Literacy Development and Enhancement Across Orthographies and Cultures
LITERACY STUDIES
VOLUME 2
Series Editor: R. Malatesha Joshi, Texas A&M University,USA Editorial Board: Linnea Ehri, CUNY Graduate School, USA George Hynd, Purdue University, USA Richard Olson, University of Colorado, USA Pieter Reitsma, Vrije University Amsterdam, the Netherlands Rebecca Treiman, Washington University in St. Louis, USA Usha Goswami, University of Cambridge, UK Jane Oakhill, University of Sussex, Brighton, JK Philip Seymour, University of Dundee, UK Guinevere Eden, Georgetown University Medical Center, USA Catherine McBride Chang, Chinese University of Hong Kong, China While language defines humanity, literacy defines civilization. Understandably, illiteracy or difficulties in acquiring literacy skills have become a major concern of our technological society. A conservative estimate of the prevalence of literacy problems would put the figure at more than a billion people in the world. Because of the seriousness of the problem, research in literacy acquisition and its breakdown is pursued with enormous vigor and persistence by experts from diverse backgrounds such as cognitive psychology, neuroscience, linguistics and education. This, of course, has resulted in a plethora of data, and consequently it has become difficult to integrate this abundance of information into a coherent body because of the artificial barriers that exist among different professional specialties. The purpose of the proposed series is to bring together the available research studies into a coherent body of knowledge. Publications in this series are intended for use by educators, clinicians and research scientists in the above-mentioned specialties. Some of the titles suitable for the Series are: fMRI, brain imaging techniques and reading skills, orthography and literacy; and research based techniques for improving decoding, vocabulary, spelling, and comprehension skills. A complete list of titles published in this series can be viewed by going to the following URL: http://www.springer.com/series/7206
Dorit Aram Ofra Korat ●
Editors
Literacy Development and Enhancement Across Orthographies and Cultures
Editors
Dorit Aram Constantiner School of Education Tel Aviv University Tel Aviv Israel
[email protected]
Ofra Korat School of Education Bar-Ilan University Ramat Gan Israel
[email protected]
ISBN 978-1-4419-0833-9 e-ISBN 978-1-4419-0834-6 DOI 10.1007/978-1-4419-0834-6 Springer New York Dordrecht Heidelberg London Library of Congress Control Number: 2009940320 © Springer Science+Business Media, LLC 2010 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)
Dedication
This book is dedicated to Professor Iris Levin on her retirement. An internationally known and influential researcher, Professor Levin is a developmental psychologist at the School of Education, Tel Aviv University, Israel. She has studied early literacy for over 25 years and has published many studies in leading journals. She is one of the leaders in early literacy policy in Israel. Before beginning her work on literacy, Dr. Levin’s work focused on children’s growing physical concepts, particularly the concepts of time and speed, within NeoPiagetian and Information Processing frameworks. In this domain, she published leading articles and edited a book with contributions by prominent researchers in cognitive development, collaborating with Robert Siegler and Friedrich Wilkening. Later, she turned to the development of early literacy and sociocultural factors that shape and enhance children’s progress in this domain. Dr. Levin focused on emergent writing and how writing initially grows from drawing. Her collaboration with Adriana Bus enriched this work. Levin also demonstrated the major role of letter knowledge in bridging oral and written words, and how it fuels early writing and initial steps of reading. This work grew in collaboration with Linnea Ehri. The study of early writing expanded to include analysis of parental mediation of writing to kindergartners and young elementary schoolchildren, and of ways to improve parental mediation. She demonstrated the significance of parents’ mediation of writing in promoting their children’s literacy, and prescribed ways to improve this mediation. Storybook reading, another common parent–child joint activity believed to be central in promoting language and literacy, was compared to joint writing, showing differential effects on children’s development. This work grew in collaboration with Dorit Aram and Ofra Korat in Israel, and with Catherine McBride Chang in China and Liliana Tolchinsky in Spain, adding a cross-cultural aspect to this line of research. As an active participant in the Price Brody Initiative in Jaffa (a poor Jewish– Arab city), she initiated, developed, and implemented literacy enhancement projects. Within this initiative, she led a line of pioneering literacy interventions within the Jewish and the Arab communities. This work grew in collaboration with Dorit Aram, Elinor Saiegh Haddad, and Margalit Ziv. Iris Levin served on several national committees dealing with early education at large and with teaching of reading in Israel. She chaired the Levin Committee, v
vi
Dedication
which developed a program for 3- to 6-year-olds in literacy and language. The Israeli Ministry of Education endorsed this program in 2007 as a national obligatory curriculum for preschoolers and kindergartners. In addition to her academic endeavors, Iris Levin has volunteered to assist the Clinical Unit of the School of Law at Tel Aviv University on issues of human rights related to education. Her work has focused on equal opportunity for education for the impoverished Bedouin population and on the humane portrayal of the gay– lesbian community in educational television targeting youth. Her partner in this work was the social activist and lawyer, Dori Spivak. Not only an outstanding scientist who has maintained the highest standards for research, Iris Levin is also an excellent and dedicated teacher who has nurtured many students, teachers, and researchers. She is a model to many as a researcher and as a human being. As her former students and her current colleagues, we are honored to edit this book and to dedicate it to her.
About the Editors
Dr. Dorit Aram is an Associate Professor and head of the Special Education Program in Tel Aviv University’s School of Education, Israel. Her research focuses on parent–child literacy interactions and their implications for early literacy and socioemotional development, in Hebrew and Arab communities. Addressing needs in special populations, her work examines these interactions in parent–child dyads of children with hearing loss, ADHD, and low socioeconomic background, and among precocious readers. Dr. Aram has conducted early interventions aimed at improving kindergarten and preschool teachers’ as well as parents’ mediation and children’s early literacy. Dr. Aram has been a board member of the Israel Literacy and Language Society from its inception and is now its president (in collaboration with Dr. Ofra Korat). Dr. Ofra Korat is an Associate Professor and the head of Early Childhood Program in the School of Education at Bar-Ilan University, Israel. Her research focuses on early literacy experience in school and family, with special attention to the role of parental beliefs on literacy development and education and parent–child interactions in facilitating literacy development. She implements this line of studies among diverse cultures in Israel, including Ethiopian and Arab families. She conducts intervention studies to facilitate literacy development among children at risk in language and literacy development. Her recent scholarly interests focus on the relationship between technology and literacy. Dr. Korat has been a board member of the Israel Literacy and Language Society from its inception and is now its president (in collaboration with Dr. Dorit Aram).
vii
Preface
It has become increasingly clear that a country’s prosperity and cultural progress largely depend on its level of literacy. Consequently, literacy enhancement has become a public issue that interests readers across the world. Integrating innovative insights from different perspectives, this book promotes cross-fertilization of ideas and practices and explores how we might provide children with stable literacy foundations. For many years, the study of literacy development and enhancement was generally based on English language learners. This picture has changed dramatically in the last two decades, and today we benefit from the intensive work of researchers from different countries and languages. Worldwide research increasingly highlights the role of early literacy in young children’s development and the increasing value of literacy in general development. Moreover, massive immigration around the world has increased the significance of studying bilingual literacy, acquisition of a second language, and foreign language learning. Researchers from different countries and cultures have studied the relationships between their languages’ orthographies and the acquisition of literacy. In the 16 chapters included in this book, we bring together studies and thoughts of researchers from eight countries across the world, focusing on literacy acquisition and advancement in Cantonese, Catalan, Chinese, Dutch, English, French, Hebrew, Kannada, Mandarin, Portuguese, and Spanish. The studies presented in this book employ a variety of quantitative and qualitative methodologies. Some of these are naturalistic, while others are experimental. All of them aim to present a reliable picture of the state of the art in the arena of literacy development and enhancement. The chapters in this book are grouped into four parts: (1) Literacy Development, (2) Promotion of Literacy at Home and in School, (3) Bilingualism – Acquiring a Second Spoken and Written Language, and (4) Beyond Literacy. Part 1 consists of five chapters focusing on Literacy Development. It includes analyses of acquisition of phonological awareness, print awareness, letter knowledge, spelling, and writing. The studies in this section refer to the significance of these skills in different orthographies. In Chap. 1, Anna Both-de Vries and Adriana Bus explore the role of name writing in children’s acquisition of alphabetic knowledge. The chapter presents a series of studies on early spelling in Dutch. The researchers assert that the first letter in their own name is the first letter that children ix
x
Preface
write phonetically. They conclude that name writing functions as the stepping-stone to the alphabetic strategy. Chapter 2 by Sarah Robins and Rebecca Trieman also refers to the significance of children’s own names in informal early literacy experiences among English-speaking children. They investigate the informal learning of surface features of writing (what letters and words look like) and deeper features of written language (the fact that print symbolizes spoken language). They conclude that informal learning about the surface and deep features of writing begins at an early age, and it can help prepare children for the formal reading and writing instruction at school. Chapter 3 by Cláudia Cardoso-Martins and Marcela Fulanete Corrêa centers on the development of spelling skills among Brazilian Portuguesespeaking children. It explores the similarities in which English-, Hebrew-, and Portuguese-speaking children take advantage of their letter name knowledge to connect print to speech. The researchers conclude that Portuguese-speaking children rely on their letter names knowledge to spell words, and in this they resemble Hebrew- and English-speaking children. Chapter 4 by Mary Ann Evans and Jean Saint-Aubin focuses on the development of print awareness. Using eye tracking methods, the researchers examine where children are looking during shared reading. The results indicate that, during shared book reading, children engage in minimal exploration of the print, and that their parents rarely draw their attention to it. The researchers suggest that print awareness may be developed by activities in which adults teach children to print letters and words and to learn letters and sounds as they take part in shared book reading. Although Hebrew letter names have been claimed to be less useful in supporting children’s literacy acquisition, in Chap. 5, Iris Levin presents two studies that assessed the impact of the nature of Hebrew letter names on children’s letter naming, letter writing, and first letter isolation. Hebrew letter names (e.g., gimel for /g/) are longer than English letter names (di for /d/), and, unlike English, they incorporate sounds that are not relevant to the focal letter sound. Despite this, Levin found that they are highly effective cues in helping children to report, spell, and sound out letters within a word. The author concludes that letter names in Hebrew are more effective than in English. The fact that Hebrew letter names are relatively long does not limit their usefulness in promoting children’s early literacy acquisition. Levin suggests that these results can be explained by the cacophonic nature of Hebrew letter names and the greater length of full name cues in Hebrew. Part 2, Promotion of Literacy at Home and in School, examines the sociocultural context of literacy acquisition. This section includes studies of storybook reading, parental beliefs about literacy practices, the influence of schooling practices, and the efficiency of literacy interventions. First, in Chap. 6, Susan Sonnenschein, Linda Baker, and Robert Serpell describe a 5-year longitudinal investigation of children’s literacy development in a sociocultural context. They provide the readers with an overview of cultural and environmental factors that are central to children’s literacy development at home. This chapter studies parental beliefs, focusing on low- and middle-income families. The authors conclude that parental beliefs and children’s home experiences make a significant difference in children’s literacy development. Although low- and middle-income children had different home
Preface
xi
literacy experiences, the intimate culture of the home was a more powerful predictor of literacy development than demographic factors such as family income and ethnicity. In Chap. 7, Catherine McBride-Chang, Yvonne Chow, and Xiuli Tong focus within the home environment context on parent–child reading and writing activities. While Chap. 6 studies variations of home contexts in different socioeconomic status (SES) groups within one language, Chap. 7 provides data regarding home literacy practice in different orthographies and cultures. The authors define two major facets affecting children’s literacy development. One is implicit and includes variables such as parents’ educational levels or beliefs regarding literacy. The other is explicit and comprises parents’ scaffolding of children’s writing and shared book reading. The authors conclude that there are distinctive features in parent–child reading and writing activities that have a clear impact on children’s language and literacy across cultures. Within the arena of literacy activities in Chap. 8, Monique Sénéchal centers on storybook reading. She asks what reading books to young children does and does not do in children’s language and literacy development. The chapter presents an overview of six studies. These studies show that the number of times, as well as the manner in which the adult reads to children, affects children’s acquisition of comprehension and spoken vocabulary. The studies also provide evidence that children’s vocabulary is a strong predictor of reading comprehension. To enhance language and literacy of preschoolers, Chap. 9 calls upon us to build a bridge between home and school. Linda Phillips presents a critical view about the ways in which researchers and educators sometimes perceive low SES parents’ attitudes regarding literacy education. In her chapter, she describes a line of studies demonstrating the disconnection between teachers and parents. These studies show that parents from low SES do care deeply about their children’s literacy development, and that educators occasionally overlook these parents’ beliefs. She declares that it is critical to forge informed collaboration between homes and schools for the advancement of children from low SES background. The author presents a 5-year longitudinal study of a project in the community. This project demonstrates effective ways of bridging between parents and teachers to the benefit of children. The last two chapters in this part relate to schoolchildren. Linnea Ehri and Julie Rosenthal in Chap. 10 provide an overview of studies of vocabulary acquisition, which plays a central role in reading ability and academic achievements. Vocabulary is traditionally studied in the context of oral language. Ehri and Rosenthal describe studies that have assessed the contribution that spelling makes to vocabulary learning in English. They conclude that the field of vocabulary learning and instruction has neglected an important skill that facilitates vocabulary growth – spelling. Concluding this part in Chap. 11, Frederick Morrison, Carol MacDonald Connor, and Annemarie Hindman present an innovative approach toward effective reading instruction. This approach acknowledges the early individual differences among children in cognitive, language, literacy, and social skills and prescribes effective instruction that takes into account each child’s characteristics. In accordance with their theories, the researchers developed an intervention program that assessed the children’s literacy. Based on the child’s scores, a computer program specified the amounts and types of instruction
xii
Preface
(i.e., teacher/child management and code/meaning focus) that a child should receive in order to reach his/her optimal level. The researchers report that children in the intervention group showed greater growth in reading skills than did children in the control group. In addition, the closer the instruction received by the child matched the duration and type of activity that the software recommended, the more children’s reading skill grew over the year. The four chapters of Part 3, Bilingualism – Acquiring a Second Spoken and Written Language, illuminate how different languages and orthographic systems affect children’s literacy acquisition. Literacy acquisition in different languages puts distinct cognitive and metalinguistic demands on readers. These differences may be related to the specific typology of the orthography and to specific language features. On the other hand, common cognitive processes might underlie the reading process across languages. Chapter 12 by Malatesha Joshi presents an integration of two studies, exploring both commonalities and differences of literacy acquisition in different languages. In the first study, reading comprehension of monolinguals is shown, in English and Spanish alike, to be strongly explained by decoding and listening comprehension. But decoding contributed more to reading comprehension in English, the language with more opaque orthography. In the second study, an analysis was carried out on reading and spelling of three bilingual children: one with dyslexia, another with hyperlexia, and the third a normal reader. These children spoke English and Kannada. Similar differences in decoding and comprehension emerged between the dyslexic and the hyperlexic children across the two languages, again showing commonalities between languages. Similarities and differences in second language acquisition were also studied by Liliana Tolchinsky. In Chap. 13 she describes a study that examined the effects of first language (L1) on the acquisition of the second language (L2). She compared morphological transformations in Catalan by children recruited from Moroccan and Chinese origins living in Spain. These two groups are particularly revealing because of the richer morphology in L1 of children originating from Morocco rather than China. Whereas the two groups exhibited similar levels of successful performance, they differed in the kind of mistakes they produced, reflecting the features of their L1. In the same vein, referring to similarities and differences of literacy acquisitions in different languages, Esther Geva and Dana Shafman in Chap. 14 examine the development in morphology, oral comprehension, and vocabulary in Hebrew of English-speaking children. These first- and second-grade children studied Hebrew as a foreign language (HFL) for about two and a half hours a day, 5 days a week. The researchers found that grade 2 children had somewhat better vocabulary, morphological, and syntactic skills than their counterparts in grade 1. The researchers claim that vocabulary is significant for the emergence of basic morphological knowledge in HFL children. A follow-up study is presented in Chap. 15 by Orly Lipka and Linda Siegel. The authors show how a balanced literacy program, consisting of early literacy skills and language support, can close gaps in literacy knowledge between children of L1 and English as a second language (ESL) background. The children, who received a balanced literacy program in kindergarten, were followed from kindergarten to fifth grade. Whereas in kindergarten, L1 speakers
Preface
xiii
outperformed their ESL counterparts on a wide range of measures (phonological processing, syntactic awareness, memory for sentence, and spelling), by grade 5, the same groups of children exhibited a similar level of literacy almost across the board. Part 4, Teaching and Learning, presents Chap. 16 by David Olson, who approaches the important issue of teaching. This chapter provides an overarching critical review of theories of teaching pedagogical theories. The author presents his own perspective on pedagogy, based on intentionality, responsibility, accountability, and the earning of entitlement. In Olson’s words: “What one knows, understands, and infers is subordinated to one’s ability to take on and successfully meet one’s obligations.” He calls for developing a new science of pedagogy. While this chapter has a broader perspective than usually met in literature on literacy, the examples utilized refer to teaching of literacy. We would like to thank the authors of the chapters in this book for their dedication and scholarship and for their enthusiasm in contributing to a book dedicated to Professor Iris Levin. The research literature on the development and enhancement of literacy across orthographies and cultures is expanding. We hope that the ideas set forth in this book will stimulate researchers around the world to broaden this line of studies and include more languages and orthographies in different countries.
Dorit Aram Ofra Korat
Contents
Part 1 Literacy Development 1 It’s All in the Name.................................................................................... Anna C. Both-de Vries and Adriana G. Bus
3
2 Learning About Writing Begins Informally............................................ Sarah Robins and Rebecca Treiman
17
3 The Role of Letter Name Knowledge in Early Spelling Development: Evidence from Brazilian Portuguese............................... Cláudia Cardoso-Martins and Marcela Fulanete Corrêa
31
4 An Eye for Print: Child and Adult Attention to Print During Shared Book Reading................................................................................ Mary Ann Evans and Jean Saint-Aubin
43
5 The Role of Hebrew Letter Names in Early Literacy: The Case of Multi-phonemic Acrophonic Names................................... Iris Levin
55
Part 2 Promoting Literacy at Home and in School 6 The Early Childhood Project: A 5-Year Longitudinal Investigation of Children’s Literacy Development in Sociocultural Context............................................................................ Susan Sonnenschein, Linda Baker, and Robert Serpell 7 Early Literacy at Home: General Environmental Factors and Specific Parent Input.......................................................................... Catherine McBride-Chang, Yvonne Y. Y. Chow, and Xiuli Tong
85
97
xv
xvi
Contents
8 Reading Books to Young Children: What It Does and Does Not Do....................................................................................... 111 Monique Sénéchal 9 The Making of Literate Families: Considerations of Context and Misconceptions............................................................... 123 Linda M. Phillips 10 Spellings of Words: A Neglected Facilitator of Vocabulary Learning........................................................................... 137 Linnea C. Ehri and Julie Rosenthal 11 Early Schooling and Growth of Literacy in the Transition to School.................................................................................................... 153 Frederick J. Morrison, Carol MacDonald Connor, and Annemarie Hindman Part 3 Bilingualism – Acquiring a Second Spoken and Written Language 12 Role of Orthography in Literacy Acquisition and Literacy Problems Among Monolinguals and Bilinguals............. 167 R. Malatesha Joshi 13 Becoming Literate in a Foreign Language............................................ 177 Liliana Tolchinsky 14 Rudiments of Inflectional Morphology Skills in Emergent English–Hebrew Biliterates.............................................. 191 Esther Geva and Dana Shafman 15 Early Identification and Intervention to Prevent Reading Difficulties................................................................................................. 205 Orly Lipka and Linda S. Siegel Part 4 Teaching and Learning 16 Whatever Happened to Pedagogical Theory?....................................... 223 David R. Olson Index.................................................................................................................. 235
Contributors
Linda Baker University of Maryland, Baltimore, MD, USA Anna C. Both-de Vries Leiden University, Leiden, The Netherlands Adriana G. Bus Leiden University, Leiden, The Netherlands Cláudia Cardoso-Martins Universidade Federal de Minas Gerais, Belo Horizonte, MG, Brazil Yvonne Y.Y. Chow The Chinese University of Hong Kong, Hong Kong Marcela F. Corrêa Universidade Federal de Minas Gerais, Belo Horizonte, MG, Brazil Linnea C. Ehri City University of New York Graduate Center, New York, NY, USA Mary A. Evans University of Guelph, Guelph, ON, Canada Esther Geva University of Toronto, Toronto, ON, Canada Annemarie Hindman University of Michigan, Ann Arbor, MI, USA Malatesha R. Joshi College of Education and Human Development, Texas A&M University, College Station, TX, USA Iris Levin Tel Aviv University, Tel Aviv, Israel
xvii
xviii
Orly Lipka University of British Columbia, Vancouver, BC, Canada Carol MacDonald-Connor University of Michigan, Ann Arbor, MI, USA Catherine McBride-Chang The Chinese University of Hong Kong, Hong Kong Frederick J. Morrison University of Michigan, Ann Arbor, MI, USA David R. Olson University of Toronto, Toronto, ON, Canada Linda M. Phillips University of Alberta, Edmonton, AB, Canada Sarah Robins Washington University, St. Louis, MO, USA Julie Rosenthal William Paterson University, Wayne, NJ, USA Jean Saint-Aubin Université de Moncton, Moncton, NB, Canada Monique Sénéchal Carleton University, Ottawa, ON, Canada Rebecca Treiman Washington University, St. Louis, MO, USA Robert Serpell University of Zambia, Lusaka, Zambia Dana Shafman University of Toronto, Toronto, ON, Canada Linda S. Siegel University of British Columbia, Vancouver, BC, Canada Susan Sonnenschein University of Maryland, Baltimore, MD, USA Liliana Tolchinsky University of Barcelona, Barcelona, Spain Xiuli Tong The Chinese University of Hong Kong, Hong Kong
Contributors
Part 1 Literacy Development
Chapter 1
It’s All in the Name Anna C. Both-de Vries and Adriana G. Bus
Abstract From Levin’s seminal research, it appears that name writing, a natural focus of most young children, is the first stable written form with meaning. Inspired by Levin’s finding, we began to investigate the role of name writing in children’s acquisition of alphabetic knowledge. We present here a series of three Dutch studies from which it appears that children’s understanding of writing as a symbolic device starts with the first letter of the proper name. Taking into account the finding that name writing in the preschool stage is one of the best predictors of conventional literacy in school age (Strickland & Shanahan, 2004), children’s own name, often the first word that they learn to read and write, may play a special role when they develop an understanding of the symbolic function of writing. Badian (1982), for instance, found name writing to be one of the top three predictors of both first- and second-grade reading achievement, using the Stanford Achievement Test Total Reading Score as the outcome measure. Furthermore, there is evidence for the assumption that name identification is strongly related to phonological skills and letter-sound knowledge, whereas recognition of environmental print (e.g., Exit, McDonalds or Coca Cola) is unrelated to phonological awareness or letter-sound knowledge (Blair & Savage, 2006). Learning to understand language, infants demonstrate significant preference for their own name when compared with foils (Mandel, Jusczyk, & Pisoni, 1995). Likewise, children’s writing of their own name is identifiable as writing prior to other words (Levin, Both-de Vries, Aram, & Bus, 2005). Besides, their proper name is among the first words that children can write conventionally (Levin et al., 2005).When children have learnt to spell their proper name, they follow it up by other names as Sarah’s case demonstrates: Once she was able to write her own name, she immediately began working intently on learning to write Matthew, Mommy, and Daddy (Martens, 1999). As the name is the first stable written form with meaning, it is assumed to represent a singularly important benchmark in early literacy development
A.G. Bus () Leiden University, Leiden, The Netherlands e-mail:
[email protected] D. Aram and O. Korat (eds.), Literacy Development and Enhancement Across Orthographies and Cultures, Literacy Studies 101, DOI 10.1007/978-1-4419-0834-6_1, © Springer Science+Business Media, LLC 2010
3
4
A.C. Both-de Vries and A.G. Bus
(Ferreiro & Teberosky, 1982; Welsch, Sullivan, & Justice, 2003). In line with this finding, we have begun to study how the proper name influences the spelling of other words, and the skills that are basic to spelling simple words.
Study I: Random Versus Ambiguous Use of Letters from the Name to Symbolize Meaning Following Ehri’s developmental portrayal, children’s emerging spelling skills can be described in terms of an increasing ability to parse spoken words and decide what units align with the written form (Ehri & Wilce, 1985). The name may be the start of this development. Exposure to their written name affords children many early learning opportunities by which graphemes and sounds become connected (Treiman & Broderick, 1998; Villaume & Wilson, 1989). More than other words, the name may elicit responses from parents and teachers like: “That’s /pi/ of Peter.” Children are thus faced with an important feature of written language, namely that a familiar letter form refers to a sound in a spoken word (Byrne, 1998; Molfese, Beswick, Monlar, & Jacobi-Vessels, 2006).We hypothesize therefore that children’s understanding of writing as a symbolic device starts with letters from their names. Following this line of argument, we may expect that phonetic writing starts with the letters of a child’s own name, whatever those letters are. However, so far the literature does not provide unanimous support for this hypothesis. Treiman and Broderick (1998) found that English-speaking children do not necessarily know the letter sound for the first letter of their own name, even when they show a relatively good knowledge of the conventional label of this letter. That is, a child named Victor is likely to be better than a child named Susan at naming the letter V correctly, but not at saying the corresponding sound /v/. If we assume that knowledge of the letter-sound relationship is essential, this hypothesis predicts that Victor wouldn’t spell V better than other letters when making attempts to write words. On the other hand, there is evidence that young speakers of Hebrew show elevated levels of letter-sound knowledge for the first letter of their own name (Levin & Aram, 2004). This finding would predict that a child’s spelling might be stronger for that letter in particular. Studying kindergarten children and first and second graders, Treiman and colleagues (Treiman, Kessler, & Bourassa, 2001) found that early phonetic spellings are not limited to letters from the name. In spite of Treiman’s results, we hypothesize that letters of the child’s name are incentives for phonetic sensitivity. Assuming that sounding out is first practiced with letters of the name (grown-ups may often say: “That’s /pi/ of Peter”, but not too often: “That’s /bi/ of bear” or something similar), we expect that for a short period, phonetic writing is restricted to the letters of a child’s name. In Treiman et al.’s sample (2001), the effect of name letters may not have been identified because a group of children who had just started to write phonetically was mixed with a some what more advanced group, thus distorting effects of the name at the very beginning.
1 It’s All in the Name
5
Several studies have reported that young children select letters from their own name when they compose texts or write down words that are dictated (Aram & Levin, 2001; Bloodgood, 1999; Treiman et al., 2001). Bloodgood (1999), for instance, concluded after a close examination of the letters used by 30 four- and five-year-old children in writing samples collected over a year’s time that 41% of the letters were letters from their own name. Children may prefer these letters to other letters from the alphabet because they know how to print letters from the name and probably also because they are aware that these letters symbolize meaning (Sulzby, Barnhart, & Hieshima, 1989). Most children have experienced that grownups are able to recognize their name writing. Another possibility is that the sounds of name letters are recognized in spoken words prior to other letter sounds because children often practice sounding out name letters when grown-ups sound out letters of the name. Adults may often say: “That’s your letter, /di/ of Daisy.” However, as letters from the name are known to be utilized randomly when writing unpracticed words, it is hard to decide when these letters are used phonetically. Letters of the name may indeed match sounds of dictated words, but they may be selected purely by accident. Treiman and colleagues (2001) characterized those letters therefore as being used ambiguously, leaving open the possibility that letters from the child’s name were selected by chance even where sounds did match. To test phonetic use of any letter it is essential to contrast words including the letter and words not including the letter. When children use a letter equally often in both selections, it is plausible that the letter is selected by chance. If however a letter is more often used ambiguously than randomly we may assume that children use this letter not purely by chance but that they can identify the letter sound in the spoken word. We designed a study (Both-de Vries & Bus, 2009a) to test whether inventing alphabetic–phonetic spelling begins with letters from the name by showing for the earliest stage of alphabetic–phonetic writing that children are best at representing letters from the proper name. This study included 35 children in the age range of 3½ to 5 using conventional letters in more than half of 16 dictated words, but the products rarely were conventional or readable invented spellings. Besides their proper name, we dictated 16 randomly chosen words like DROPJES [liquorices], SNEEUW [snow], WIEL [wheel], DOOS [box], BOOM [tree], and DRIE BOMEN [three trees]. First we coded per child which conventional letters occurred in one or more dictated words, then for each letter we coded whether the letter was the first letter from the name, another letter from the name or a letter not included in the name (referred to here as non-name letter). In addition, for each of the letters we coded per word whether the word included the letter, and whether the letter was used ambiguously (it matched a sound in the dictated word) or randomly (it did not match a sound in the dictated words). Because per child the number of dictated words varied that included the first letter of the name, another letter from the name, or a non-name letter, we divided the number of words in which a child had used the focal letter ambiguously or randomly by the total number of words that did or did not include this letter. This resulted in two percentages per letter: random and ambiguous use of the focal letter. For example: Sandra used the first letter of her name in her written
6
A.C. Both-de Vries and A.G. Bus
representation of two out of four dictated words that indeed include an s/z-sound (i.e., dropjeS [liquorices], Sneeuw [snow], dooS [box], and Zon [sun]). This score resulted in a 50% score for ambiguous use of the first letter of her name. Sandra also used S in ten of the 12 other words that do not include S (e.g., moeder [mother], baby [baby], etc.) resulting in an 83% score on random use of the letter S. The same coding was done for each of the other letters in Sandra’s name, that is, for A, N, D and R, and for the three non-name letters M, O, and F that she used in her written representations of dictated words. We thus tested the difference between ambiguous and random use of letters for (1) the first letter of the name (here: S), (2) the other letters of the name (here: A, N, D, and R), and (3) non-name letters appearing in the spelling of words (here: M, O, and F). The children in this study often selected letters from their proper name to write dictated words; on average 52% (SD = 21) of all written letters were letters from the proper name. The first letter of the name occurred as often as other letters of the name. In so far as they used non-name letters, simple forms like O and I dominated. On analyzing the data, we separated children not yet writing phonetically from those who had just started to produce some phonetic spelling. The group not yet writing phonetically (N = 17) characteristically mainly produced random letter strings for the dictated words rarely selecting correct letters (they wrote at most one correct letter in two out of 16 dictated words). The other group who produced some phonetic writing (N = 18), by contrast, chose one or more correct letters in six or more words out of 16 words. Apart from very few exceptions these children did not produce readable invented spellings. Children from both groups were on average 4 years and 6 months old. Ambiguous use of letters was contrasted with random use in both spelling groups. The findings corroborated our hypothesis that phonetic spelling starts with the first letter of the proper name. The percentage of ambiguous use of the first letter from the name (M = 55.4; SD = 39.9) exceeded the percentage of random use among children who had started to write phonetically (M = 37.5; SD = 37.3) The contrast was statistically significant. Hence the ambiguous first letters were not chance hits but indicated that these children often selected the first letter of the name because it represents a sound in the word. By way of contrast, the difference between random and ambiguous first letters of the name was not statistically significant for the group mainly writing random letter strings; mean percentages for random and ambiguous first letters were M = 38.7 (SD = 33.8) and M = 34.5 (SD = 33.2), respectively. These children used the first letter of their name as often randomly as ambiguously, which indicates that at this writing level correctly selected first letters from the name were chance hits and not chosen by the child because he or she had noticed similarity between the proper name and dictated words. All children used one of the other name letters in half of the 16 dictated words. However, both spelling groups used these name letters as often ambiguously as randomly. Apparently, other name letters are known as visual forms that belong to written text but children do not notice similarities between the name and dictated words that motivate them to select these letters and they use them purely randomly. The same was true for non-name letters. By averaging scores for all name letters (except the first letter) and all non-name letters any contrasts between ambiguous and random use of some letters may have been evened out. To test whether other
1 It’s All in the Name
7
letters than the first letter of the name were used phonetically, we selected per child another letter, namely one that apart from the first letter of the name most frequently occurred in the 16 dictated words. It could be a non-name letter or a name letter. In many cases O or I frequently occurred probably due to their simplicity. Testing revealed that children from both spelling groups used the thus selected letter as often ambiguously as randomly; the difference between ambiguous and random use was not statistically significant according to a matched-pair Wilcoxon test. These findings suggest that phonetic writing starts with the first letter from the name whichever it is. Ahead of all other letters, children are able to recognize the sound of this letter in dictated words and to correctly represent this letter in the spellings that they make up to represent referents. It is not very plausible that children select the first letters of the name because they are “easy” letters. Inspecting the set of first letters it is striking that this set includes most letters of the alphabet (18 out of 26) and not just “easy forms” like O and I or acrophonic letter names like P or T where the CV format of the name may facilitate acquisition of letter sounds (Foulin, 2005). As yet, the most plausible explanation for phonetic use of the first letter of the name preceding other name letters is that this letter is practiced more than any other letter as a form that relates to a name or sound. Considering that grown-ups often say things like: “That’s your letter, the /ti/ from Tom,” we hypothesize that children start to use the first letter of the name phonetically because they are familiar with this letter and how it sounds in words. A second study (Both-de Vries & Bus, 2009a) was designed to test whether more advanced children generalize phonetic wriiting to other letters from the name. When supporting their children as they try to write unpracticed words parents may use other letters from the name as cues: “It’s /r/ like in Peterrr” (Aram & Levin, 2001). Alternatively, children with an emerging understanding of the alphabetic–phonetic principle (they are able to recognize and write the first letter of the name in spoken words) may, while writing new, not yet practiced words wonder how to represent other unknown sounds. Children may thus elicit instruction in non-name letters and sounds. In the first study children who were just starting phonetic writing, wrote other letters from their name more often than non-name letters, but phonetic writing of these other name letters was a chance hit. We wondered if somewhat more advanced children first expand phonetic writing to other letters from their name and use the other letters from their name more often ambiguously than non-name letters.
