It also entails how “A influences B, and that B, made different by A’s influence, influences A, which means that A’s own activity enters into the stimulus which is causing his activity”.

204

P. Gronn

French and Snyder A common strand in the both Gibb’s and Benne & Sheats’ observations was their focus on small group processes. Two of Gibb’s contemporaries, French and Snyder (1959), adopted a similar focus. Drawing on power as the core attribute of leadership (but also treating power and influence as synonymous!), these writers defined leadership as “the potential social influence of one part of the group over another”, so that if one person had power over fellow group member, then that person “has some degree of leadership” (French and Snyder, 1959, p. 118, original italicised). Despite these differences of power and influence, the group norm was one of distribution (ibid): usually every member has some degree of influence over others in an informal group; in other words the leadership is widely distributed throughout the group.

Perhaps slightly confusingly, in this view of groups followers were defined as “members with less leadership” (ibid). This was claimed to be due to their personal qualities or their subordinate roles.

Katz and Kahn For Katz and Kahn (1978), in The Social Psychology of Organizing, the focus on leadership migrated upwards from the group to the organisation as a whole. They defined leadership as “the exercise of influence on organizationally relevant matters by any member of the organization”, with organizations more likely to be effective when the leadership “function” was distributed or shared (Katz and Kahn, 1978, p. 571). The distribution of leadership was thought to be desirable because it increased the likelihood of two preferred outcomes: first, strengthened decisionmaking commitment; second, improved decision-making quality. The justification for this second line of argument was not dissimilar to current claims about the significance of workplace knowledge: All the knowledge of the world within and outside the organization is not located in the formal chain of managerial command, much less at the upper end of that chain. . . The sharing out of the leadership function means using more fully the resources of the organization (ibid).

Likewise, the strategies which Katz and Kahn envisaged that formal leaders would use to distribute leadership (i.e., delegation, shared decision-making, openness to subordinates’ influence and provision of information) align closely with about half of the range of current practitioners’ understandings of distributed leadership synthesized by MacBeath (2005).

Schein Finally, for Schein (1988), leadership was once again viewed from within the language of functions. Apart from four of these functions which formal leaders and managers alone could perform (e.g., translation of directives between levels), a wide

10 From Distributed to Hybrid Leadership Practice

205

range of other functions was distributed among fellow group members. Moreover, once a broad set of functions was identified at a group level, “it becomes clear that any member of a given group can perform them”. Indeed, an effective group was one in which such functions were “optimally distributed” (Schein, 1988, p. 133, original emphasis).

Numbers Count This listing is clear evidence of an impressive scholarly pedigree, yet none of these authors provided a basis for differentiating between unit sizes (i.e., groups and organizations). As a consequence, leadership could be distributed within a dyad or among an entire membership. Without some means of differentiation, however, specification of the patterns and dynamics of leadership distribution proves extremely difficult. Simmel’s (1902) pioneering work on the dynamics of group unity and cohesion, and impact of membership changes, indicated how size makes a difference. With large groups, in which “a decisive division of labor” is sufficient for these to cohere, otherwise such a “widely extended group would break apart on every occasion” (Simmel, 1902, I, p. 3), the effects of membership changes can be absorbed more readily. The reverse outcome applies with smaller groupings of up to four members (where there is also a division of labour) in which group members’ relations and their sense of unity are vulnerable to loss of membership continuity (Simmel, 1902, I, p. 3). Numbers, it would appear then, count, with small numbers perhaps counting more than large numbers. In this next section, I indicate that studies of small number groupings are the genuinely interesting and significant ones, with their importance being three-fold. First, as with the preceding discussion of early generation writings, there is a parallel legacy of relevant scholarship, much of which pre-dates the current burst of distributed leadership commentary, yet its existence and its messages are mostly unacknowledged. Second, while small number group formations may amount to little more than aggregates of individuals, they may also germinate into holistically expressed forms of influence. Once more, this possibility is a largely untapped resource and its emergent potential unrealised. Third, in a number of empirical investigations of distributed leadership situations, there is persistent evidence of the continuing influence of key individuals in parallel with evidence of partnering and teaming. Use of the label “distributed” to try to encompass such eventualities may be a possible source of confusion. A more helpful designation, as was foreshadowed earlier, might be to subsume all of these variations under the rubric of hybrid leadership practice.

The Legacy of Small Numbers While leadership is far from an exclusive privilege of persons occupying hierarchically defined roles, the work of executive level leaders should be understood as part of a role set, rather than in isolation (Fondas and Stewart, 1994). Role sets

206

P. Gronn

materialize because the pressures of complex, uncertain and rapidly changing task environments require the co-ordination of collective intelligence to try to solve problems. This is why managers and leaders frequently find themselves improvising their work patterns and routines on an as-needs-be basis, motivated for the most part by self-interest and survival rather than the observation of textbook niceties (Whisler, 1960, p. 2). In this sense, executives are far from as isolated as suggested in the popularly understood idea of lonely at the top. The reality, by contrast, is dualistic in that “power holders have independent self construals and interdependent relational structures” (Lee and Tiedens, 2001, p. 44). Popular awareness of dyads tends to concentrate on their manifest tensions. A common topic of gossip journalism, for example, is the battle of wills between members of a celebrity couple. Political journalists often draw attention to the tensions in leadership dyads. Here in Britain, the transition in 2007 between the former Prime Minister, Tony Blair, and his successor the former Treasurer, Gordon Brown, where each needed the other for them to be a collective political force, and yet their rivalry for high office put them at odds with one another, received a significant amount of airplay. There are similar examples in Australia of publicly acknowledged pairings between Prime Ministers, Bob Hawke and John Howard, and their respective Treasurer deputies and rivals, Paul Keating and Peter Costello. Building on Simmel’s work, Becker and Useem (1942, p. 15) showed how the joint interests of both dyad members in patterned mutual action (a defining attribute of dyads) manage to accommodate this seeming paradox of attraction and antipathy. Thus, effective dyads may be odd couples, in which contrasts of personality and outlook strengthen their bond (Powell, 1997; Gronn, 1999). Leadership researchers have investigated the properties of dyads, one of which is complementarity. Here different attributes combine to strengthen a pair’s overall effectiveness (Powell, 1997, p. 194). An informant of Whisler’s (1960, p. 199), a US city commissioner, described how this was possible in the relationship with his assistant: [The commissioner] explains that long association has made the two individuals so familiar with each other’s thinking that they simply divide the work more or less at random, handling incoming work and information independently for long periods of time and checking with one another at intervals.

Three senses of complementarity may be distinguished. First, both dyad members may pool their separate specialisms, as in Etzioni’s (1965) example of dual leadership that combines both expressive and instrumental functions, and in Senger’s (1971) survey of a number of post-World War 2 initiatives in US firms and the military. Second, as indicated in Gronn and Hamilton’s (2004) case study of coprincipals, each person’s expertise overlaps with the partner so that both reinforce one another. Third, the members’ attributes might be duplicated, which means that, when required, the partners may substitute for one another. When a third party augments the dyad to form a triad, maintenance of the reciprocity implicit in dyadic complementarity requires significant interpersonal adjustment. Third members always risk being in a minority of one, either experiencing isolation or being a vehicle, Oedipal-like, for the routing of conflicts between the other

10 From Distributed to Hybrid Leadership Practice

207

two members. A triadic role set will integrate to become a constellation, as distinct from a mere aggregate, provided that the members negotiate a division of labour based on specialized formal (and psycho-emotional) role sharing. Specialization of functions in turn begets a second property of constellations, their individual differentiation of roles, which in turn begets a third, the afore-mentioned complementarity (Hodgson et al., 1965, p. 489). One of the key differences between dyadic pairings and triads, according to Alvarez and Svejenova (2005, p. 150), following their extensive review, is the greater likelihood of the emergence of a supra-individual sense of identity in a trio, a dimension which raises the possibility of holism.

Emergence and Holism As a standpoint that is antipathetic to methodological individualism, holism posits that, in the relations between structural levels of phenomena, higher level social phenomena are more significant than lower level elements. They may, for example, have causal power. The derivation of holistic structures opens up a whole series of complex questions concerned with emergence, relations between levels and reductionism (see Sawyer, 2005, pp. 52–61). For the present discussion, it suffices that, in a group or small number context, provided collective or plural member entities achieve the constellation status just discussed, then they amount to more than the mere sum of their parts. This is a distributed leadership possibility for which I allowed with the idea of conjoint agency expressed in concertive actions (Gronn, 2002a, p. 429), whereas Gibb made just two oblique passing references to the holistic potential of small groups. At one point he referred to “the multiplicity or pattern of group functions performed” by leaders and “the many roles of which the leadership complex is constituted”, while also speculating, intriguingly, that “the concept of the leader will be of no further value to us” when this complex had been differentiated (Gibb, 1968a, p. 215). The practical import of holism for working in concert struck me most forcefully while listening to an address to the 2007 Conference of the British Educational Leadership Management Administration Society (BELMAS), by Hilary Emery (Training and Development Agency for Schools). In a review of workforce modernization in English schools, Emery claimed there are about 40 school roles supporting leadership for learning, grouped into seven key areas: teaching assistance, student welfare, technical assistance, site, facilities, administration and other pupil support. Currently, just over 40% of staff in schools are support staff. When visual representation of these seven clusters was combined in a power point slide with the range of integrated children’s services and program initiatives in schools (e.g., health, police, youth justice, child care provision) the effect was to create a powerful image of a complex division of school leadership labour. It takes little effort to see how the learning-related problem solving effectiveness of these emergent role sets might be enhanced by a dose of constellation-style membership understandings.

208

P. Gronn

Hierarchy and Heterarchy A sure sign of hybridity is the intermingling of both hierarchical and heterarchical modes of ordering responsibilities and relations. In hierarchical ordering, each level is successively implicated in the higher level, whereas with heterarchy, “various levels exert a determinate influence on each other in some particular respect” (Kontopoulos, 1993, p. 55). Situations lacking any clear ordering of elements, on the other hand, as in Buchanan et al.’s (2007) example of aspects of distributed decisionmaking in health care, are quintessentially heterarchical. Indicative evidence of hybrid leadership practice, and the mingling of heterarchical and hierarchical ordering modes are found in recent research by Spillane et al. (2007), Leithwood et al. (2007) and Timperley (2005). Using an experience sampling method (ESM), Spillane et al. (2007) investigated leadership practice by tracking the daily activities engaged in by 42 US principals and their colleagues over a 6-day period. The principals were randomly beeped 15 times daily, during which they each completed a 10-item questionnaire on handheld computers. For four categories of activities, principals were required to report whether they were leading, co-leading, with whom they were co-leading or who was leading if they were not, and with leaders defined as those who “have responsibility for executing the activity” (Spillane et al., 2007, p. 110). The results display both hierarchical and heterarchical elements. First, in “Cloverville” there was evidence of a combination of solo performed and co-performed work. Thus, leadership was hierarchical, with principals thinking they were leading for about two-thirds of the time, about half of which as sole leader (Spillane et al.’s Table 3). Leadership was also heterarchical, because for the other half of the time principals led in conjunction with a colleague, mostly teachers, and teachers (by an overwhelming margin) also led for about a third of the time principals were not leading (Spillane et al.’s Tables 4 & 5)—a result providing some comfort to proponents of teacher leadership. Leithwood et al. (2007) adapted my three suggested forms of concertive action (Gronn, 2002a, pp. 430–431): spontaneous collaboration, intuitive working relations and institutionalised practices. These were aligned with four leadership functions: direction setting, people development, organisational re-design and instructional management. With an eye to distributed leadership’s “potential to leverage organizational expertise”, Leithwood et al. (2007, p. 47) were particularly interested in the engagement of informal “nonadministrator leaders”. Once more (this time in southern Ontario, Canada) a mixed pattern of solo and shared leading was evident. Based on interviews with school and district informants from eight schools, Leithwood and colleagues found evidence of the aforementioned differentiation and specialization of functions, more colleagues working together on complex tasks rather than simple ones and attributions of leaders tending to match those with formally designated leader status. Informal leaders made a high contribution to three of the four leadership functions (Leithwood et al.’s Table 1). Thus, in relation to direction setting, once a vision was in place, these informal leaders tended to perform the on-the-ground battle for the hearts and minds of colleagues. If, however, informal leaders were to perform in this way (singly or collectively), there still had

10 From Distributed to Hybrid Leadership Practice

209

to be regular monitoring by principals. Distributed leadership, it seemed, depended on “effective forms of focused leadership—leading the leaders” (Leithwood et al., 2007, p. 55). Timperley sought evidence of distributed leadership during a governmentsponsored four-year school improvement initiative in Auckland, New Zealand—an intervention aimed at improving disadvantaged children’s literacy skills following the provision of external professional development for teachers. She observed and tape-recorded literacy team meetings in seven schools, analysed the transcripts and interviewed the teachers about their meetings, during which they utilised students’ achievement data from their own and other classes, along with national benchmark summaries of reading achievement scores. Although teachers were meant to collaborate to identify poorly performing children with a view to collectively devising strategies to improve their progress, in Year 2 of the project some teachers were reluctant to discuss particular children as this would reflect poorly on their own sense of professionalism. In these instances of unproductive engagement, meeting talk was about whole classes or year levels rather than evidence-based strategies for needy children. By Year 3, however, in these same schools the dialogue had become much more productive, with the crucial ingredient for change being the literacy teacher leader’s resolve to focus on teachers’ beliefs and practices, with a view to changing them (Timperley, 2005, p. 414). This strategy proved to be successful, so much so that, when interviewed, one teacher who had previously been reluctant to examine her practice closely now knew clearly what she had to do (using graphed evidence of individuals’ progress) if she was to improve the performance of needy children in her class against the national benchmarks. Once more, then, leadership practice comprised collaborative work (teacher teams), but with significant roles also performed by designated individual leaders who spanned the boundaries between principals and teacher teams (Timperley, 2005, p. 410). None of the authors of these examples use “hybrid”. In my view, however, the mix of the combined work of solo, dyad and team leadership formations is a more accurately described as hybrid than distributed. In retrospect, then, even though I allowed originally for distributed leadership to encompass, in addition to holistic groupings (Gronn, 2002a, p. 429), the idea of a potentially large number of leaders in schools (i.e., 1+ leaders), this possibility underplays the significance of the contributions of highly influential individuals working in parallel with collectivities (i.e., from pairings through to small size teams). That is, an emphasis on the number and multiplicities of individual leaders ignores qualitative variations and combinations. To strengthen the point: consider a hypothetical case in which schools with numerous solo performers, each of whom might fit a classic charismatic or transformational prototype, to which might be added a couple of teams and teacher networks, all of which adds up to a critical mass of leaders. To characterise such an overall leadership configuration as “distributed”, particularly when it is dominated by high profile prima donnas, would not necessarily be an accurate representation of reality. In hindsight, then, it may have been better to confine “distributed” to instances of conjoint agency (Gronn 2002a, p. 431).

210

P. Gronn

Unfinished Business The previous discussion has raised a number of issues. Here I concentrate on three: two that have been recently highlighted by Hatcher (2005) and the third in which I attempt to clinch the argument about hybrid practice. First, is it leadership we are talking about or is it something else, and how do we know? Second, is distributed leadership merely another way of talking about democratic leadership? Third, what does it mean to describe leadership practice as “hybrid”?

Leadership or Something Else? Hatcher’s (2005) answer to this question is something else: i.e., power. As I suggested at the outset of the chapter, there have always been close conceptual relations between power, influence, authority and leadership. For this reason, discussions of leadership are bound to be plagued by demarcation disputes. What, then, is the relationship between these terms? Are some of them synonyms, can they co-exist in schools and organizations generally or might one term subsume the others? Whereas I have allowed in my notion of distributed leadership for the possibility of the influence of school leaders (e.g., teachers) to trump the authority of head teachers, Hatcher’s hard line objection to this possibility is the reverse: the managerial prerogative will always and everywhere trump influence. On this point, Hatcher (2005, p. 256) claims that: authority (power) is not simply one of a number of dimensions of activity in organisations; it is a different category of phenomenon, because it over-determines all the other dimensions.

Does it? Always? Such triumphalism assumes that English schooling managerialism is by definition malign, except that there is a cultural conditional or overlay operating here: so tight is the coupling between centrally determined targets and the surveillance of schools that, as far as Hatcher is concerned, head teachers reduce to mere state functionaries who, stripped of agency, implement state policy automatonlike and enforce teacher compliance. In other cultural contexts, and in previous times in England (Hatcher, 2005, pp. 261–265), the coupling between system and schools has been much looser and, to paraphrase Mao Zedong, if not 1000 or 100, then certainly at least some democratic, flowers have managed to bloom. There are three points here. First, in my own case the cultural overlay that shaped what I have written on distributed leadership has been Antipodean, not English. In Australia, particularly in Victoria, where school inspection (a crucial weapon of state power for Hatcher) has been dead and buried for about 25 years, it is still possible for authority/power to be trumped and where it is not then to be at least significantly mitigated by influence/leadership (see Gronn, 2008). Second, despite Hatcher’s managerialist pessimism, mitigation or even occasional trumping may even be possible in England. Head teachers need not be na¨ıve bearers of state edicts and can, in fact, resist aspects of state policy (and what is more get away with it), as Thomson’s (2008) examples of head teachers behaving badly show. The third reason

10 From Distributed to Hybrid Leadership Practice

211

why all hope may not be lost is evidence of target gaming by service providers— ratchet effects, threshold effects, output distortion and creative categorization (the first three of which also occurred under earlier state managerialism in the former USSR)—across the UK public services (Hood, 2006). The good news, then, in face of Hatcher’s pessimistic, structurally determined managerial shut-down is that slippages of power are real. While Hatcher’s resolution of these points about power, leadership and the relationship between the two may be suspect, he is right to raise them, principally because they invite a number of wider questions. These include: What are the possible ways with which to characterize collaborative human conduct, interaction and action in organizations? Instances, perhaps, of Lukes’ (2005) significant affecting group of phenomena? What might be the justification for adopting one such mode of characterization (such as persuasion, coercion, manipulation, leadership or power) rather than another? In the late 1970s and early 1980s, there was a strong academic focus by scholars on micro-politics and power in what was then known as the field of educational administration (or management). More recently, that language has been superseded and leadership is now the default option. While power may lack the currency and mystique of leadership, these are insufficient reasons for ignoring or marginalising it. Specifically regarding whether leadership may or may not be the appropriate characterisation, in a significant admission Spillane et al. (2007, p. 105) claim not to have known whether the principals’ reported activities in their distributed leadership study counted as management or leadership, as these “happen in tandem and are often intertwined”. Moreover, while Sawyer et al. (2005) proclaim the virtues of conversation analysis as a technique for studying distributed leadership, it is unclear (in the absence in their research of any explicit justification) why their numerous examples of interesting dialogue may count as instances of distributed leadership rather than other more traditional phenomena such as collective decision-making or delegated authority. Clearly, then, important conceptual surgery still needs to be performed on leadership and terms closely associated with it.

Is Distributed Leadership Democratic? Hatcher (2005, pp. 257–261) is also critical of the equation of distributed leadership with organizational democracy. Similarly, Maxcy and Nguyeˆ˜ n (2006, p. 180) claim that the rhetoric of leadership distribution masks an anti-democratic managerial bias. Such criticisms, I think, rest on much firmer ground. There are a number of reasons for the attractiveness of the idea that, even if it may not be synonymous with democratic leadership, distributed leadership at least lays the groundwork for it—a connection that is considered in more detail by Woods and Gronn (2008). One reason is the idea of “voice”. That is, by de-monopolising leadership and potentially increasing the sources and voices of influence in organizations beyond just one, distributed leadership has helped widen the span of employee and member participation. Part of the issue about the validity of this claim turns

212

P. Gronn

on whether or not a mechanism such as voice is sufficient to realize, in Hatcher’s (2005, p. 261) percipient phraseology, an “authentically participative professional culture”, or whether what is required for this is something stronger such as a veto. In this respect Hirschman’s (1970, p. 4) distinction between voice and “exit”, the two main ways in which citizens or consumers exercise choices in respect of their range of commitments and loyalties (e.g., to organisations, products, pursuits), may be helpful. Voice, the quintessential mode of mass political engagement in a liberal democracy, entails the articulation of affirmative or negative claims, either directly by virtue of membership or indirectly through some mode of representation, to managers and power-holders. Exit, by contrast, which is Hirschman’s shorthand term for consumer market behaviour, means action in which one votes with one’s feet, as it were, by withdrawal, disengagement, non-renewal of membership or by switching of allegiances. Exit, typically, is adopted as the option of last resort when voice fails. Adapting this idea slightly, a collective sense of veto or, as Hatcher (2005, p. 262) so ably expresses it, “collective democratic control through elected management bodies subject to the membership of the organization” is a world (or perhaps even a galaxy) away from the promise and practice of distributed leadership. As mentioned at the outset of this chapter, numerous public agencies in education are trumpeting the virtues of distributed leadership (e.g., the National College for School Leadership), but to the best of my knowledge none of them are espousing doctrines that come anywhere near such collectively determined democratic leadership. If the experiences of teacher-elected principals in Spain (B´olivar and Moreno, 2006) and teacher-elected departmental heads in Portugal (de Lima, 2008) attest to the operational tensions associated with school democracy, then perhaps this shyness about anything beyond distributed leadership is understandable. In addition to Hatcher’s examples of democratic school leadership (such as the well-known one of Counteshorpe College), there are historical precedents for school democracy. Knox (1953, pp. 39–40), for example, provides a number of instances of secondary education in nineteenth century Scotland in which rectors had limited managerial capacity to control schools and, in some cases, these were controlled by councils of masters. Then, in the 1970s in Victorian state high schools, the secondary teacher’s union advocated an elected staff executive to control schools, to which principals would be responsible and of which they would also be members (Gronn, 2002b)—a model akin to that adopted in Spain from 1970. These instances also go the heart of Hatcher’s concerns about power and leadership, as do those of Maxcy and Nguyeˆ˜ n (2006, p. 87), who show that while teachers at their two case study schools seized the opportunities to exercise leadership, their distributed leadership was “tethered” or confined to domains legitimated by the managers of the schools and the district. The long and the short of these examples, is that the distinction between distributed and democratic leadership is genuine. Thus, having regard to the interests of policy-makers and the powers that be, distributed leadership continues to be politically palatable as a normative possibility, perhaps for the kinds of reasons given by Hartley (2007), whereas democratic leadership, for the most part, is still considered outr´e.

10 From Distributed to Hybrid Leadership Practice

213

What Does it Mean to Map Leadership Practice? To “map” practice is simply shorthand for describing and explaining patterns of leadership manifest in particular contexts. In the main part of the discussion in this chapter, and elsewhere on the basis of the findings of an Australian case study (Gronn, 2008), I have been proposing that the leadership practice to be mapped, regardless of situations or circumstances, is best thought of as varied, hybrid and diverse in character. I now want to develop this point. What does “hybrid” mean? What is the justification on which hybridity rests? The reasoning in support of this claim is perhaps a little bit abstract and technical, and as such it no doubt warrants more extended treatment in a full length article. Moreover, in what I say now, I probably run the risk of being seen as creating false binaries, although that is not my intention, and offending vested interests. The starting point for what I want to argue is to begin by contrasting hybridity with the notion of “one-size-fits-all”. A one-size-fits-all paradigm implies that every (or virtually every) instance of a general class of objects, events, circumstances or occurrences is, to all intents and purposes, a clone or replica of the others. That is, in the uniform (or near uniform) likeness or similarity of their features they represent standard manifestations of a type. Each instance lacks singularity and this absence of singularity may be characterized as evidence of an overall pattern of convergence. The opposite of such a pattern is divergence, in which the totality of the units in question is heterogeneous (i.e., the differences between the units is far greater than the similarities). In making a case for hybridity, I do not have heterogeneity in mind. Rather, if we go return for a moment to Gibb’s, (1954, 1968a) distinction between focused and distributed leadership, then we can view these terms as contrasting points of possible convergence. Thus, as part of the group dynamics he was researching, focused leadership applied as a description when and if (in both snap-shot terms and longitudinally) one individual’s influence could be shown to predominate. On other occasions and in other locations when pluralities of persons were influential, however, then the leadership was distributed. The hybridity that I am calling for is a focused-distributed mix, in which varying degrees of each co-exist, and then a range of formations within the idea of “distribution” (MacBeath, 2005; Bolden et al. 2008, p. 39). One of my longstanding criticisms of the field of leadership has been that, notwithstanding the insights of Gibb and the other early-generation writers documented above, leadership scholars have been fixated overwhelmingly on focused understandings to the detriment of distributed formations. Thus, they have adopted an a priori division of labour of one “leader” and many “followers”. Recently, however, with the discovery (or more appropriately, as I have suggested, re-discovery) of distribution, there has been a surge of interest in, and a shift in focus to, an alternative possibility. Yet, this is a situation that is as equally unreflective (or even potentially mindless) as the one to which it has arisen as a reaction: a focused monolithic paradigm suddenly incurs the risk of being supplanted by an equally monolithic distributed one. If scholars have been just as guilty of romanticizing individual leadership heroics as the informants whose leadership attributions they