Study II: Do Children Generalize Phonetic Writing First to Other Letters from Their Name or Directly to Other Non-name Letters? A second study tested: (1) whether older and more advanced children than those in the first study generalize phonetic writing first to other letters from their name or directly to other non-name letters; and (2) whether in a group of children from low SES families we can replicate that phonetic writing starts with the first letter of the name. Similar to the first study we selected children who used conventional letters to represent their name and other words and we excluded children who merely
8
A.C. Both-de Vries and A.G. Bus
produced drawings or writing-like products. Thus, 79 Dutch children aged 4–6 years old (M = 5; 8, range: 4; 3–6; 5) were selected from a complete sample of 88 children from three classrooms of a school in a small town in the Netherlands. Overall these children came from families with low-educated parents. Similar to the first study, in addition to their name the children wrote nine other words: KAAS [cheese], ZAK, [sac], WIP [seesaw], POP [doll], VIS [fish], ZON [sun], PAPA [daddy], MAMA [mommy], and FLIPPO [pog]. The instructions were straightforward: “Write your name” and “Write X” (i.e., ‘vis’ [fish]). PAPA and MAMA were not included in the analyses, because we noticed that children knew these words better than other words (Bus & Out, 2009; Martens, 1999). On average 23% (N = 18) of the group wrote papa and/or mama conventionally, whereas only 13% (N = 10) wrote one or more of the other dictated words conventionally. Per conventional letter we examined: (1) whether the letter was the first or another letter from the name or a letter not from the name; and (2) whether the letters were used randomly or ambiguously. This resulted in two scores per letter: random and ambiguous use. In this study we distinguished three spelling groups. Similar to Study I, some children (N = 26) made random letter strings rarely selecting correct letters (at most one letter in one or two of all dictated words), whereas other somewhat more advanced children (N = 37) wrote one letter correctly in three or more dictated words. Unlike the first study the sample included a third group of more advanced pupils (N = 16) who scored 2 or more correct letters per word resulting in a substantial number of somewhat readable invented spellings. They would write KS instead of KAAS [cheese] or VS instead of VIS [fish]. Quality of name writing differed among the three writing levels; according to a Kruskal Wallis test, this difference was statistically significant, c2 (2, N = 79) = 9.65, p < .008. Children who just started to write phonetically wrote almost all letters of their name correctly (correct: 87%), whereas almost half of the least advanced children wrote only one or two letters of their name correctly (correct: 54%). All children in the most advanced group wrote their name correctly. In this sample including older Dutch kindergarten children from low SES families, we were able to replicate the main result of the first study: invented spelling begins with the first letter of the proper name. The first letter of the name was the letter most often used correctly when children just started to write some correct letters to represent referents; the percentage of ambiguously written first letters exceeded that of randomly written first letters indicating that the correctly used first letters were no chance hits selected by children because they are familiar with the letter form. Other letters from their own name often appeared in their writing (in about half of the dictated words) but apparently not because of the match with sounds in the word. In so far as children were familiar with the spelling of other well-known words such as MAMA [mommy] and PAPA [daddy] the findings offered some support for the hypothesis that first letters of these words may also be among the first to be written phonetically. These effects were not found in the least advanced group who rarely selected correct letters. The results of this second study are unique because we were able to reconstruct how invented spelling develops after children have begun to represent the first letter of the
1 It’s All in the Name
9
name correctly in dictated words. Unlike the first study, the data of this study enabled us to test whether somewhat more advanced children (the ones who represent one or two letters correctly in most dictated words) use other letters from the name correctly before non-name letters. Surprisingly we found that in this more advanced group phonetic writing was not limited to the letters of the name. This group used all letters from their proper name more often ambiguously than randomly but also letters not from the name. In other words, only at the very start is invented spelling restricted to the first letter from the name. The alphabetic–phonetic principle is transferred to a variety of letters after children have discovered by means of the first letter of the name that letters relate to sounds in spoken words. We can imagine that after children have become proficient in writing their own name, they ask for support in writing other names and referents. In response to that, grown-ups expand instruction to other letters; they demonstrate how letters other than the first letter of the child’s name are written and sound in words thereby stimulating that children also use other letters in their invented spellings.
Study III: How Does Name Writing Influence Children’s Alphabetic–Phonetic Knowledge? The results so far indicate that the first letter of the name, whichever it is, is the one and only letter that is written phonetically at the very start of phonetic writing. A plausible explanation is that knowing how to write the name boosts instruction in naming the first letter of the name and in phonetic sensitivity for this letter (Foulin, 2005). We can imagine that instruction elicited by the first letter of the proper name mediates between name writing and inventing phonetic spelling. By naming the first letter, children learn the name of this letter before any other letters (cf. Treiman, Levin, & Kessler, 2007). Moreover, young children may thus practice phonetic sensitivity, i.e., to identify the sound of the first letter of their name in a spoken word (Molfese et al., 2006); grown-ups may often make comments like: “It’s /pi/ of Peter.” According to this line of reasoning, we expected: (1) that name writing promotes familiarity with the first letter of the name and that young children know the first letter of their name better than any other letter, (2) that phonemic sensitivity for the first letter of their name precedes phonemic sensitivity for other letters (cf. Vernon & Ferreiro, 1999), and (3) that both skills benefit the spelling of words that include the first letter. Alternatively, it is conceivable that knowing the first letter of their name is sufficient: the first letter, and not phonemic sensitivity for this letter, predicts why spelling starts with the first letter of the name.
Writing the First Letter of the Name Phonetically We designed another study (Both-de Vries & Bus, 2009b) to test these hypotheses. It was our intention to preclude that, like in our previous studies (Both-de Vries & Bus, 2009a), some name letters fail to occur in the dictated words or very few
10
A.C. Both-de Vries and A.G. Bus
dictated words include the first letter of the name. In the previous studies it was possible that the first letter of the name occurred in 1 or 2 dictated words, thus making scores more susceptible for chance hits. In Study III this problem was eliminated by composing per child two sets of eight words, one with the first letter of their name and one without. Assuming that letters at the start of words are easier than letters at the end we presented the first letter of the name four times as first letter and four times as last letter (Ehri & Wilce, 1985). For instance, Nico’s set with the name letter included eight words with N at the start (Neus [nose], Net [net], Noot [nut], Nagel [nail]) or at the end (peN [pen], maN [man], teeN [toe] en zoN [sun]). Similar sets were composed for all 60 participants. Furthermore, the study was designed such that we could test effects of other characteristics of the first letter of the child’s name. Most Dutch consonant names contain the corresponding letter sounds, either as initial phoneme, i.e., following the acrophonic principle (CV letter names: B, D, or P) or as final phoneme (VC letter names: F, L, M, N, R, or S). Children may have more to gain from acrophonic first letters than from nonacrophonic letters since it is easier to recognize /b/ in ‘bear’ than /es/ in ‘stone.’ To test this effect we included participants with both types of first letters. About half (N = 25) of the names of the participants started with a CV letter and half (N = 35) with a VC letter. Furthermore, the pronunciation of consonants may influence letter-sound learning (Stuart & Coltheart, 1988). To test effects of this characteristic about half (N = 36) of the participants’ names started with an obstruent consonant (e.g., B, C, D) and half with a less pronounced sonorant consonant (e.g., L, M, R). The natural consequence of these aims was that children were only qualified as participants if their names started with consonants. Of the 60 participants, all 4–5½ years old, 39 (65%) wrote their name readably producing at least invented spelling, i.e., all consonants and vowels were represented by appropriate letters. The rest wrote a few correct letters from their name at most. Children used the first letter of their name quite frequently, namely in 41% of 16 dictated words. Overall, children wrote the first letter of their name more often when it indeed appeared in the dictated word. When children could write their name, ambiguous use of the first letter outnumbered random use, F (1, 58) = 13.58, p < .001, h ² = .19. This group used the first letter of their name twice as often in ` words that indeed included this letter: The first letter of the name was written in 5.2 out of eight words with this letter and in 2.6 out of eight words without this letter. Apparently, this group used this letter phonetically. On the other hand, children not able to write their name used the first letter of the name as often randomly as ambiguously; the letter appeared on average in two out of eight words in both sets of words. The most advanced children who wrote other letters than the first letter of their name phonetically may have been accountable for the difference between ambiguous and random writing of the first letter of the name. We therefore redid the analyses including only children who exclusively used the first letter of the name correctly. The outcomes for the subsample were similar to the outcomes for the complete group. A Wilcoxon Signed Rank Test contrasting the two sets of words revealed a statistically significant effect, Z = −2.58, p < .01. Furthermore,
1 It’s All in the Name
11
according to insignificant t-tests, CV versus VC letters as well as obstruent consonants versus sonorant consonants revealed similar outcomes on ambiguous writing of the first letter of the name, thus indicating that neither letter name type nor consonantal strength influences the spelling.
Letter Name Knowledge If writing the proper name is an incentive to learn the first letter of the name we could expect that children who are able to write their name know the first letter of their name. We may also expect that being able to name the first letter precedes naming other name or non-name letters. To investigate these hypotheses children were tested on naming 37 letters. The letters were presented on seven sheets each including five or six different letters. Children were invited to name the letters they know; both the letter name and sound were awarded. The first letter of the name was the best-known letter when children were able to write their name. Eighty percent of the children who wrote their name legibly were able to name the first letter of the name correctly, whereas only 19% of the group who was unable to write the name correctly could name the first letter, resulting in a high correlation between name writing and knowing the first letter of the name (j = .59, p < .001). The two groups (able to write their proper name or not) did not differ statistically significantly in naming the second letter of the name correctly; overall about one-third of the children (37%) named the second letter correctly. And the groups did not differ in naming other non-name letters; the children who were able to write their name and the children who were not knew on average 25% (SD = 31) and 14% (SD = 22) of the non-name letters, respectively.
Phonetic Sensitivity Does knowing the name of the first name letter stimulate children so they can isolate the sound of this letter in spoken words? Does phonetic sensitivity start with the bestknown letter, the first letter of the name, or is it a skill that develops independent of letter knowledge? We asked children to isolate the first or last sound of spoken words. In one set of six words the first or last sound was the sound of the first letter of the name and in the other set the focal sound was a randomly chosen non-name sound. In all, the task included 12 three- or four-phoneme words. Children isolated the sound of the first name letter more often correctly (M = 2.60, SD = 2.52) than a random sound (M = 1.98, SD = 2.14). Children who knew the first letter of their name were more successful in recognizing this letter’s sound (M = 3.66, SD = 2.41) than in isolating sounds of other letters (M = 2.54, SD = 2.32), F (1, 58) = 5.55, p < .02, h2 = .09. They isolated the sound of the first letter of their name 1.5 times more often correctly than a randomly chosen sound. On the other hand, children who did not know the
12
A.C. Both-de Vries and A.G. Bus
first letter of their name scored not better on the sound of the first letter of the name than on a non-name letter; overall their scores on the sound of the name letter (M = 1.12, SD = 1.86) and on the sounds of non-name letters (M = 1.20, SD = 1.61) were low. According to insignificant t-tests, CV versus VC letters revealed similar outcomes, indicating that this characteristic of letter names is not important. Consonantal strength, on the other hand, made some difference; children with an obstruent first letter (M = 4.42, SD = 2.12) named more letters in six words correctly than children with a sonorant (M = 2.75, SD = 2.49) first letter of the name, t(33) = 2.15, p < .04.
How Does the Ability to Write the Proper Name Promote Alphabetic–Phonetic Writing? Is the relationship between name writing and writing the first name letter phonetically mediated by both phonemic sensitivity for the sound of the name letter and familiarity with the first letter’s name? Or is it enough if children know the first letter’s name? To choose between these options we carried out hierarchical regression analysis with: (1) name writing (coded on a 6-point-scale), (2) children’s ability to name the first letter of the proper name, and (3) children’s ability to isolate the sound of this letter in six spoken words including this sound, as predictors. The variables were entered in this order. The dependent variable was their ability to represent the first letter of the name phonetically in dictated words that included this letter. Name writing did not explain any variance when knowledge of the first letter of the name and phonetic sensitivity were entered; see model 2 and 3 in Table 1.1. This proves that name writing does not trigger other relevant concepts of writing except for knowledge of the first letter’s name and phonetic sensitivity for the sound of this letter. This means it is plausible that name writing promotes Table 1.1 Prediction of phonetic writing of the first letter of the name by knowledge of the first letter’s name, phonetic sensitivity for the first letter of the name, and ability to write the proper name Variable Model 1 Model 2 Model 3 Standardized estimate (b) Able to write the name .45*** .20 .14 Knowing first letter of the name .42** .30* Phonemic sensitivity for name letter .32* Model statistics F (change) 14.65*** 9.7** 6.6* Error df 58 57 56 R² .20 .32 .39 .12 .07 ∆R² *p < .05 **p < .01 ***p < .001
1 It’s All in the Name
13
alphabetic knowledge by boosting instruction in basic skills. The results also show that it is not sufficient to know the first letter’s name or sound. Phonetic sensitivity, i.e., being able to identify the sound of the first letter of the name in spoken words, explains additional variance after we controlled for knowledge of the first letter’s name. In addition to the first letter’s name that explained 9% of children’s spelling ability (b = .30), phonetic sensitivity for the first letter of the name explained an additional 10% (b = .32). Knowing the first letter’s name is important but children also need to know how this letter sounds in spoken words.
In Conclusion In summary, in three subsequent studies we found that the first letter of the child’s name, whichever it is, is the first letter that is written phonetically as children begin to create invented spellings. We have also found evidence for the hypothesis that phonetic sensitivity skills start with the first letter of the name; young children are successful in recognizing this letter in spoken words before any other letter. By promoting knowledge of the first letter’s name and phonetic sensitivity for this letter, name writing functions as the stepping stone to the alphabetic strategy (Frith, 1985). We found no support for the alternative hypothesis that from the very beginning phonetic sensitivity is not restricted to the first letter of the name and that invented spelling mainly depends on letter knowledge (cf. Vernon & Ferreiro, 1999). The way grown-ups react to name writing (for instance, recognizing the name and reading it aloud) may stimulate the shifts in children’s procedural knowledge of writing, described above (Levin & Aram, 2004; Levin & Bus, 2003). Interactions surrounding name writing happen at home from an early age and are carried out by family members as routine activities (Gillanders & Jimenez, 2004). In joint attentional scenes like name writing grown-ups may stimulate children to reflect on their rudimentary writing activities, which may improve children’s understanding of basic concepts of writing (Tomasello, 1999). As grown-ups read the name, children begin to reflect on what makes writing readable and they may isolate features of their performance relevant to that success. As a result, children may become aware of letters as symbols and use these letters (“my a”) when they write other unpracticed words. By sounding out letters (/t/ of Taco) grown-ups initiate children’s reflection on imitative writing and pave the way to a higher level of understanding of how the written form represents a referent. They focus children’s attention on letter units and how they sound in spoken words thus promoting alphabetic–phonetic writing that goes beyond imitation of the form. Grown-ups provide children with fairly substantial amounts of direct instruction about letters as symbols by talking about children’s own or other people’s letters and how they sound in words: “Look, that’s your letter” or “That’s the ‘m’ from mama” (Levin & Aram, 2004; Welsch et al., 2003). Grown-ups thus unintentionally instruct children on how letters of the name sound in words, thus stimulating phonetic sensitivity for the first letter of the name (Ehri & Wilce, 1985; Frost, 2001). Bus and colleagues concluded from a series of
14
A.C. Both-de Vries and A.G. Bus
case studies (Bus, Both-de Vries, de Jong, Sulzby, de Jong, & de Jong, 2001) that without instruction by grown-ups 4- and 5-year-old children are not actively seeking to expand their elementary alphabetic knowledge; on the contrary, they then revert to simpler strategies like random letter strings or pseudocursive scribble even when they are able to produce some invented spelling. Even though there is evidence suggesting that early writing development is qualitatively similar for various countries (Levin & Bus, 2003) we wonder whether the present findings for Dutch samples also apply in a similar way to countries where teaching of reading starts earlier and kindergarten children are taught the letter names. In the Netherlands where teaching letter names is not part of the kindergarten curriculum, learning may depend more than in other countries on informal print experience like name writing for the inducement of some letter knowledge and phonetic sensitivity. On the other hand, it is conceivable that practice of letter names and phonetic sensitivity is a waste of time until children develop some rudimentary understanding of the basic concepts of alphabetic writing.
References Aram, D., & Levin, I. (2001). Mother-child joint writing in low SES: sociocultural factors, maternal mediation, and emergent literacy. Cognitive Development, 16, 831–852. Badian, N. A. (1982). The prediction of good and poor reading before kindergarten entry: a 4-year follow-up. The Journal of Special Education, 16, 309–318. Blair, R., & Savage, R. (2006). Name writing but not environmental print recognition is related to letter-sound knowledge and phonological awareness in pre-readers. Reading and Writing, 19, 991–1016. Bloodgood, J. W. (1999). What’s in a name? Children’s name writing and literacy acquisition. Reading Research Quarterly, 34, 342–367. Both-de Vries, A. C., & Bus, A. G. (2009a). Name writing: a first step to phonetic writing? Does the name have a special role in understanding the symbolic function of writing? Literacy Teaching and Learning, 12, 37–55. Both-de Vries, A.C., & Bus, A.G. (2009b). The proper name as starting point for basic reading skills. Reading and Writing. DOI 10.1.007/s11145-008-9158-2. Bus, A. G., Both-de Vries, A. C., de Jong, M.T., Sulzby, E., de Jong, W., & de Jong, E. (2001). Conceptualizations underlying emergent readers’ story writing. Ann Arbor, MI: Ciera report #2.015. Bus, A. G., & Out, D. (2009). Unraveling genetic and environmental components of early literacy; a twin study. Reading and Writing, 22, 293–306. Byrne, B. (1998). The foundation of literacy. The child’s acquisition of the alphabetic principle. East Sussex: Psychology Press Ltd. Ehri, L. C., & Wilce, L. S. (1985). Movement into reading: is the first stage of printed word learning visual or phonetic? Reading Research Quarterly, 20, 163–179. Ferreiro, E., & Teberosky, A. (1982). Literacy before schooling. Portsmouth, NH: Heinemann. Foulin, J. N. (2005). Why is letter-name knowledge such a good predictor of learning to read? Reading and Writing, 18, 129–155. Frith, U. (1985). Beneath the surface of dyslexia. In K. E. Patterson, J. C. Marshall & M. Coltheart (Eds.), Surface dyslexia. neuropsychological and cognitive studies of phonological reading (pp. 301–330). London: Erlbaum.
1 It’s All in the Name
15
Frost, J. (2001). Phonetic awareness, spontaneous writing, and reading and spelling development from a preventive perspective. Reading and Writing, 14, 487–513. Gillanders, C., & Jimenez, R. T. (2004). Reaching for success: a close-up of Mexican immigrant parents in the USA who foster literacy success for their kindergarten children. Journal of Early Childhood Literacy, 4, 243–269. Levin, I., & Aram, D. (2004). Children’s names contribute to early literacy: a linguistic and social perspective. In D. Ravid & H. Bat-Zeev Shyldkrot (Eds.), Perspectives on language and language development. Dordrecht: Kluwer. Levin, I., & Bus, A. G. (2003). How is emergent writing based on drawing? Analyses of children’s products and their sorting by children and mothers. Developmental Psychology, 39, 891–905. Levin, I., Both-de Vries, A. C., Aram, D., & Bus, A. G. (2005). Writing starts with own name writing: from scribbling to conventional spelling of Israeli and Dutch children. Applied Psycholinguistics, 26, 463–477. Mandel, D. R., Jusczyk, P. W., & Pisoni, D. B. (1995). Infants’ recognition of the sound of their own names. Psychological Science, 6, 314–317. Martens, P. A. (1999). Mommy, how do you write “Sarah?”: the role of name writing in one’s child literacy. Journal of Research in Childhood Education, 14, 5–15. Molfese, V. J., Beswick, J., Molnar, A., & Jacobi-Vessels, J. (2006). Alphabetic skills in preschool: a preliminary study of letter naming and letter writing. Developmental Neuropsychology, 29, 5–19. Strickland, D. S., & Shanahan, T. (2004). Laying the groundwork for literacy – Preliminary findings from the National Early Literacy Panel. Educational Leadership, 6, 74–77. Stuart, M., & Coltheart, M. (1988). Does reading develop in a sequence of stages? Cognition, 30, 139–181. Sulzby, E., Barnhart, J., & Hieshima, J. (1989). Forms of writing and rereading from writing: a preliminary report. In J. Mason (Ed.), Reading-writing collection (pp. 31–63). Neeham Heights, MA: Allyn & Bacon. Tomasello, M. (1999). The cultural origins of human cognition. Cambridge, MA: Harvard. Treiman, R., & Broderick, V. (1998). What’s in a name? Children’s knowledge about the letters in their own names. Journal of Experimental Child Psychology, 70, 97–116. Treiman, R., Kessler, B., & Bourassa, D. (2001). Children’s own names influence their spelling. Applied Psycholinguistics, 22, 555–570. Treiman, R., Levin, I., & Kessler, B. (2007). Learning of letter names follows similar principles across languages: evidence from Hebrew. Journal of Experimental Child Psychology, 96, 87–106. Vernon, S. A., & Ferreiro, E. (1999). Writing development: a neglected variable in the considerations of phonological awareness. Harvard Educational Review, 69, 395–415. Villaume, S. K., & Wilson, L. C. (1989). Preschool children’s explorations of letters of their own names. Applied Psycholinguistics, 10, 283–300. Welsch, J. G., Sullivan, A., & Justice, L. M. (2003). That’s my letter!: what preschoolers’ name writing representations tell us about emergent literacy knowledge. Journal of Literacy Research, 2, 757–776.
Chapter 2
Learning About Writing Begins Informally Sarah Robins and Rebecca Treiman
Abstract The first few years of formal schooling focus on teaching children how to read and write. In this chapter, we explore what children may learn about written language prior to receiving formal instruction. Specifically, we review evidence suggesting that children acquire information about the surface and deeper features of language through informal interactions with their parents and other adults. For example, children learn about the surface features of language (the general features of print, as well as features of words and letters) through exposure to print in their environment and through activities such as learning about the alphabet. Children may also begin to learn about the deeper features of print (how writing symbolizes language) through interactions with their parents that suggest that written language is in some ways similar to spoken language. The research findings suggest that children learn a good deal about written language prior to instruction in school but that their knowledge may be restricted to elements of the language that are particularly salient to them, such as their own names. Recognizing the role that these experiences play in shaping young children’s knowledge of written language offers new avenues of research, as well as suggestions for how to improve formal education in reading and writing.
Introduction Learning to read and write is a difficult task for young children. For example, a child needs to understand that the printed word dog symbolizes a spoken word (which is itself a symbol), and that it does so in virtue of the sounds that correspond to each of the letters in the word. How do children acquire this ability? Most children do not fully understand the spelling–sound relations that underlie reading R. Treiman () Washington University, St. Louis, MO, USA e-mail:
[email protected] S. Robins Washington University, St. Louis, MO, USA e-mail:
[email protected] D. Aram and O. Korat (eds.), Literacy Development and Enhancement Across Orthographies and Cultures, Literacy Studies 101, DOI 10.1007/978-1-4419-0834-6_2, © Springer Science+Business Media, LLC 2010
17
18
S. Robins and R. Treiman
and writing without formal instruction. In many countries, this instruction proceeds by teaching children the sounds of the individual letters and then teaching them to pronounce and spell written words. Much research has focused on this explicit instruction, asking which methods best help children learn about printed language (e.g., Snow & Juel, 2005). Becoming fully literate may require formal instruction, but children learn about some aspects of print informally. This learning may occur at home, through interactions with parents, or in preschools and daycare centers. Indeed, research has shown that children in literate societies often receive information about the printed language in such ways. These findings suggest that, even before they can read or write, children may know a good deal about how the printed language works. This general point has been urged by several theorists (e.g., Ferreiro & Teberosky, 1982; Sulzby, 1985). In this chapter, we identify two classes of information about written language that children often receive before formal literacy instruction begins and discuss how this information may help children learn about reading and writing. The first type of information concerns the surface features of writing: what letters and words look like. A second concerns the deeper features of the written language, specifically the fact that print symbolizes spoken language. Informal learning about the surface and deep features of writing begins at an early age, and it can help prepare children for the formal instruction on the specifics of the written language that takes place in school.
Surface Features of Language Children begin to learn about written language by learning what print looks like. Children in literate societies see print on cereal boxes, traffic signs, notes around their homes, and in many other contexts. Through such exposure, children gain familiarity with the general visual characteristics of print and begin to learn how it differs from other symbol systems. From here, children can begin to identify particular units of print that they encounter frequently and may come to recognize features of the symbols that comprise those units. Knowledge of the surface features of written language can be described at three levels: general surface features of print, words, and letters.
Children’s Knowledge of the General Surface Features of Print Understanding written language requires differentiating it from other symbol systems. Graphically, writing differs from other symbol systems along a variety of dimensions. Writing differs from drawing in two central ways. First, the symbols of writing are organized into straight lines, whereas drawing allows for nonlinear arrangements. Second, the symbols used in the written language rarely look like
2 Learning About Writing Begins Informally
19
what they represent. In the few cases that they do, as in the case of the two circles in look that look rather like eyes, the resemblance is usually incidental. In drawing, on the other hand, the symbols are intended to look like the objects they symbolize. These two differences between writing and drawing correspond to two general features of the printed language: linearity and lack of iconicity, respectively. Written language differs from another symbol system, numerals, in that the symbols of printed language are not often repeated. For example, 8888 is a number, and one that a child might see, as on a mailbox. In contrast, pppp cannot be a word. Additional characteristics of the written language are specific to particular systems. One such characteristic is directionality: whether a language is read from left to right or from right to left. For example, Hebrew goes from right to left while English and Spanish go from left to right. Languages also differ in the orientation of writing: Although all are linear, some – such as German – arrange their symbols in horizontal lines, while others – such as Japanese – traditionally use vertical lines. Written languages also differ in their systems of symbol shapes, as can be seen from the differences between the letters of the Latin alphabet and Chinese characters. Studies suggest that children learn about the surface characteristics of print through exposure to print in their environments. Before formal reading instruction begins (around age six in the U.S. and Israel), children show a preference for print that is compatible with the general characteristics of the written language to which they have been exposed. In one study, English-speaking 3-year-olds were more likely to select sets of symbols from their own language as instances of writing than Chinese characters or Mayan designs (Lavine, 1977). In another study, differentiation based on symbol shape did not occur until closer to four and a half years of age (Levy, Gong, Hessels, Evans, & Jared, 2006). Either way, a preference for the symbols of the child’s own printed language develops prior to formal instruction about word reading and spelling. Additionally, when asked to choose items that are appropriate for reading, children as young as 3 years of age tend to select linear arrangements of symbols (Ganopole, 1987). Children tend to reject strings of identical symbols (e.g., pppp) as appropriate for writing (Lavine, 1977). Children display an understanding of the general surface characteristics of print in their own attempts at writing. Four-year-olds from the U.S. who cannot write actual words tend to arrange their scribbles horizontally (Brenneman, Massey, Machado, & Gelman, 1996), and some studies report a correspondence between children’s scribbling movements when they attempt to write and the directionality of their written language (Pontecorvo, 1985; Sulzby, 1996). In a study with Hebrewand Dutch-speaking children, Levin and Bus (2003) showed that, starting after around their third birthday, children’s writings and drawings began to look subtly different from one another. The writings tended to be organized linearly and divided into discrete segments, whereas drawings were larger, nonlinear, and more continuous. Children showed some ability to tell the difference between their writings and drawings when asked about them later, as did their parents. Similar findings have been reported for Japanese-speaking children (Yamagata, 2007). However, the differences between children’s writings and drawings are subtle, and neither the writing nor the drawing looks much like what it is intended to represent. This is
20
S. Robins and R. Treiman
reflected in the fact that parents typically refer to both as scribbling. Although it takes children some time to make more prominent distinctions between the two systems, they seem to possess some general understanding of the distinction between writing and other symbol systems prior to explicit instruction about writing.
Children’s Knowledge of the Surface Features of Words Exposure to print helps children become familiar with the general features of written language, but this does not always result in knowledge about the specific units of print, such as words and letters. Parents often report that 3- and 4-year-olds can recognize a number of words in their environments, such as the names on labels of products. Parents may take this as reading, but children’s ability to recognize these words depends heavily on context, such as the color of the label, the presence of pictures or graphics, or the presence of the item denoted. Thus, children are unlikely to notice when letters of these words have been altered, and they have difficulty identifying the words outside their usual contexts (Masonheimer, Drum, & Ehri, 1984). Evidence that children’s knowledge of writing can be restricted to general features of print has led Tolchinsky (2003) to argue that children’s understanding of written language moves from general to more specific features through a process of differentiation. Children first learn about the features of written language that are common to all languages – linearity, lack of iconicity, and so on – and only later learn about features that are specific to the particular written language to which they are exposed, such as directionality. This hypothesis has been challenged by recent findings that suggest that, at least for certain types of words, children show a preference for certain language-specific features quite early, earlier than they show preferences for certain general features (Treiman, Cohen, Mulqueeny, Kessler, & Schechtman, 2007). These findings were obtained with words that are highly salient to young children: their own names. Three- and four-year-olds exposed to English were especially sensitive to the shapes of the letters at the beginnings of their names, suggesting that they pay special attention to the beginnings of words (the leftmost letter in the case of English) and that they know about the properties of the letters in at least some specific words. For example, a child named Brendan would often reject a misspelling of his name even when the first letter was of a similar shape to the one actually in his name (e.g., PRENDAN), showing good knowledge of the initial letter shape. Children were less knowledgeable about the exact shapes of their names’ medial and final letters, so that Brendan sometimes accepted BREMDAN as a correct rendition of his name, while rejecting BREXDAN. The results from Treiman, Cohen et al. (2007) further suggest that we must consider children’s early opportunities for learning about print when assessing their knowledge about writing. Children have many opportunities to learn about and recognize their own names at home and at preschool, and this knowledge plays an
2 Learning About Writing Begins Informally
21
important role in learning to read and write (Levin & Aram, 2004; Treiman, Kessler, & Pollo, 2006). Indeed, a survey of the parents of U.S. preschool children found that 70% of the words that 3- and 4-year-old children knew were personal names (Treiman, Cohen, et al., 2007). Only 15% of the words that parents reported their children as knowing came from a list of preprimer and primer words (e.g., me, help, dog, stop) that commonly appear in children’s books (Harris & Jacobson, 1972) and that are typically used in formal reading instruction with beginners. Attention to the words that young children actually attend to and learn about can help us understand what children know about the written language. The results of Treiman, Cohen, et al. (2007) indicate that, as U.S. children progress through the preschool period, they develop a preference for depictions of their names in which all the letters are capitalized. This result is at first puzzling, as the children’s developing preferences are not aligned with the conventional patterns of English (names written in English have the first letter capitalized, as in Steve). However, upon further reflection, this result provides further support for the claim that children’s exposure to print should be taken into consideration when considering their understanding of printed language. This is because, in the above-mentioned survey by Treiman, Cohen et al. parents reported that they tended to use uppercase letters when writing for their children. The finding that U.S. children prefer words in all-uppercase letters helps to specify the types of early experiences that are the most formative for knowledge about print. Children are exposed to print when adults are not writing for them, such as when parents and children read storybooks. Children’s books in English typically have capital letters only at the beginning of proper names and sentences. Given that children’s preferences for forms like STEVE are compatible with their parents’ writing, and not with the print in books, we can infer that children are more attentive to writing during the former experiences than the latter. This conclusion fits with research showing that young children spend most of their time looking at the pictures, not the print, during storybook reading (Evans & SaintAubin, 2005). It also fits with the research suggesting that writing with parents, with the parents guiding the children, is important in literacy development (Aram, 2006; Aram & Biron, 2004). Children’s attentiveness to their own first names is revealed in their attempts at writing. For example, studies with Hebrew- and Dutch-speaking children have shown that preschoolers are better at writing their own names than other words (Levin, Both-De Vries, Aram, & Bus, 2005). Levin et al. suggest that this reflects the attention that both children and parents pay to the child’s name, and that children’s attention to their own name leads to an understanding of print that they can then use more widely. In the U.S., most children can write their name by 4½ years of age, although they may make errors on some of the letters (Treiman & Broderick, 1998). Name writing appears to correlate highly with other print-related skills, such as letter naming and general print knowledge (Welsch, Sullivan, & Justice, 2003). Although young children often know a good deal about the visual characteristics of their own printed names, their knowledge of their names is rather limited in other
22
S. Robins and R. Treiman
ways. Although the children may recognize a particular set of symbols as standing for their name, they may have no idea about how the string does so. They may not know why that particular sequence of letters composes their name, as opposed to any other, or how the letters in their name are used in other words with similar sounds (Villaume & Wilson, 1989).
Children’s Knowledge of Letters In many societies, preschoolers learn about the letters of the alphabet through activities such as singing alphabet songs, hearing stories with letters as characters, and reading alphabet books (Aram & Levin, 2002). Also, children are encouraged to identify the letter shapes by their names. Letter naming is related to a child’s understanding of written language more generally (Molfese, Beswick, Molnar, & JacobiVessels, 2006). In the U.S., as in a number of other countries that use the Latin script, many of children’s early experiences with letters use uppercase letters (Treiman, Cohen, et al. 2007). Correspondingly, U.S. children are more familiar with the uppercase forms of letters than the lowercase forms (Worden & Boettcher, 1990). During the preschool period, children’s alphabetic knowledge remains focused on the surface features of written language: Children are usually more knowledgeable about the physical shapes of letters and the letters’ names than about other features of letters, such as the sounds that correspond with them (Levin, Shatil-Carmon, & Asif-Rave, 2006; Treiman et al., 2006). The importance of shape is further supported by the finding that many of the errors that children make in identifying letters involve letters with similar shapes (Treiman, Levin, & Kessler, 2007). Attention to shape in the learning of letter names may reflect the role of shape in vocabulary learning more generally (Treiman et al., 2006). However, the focus on identifying shapes and memorizing letter names may distract children from thinking about the function that letters play. Children may not be aware that letters function as symbols for sounds – an important aspect of learning to read and spell. Thus, although children often come to recognize the names and shapes of the symbols of their written language from their early informal experiences, it may be some time before they put this knowledge toward the learning of the spelling–sound correspondences that are required to read and write.
Deeper Features of Language Children may begin to learn about the deeper nature of written language through informal conversations with adults. To understand the deeper nature of written language, children need to understand that written language is glottographic (Sampson, 1985). That is, writing is based on the sounds of speech. Children must learn that printed language symbolizes something other than itself, and in particular
2 Learning About Writing Begins Informally
23
that it symbolizes spoken language, which is itself a symbol. Writing thus differs from drawing, which symbolizes objects directly. A picture looks like what it represents: A drawing of a dog bears a surface resemblance to the dog it depicts. The written word dog, in contrast, symbolizes a spoken word in English, which in turn represents a dog. The printed word dog does not look like a dog; the features of the word do not correspond to features of the object. To go beyond their understanding of the surface features of print – to learn how to read and write – children need to understand the deeper nature of written language. Young children appear to lack an understanding of how meaning is conveyed in print. One common misunderstanding is to think of writing as a form of pictorial representation. That is, young children often assume that words get their meaning through symbolizing features of the object directly. Although children may distinguish writing and drawing on the basis of their visual features, as the research mentioned above suggests (e.g., Levin & Bus, 2003), children appear to think that the manner of representation is similar across these two symbol systems. Research suggests that children think that several features of the object referred to, especially size, are part of the way in which it is symbolized in print. For example, young children sometimes assume that the length of a printed word should relate to the size of the object that the word represents (Bialystok, 1991). When shown the written words cat and caterpillar, English-speaking children often presume that the larger word, caterpillar, goes with the larger object, a cat. Swedish-speaking children have been reported to do the same (Lundberg & Tornéus, 1978). Anecdotal reports suggest that some young children believe that this length–size correspondence should be reflected in their own names as well, requesting that their names be written with more letters when they get older (Ferreiro & Teberosky, 1982). Children’s tendency to prefer more letters for larger objects also appears in their own attempts at writing. When attempting to write two words, say snake and butterfly, Hebrew-speaking children reportedly sometimes use more symbols, or space their symbols farther apart, when writing snake because snakes are longer than butterflies (Levin & Tolchinksy Landsmann, 1989). Similarly, an Italian 5-year-old wrote cane (dog) and cagnolino (puppy) with the same three letters, but made the letters smaller in the latter case (Pontecorvo, 1985). Hebrew-speaking 4- and 5-year-olds have also been reported to use more letters when writing words that denote a collection of objects, such as forest, than words that denote a single object, such as tree (Levin & Korat, 1993). Some reports suggest that children may attempt to symbolize the color, location, and number of objects in their attempts at writing. For example, children may use a green crayon to write cucumber and a red one to write tomato (Levin & Tolchinksy Landsmann, 1989). Children’s attempts at writing may also show a tendency for correspondence between the object and its location on the page. For example, when given a blank sheet of paper, some children will write grass at the bottom and bird at the top (Levin & Bus, 2003). Homer and Olson (1999), surveying children’s developing ability to write different types of phrases, showed that some children thought that the number of objects being discussed should be directly reflected in the words used to symbolize those objects. They asked children to write two dogs,
24
S. Robins and R. Treiman
red dogs, and no dogs. In response, some children made two similar symbols to represent two dogs and nothing for no dogs, insisting that no dogs cannot be written. Also, when the researchers showed children the printed phrase two dogs, and then covered up the word two, some children reported that the writing said one dog. Although many of these reports are anecdotal, collectively they suggest that children’s early attempts at writing are influenced by thinking of writing as a form of pictorial symbolization.These findings are particularly interesting given the previously discussed research suggesting that children make some differentiations between writing and drawing in recognition and production. Together, these results indicate that children recognize a distinction between these systems but do not understand the underlying reasons for it. Children who see that there is no straightforward mapping between features of objects and the written words that symbolize them are left without a way of making sense of writing. They may default to thinking of writing as some kind of system of pictorial symbolization because drawing is a symbol system that is more familiar to them. Indeed, research has shown that, by the age of three, children possess a general understanding of drawings as intentional objects (Gelman & Ebeling, 1998; Bloom & Markson, 1998). Young children also appear to have difficulty understanding that printed words have stable meanings. Children often take cues for the print’s meaning from objects near the print. Thus, in studies where a puppet moves a card with a printed word from being near one object to being near another in a seemingly accidental way, 3-year-old children often change their reports of what the word says to reflect the object closest to the card (Bialystok, 1991, 2000; Bialystok, Shenfield, & Codd, 2000). For example, children report that a card with the word doll says doll when closest to the picture of a doll, but then report that the word says cat when placed under the picture of a cat. Children do not make the same mistake as often with pictures, even when the pictures do not look like the objects in question. Children appear to use the intention of the person who drew the picture to fix the picture’s meaning (Apperly, Williams, & Williams, 2004). Children do not use the same method to determine the meaning of words. Even when an adult writes the word on the card while the child watches, presumably conveying the intention to write a particular word, the child often assumes that the word says different things depending on the object to which it is closest (Bialystok & Martin, 2003). How do children come to understand that words symbolize speech? Bialystok (1992) argued that children go through a lengthy developmental process, taking place in three stages, to reach the appropriate understanding of the nature of written language. The first stage involves rote memorization of the alphabetic sequence, as done through singing the alphabet song. This ability to recite the alphabet does not entail that children can recognize the individual letters. Transition to the second stage is marked by the ability to recognize and produce individual letters. However, even at this stage children lack a full conception of how these basic elements combine to make meaning. Children can recognize d, o, and g without the additional understanding that, when placed together in that order, those letters represent a distinct, stable meaning (much less that the meaning they represent is dog). This further level of understanding appears in the third stage, when children master the
2 Learning About Writing Begins Informally
25
significance of the sequence of letters in a word and how this conveys meaning. However, the research discussed in this chapter suggests that the last stage of this process may need to be refined: young children may not recognize that writing is symbolic and how it is symbolic at the same time. Children appear to understand that writing is distinct from other symbol systems and that it conveys meanings prior to understanding how it does so. How might children begin to learn about the glottographic nature of written language? In a recent study (Robins & Treiman, 2009), we proposed that parent speech about writing helps young children to do this. In various domains, there are patterns in parent speech that serve as cues for how to understand aspects of the world (Bloom & Wynn, 1997; Gelman, 2003). For example, parents’ use of generic noun phrases such as birds lay eggs and dogs have fur may invite children to presume that laying eggs and having fur are properties that refer to the birds and dogs generally, and not just some individual birds and dogs, even though these are not explicit statements about the essential features shared by members of a species. We propose that similar clues about the nature of print may be available in parents’ speech. To test the idea that children begin to learn about the nature of print through implicit clues in everyday conversation with their parents, we conducted a series of analyses using CHILDES, a database containing child language transcripts from many researchers (MacWhinney, 2000). All of the transcripts we analyzed involved parent–child conversations in English, most of them in the U.S. The conversations occurred when the children were between 18 months and 5 years of age, and transcripts from a total of 607 children were included in our analyses. These transcripts were collected by many researchers for a variety of purposes. We found that parents do speak about writing with their children, even when the children are quite young. Given this, we explored the specifics of parent speech about writing to better understand the types of information that children receive at home. We began by examining the way in which parents speak about speech and writing as similar. For example, a parent may use the word say to refer to both language systems (e.g., What did you say? or What does that sign say?). The words name and word can also be used to refer to units of the written language, much as they can be used to refer to units of the spoken language (e.g., I just wrote my name and that’s not a nice word to use). Using say to refer to both speech and writing may help children to realize that the two systems symbolize the same thing (namely, language). Further, understanding that the same items – names and words – can be objects of speech and print may guide children toward understanding print’s glottographic nature. From the earliest age we examined, parents use say to refer to spoken and written language and talk about both names and words as objects of writing. However, parents tend to focus on names as objects of writing, which may be related to other findings, discussed above, that children’s own names play an important role in their early understanding of print (Levin & Aram, 2004; Treiman et al., 2006, 2007). In learning about print, children need to understand how written language symbolizes spoken language. Print does not represent the ideas of spoken language in
26
S. Robins and R. Treiman
an arbitrary or loose fashion. Rather, written language represents spoken language precisely, symbolizing specific units of speech. The same three letters, in the same order, must be used every time dog is written, whereas pictures of dogs may vary in how they depict dogs. Children need to know that writing and drawing are different in this way. Parents could help children do this by making syntactic distinctions in their discussions of the two systems. In English, the verbs draw, spell, and write differ in whether they require a determiner when taking concrete, singular terms as their objects. Draw does; write and spell do not. For example, it is appropriate to say draw a kitty but not spell a kitty or write a kitty (unless one is specifically asking a person to write both the word a and the word kitty). We found that parents almost always use a determiner with draw and almost never with write or spell. This result is particularly interesting given that children’s writings and drawings look rather similar during these early years and that children themselves often confuse the two systems, saying things like draw my letters. The syntactic distinction in parents’ speech could potentially help children to understand that the two symbol systems are different, prompting the children to investigate further how the written language works. Of course, children can learn to differentiate these two systems without a distinction in determiner use, given that there are a number of languages that lack this distinction, including Hebrew. However, in languages where such a distinction exists, it may serve as a clue for children to begin differentiating print from pictures. These results suggest that, as in many other domains, parent speech provides information that children can use in understanding the world around them. A good deal of information about the written language is available to children in literate societies prior to their receiving formal instruction. Learning about the deeper features of writing may also begin informally.
Conclusions and Implications We have discussed various ways in which children can learn about writing prior to receiving formal instruction in school. Although explicit instruction about spelling– sound relationships may be required for children to master reading and spelling, children learn a great deal about how writing works through informal interactions with their parents and other adults. These interactions take a variety of forms. Through exposure to print, children become aware of a number of its salient surface features. For words that are particularly important to young children, such as their own names, children may even learn about language-specific properties from these experiences. Other activities, such as singing songs that include the letters of the alphabet and watching their parents write letters, teach children about the names and shapes of letters. All these experiences teach children about the surface features of writing. Children may also begin to learn about the deeper features of writing through informal interactions with adults. We have discussed preliminary research that suggests that children could potentially gain information about the symbolic
2 Learning About Writing Begins Informally
27
function of writing through the conversations in which they engage at home. Through implicit cues in parent speech, children could begin to understand that writing works differently than drawing, and even that it symbolizes speech, before learning how it is that writing does so. Thus, a host of information about the written language is available to children before formal schooling begins. Although formal instruction is required for many of the particular spelling–sound relations, early informal experiences may provide the foundation for this further mastery. The research we have discussed has focused on children in literate societies. These children are similar to one another in that written language is a prominent feature of their environments. However, there may be important differences between the print environments of young children, as in how much time their parents and preschool teachers spend writing with and for them. Understanding these individual differences, and how they affect preschoolers’ knowledge of print, is an important topic for future research. Previous studies have tended to focus on how often parents read books to their young children as an important determinant of later literacy skill. However, given that young children do not spend much time looking at the print in books and do not get many of their ideas about print from books, book reading may not be as important for print knowledge as was previously thought. By examining the other opportunities that children have to learn about print in homes and preschools, and how these opportunities may differ across groups, we can work to better prepare young children for formal instruction in reading and writing. Acknowledgment Work of this chapter was supported in part by NIH grant HD051610.
References Aram, D. (2006). Early literacy interventions: the relative roles of storybook reading, alphabetic activities, and their combination. Reading and Writing: An Interdisciplinary Journal, 19(5), 489–515. Aram, D., & Biron, S. (2004). Joint storybook reading and joint writing interventions among low SES preschoolers: differential contributions to early literacy. Early Childhood Research Quarterly, 19, 588–610. Aram, D., & Levin, I. (2002). Mother-child joint writing and storybook reading: relation with literacy among low SES kindergartners. Merrill-Palmer Quarterly, 48, 202–224. Apperly, I. A., Williams, E., & Williams, J. (2004). Three- to four-year-olds’ recognition that symbols have a stable meaning: pictures are understood before written words. Child Development, 75, 1510–1522. Bialystok, E. (1991). Letters, sounds, and symbols: changes in children’s understanding of written language. Applied Psycholinguistics, 12, 75–89. Bialystok, E. (1992). Symbolic representation of letters and numbers. Cognitive Development, 7, 301–316. Bialystok, E. (2000). Symbolic representation across domains in preschool children. Journal of Experimental Child Psychology, 76, 173–189. Bialystok, E., & Martin, M. M. (2003). Notation to symbol: development in children’s understanding of print. Journal of Experimental Child Psychology, 86, 223–243.
28
S. Robins and R. Treiman
Bialystok, E., Shenfield, T., & Codd, J. (2000). Languages, scripts, and the environment: factors in developing concepts of print. Developmental Psychology, 36, 66–76. Bloom, P., & Markson, L. (1998). Intention and analogy in children’s naming of pictorial representations. Psychological Science, 9, 200–204. Bloom, P., & Wynn, K. (1997). Linguistic cues in the acquisition of number words. Journal of Child Language, 24, 511–533. Brenneman, K., Massey, C., Machado, S. F., & Gelman, R. (1996). Young children’s plans differ for writing and drawing. Cognitive Development, 11, 397–419. Evans, M. A., & Saint-Aubin, J. (2005). What children are looking at during shared storybook reading: evidence from eye movement monitoring. Psychological Science, 16, 913–920. Ferreiro, E., & Teberosky, A. (1982). Literacy before schooling. New York: Heinemann. Ganopole, S. J. (1987). The development of word consciousness prior to first grade. Journal of Reading Behavior, 19, 415–436. Gelman, S. A. (2003). The essential child. New York: Oxford University Press. Gelman, S. A., & Ebeling, K. S. (1998). Shape and representational status in children’s early naming. Cognition, 66, B35–B47. Harris, A. J., & Jacobson, M. D. (1972). Basic elementary reading vocabulary. London: Macmillan. Homer, B. D., & Olson, D. R. (1999). The role of literacy in children’s concept of word. Written Language and Literacy, 2, 113–140. Lavine, L. O. (1977). Differentiation of letterlike forms in prereading children. Developmental Psychology, 13, 89–94. Levin, I., & Aram, D. (2004). Children’s names contribute to early literacy: a linguistic and a social perspective. In D. Ravid & H. Bat-Zeev Shyldkrot (Eds.), Perspectives on language and language development (pp. 219–239). Dordrecht, The Netherlands: Kluwer. Levin, I., Both-De Vries, A., Aram, D., & Bus, A. (2005). Writing starts with own name writing: from scribbling to conventional spelling in Israeli and Dutch children. Applied Psycholinguistics, 26, 463–477. Levin, I., & Bus, A. (2003). How is emergent writing based on drawing? Analyses of children’s products and their sorting by children and mothers. Developmental Psychology, 39, 891–905. Levin, I., & Korat, O. (1993). Sensitivity to phonological, morphological, and semantic cues in early reading and writing in Hebrew. Merrill-Palmer Quarterly, 39, 213–232. Levin, I., Shatil-Carmon, S., & Asif-Rave, O. (2006). Learning of letter names and sounds and their contribution to word recognition. Journal of Experimental Child Psychology, 93, 139–165. Levin, I., & Tolchinksy Landsmann, L. (1989). Becoming literate: referential and phonetic strategies in early reading and writing. International Journal of Behavioral Development, 12, 369–384. Levy, B. A., Gong, Z., Hessels, S., Evans, M. A., & Jared, D. (2006). Understanding print: early development and the contributions of home literacy experiences. Journal of Experimental Child Psychology, 93, 63–93. Lundberg, I., & Tornéus, M. (1978). Nonreaders’ awareness of the basic relationship between spoken and written words. Journal of Experimental Child Psychology, 25, 404–412. MacWhinney, B. (2000). The CHILDES project: tools for analyzing talk (3rd ed.). Mahwah, NJ: Erlbaum. Masonheimer, P. E., Drum, P. A., & Ehri, L. C. (1984). Does environmental print identification lead children into word reading? Journal of Reading Behavior, 16, 257–271. Molfese, V. J., Beswick, J., Molnar, A., & Jacobi-Bessels, J. (2006). Alphabetic skills in preschool: a preliminary study of letter naming and letter writing. Developmental Neuropsychology, 29, 5–19. Pontecorvo, C. (1985). Figure, parole, numeri: un problema di simbolizzazione. [Pictures, words, numbers: a problem of symbolization]. Età Evolutiva, 22, 5–33. Robins, S. & Treiman, R. (2009). Talking about writing: What we can learn from conversations between parents and their young children. Applied Psycholinguistics, 30, 463–484.
2 Learning About Writing Begins Informally
29
Sampson, G. (1985). Writing systems. Stanford, CA: Stanford University Press. Snow, C. E., & Juel, C. (2005). Teaching children to read: what do we know about how to do it? In M. J. Snowling & C. Hulme (Eds.), Science of reading: a handbook (pp. 501–520). Malden, MA: Blackwell. Sulzby, E. (1985). Children’s emergent reading of favorite storybooks: a developmental study. Reading Research Quarterly, 20, 458–481. Sulzby, E. (1996). Roles of oral and written language as children approach conventional literacy. In C. Pontecorvo, M. Orsolini, B. Burge & L. B. Resnick (Eds.), Children’s early text construction (pp. 25–46). Hillsdale, NJ: Erlbaum. Tolchinsky, L. (2003). The cradle of culture and what children know about writing and numbers before being taught. Mahwah, NJ: Erlbaum. Treiman, R., & Broderick, V. (1998). What’s in a name: children’s knowledge about the letters in their own names. Journal of Experimental Child Psychology, 70, 97–116. Treiman, R., Cohen, J., Mulqueeny, K., Kessler, B., & Schechtman, S. (2007). Young children’s knowledge about printed names. Child Development, 78, 1458–1471. Treiman, R., Kessler, B., & Pollo, T. (2006). Learning about the letter name subset of the vocabulary: evidence from U.S. and Brazilian preschoolers. Applied Psycholinguistics, 27, 211–227. Treiman, R., Levin, I., & Kessler, B. (2007). Learning of letter names follows similar principles across languages: evidence from Hebrew. Journal of Experimental Child Psychology, 96, 87–106. Villaume, S. K., & Wilson, L. C. (1989). Preschool children’s explorations of letters in their own names. Applied Psycholinguistics, 10, 283–300. Welsch, J. G., Sullivan, A., & Justice, L. M. (2003). That’s my letter!: what preschoolers’ name writing tells us about emergent literacy knowledge. Journal of Literacy Research, 35, 757–776. Worden, P. E., & Boettcher, W. (1990). Young children’s acquisition of alphabet knowledge. Journal of Reading Behavior, 22, 277–295. Yamagata, K. (2007). Differential emergence of representational systems: drawings, letters, and numerals. Cognitive Development, 22, 244–257.
Chapter 3
The Role of Letter Name Knowledge in Early Spelling Development: Evidence from Brazilian Portuguese Cláudia Cardoso-Martins and Marcela Fulanete Corrêa
Abstract This chapter describes the results of studies investigating early spelling development among Portuguese-speaking individuals in Brazil. Similar to what has been found for English- and Hebrew-speaking children, we present evidence that 1) non-literate speakers of Brazilian Portuguese – both children and adults – use their knowledge of letter names to connect print to speech; and 2) reliance on letter names in beginning spelling development in Brazilian Portuguese accounts for the so-called syllabic spellings, i.e., spellings in which letters stand for whole syllables in the pronunciation of words. Taken together, these findings strongly suggest that an adequate theory of early spelling development in Brazilian Portuguese has to grant a major role for children’s knowledge of letter names and sounds. We argue that Ehri’s phase theory, developed on the basis of her observations of Englishspeaking children’s spellings, can be successfully extended to the case of early spelling development in Brazilian Portuguese.
Introduction Research in the past few years has underscored the importance of letter names in early literacy acquisition. In particular, Treiman and her colleagues (e.g., Treiman & Kessler, 2003; Treiman, Tincoff, & Richmond-Welty, 1996) have suggested that knowledge of letter names helps children to connect print and speech (Treiman & Kessler, 2003; Treiman et al., 1996). For example, Treiman et al. (1996) have shown that young, preschool children are more likely to spell the initial consonant sound in words when the entire letter name is clearly detected in the word’s pronunciation (as in beach) than when it is not (as in bone). The work conducted by Professor Iris Levin (Levin, 2007; Levin, Patel, Margalit & Barad, 2002) has extended Treiman’s work in three important ways. First, it has shown that the same phenomenon is true of Hebrew-speaking children, even though letter C. Cardoso-Martins (*) Universidade Federal de Minas Gerais, Belo Horizonte, MG, Brazil e-mail:
[email protected]
D. Aram and O. Korat (eds.), Literacy Development and Enhancement Across Orthographies and Cultures, Literacy Studies 101, DOI 10.1007/978-1-4419-0834-6_3, © Springer Science+Business Media, LLC 2010
31
32
C. Cardoso-Martins and M.F. Corrêa
names in Hebrew are longer than in English, and, as a result, less often heard in the pronunciation of words. Second and, perhaps more importantly, it has shown that young children are able to use their knowledge of letter names to connect print to speech, even when only part of the letter name can be detected in the pronunciation of the word. Finally, Professor Levin’s work has shown that reliance on the knowledge of letter name is particularly strong among children with low levels of phonological awareness. Our work about the development of spelling skills among Brazilian Portuguesespeaking children has been greatly influenced by Professor Levin’s work, and has benefited enormously from her generous suggestions and comments over the years. It is thus a great honor to describe part of this work in a book dedicated to her. This chapter begins with a brief demonstration of the similarity between what has been found for English- and Hebrew-speaking children, and non-literate speakers of Brazilian Portuguese –children as well as adults – both take advantage of their knowledge of letter names to connect print to speech. We then discuss the implications of these findings for a theory of the development of early spelling skills. In particular, we argue that an adequate theory of early spelling development in Brazilian Portuguese has to grant a major role for children’s knowledge of letter names and sounds.
Letter Name Knowledge and Spelling Skills in Brazilian Portuguese: Evidence from Non-literate Children and Adults There is evidence that letter names in general, and vowel names in particular, are more abundant in the pronunciation of words in Portuguese than in English (CardosoMartins, Resende, & Rodrigues, 2002; Pollo, Kessler, & Treiman, 2005). For example, Cardoso-Martins et al. (2002) observed that while only 17 of the 56 regular and irregular words in Seymour’s (1986; list 2) list of high frequency English words contained letters whose names can be heard in the pronunciation of the words, 51 of the 56 most frequent words in Brazilian books for kindergarten children (Pinheiro, 1996) contained at least one letter whose entire name can be heard in the pronunciation of the word. Brazilian-Portuguese speaking children should therefore be able to take advantage of their knowledge of letter names to connect print to speech. Cardoso-Martins and Batista (2005) have indeed shown that this is the case. These researchers asked Brazilian Portuguese-speaking children to spell pairs of words beginning with the same consonant letter and sound. For each pair, the entire name of the beginning consonant could be detected in the pronunciation of the experimental word (for example, the word telefone ‘telephone’ /tele’foni/ in which the name of the letter t (/te/) can be heard in the beginning of the letter), but not in the pronunciation of the control word (for example, the letter t in the beginning of the word tartaruga ‘turtle’ / tahta’ruga/ stands for the sound /t/ which is contained in the name of the letter). Twenty-five children varying from 4 to 5 years of age participated in the study. None of them had started to read at the occasion of the study. The children were asked to spell the words as well as they could, but only the initial letters in the children’s spellings were considered in the analyses reported below.