214

P. Gronn

have documented, then the last thing the field needs is to substitute the late James Meindl’s (1995) “Romance of leadership” idea with something akin to a “Romance of distributed leadership”. Quite apart from the fact that adherence or allegiance to either of these understandings would be blindly unreflective, each understanding inaccurately represents reality. This is because the accumulating empirical evidence, as summarized in this chapter (and elsewhere), points increasingly to the existence of hybrid mixes of leadership: numerically distributed individuals acting alone or loosely in concert (Bolden et al., 2008; de Lima, 2008; Firestone & Martinez, 2007), numerically distributed individuals, couples and partners co-existing with other holistically distributed forms (Gronn, 2008) and varying degrees of embedded or emerging shared practice (Ritchie and Woods, 2007). The major implication of these hybrid trends for researchers, as I see it, therefore, is that what gets mapped in particular contexts or situations is simply “leadership” as it is practiced. Any endorsement of particular adjectival brands or typifications of leadership (e.g., charismatic, transformational, visionary, authentic) is unnecessary and bogus, in which case the work of scholars would become one of describing and accounting for patterns manifesting varying degrees of both concentrated and shared influence. One consequence of this disavowal of advocacy of particular leadership brands might be to better align the work of leadership commentators with that of their peers working in the field of power, where such advocacy is anathema. The kind of commitment I am proposing would be legitimate in itself as an intellectual endeavour or, if scholars’ interest is in leadership impact, they could choose (as many of them currently do) to go the next step and seek evidence (if any) of causal effects.

Conclusion In this chapter I have paid tribute to the knowledge and insight that the recent interest in a distributed leadership perspective has brought to bear. But I also expressed some caveats. My particular concerns included a scholarly ignorance that denies distributed leadership’s continuity with intellectual precedence; an absence of recognition of parity of status accorded parallel forms of distributed human conduct and a permissive discursiveness that is inclined to hoover up indiscriminately as “leadership” numerous aspects of practice. While distributed leadership will no doubt continue to assist in understanding school-level decision-making practice, I have also articulated a need to move beyond distributed leadership and have argued a case for hybridity as a more accurate representation of diverse patterns of practice which fuse or coalesce hierarchical and heterarchical elements of emergent activities. For all of these reasons, I raised the possibility of slightly refining current meanings of distributed leadership along with the need to better think through its relationship to two closely allied conceptual domains, power and democratic leadership in organizations and also to begin thinking in “post”-distributed leadership terms. As was hinted earlier, to the extent that conceptual analysis of an idea may be said to precede its empirical investigation, which in turn may precede some evidential measure of its impact (and therefore, presumably, some estimate of its utility), there

10 From Distributed to Hybrid Leadership Practice

215

is still much to do both conceptually, empirically and impact-wise with distributed leadership. On the other hand, practice is rarely as neat and discrete as is implied by this linear ordering of developmental stages. With the kinds of misgivings and adjustments that I have suggested in mind, it is not clear where distributed leadership (and leadership more generally for that matter) goes from here—the tide of popularity, for example, may soon ebb—and, in particular, whether it follows the preferred line of development I have articulated. Whatever the future might hold for this particular idea, however, the opportunity to reflect on its career history and scholarly uptake has reinforced the claim that distributed leadership’s contribution to better understanding and appraising the work of organizations, especially schools, has so far been both insightful and productive.

References ALVAREZ, J.L. & SVEJENOVA, S. (2005) Sharing Executive Power: Roles and Relationships at the Top (Cambridge: Cambridge University Press). BECKER, H. & USEEM, R.H. (1942) Sociological analysis of the dyad, American Sociological Review, 7(1): 13–26. BENNE, K.D. & SHEATS, P. (1948) Functional roles of group members, Journal of Social Issues, 4(2): 41–49. BOLDEN, R., PETROV, G. & GOSLING, J. (2008) Developing Collective Leadership in Higher Education, Final Report (London: Leadership Foundation for Higher Education). BOL´lVAR, A. & MORENO, J.L. (2006) Between transaction and transformation: The role of school principals as education leaders in Spain, Journal of Educational Change, 7(1): 19–31. BRYMAN, A. (1996) Leadership in organizations, in S.R. CLEGG, C. HARDY & W.R. NORD (Eds), Handbook of Organization Studies (London: Sage), pp. 276–292. BUCHANAN, D.A., ADDICOTT, R., FITZGERALD, L., FERLIE, E. & BAEZA, J.I. (2007) Nobody in charge: Distributed change agency in healthcare, Human Relations, 60(7): 1065–1090. CAPLOW, T. (1956) A theory of coalitions in the triad, American Sociological Review, 21(4): 489–493. CROZIER, M. (1964) The Bureaucratic Phenomenon (Chicago: University of Chicago Press). DAHL, R.A. (1961) Who Governs?: Democracy and Power in an American City (New Haven: Yale University Press). de LIMA, J.A. (2008) Department networks and distributed leadership in schools, School Leadership & Management, 28(2): 159–187. ETZIONI, A. (1965) Dual leadership in complex organizations, American Sociological Review, 30: 688–698. FIRESTONE, W.A. (2007) Districts, teacher leaders, and distributed leadership: Changing instructional practice, Leadership and Policy in Schools, 6(1): 3–35. FOLLETT, M.P. (1973 [1927]) Dynamic Administration: The Collected Papers of Mary Parker Follett, E.M. Fox & L. Urwick (Eds) (London: Pitman & Sons). FONDAS, N. & STEWART, R. (1994) Enactment in managerial jobs: A role analysis, Journal of Management Studies, 31(1): 83–103. FRENCH, Jr, J.R.P. & SNYDER, R. (1959) Leadership and interpersonal power, in CARTWRIGHT, D. (Ed.), Studies in Social Power (Ann Arbor: Institute for Social Research, University of Michigan), pp. 118–149. GIBB, C.A. (1954) Leadership, in G. LINDZEY (Ed.), Handbook of Social Psychology, Vol 2 (Reading, MA: Addison-Wesley), pp. 877–917.

216

P. Gronn

GIBB, C.A. (1958) An interactional view of the emergence of leadership, Australian Journal of Psychology, 10(1): 101–110. GIBB, C.A. (1968a) Leadership, in G. LINDZEY & E. ARONSON (Eds), The Handbook of Social Psychology, 2nd Ed, Vol 4 (Reading, MA: Addison-Wesley), pp. 205–283. GIBB, C.A. (1968b) Leadership: Psychological aspects, in D.L. Sills (Ed.), International Encyclopedia of the Social Sciences, Vol 9 (New York: MacMillan/Free Press), pp. 91–101. GRONN, P. (1999) Leadership from a distance: Institutionalising values and forming character at Timbertop, 1951–1961, in P.T. BEGLEY & P. LEONERD (Eds), The Values of Educational Administration (London: Falmer), pp. 140–168. GRONN, P. (2002a) Distributed leadership as a unit of analysis, Leadership Quarterly, 13(4): 423–451. GRONN, P. (2002b) Distributed leadership, in K. LEITHWOOD & P. HALLINGER (Eds), Second International Handbook of Educational Leadership and Administration (Dordrecht: Kluwer), pp. 653–696. GRONN, P. (2003) Distributing and intensifying school leadership, in N. Bennett & L. Anderson (Eds), Rethinking Educational Leadership: Challenging the Conventions (London: Sage), pp. 60–73. GRONN, P. (2008) Hybrid leadership, in K. LEITHWOOD, B. MASCALL & T. STRAUSS (Eds), Distributed Leadership according to the Evidence (Mahwah, NJ: Erlbaum), pp. 17–40. GRONN, P. & HAMILTON, A. (2004) ‘A bit more life in the leadership’: Co-principalship as distributed leadership practice, Leadership and Policy in Schools, 3(1): 3–35. HARTLEY, D. (2007) The emergence of distributed leadership: Why now?, British Journal of Educational Studies, 55(2): 202–214. HATCHER, R. (2005) The distribution of leadership and power in schools, British Journal of Sociology of Education, 26(2): 253–267. HIRSCHMAN, A.O. (1970) Exit, Voice and Loyalty: Responses to Decline in Firms, Organizations, and States (Cambridge, Mass: Harvard University Press). HMIE (2007) Leadership for Learning: The Challenges of Leading in a Time of Change (Livingstone, Scotland: HM Inspectorate of Education). HODGSON, R.C., LEVINSON, D.J. & ZALEZNIK, A. (1965) The Executive Role Constellation: An Analysis of Personality and Role Relations in Management (Boston: Graduate School of Business Administration, Harvard University). HOOD, C. (2006) Gaming in targetworld: The targets approach to managing British public services, Public Administration Review, 66(4): 515–521. KATZ, R.L. & KAHN, D. (1978) The Social Psychology of Organizing, 2nd Edition (New York: Wiley). KNOX, H.M. (1953) Two Hundred and Fifty Years of Scottish Education (Edinburgh: Oliver & Boyd). KONTOPOULOS, K.M. (1993) The Logics of Social Structure (Cambridge: Cambridge University Press). LEE, F. & TIEDENS, L.Z. (2001) Is it lonely at the top?: The independence and interdependence of powerholders, in B.M. STAW & R.I. SUTTON (Eds) Research in Organizational Behavior, 23 (Amsterdam: JAI Press), pp. 43–91. LEITHWOOD, K., MASCALL, B., STRAUSS, T., SACKS, R., MEMON, N. & YASHKINA, A. (2007) Distributing leadership to make schools smarter: Taking the ego out of the system, Leadership and Policy in Schools, 6(1): 37–67. LUKES, S. (2005) Power: A Radical View, 2nd edition (London: Palgrave). MACBEATH, J. (2005) Leadership as distributed: A matter of practice, School Leadership & Management, 25(4): 349–366. ˜ˆ T.S.S. (2006) The politics of distributing leadership: Reconsidering MAXCY, B.D. & NGUYEN, leadership in two Texas elementary schools, Educational Policy, 20(1): 163–196. MEINDL, J.R. (1995) The romance of leadership as a follower-centric theory: A social constructionist approach, Leadership Quarterly, 6: 329–341.

10 From Distributed to Hybrid Leadership Practice

217

O’LEARY, M., ORLIKOWSKI, W. & YATES, J. (2002) Distributed work over the centuries: Trust and control in the Hudson’s Bay Company, 1670–1826, in P. HINDS & S. KIESLER (Eds), Distributed Work (Cambridge, Mass: MIT Press), pp. 27–54. POWELL, M. (1997) The Whitlam Labour government: Barnard and Whitlam: A significant historical dyad, Australian Journal of Politics and History, 43(2): 183–199. REID, I., BRAIN, K. & BOYES, L.C. (2004) Teachers or learning leaders?: Where have all the teachers gone? Gone to be leaders, everyone, Educational Studies, 30(3): 251–264. RITCHIE, R. & WOODS, P.A. (2007) Degrees of distribution: Towards an understanding of variations in the nature of distributed leadership in schools, School Leadership & Management, 27(4): 363–381. SAWYER, R.K. (2005) Social Emergence: Societies as Complex Systems (Cambridge: Cambridge University Press). SAWYER, R.K., SCRIBNER, J.P., WATSON, S.T. & MYERS, V.L. (2005) Talking leadership: Conversation analysis and distributed leadership, in W. HOY & C. MISKEL (Eds), Educational Leadership and Reform (Greenwich, CT: Information Age), pp. 169–185. SCHEIN, E.H. (1988) Organizational Psychology (Englewood Cliffs, NJ: Prentice-Hall). SENGER, J. (1971)The co-manager concept, California Management Review, 13(3): 77–83. SIMMEL. G. (1902) The number of members as determining the sociological form of the group. 1 & II, American Journal of Sociology, 8(1): 1–46, 158–196. SPILLANE, J. P., CAMBURN, E.M. & PAREJA, A.S. (2007) Taking a distributed perspective to the school principal’s workday, Leadership and Policy in Schools, 6(1): 103–125. THOMSON. P. (2008) Headteacher critique and resistance: A challenge for policy, and for leadership/management scholars, Journal of Educational Administration and History, 40(2): 85–100. TIMPERLEY, H.S. (2005) Distributed leadership: Developing theory from practice, Journal of Curriculum Studies, 37(4): 395–420. WHISLER, T.L. (1960) The “assistant-to” in four administrative settings, Administrative Science Quarterly, 5(2): 181–216. WOODS, P.A. & GRONN, P. (2008) Nurturing democracy: The contribution of distributed leadership to a democratic organisational landscape, Educational Management, Administration & Leadership (accepted and in press).

Chapter 11

Fit for Purpose: An Educationally Relevant Account of Distributed Leadership V.M.J. Robinson

Currently, there is intense interest in the relationship between educational leadership and student outcomes. The interest is evident in the number of recent reviews, syntheses and meta-analyses of the available published evidence (Leithwood et al., 2008; Marzano et al., 2005; Mulford, 2008; Robinson et al., 2008). These publications provide an opportunity for sober reflection on the state of research on educational leadership. Perhaps the most important conclusion to be drawn from them is that there is a radical disconnection between research on educational leadership and the core purpose of schooling – the education of children. The disconnection is most compellingly demonstrated by the miniscule proportion of publications that have empirically tested the relationship between aspects of educational leadership and student achievement and well-being. Of the thousands of published studies of educational leadership, less than 30 have empirically tested this relationship (Robinson et al., 2008). Given the much shorter history of research on distributed leadership in education, the number of such studies is, understandably, substantially fewer. Another indicator of the disconnection between leadership research and teaching and learning is the popularity of generic theories of leadership. By “generic” I mean those theories which could apply to the leadership of any type of organisation. Transformational leadership theory, for example, is an attempt to explain why some leaders are able to engender exceptional levels of effort, commitment and loyalty from staff in pursuit of a shared vision of change and transformation (Bass, 1985; Burns, 1978). Transformational leadership is generic in the sense that it is based on studies of leadership in corporations, governmental and not-for-profit organisations. This means that theories of transformational leadership do not capture the educationspecific knowledge and skills that are required to establish the conditions that make it possible for teachers to have a bigger positive impact on students. The more generic nature of transformational leadership may explain why this type of leadership seems to have substantially less impact on student outcomes than the more education-specific approach known as instructional leadership (Robinson et al., 2008). V.M.J. Robinson (B) The University of Auckland, Auckland, New Zealand A. Harris (ed.), Distributed Leadership, Studies in Educational Leadership 7, C Springer Science+Business Media B.V. 2009 DOI 10.1007/978-1-4020-9737-9 12, 

219

220

V.M.J. Robinson

Given the burgeoning interest in distributed leadership in education, it is timely to consider how research on this topic could make stronger and more rapid connections with student outcomes than has been evident in the history of the parent field of educational leadership (Harris, 2008). The purpose of this chapter, therefore, is to develop an account of distributed leadership that is appropriate for research on this relationship. Several arguments have been put forward about why strong distributed leadership might enhance student outcomes. One argument is based on the assumption that under a pattern of distributed leadership, more of the expertise and talent of staff will be identified, developed and utilised than under a more traditional hierarchical pattern (Camburn et al., 2003; Fullan, 1993; Harris, 2005a). As Harris puts it, “distributed leadership means multiple sources of guidance and direction, following the contours of expertise in an organisation, made coherent through a common culture” (Harris, 2005a, p. 258). This argument seems particularly compelling given the breadth and depth of pedagogical expertise required to meet today’s ambitious goal of all students succeeding on intellectually challenging curricula. A second and related argument is to do with the sustainability of efforts to improve teaching and learning. Schools with stronger distributed leadership will, it is argued, have more staff that are knowledgeable about, and take responsibility for, the improvement of educational outcomes. Such distribution of knowledge, responsibilities and instructional leadership roles should protect a school improvement effort against the consequences of a loss of key personnel. Such losses, together with a failure to develop a broad base of strong instructional leadership, have been identified as key reasons for the stalling of improvement efforts (Camburn et al., 2003; Elmore, 2004). Although conceptual analyses of distributed leadership have already been published, these analyses have not been constrained by the goal of forging links with educational outcomes. Indeed, much of the research has been motivated by interest in how distributed leadership alters the flow of information and the power relationships between the adults, rather than by interest in its impact on teaching and learning (Goldstein, 2004; Hatcher, 2005; Urbanski and Nickolaou, 1997). I argue that if research programmes on distributed leadership are to investigate its impact on student outcomes, the concept and its associated measures need to discriminate between leadership practices in general and those particular leadership practices that develop more effective teaching and learning. Developing an account of distributed leadership that makes this kind of discrimination requires an integration of research on leadership with pedagogical theory and evidence – an integration that is not clearly apparent in existing conceptual and empirical research on distributed leadership in education. My argument proceeds as follows: First, I develop an account of both leadership and distributed leadership that is defensible in terms of the social psychology of leadership. Second, I discuss how those accounts need to be modified to reflect the educational evidence about the conditions under which teaching and teacher learning makes a positive difference to student outcomes. Third, I apply

11 An Educationally Relevant Account of Distributed Leadership

221

this educationally informed concept of distributed leadership to several of the existing research programmes on distributed leadership in education. The purpose of this application is to consider the extent to which the concepts and measures of distributed leadership used in these research programmes are suited to inquiry into the links between distributed leadership and student outcomes. It should be noted that since the research under discussion was not designed to investigate those links, this section does not constitute a critique of the existing studies. Rather it seeks to address the question of the shifts that might be required if we are to discover the types of distributed leadership practice that have more and less impact on student outcomes.

What is Leadership? There are at least three different concepts of leadership in the research literature (Katz and Kahn, 1966). First, leadership is associated with particular types of position. A head teacher, by virtue of being in that position, is expected to set the direction for the school; a head of department is expected to provide direction, coordination and oversight of the activities of the subject teachers within the department. Yet head teachers and heads of department vary in their capacity to perform such functions – hence the recognition that leadership is also a personal quality and that some individuals exhibit more of the requisite knowledge, skills and dispositions than others. This second sense of leadership as a personal quality needs to be qualified by recognition of the interaction between leadership and context. Too strong an emphasis on the personality traits of leaders denies the intimate connection between the exercise of leadership and the characteristics of task, followers and situation. A head who is a successful leader in one school may be far less successful in a subsequent appointment where the school community and teacher culture are very different from those experienced in the prior position. The third concept of leadership emphasises the behaviours or practices involved in leadership. This approach decouples leadership from formal position as it recognises that those who hold such positions may or may not exhibit either the qualities of a leader or the behaviours that would normally be associated with holders of such positions. Distributed leadership is most strongly associated with this third concept as it is inclusive of the leadership of those in both formal and informal leadership roles. Leadership as a contribution to group purposes. The third concept of leadership associates it with the performance of particular functions or behaviours rather than with particular positions or personal qualities. This approach requires a strategy for distinguishing between leadership and non-leadership functions or behaviours. The distinction is typically drawn by contextualising leadership within groups or organisations. One of the defining characteristics of such entities is the interdependence of their members in pursuit of common goals (Gibb, 1969). Leadership is attributed

222

V.M.J. Robinson

to those members who are seen to influence others in ways that advance the group’s progress towards their goals (Katz and Kahn, 1966; Robinson, 2001). Katz and Kahn (1966, p. 303) make the connection between leadership and advancing group purposes as follows: “. . .every act of influence on a matter of organisational relevance is in some degree an act of leadership.” They elaborate the phrase “in some degree” by explaining that some acts of influence are little more than routine role performance. Imagine a school where there are well established routines for the collaborative review of student work. In such a school, a Head of Department who routinely receives samples of students’ work from teachers is exercising only a minimal level of leadership. Now imagine obtaining the same degree of cooperation in a school that is just beginning to establish a culture of collective review of student work. In this latter case a considerably greater degree of leadership is being exercised. Leadership is associated more with the exceptional rather than the routine end of the influence continuum. As Katz and Kahn explain “we consider the essence of organizational leadership to be the influential increment over and above mechanical compliance with the routine directives of the organization” (1966, p. 302). The problem of distinguishing ordinary from more significant acts of leadership is tackled slightly differently by Brian Fay (1987) in his account of leadership. For Fay, leadership is not just goal-relevant influence but influence that causes others to think or act differently. The concept of leadership is thus closely allied to the achievement of change – leadership by definition, for Fay at least, is the disruption of the routine and the initiation of new ways of thinking and acting. Leadership as a particular type of influence process. A further distinction is critical to an adequate concept of leadership. Leadership must be distinguished from other types of goal-directed influence process such as force, coercion and manipulation. While these are clearly influence processes, we do not call them leadership because the source of the influence is not one that is associated with leadership. Leadership influence is based on followers’ judgments that “the leaders occupy a position which gives them the right to command a course of action, or that they possess the requisite personal characteristics of leaders, or that they seek an action which is correct or justifiable” (Fay, 1987, p. 121). The first of Fay’s three sources of leadership influence is legitimate use of positional authority. If that authority is not seen as legitimate, then it does not count as leadership. On the other hand, as Gibb makes clear, if a particular use of authority is seen as legitimate, it can also be an act of leadership. While positional authority is granted by virtue of position, leadership is earned through follower perception that the directive attempts of the position holder are legitimate. “. . .many [organisation] heads are recognized by their subordinates as making very positive contributions to group progress and are therefore accorded willing cooperation and, through it, leadership status” (Gibb, 1969, p. 213). Fay’s reference to personal qualities is similar to French and Raven’s classic account of referent power, in which the source of influence is followers’ personal liking of or identification with the leader(French and Raven, 1960).

11 An Educationally Relevant Account of Distributed Leadership

223

The third source of leadership influence is allied to the notion of relevant expertise – the influence attempt is accepted as being the right thing in the circumstances or as making a relevant contribution. As indicated above, the distinction between leadership and other forms of power rests on the source of the influence – hence the insufficiency of a definition which describes leadership as an influence process. The basis of influence in force is removal of choice; in coercion it is the threat of deprivation and in manipulation it is deception. These are very different influence processes from those involved in leadership. Fay’s definition makes the relationship between leader and follower very clear. Whether or not leadership is exercised is determined by the acceptance of the influence attempt and by the basis of the acceptance. In summary, the concept of leadership I propose has two main elements. First, leadership comprises goal-relevant influence – that is, those acts which take a group or organisation closer to its goals. Second, the source of the influence is follower’s personal liking or identification with the leader, the leader’s goal relevant expertise or the perceived legitimacy of his or her authority.

From Leadership to Distributed Leadership According to Gronn (2000; 2008) the idea of distributed leadership was first discussed by Gibb in 1954. The many variations in the concept reflect variations in the parent concept of leadership itself. It can refer to distribution of formal leadership roles; to the distribution of leadership task performance among organisational members and to the distribution of influence. I shall discuss each of these in turn while recognising the overlap between them. When leadership is understood in terms of a formal role, the indicator of its distribution becomes the extent to which formal leadership roles and responsibilities are spread across members of the organisation. For example, in a recent study of distributed leadership in 52 schools in the north-east of the United States, Spillane et al. (2008) used percentage of staff with formal leadership roles as one measure of distributed leadership. On average, 30% of all school personnel reported spending a portion of their time in a formally designated leadership role, with over 25% reporting it to be in a full-time position. Previous research on a sample of English secondary schools also suggests that the majority of teachers either hold formal leadership roles or are nominated by colleagues as leaders (Reid et al., 2004). If these two samples are at all typical of urban schools, school leadership, in the sense of formal leadership roles at least, is widely distributed. Formal role designation is, however, only one indicator of distributed leadership. A second indicator of distribution involves measures of who performs those activities, functions or tasks that are involved in leading a group or organisation. In this case the research question is not “How are the formal roles and responsibilities distributed?” but “Who actually does the leadership work?”