3 The Role of Letter Name Knowledge in Early Spelling Development
33
As expected, the children wrote the correct consonant significantly more often for the experimental words than for the control words. In contrast, the beginning letter in the children’s spellings more often corresponded to the first vowel in the word for the control than for the experimental words. As Cardoso-Martins and Batista noted, this is easily understood when we take into account that the entire name of the first vowel in the word could be clearly detected in the pronunciation of the control words (e.g., the name of the letter a, /a/, can be clearly heard in the pronunciation of the first syllable of the word /tahta’ruga/, i.e., /tah/) but not in the pronunciation of the experimental word (e.g., the name of the letter e, i.e., /e/ does not appear in the pronunciation of the word /tele’foni/, neither in the first syllable /te/, nor in the second syllable /le/). An interesting question is whether the use of letter names as a bridge between spoken and written words is a developmental phenomenon, typical only of children, or whether it is a strategy used by pre-readers in general, regardless of their age. To address this question, Corrêa, Cardoso-Martins and Rodrigues (in press) analyzed the spellings produced by a group of low SES Brazilian adults who were participating in a study investigating the strategies nonliterate adults use to interpret print (see Cardoso-Martins, Rodrigues, & Ehri, 2003). Only the 19 adults who had not started to read or to spell alphabetically were included in the analyses reported below. At the time of the study, they ranged in age from 20 to 74 years. Participants were asked to spell 14 words. For half of the words, the name or part of the name of the initial consonant could be clearly heard in the pronunciation of the word. For the other half, only a phoneme contained in the name of the initial consonant could be detected in the pronunciation of the word. Similar to what has been found for preliterate children, spellings of the words’ beginning with a consonant were more often correct when the consonant name could be clearly heard in the pronunciation of the word than when only the phoneme typically represented by the consonant letter was heard in the beginning of the word’s pronunciation. These results suggest that knowledge of the letter name is a key factor contributing to the understanding of the connection between print and speech for adults, as it is for children.
Two Alternative Views of Early Spelling Development in Brazilian Portuguese According to one prominent theory of spelling development (Ferreiro, 1989; 1990, Ferreiro & Gomez Palacio, 1982), children go through three main stages in their attempt to understand their writing system. In Brazil, these stages are often called pre-syllabic, syllabic, and alphabetic. During the pre-syllabic stage, children are oblivious to the fact that print represents sounds. They are nonetheless aware of the representative nature of print, and spend a lot of cognitive effort trying to pinpoint its distinctive features. As a matter of fact, one of Ferreiro’s (1989, 1990) most famous principles, the principle of “minimum
34
C. Cardoso-Martins and M.F. Corrêa
quantity” (according to which a spelling needs to have a minimum number of letters, usually three or four, to qualify as a word), develops during this stage. Having constructed this and other principles, children begin to search for principles that enable them to distinguish the spellings of different words. Sooner or later, this search leads the way towards the discovery that print represents speech. Initially, however, children believe that the letters they see in print stand for whole syllables in the pronunciation of words. As a consequence, they spell one and only one letter for each syllable in the word, writing, for example, the letters XYS for the three-syllable Spanish word caballo ‘horse’ (Ferreiro & Gomez Palacio, 1982). This hypothesis, known as the syllabic hypothesis, is destabilized by the spellings children see in the environment, and eventually it gives way to the alphabetic stage. Now letters are conceptualized as what they really are: symbols for individual phonemes. One major problem of Ferreiro’s theory is that it does not take into account the widespread finding that children’s knowledge of letter names and/or sounds is instrumental in children’s discovery of the relation between print and speech. As a matter of fact, Ferreiro has systematically downplayed the role played by the knowledge of the letter in the discovery of that connection. For example, Ferreiro and Teberosky (1986) suggested that syllabic spellings are observed even among children who don’t know any letters and, as a result, draw little circles or dashes when requested to write as well as they can. In contrast to Ferreiro’s psychogenetic explanation, Cardoso-Martins and Batista (2005) have argued that syllabic spellings may be more appropriately described as an incidental result of young children’s incipient understanding that letters represent sounds, and from their attempt to represent the sounds they can detect in the pronunciation of words. In Spanish, as well as in other languages in which syllabic spellings have been described (for example, Portuguese and Italian), those sounds often correspond to vowel names. Since by definition there is only one vowel per syllable, young children’s reliance on letter names to spell words will sometimes result in syllabic spellings. In support of their hypothesis, Cardoso-Martins and Batista (2005) observed a strong correlation between the incidence of syllabic spellings and young children’s incipient ability to represent sounds in the pronunciation of words by phonetically appropriate letters, that is, letters that typically represent the target sounds in the writing system, even if not in the particular word under consideration (e.g., the letter u in the syllabic spelling DU for the word dedo ‘finger’ /’dedu/ is a phonetically appropriate representation of the sound /u/ at the end of the word, since it is often used to represent this sound in other words). Furthermore, they observed clear signs of the understanding that speech represents sounds in the so-called pre-syllabic spellings, that is, spellings that are not syllabic or alphabetic. For example, the first letter in those spellings often corresponded to the first sound in the pronunciation of the word, particularly when that sound corresponded to a letter name (e.g., TAP for telephone ‘telephone’ /tele’foni/). On the basis of these findings, Cardoso-Martins and Batista (2005) suggested that the development of Brazilian Portuguese-speaking children’s spelling skills might be more aptly described in terms of children’s increasing ability to map sounds in the pronunciation of words with phonetically appropriate orthographic
3 The Role of Letter Name Knowledge in Early Spelling Development
35
units. Ehri (1992, 1998) has provided such an account of the development of English-speaking children’s ability to read and spell words. In Ehri’s view, children progress through four phases, each characterized by a predominant, although by no means exclusive, process: (1) pre-alphabetic, (2) partial alphabetic, (3) alphabetic or full alphabetic, and (4) consolidated alphabetic. Initially, when children do not know the names and sounds of the letters, they produce more or less arbitrary letters, which do not bear any relationship to the sounds in the pronunciation of the words. As children learn about letter names and sounds, they begin to understand the sound symbolizing function of letters in spellings. At the dawn of this understanding, children are able to represent only a few sounds in the word, generally a sound in the beginning or a sound in the beginning and at the end of the word, for example, the spelling JL for the word jail (Ehri & Wilce, 1985). In addition, as this example shows, the sounds children initially are able to represent quite often correspond to letter names in the pronunciation of words. Ehri has called this type of spelling partial alphabetic to distinguish it from the full graphophonemic processing that is characteristic of the full alphabetic phase. For most children, explicit instruction in print–speech correspondences at the sublexical level seems necessary for movement into the full alphabetic phase. Children are now able to represent all sounds in the pronunciation of the words by phonetically appropriate letters. In the final phase, i.e. the consolidated alphabetic phase, children are able to operate with multiletter units that correspond to syllables or parts of syllables in words (for example, the letter sequence “ight” to represent the phoneme blend /ait/ in words like light, fight, night). In what follows, we describe the results of a short-term longitudinal study conducted with the aim of testing Cardoso-Martins and Batista’s (2005) proposal. Specifically, the study tested the hypothesis that Ehri’s partial alphabetic phase might provide a better explanation of early spelling development in Brazilian Portuguese than Ferreiro’s syllabic stage.
A Longitudinal Study of Early Spelling Development in Brazilian Portuguese One hundred and twenty-four children enrolled in preschool classrooms in private and public schools of a major Brazilian city were recruited to participate in the study. At the beginning of the study, they ranged in age from 4 to 7 years of age, and the immense majority had not started to read. Specifically, except for 38 children, who read between one and three words, as many as 86 children could not read any words in a list of 30 words frequently found in books for children (Pinheiro, 1996). Children’s ability to spell words was tested at two different times, with an interval of approximately 6 months. Only words commonly observed in children’s vocabularies (e.g., flor ‘flower’ /floh/, dedo ‘finger’ /’dedu/, cavalo ‘horse’ /ka’valu/, bicicleta ‘bike’ /bisi’kleta/) were used. Words were dictated by the examiner one at a time. We told the children that it was okay for them to make mistakes but that they
36
C. Cardoso-Martins and M.F. Corrêa
should try to write the words as well as they could. Twelve words were dictated the first time: three monosyllabic, three two-syllable, three three-syllable, and three four-syllable words. In the second round, we excluded one monosyllabic word from the list, and included one new word in each of the other syllable sizes. This procedure was followed in an attempt to increase the number of syllabic spellings, since the principle of “minimum quantity” (see p. 5) might work against children’s proclivity to spell monosyllabic words syllabically.
Early Development of Spelling Skills: Ferreiro’s Stage Model To assess Ferreiro’s model, children’s spellings at Times 1 and 2 were classified as pre-syllabic, syllabic, or alphabetic. A child was classified in one of these levels when more than half of his productions corresponded to spellings at that level. In the few cases where this criterion could not be used, the child was assigned to the level containing more spellings. An exception to this rule occurred when half of a child’s spellings had been coded at one level and half at another. In these cases, the child was assigned to the less mature level. The pre-syllabic level consisted of spellings that could not be coded either as syllabic or as alphabetic. Examples include AAO for bicicleta ‘bike’ /bisi’kleta/, ACOT for dedo ‘finger’ /’dedu/, and RAINA for telefone ‘telephone’ /tele’foni/. Syllabic spellings were spellings in which the number of letters corresponded to the number of syllables in the word. The spellings MNA for barata ‘roach’ /ba’rata/, COA for cigarro ‘cigarette’ /si’gahu/, and IH for bico ‘beak’ /’biku/, are examples of spellings coded as syllabic. Alphabetic spellings consisted of spellings in which all sounds in the pronunciation of the word were represented by a phonetically appropriate, even if not conventional, letter. This was the case of the spellings DEDU for dedo ‘finger’ /’dedu/, TELEFONI for telefone ‘telephone’ /tele’foni/, BIQO for bico ‘beak’ /’biku/, and SIGARO for cigarro ‘cigarette’ /si’gahu/. Spellings in which the number of sounds represented by a phonetically appropriate letter was greater than the number of sounds not represented or misrepresented were also coded as alphabetic (e.g., TATARUGA for tartaruga ‘turtle’ /tahta’ruga/, BICIQUETA for bicicleta ‘bike’ / bisi’kleta/, and CVALO for cavalo ‘horse’ /ka’valu/). In all cases, a letter was considered phonetically appropriate if it represented the target sound in other words, even if only in other orthographic contexts. Figure 3.1 illustrates the number of children classified in each of the three levels, separately for each time. The figure also illustrates the changes that occurred between Time 1 and Time 2. Horizontal arrows indicate the number of children that remained in the same level and oblique arrows indicate the number of children that changed levels from one time to the other. Descending arrows represent changes to more advanced levels; ascending arrows, on the other hand, indicate regressions, that is, changes to less advanced spelling levels. Bold arrows were used to depict unexpected changes.
3 The Role of Letter Name Knowledge in Early Spelling Development Time 1
Pre-syllabic
115
37 Time 2
79
81 2
7 Syllabic
1
5
8 29
2
Alphabetic
4
4
N = 124
35
N = 124
Fig. 3.1 Spelling development according to Ferreiro’s Stage Model
As illustrated in Fig. 3.1, relatively few children were classified as syllabic: one child was classified as syllabic both the first and the second time, and four and seven children were classified as syllabic the first and the second time, respectively. Furthermore, many children seemed to skip the syllabic stage. Specifically, of the 36 children classified as pre-syllabic at the first time and who progressed towards more advanced levels at the second time, only seven progressed to the syllabic stage. Twenty-nine children moved straight to the alphabetic stage, a difference that is statistically significant according to the binomial test, p < .001(two-tailed).
Early Development of Spelling Skills: Ehri’s Phase Model To assess the appropriateness of Ehri’s model to children learning to spell in Portuguese, children’s spellings were classified according to three levels: prealphabetic, partial alphabetic, and alphabetic. As in the case of Ferreiro’s model, with very few exceptions, a child was assigned to one of these levels when more than half of his or her productions corresponded to spellings at that level. Pre-alphabetic spellings did not reveal any understanding that letters represent sounds. Examples included PEISF for cavalo ‘horse’ /ka’valu/, AFNUL for bico ‘beak’ /’biku/, and EJHDO for bicicleta ‘bike’ /bisi’kleta/.
38
C. Cardoso-Martins and M.F. Corrêa Time 1 PreAlphabetic
82
Ti 42
me 2 44
2 30 Partial Alphabetic
15
38
45
10 21
Alphabetic
4
N = 124
4
35
N = 124
Fig. 3.2 Spelling development according to Ehri’s Model of Literacy Development
Partial alphabetic spellings, on the other hand, showed the beginning of that understanding. A spelling was coded as partial alphabetic if its first letter represented the first consonant or vowel in the word. Examples include DIEZT for dedo ‘finger’ /’dedu/, IZAOBH for cigarro ‘cigarette’ /si’gahu/, and TOA for telefone ‘telephone’ /tele’foni/. As in the case of Ferreiro’s model, alphabetic spellings corresponded to spellings in which the majority of the sounds in the pronunciation of the word were represented by a phonetically appropriate letter. As can be seen in Fig. 3.2., Ehri’s phase model seems to offer a more adequate description of children’s early spelling development than Ferreiro’s stage model. In particular, the partial alphabetic level appeared as an intermediate level between the pre-alphabetic and alphabetic levels much more often than was the case for the syllabic level. Specifically, 68 children were classified at Ehri’s partial alphabetic level: 15 children both the first and the second time, and 53 children at least one time/once. In striking contrast to the results obtained for the syllabic stage, relatively few children seemed to skip the partial alphabetic level. In fact, among the 40 children classified as pre-alphabetic the first time and who progressed to a more advanced level the second time, the number who moved to the partial alphabetic level were significantly more (N = 30) than the number who moved to the alphabetic level (N = 10), binomial test, p < .01(two-tailed).
3 The Role of Letter Name Knowledge in Early Spelling Development
39
Are Syllabic Spellings Partial Alphabetic Spellings? Cardoso-Martins and Batista (2005) suggested that the syllabic spellings observed among young speakers of Spanish, Portuguese and Italian result from children’s reliance on letter names in the early phases of spelling development (Levin et al., 2002; Treiman & Kessler, 2003; Treiman et al., 1996). This hypothesis predicts a strong connection between children’s syllabic spellings and their incipient ability to represent sound-letter relations in spellings. Three analyses were conducted in an attempt to test this prediction. First, we looked at whether children classified as syllabic spellers, according to Ferreiro’s model, had also been classified as partial alphabetic spellers, according to Ehri’s model. Only one of the 13 syllabic spellers was coded as pre-alphabetic in Ehri’s classification scheme. All other cases were coded as partial alphabetic spellers, a difference highly significant by the binomial test, p < .001 (two-tailed). Second, if syllabic spellings are an incidental result of young children’s tendency to spell the letters whose sounds they can identify in the pronunciation of words, syllabic spellings should be more prevalent among partial alphabetic children than among pre-alphabetic children. We therefore calculated the mean number of syllabic spellings for the pre-alphabetic and the partial alphabetic children, separately for Times 1 and 2. As predicted, children classified as partial alphabetic tended to produce a greater number of syllabic spellings than children classified as pre-alphabetic, t(118) = 2.97, p <. 01(two-tailed), for the first time and t(87) = 1.89, p < .10 (two-tailed), for the second. Finally, we coded all letters in the syllabic spellings of the syllabic children as arbitrary or phonetically appropriate. A letter was coded as arbitrary if it did not bear any relationship to a sound in the syllable it stood for. For example, the letter p in the spelling APA for barata ‘roach’ /ba’rata/ was coded as arbitrary since it never stands for either the consonant or the vowel sound in the syllable /ra/. Phonetically appropriate letters were coded as either letter name or letter sound. A letter was coded as a letter name when its entire name could be heard in the pronunciation of the word (for example, both the letter d and the letter u in the spelling DU for dedo ‘finger’ /¢dedu/ were coded as letter names since their names can be heard in the pronunciation of the word, the name of the letter d, in the first syllable and the name of the letter u, in the last). The remaining phonetic letters were coded as letter sounds. Most of the letter sounds stood for a phoneme contained in the name of the letter (e.g., the letter t in the spelling TOA for telephone ‘telephone’ /tele’foni/, apparently stands for the phoneme /t/, which can be detected in the name of the letter t). Exceptions to this pattern include the letters e and o to represent the phonemes /e/ and /o/, respectively, which are not contained in the names of the letters (/e/ and /ɔ/, respectively). The majority of the letters (Mean proportion = .80) in the syllabic spellings of the children assigned to the syllabic stage were phonetically appropriate rather than arbitrary letters. As a matter of fact, only one child wrote more arbitrary letters than phonetically appropriate letters, a difference significant by the binomial test, p < .01
40
C. Cardoso-Martins and M.F. Corrêa
(two-tailed). Letter names, particularly vowel names, occurred more often than letter sounds. This was the case for all syllabic spellers.
Discussion and Conclusions The results of our longitudinal study raise questions for Ferreiro’s psychogenetic account of early spelling development. Instead, they suggest that young children’s reliance on letter names to spell words early on may offer a more parsimonious explanation for the so-called syllabic spellings than Ferreiro’s conceptual explanation. We have found hardly any evidence that children attend only to the quantitative dimension when they spell syllabically. Indeed, most of the letters in the syllabic spellings of the children assigned to the syllabic stage in the present studies consisted of letters that stood for sounds in the pronunciation of the words. In a great number of cases, the names of the letters could be clearly detected in the pronunciation of the words. In fact, in many cases (e.g., DU for dedu ‘finger’ /’dedu/, AAA, for barata ‘roach’ /ba’rata/, IIEA, for bicicleta ‘bike’ /bisi’kleta/, AAUA, for tartaruga ‘turtle’ /tahta’ruga/, etc.), the names of all letters in a syllabic spelling could be detected in the pronunciation of the target word, each name corresponding to a different syllable in the word (either to the whole syllable or to a sound contained in the syllable). Furthermore, in contrast to Ferreiro’s (e.g., 1989, 1990) claim that the syllabic hypothesis constitutes the first manifestation of the understanding that print maps onto speech and, as such, constitutes a major landmark in the development of spelling, we found that a large number of children who had been classified as pre-syllabic according to Ferreiro’s model were classified as partial alphabetic according to Ehri’s model. Specifically, between 29 and 45% of the pre-syllabic children were assigned to Ehri’s partial alphabetic phase. Together with the evidence that the letters in the syllabic spellings of the children assigned to the syllabic stage corresponded to sounds in the pronunciation of the words, these results strongly suggest that, rather than the result of children’s construal that letters represent whole syllables, young children’s syllabic spellings are an incidental result of their beginning to understand that letters stand for stable sounds in the pronunciation of words. Perhaps more damaging to Ferreiro’s model of early spelling development was the finding of a very low incidence of syllabic spellings. Nor have we found evidence for a more lenient version of the syllabic hypothesis, that is, that children relate the number of letters in the spelling of words to the length of the words’ pronunciation before they are able to process letter-sound relations. In contrast to what would be expected on the basis of this hypothesis, the pre-alphabetic children produced about the same number of letters regardless of the number of syllables in the target words. Taken together, our findings strongly favor Ehri’s (e.g., 1992, 1998) phase model of early spelling development. Ehri’s model indeed offers a more parsimonious explanation of early spelling development than Ferreiro’s stage model. In fact,
3 The Role of Letter Name Knowledge in Early Spelling Development
41
while it seems capable of accommodating data from languages as diverse as English and Brazilian Portuguese, most evidence for Ferreiro’s syllabic stage comes from observations of young speakers of Romance languages (e.g., Grossi, 1990; Pontecorvo & Zucchermaglio, 1990; Weisz, 2000). Ferreiro (1990) suggested that the paucity of syllabic productions among English-speaking children results from the relatively high incidence of monosyllabic words in English. Presumably, this characteristic of the English language prevents young speakers of English from exploring the syllabic hypothesis. It is nonetheless possible that the absence of syllabic spellings among English-speaking children results from the fact that, relative to Romance languages like Brazilian Portuguese, English presents a larger number of vocalic sounds only a minority of which corresponds to vowel names. As mentioned previously, there is indeed evidence that letter names in general, and vowel names in particular, are more abundant in the pronunciation of words in Brazilian Portuguese than in English (Cardoso-Martins et al., 2002; Pollo, Treiman, & Kessler, 2008). As there is only one vowel per syllable, and young children seem to rely on letter names to spell words (e.g., Levin et al., 2002; Treiman et al., 1996), it makes sense that syllabic spellings should be more prevalent in Brazilian Portuguese than in English.
Final Remarks In line with the results of Treiman’s (1996, 2003) and Levin’s (2002) studies, the findings discussed in this chapter suggest that knowledge of letter names also helps speakers of Brazilian Portuguese to connect print to speech. An adequate theory of spelling development in Brazilian Portuguese needs thus to incorporate these findings. The results of our work (see also Cardoso-Martins & Batista, 2005; Cardoso-Martins, Corrêa, Lemos, & Napoleão, 2006) suggest that Ehri’s phase model, developed on the basis of her observations of reading and spelling development in English, may be fruitfully extended to spelling development in more transparent writing systems such as the Portuguese orthography. They also suggest that teaching about letter names should be an essential ingredient of early literacy instruction. Acknowledgments The research described in this chapter was funded by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Brazil. We thank the children and the schools for their cooperation.
References Cardoso-Martins, C., Resende, S. M., & Rodrigues, L. A. (2002). Letter name knowledge and the ability to learn to read by processing letter-phoneme relations in words: evidence from Brazilian Portuguese-speaking children. Reading and Writing, 15, 409–432. Cardoso-Martins, C., Rodrigues, L. A., & Ehri, L. C. (2003). Place of environmental print in reading development: evidence from nonliterate adults. Scientific Studies of Reading, 7, 335–355.
42
C. Cardoso-Martins and M.F. Corrêa
Cardoso-Martins, C., & Batista, A. C. E. (2005). O conhecimento do nome das letras e o desenvolvimento da escrita: evidência de crianças falantes do português. Psicologia: Reflexão e Crítica, 18, 330–336. Cardoso-Martins, C., Corrêa, M. F., Lemos, L. S., & Napoleão, R. F. (2006). Is There a Syllabic Stage in Spelling Development? Evidence From Portuguese-Speaking Children. Journal of Educational Psychology, 98, 628–641. Corrêa, M. F., Cardoso-Martins, C., & Rodrigues, L. A. (in press). O conhecimento do nome das letras e a sua relação com o desenvolvimento da escrita: evidência de adultos iletrados. Psicologia: Reflexão e Crítica. Ehri, L. (1992). Reconceptualizing the development of sight word reading and its relationship to recoding. In P. Gough, L. Ehri & R. Treiman (Eds.), Reading acquisition (pp. 107–143). Hillsdale, NJ: Erlbaum. Ehri, L. (1998). Learning to read and learning to spell are one and the same, almost. In C. Perfetti, L. Rieben & M. Fayol (Eds.), Learning to spell: research, theory, and practice across languages (pp. 237–269). Mahwah, NJ: Erlbaum. Ehri, L., & Wilce, L. (1985). Movement into reading: is the first stage of printed word learning visual or phonetic?. Reading Research Quarterly, 20, 163–179. Ferreiro, E. (1989). Reflexões sobre a alfabetização. São Paulo, Brazil: Cortez. Ferreiro, E. (1990). Literacy development: psychogenesis. In Y. M. Goodman (Ed.), How children construct literacy: Piagetian perspectives (pp. 12–25). Newark, DE: International Reading Association. Ferreiro, E., & Gomez Palacio, M. (1982). Analisis de las perturbaciones en el proceso de aprendizaje escolar de la lectura y de la escritura. Mexico City, Mexico: Siglo XXI. Ferreiro, E., & Teberosky, A. (1986). Psicogênese da língua escrita. São Paulo, Brazil: Artes Médicas. Grossi, E. P. (1990). Applying psychogenesis principles to the literacy instruction of lower-class children in Brazil. In Y. M. Goodman (Ed.), How children construct literacy: Piagetian perspectives (pp. 99–114). Newark, DE: International Reading Association. Levin, I. (2007). The role of Hebrew letter names in early literacy: the case of multi-phonemic acrophonic names. Journal of Experimental Child Psychology, 98, 193–216. Levin, I., Patel, S., Margalit, T., & Barad, N. (2002). Letter names: effect on letter saying, spelling, and word recognition in Hebrew. Applied Psycholinguistics, 23, 269–300. Pinheiro, A. M. V. (1996). Contagem de freqüência de ocorrência de palavras expostas a crianças na faixa pré-escolar e séries iniciais do 1º. grau. São Paulo: Associação Brasileira de Dislexia. Pollo, T. C., Kessler, B., & Treiman, R. (2005). Vowels, syllables, and letter names: differences between young children’s spelling in English and Portuguese. Journal of Experimental Child Psychology, 92, 161–181. Pollo, T. C., Treiman, R., & Kessler, B. (2008). Preschoolers use partial letter names to select spellings: evidence from Portuguese. Applied Psycholinguistics, 29, 195–212. Pontecorvo, C., & Zucchermaglio, C. (1990). A passage to literacy: learning in a social context. In Y. Goodman (Ed.), How children construct literacy (pp. 55–78). New York, NY: International Reading Association. Seymour, P. (1986). Cognitive analysis of dyslexia. New York, NY: Routledge & Kegan Paul. Treiman, R., Tincoff, R., & Richmond-Welty, D. (1996). Letter names help children connect print and speech. Developmental Psychology, 32, 505–514. Treiman, R., & Kessler, B. (2003). The role of letter names in the acquisition of literacy. In R. Kail (Ed.), Advances in child development and behavior (Vol. 31, pp. 105–135). San Diego, CA: Academic. Weisz, T. (2000). O diálogo entre o ensino e a aprendizagem. São Paulo, Brazil: Ática.
Chapter 4
An Eye for Print: Child and Adult Attention to Print During Shared Book Reading Mary Ann Evans and Jean Saint-Aubin
Abstract According to a simple view of reading, skilled reading is the product of decoding or the ability to turn the printed word into its spoken counterpart, and comprehension or the ability to understand the vocabulary, syntax, morphology, expression, and conventions of the spoken language in which the printed text is written. While an association between shared book reading and language development has been fairly consistently established, the association with reading development is more controversial. The more modest association and even disassociation with the latter may well be a function of the extent to which parents and children jointly attend to the printed text during shared book reading. The purpose of this chapter is to review and integrate previous naturalistic, experimental, and intervention literature with this specific focus. Once one becomes a skilled reader, print that is fixated by the eye cannot help but be processed, and the individual words are recognized as readily as common, everyday objects. This phenomenon has been well illustrated by the Stroop test (Stroop, 1935). Here, when a word is the same as the color of the ink in which it is printed, skilled readers more slowly and erroneously name the ink color than when there is a correspondence. The printed words command their attention; these are automatically processed to a sound-based code and label which must be suppressed to execute the task (e.g., Bub, Mason, & Lalonde, 2006; Carter, Mintun, & Cohen, 1995). Such interference is generally not observed until age 7 when children are at about a grade 2 reading level (Comalli, Wapner, & Werner, 1962; Golden, 1978), and moving into Chall’s (1983) confirmation of the fluency stage of reading. Until then, the pictorial representations on the pages and their attributes are of equal or greater salience. This chapter will outline the development of an “eye for print” in young children, and the way parents coach them to process print within the context of one-to-one M.A. Evans (*) University of Guelph, Guelph, ON, Canada e-mail:
[email protected] D. Aram and O. Korat (eds.), Literacy Development and Enhancement Across Orthographies and Cultures, Literacy Studies 101, DOI 10.1007/978-1-4419-0834-6_4, © Springer Science+Business Media, LLC 2010
43
44
M.A. Evans and J. Saint-Aubin
shared reading as revealed by studies in our laboratories. A discussion of adult-guided child writing activities appears in other chapters within this volume and so, while relevant to this paper, it is not incorporated here. The chapter will proceed in two main sections. The first will review and integrate research examining where children look during shared reading, and the second will look at parental coaching in this activity. These are preceded by a brief outline of the evolving nature and value of shared reading to provide a background to understanding children’s and adults’ print-related behavior and the effects. It will close by highlighting the key findings and implications.
Evolving Nature and the Value of Shared Reading It is commonly thought that being read to by parents during the preschool years helps children develop positive attitudes to reading, increased motivation to learn to read and to read, and higher reading and writing achievement (see for example, statements by Epstein, 1987; Grimmet & McCoy, 1980). In fact, not being read to in the preschool years has been described as having pernicious consequences for low-income children’s long-term success in schooling (Neuman, 1999). Public campaigns of various kinds encourage parents to read to their children as the best way to help them learn to read, and parents have enthusiastically taken up this advice. In the developed world, reading storybooks with children as a daily activity has become common, virtually on par with personal hygiene, among middle-class families. In a sample of over 650 families in the provinces of Ontario and New Brunswick in Canada, we find that 72% of children aged 4–6 are read to 5 or more days a week. Even among socioeconomically disadvantaged families, the majority of parents report reading to their children. For example, among lowincome families in the United States, Dickinson and Tabors (2001) found that 96% of children aged 3 and 4 were read to, of whom 66% were read to daily (see Raikes et al., 2006 for similar findings). The reasons why parents read books with their children undoubtedly center around its positive characteristics of being entertaining, stimulating, and intimate. In fact in a recent study of the goals that parents of children aged 3–9 have for shared book reading, the goals of enjoyment and having quality time with their child emerged as primary and highest rated (Audet, Evans, Williamson, & Reynolds, 2008). These goals were followed by helping their child learn to read and stimulating cognitive development, parental behaviors toward which have been increasingly documented in studies of shared book reading with children from infancy through the primary grades (see Bus, Van IJzendoorn, & Pellegrini; 1995; Fletcher & Reese, 2005; Scarborough & Dobrich, 1994; van Kleeck, Stahl, & Bauer, 2003). In some of the earliest studies of parents and children reading together, Ninio and collaborators (Ninio, 1983; Ninio & Bruner, 1978) effectively described how adults initially point to and label illustrations in books when reading to young children, and ask them to do the same as a mechanism to teach and test vocabulary.
4 An Eye for Print: Child and Adult Attention to Print During Shared Book Reading
45
Gradually parents move toward more abstract language involving inferential questions and comments about superordinate concepts, reasons, predictions, and effects (e.g., Snow & Goldfield, 1983; van Kleeck, Gillam, Hamilton, & McGrath, 1997; Wheeler, 1983). Thus from the earliest parent–child interactions with books, children are encouraged to attend to and develop an understanding of the meaning of the book’s text and pictures. These conversations, however, should not be construed as totally parent directed, for even at 10 months of age, parent–child interactions are of a reciprocal nature with parents responsive to the child’s communicative contributions (Lewis & Gregory, 1987). In addition, as children begin to take their first steps in the reader role, parents actively support them by letting them “read” predictable words and helping them to accurately decode. Thus both parental goals and reciprocity with the child’s interests and emerging skills are important for understanding the nature of shared book reading and attention to print within it.