224

V.M.J. Robinson

Spillane’s study included this second type of measure in deliberate recognition of the likely difference between formal organisation design and lived leadership realities (Spillane et al., 2008). One of the measures was derived from principals’ electronic recordings of their activities at randomly selected intervals over six school days. They were asked if they were leading the activity they were engaged in at the time and if so, whether they were leading alone or sharing the leadership with someone else. The electronic records showed that the most frequently reported coleader was a classroom teacher with no formal leadership role. While there are some limitations in this indicator – it is anchored in the work flow of the principal and thus does not capture those leadership tasks which did not involve the principal – it does suggest that a great deal of leadership in schools is done by those without a formal leadership designation. Under the third variation of the concept, leadership distribution is the distribution of task-relevant influence. The challenge for researchers under this concept is how to identify, those occasions when (a) people exercise influence and (b) the influence has its origins in one of the three sources of leadership influence and (c) the influence progresses the task of the group. The challenge of measuring leadership in a way that is true to this conception is considerable. That it is not insurmountable, however, is evident from classical empirical literature on the social psychology of leadership. Gibb (1969) concludes that there “is good evidence that members of a group can identify reliably those persons who exert most influence upon them and that leaders defined this way are closely correlated with leaders identified by external observers and by other criteria” (p. 211). Some of the features of this third concept are very different from those of the first two. Rather than assume that people in particular roles exercise influence, or that the performance of particular tasks involves, by definition, the exercise of influence, this concept requires a much more direct measure of actual influence. Followers themselves report who has influenced them or observers report on this after intensive observations of group activities. The question is not “Who has leadership roles?” or “Who performs functions that are assumed to be influential?” but “Who has actually had an impact on others?” The source of the relevant information is the followers, through their self reports or their observed reactions to other group members. In some studies, others’ influence has been assessed by asking teachers whom they turn to for advice or ideas (Friedkin and Slater, 1989) or whom they nominate as leaders (Leithwood et al., 2007; Spillane et al. 2008). In the study of Spillane et al. (2008), teachers were asked to name those whom they turned to for advice in reading and mathematics. Those who were nominated by at least three staff were judged to be leaders in the curriculum area. I should acknowledge at this point that seeking advice is not quite the same as taking it, and it is the latter that comes closer to the concept of leadership as consensually accepted task-related influence. It is not unreasonable to assume, however, that teachers learn from experience about whose advice is more likely to be helpful, and that those teachers from whom advice is sought are likely to be the more influential with respect to the relevant task. Once again, results from indicators based on teachers’ nominations of those from whom they seek advice and ideas show how very different patterns of distribution

11 An Educationally Relevant Account of Distributed Leadership

225

emerge when different indicators of distributed leadership are used. In the study of Spillane et al. (2008), only 45% of those nominated as sources of advice in math and 44% of those in reading had formal leadership roles. Even more significant was the finding that approximately two thirds of those with a specialist leadership role in reading were not nominated as a source of advice by at least three teachers. In math the equivalent figure was “over 50%” (Spillane et al., p. 208). These findings suggest the limitations of equating leadership with formal position and the importance of comparing evidence gained from multiple indicators of distributed leadership. In summary, there are three broad indicators of distributed leadership; those based on the distribution of formal leadership positions or role, on the distribution of so-called leadership tasks and those based, directly or indirectly, on the distribution of interpersonal influence. The first two types of measure assume rather than directly test the exercise of influence – it is assumed that the holders of certain positions or that those who exercise certain functions will turn out to be influential and that, therefore, it is appropriate to attribute leadership to them. The third measure makes fewer such assumptions because it seeks, through measures of follower reaction, to test the extent to which others have actually been influenced by position holders or by those performing tasks or functions which are assumed to be influential. The empirical evidence so far suggests that the distribution of positional leadership is not readily mapped onto the distribution of actual influence. If we are to trace the impact of leadership, whether distributed or centralised, over student outcomes, then we need to test rather than assume the link between leadership position, role or task and the exercise of actual influence.

Distributed Leadership – A Richer Story The concept of leadership developed earlier admits both centralised and distributed forms because it extends beyond positional authority. While positional authority may be available to a limited number of organisational members, the other two sources of leadership influence – personal qualities and relevant expertise – are open to any member of the organisation. Leaders emerge as a function of the task, the characteristics of other group members, the constraints imposed by formal authority relations and the inclination and skill of the group members. These contingencies mean that both leaders and followers have a transient status “. . . either leader or follower at any one time may, at another time, be in the other role. . .” (Gibb, 1969, p. 252). Rather than playing a subordinate role, followers are collaborators in the accomplishment of group tasks. In the remainder of this section I enrich this discussion of distributed leadership in two ways. First, I illustrate the multiple sources of leadership influence through an analysis of a hypothetical school meeting. Second, I explain how my account of distributed leadership, which has so far referred only to interpersonal influence processes, needs to be supplemented with an account of the role of impersonal influence processes.

226

V.M.J. Robinson

The following hypothetical scenario describes a meeting of science teachers concerned to review recent student assessments.1 Mary, the Head of Science, is chairing a meeting in which her staff are reviewing the results of the assessment of the last unit of work. She circulated the results in advance, with notes about how to interpret them, and asked the team to think about their implications for next year’s teaching of the unit. The team identifies common misunderstandings and agrees they need to develop resources which help students to overcome them. Julian, a second year teacher, was pretty unhappy with the assessment protocol used this year, and suggests revisions which he thinks will give more recognition to students who have made an extra effort. Most of his suggestions are adopted. Lee, who teaches information technology as well as science, shows the group how the results have been processed on the computer so that they can be combined with other assessments and used in reports to parents and the Board. Several team members express nervousness about reporting to the Board so they decide to review a draft report at the next meeting. There are several instances where the science teachers appear to have influenced each other. They follow Mary’s request for meeting preparation, and the meeting structure that she has prepared. Julian and Lee also change how the task is done through their ideas about how to improve the assessment and reporting procedures. Leadership is distributed in this meeting in the sense that it has emerged in the course of task performance from different participants, including those with no positional authority. If, however, we see the leadership influence of Mary, Lee and Julian as an entirely interpersonal process we miss the impersonal sources of influence that were also important in this meeting. People exercise influence indirectly by creating the conditions which enable group members to contribute to the achievement of shared purposes – a type of leadership that is often called empowerment (Fay, 1987). This involves designing routines, tools and task-related infrastructure that structures how the work is done. Recent concepts of distributed leadership, largely influenced by developments in cognitive science, recognise that it involves more than the exercise of interpersonal influence, however fluid and widely distributed (Hutchins, 1995). In the meeting of science teachers discussed earlier, Mary exercised influence by not only requesting staff preparation and chairing the meeting. In addition to these interpersonal influence processes, she structured the task through the provision of notes about the assessment results. Similarly, while Lee’s leadership involved some face-to-face persuasion, his interpersonal influence was greatly enhanced by his presentation of information and ideas via computer-generated tables and graphs. If those tools are shared with staff in other departments, and they change aspects of their assessment practice as a result, then Lee’s leadership of those staff would have required no face-to-face interaction with them at all.

1 The following section is based on Robinson, V.M.J. (2001). Embedding leadership in task performance. In K. Wong & C. Evers (Eds.), Leadership for quality schooling: International perspectives (pp. 90–102). London: Falmer Press.

11 An Educationally Relevant Account of Distributed Leadership

227

It is now clear why Spillane insists that leadership is more than the aggregation of the leadership acts of either individuals or interacting individuals (Spillane, 2006; Spillane et al., 2004). Such a focus is insufficient because it neglects the ways in which leader-follower interactions are structured by aspects of the situation, including the tools that structure the work and communicate task-relevant knowledge. For Spillane, leadership is inevitably distributed because it takes place in the interaction between leaders, followers and aspects of the situation including the tools that structure task performance. “In this scheme” he writes, “what is critical are the interdependencies among the constituting elements – leaders, followers and situation – of leadership activity” (Spillane et al., 2004, p. 16). For Gibb, Gronn and Spillane, therefore, distributed leadership is not a trendy new category of leadership, but a defining characteristic of leadership itself. The distribution involves more than shared interpersonal influence and the exchange of leader and follower roles. It also involves the recognition that particular acts of leadership are scaffolded by tools and artefacts that structure the way we think about and perform particular leadership tasks. This means that much of the knowledge that enables leadership work to be done is not represented in the heads of leaders but in the social and material artefacts (e.g., forms, templates, software and schedules) with which they interact (Hutchins, 1995). These artefacts are the products of prior problem-solving activities – and the quality of the solutions they represent will have a considerable influence on the effectiveness of any particular leadership attempt (Robinson, 2001). Empirical research on the role of tools in distributed leadership practice is in its infancy. In large organisations and systems, tools and their associated routines are crucial to goal achievement, since they free leadership influence from the space and time constraints associated with face-to-face engagement. The importance of tools in the leadership of the improvement of teaching and learning was a key finding of a recent synthesis of 17 evaluations of school improvement initiatives that had resulted in positive gains in student academic and social outcomes (Robinson and Timperley, 2007). A careful analysis of all the leadership practices reported in the evaluations resulted in the derivation of six leadership dimensions, one of which involved the design and use of “smart tools”. Robinson and Timperley (2007) describe smart tools as having two distinguishing features: they incorporate a valid theory of the task for which they were intended and they are well designed with respect to such qualities as addressing the limits of users’ working memory (Mayer and Moreno, 2003). Given that leadership invariably takes a distributed form, descriptive research involves studying how it is distributed in particular contexts and the antecedents and consequences of such distribution. Some important research questions include: To what extent does the pattern of distribution follow the contours of task-relevant expertise? What do teachers report about the sources of influence on selected aspects of their practice? To what extent do tools feature in those reports? Are the influential tools smart tools? Having reviewed the key elements of the concept of three different concepts of leadership and its distribution, I now turn to the discussion of distributed leadership

228

V.M.J. Robinson

and educational outcomes. What shifts, if any, in these concepts are needed in order to identify how distributed leadership is related to student outcomes?

From Distributed Leadership to Educational Outcomes Leadership is attributed to those who contribute in sufficient degree to the achievement of group goals. Gibb expresses the point as “a group member achieves the status of a group leader for the time being in proportion as he [sic] participates in group activities and demonstrates his capacity for contributing more than others to the group achievement of the group goal” (Gibb, 1969, p. 247). It is possible, therefore, that the particular behaviours involved in leadership will vary by goal type, and that, as a consequence, those group members who are identified as leaders in one task situation may not be so identified in another. The research evidence reviewed by Gibb (1969) suggests that this is the case – leadership is neither wholly generic nor entirely situation specific. If, as Gibb suggests, what counts as leadership is partly goal-dependent, then measures of leadership must not only reflect a defensible generic account of leadership, they must also be tailored to reflect existing theory and evidence about what is required to achieve the particular goals in question. In the case of schools, this means linking measures of leadership to our existing knowledge about the school and classroom conditions required to promote the social and academic well-being of students. How well suited are the three concepts and measures of distributed leadership discussed earlier to the task of investigating the links between leadership and student outcomes? What adaptations are needed to more precisely test the extent to which distributed leadership is creating those particular conditions that enhance student achievement and well-being? In the remainder of this section I address these two questions using the three previously discussed concepts of distributed leadership as an organising framework. I make detailed reference to existing empirical research on distributed leadership. In some cases these programmes have not yet tested the relationship to student outcomes but they are useful, nevertheless, because they provide valuable insights into the methodological and empirical implications of adopting particular concepts and measures of distributed leadership.

Distributed Leadership Positions and Educational Outcomes We know from the research of Reid and Spillane that formal leadership roles are probably widely distributed in schools. What we do not know is whether the degree and pattern of formal leadership distribution makes any difference to student outcomes. In the absence of direct evidence about this question, we can only speculate on the likely fruitfulness of using patterns of formal leadership distribution to test leadership-outcome relationships. It is likely that this approach will not yield

11 An Educationally Relevant Account of Distributed Leadership

229

clear findings because leadership structures, such as the proportion of staff that hold formal leadership positions, are poor predictors of actual leadership practice. The research of Spillane, for example, showed that many of those who held formal curriculum leadership roles were not reported by teachers to be a source of curriculum advice. Similarly, Timperley (2005) discovered, in an evaluation of a literacy leadership initiative in a small sample of New Zealand primary schools, that those who held the position of literacy leader demonstrated widely varying leadership practices and those variations were predictive of differential gains in student reading gains. She concluded that identical structures of leadership distribution can have very different consequences for students depending on the cultural norms of the group, and the knowledge and skills of those in the leadership roles. “Distributing leadership over more people is a risky business and may result in the greater distribution of incompetence. I suggest that increasing the distribution of leadership is only desirable if the quality of the leadership activities contributes to assisting teachers to provide more effective instruction to their students, and it is on these qualities that we should focus” (Timperley, 2005, p. 417). Given the uncertain and possibly weak relationship between the distribution of leadership positions, the work of leadership, and student outcomes, one could ask why a greater spread of leadership positions is considered desirable by many writers on distributed leadership. The basis for this normative claim is often, not the educational benefits of such structural arrangements, but a theory of power. Distributed leadership is seen as desirable because it counters a concentration of power and authority in the hands of the principal or senior management team. Goldstein, for example, in an empirical study of a mandated shift to peer-assisted summative teacher evaluation in a United States school district, uses the wider distribution of power as the normative justification for a more distributed approach to teacher evaluation (Goldstein, 2004). Her data showed that rather than take full responsibility for the summative evaluation, peer evaluators collaborated with principals in making the final judgment about the staff they had evaluated. She is critical of this failure to realise the potential of the evaluation policy for full transfer of power to the peer evaluators: While collaboration is a legitimate approach to leadership, the term itself is pregnant with ambiguity, and allowed for a drift away from teacher jurisdiction for teacher evaluation. However attractive the shared or collective model may be, institutional theory and prior research on teacher leadership policies suggest that the shared model may be just a stop on the way back to principal jurisdiction for teacher evaluation [in this school district]. This possibility highlights the ongoing challenge to distributing leadership in public education (Goldstein, 2004, p. 192).

Harris is also a strong advocate of the potential of distributed leadership for power equalisation and pays particular attention to the features of school culture and micro politics that may impede such shifts (Harris, 2005b). If distributed leadership is to be educationally credible, it needs a normative theory that is firmly grounded in its educational consequences for students, rather than in its consequences for staff relationships.

230

V.M.J. Robinson

In summary, the empirical work on the distribution of leadership positions has provided useful reminders of the disjunction between leadership structures and leadership work. Such disjunctions suggest that investigations of the links between leadership and student outcomes need to include measures of who does the goalrelevant leadership work and such measures can not be based on determining the occupiers of leadership positions. Distributed Leadership Tasks and Educational Outcomes. The second concept of distributed leadership is based, not on the distribution of leadership positions, but on the distribution of so-called leadership tasks and functions. In this section I discuss the implications of this approach for research on the links between distributed leadership and student outcomes, and examine, in particular, how researchers have justified their selection of leadership tasks and functions. Camburn, Rowan and Taylor have taken this second approach in their study of distributed leadership in United States schools participating in a comprehensive school reform program (Camburn et al., 2003). For these authors, leadership is “a set of organizational functions that leaders might be expected to perform – including not only instructional leadership functions, but also functions related to broader school and building management, as well as boundary-spanning functions entailing the acquisition of resources and the establishment or maintenance of relationships with external constituents” (Camburn et al., 2003, p.349). Camburn justifies this definition of leadership by claiming to follow “a long line of research and theory that conceptualizes leadership in terms of organizational functions and then examines who within an organization performs these functions” (Camburn et al., 2003, p. 349). A similar approach to the definition and measurement of distributed leadership is seen in some of the empirical research of Spillane and his colleagues. Spillane defines leadership as, “the activities engaged in by leaders, in interaction with others in particular contexts around specific tasks” (Spillane et al., 2004, p. 5). He then goes on to specify the tasks involved in school leadership as the “identification, acquisition, allocation, co-ordination, and use of the social, material, and cultural resources necessary to establish the conditions for the possibility of teaching and learning” (p. 11). Organisation theory does not provide a sufficient basis for the development of leadership indicators that measure the particular educational practices that are likely to be associated with improved student outcomes. The purpose of organisation theory is to identify functions associated with organisational survival, and the leadership tasks required for that purpose are different from those required for the much narrower purpose of achieving particular goals (improved student outcomes) in a particular type of organisation (schools). A more useful resource than generic organisational theories is the recent evidence that identifies the relative impact of particular leadership practices. Broadly speaking, the leadership tasks with the strongest impact on student outcomes are those involved in instructional leadership. A recent meta-analysis of published studies of the impact of leadership on student outcomes showed that the impact of instructional

11 An Educationally Relevant Account of Distributed Leadership

231

leadership was between two and three times greater than that of transformational leadership (Robinson et al., 2008). Within instructional leadership itself, the relative impact of five different sets of leadership practices was calculated. Small effects were found for establishing goals, strategic resourcing and establishing an orderly and supportive environment, moderate effects for planning, coordinating and evaluating teaching and the curriculum and large effects for promoting and participating in teacher learning and development (Robinson et al., 2008). Paradoxically, even though Camburn makes no reference to theory and research on instructional leadership, his leadership measures include twenty-one indicators, grouped into four scales, about detailed aspects of instructional leadership. Holders of leadership positions were asked to indicate the priority they gave to such activities as “clarifying expectations for students’ academic performance” and “promoting instructional coordination across grade levels in the school.” On other instructional leadership scales they were asked to indicate how often they did such things as “share information or advice about classroom practice with a teacher” and “demonstrate instructional practices and/or the use of curriculum materials in a classroom” (Camburn, et al., p. 368). While the purpose of Camburn’s study was to describe the distribution of instructional leadership rather than to test its impact, his indicators of instructional leadership would be well suited to such testing as they are closely aligned to the research evidence about the school and classroom conditions required to enhance student achievement and well being. As this knowledge base changes, leadership indicators should also change so that the measures are focussed on the practices that matter most. In short, I am advocating the development of concepts and indicators of distributed leadership which infuse a defensible account of leadership with outcomes-based educational content. The following example provides a concrete illustration of what I mean. We know that leadership through promoting and participating in teacher learning and professional development has a relatively large impact on student outcomes (Robinson et al., 2008). Yet we also know from a recent meta-analysis of the impact of professional learning opportunities on the students of participating teachers, that aspects of the context, content, learning activities and learning processes associated with these opportunities, make a discernible difference to their effectiveness for students (Timperley and Alton-Lee, 2008). With respect to context, for example, what is critical to effectiveness is not whether or not teachers volunteer for the learning opportunity, but whether they engage with the ideas at some point in the process. This evidence suggests, therefore, that we need leadership indicators that go beyond measurement of such things as the frequency of leaders’ involvement in teacher professional learning and development, to the measurement of such things as leaders’ knowledge of the conditions required for effective professional development and their success in ensuring that the development opportunities for which they have responsibility embody those qualities.

232

V.M.J. Robinson

The same point can be made with respect to leaders’ use of data. Studies of leadership impact on student outcomes frequently include indicators about how leaders promote, model or require the use of student data to improve teaching and learning (Heck, 2000; Heck et al., 1990, 1991). These indicators need to reflect recent outcomes-linked evidence about the particular types of use that are more or less likely to help teachers make a difference to the achievement of their students. In a recent study of distributed leadership in seven New Zealand primary schools participating in a literacy initiative, Timperley (2005) found that in the high gain schools, leaders used disaggregated data to help teachers make connections between how they taught reading and the achievement of their own students. In low gain schools, data were aggregated in ways that protected the privacy of each teacher’s class results. This evidence suggests that indicators that assess how leaders use data, rather than the amount or frequency of use, will be more productive in detecting impacts on student outcomes. The challenge is, therefore, to infuse leadership concepts and indicators with educationally relevant content. If the goal is to test the link between distributed leadership and student outcomes, then indicators of leadership must assess the extent to which leaders are creating the conditions that existing evidence suggests are required. The knowledge required to specify those conditions is located in educational research on effective schooling and teaching, not in generic leadership theory and research. We need the former literature, as has been argued in the first section of this paper, to develop a defensible concept of leadership, but beyond that point we need to infuse relevant educational content. In summary, I have argued that if leadership is taken as the performance of certain tasks and functions, then the associated indicators will be more strongly predictive of student outcomes if they assess the qualities associated with effective performance of those tasks. The same argument can be applied to the study of leadership exercised through the design and use of tools and their associated routines. If links are to be made with student outcomes, distributed leadership researchers must go beyond analyses of tools as a constituent of distributed leadership activity and ask normative questions about the adequacy of the tools which are shaping teaching and leadership practice. Leadership through the design and use of tools, like any form of leadership influence, may or may not contribute to improved student outcomes – hence my earlier discussion of the concept of a “smart tool” as one that has the qualities required to advance the purpose for which it is being employed. A defensible distinction between smart and dumb tools is needed if we are to learn more about how distributed leadership contributes to student achievement and well being. For example, there is some empirical evidence to support the argument that many of the tools used in teacher evaluation (e.g. evaluation policies and classroom observation checklists), lack validity in terms of the goal of improved teaching and learning, because they incorporate faulty theories of effective teaching (Davis et al., 2002; Ellett and Teddlie, 2003; Sinnema and Robinson, 2007). Thus, while the inclusion of tools and other aspects of the situation in the concept of distributed leadership is to be welcomed, a stronger normative framework for evaluating their educational merits is needed if analyses of distributed leadership

11 An Educationally Relevant Account of Distributed Leadership

233

through tool design, adaptation and use are to contribute to a greater understanding of its role in the improvement of teaching and learning. Distributed Leadership Influence and Educational Outcomes. In my earlier discussion of the concept of leadership, I argued that the exercise of particular sorts of influence was a defining feature of leadership. Strictly speaking then, the assessment of leadership should involve demonstrating that followers have been influenced by changing their ideas, intentions or actions. It is the exercise of influence that links leader and follower and that establishes the social nature of leadership acts. This means that the intended recipients of leadership influence, rather than the intended influencing agents should be the focus of leadership indicators. An examination of empirical research on educational leadership, however, shows that few studies have assessed leadership by establishing shifts in follower ideas, attitudes or actions. Instead, researchers have tended to rely on indicators which directly or indirectly establish the intention to influence others. Intentions may be measured indirectly by observing or reporting leaders’ engagement in practices which are intended to change the practices of followers. It is assumed that leaders’ intention to influence is signalled by their engagement in such practices as establishing goals or discussing data with staff. The link between intended and actual influence may be eventually tested, not in the context of any given influence attempt, but through the development of research generalisations about whether, on average, such leadership practices do make a difference to teacher practice and/or the outcomes of their students. A few studies of distributed leadership in education take a second more direct approach to the assessment of the intention to influence. Among his many different measures of leadership, Spillane has included principal self reports of their intentions to motivate, develop or increase the knowledge of others (Spillane et al., 2007a). These measures reflect his concern to capture the how as well as the what of leadership. For Spillane the how of leadership involves “mobilizing school personnel and clients to notice, face, and take on the tasks of changing instruction as well as harnessing and mobilizing the resources needed to support the transformation of teaching and learning (Spillane et al., 2004, p. 11–12)” (p. 11). In his study of 52 Chicago school principals he recorded the how of leadership by asking principals to record their primary intention by choosing from a list which included increasing knowledge, monitoring teaching and the curriculum, developing common goals, motivating or developing others or redesigning the teaching and learning. Some of these options capture the principal’s intention to exercise direct or indirect influence over staff (Spillane et al., 2007a). Spillane’s approach is worth considering in more detail as his indicators of leadership intent are grounded in a concept of leadership that, unlike my own, is based on intended rather than actual influence. For Spillane, “Leadership refers to activities tied to the core work of the organisation that are designed by organisational members to influence the motivation, knowledge, affect or practices of other organizational members, or that are understood by other organizational members as intended to influence their motivation, knowledge, affect or practices” (Spillane, 2006, p. 12–13). For Spillane it is the intention, or the perception of the