Where Children Are Looking During Shared Reading In what was perhaps the first report of young children’s attention to the different elements of storybooks, Yaden, Smolkin, and Conlon (1989) audio taped the interactions of two preschool-age children. The children most often asked questions about the illustrations, followed by questions about the meaning of the story and words within it, then questions about the printed forms on the page, and lastly book conventions. Children’s questions can be taken as an indirect window onto the objects of their attention. However, this index is flawed, for it may be easier for young children to formulate and express comments about the storyline and accompanying illustrations than about the appearance of lines of text. Children finger pointing to the text may underestimate attention to print, for accurate finger pointing requires phonological awareness and matching of print to sound (Ehri & Sweet 1991). The advent of head-mounted eye trackers that allow children to sit naturally and move their heads has afforded more direct access to and accurate measurement of what they are looking at. They allow researchers to monitor children’s eye movements, superimpose these directly on the visual stimuli, and determine the frequency, duration, and latencies of fixations to various parts of the visual display. Three research laboratories have independently shown that it is the pictures that are the objects of preschool children’s regard. In the first published report, Evans and Saint-Aubin (2005) described two studies with children aged 48–61 months. In the first study, five short storybooks having a variety of page layouts were read to children. In one, the illustration occupied the center of each page with one line of text above and below. In another, the illustration was on the right page and the text was on the left page along with one large decorated letter and a small black-and-white drawing. In a third, the text was presented in speech bubbles within the illustrations. Finally two books were included with simple monochromatic drawings and a minimal amount of text. Results revealed that children rarely looked at the print (about 7% of their viewing time) regardless
46
M.A. Evans and J. Saint-Aubin
of the book design. In the second study, two text versions of a storybook were created. For each page, each version specifically referred to a different element of the illustration. For instance, for the page showing the characters in a passenger train on a bridge, one version referred to a small boat on the river and the other to the smoke of the locomotive. Results revealed that children spent more time on the part of the illustration highlighted by the text, suggesting that the narration of the story guided their attention. In the same year, Justice, Skibbe, Canning, and Lankford (2005) independently ran a similar study. They monitored eye movements of children aged 50–69 months when mothers read them a print-salient and a picture-salient book. The printsalient book differed from the picture-salient one by having larger print size and fewer words per page. In both cases children rarely looked at the print and the design manipulation had only a limited impact. Most recently, Guo and Feng (2007) in a study with 18 Chinese-speaking 6-year olds also found that the children rarely looked at Chinese characters. Further, whether the book was read verbatim or with an interactive style had no effect on the proportion of time spent on print. Finally, using a different methodology and sample of 76 children aged 3–5, Evans, Williamson, and Pursoo (2008) videotaped the direction in which children’s eyes were turned while an adult read two large-format books placed at their midline. These books included text made highly salient by printing a word on some of the pages in a large font, in a colored box, or surrounded by a decorative border. In one condition, the adult pointed to each word of the text as she read it. After each book was read, the children were presented with cards, each showing a part of the illustrations, a salient part of the text, or a picture or a corresponding foil, and asked to identify whether it was in the book. When the words were not pointed to, children looked toward the print 2, 4, and 6% of the time for the 3-, 4-, and 5-year olds, respectively. Pointing to the words increased print-looking time to 10, 11, and 24%, respectively, but increased print target recognition only for the 4-year olds. Lastly, after controlling for vocabulary, visual memory and maturation associated with these scores, children’s scores on a test of emergent orthography and letter–word identification predicted 10% of the variance in time looking at print and 16% of the variance in the recognition of the print elements. One might argue with these results. Even though an attempt was made to make the print salient, to contrast books having simple monochrome drawings with complex, colorful ones, or to read with different styles, the books had a narrative read by an adult and illustrations to which the child’s attention might naturally be drawn. In a second study reported by Yaden, Smolkin, and Conlon (1989) of seven children aged 3–4, they again found that children overwhelmingly focused on the illustrations and storyline, but that alphabet books were more likely to elicit questions about the graphic forms. This led them to speculate that books that highlight print may direct children’s attention to it at an earlier age. Yopp and Yopp (2000) concurred that alphabet books are particularly useful for helping children learn letters because the print is made salient (p. 414), and they have been frequently recommended by educators for their potential to focus attention on print (Fountas & Pinnell, 1996; Tompkins, 2001).
4 An Eye for Print: Child and Adult Attention to Print During Shared Book Reading
47
To determine the extent to which findings might deviate with an alphabet book, Evans, Saint-Aubin and Landry (2009) selected a book exemplifying characteristics recommended for alphabet books for young children. Each page displayed a single large uppercase letter; a single, simply drawn corresponding object; a word in uppercase letters naming the object; and a panda bear in different poses on each page. Children, aged 59–71 months, were asked to “read” the alphabet book to the adult. In line with previous eye movement studies of shared reading, children spent much more time on the illustrations than on the print, be it the featured letter or the printed word. Also in line with previous studies, about 95% of the time children first fixated one of the illustrations rather than one of the text elements. However, most interestingly, the more letters of the alphabet known, the faster the children were as at fixating the featured letter on the page, and the longer the printed word and its first letter were looked at. This latter finding is even more remarkable given that, on average, children were unable to read 97.3% of the presented words. Together, these studies of children’s attention to print clearly show the pre-eminence of illustrations for young children. Given these results, the potential for children to learn about print from storybooks and even alphabet books without adult mediation explicitly focusing their attention on the graphic forms seems limited unless they already have adequate print knowledge. Once children begin to read one might expect them to follow along with the text to at least some degree as the adult reads aloud. A study of children in grades kindergarten through four by Roy-Charland, Saint-Aubin, and Evans (2007) found this to be partly the case. Here, the parent read three texts of various difficulty levels to the child. Once all books had been read, the child read to the parent two pages from each book, not included in the parent-to-child reading activity. Based on how well they read, children were classified as able to read the storybook on their own or not. This proved to be an important distinction: Children who could not read the text looked at the print between 4 and 9% of the time when the parent was reading, compared to 46 and 59% for children who could read the text. This difference was observed when comparing children who were able to read with those unable to read the storybooks, as well as within children, when contrasting the storybooks each child was able to read with those s/he could not. It should be noted, however, that although children who could read a given storybook spent a fair amount of time on its text, the pattern of their eye movements suggested that they were only sporadically and not closely reading along. In short, the overall lesson to be learned from these studies is that just because print is close at hand in books, does not mean that young children look at it or follow it. Some letter and decoding knowledge to facilitate processing of the print appears to be a prerequisite for fixating individual words appearing singly on a page. Similarly reading skill that matches the difficulty level of the text is a prerequisite for children to fixate parts of it when being read to. Otherwise, young children have an eye for the illustrations while listening to the story. This may well explain why research has found that the frequency of shared book reading with young children is only modestly related to the development of print specific (see reviews by Bus et al., 1995; Scarborough & Dobrich, 1994; van Kleeck et al., 2003).
48
M.A. Evans and J. Saint-Aubin
Parental Coaching Behavior As noted earlier, when reading books to young children, the text of the book provides a script that parents modify, extend, and elaborate. For example, Martin (1998), through studying mother–child dyads from age 6 months to 4 years, observed that mothers omitted entire sentences, substituted simpler words for more complex ones, inserted animal sounds, issued attentional markers (e.g., gasps and exclamations), and discussed the text, all to focus the child’s attention on the content and enhance understanding. This is, after all, what reading to young children is all about: parents and children having an enjoyable and intimate time together, in keeping with the primary goals parents express for this activity (Audet et al., 2008). However as a context for stimulating cognitive development and helping children learn to read, one might also expect parents to substantially deviate from reading the text to talk about the print. To examine this, van Kleeck et al. (1997) coded the utterances of 33 parents reading a familiar and an unfamiliar book to children aged 3–4. They observed many more story-related utterances than print/book-related utterances, outnumbering the latter by a ratio of 6:1 for each of the books. The picture changes only slightly when alphabet books are read to children. Alphabet books do encourage more talk than storybooks about phonemic awareness and letter–sound relationships, but this is still overwhelmingly outnumbered by other comments. In a study by Stadler and McEvoy (2003), parents of 55 normally developing children and 17 children with language delay (average age: 5) read an alphabet book and a storybook with their child. While the alphabet book elicited more parent talk about print concepts in both language-delayed and normal dyads than the storybook, parents made few comments related to reading per se with either book, just to 2–3 in a 5- to 10-min period. Instead talk centered around the illustrations. In fact, despite the fact that the alphabet book did not contain a story with characters or events, parents commented as much about the contents of the alphabet book as when reading the narrative book. Bus and van IJzendoorn (1988) also observed that compared to a storybook, alphabet books elicited more print-related comments: Parents were more likely to name letters, try to make children recognize sounds in words, and connect letters to words. Importantly, however, they found that mothers made more of these sorts of reading instructional comments to children with higher levels of emergent literacy. Similarly, Hayden and Fagan (1987) observed that preschoolers with high print awareness received more high-demand comments, which required not only explaining things and making inferences, but also talk about print concepts and the process of reading. Both studies demonstrate the reciprocal nature of parent–child interactions in shared reading, with parents appearing to be sensitive to the child’s print awareness/ literacy level and adjusting their input, and children potentially benefiting from their comments. The influence of children’s print-specific knowledge on parental behavior becomes even clearer in longitudinal studies when children begin to read, in that parents shift the nature of their supporting comments and clues in helping the child to decode. We have examined this in two ways. First, following neo-Vygotskian
4 An Eye for Print: Child and Adult Attention to Print During Shared Book Reading
49
notions of scaffolding and mediation (Griffin & Cole, 1984; Wood & Middleton, 1975) parental comments after children’s reading errors were coded according to the degree of support (Evans, Moretti, Shaw, & Fox, 2003). Second, they were coded according to focus of attention (Evans, Baraball, & Eberle, 1998; Mansell, Evans, & Hamilton-Hulak, 2005). Degree of support refers to how transparent the clue is or close it is to the word to be read, regardless of its type. Thus giving the word is the most supportive, followed by almost giving the word (e.g., pointing to a clear picture of the word, slowly enunciating the word), giving a specific clue, providing a general hint (e.g., it something you eat), offering a general admonition (e.g., slow down), encouraging a second try without any help, and ignoring the misreading. Among the 46 dyads observed when the children were aged 6–7, parents provided a variety of levels of support, but importantly, they provided the greatest percentage of support at the level just below the level at which their child was most likely to correctly re-read the word. Thus parents appear to adjust the level of support to match their child’s on-going performance. Focus of attention was coded as ignoring the error, encouraging a reattempt without any additional help, supplying the word, highlighting a context clue (e.g., the picture or meaning thus far), or giving a graphophonemic clue (e.g., the initial phoneme or part of the word). Evans et al. (1998) showed that mothers were less likely to ignore the error, less likely to provide context clues, and more likely to provide graphophonemic feedback to 5- and 6-year olds if the child had some word identification skill. In a longitudinal study of 50 normally progressing children using this same coding scheme, Mansell et al. (2005) found that from kindergarten to grade 2, parents increased requests to just try again and decreased context clues and supplying the word after the child made an error. They also gave fewer hints before the child attempted a word. Providing graphophonemic clues peaked in first grade. Thus parents explicitly coach the application of print-specific skills, especially in grade 1, and increasingly encourage more independence in decoding. These various types of feedback were not evenly distributed in parents. Through cluster analysis, Mansell et al. (2005) delineated two main clusters: Word Suppliers were more likely to give the miscued word, less likely to signal that an error had been made, and less likely to give a graphophonemic clue than Code Coaxers. These styles for the most part persisted across the grades. A key question is whether feedback style makes any contribution to the development of print-specific skills. To address this, Evans, Mansell, and Shaw (2006) examined the progress of 44 normal children and 19 children at risk for poor reading development. After controlling for children’s earlier reading skill, a Code Coaxer style across kindergarten and grade one positively predicted word identification skill in grades one and two in both skilled and less skilled readers.
Key Findings and Their Implications The various findings outlined above suggest that the common wisdom that reading to your child will develop their print-specific skills needs to be more finely articulated to differentiate what in the interaction affects what in the child’s development.
50
M.A. Evans and J. Saint-Aubin
According to a “simple view of reading” (Gough & Tunmer, 1986), skilled reading is a product of decoding or the ability to turn the printed word into its spoken counterpart, and comprehension or the ability to understand the text. While a modest association between shared book reading and language development has been fairly consistently established which benefits the comprehension side of the equation, the association with decoding is less likely. Several recent studies challenge the benefits of shared reading for the development of print-specific skill (Evans, Shaw, & Bell, 2000; Frijters, Barron, & Brunello, 2000; Levy, Gong, Hessles, Evans, & Jared, 2006; Sénéchal & LeFevre, 2002; Sénéchal, LeFevre, Thomas, & Daley, 1998). These studies have found that parent–child shared book reading, while predicting a small amount of variance in oral language skill, did not predict print-specific knowledge or word reading skill, while explicit parent teaching did. Similarly, Meyer, Stahl, Linn, and Wardrop (1994) found that the amount of time kindergarten teachers spent reading to their students was negatively correlated with reading achievement, but positively correlated with listening comprehension. On the basis of data from interviews with the parents of these children, they also found that the relationship between parent reading to the child and child reading achievement was weak. Rather, the child’s own participation in reading and extent to which he/she read to the parent was positively related to achievement. Meyer et al. (1994) concluded that “there is no magic in just reading to children. Instead the magic comes as you as engage them with print” (p. 83). Our work studying children’s eye movements in concert with others’ research on the nature of their comments suggests a reason for the modest association at best, and even disassociation, found between reading to children and print-specific skills. It may be a function of how little attention to print there is in this activity. The positive effects found in some studies may well have been carried by home activities coexisting with shared book reading, activities in which parents actively teach children to print letters/words, to learn letters sounds and to read words. These undoubtedly include writing activities, the effectiveness of which has been shown by Levin and her colleagues (e.g., Aram & Levin, 2002; Shatil, Share, & Levin, 2000). In addition, most recently with the emergence of digital books, research has also shown that e-books that can animate print and direct the child’s attention to it in engaging ways appear to produce gains in children’s print knowledge (see chapter by Shamir & Korat, in press). If the parent’s goal is to promote reading acquisition during shared book reading, the studies reviewed here have several implications. First, children must have a critical amount of letter knowledge to attend to print. Without it they are unlikely to consolidate or add to this knowledge through shared book reading without explicit adult direction. Thus shared reading activities and explicit teaching should coexist. Second, reading materials should be adapted to the child’s reading skills. Alphabet books are often used as picture books with very young children but can be effective in encouraging print-specific skills and attention to words with older children who have some letter knowledge, even if they are not yet readers. Similarly when reading to children, books having text within their reading level may facilitate some reading along on their part and fluency development. Finally, parent comments
4 An Eye for Print: Child and Adult Attention to Print During Shared Book Reading
51
and interactions should include a focus on print as shown by Justice and Ezell (2002) with at-risk children. In shared book reading attention to the print should be done in a way that will not alter the intimate and entertaining quality that this parent–child activity should have. Thus, direction of the child’s attention to print will likely be most effective when provided sensitively, frequently, positively, and in a variety of activities.
References Aram, D., & Levin, I. (2002). Mother-child joint writing and storybook reading: relations with literacy among low SES kindergartners. Merrill-Palmer Quarterly, 48, 202–224. Audet, D., Evans, M. A., Williamson, K., & Reynolds, K. (2008). Goals for shared reading across the primary grades and their relationship to the parents’ behaviour with 4-year-olds. Early Education and Development, 19, 113–138. Bub, D. N., Mason, M. E. J., & Lalonde, C. E. (2006). Cognitive control in children: Stroop interference and suppression of word reading. Psychological Science, 17, 351–357. Bus, A. G., & van IJzendoorn, M. H. (1988). Mother-child interactions: attachment and emergent literacy: a cross-sectional study. Child Development, 59, 1262–1272. Bus, A. G., van IJzendoorn, M. H., & Pellegrini, A. (1995). Joint book reading makes for success in learning o read: a meta-analysis on intergenerational transmission of literacy. Reading Research Quarterly, 30, 998–1015. Carter, C. S., Mintun, M., & Cohen, J. D. (1995). Interference and facilitation effects during selective attention: an H2150 PET study of Stroop task performance. Neuroimage, 2, 264–272. Chall, J. S. (1983). Stages of reading development. New York: McGraw Hill. Comalli, P. E., Jr., Wapner, S., & Werner, U. (1962). Interference effects of the Stroop colour-word testing in childhood, adulthood and aging. Journal of Genetic Psychology, 100, 47–53. Dickinson, D. K., & Tabors, P. O. (2001). Beginning literacy with language. Baltimore: Brookes. Ehri, L. C., & Sweet, J. (1991). Fingerpoint-reading of memorized text: what enables beginners to process the print. Reading Research Quarterly, 26, 442–462. Epstein, J. (1987). Parental involvement: what research says to administrators. Education and Urban Society, 19, 119–136. Evans, M. A., Baraball, L., & Eberele, T. (1998). Parental responses to miscues during child-to parent book reading. Journal of Applied Developmental Psychology, 19, 67–84. Evans, M.A., Mansell, J., & Shaw, D. (2006). Parental coaching of normally and slowly progressing young readers. In D. Aram (Chair) Parenting and early literacy. Vancouver: Symposium conducted at Society for the Scientific Study of Reading. Evans, M. A., Moretti, S., Shaw, D., & Fox, M. (2003). Parent scaffolding in children’s oral reading. Early Education and Development, 14, 363–388. Evans, M. A., & Saint-Aubin, J. (2005). What children are looking at during shared storybook reading. Psychological Science, 16, 913–920. Evans, M. A., Saint-Aubin, J., Landry, N. (2009). Letter names and alphabet book reading by senior kindergarteners: An eye-movement study, Child Development, 80, 1824–1841. Evans, M. A., Shaw, D., & Bell, M. (2000). Home literacy activities and their influence on early literacy skills. Canadian Journal of Experimental Psychology, 54, 65–75. Evans, M. A., Williamson, K., & Pursoo, T. (2008). Preschoolers’ attention to print during shared book reading. Scientific Studies of Reading, 12, 106–129. Fletcher, K., & Reese, E. (2005). Picture book reading with young children: a conceptual framework. Developmental Review, 25, 64–103.
52
M.A. Evans and J. Saint-Aubin
Fountas, I. C., & Pinnell, G. S. (1996). Guided reading: good first teaching for all children. Portsmuth, NH: Heineman. Frijters, J. C., Barron, R. W., & Brunello, M. (2000). Direct and mediated influences of home literacy and literacy interests on prereaders’ oral vocabulary and early written language skill. Journal of Educational psychology, 92, 466–477. Golden, C. J. (1978). Stroop colour and word test: a manual for clinical and experimental uses. Chicago, IL: Stoelting. Gough, P. B., & Tunmer, W. E. (1986). Decoding, reading, and reading ability. Remedial and Special Education, 7, 6–10. Griffin, P., & Cole, M. (1984). Current activity for the future: The Zoped. In R. Rogoff & J. V. Wertsch (Eds.), Children’s learning in the “Zone of proximal development”. New Directions for Child Development, No. 23 (pp. 45–65). San Francisco, CA: Jossey-Bass. Grimmet, S., & McCoy, M. (1980). Effects of parental communication on reading performance of third grade children. The Reading Teacher, 33, 303–308. Guo J., & Feng, G. (2007). Eye movements during shared storybook reading: do children look at the text? Poster presented at the meeting of Scientific Studies of Reading, Prague. Hayden, H. M. R., & Fagan, W. T. (1987). Keeping it in context: strategies for enhancing literacy awareness. First Language, 7, 159–171. Justice, L. M., & Ezell, H. K. (2002). Use of storybook reading to increase print awareness in at-risk children. American Journal of Speech-Language Pathology, 11, 17–29. Justice, L. M., Skibbe, L., Canning, A., & Lankford, C. (2005). Preschoolers, print and storybooks: an observational study using eye movement analysis. Journal of Research in Reading, 28, 229–243. Levy, B. A., Gong, Z., Hessles, S., Evans, M. A., & Jared, D. (2006). Understanding print: early reading development and the contributions of home literacy experiences. Journal of Experimental Child Psychology, 93, 63–93. Lewis, C., & Gregory, S. (1987). Parents talk to their infants: the importance of context. First Language, 7, 201–216. Mansell, J., Evans, M. A., & Hamilton-Hulak, L. (2005). Developmental changes in parents’ use of miscue feedback during shared book reading. Reading Research Quarterly, 40, 294–317. Martin, L. E. (1998). Early book reading: how mothers deviate from printed text for young children. Reading Research and Instruction, 37, 137–160. Meyer, L. A., Stahl, S. A., Linn, R., & Wardrop, J. L. (1994). Effects of reading storybooks aloud to children. Journal of Educational Research, 88, 69–85. Neuman, S. B. (1999). Books made a difference: a study of access to literacy. Reading Research Quarterly, 34, 286–311. Ninio, A. (1983). Joint book reading as a multiple vocabulary acquisition device. Journal of Child Language, 5, 1–15. Ninio, A., & Bruner, J. (1978). The achievement and antecedents of labeling. Journal of Child Language, 5, 1–15. Raikes, H. H., Raikes, H. A., Pan, B. A., Luze, G., Tanis-Lemonda, C. S., Rodrigues, E. T., et al. (2006). Mother-child book-reading in low-income families; Correlates and outcomes during the first three years of life. Child Development, 77, 924–953. Roy-Charland, A., Saint-Aubin, J., & Evans, M. A. (2007). Eye movements in shared book reading with children from kindergarten to grade 4. Reading and Writing: An Interdisciplinary Journal, 20, 909–931. Scarborough, H., & Dobrich, W. (1994). On the efficacy of reading to preschoolers. Developmental Review, 14, 245–302. Sénéchal, M., & LeFevre, J. A. (2002). Parent involvement in the development of children’s reading skill: a five-year longitudinal study. Child Development, 73, 445–460. Sénéchal, M., LeFevre, J., Thomas, E. M., & Daley, K. E. (1998). Differential effects of home literacy experiences on the development of oral and written language. Reading Research Quarterly, 33, 96–116.
4 An Eye for Print: Child and Adult Attention to Print During Shared Book Reading
53
Shamir, A. & Korat, A. (in press). Activity with e-books as a support for emergent literacy in different learning contexts. In O. Korat & D. Aram (Eds.), Literacy and language: relationships, bilingualism and difficulties. Jerusalem: Magness. Shatil, E., Share, D. L., & Levin, I. (2000). On the contribution of writing to grade 1 literacy: a longitudinal study in Hebrew. Applied Psycholinguistics, 21, 1–21. Snow, C. E., & Goldfield, B. A. (1983). Turn the page please: situation-specific language acquisition. Journal of Child Language, 10, 551–569. Stadler, M. A., & McEvoy, M. A. (2003). The effect of text genre on parent use of joint book reading strategies to promote phonological awareness. Early Childhood Research Quarterly, 18, 373–395. Stroop, J. P. (1935). Studies of interference in serial verbal reactions. Journal of Experimental Psychology, 18, 643–662. Tompkins, G. (2001). Literacy for the 21st century. New York: Prentice Hall. van Kleeck, A., Gillam, R. B., Hamilton, L., & McGrath, C. (1997). The relationship between middle-class parents book sharing discussion and their preschoolers abstract language development. Journal of Speech, Language, and Hearing Research, 40, 1261–1271. van Kleeck, A., Stahl, S. S., & Bauer, E. B. (2003). On reading books to children. Mawah, NJ: Erlbaum. Wheeler, M. P. (1983). Context-related age changes in mothers speech: joint book-reading. Journal of Child Language, 10, 259–263. Wood, D., & Middleton, D. (1975). A study of assisted problem solving. British Journal of Educational Psychology, 66, 181–191. Yaden, D. B., Smolkin, L. B., & Conlon, A. (1989). Preschoolers’ questions about pictures, print conventions, and story text during reading aloud at home. Reading Research Quarterly, 24, 189–214. Yopp, R. H., & Yopp, H. K. (2000). Sharing informational text with young children. The Reading Teacher, 53, 410–423.
Part 2
Promoting Literacy at Home and in School
Chapter 5
The Role of Hebrew Letter Names in Early Literacy: The Case of Multi-phonemic Acrophonic Names Iris Levin
Abstract English-speaking children more often spell letters correctly when the letters’ names are heard in the word (e.g., B in beach vs. bone). Hebrew letter names have been claimed to be less useful in this regard. In Study 1, kindergartners were asked to report and spell the initial and final letters in Hebrew words, including full (CVC), partial (CV), and phonemic (C) cues derived from these letter names (e.g., kaftor, kartis, kibel, spelled with /kaf/). Correct and biased responses increased with the length of congruent and incongruent cues, respectively. In Study 2, preschoolers and kindergartners were asked to report the initial letters of monosyllabic or disyllabic names (e.g., /kaf/ and /samex/, respectively) that included the cues described above. Correct responses increased with cue length; the effect was stronger with monosyllabic over disyllabic letter names, probably because the cue covered a larger ratio of the letter name. Phonological awareness was linked to the use of letter names. Letter knowledge is indisputably one of the basic foundations for the acquisition of reading and spelling skills in alphabetic orthographies (Byrne, 1998; Ehri, 1992, 1998). The recognition of letters has repeatedly been found to be one of the best predictors for reading and spelling acquisition later in school, across alphabets and educational systems (Bruck, Genesee, & Caravolas, 1997; de Jong & van der Leij, 1999; Lonigan, Burgess, & Anthony, 2000). The predictive value of letter knowledge often does not fall behind that of phonological awareness (Caravolas, Hulme, & Snowling, 2001; Cardoso-Martins, 1995; Muter, Hulme, Snowling, & Taylor, 1997; Scarborough, 1998) and is greater than that of vocabulary or home literacy (Caravolas et al., 2001; Schatschneider, Fletcher, Francis, Carlson, & Foorman, 2004; Share, Jorm, MacLean, & Matthews, 1984; Storch & Whitehurst, 2002). Furthermore, it has been reported that school-age children who appeared to be lagging behind their peers in reading acquisition had also been found to fall behind in letter knowledge when in kindergarten (de Jong & van der Leij, 1999; Elbro, Borstrom, & Petersen, 1998; Pennington & Lefty, 2001; Snowling, Gallagher, & Frith, 2003). I. Levin (*) Tel Aviv University, Tel Aviv, Israel e-mail:
[email protected] D. Aram and O. Korat (eds.), Literacy Development and Enhancement Across Orthographies and Cultures, Literacy Studies 101, DOI 10.1007/978-1-4419-0834-6_5, © Springer Science+Business Media, LLC 2010
55
56
I. Levin
Children who know more letters by name and fewer, if any, letters by sound can use letter name knowledge as leverage in learning to recognize printed words or spell them. This conclusion, based on the seminal research by Ehri (Ehri & Wilce, 1985; Scott & Ehri, 1990), was expanded by Treiman (Bowman & Treiman, 2002; McBride-Chang & Treiman, 2003; Treiman, & Rodriguez, 1999; Treiman, Sotak, & Bowman, 2001). Treiman compared the way children learned to recognize or spell two series of printed words in abbreviated spellings in English: name series and sound series. The spellings in the name series included a letter whose name could be detected in the pronunciation of the word, like TM for “team,” which includes the sound /ti/. In the sound series the spellings included letters whose sounds, but not their name, could be detected in the pronunciation of the word, like TM for “tame.” They found that not-yet-readers learned to recognize and spell the words in the name series faster than they did the words in the sound series (see also de Abreu & Cardoso-Martins, 1998; Levin, Patel, Margalit, & Barad, 2002). Since this research was carried out in Hebrew, a short description of Hebrew letters is warranted. The Hebrew alphabet is composed of 22 basic letters. In many alphabets, like English or Russian, almost all letter names are made up of two phonemes so that the iconic part of a letter name (e.g., /b/ of /bi/) amounts to a phonemic sound that is often represented by the letter. However, in Hebrew, as in Arabic, Greek, and Portuguese, where all or some of the letter names are longer, the iconic sound in the name is one of several phonemes. About two-thirds of Hebrew letter names are monosyllabic, mostly of a CVC structure (e.g., /kaf/) and about one third are disyllabic, mostly of a CVCVC structure (e.g., / lamed /). All Hebrew letter names are stressed on the first syllable. In some alphabets, for example, in Arabic, the shape of a letter depends on its position in the word. In Hebrew, five letters have special shapes when they appear at the end of a word, and the name of such a final letter is even longer, as it is qualified by the adjective “sofit,” meaning final (e.g., kaf sofit). Letter name knowledge can help children grasp the concept that letters stand for sounds if they frequently encounter words in print that allow them to notice the correspondence between their spoken forms and the names of the letters that appear in their print form (Foulin, 2005). Bowman and Treiman (2002) estimated that around 43% of English words in print targeted for kindergartners and first graders include a letter whose name can be heard in the word. Cardoso-Martins (Cardoso-Martins, Lemos, Corrêa, & Napoleão, 2006; Cardoso-Martins, Resende, & Rodrigues, 2002) suggested that this phenomenon is more prevalent in Romance languages, like Portuguese, Spanish, or Italian, than in English because the names and the sounds of vowels very often coincide in the former. Due to the length of Hebrew letter names, it has been claimed that they are less useful in bridging the gap between the spoken and the written words. Tolchinsky (2003) argued that whereas English letter names facilitate the grasping of their sounds, this is not a cross-linguistic phenomenon: “In Hebrew, the name of the letters differs greatly from their sound and studies such as that performed by Treiman would be difficult to carry out in Hebrew because there is no possibility of parts of words coinciding with letter-names” (p. 82). Shimron (2007, p. 61) reported that out of 200 words spelled with the letter /kaf/ in a Hebrew dictionary, only 10 words
5 The Role of Hebrew Letter Names in Early Literacy: The Case of Multi-phonemic
57
started with this sequence of sounds. He concluded that teaching letter names in preschool or kindergarten is, at best, useless. Hebrew words can include letter name cues, but most often they include part of the name rather than its whole. We carried out an analysis of all the printed words in a corpus of 137 popular and canonic books for preschoolers that amounted to a total of 62,375 words. We counted how many word tokens contained any two-to-five phoneme sequence from the name of any of the letters in the word. About 48% of all tokens contained a cue for one of the 37 possible letter names in Hebrew (including 22 basic letter name and 15 letter name varieties; see, varieties in Appendix of Treiman, Levin, & Kessler, 2007), but only 2% of these included a whole letter name; most of them (97%) were monosyllabic (e.g., /kaf/) rather than disyllabic names (e.g., /lamed/). Most of the cues found in the words (95%) were the initial CV string of letter names. For example, while we hardly found any word that includes the sequence of the letter name /lamed/, there were many words that include the initial CV of this name, /la/. In addition, letter name cues can appear in any position in the word. In our data base, 41% of all cues were word-initial, 49% medial or final, showing that such cues are not dependent on word position. Moreover, about 19% of words with letter name cues included more than a single cue (e.g., tanur (oven) includes the strings /ta/ and /nu/ that appear in the letter names /taf/ and /nun/). A previous study on the effects of Hebrew letter names in helping kindergarteners to connect print and speech focused mainly on children’s reporting the initial letter of spoken words (Levin et al., 2002). It showed that both full as well as partial letter names embedded in spoken words increased children’s ability to report these letters in comparison with words that included only the initial phoneme of the letter name. In the present research including two studies, the effects of full and partial letter names are compared to each other, with the former expected to be more helpful than the latter. Since Hebrew words contain partial letter names much more often than they do full letter names, determining the relative effect of these two cues is of great significance. Study 1 expands upon Levin et al.’s (2002) study by comparing the use of letter names in three tasks: letter reporting, letter spelling, and phoneme isolation. Sounds of letter names are expected to facilitate the tasks of letter reporting and letter spelling substantially and to a similar extent. It is still an open question whether sounds of letter names might facilitate phoneme isolation. On the one hand, it has been argued that letters draw the children’s attention to the fact that words can be segmented into sounds (Foulin, 2005). Hence, in detecting the sounds of a letter name it can help the children to come up with its sound. On the other hand, the demand to isolate sounds directly alerts children to the sound structure of the word rather than to that of the letters it is spelled with. Consequently, the task demand can divert the children’s attention from the letter name. Also compared in Study 1 were congruent versus incongruent cues, the former expected to facilitate and the latter to hamper performance. Two types of incongruent cues were used: one referring to the beginning and one to the end of the word. Finally, the effect of children’s phonological awareness in their use of letter names was examined. Individual differences in the use of letter names in reporting or spelling letters have not been studied before in any language, to the best of our knowledge.
58
I. Levin
The scope of the research was further increased in Study 2 by comparing the effects of monosyllabic and disyllabic letter names. This design afforded the opportunity to disentangle the effects of cue length per se from that of the ratio of the letter name that this cue encompassed. Note that the same cue length, say CVC, can encompass the entire letter name for monosyllabic letter names like /kaf/, but only a part of the letter name for disyllabic names, like /gimel/.
Study 1 The first aim of this study was to compare the facilitation enabled by full and partial letter names heard in words on kindergarteners’ ability to connect print and speech. Kindergartners were asked to deal with triads of matched words: (1) one starting with a full letter name cue composed of three phonemes (CVC), (2) one starting with a partial letter name cue composed of the first two phonemes (CV), and (3) the last starting with the letter’s initial phonemic cue (C). We expected that the greater the phonological correspondence between the letter name and the sounds detected in the word, the easier it would be for children to come up with the cued letter. For example, children asked to spell three words that start with the letter named /kaf/ כ should have greatest success in spelling the initial letter of kaftor, כפתורsomewhat less success with kartis כרטיסand the least success with kivsa ( כבשהbutton, ticket, and ewe, respectively). Note that Hebrew words are spelled from right to left. Unless indicated otherwise, Hebrew words are stressed on the last syllable. The effect of letter names on children’s ability to connect print and speech was studied in three tasks: letter reporting, letter spelling, and phoneme isolation. Treiman, Tincoff, and Richmond-Welty (1996) asked children to report the initial or final letter of spoken words, and found that letter names detected in spoken words facilitated initial and final letter reporting. In addition, children were found to be more successful at reporting initial rather than final letters, just like in spelling (Cardoso-Martins et al., 2002; Stage & Wagner, 1992; Treiman, Berch, & Weatherston, 1993). Finally, letter names also hampered initial letter reporting when the sound of a letter different from the initial could be detected in the spoken word. Thus, children at times reported the letter Y for wife but rarely did so for the control word wait. Hebrew letter names differ phonologically from English letter names (Treiman et al., 2007). Hebrew letter names are acrophonic, that is, they start with the sound that the letter stands for. This is the case for some English letter names (e.g., B, D), but not for other letter names that end with the sound that they stand for (e. g., F, R) or do not include their typical sounds (e.g., W, H). Therefore, words like wife that start with one letter (W) but with sounds that make up the name of another letter (Y) do not exist in Hebrew, except for homophonous letters. Hebrew words can start with a sequence of sounds corresponding to one letter name but the word might be spelled with its homophonous letter. For instance, in kafats קפץ (jumped) the letter כnamed /kaf/ can be detected, but it is spelled with ק/kuf/.
5 The Role of Hebrew Letter Names in Early Literacy: The Case of Multi-phonemic
59
In English, reporting incorrect initial letters, like Y for wife, occurred infrequently, probably because it contradicted children’s knowledge that Y stands for /j/. In Hebrew, the error of reporting a homophonous letter may be more frequent because phonologically it is appropriate. Only the specific knowledge of the word’s orthography can help decide that kafats is spelled with /kuf/, not with /kaf/, and this knowledge is beyond kindergartners (Ravid, 2001, 2004). English words at times end with a sequence of sounds that corresponds to its last letter, like deaf that ends in /ef/. Due to the acrophonic nature of Hebrew letter names, in a word that ends with a sequence of sounds that corresponds to a letter name this letter is typically not its last letter. A word ends in the sounds of its last letter’s name only if the name starts and ends with the same phoneme, and there are only four such letter names (/vav/, /xet/, /mem/, and /nun/). In most letters, the name of the last letter cannot be detected in the word. For instance, Sakuf ( שקוףtransparent) ends in the letter ף/fei/ that marks /f/, but the sounds that can be detected at the end are /kuf/. Note that incongruent partial letter names in Hebrew differ phonologically depending on whether they are embedded in the initial or in the final position of words. In the initial position, a partial letter name is the body of the name, whereas in the last position it is the rime. For instance, a word that starts with part of the letter name /kuf/ has to begin with the sound /ku/, like kursa (כורסאarmchair) spelled with /kaf/ (כnot /kuf/ )ק. In contrast, a word that ends with part of this letter name has to end with the rime of the letter name /kuf/, /uf/, like saruf ( שרוףburnt) where /kuf/ is not included in its spelling. We expect that children would erroneously respond that kursa כורסאbegins with /kuf/ קmore often than responding that saruf ורשףends with /kuf/ ק. The body of the letter name would serve as a clue to the syllable forming this name more usefully than the rime of the letter name because of two reasons. First, due to the acrophonic nature of Hebrew letter names, the body of the name includes the sound of the letter, while the rime does not. Second, because Hebrewspeaking children segment CVC syllables primarily to body–coda (CV–C) rather than to onset–rime (C–VC) (Share & Blum, 2005), the phonological connection between the body and the syllable may be more salient than between the rime and the syllable. This implies that children may detect that /ku/ is part of /kuf/ to a greater extent than detecting that /uf/ is part of /kuf/. Consequently, it was assumed that partial letter names would serve as a clue to letter reporting and spelling more in initial than in final positions. The third task used in this study was phoneme isolation. It has recently been argued that letter name knowledge helps children develop phonemic awareness, because alphabetic letters function as visual-graphic symbols of phonemes (Foulin, 2005). This claim is supported by the correlations found between letter naming and phonemic awareness in preschool years (e.g., Evans, Shaw, & Bell, 2000; Sénéchal & LeFevre, 2002) and the reciprocal predictive relationship between them in longitudinal studies from preschool to school (Burgess & Lonigan, 1998; McBride-Chang, 1999; Wagner et al., 1997). The best support comes from an experimental context. Cardoso-Martins et al. (2002) found that children who knew many letters by name and few by sound utilized letters’ phonemic sounds in learning to recognize
60
I. Levin
phonetic abbreviated spellings that did not include any letter name sequence. In the current study, the possibility was explored that when children detect in a word the sounds of a letter name it helps them not only to report the letters but also to isolate the phoneme that this letter stands for. It was expected that this new hypothesis might start explaining the connection between letter name knowledge and phonemic awareness. Individual differences in using letter names for letter reporting or spelling have not been studied. However, children who perform well only when letter name sounds can be detected in the word are relatively poor on phonemic awareness. For this reason, the use of letter names in letter reporting and spelling in groups differing on phonemic isolation was examined. In conclusion, children were tested on letter reporting, letter spelling, and phoneme isolation in words that include full, partial, and phonemic cues of letter names. The letter names detected in the words appeared at the beginning or the end of the words and corresponded either to the congruent or to the incongruent letters. Finally, the impact of letter names was studied in groups of kindergarteners with high, middle, or low phonological awareness.
Method Participants The participants were 62 kindergartners from Tel Aviv, Israel. They included 32 boys and 30 girls with a mean age of 5 years and 11 months (range: 66–78 months). Classroom instruction in letter knowledge, phonological awareness, as well as word reading and spelling were not a part of the curriculum and there was no evidence that teachers paid special attention to these skills while data were being collected. No child who participated was described by the teacher as having special needs; all spoke Hebrew as their mother tongue. Parental written consent was obtained.
Tasks 1. Naming Letters. The children were individually presented with all 27 Hebrew letters, printed in Aharoni font, each displayed on a separate card. Children were then instructed to name each letter. The score was the number of letters named correctly (0–27). 2. Writing Letters. The children were asked to write each of the 27 letters. The score was the number of letters written correctly (0–27). 3. Reporting of Initial Letters in Words. The children listened to 18 words presented to them orally and were asked to report their initial letters. Of the 18 words,
5 The Role of Hebrew Letter Names in Early Literacy: The Case of Multi-phonemic
61
9 started with congruent letter name cues and 9 with incongruent letter name cues. Each group of nine words was composed of three triads of words that started with the same letter but differed by cue length. One word in the triad started with a full letter name cue (CVC), the second word started with a partial letter name cue (CV), and the third word with a phonemic cue (C). To illustrate, one triad of words with congruent letter name cues was spelled with the letter תnamed /taf/. The full letter name cue word was תפקידtafkid (role), which starts with the sound /taf/ (CVC) like the name of the initial letter ת. The partial letter name cue was תלמידtalmid (pupil), which starts with /ta/ (CV) which is a part (the body) of the name of its initial letter. The phonemic cue word תקרהtikra (ceiling) begins with /t/, the onset of the letter name /taf/ (C). The incongruent name cues at the initial position used homophonous letters. For example, a full letter name cue word was קפץkafats (jumped), which starts with the sound /kaf/ (CVC), like the name of the letter כ/kaf/ which is homophonous to the initial letter, named /kuf/ ק. The partial letter name cue was קשרkaSar (tied) which starts with /ka/ (CV), which is the body of the name of a letter homophonous to its initial letter. The phonemic cue word קיבלkibel (received) begins with /k/ (C), the onset of the letter name /kaf/. Note that phonemic cues stand for the initial sounds of both the correct and the homophonous letter, for example, /k/ in /kuf/ and /kaf/, so it facilitates to a similar extent the derivation of either. 4. Reporting of Final Letters in Words. The children listened to 18 words presented to them orally and were asked to report their final letters. Because of the acrophonic nature of Hebrew letter names, words cannot end with a letter whose name is heard at the end of the word, like “deaf” ending with the sound of the last letter /ef/. Therefore, to select words that end with congruent letter name cues we used words that end in letters with symmetrical names, that is, names that start and end with the same phoneme, like /mem/. The Hebrew alphabet contains only four such letters. Nine words ended with congruent letter name cues and nine with incongruent letter name cues. Each group of nine words included three triads of words that ended with the same letter, and differed by cue length, as in the previous task. To illustrate, one triad of words, including congruent letter name cues, was spelled at the end with ם /mem/. The full name cue word was חימםximem (warmed) which ends with the sounds /mem/ (CVC) like the name of the last letter. The partial name cue was שילם /Silem/(paid), ending with /em/ (VC) which is the rime of the name of the letter / mem/. The phonemic cue word עולםolam (world) ends with /m/ (C) the coda of the letter name /mem/. Nine words ended with incongruent letter name cues. Each group of nine words included three triads of words that ended with the same letter but differed by cue length. To illustrate, one triad of words, including incongruent letter name cues, was spelled at the end with ף/fei/ but the sounds of /kuf/ were heard at its end. The full letter name cue word was שקוףSakuf (transparent) which ends with the sound /kuf/ (CVC) unlike the name of the last letter. The partial letter name cue was שרוףsaruf
62
I. Levin
(burnt), ending with /uf/ (VC) which is the rime of the name of the letter /kuf/. The phonemic cue word מגףmagaf (boot) ends with /f/ (C), the coda of the letter name /kuf/. Table 5.1 presents examples of words included in this and in the following tasks. When the last letter in a word was a final letter, like magaf , spelled at the end with /fεi sofit/, the answers /fεi/ and /fεi sofit/ were both accepted. 5. Spelling words. The children were asked to spell 36 words included in the two former tasks of letter reporting. The score was arrived at by calculating the first and the last letter spelled with correct or biased letters in each word, according to the goal of analysis. 6. Isolating Initial Sounds in Words. The children listened to the 18 words used on the task of Reporting of Initial Letters and were asked to isolate their initial phoneme. The responses to the first two items were corrected, if the child erred, with an explanation. For the remaining items correct responses were provided following the children’s error without any explanations. This process was followed because Hebrew-speaking kindergartners often produce CV when asked for the phoneme (Levin, Shatil-Carmon, & Asif-Rave, 2005; Share & Blum, 2005). Two scores were given, strict scores of number of correct phonemes and lenient scores of number of correct phonemes or correct bodies (CV). 7. Isolating Final Sounds in Words. The children listened to the 18 words used on the task of Reporting of Final Letters and were asked to isolate their final phoneme. In all other respects, this task was designed like the previous task described above. Two scores were given, strict scores of number of correct phonemes and lenient scores of number of correct phonemes or correct rimes (VC). The words included in the tasks of letter reporting, spelling, and phoneme isolation are presented in Appendix 5.A. An attempt was made to include only words familiar to kindergartners, but a few unfamiliar words were included when we failed to find familiar words that fulfilled the task demands. All words were Table 5.1 Examples of Words Used for Tests of Letter Reporting, Letter Spelling, and Phoneme Isolation Cue function Cue length – initial position Cue length – final position Congruent Incongruent Congruent Incongruent Letter name /taf/ /kaf/spelled / /mem/ /kuf/spelled /pei/ cue kuf/ tafkid (role) kafats ximem Sakuf (transparent) Full letter (jumped) (warmed) name (CVC) Partial letter talmid (pupil) kaSar (tied) Silem (paid) saruf (burnt) name (CV) Phonemic (C) tikra (ceiling) kibel (received) olam (world) magaf (boot) Number of 3 3 3 3 triads Note: All Hebrew words included here are stressed on the last syllable; Hebrew words are spelled from right to left
5 The Role of Hebrew Letter Names in Early Literacy: The Case of Multi-phonemic
63
disyllabic, except two that were tri-syllabic. The order of words within each task was scrambled for each child.
Design and Procedure The children were tested individually in the classroom, in a quiet atmosphere. Testing was carried out in five sessions, each lasting about 25 min. The children were asked to name letters in session 1. Since writing letters and spelling words can be cumbersome for kindergartners, these tasks were conducted gradually across sessions, as follows: the children were asked to write seven letters in session 1, and five letters in each of the following sessions. The children spelled seven words in each of the first four sessions, and eight words in the last session. The tasks of isolating initial and final sounds in words were carried out in the sessions 2 and 3, in a counterbalanced order. The tasks of reporting initial or final letters in words were carried out in sessions 4 and 5, in a counterbalanced order. Sessions were separated by anywhere from one to a few days.
Results Descriptive Statistics The means, standard deviations, ranges, and reliabilities of the kindergarteners’ performance on all tasks appear in Table 5.2.
Table 5.2 Means, standard deviations, and ranges of performance on all tasks (in percentages) Actual range of performance (Min.–Max.) Cronbach a Task M (SD) Letter naming 77.9 (17.3) 29.6–100 0.88 Letter writing 69.8 (16.0) 33.3–100 0.80 Initial letter reporting: correct/homophonousa 79.2 (17.2) 44.4–100 0.80 Final letter reporting: correct/homophonousa 40.6 (29.7) 0–100 0.94 Initial letter spelling: correct/homophonousa 81.4 (19.0) 33.3–100 0.84 Final letter spelling: correct/homophonousa 39.1 (31.4) 0–100 0.94 Initial phoneme isolation 38.6 (34.3) 0–100 0.94 Final phoneme isolation 48.6 (35.9) 0–100 0.95 Since letter reporting and letter spelling were examined both on correct responses and on homophonous responses, and because responding with either is phonologically justifiable and can be considered appropriate for the age range studied, the general level of performance presented here is based on summed scores of reporting and spelling either the correct or the homophonous letter
a
64
I. Levin
On letter naming and letter writing, mean scores at the lower end of the range were higher than zero because only those children who were able to name and write at least six out of nine particular letters were selected as participants in the study. Concomitantly, mean scores at the lower end were also somewhat higher than zero on the following tasks: initial position of the word, in letter reporting, letter spelling, and body isolation. In contrast, the ceiling or highest possible score in the range was reached across the board, and the variances were wide. These results indicate that the sample varied widely on all tasks.
Letter Reporting, Letter Spelling, and Phonemic Isolation: Correct Responses A four-way ANOVA was carried out on correct responses on the three tasks (letter reporting, letter spelling, and phoneme isolation) of 2 (cue function: congruent and incongruent) × 3 (cue length: full letter name cue, partial letter name cue, and phonemic cue) × 2 (position: initial and final position in word) × 3 (task: letter reporting, letter spelling, and phoneme isolation). Almost all main effects were significant, including a four-way interaction, F (4, 244) = 6.65, p < .001, Eta2 = .098. To simplify the description, a three-way ANOVA was carried out twice. It was decided to analyze the scores on the initial and final word positions tasks separately because the incongruent cues were different in the two different positions. The ANOVAs reported below were carried out twice: once by participants and once by items, followed by Bonferroni tests to clarify the source of the effects. Results almost always coincided. When the two ANOVAs revealed a significant effect, the effect displayed related to participants. When only one analysis showed a significant effect, it was considered nonsignificant and ignored. Initial Position in a Word A three-way ANOVA on correct responses was carried out on words including letter name cues at the initial position in a word. Performance was significantly higher for words including congruent rather than incongruent cues, F (1, 61) = 473.93, p < .001, Eta2 = .886, indicating that congruent cues of letter names facilitate and incongruent cues of letter names hamper performance. The interaction of cue function by task, F (2, 122) = 144.64, p < .001, Eta2 = .703, examined by Bonferroni tests showed that performance was significantly higher for words including congruent rather than incongruent cues on letter reporting and letter spelling but not on phoneme isolation. In fact, phoneme isolation was almost identical on words with congruent and incongruent cues. Performance improved with longer congruent cues: Scores declined significantly from full cues to partial cues and further to phonemic cues. But the effect of cue length was weaker on incongruent cues, reaching significance only on spelling.
5 The Role of Hebrew Letter Names in Early Literacy: The Case of Multi-phonemic
65
These trends were exhibited on the interaction of cue length and cue function, F (2, 122) = 120.86, p < .001, Eta2 = .665. In sum, the results presented above indicate that children’s success on initial letter reporting was close to that on initial letter spelling, and level of performance increased with congruent and decreased with incongruent cues. Cue Length affected similarly performance on these tasks, increasing with longer congruent cues. In contrast, phoneme isolation remained stable on words involving congruent and incongruent cues, and did not change with cue length. Final Position in a Word Results mainly coincided with those on initial position. Performance was higher on words including congruent rather than incongruent cues, F (1, 61) = 125.22, p < .001, Eta2 = .672. Performance on words including congruent cues decreased from full name cue to both partial cue and phonemic cues. Performance was much lower and stable on words including incongruent cues of any length. These trends were exhibited in significant effect of cue length, F (2, 122) = 57.20, p < .001, Eta2 = .484, and the interaction of cue function by cue length, F (2, 122) = 63.07, p < .001, Eta2 = .508. The significant effect of task, F (2, 122) = 16.01, p < .001, Eta2 = .208, the interaction of task by cue function, F (2, 122) = 85.86, p < .001, Eta2 = .585, and the three way interaction, F (2, 122) = 31.65, p < .001, Eta2 = .342, revealed the following picture. In words with congruent cues, cue length affected performance in letter reporting and in letter spelling, decreasing from full name cue to partial name cue and further to phonemic cue. These differences were all significant except on spelling words with partial versus phonemic cues. On words with incongruent cues, cue length had no effect on letter reporting or letter spelling. On phoneme isolation task performance was relatively stable on words with congruent or incongruent cues, on each cue length. In sum, the results presented above indicate that performances on final letter reporting and spelling were higher on congruent than incongruent cues. Cue length affected similarly performance on these tasks, increasing with longer congruent cues. In contrast, phoneme isolation remained stable on words involving congruent and incongruent cues, and did not change with cue length. The Position of Cues To examine the effect of position of cues on children’s performance on the three tasks, a three-way ANOVA of 2 (cue position) × 3 (cue length) × 3 (task), was carried out on congruent and on incongruent cues. The data for these can be found in Tables 5.3 and 5.4. Only results pertaining to cue position are presented. On words including congruent cues, children’s performance on letter reporting and letter spelling was higher when the letters were in the initial rather than the final position.
66
I. Levin
Table 5.3 Percentage of correct responses, on initial position in three tasks for word including congruent and incongruent cues (means and standard deviations, and Bonferroni test results) Cue length Partial name Phonemic Full name (CV) (C) (CVC) Bonferroni test M (SD) M (SD) M (SD) Task Cue function results a a a Initial letter Congruent 96.2 77.4 37.6 CVC > CV > C reporting (12.2) (25.4) (29.2) Incongruent 9.7b 16.7b 24.2b n.s. (18.5) (24.7) (27.1) Initial letter Congruent 88.7a 72.6a 46.8a CVC > CV > C spelling (17.0) (27.3) (25.2) Incongruent 7.5b 20.4b 30.6b C > CVC (15.3) (24.4) (28.5) Congruent 40.9 38.7 36.6 n.s. Initial phoneme isolation (39.3) (36.3) (38.5) 36 40.3 39.2 n.s. Incongruent (40.6) (38.7) (41.2) Note: Pairs with different superscript letters differ significantly between congruent and incongruent cues based on results of the Bonferroni test
Table 5.4 Percentage of correct responses in final position in three tasks on words including congruent and incongruent cues (means, standard deviations, and Bonferroni test results) Cue length Full name Partial Phonemic (CVC) name (CV) (C) Bonferroni test M (SD) M (SD) M (SD) results Task Cue function Final letter Congruent 90.9a 55.4a 22.6 CVC > CV > C reporting (23.5) (33.0) (33.5) Incongruent 14.5b 20.4b 19.4 n.s. (29.3) (32.7) (33.9) Final letter Congruent 72.0a 37.1a 26.8 CVC > CV, C spelling (31.5) (33.6) (35.1) Incongruent 20.4b 21.0b 17.7 n.s. (28.5) (30.3) (28.8) Congruent 42.5 46.2 48.4 n.s. Final phoneme (41.0) (41.1) (40.8) isolation 50.0 55.9 48.4 n.s. Incongruent (39.9) (42.2) (42.5) Note: Pairs with different superscript letters differ significantly between congruent and incongruent cues based on results of the Bonferroni test
This difference was significant on five out of six comparisons: reporting initial letters in words including partial and phonemic cues and on spelling initial letters in words including cues of all lengths. On the phoneme isolation task there was no
5 The Role of Hebrew Letter Names in Early Literacy: The Case of Multi-phonemic
67
difference in performance between initial and final phonemes. These results led to position effect, F (1, 61) = 20.19, p < .001, Eta2 = .249, and position by task interaction, F (2, 122) = 40.11, p < .001, Eta2 = .397. The ANOVA with incongruent cues revealed a host of significant effects: position, F (1, 61) = 4.44, p < .04, Eta2 = .068, interaction of cue position by task, F (2, 122) = 5.01, p < .01, Eta2 = .076, and cue position by cue length F (2, 122) = 16.18, p < .002, Eta2 = .210. However, the discrepancies between performances on initial and final cue positions on letter reporting and spelling were slight and the comparisons between them were all insignificant. Hence, position effects on letter reporting and spelling for words including incongruent cues could not be established with certainty. On phoneme isolation task, final phonemes were more often isolated than initial phonemes on words with full name cue and with partial name cues. In sum, children’s performance on words including congruent cues was higher on letter reporting and letter spelling when the cues were in the initial rather than in the final position. On words including incongruent cues, no clear position effect occurred on letter reporting or spelling, but final phonemes were more often isolated than initial phonemes.
Letter Reporting and Spelling: Biased Responses A direct way to assess the effect of incongruent cues is to analyze children’s responses biased by the incongruent cues. For cues in the initial position, biased responses amounted to producing letters homophonous to the correct letters (e.g., reporting or spelling /kaf/ כas the initial letter of kafats קפץ, though this word is spelled with the letter /kuf/ ק. For cues in the final position, biased responses amounted to producing letters other than the last ones, but whose sounds are heard at the end of the words. For example, a biased response would be providing /kuf/ קas the final letter of /Sakuf/ שקוף, although the word ends with the letter /fei/ ף. On a three-way ANOVA of 3 (cue length: CVC, CV, C) by 2 (cue position: initial and final) by 2 (task: letter reporting and spelling) carried out on biased responses, all main effects and two way interactions were significant. Biased responses decreased significantly from words with full name cues, to words with partial cues and further to words with phonemic cues, F (2, 122) = 213.38, p < .001, Eta2 = .778. The decrease in biased responses with length of cues differed in pace for words with initial versus final cues. For initial cues the decrease looked gradual and was significant between words with full name cues to phonemic cues. For final cues partial name cues led to few biased responses, so that performance decreased significantly from full name cue to both partial and phonemic cues, which were very low and did not differ from each other. These were exhibited in the effect of cue position, F (2, 122) = 259.69, p < .001, Eta2 = .810, and the interaction of cue position and cue length, F (2, 122) = 24.40, p < .001, Eta2 = .286.
68
I. Levin
Table 5.5 Percentage of biased responses on initial and final positions in two tasks (reporting and spelling) for words including incongruent cues Cue length Full name Partial name Phonemic (CVC) (CV) (C) Bonferroni test M (SD) M (SD) M (SD) results Task Cue position Letter Beginning 88.2a 63.4a 28.5a CVC > CV > C reporting (20.1) (31.2) (28.8) End 70.4b 23.7b 1.6b CVC > CV > C (42.3) (29.8) (7.2) Beginning 80.6a 65.1a 31.2a CVC > CV > C Letter spelling (28.0) (31.0) (28.9) 54.3b 3.8b 2.2b End CVC > CV, C (35.8) (10.6) (8.3) Note: Pairs with different superscript letters differ significantly between beginning and end based on results of the Bonferroni test
The effect of Task, F (1, 61) = 14.4, p < .001, Eta2 = .187, indicated that biased responses were slightly more prevalent on letter reporting than on letter spelling. The task by position interaction, F (2, 122) = 9.42, p < .01, Eta2 = .134, and the task by cue length interaction, F (2, 122) = 6.29, p < .01, Eta2 = .094, were hard to interpret since Bonferroni tests revealed no significant difference between tasks either on the initial or on the final position, or on any cue length (Table 5.5). In sum, biased responses appeared to a similar extent on the tasks of letter reporting and letter spelling. Biased responses were more frequent when incongruent cues appeared at the beginning rather than the end of words, and generally decreased with decreasing length of incongruent cue.
Individual Differences: Phonemic Isolation and the Effect of Cue Length When cues were in the initial position of words, the longer the congruent cues the higher was the child’s performance on letter reporting and letter spelling (see Table 5.3). It is possible, however, that the effect of cue length might be mediated by the children’s ability to isolate phonemes. We expected that success on letter reporting and letter spelling would be facilitated by letter name cues, particularly among children who scored lower on phonemic isolation. Children poor on phonemic isolation may fail to report or spell the letter /kaf/ on the word kivsa, a task requiring phoneme isolation, but might succeed on the word kaftor, a task that can be solved by syllable isolation. In contrast, children who are more advanced on phonemic isolation might succeed to a similar extent when performance requires sensitivity to phonemes, sub-syllables, or syllables.
5 The Role of Hebrew Letter Names in Early Literacy: The Case of Multi-phonemic
69
Phonemic isolation was scored by mean performance on initial and final phoneme isolation together, as these scores were highly correlated (r = .74). Children were divided into three almost equal groups on phonemic isolation performance. To pinpoint the interaction between phonemic isolation and the effect of cue length, a three-way ANOVA of 3 (phonemic isolation group: high, middle, and low) × 3 (cue length: CVC, CV, C) × 2 (task: initial letter reporting and initial letter spelling) was carried out on reporting and spelling initial letters either by correct or by homophonous letters – both of which are responses which are justified phonologically. The high group on phonemic isolation outperformed the middle or low group, F (2, 61) = 9.65, p < .001, Eta2 = .247. Cue length interacted significantly with phonemic isolation group, F (4, 118) = 7.35, p < .001, Eta2 = .199, and with task, F (2, 118) = 9.00, p < .001, Eta2 = .132. All other effects were insignificant. Since children’s performance on phonemic isolation did not interact significantly with type of task, comparisons between groups were carried out on children’s scores on the letter reporting and letter spelling tasks combined. Table 5.6 presents the data by different cue lengths and the three phonemic isolation groups across the two tasks of letter reporting and spelling. The high-scoring group on phonemic isolation performed quite similarly on words with different cue lengths; the means were statistically indistinguishable. The middle- and low-scoring groups on phonemic isolation performed significantly better on words with full or partial name cues than on those with phonemic cues. The high-scoring group significantly outperformed the middle- and low-scoring groups only on words with phonemic cues. These results indicate that the middle- and low-scoring groups on phonemic isolation utilized full and partial name cues. Without these cues, their performance substantially decreased. The high-scoring group on phonemic isolation did not seem to need these cues to perform well. Only congruent cues had an effect on letter reporting and spelling final letters (see Table 5.4). Children poor on phonemic isolation may be able to report or spell the letter /mem/ at the end of /ximem/ but fail to do so for /olam/, the former requiring syllabic segmentation and the latter phonemic isolation. Children scoring higher on phonemic isolation may be able to succeed irrespective of cue length.
Table 5.6 Percentage of correct or homophonous responses on initial position in words including congruent and incongruent cues, on letter reporting and letter spelling combined, by phoneme isolation group Cue length Phoneme isolation Full letter Partial letter Phonemic (C) Bonferroni test group name (CVC) name (CV) results High 96.8(8.1) 92.5(12.1) 83.7a(25.3) n.s. scorers(n = 21) CVC, CV > C Middle 97.5(4.8) 84.6(17.6) 50.8 b(29.0) scorers(n = 20) 93.3(8.6) 75.8(24.6) 47.2b(28.4) CVC, CV > C Low scorers(n = 21) Note: Pairs with different superscript letters differ significantly between phoneme isolation groups based on results of the Bonferroni test
70
I. Levin
Table 5.7 Percentage of correct or homophonous responses on final position in words including congruent cues, on letter reporting and letter spelling combined, by phoneme isolation group Cue length Phoneme Full letter Partial letter Phonemic (C) Bonferroni test isolation group name (CVC) name (CV) results High scorers(n = 21) Middle scorers(n = 20) Low scorers(n = 21)
95.2a(9.3)
78.6a(24.8)
72.2a(36.6)
n.s.