234

V.M.J. Robinson

intention to influence, rather than actual influence that is the defining characteristic of leadership. What are Spillane’s reasons for basing his account of leadership in design or intention rather than actual influence, and what are the consequences of this conceptualisation for studying the links between leadership and student outcomes? Spillane gives three reasons for using intended rather than actual influence. He claims that (a) assessment of actual influence will capture only a subset of total leadership acts; (b) even if staff ignore influence attempts they still perceive those attempts as designed to influence them, and (c) definitions of leadership that are tied to outcomes are circular. In reply to Spillane’s first argument, I argue that the problem of capturing a subset of leadership acts begs the question of how leadership should be defined – if intended but unsuccessful influence attempts are not to be included then a definition based on actual influence captures all acts of leadership rather than a subset. If leadership is tied to intentions to influence then Spillane is right – that is the issue to be decided. With regard to the second argument the fact that some staff may perceive influence attempts as leadership has to be taken into account but it is not definitive. If the intention to influence, rather than actual influence, is taken as characteristic of leadership, then, in theory, leadership is attributed equally to those who succeed and those who fail in their attempts. If leadership, as Gibb, Fay and many others have argued, is centrally concerned with the voluntary accord of authority by followers, then a definition based on the intentions of the influencing agent ignores the role of followership in the attribution of leadership. Followers become irrelevant to both the concept and assessment of leadership and no distinction is made, in the determination of leadership itself, between those whose influence attempts are accepted or rejected by followers. This is inconsistent with Spillane’s own account of distributed leadership as centrally concerned with the interactions between leaders and followers (Spillane, 2006, p. 4). If actual rather than intended influence is taken as one of the defining elements of leadership, then questions about the impact of leadership need to be framed carefully to avoid the circularity that Spillane warns about. The fact that leadership has an effect does not preclude important questions being asked about the nature and value of the effects. It is important to know, for example, whether those who lead teachers do so in ways that directly or indirectly produce better educational outcomes for students. Leadership influence over teachers, whether exercised by teacher leaders, school leaders, district or national policies or researchers themselves, may or may not result in educational benefit for students. It is important therefore, to study the content and effectiveness of leadership influence and this is not precluded by an account of leadership that makes some type of influence a defining characteristic of leadership itself. In the earlier discussion of the concept of leadership, I referred to the study by Friedkin and Slater (1989) which identified leaders by asking staff who they turned to for advice. This is an important study because the measure of leadership is derived from followers’ reaction to it (they seek advice), it investigates perhaps the most important source of leadership influence in schools (attributed expertise) and it links leadership to student outcomes. These authors assessed expertise as a

11 An Educationally Relevant Account of Distributed Leadership

235

source of leadership by asking staff in a sample of 20 California primary schools to nominate those persons in the school to whom they turn for advice on events or issues that arise in the school. As noted in my earlier discussion of this study, the fact that teachers turn to others for advice does not guarantee that they are influenced by them, but it is probably a reasonable substitute for a more direct measure of influence. The network analysis of teacher nominations distinguished between nominations of other teachers and nominations of the principal. The relationship between these nominations and schools’ average performance, over a four year period, on standardised tests of reading, language and mathematics was calculated. There was a strong association between the degree to which principals were central in teachers’ advice networks and school performance. In contrast, there was no independent association between a school’s performance and the extent to which teachers reported other teachers as sources of advice (Friedkin and Slater, 1989). The authors’ discussion of these findings provides some insights into why distributed leadership may require strong principalship rather than be a substitute for it. In schools where principals are at the centre of instructional advice networks there are more opportunities for the advice to be widely communicated, incorporated into relevant routines and followed up, than in schools where principals play a less central role in the advice network. In other words, principal centrality in an instructional advice network increases the chances that any distributed instructional leadership contributes to a coherent instructional programme. The link to better student outcomes is explained by the independent positive effect of instructional coherence on student outcomes (Newmann et al., 2001). It is important to acknowledge the considerable practical difficulties in employing indicators of leadership based on the reactions of followers. No doubt those difficulties are responsible for the use of indicators which assess the intention rather than the actuality of influence. One difficulty not mentioned so far is the need to specify the time period within which influence is demonstrated (Katz and Kahn, 1966). As Gronn explains, the effect of the influence may be immediately apparent or may not be felt until a considerable period of time has elapsed. “The absence of evidence of immediate causal effects at any point in time, therefore, should not be interpreted as absence of influence or leadership” (Gronn, 2000, p. 331).

Discussion The purpose of this chapter was to develop a concept of distributed leadership that is fit for the purpose of conducting research on the relationship between distributed leadership and a range of valued student outcomes. To be fit for this purpose, I argued that the concept and any associated indicators of leadership needed to incorporate a) a defensible concept of leadership, i.e. one which recognised it as a particular type of social influence process and b) focus on those particular leadership practices which were most likely to have an impact on student outcomes. On both

236

V.M.J. Robinson

counts there remains substantial work to be done. While there are several important analyses of distributed leadership already available, they have not been carried out with this purpose in mind. With respect to the first consideration – a defensible account of leadership – shifts are needed in the concept of leadership that has informed many empirical studies of distributed leadership in educational settings. Leadership is a particular type of influence process, and the most direct indicator of such influence is shifts in the ideas, attitudes or practices of followers. Despite this, very few studies use follower reactions as indicators of leadership. Rather than directly assess impact, researchers have either asked respondents about their intention to influence others (Spillane et al., 2007a) or, more commonly, asked teachers to report the extent to which their leaders engage in those “leadership” practices which are intended to change the way teachers do their work. The fact that leaders engage in such practices as modelling the analysis of student achievement data, or setting academic goals, does not ensure, however, that teachers are influenced as intended. I have argued that a concept of leadership based on the intention to be influential is problematic and does not substitute for indicators of actual influence. We need more quantitative research that identifies the sources of influence on teacher practice, for that is where leadership lies. Indicators that ask teachers whether they have noticed an intended influence attempt, and trace their reaction to it, would provide more direct evidence about the sources of both interpersonal and impersonal influence. Similarly, if leaders intend teachers to learn new practices, teachers could be assessed in term of their use of those practices and the influence processes that led them to do so. Some of the qualitative cases studies on distributed leadership offer more insights than the quantitative studies, into the influence processes that lie at the heart of leadership (Spillane and Diamond, 2007; Timperley, 2005). If distributed leadership is to serve students, then more research is needed on the conditions under which teachers, especially those without positional authority, succeed in influencing their colleagues in ways that benefit students. What shifts in school and teacher culture are needed to support the wider distribution of those leadership tasks that are critical to sustained improvement in learning and teaching (Harris, 2005b; Little, 1982)? How do those in senior leadership positions authorise and develop a more distributed leadership approach? What influence processes are involved in shifting a privatised teacher culture to one in which more teachers are more willing to take collective responsibility for the quality of teaching and learning experienced by all their students (Newmann, 1994)? What influence processes increase the willingness of teachers to influence one another, and what conditions encourage teachers to exercise such influence in the areas that matter for students? It is important to remember that two different influence targets are involved in the relationship between leadership and student outcomes. The first target is those other adults, usually teachers, who mediate leader-student influence. The second is the students – the intended beneficiaries of leader-teacher influence. This brings us to the second of our two conditions for establishing the relationship between leadership and student outcomes – namely showing that the content or direction of

11 An Educationally Relevant Account of Distributed Leadership

237

influence on teachers, does in fact deliver benefit for students. Researchers need assessments of leadership that target those practices that have been shown to have the greatest magnitude of impact on student outcomes. Recent research on effective teaching and teacher learning has provided detailed evidence about the particular qualities of teaching, teacher evaluation and teacher learning that are associated with positive impacts on students (Brophy, 2000; Timperley and Alton-Lee, 2008). The probability of finding significant leadership impacts is greatly increased by basing indicators on such outcomes-linked evidence. In summary, we need a concept and indicators of distributed leadership that recognise that leadership is the exercise of task-relevant influence. If the task is the improvement of student achievement and well-being, then the relevant leadership practices are those that create the conditions that enable and require teachers to improve those outcomes. The knowledge base for identifying those conditions is not found in the literature on leadership, or in organisational theories. Rather, it is found in recent research on effective teaching and teacher learning. Hence there is a need for a far closer alignment of research on leadership with outcomes-linked evidence about teaching and learning. Several shifts are needed in the field of distributed leadership before this alignment will be achieved. One is resolving the confusion between distributed leadership as a descriptive and a normative concept. It is descriptive in the sense that, as I have already discussed, leadership is inevitably distributed across fluid and taskcontingent configurations of leaders, followers, and aspects of the situation. It is also a normative concept in the sense that one can argue that distributed leadership is a desirable form of organisational leadership. Of course any such normative claim immediately invites questions about why it is seen as desirable. Two types of normative argument are found in current writing on distributed leadership. Distributed leadership is considered desirable by some authors because of its potential educational benefits. The conceptual and methodological shifts that are required to research and develop this potential have been extensively discussed in this chapter. The second normative theory that is employed in research on distributed leadership is a particular theory of power. Distributed leadership is considered desirable by some authors because it is considered more democratic, or less managerial than more hierarchical structures of school leadership (Hatcher, 2005; Urbanski and Nickolaou, 1997). The confusion comes when it is assumed that these two normative theories are interchangeable. They are not. Any benefits accrued to teachers from more distributed patterns of leadership do not necessarily flow though to students. If it is shown that greater distribution of leadership brings a more collegial and satisfied staff culture, then it can not be assumed that such cultures bring improvement on valued student outcomes. Whether or not particular forms of leadership distribution promote such outcomes is an open empirical question to be addressed through both context-specific inquiry and research generalisation. The relationship between distributed leadership and educational outcomes must be directly tested using concepts and indicators that are fit for purpose. Shifts in the power relationships between the adults are orthogonal to questions about educational impacts on students.

238

V.M.J. Robinson

The research on the impact of transformational leadership on teachers and students provides a good starting point for inquiry into these issues as there is some overlap between the concept of distributed leadership and transformational leadership. In the two main research programmes on transformational leadership in education, teachers are asked, among other questions, to report on the extent to which their leaders build cultures and structures to foster collaboration. Some of the survey items used to assess this aspect of transformational leadership include “delegates leadership for activities critical to achieving goals” and “ensures we have adequate involvement in decision making” (Leithwood and Jantzi, 1999, 2000; Mulford et al., 2004). The outcomes-linked evidence that is available suggests that while transformational leadership has a moderate impact on teacher attitudes and satisfaction, its impact on students’ affective and achievement outcomes is very small. A recent meta-analysis of five studies which included measures of both transformational leadership and student outcomes showed that the average effect of transformational leadership on student outcomes is less than 0.2 – a result that is usually interpreted as indicating a very small effect (Robinson et al., 2008). One of these five studies also included measures of both principals’ and teachers’ leadership, enabling comparison of the two types of leadership (Leithwood and Jantzi, 2000). With student background factors controlled, both forms of leadership had very small impacts on student outcomes and the effect of teacher leadership was half that of principals. While such limited evidence does not justify a rejection of the possibility that distributed leadership delivers educational benefits, it does suggest that the link between the two is by no means assured. The question to ask is not “Does distributed leadership make a difference to student outcomes?” but “What are the relative impacts of particular types of distributed leadership practice?” Recent empirical research on distributed leadership is moving in this direction, as it has employed indicators which identify who engages in particular types of leadership practice. It is critical however, that such indicators discriminate the particular qualities that recent evidence on effective teaching and teacher learning suggest are more and less effective in improving valued student outcomes. Without concepts and indicators of distributed leadership that are infused with such educational content we will learn more about who does what in schools and very little about the difference it makes to the achievement and well being of students.

References Bass, B. M. (1985). Leadership and performance beyond expectations. New York, NY: Free Press Inc. Brophy, J. (Ed.). (2000). Advances in research on teaching: Subject-specific instructional methods and activities (Vol. 8). Oxford: JAI Press, Elsevier Science. Burns, J. M. (1978). Leadership. New York: Harper & Row. Camburn, E., Rowan, B., & Taylor, J. E. (2003). Distributed leadership in schools: The case of elementary schools adopting comprehensive school reform models. Educational Evaluation and Policy Analysis, 25, 347–373.