88.8a(13.4)
39.2b(20.4)
13.3b(16.8)
69.8b(34.4)
36.5b(32.8)
4.8b(9.3)
CVC > CV, C CVC > C
Note: Pairs with different superscript letters differ significantly between phoneme isolation groups based on results of the Bonferroni test
A three-way ANOVA of 3 (phonemic isolation group) × 3 (cue length) × 2 (task) was carried out on reporting and spelling final letters either by correct or by homophonous letters – both of which are responses which are justified phonologically. Table 5.7 presents mean scores by cue length in the three phonemic isolation groups. The high group on phonemic isolation outperformed the middle or low group on final letter reporting and spelling task, F (2, 61) = 35.73, p < .001, Eta2 = .548. Performance decreased from full name cue to partial name cue and further to phonemic cue, F (2, 118) = 117.95, p < .001, Eta2 = .667. The decreasing performance from full name cue to phonemic cue was significant in the middle- and low-scoring groups on phonemic isolation, but not on the high scoring group, leading to cue length by phonemic isolation interaction, F (4, 118) = 10.52, p < .001, Eta2 = .263. This shows that the high group, unlike the others, did not need the letter name cue to succeed on letter reporting or spelling. The groups differed on all cue lengths, with the high-scoring group on phonemic isolation outperforming the low-scoring group across the board on all cue lengths, and outperforming the middle-scoring group on partial and phonemic cues. These results suggest that, in general, the effects of letter name cues increased with decreasing phonemic awareness. Children’s performance was significantly affected by task, F (1, 59) = 8.90, p < .01, Eta2 = .131, and task interacted significantly with phonemic isolation, F (2, 59) = 5.96, p < .01, Eta2 = .168. However, the discrepancies between tasks were all low and insignificant. Therefore, these findings should be viewed as uncertain. In sum, phonological awareness measured by phoneme isolation (granting partial scores to two-phoneme isolation) had an impact on letter reporting and spelling. In general, children higher on phoneme isolation performed better. Cue length had an effect on performance on letter reporting and spelling – the longer the cue the higher the performance. This effect emerged in two groups – those in the middleand low-scoring groups on phonological awareness. Those high on phonological awareness did not need letter name cues to succeed on the tasks.
5 The Role of Hebrew Letter Names in Early Literacy: The Case of Multi-phonemic
71
Study 2, a follow-up study, is presented below prior to the discussion of the integrated results of both Studies 1 and 2.
Study 2 A major feature of Hebrew letter names is their length. Most names are either monosyllabic, of a CVC structure (e.g., /kaf/), or disyllabic, of a CVCVC structure (e.g., /samex/). In Study 1, only letters with monosyllabic names were used, so that words that included the CVC sounds of a letter name did include, in fact, the full letter name (e.g., kaftor spelled with /kaf/). In Study 2, ten monosyllabic and seven disyllabic letter names were used to derive the words that the children were required to deal with. This design afforded the opportunity to disentangle the effects of cue length per se from that of the ratio of the letter name that this cue encompassed. To clarify, the triad based on the monosyllabic name /kaf/ (kaftor, kartis, and koxav – button, ticket, and star, respectively) included CVC, CV, and C cues that covered 3/3, 2/3 and 1/3 of that letter name’s phonemes, respectively. In contrast, the triad based on the disyllabic letter name /samex/ (samal, salat, and sipur – sergeant, salad, and story, respectively) similarly included CVC, CV, and C cues, respectively, but covered only 3/5, 2/5, and 1/5 of that letter name’s phonemes. The two factors – the cue’s length and its ratio of the letter’s name – were expected to affect children’s ability to derive the letter name. First, children’s performance was expected to decrease from kaftor to kartis to koxav and, likewise, from samal to salat to sipur (as in Study 1). Second, children’s performance was expected to be at a higher level on words where the ratio of the letter’s name was greater; thus, it was hypothesized that children would perform better on kaftor than on samal, on kartis than on salat, and on koxav than on sipur. This hypothesis had never been tested before. It should be noted, however, that comparisons across triads are not as well controlled because children may succeed with one letter name than the other because they are more familiar with the former. To try to control for this, we used as many letter names as possible (17 out of 22 Hebrew letter names), since it has been found that, in general, children are as good at learning to name monosyllabic Hebrew letter as they are at naming disyllabic Hebrew letters (Levin et al., 2005; Treiman et al., 2007). We did not use the remaining five Hebrew letter names because we failed to find appropriately matched triads. It is important to point out that in Study 1 only 3 letters names were involved in each task, whereas in Study 2 the number was expanded to 17, thereby testing the generalizability of the conclusion that could be drawn regarding effect of cue length. A single task was used in Study 2: children were asked to report the initial letter of words including congruent cues. This task was considered representative because the results of Study 1 indicated that children’s performance was higher when the congruent letter name cue was longer across all tasks of letter reporting and spelling. We predicted that the length of letter name cue and the ratio it covers in the letter’s name would affect children’s ability to report the initial letter of spoken words.
72
I. Levin
Method Participants The participants were 34 preschoolers and kindergartners with a mean age of 5 years and 3 months (range: 51–76 months). No participant took part in Study 1. They were recruited from different neighborhoods, ranging from poor to affluent. According to maternal reports, the schools and the homes varied in the significance attributed to teaching children alphabetic skills, including letter knowledge. No child was described by the mother as having special needs and all spoke Hebrew as their mother tongue.
Task Reporting of Initial Letters in Words The children listened to 51 words presented to them orally and were asked to report their initial letters. The words included 17 triads of words that started with the same letter but differed by cue length. Of the 17 triads, 10 started with monosyllabic letter names and 7 with disyllabic letter names. The sounds of the name of the initial letters in each triad were as follows: one word started with CVC sounds, the second with CV sounds, and the third word with a C sound. The words included in each triad were matched on number of syllables and position of stress. Overall, 6 of the 17 triads were stressed on the first syllable and 11 on the second syllable. An attempt was made to include only words familiar to kindergartners, but a few unfamiliar words were included when we failed to find familiar words that fulfilled the task demands. Children who asked what a word meant were responded to with a synonym or a simple definition. The words included in the task of letter reporting, and the percentage of correct responses per word, are presented in Appendix 5.B. To introduce the task, the children were asked to tell the first letter of their own name and were praised for their correct response. Then the experimenter presented orally the list of words, each with normal pacing and stress, and asked the children to tell the first letter of each word. No feedback was provided, except for occasional encouragement. The 51 words appeared in different random orders across children, but we made sure that words from the same triad that started with the same letter never followed each other. Testing took place in the child’s home, in a quiet room, with no one else present in the room except the child and the experimenter. The task was completed in 10–15 min. A pleasant atmosphere was preserved.
5 The Role of Hebrew Letter Names in Early Literacy: The Case of Multi-phonemic
73
Results Table 5.8 presents means (and standard deviations) of correct responses on reporting initial letters of words. As can be seen from the data in Table 5.8, cue length had a substantial effect on children’s performance, which was higher by 39% on words with a CVC cue compared to words with a C cue. The effect of letter name length was smaller, with performance on monosyllabic names surpassing that on disyllabic names by 8%. A two-way ANOVA on 3 (cue length: CVC, CV, C) × 2 (letter name: monosyllabic and disyllabic) was performed twice, once by participants and once by items. The two analyses revealed a significant main effect for cue length, participants: F (2, 66) = 74.11, p < .001, Eta2 = .692, items: F (2, 45) = 26.76, p < .001, Eta2 = .543. Children succeeded in reporting the initial letter most often on words with CVC cues, next on words with CV cues, and least often on words with a C cue, with all differences being significant. The effect of ratio of letter name covered by the cue was less clear, but also appeared in the two analyses. In the analysis by participants, performance was significantly higher on words that started with monosyllabic than with disyllabic letter names, F (1, 33) = 10.34, p < .01, Eta2 = .239; the same effect approached significance on the analysis by items, F (1, 45) = 3.35, p < .07, Eta2 = .069. This difference emerged on words with CVC cues as well as on words with CV cues. A significant interaction between cue length and letter name, F (2, 66) = 7.27, p < .001, Eta2 = .181, emerged only on the analysis by participants. Taking the results of both ANOVAs together, it appears that cue length had a greater effect on words spelled with monosyllabic rather than disyllabic initial letters. This may be so because a cue of the same length is covering a greater ratio of the letter name, with respect to monosyllabic than to disyllabic letter names.
Discussion The results of both studies presented above provide ample evidence that Hebrew letter names are powerful cues for letter reporting and letter spelling in preschoolers and kindergartners. Performance on these two tasks was highly similar with respect to initial letters (see Tables 5.3 and 5.5). Performance improved with longer congruent letter name cues. It was highest on full name cue (CVC), second best on partial name cue formed by the body of the name (CV), and lowest on phonemic cues (C). To illustrate, children responded (orally or in writing) with the correct /taf/ most often on /tafkid/, less so on /talmid/, and least often on /tikra/. In Study 1, which used cues only of monosyllabic letter names, the kindergarteners’ successful reporting or spelling of correct letters increased by about 50% with words that included full name cues relative to phonemic cues (see Table 5.3). In Study 2, children’s reporting of initial letters increased by 46% on monosyllabic letter names, and by 32% on disyllabic letter names (see Table 5.8) probably
74
I. Levin
Table 5.8 Percentage of correct responses, by cue length and letter-name length (means and standard deviations, and Bonferroni test results) Cue length CVC CV C Letter-name M (SD) M (SD) M (SD) Bonferroni test results a a 58.24 32.94 CVC > CV > C Monosyllabic 79.12 (23.01) (28.01) (23.81) 66.39b 45.80b 34.03 CVC > CV > C Disyllabic (26.29) (33.32) (28.79) Note: Pairs with different superscript letters differ significantly between monosyllabic and disyllabic letter-names based on results of the Bonferroni test
because the cue covered a larger ratio of the letter name in monosyllabic letter names. These results taken together suggest that Hebrew letter name cues are highly effective in facilitating letter reporting and spelling, and that their effectiveness depends not only on the cue’s length but also, to some extent, on the ratio of coverage of the letter name. Disyllabic letter names could also have been highly effective had the cue covered the entire letter name. Homophonous responses triggered by incongruent letter name cues of monosyllabic names increased by about 54% in Study 1 (see Table 5.5). This implies that kindergartners have no clue which of two homophonous letters is correct, and they frequently give responses that correspond to the sounds of a letter name detected in the word. This finding expands on Ravid’s (2001, 2004) conclusion that the correct spelling of letters that have homophonous counterparts is mastered only in grade school. The advantage in letter reporting gained by letter name cues in Hebrew seems to be higher than that found in English. For example, Treiman et al. (1996) found that English-speaking preschoolers and kindergartners improved in correct letter reporting by about 11–18%, when asked to provide the initial letter of, say, beach or beaver that start with the sounds of the letter name /bi/, than of bone, that starts with the phoneme /b/. They also found that the reporting of incorrect letter names, for example, Y for wife, was higher than for the control word wait by about 21%. The probable stronger effects of letter names in Hebrew than in English should be accepted cautiously, because it is based on different experiments, with unmatched samples in terms of letter knowledge or phonological awareness. If accepted as valid, the difference in findings could stem from the greater length of full name cues in Hebrew. In this case, the discrepancy in English between full name cue (CV) and phonemic cue (C) might be close to the discrepancy in Hebrew between partial letter name cue (CV) and phonemic cue (C). However, this discrepancy in Hebrew between partial letter name cues and phonemic cues (about 34%) still seems higher than that found in English. The apparent stronger effects in Hebrew may also stem from the acrophonic nature of letter names. Hebrew-speaking children may readily discover the correspondence between the name and sound of a letter because the sound appears in the salient initial position of the name (Levin et al., 2005). According to this argument, alphabets with
5 The Role of Hebrew Letter Names in Early Literacy: The Case of Multi-phonemic
75
acrophonic letter names are advantageous for the discovery of the alphabetic principle because the correspondence between names and sounds of letters are more transparent. Now we turn to final letter names cues. Congruent letter name cues when positioned at the end of the word had the same effect as in the initial position. Children responded with the correct letter (orally or in writing) more often with increased length in the letter name cue. To illustrate, success was highest on ximem, less so on Silem, and lowest on olam, all spelled at the end with the letter /mem/. The discrepancy between performance on the letter reporting and spelling tasks, combined with words having full name cues and words having phonemic cues, was high, about 57%. However, because of the acrophonic nature of Hebrew letter names, when a letter name is detected at the end of a word, this letter can, indeed, only be the last letter in cases where it has a symmetric name, like /mem/. Since only four Hebrew letters names are symmetrical, this occurrence is probably rare. Mostly, when letter names can be detected at the end of a word, the words can include these letters, but they are not the word’s last letters. For instance, Sakuf that ends with the sounds of /kuf/, includes the letter /kuf/ in its spelling but not in the final position. Consequently, incongruent letter name cues at the final position can be viewed as not entirely misleading. Children exposed to the print of /Sakuf/ can detect the sound of /kuf/ and discover this letter in the spelling of the word, though the oral and written forms of /kuf/ are not in exactly the same position. Partial letter name cues strongly affected letter reporting and spelling when in the initial position of the word, regardless of whether the cue was congruent or incongruent. In a slightly different pattern, partial letter name cues strongly affected performance when in the final position of the word if the cue was congruent; however, when the cue was incongruent, like /uf/ corresponding to the rime of /kuf/ in the word saruf, the effect was weak. In such cases, children responded rarely with the biased letter /kuf/ (about 14%). Moreover, writing /kuf/ as the final letter of saruf might also stem from incomplete spelling of the word. Indeed, ending the spelling of a word with a letter marking the consonant preceding the last, like /reish/ for saruf was just as common (about 16%) across tasks (unreported above). The lack of effect of partial wrong letter names in the final position of the word probably stems from two factors: (1) because Hebrew letter names are acrophonic, the rime of a letter name (VC), unlike the coda (CV), does not include the sound that the letter stands for. Listening to saruf, a child cannot detect /k/, the sound of the letter /kuf/, unlike /ef/ that can be detected in “deaf.” Moreover, as persuasively shown by Share and Blum (2005), Hebrew speakers split CVC syllables primarily to body–coda (CV–C) rather than to onset–rime (C–VC), hence the phonological connection between the body and the syllable may be more salient than between the rime and the syllable. Letter names had no effect on phoneme isolation. Children’s success in isolating the initial phoneme was independent of whether the word started or ended with full, partial, or phonemic cues. It is possible that this occurred because Israeli kindergartners are poor at recalling the phonemic sound that letters stand for (Levin et al., 2005) and, therefore, the name of a letter is not a useful clue for its phonemic sound. If this argument holds water, letter name cues can facilitate phonemic
76
I. Levin
isolation in contexts teaching children letters primarily by their sounds, like they do currently in Great Britain (Caravolas et al., 2001). This hypothesis deserves to be further tested to deepen our understanding of the function of letter names. Still, phonemic isolation explained individual differences in using letter names for reporting or spelling letters. Children high on phonemic isolation were as good in responding with the correct letter on words including full, partial, or phonemic cues, whether at the beginning or the end of the word (see Tables 5.6 and 5.7). Children who fell in the middle- or low-scoring levels in phonemic isolation were more successful when scaffolded by full or partial letter names. This finding deserves to be studied in other languages as well. It implies that preschoolers and kindergartners, often limited in phonemic awareness, could benefit from working on words including letter name cues. The results of this study show that, by and large, Hebrew letter names are highly effective cues in helping children to report and spell letters. The fact that Hebrew letter names are relatively long does not limit their usefulness in helping children to embark on literacy. On the contrary, longer names are probably advantageous. In addition, the acrophonic nature of Hebrew names most likely enhances their contribution. These conclusions call for extending the type of research conducted here to other alphabets with different qualities. Acknowledgments I am very grateful to Bracha Nir-Sagiv for her contribution to the analysis of a corpus of words in children’s books, for counting letter name cues. Thanks are also extended to Yafit Melamed and Danit Ben-Amar for their involvement in the empirical study.
קשר tevakeS (Will ask)
קפץ tetaken (Will repair)
קיבל torid (Will bring down)
(C) Saxor (Black) שחור kivsa (Ewe) כבשה tikra (Ceiling) תקרה kibel (Received)
(CV) Siul (Cough) שיעול kartis (Ticket) כרטיס talmid (Pupil) תלמיד kaSar (Tied)
(CVC) Sinuj (Change) שינוי kaftor (Button) כפתור tafkid (Role) תפקיד kafats (Jumped)
Cue length – final position
צבט Sotef (Fluent) שוטף
ענף
מגף tsavat (Pinched)
(C) olam (World) עולם aSan (Smoke) עשן paSut (Simple) פשוט magaf (Boot)
Phonemic
בועט anaf (Branch)
Letter name Full letter Partial letter cue name name (CVC) (VC) /mem/ ximem Silem (Warmed) (Paid) ם מחמם שילם /nun/ tsinun tsijun (Cold) (Score) ן צינון ציון /tet/ xitet Saket (Dug) (Quiet) ט חיטט שקט /kuf/ Sakuf saruf Spelled /fei/ (Trans-parent)(Burnt) ()פי ף שקוף שרוף /xet/ Soxet boet Spelled /tet/ (Butcher) (Kicks)
()תף ()טת ת תתקן תבקש תוריד ט סוחט /taf/ tafsan tardan tilon /taf/ kataf Spelled /tet/ (Scaffolder) (Trouble-maker) (Popsicle) Spelled /fei/ (Picked) ()טת ()פי טפסן טרדן טילון קטף ט ף Note: All these Hebrew words are stressed on the last syllable; Hebrew words are spelled from right to left
Incongruent
/Sin/ ()שין ש /kaf/ ()כף כ /taf/ ()תף ת /kaf/ Spelled /kuf/ ()קוף ק /tet/ Spelled /taf/
Partial letter name Phonemic
Letter name cue
Congruent
Full letter name
Cue length – initial position
Cue function
Appendix 5.A. Words Used for Tests of Letter Reporting, Letter Spelling, and Phoneme Isolation, in Study 1
5 The Role of Hebrew Letter Names in Early Literacy: The Case of Multi-phonemic 77
Mono-syllabic
Letter name length
יוד /kaf/ ()כף כ /mem/ ()מם מ /kuf/ ()קוף ק /reS/ ()ריש
Letter name cue /bet/ ()בית ב /vav/ ()וו ו /xet/ ()חית ח /jud/ ()יוד
CVC ‘beten (Belly) בטן vavim (Hooks) ווים xetef’ (Snatch) חטף judrax (Will-beguided) יודרך kaftor (Button) כפתור memSala (Government) ממשלה kufsa (Box) קופסה ‘reSet (Net)
Cue length – initial position CV ‘berez (Faucet) ברז varod (Pink) ורוד xevel’ (Rope) חבל jugaS (Will-beserved) יוגש kartis (Ticket) כרטיס medina (State) מדינה kulmus (Pen) קולמוס ‘rexev (Car) ישתה koxav (Star) כוכב miSpaxa (Family) משפחה kotzim (Thorns) קוצים ‘ra’al (Poison)
C ‘boker (Morning) בוקר vilon (Curtain) וילון xoSex’ (Darkness) חושך jiSte (Will drink)
70.6
26.5
61.8
73.5
79.4
76.5
82.4
91.2
58.8
73.5
38.2
79.4
91.2
67.7
CV 64.7
CVC 73.5
Percentage of correct responses
Appendix 5.B. Words Used for Test of Letter Reporting, in Study 2 (n = 34)
47.1
20.6
44.1
17.7
5.9
29.4
61.8
C 35.3
78 I. Levin
ר רשת רכב רעל /Sin/ Sinuj Sikun Sa’on 85.3 73.5 55.9 שין (Change) (Neighborhood) (Watch) ש שינוי שיכון שעון /taf/ tafkid targil tinok 73.5 32.4 11.8 ()תף (Role) (Exercise) (Baby) ת תפקיד תרגיל תינוק Di-syllabic /gimel/ gimur giSur gozal 58.8 50.0 23.5 ()גימל (Finish) (Bridging) (Chick) ג גימור גישור גוזל /dalet/ dalik dagim doker 70.6 38.2 26.5 ()דלת (Flammable) (Fish) (Pricking) ד דליק דגים דוקר /zajn/ ‘zajit ‘za’am ‘zevel 70.6 64.7 47.1 ()זין (Olive) (Rage) (Trash) ז זית זעם זבל /lamed/ ‘lama ‘laka ‘lexem 76.5 61.8 41.2 ()למד (Why) (Nail polish) (Bread) ל למה לכה לחם /samex/ samal salat sipur 47.1 50.0 38.2 ()סמך (Sergeant) (Salad) (Story) ס סמל סלט סיפור /ajin/ ajir’ Ako’ ovesh’ 64.7 20.6 8.8 ()עין (Donkey) (Acre) (Mold) ע עיר עכו עובש /tsazik/ tsadu tsamu tzemer 76.5 35.3 52.9 ()צדיק (Hunted) (Fasted) (Wool) צ צדו צמו צמר Note: All Hebrew words marked by ‘ at the beginning are stressed on the first syllable. Other words are stressed on the last syllable; stress had no significant effect on frequency of correct responses either as a main effect or in interaction with cue length. Hebrew words are spelled from right to left.
5 The Role of Hebrew Letter Names in Early Literacy: The Case of Multi-phonemic 79
80
I. Levin
References Bowman, M., & Treiman, R. (2002). Relating print and speech: the effects of letter naming and word position on reading and spelling performance. Journal of Experimental Child Psychology, 82, 305–340. Burgess, S. R., & Lonigan, C. J. (1998). Bidirectional relations of phonological sensitivity and prereading abilities: evidence from preschool sample. Journal of Experimental Child Psychology, 70, 117–141. Bruck, M., Genesee, F., & Caravolas, M. (1997). A cross-linguistic study of early literacy acquisition. In B. Blachman (Ed.), Foundations of reading acquisition and dyslexia: implications for early interventions (pp. 145–162). Mahwah, NJ: Erlbaum. Byrne, B. (1998). The foundations of literacy: the child’s acquisition of the alphabetic principle. Hove, United Kingdom: Psychology Press. Caravolas, M., Hulme, C., & Snowling, M. J. (2001). The foundations of spelling ability: evidence from a 3-year longitudinal study. Journal of Memory and Language, 45, 751–774. Cardoso-Martins, C. (1995). Sensitivity to rhymes, syllables, and phonemes in literacy acquisition in Portuguese. Reading Research Quarterly, 30, 808–828. Cardoso-Martins, C., Lemos, L. S., Corrêa, M. F., & Napoleão, R. F. (2006). Is there a syllabic stage in the development of spelling? Evidence from Brazilian Portuguese-speaking children. Journal of Educational Psychology, 98, 628–641. Cardoso-Martins, C., Resende, S. M., & Rodrigues, L. A. (2002). Letter name knowledge and the ability to learn to read by processing letter-phoneme relations in words: evidence from Brazilian-Portuguese-speaking children. Reading and Writing, 15, 409–432. de Abreu, M. D., & Cardoso-Martins, C. (1998). Alphabetic access route in beginning reading acquisition in Portuguese: the role of letter-name knowledge. Reading and Writing, 10, 85–104. de Jong, P. F., & van der Leij, A. (1999). Specific contributions of phonological abilities to early reading acquisition: results from a Dutch latent variable longitudinal study. Journal of Educational Psychology, 91, 450–476. Ehri, L. (1992). Reconceptualizing the development of sight word reading and its relationship to recoding. In P. Gough, L. Ehri & R. Treiman (Eds.), Reading acquisition (pp. 107–143). Hillsdale, NJ: Erlbaum. Ehri, L. (1998). Learning to read and learning to spell are one and the same, almost. In C. Perfetti, L. Rieben & M. Fayol (Eds.), Learning to spell (pp. 237–269). Mahwah, NJ: Lawrence Erlbaum Associates. Ehri, L., & Wilce, L. (1985). Movement into reading: is the first stage of printed word learning visual or phonetic? Reading Research Quarterly, 20, 163–179. Elbro, C., Borstrom, I., & Petersen, D. K. (1998). Predicting dyslexia from kindergarten: the importance of distinctness of phonological representation of lexical items. Reading Research Quarterly, 33, 36–60. Evans, M. A., Shaw, D., & Bell, M. (2000). Home literacy activities and their influence on early literacy skills. Canadian Journal of Experimental Psychology, 54, 65–75. Foulin, J. A. (2005). Why is letter-name knowledge such a good predictor of learning to read? Reading and Writing, 18, 129–155. Levin, I., Patel, S., Margalit, T., & Barad, N. (2002). Letter-names: effect on letter saying, spelling, and word recognition in Hebrew. Applied Psycholinguistics, 23, 269–300. Levin, I., Shatil-Carmon, S., & Asif-Rave, O. (2005). Learning of letter names and sounds and contribution to word reading. Journal of Experimental Child Psychology, 93, 139–165. Lonigan, C. J., Burgess, S. R., & Anthony, J. L. (2000). Development of early literacy and early reading skills in preschool children: evidence from a latent-variable longitudinal study. Developmental Psychology, 36, 596–613. McBride-Chang, C. (1999). The ABC’s of the ABC’s: the development of letter-name and lettersound knowledge. Merrill-Palmer Quarterly, 45, 285–308.
5 The Role of Hebrew Letter Names in Early Literacy: The Case of Multi-phonemic
81
McBride-Chang, C., & Treiman, R. (2003). Hong Kong Chinese kindergartners learn to read English analytically. Psychological Science, 14, 138–143. Muter, V., Hulme, C., Snowling, M., & Taylor, S. (1997). Segmentation, not rhyming, predicts early progress in learning to read. Journal of Experimental Child Psychology, 65, 370–396. Pennington, B. F., & Lefty, D. L. (2001). Early reading development in children at family risk for dyslexia. Child Development, 72, 816–833. Ravid, D. (2001). Learning to spell in Hebrew: phonological and morphological factors. Reading and Writing, 14, 459–485. Ravid, D. (2004). Hebrew orthography and literacy. In R. M. Joshi & P. G. Aaron (Eds.), Handbook of orthography and literacy (pp. 339–363). Mahwah, NJ: Lawrence Erlbaum Associates. Scarborough, H. S. (1998). Early identification of children at risk for reading disabilities: phonological awareness and some other promising predictors. In B. K. Shapiro, P. J. Accardo & A. J. Capute (Eds.), Specific reading disability: a view of the spectrum (pp. 75–119). Timonium, MD: York Press. Schatschneider, C., Fletcher, J. M., Francis, D. J., Carlson, C. D., & Foorman, B. R. (2004). Kindergarten prediction of reading skills: a longitudinal comparative study. Journal of Educational Psychology, 96, 265–282. Scott, J., & Ehri, L. (1990). Sight word reading in prereaders: use of logographic vs. alphabetic access routes. Journal of Reading Behavior, 22, 149–166. Senechal, M., & LeFevre, J.-A. (2002). Parental involvement in the development of children’s reading skill: a five-year longitudinal study. Child Development, 73, 445–460. Share, D. L., & Blum, P. (2005). Syllable splitting in literate and pre-literate Hebrew speakers: onsets and rimes or bodies and codas. Journal of Experimental Child Psychology, 92, 182–202. Share, D. L., Jorm, A. F., MacLean, R., & Matthews, R. (1984). Sources of individual differences in reading acquisition. Journal of Educational Psychology, 76, 1309–1324. Shimron, J. (2007). Tnu lahem likro. Hed Hachinuch, 81, 58–63. Let them read, The Echo of Education, 81, 58–63. (Hebrew) Snowling, M. J., Gallagher, A., & Frith, U. (2003). Family risk of dyslexia is continuous: individual differences in precursors of reading skills. Child Development, 74, 358–373. Stage, S. A., & Wagner, R. K. (1992). The development of young children’s phonological and orthographic knowledge as revealed by their spellings. Developmental Psychology, 28, 287–296. Storch, S. A., & Whitehurst, G. J. (2002). Oral language and code-related precursors to reading: evidence from a longitudinal model. Developmental Psychology, 38, 934–947. Tolchinsky, L. (2003). The cradle of culture and what children know about writing and numbers before being taught. Mahwah, NJ: Lawrence Erlbaum. Treiman, R., Berch, D., & Weatherston, S. (1993). Children’s use of grapheme-phoneme correspondences in spelling: roles of position and stress. Journal of Educational Psychology, 85, 1–12. Treiman, R., Levin, I., & Kessler, B. (2007). Learning of letter names follows similar principles across languages: evidence from Hebrew. Journal of Experimental Child Psychology, 96, 87–106. Treiman, R., & Rodriguez, K. (1999). Young children use letter names in learning to read words. Psychological Science, 10, 334–338. Treiman, R., Sotak, L., & Bowman, M. (2001). The roles of letter names and letter sounds in connecting print and speech. Memory and Cognition, 29, 860–873. Treiman, R., Tincoff, R., & Richmond-Welty, E. D. (1996). Letter names help children connect print and speech. Developmental Psychology, 32, 505–514. Wagner, R. K., Torgesen, J. K., Rashotte, C. A., Hecht, S. A., Barker, T. A., Burgess, S. R., et al. (1997). Changing relations between phonological processing abilities and word-level reading as children develop from beginning to skilled readers: a 5-year longitudinal study. Developmental Psychology, 33, 468–479.
Chapter 6
The Early Childhood Project: A 5-Year Longitudinal Investigation of Children’s Literacy Development in Sociocultural Context Susan Sonnenschein, Linda Baker, and Robert Serpell
Abstract The Early Childhood Project examined the intimate culture of young children’s homes, defined by a confluence of parental beliefs, recurrent activities, and interactive processes, and the relation between that intimate culture and the children’s literacy development. The children in the 5-year longitudinal study came from low- and middle-income families of European-American and AfricanAmerican heritage. Families joined the project when the children were in prekindergarten or kindergarten. Results showed the importance of parental beliefs and children’s home experiences to children’s literacy development. Although low and middle income children had somewhat different home literacy experiences, the intimate culture of the home was a more powerful predictor of literacy development than demographic factors such as family income and ethnicity. Children coming from an intimate culture that emphasized an entertainment approach, one that focuses on making interactions enjoyable and engaging for young children, were more likely to be involved in affectively positive reading interactions with their caregivers, to engage in frequent interactions with age-appropriate text, and, consequently to benefit in the development of literacy competencies. Many children in literate societies such as the United States struggle to learn to read, and a sizeable percentage fail to master all but the most basic skills. A recent assessment using a nationally representative sample from the United States found that over 30% of fourth graders do not read well enough to understand grade level texts (National Assessment of Educational Progress, 2005). Children from lowincome or minority backgrounds are more often found among those who are not successful (Snow, Burns, & Griffn, 1998).