11 An Educationally Relevant Account of Distributed Leadership

239

Davis, D. R., Ellett, C. D., & Annunziata, J. (2002). Teacher evaluation, leadership and learning organizations. Journal of Personnel Evaluation in Education, 16(4), 287–301. Ellett, C. D., & Teddlie, C. (2003). Teacher evaluation, teacher effectiveness and school effectiveness. Journal of Personnel Evaluation in Education, 17(1), 101–128. Elmore, R. F. (2004). School reform from the inside out: Policy, practice, and performance. Cambridge, Mass.: Harvard Education Press. Fay, B. (1987). Critical social science: Liberation and its limits. Cambridge: Polity. French, J. R. P., & Raven, B. H. (1960). The bases of social power. In D. Cartwright & A. Zander (Eds.), Group dynamics: research and theory (2nd ed., pp. 607–623). London: Tavistock. Friedkin, N. E., & Slater, M. R. (1989). School leadership and performance: A social network approach. Sociology of Education, 67(2), 139–157. Fullan, M. (1993). Change forces: Probing the depths of educational reform. Bristol, Pa.: Falmer Press. Gibb, C. A. (1954). Leadership. In G. Lindzey (Ed.), Handbook of social psychology (Vol. 2, pp. 877–917). Reading, MA: Addison-Wesley. Gibb, C. A. (1969). Leadership. In G. Lindzey & A. Aronson (Eds.), The handbook of social psychology (2nd ed., Vol. 4, pp. 205–282). Reading, MA: Addison-Wesley. Goldstein, J. (2004). Making sense of distributed leadership: The case of peer assistance and review. Educational Evaluation & Policy Analysis, 26(4), 397–421. Gronn, P. (2000). Distributed properties: A new architecture for leadership. Educational Management and Administration, 28(3), 317–338. Gronn, P. (2004). The future of distributed leadership Journal of Educational Administration, 46(2), 141–158. Harris, A. (2005a). Leading or misleading? Distributed leadership and school improvement. Journal of Curriculum Studies, 37(3), 255–265. Harris, A. (2005b). Teacher leadership: More than just a feel-good factor? Leadership and Policy in Schools, 4(3), 201–219. Harris, A. (2008). Distributed leadership: According to the evidence. Journal of Educational Administration, 46(2), 172–188. Hatcher, R. (2005). The distribution of leadership and power in schools. British Journal of Sociology of Education, 26(2), 253–267. Heck, R. H. (2000). Examining the impact of school quality on school outcomes and improvement: A value-added approach. Educational Administration Quarterly, 36(4), 513–552. Heck, R. H., Larsen, T. J., & Marcoulides, G. A. (1990). Instructional leadership and school achievement: Validation of a causal model. Educational Administration Quarterly, 26(2), 94– 125. Heck, R. H., Marcoulides, G. A., & Lang, P. (1991). Principal instructional leadership and school achievement: The application of discriminant techniques. School Effectiveness and School Improvement, 2(2), 115–135. Hutchins, E. (1995). Cognition in the wild. Cambridge, MA: MIT Press. Katz, D., & Kahn, R. L. (1966). The social psychology of organizations. New York: John Wiley Leithwood, K., Harris, A., & Hopkins, D. (2008). Seven strong claims about successful school leadership. School Leadership & Management, 28(1), 27–42. Leithwood, K., & Jantzi, D. (1999). Transformational school leadership effects: A replication. School Effectiveness and School Improvement, 10(4), 451–479. Leithwood, K., & Jantzi, D. (2000). Principal and teacher leadership effects: A replication. School Leadership and Management, 20(4), 415–434. Leithwood, K., Mascall, B., Strauss, T., Sacks, R., Memon, N., & Yashkina, A. (2007). Distributing leadership to make schools smarter: Taking the ego out of the system. Leadership and Policy in Schools, 6(1), 37–67. Little, J. W. (1982). Norms of collegiality and experimentation: Workplace conditions of school success. American Educational Research Journal, 19(3), 325–340. Marzano, R. J., Waters, T., & McNulty, B. (2005). School leadership that works: From research to results. Auroroa, CO: ASCD and McREL.

240

V.M.J. Robinson

Mayer, R., & Moreno, R. (2003). Nine ways to reduce cognitive load in multimedia learning. Educational Psychologist, 38(1), 43–52. Mulford, B. (2008). The leadership challenge: Improving learning in schools. Melbourne: Australian Council for Educational Research. Mulford, W., Silins, H., & Leithwood, K. (2004). Educational leadership for organisational learning and improved student outcomes. Dordrecht: Kluwer Academic. Newmann, F. (1994). School-wide professional community: Issues in restructuring schools (Issue Report No. 6). Madison,WI: Center on Organisation and Restructuring of Schools, University of Wisconsin. Newmann, F. M., Smith, B., Allensworth, E., & Bryk, A. S. (2001). Instructional program coherence: What it is and why it should guide school improvement policy. Educational Evaluation and Policy Analysis, 23(4), 297–321. Reid, I., Brain., K., & Boyes, L.C. (2004). Teachers or learning leaders?: Where have all the teachers gone? Gone to be leaders, everyone. Educational Studies, 30(3), 251–264. Robinson, V. M. J. (2001). Embedding leadership in task performance. In K. Wong & C. Evers (Eds.), Leadership for quality schooling: International perspectives (pp. 90–102). London: Falmer Press. Robinson, V. M. J., Lloyd, C., & Rowe, K. J. (2008). The impact of leadership on student outcomes: An analysis of the differential effects of leadership type. Educational Administration Quarterly, 44(5), 635–674. Robinson, V. M. J., & Timperley, H. (2007). The leadership of the improvement of teaching and learning: Lessons from initiatives with positive outcomes for students. Australian Journal of Education, 51(3), 247–262. Sinnema, C., & Robinson, V. M. J. (2007). The leadership of teaching and learning: Implications for teacher evaluation. Leadership and Policy in Schools, 6(4), 319–343. Spillane, J. P. (2006). Distributed leadership. San Francisco, CA: Jossey-Bass. Spillane, J. P., Camburn, E. M., & Pareja, A. S. (2007a). Taking a distributed perspective to the school principal’s workday. Leadership and Policy in Schools, 6(1), 103–125. Spillane, J. P., Camburn, E. M., Pustejovsky, J., Pareja, A. S., & Lewis, G. (2008). Taking a distributed perspective: Epistemological and methodological trade-offs in operationalizing the leader-plus aspect. Journal of Educational Administration, 46(2), 189–213. Spillane, J. P., & Diamond, J. B. (Eds.). (2007b). Distributed leadership in practice. New York: Teachers College Press. Spillane, J. P., Halverson, R., & Diamond, J. B. (2004). Towards a theory of leadership practice: A distributed perspective. Journal of Curriculum Studies, 36, 3–34. Timperley, H. (2005). Distributed leadership: Developing theory from practice. Journal of Curriculum Studies, 37(4), 395–420. Timperley, H., & Alton-Lee, A. (2008). Reframing teacher professional learning: An alternative policy approach to strengthening valued outcomes for diverse learners. In G. Kelly, A. Luke & J. Green (Eds.), Review of Research in Education (Vol. 32). Washington, D.C.: Sage. Urbanski, A., & Nickolaou, M. B. (1997). Reflections on teachers as leaders. Educational Policy 11(2), 243–254.

Chapter 12

Coda Alma Harris

The hope of transforming schools or school systems through the personality or actions of individual leaders is quickly fading. Strong leaders with exceptional vision do exist but unfortunately they do not come in sufficient numbers to meet the demands and challenges of today’s schools. An alternative conceptualization of leadership is one that is distributed and premised on the quality of shared activity and interaction. There is increasing research evidence which suggests a positive relationship between distributed leadership and organizational change, as many of the chapters in this book have shown. The challenge is to build on this empirical platform and extend the knowledge base even further. We need to know much more about the barriers, unintended consequences and limitations of distributed leadership before offering any advice, giving prescription or advocating theories of action. We need to know the limitations and pitfalls of distributed leadership as well as the potential and possibilities of this form or forms of leadership practice. If, as Leithwood and colleagues suggest, “planful alignment” is the most potentially useful form of distributed leadership practice, how does this occur and exactly what difference does this particular configuration or pattern of distribution make to learning? We also need to know more about the nature of trust relationships that hold certain forms distributed leadership together. Is trust simply a by-product of distributed leadership or an essential pre-requisite? The chapter by Seashore Louis and her colleagues suggest that sense-making and trust are pivotal components within the amalgam of distributed leadership practices. Most importantly we need to know the factors that influence the nature and extent of distributed leadership in schools, as well as the consequences of different distributed patterns of leadership for schools and students. Some have argued that measuring the impact of distributed leadership is not possible, that the methodological challenges are too great and the empirical rewards too small. However there are now ways of operationalising distributed leadership that will allow us, at least, to know what we are measuring (Spillane et al in this book). Also there are sophisticated forms of measurement that can be used to explore the issue of

A. Harris (B) Institute of Education, 20 Bedford Way, London, WCIH OAL

A. Harris (ed.), Distributed Leadership, Studies in Educational Leadership 7, C Springer Science+Business Media B.V. 2009 DOI 10.1007/978-1-4020-9737-9 13, 

241

242

A. Harris

impact and effect. The chapters in this book offer insightful and important empirical findings about the potential influence and impact of distributed leadership. They also provide rich methodologies that can assist and inform the design of future research studies. There is evidence to suggest a correlation between multiple leadership functions and school improvement (Day et al., 2007 & 2009) The dynamic model outlined by Hallinger and Heck (in this book) begins to validate the viability of a set of key leadership processes that can be linked to school improvement. More specifically, the findings support the active building of professional and leadership capacity in schools. Their study adds to a growing body of empirical research that finds positive effects of collaborative or distributed leadership on school improvement processes and outcomes. Mayrowetz et al. (2009) note that, it is important for researchers to continue to investigate the linkage between multiple leadership functions and school improvement. They urge researchers to provide further explanation as to why distributed leadership can end in positive results for schools and students (p191). The evidence we have, to date, suggests that distributed leadership is undoubtedly worth further investigation and scrutiny. So too, is some consideration of the interplay between the formal and informal leadership structures and processes. As Hallinger and Heck (in this book) acknowledge, the nature of the relationship between principal and teacher leadership needs to be investigated further. More needs to be understood about the formal and informal leadership interdependencies and interconnections that make school improvement more possible or more fragile. Evidence suggests that principal leadership remains a key success factor in school improvement, especially in contexts where the challenges are greatest. It also suggests that purposeful distributed leadership is a common denominator of schools that are effective and sustain improvement (Day et al., 2009). It is clear that the task of building professional capacity and distributed leadership requires principal support and the leadership of principals and teachers in securing and sustaining school improvement (Day et al., 2009). Greater distribution of leadership outside the formal structure requires intervention on the “part of those in formal leadership roles” (Leithwood et al., 2009a:279). In short, distributed leadership necessitates some external orchestration and intervention. A few years ago it would have been unwise to attempt research that focused exclusively and directly on the impact or outcomes of distributed leadership (Leithwood et al., 2009b:281). The definitional, operational and empirical platform was simply not robust enough. Now there is a clearer conceptualization and operationalisation of distributed leadership practice. Empirical work has also started to map its direction and influence. There is now more research activity addressing the issue of leadership outcomes and effects. For policy makers, practitioners and researchers, anxious for research findings about the impact of distributed leadership (or lack of it) in relation to student learning, the evidence is long overdue. However the considerable theoretical, conceptual and empirical progress made by the field has ensured that impact studies are not only desirable but also possible.

12 Coda

243

References Day, C. Leithwood, K. Sammons, P. Harris, A. and Hopkins, D. (2007) Leadership and Student Outcomes, London, DCSF Interim Report. Day, C. Leithwood, K. Sammons, P. Harris, A. and Hopkins, D. (2009) Leadership and Student Outcomes, London, DCSF Final Report. Leithwood, K. Mascall, B. and Strauss, T. (2009a) Distributed Leadership According to the Evidence, London, Routledge. Leithwood, K. Mascall, B. Strauss, T. Sacks, R. Memon, N. and Yashkina, A. (2009b) Distributing Leadership to Make Schools Smarter: Taking Ego out of the System in Leithwood, K. Mascall, B. and Strauss, T. Distributed Leadership According to the Evidence, London, Routledge. Mayrowetz, D. Murphy, J. Seashore Louis, K. and Smylie, M. (2009) Conceptualizing Distributed Leadership as School Reform in Leithwood, K. Mascall, B. and Strauss, T. (2009) Distributed Leadership According to the Evidence, London, Routledge.