S. Sonnenschein (*), L. Baker, and R. Serpell University of Maryland, Baltimore County, Baltimore, MD, USA e-mail:
[email protected] D. Aram and O. Korat (eds.), Literacy Development and Enhancement Across Orthographies and Cultures, Literacy Studies 101, DOI 10.1007/978-1-4419-0834-6_6, © Springer Science+Business Media, LLC 2010
85
86
S. Sonnenschein et al.
The Early Childhood Project was motivated by the desire to extend our understanding of literacy development in children from different sociocultural backgrounds. This chapter focuses on one aspect of the project – the relation between children’s home experiences and their literacy development. Home experiences included the beliefs parents held about how their children learn and their own role in such learning, as well as the nature and frequency of children’s engagement in literacy-relevant activities. Other components of the project addressed teachers’ beliefs about how to foster literacy development, the nature and frequency of children’s activities at school, the relations between teachers’ and parents’ beliefs about learning, and the similarities and differences in how teachers and parents viewed the children. A complete description of the Early Childhood Project was published in our monograph, Becoming Literate in the City (Serpell, Baker, & Sonnenschein, 2005).1 The study took place in Baltimore, Maryland, from 1992 to 1997. Baltimore, located on the eastern seaboard of the United States, is the largest city in Maryland with a population of about 743,000. According to the 1990 US census data, 59% of Baltimore residents were black, 39% were white. Forty percent of the adult residents of the city reportedly did not graduate from high school. The Early Childhood Project followed the lives of a cohort of children enrolled in Baltimore’s public schools. The majority of the children attending Baltimore City Public Schools were black. There is a fairly high mobility rate for students enrolled in the city’s public school system – about 25% of children in low-income schools transfer to another school or withdraw from the system each year. As is common in many city school systems, a large percentage of children attending the Baltimore City Public Schools earn low scores on state-administered reading and math assessments. Many do not graduate. For example, 63% of the students who should have graduated in 1994 did not. The children participating in the project were growing up in low- and middleincome families of European-American and African-American heritage. Children were in prekindergarten at the start of the project and in third grade at the end. Of particular interest were the literacy-related home experiences of children traditionally more at-risk for educational difficulties, that is, low-income or African-American children. The theoretical framework was influenced by systems theory (Bronfenbrenner, 1979), conceptualizations of the eco-cultural niche of child development (Super & Harkness, 1997), neo-Vygotskian theory (Rogoff, 1990), and emergent literacy perspectives (e.g., Sulzby & Teale, 1991). We hypothesized that the historical legacy of racial and economic oppression in America has placed low-income and AfricanAmerican children at risk for school failure by isolating their families from cultural practices and beliefs that match the demands of the school curriculum. We sought to document in detail how the developmental niche of different social addresses within the city of Baltimore structures children’s early opportunities for participatory appropriation of the cultural practice of literacy. Although income level and ethnicity were taken into account in sample selection and data analyses, social address was found to
Several of the findings described in this chapter appeared in separate publications as well as in our monograph.
1
6 The Early Childhood Project: A 5-Year Longitudinal Investigation of Children’s
87
carry less explanatory power than the intimate culture of a child’s home, defined as a confluence of parental beliefs, recurrent activities, and interactive processes. This chapter begins with an overview of the techniques used to understand parents’ beliefs about their role in their children’s learning and to document the nature of children’s experiences and literacy development. The Intimate Culture section consists of two parts. The first highlights the nature of parents’ beliefs about how to foster literacy development and the types of activities children engaged in. The second part covers qualitative aspects of children’s literacy interactions with their family members. The Children’s Literacy Competencies section begins with a summary of our findings of literacy development of the children in our study and goes on to relate growth in competencies to the children’s home experiences.
Methods Used in the Early Childhood Project Our recruitment strategy was designed to maximize the chance of recruiting those families with children traditionally most at risk for educational difficulties. Such families often are reluctant to participate in research and are difficult to recruit. In addition to the common practice of sending home letters to parents of children enrolled in the target schools, we scheduled in-person meetings before and after school. We also tried to meet informally with parents when they dropped their children off or picked them up from school. We reasoned that we would be more successful in recruiting the historically most reluctant parents through less formal means of contact than in large, less individualized group meetings. To make families feel comfortable, we used recruiters who were matched on ethnicity with the target family. We succeeded in recruiting 81 African-American and European-American families from low-income or middle-income backgrounds. Due to the timing of our funding, about half the families were recruited when children were in prekindergarten (N = 40) and the remainder (N = 41) when the children were completing kindergarten. A subset of 63 children and their families remained in the study through the end of third grade. The percentage of families who did not continue is consistent with attrition rates in other longitudinal studies (e.g., Leseman & de Jong, 1998). We conducted annual in-depth interviews with children’s parents about their beliefs and practices. We observed literacy-related interactions between children and parents, children and their siblings.2 Each spring we assessed children’s literacy competencies using a range of measures. When children were in prekindergarten and kindergarten we assessed their orientation toward print, phonological awareness, and narrative competencies. When they began first grade, we included standardized measures of decoding, word recognition, and reading comprehension (the We discuss here only methods used for collecting information about children’s home experiences and their literacy competencies. See Serpell et al. (2005) for methods used to document teachers’ beliefs and practices. 2
88
S. Sonnenschein et al.
latter was assessed only in third grade). Included in our measures were several that assessed children’s functional knowledge of print. These latter tasks allowed us to investigate how children fared on tests more tailored to their individual experiences. The scope of the study was noteworthy in several respects. We assessed a wider array of literacy competencies than is common. We documented experiences children had involving oral and written forms of language. We noted with whom children interacted when they engaged in various activities and observed several literacy interactions. We coded not only the cognitive–linguistic aspects of these literacy interactions (e.g., the nature of the talk) but also the affective quality of the interactions. We extensively probed parents’ beliefs about children’s literacy development, an area that has been understudied. In doing so, we took great care to make our interviews relevant and understandable for parents whose own literacy skills were limited. We also focused on children’s reading motivation, another understudied topic. Throughout the study we interwove qualitative and quantitative approaches to data collection and analysis.
The Intimate Culture of a Child’s Home Parents’ Beliefs and Children’s Activities The intimate culture of a child’s home was documented through in-depth interviews with parents, typically the mothers, about their beliefs and practices. Immediately after enlisting a family, parents were asked to keep a 1-week semi-structured diary of their children’s activities. Parents were asked to record all activities that occurred within a time period (e.g., getting up activities and so on). We purposely did not tell them to include only literacy activities so as not to bias responses. Subsequent interviews began by grounding questions in practices noted in the diaries. Parents completed at least one in-depth semi-structured interview each year addressing their beliefs about children’s literacy appropriation. For example, during one of the interviews from the first year of the project, parents were asked a series of questions about what they can do to assist their children in acquiring age-appropriate cognitive, academic, and social competencies. Of particular interest for this chapter were responses to the question, “What is the best way to help your child learn to read?” Each year parents also completed an ecological inventory probing the nature and frequency of their child’s participation in a broad range of literacy-relevant activities. The specific activities were modified as the children got older to reflect age-appropriate experiences. All of the interviews and questionnaires were administered in person with the parents. The diaries were quite variable in length and content. Some contained few entries with limited descriptive information, whereas others were very detailed. The diaries were coded for the types of literacy activities mentioned as well as whether they reflected an entertainment or a skills perspective in context. The
6 The Early Childhood Project: A 5-Year Longitudinal Investigation of Children’s
89
coding was adapted from work by Teale (1986) and Goldenberg, Reese, and Gallimore (1992) who observed directly the print-related activities that occurred in the homes of children from low-income backgrounds. Entertainment activities included shared book reading, independent initiating of looking at or reading books, playing with print, and so on. Skills activities were those that focused on practicing literacy skills by using flashcards or workbooks for learning letters, letter sounds, or words. Responses to the question “What is the best way to help your child learn to read” were similarly coded for an entertainment or skills perspective. An entertainment perspective focused on making the interaction enjoyable for the child, whereas a skills perspective emphasized inculcating skills. Scores for an entertainment home orientation and a skills home orientation were computed for each family by averaging scores from the diary and interview question. Note that we also used a second means of assessing parent’s beliefs about fostering literacy. We asked parents to rate the importance for young children’s literacy development of participating in various common, everyday activities. Some of the activities were more consonant with an entertainment approach and others with a skills approach. The pattern of findings for these two means of analyzing parents’ beliefs (responses to the “What is the best way …” question and rating of activities) about how to facilitate children’s literacy development was similar. Parents expressed consistent beliefs about how to help their children become literate. Some parents emphasized the importance of making reading experiences enjoyable and allowing the child to play a role in initiating activities, whereas others focused more on teaching skills directly. More low-income parents than middle-income ones emphasized skills inculcation. These findings are consistent with those of other researchers who have found systematic differences in literacy-learning beliefs among parents from different sociocultural backgrounds (e.g., Fitzgerald, Spiegel, & Cunningham, 1991; Reese & Gallimore, 2000). For example, Goldenberg et al. (1992) interpreted their findings with low-income Hispanic families in Los Angeles as demonstrating that parents believed that “learning to read (not necessarily the act of reading itself) is a process of learning decoding and other skills, not a process driven by children’s interest in making meaning from texts ….” (p. 504). The types of literacy activities parents offered their children reflected their beliefs about how to facilitate children’s literacy development. Parents who favored an entertainment orientation focused on making reading experiences enjoyable and allowing the child to play a role in initiating activities. For example, a middleincome European-American mother stated, “If I picked out the book, I picked out a book that I thought was a good story, something that I would enjoy too at that age. You know things we could enjoy together.” These parents reported more frequent engagement with activities consonant with such an approach, such as reading storybooks, chapter books, playing games involving print, and telling stories. Parents who advocated an approach more oriented toward skills talked about having their children use workbooks and flash cards. The parent of a low-income EuropeanAmerican child talked about “Just getting him to practice.”
90
S. Sonnenschein et al.
A similar relation between the nature of beliefs espoused by parents and children’s activities was found by Lynch, Anderson, Anderson, and Shapiro (2006) among parents of preschoolers in western Canada. Parents who endorsed what Lynch and colleagues called holistic beliefs were more likely to report their children played literacy-related games and read with them; parents with skills-based beliefs were more likely to report they taught their children to recognize letters and write their names and that their children used workbooks. We turn next to findings from the Early Childhood Project about the types of literacy activities children engaged in. Several years ago, Purcell-Gates (2000) noted that almost all children growing up in the United States (and presumably other industrialized countries) have frequent exposure to print. Exposure itself is not the issue. What is important is the nature and frequency of the exposure and how it changes with development, because reading challenging and age-appropriate text is important for growth in reading competencies (Baker, Dreher, & Guthrie, 2000; Byrnes, 2000). The results from our study confirm the need to document the type of exposure children have, given the consistent sociocultural differences we found. Children from middle-income European-American families in the Early Childhood Project had more frequent experiences than low-income or AfricanAmerican children with texts that are more conducive to literacy development, that is, storybooks and chapter books. In contrast, African-American children more frequently read less developmentally advanced text, such as picture books, even when they were in third grade. Differences in the frequency with which children read certain types of text were evident in prekindergarten. Almost all the parents reported that their prekindergarten children read storybooks at least periodically, but families differed in the frequency with which children read such texts. One hundred percent of the middle-income children did this on a daily basis whereas 45% of the low-income children did. Although most studies of reading interactions have focused on the impact of storybook reading, reading preschool books (ABC books) has been shown to be more strongly related to children’s acquisition of knowledge about print (Bus & van IJzendoorn, 1988). More of the middleincome prekindergarteners than the low-income ones read preschool books at least weekly. Another way we considered the import of children’s home experiences was by selecting two groups of children with different literacy competencies at the end of the study and comparing their reading activities over the course of the project. We compared a group of children who were reading on or above grade level at the end of third grade with a group reading at least a year below grade level. We excluded children whose scores fell between the cut-offs for the two comparison groups. There were consistent differences in the frequency with children in the two groups read and the type of text read. Children who read on or above grade level at the end of third grade engaged in daily, age-appropriate reading activities at home. They read storybooks and preschool books (e.g., ABC books) in prekindergarten, and by second grade they
6 The Early Childhood Project: A 5-Year Longitudinal Investigation of Children’s
91
read chapter books. In contrast, children who read below grade level at the end of third grade were far less likely to have engaged in daily reading activities. More lowincome children, particularly African-Americans, were represented in the below grade level group.
The Nature of Children’s Engagement with Literacy-Relevant Activities Much of the early research on literacy-related home experiences focused on the frequency of storybook reading (Bus, van IJzendoorn, & Pellegrini, 1995). Researchers found a weak but significant relation between reading activity and literacy acquisition, such that children who were read to more frequently had higher scores on measures of emergent literacy. This type of correlation, however, does not tell us anything about the importance of what occurs during reading activities. The focus more recently has been on the nature of the adult–child interactions, most typically the type of conversation that takes place during shared storybook reading. Researchers have examined the amount of conversation that goes beyond the factual elements in the text to include the non-immediate context as well as talk about aspects of the print. Snow (1994) has suggested that non-immediate talk is pertinent for more advanced language development as well as understanding text. As previously noted talk about print facilitates decoding and word recognition skills. In the Early Childhood Project, we observed children engaging in storybook reading interactions with others on two different occasions, at the beginning of kindergarten and again toward the end of first grade. Both observations were videotaped and took place within the children’s homes. During kindergarten, children were observed reading a book of their own choosing with the person who most typically read with them. Most children read with an adult, typically their mother. Slightly less than a third of the children were read to by an older sibling. The difference in reading partners allowed us to consider any differences in the interactions as a function of the age of the reader. Most researchers studying reading interactions have observed adult–child dyads. However, siblings may play a larger role in socializing their younger siblings in low-income than in middle-income families (Heath, 1983). Thus, it is important to extend observations of reading interactions to sibling dyads. During first grade, we provided a book unfamiliar to the children and asked that the mothers or primary caregivers read it with their child. Interactions were coded for the nature of conversation about the book, including talk about immediate and non-immediate context and talk about print, and the affective quality of the interactions. Although many other researchers have considered the type of talk that occurs in reading interactions, few have considered the social-affective quality (cf., Bus & van IJzendoorn, 1988; Leseman, & de Jong, 1998). Videotapes were reviewed for observable behaviors that reflected an enjoyable, engaging interaction. The mother’s
92
S. Sonnenschein et al.
behavior was coded for reading expression, physical and visual contact with the child, appearance of engagement in the activity, and sensitivity to the child’s engagement. The child’s behavior was coded for the appearance of engagement. In addition to describing qualitative aspects of the reading interactions, we examined relations between the interactions and components of literacy development. In addition to measures already discussed, we also assessed children’s reading motivation, a predictor of frequency of engagement as well as a literacy-related outcome in its own right. Reading motivation was assessed with a measure, developed for this study, which addressed the multidimensional nature of motivation. That is, children may be motivated to read for a variety of reasons – because it is viewed as enjoyable, is a valuable skill, affords an opportunity for social interaction and receiving praise from others. Children were asked to choose which of two statements about reading motivation was more similar to their own views. That is, they were shown two puppets, Regal and Cha-Cha, read descriptions of each puppet and asked to choose which puppet they were more like. For example, “Regal likes to read but Cha Cha doesn’t like to read. Who are you more like?” After choosing one of the puppets the child was asked, “Are you a lot like Regal/Cha-Cha, or just a little?” Children were read 16 statements tapping the four components of reading motivation: enjoyment, perceived value of reading, perceived competence in reading, interest in library-related activities. The most common type of talk during the storybook reading interactions was about the immediate context (e.g., factual elements about the story). Forty-five percent of remarks were coded as being about the immediate context. Such remarks were positively related to children’s motivation for reading in first grade. Consistent with what others have found, there was little discussion about print during storybook reading interactions when children were entering kindergarten or were in first grade. Discussion about print typically occurred when children read to their mothers and made reading errors (in first grade). The affective quality of reading interactions when children were entering kindergarten was related to children’s motivation for reading in first and second grade which predicted the frequency with which children read challenging text. The affective quality of reading interactions in first grade was negatively related to the frequency with which print was discussed. Affective quality also was related to the frequency with which children engaged in chapter book reading in second and third grade which, in turn, predicted reading competencies in third grade. The affective quality of the reading interactions was positively and significantly related to an entertainment home orientation. Low-income children were less likely to experience shared storybook reading with an adult than were middle-income children; instead, they were read to by their older siblings. The difference in reading partners has implications for subsequent outcomes. The siblings were less fluent readers than the children’s parents and were less skilled at maintaining the interest of their listeners or refocusing their wandering attention. In addition, the affective quality of the interaction was poorer, with less enjoyment exhibited by reader and listener alike.
6 The Early Childhood Project: A 5-Year Longitudinal Investigation of Children’s
93
Children’s Literacy Competencies When children were in prekindergarten and kindergarten, we assessed their letter knowledge, concepts about print (knowledge of how to hold a book, turn a page), functions of print (ability to label types of print commonly found in their home and neighborhoods and state how these were used), phonological awareness, and nursery rhyme knowledge. We also assessed their ability to answer questions about stories and relate a story of their own. When children were in elementary school we either modified the measure to make it more age-appropriate or chose ones tapping more age-appropriate skills. For example, the functional print tasks now required the child to sort mail, look up on a calendar the day their birthdays occurred and complete a coupon to receive a sticker. We also administered standardized measures of decoding, word recognition, and reading comprehension from the Woodcock Johnson Tests of Achievement-Revised (Woodcock & Johnson, 1989). There was great variability in the scores earned by different children. For example, consider scores at the end of third grade, the last time we assessed children’s literacy competencies. Children’s scores on a word recognition measure administered at the end of the school year ranged from a grade equivalency of 1.0 to 11.9. Forty-seven percent of the children were struggling readers; that is, they received grade equivalency scores at least one grade below expectations. Low-income children earned significantly lower scores on standardized measures of word recognition and reading comprehension than middle-income children, particularly European-American children. The performance differences among children from different social addresses were first evident in prekindergarten, even on the functional print measures that were tailored to children’s individual experiences. Such group-related differences were even greater on more traditional literacy-related measures, that is, those not tailored to the child’s own experiences. Children’s phonological analysis skills assessed in prekindergarten and kindergarten, as well as their phonemic analysis skills assessed in grades 1–3, were related to their word recognition and reading comprehension at the end of third grade. Scores on the functional print tasks administered in elementary school also were highly related to children’s word recognition scores. A major source of difficulty for low-income children as they learn to read appears to be mastery of the skills needed to decode the text rather than comprehension. Although we found differences in children’s word recognition favoring middle-income children, particularly European-American children, we found no such differences in children’s oral story comprehension. In fact, when the same children were asked to retell a story that had just been read to them, low-income African-American children produced better retellings than low-income EuropeanAmerican and middle-income African-American children. This may be an illustration of the oral narrative competence observed by Heath (1983) that is not typically assessed or rewarded in formal schooling.
94
S. Sonnenschein et al.
The Relation Between Aspects of the Intimate Culture and Children’s Emerging Competencies We turn next to a consideration of how the intimate culture of the home, which includes parents’ beliefs and the experiences provided for their children, relates to literacy development. Indicators of the home orientation included the composite score from the diary activities and parents’ responses to the question about how best to foster young children’s literacy development. The data strongly support the notion that if the intimate culture is more consonant with the theme that literacy is a source of entertainment, children’s literacy development is facilitated. An entertainment perspective was significantly and positively related to phonological awareness (prekindergarten), orientation toward print (prekindergarten and kindergarten), narrative competence (kindergarten), word recognition (grades 1, 2, 3), reading comprehension (grade 3), and children’s use of functional print (grades 1, 2). In all cases (even when the correlation failed to reach significance), the correlation between an entertainment orientation and the specific literacy competency was positive. In contrast, the correlation between a skills orientation and the specific literacy competency was generally either near zero or negative. Although we have discussed differences in children’s outcomes related to their social addresses, such demographic factors are not causally related to children’s literacy development but rather are proxies for relevant factors. For example, we conducted a series of regression analyses of factors predicting word recognition and reading comprehension when children were in third grade. Once we entered aspects of the intimate culture into regression equations, no additional variance was accounted for by demographic factors such as income or ethnicity. It is the nature of a child’s experiences rather than the family’s social address that has a direct effect on literacy development.
Summary The Early Childhood Project was motivated by the desire to extend our understanding of the relations between children’s home experiences and their literacy development. The study examined the intimate culture of each child’s home, defined by a confluence of parental beliefs, recurrent activities, and interactive processes, and it explored the relation between that intimate culture and the child’s literacy development. The children came from low- and middle-income families of EuropeanAmerican and African-American heritage. An important contribution of the project was its demonstration that parental beliefs and qualitative aspects of children’s home experiences do make a difference in children’s literacy development. Although low- and middle-income children had somewhat different home literacy experiences, consistent with previous research, the intimate culture of the home was a more powerful predictor of literacy development than demographic factors such as family
6 The Early Childhood Project: A 5-Year Longitudinal Investigation of Children’s
95
income and ethnicity. Children coming from an intimate culture that emphasized what we have called an entertainment approach, one that focuses on making interactions enjoyable and engaging for young children, were more likely to be involved in affectively positive reading interactions with their caregivers, to engage in frequent interactions with age-appropriate text, and, consequently to benefit in the development of literacy competencies. In closing we want to note that home experiences and emergent literacy competencies prior to formal schooling are important foundations for subsequent literacy growth, but a child’s developmental potential is not fixed at school entry. We identified a group of children whose early literacy skills were in the bottom quarter of the sample but who were in the top third by the end of third grade. In contrast to those children who remained in the bottom quarter, children whose literacy trajectories changed (improved) engaged in daily and varied reading activities at home during elementary school. In addition to reading books, they more frequently participated in storytelling activities, went to the library, and played with educational toys. These findings show the important role that parents play during preschool and early elementary school in guiding their children’s literacy development. Equally importantly, the findings have implications for ways to assist families in facilitating their children’s development. What children do at home and with whom they do it is important for literacy development; it even can help overcome a slow start toward literacy acquisition. Acknowledgments This chapter is based on a portion of the material presented in our book, Serpell, R., Baker, L., & Sonnenschein, S. (2005), Becoming literate in the city: The Baltimore Early Childhood Project. NY: Cambridge. We received funding for this study from the National Institute of Child Health and Human Development (R01 HD29737-0A1) and the National Reading Research Center (contracted by the Office of Educational Research and Improvement/ PR/Award no. 117A20007). We are very grateful to the families who participated in this project and the many graduate and undergraduate students who assisted us.
References Baker, L., Dreher, M. J., & Guthrie, J. T. (2000). Why teachers should promote reading engagement. In L. Baker, M. J. Dreher & J. T. Guthrie (Eds.), Engaging young readers: promoting achievement and motivation (pp. 1–16). New York: Guilford. Bronfenbrenner, U. (1979). The ecology of human development. Cambridge, MA: Harvard University Press. Bus, A. G., & van IJzendoorn, M. H. (1988). Mother-child interactions, attachment, and emergent literacy: a cross-sectional study. Child Development, 59, 1262–1272. Bus, A. G., van IJzendoorn, M. H., & Pellegrini, A. D. (1995). Joint book reading makes for success in learning to read: a meta-analysis on intergenerational transmission of literacy. Review of Educational Research, 65, 1–21. Byrnes, J. P. (2000). Using instructional time effectively. In L. Baker, M. J. Dreher & J. T. Guthrie (Eds.), Engaging young readers: promoting achievement and motivation (pp. 188–208). New York: Guilford. Fitzgerald, J., Spiegel, D. L., & Cunningham, J. W. (1991). The relationship between parental literacy level and perceptions of emergent literacy. Journal of Reading Behavior, 23, 191–213.
96
S. Sonnenschein et al.
Goldenberg, C., Reese, L., & Gallimore, R. (1992). Effects of literacy materials from school on Latino children’s home experiences and early reading achievement. American Journal of Education, 100, 497–536. Heath, S. B. (1983). Ways with words: language, life and work in communities and classrooms. Cambridge, England: Cambridge University Press. Leseman, P. P. M., & de Jong, P. F. (1998). Home literacy: opportunity, instruction, cooperation and social-emotional quality predicting early reading achievement. Reading Research Quarterly, 33, 294–318. Lynch, J., Anderson, J., Anderson, A., & Shapiro, J. (2006). Parents’ beliefs about young children’s literacy development and parents’ literacy behaviors. Reading Psychology, 27, 1–20. National Assessment of Educational Progress. (2005). The nation’s report card. Washington, DC: National Center for Educational Statistics. Purcell-Gates, V. (2000). Family literacy. In M. L. Kamil, P. B. Mosenthal, P. D. Pearson & R. Barr (Eds.), Handbook of reading research (Vol. 3, pp. 853–870). Mahwah, NJ: Erlbaum. Reese, L., & Gallimore, R. (2000). Immigrant Latinos’ cultural model of literacy development: an evolving perspective on home-school discontinuities. American Journal of Education, 108, 103–134. Rogoff, B. (1990). Apprenticeship in thinking: cognitive development in social context. New York: Oxford University Press. Serpell, R., Baker, L., & Sonnenschein, S. (2005). Becoming literate in the city: The Baltimore Early Childhood Project. New York: Cambridge University Press. Snow, C. E. (1994). Enhancing literacy development: programs and research perspectives. In D. K. Dickinson (Ed.), Bridges to literacy: children, families, and schools (pp. 267–272). Cambridge, MA: Basil Blackwell. Snow, C. E., Burns, M. S., & Griffin, P. (1998). Preventing reading difficulties in young children. Washington, DC: National Academy Press. Sulzby, E., & Teale, W. H. (1991). Emergent literacy. In R. Barr, M. L. Kamil, P. Mosenthal & P. D. Pearson (Eds.), Handbook of reading research (Vol. 2, pp. 727–758). New York: Longman. Super, C. M., & Harkness, S. (1997). The cultural structuring of child development. In J. W. Berry, P. R. Dasen & T. S. Saraswathi (Eds.), Handbook of cross-cultural psychology, Vol. 2: Basic processes and developmental psychology (Vol. 2, pp. 1–39). Boston: Allyn & Bacon. Teale, W. H. (1986). Home background and young children’s literacy development. In W. H. Teale & E. Sulzby (Eds.), Emergent literacy: writing and reading (pp. 173–205). Norwood, New Jersey: Ablex. Woodcock, R. W., & Johnson, M. B. (1989). Woodcock-Johnson psycho-educational battery – revised. Allan, TX: DLM Teaching Resources.
Chapter 7
Early Literacy at Home: General Environmental Factors and Specific Parent Input Catherine McBride-Chang, Yvonne Y. Y. Chow, and Xiuli Tong
Abstract This chapter outlines three aspects of early literacy in relation to general environment and parenting. First, we overview some general characteristics of the home environment that can influence how (and what) children learn to read and write. Next, we overview more specifically how parents can facilitate children’s early literacy growth. These studies focus particularly on dialogic reading and shared parent-child writing. Finally, we highlight the importance of both phonological awareness and morphological awareness for children’s early reading and writing development. Children’s early cognitive skills interact with their general environment, including the central role of parents and other caregivers, in affecting early literacy growth across cultures. Iris Levin has contributed substantially to our understanding of early literacy skill development across cultures. Her research tends to borrow from the best of traditions in both cognitive development and educational studies. Although her research spans a variety of topics and developmental periods, we have benefited most from her work on literacy development in children’s transition from prereaders to readers. Our own work is just beginning to uncover some factors in parent–child interactions that may be useful for developing early literacy skills in young children and potentially help them in using these skills to facilitate literacy development as they progress in school. In this chapter, we will make three points about literacy development in children. First, some aspects of the general home environment may be important for children’s reading success. Second, the more focused characteristics of parent–child reading and writing themselves clearly contribute to this success. Third, certain metalinguistic abilities may facilitate reading across cultures, and parents can potentially help in fostering some of these.
C. McBride-Chang (*), Y.Y.Y. Chow, and X. Tong The Chinese University of Hong Kong, Hong Kong e-mail:
[email protected]
D. Aram and O. Korat (eds.), Literacy Development and Enhancement Across Orthographies and Cultures, Literacy Studies 101, DOI 10.1007/978-1-4419-0834-6_7, © Springer Science+Business Media, LLC 2010
97
98
C. McBride-Chang et al.
General Characteristics of Home Environment Several studies have underscored the importance of general home literacy characteristics for children’s reading success. These include the income level of the parents, the educational levels of parents, particularly mothers, and the quantity and types of reading materials available at home. To these, we could also add parents’ encouragement of reading and interest in books, home languages, and their literacy beliefs. On the whole, a child’s home literacy environment can be conceptualized as the experiences and attitudes about literacy a child encounters within the home (Leseman & de Jong, 1998; Whitehurst & Lonigan, 1998). The income level in households has been one focus of research by Levin and colleagues (e.g., Aram & Levin, 2001), and it is strongly linked to reading achievement even into adolescence (e.g., Chiu & McBride-Chang, 2006). The importance of income with regard to reading is multifaceted. At an extreme, poverty tends to be associated with poor nutrition, lack of health care, premature birth, and increased exposure to lead, all of which are, in turn, associated with learning difficulty (e.g., UNICEF, 2001; Valenzuela, 1997; Vernon-Feagans, Hammer, Miccio, & Manlove, 2001). Income level is also often correlated with education level of parents, and this link is important for understanding some facets of achievement and motivation within and across cultures. In many countries, mothers with low education levels tend to have children with somewhat lower reading achievement (Aram & Levin, 2001; Rauh, Lamb-Parker, Garfinkel, Perry, & Andrews, 2003). Conversely, higher levels of educational attainment in mothers tend to be associated with a particular approach to interacting with children that may stimulate the young learners through the use of questioning (e.g., Tracey & Young, 2002). Moreover, parents with higher levels of education tend to believe and transfer the concept that literacy activities are both useful and fun (Sonnenschein, Baker, Serpell, Scher, Truitt, & Munsterman, 1997), potentially stimulating more direct focus on reading and writing activities over time. Apart from these basic features of a literacy environment, the issue of books in the home is important to consider. For example, a study by Korat and Levin (2001) showed the differences in the number of books owned by relatively poorer and wealthier families. Among children from low SES families, there were approximately 124 adult books and 51 children’s books on average. In contrast, among children who were from middle-level SES families, the average number of books owned per family was 309 (adult books) and 86 (children’s books), respectively. Clearly, there was likely some effect of income on book ownership, but the contrast across families in this context is impressive. Moreover, in a large-scale study involving 41 countries by Chiu and McBride-Chang (2006), we found that the number of books in the home explained the differences in reading among adolescents, even controlling for culture, family and school, socioeconomic status, and a variety of other variables. The issue of causality in these associations is obviously a crucial point. One cannot expect that having a lot of books in the home will magically influence children’s reading performance. However, these data suggest that a higher value for reading among other activities, with the income level of the family statistically controlled, is associated with a better
7 Early Literacy at Home: General Environmental Factors and Specific Parent Input
99
reading performance even into the adolescent years. Perhaps some book ownership represents an ethos of reading in the home, which can be quite positive for children. In another of our own studies in Hong Kong (Lau & McBride-Chang, 2005), we also found that second graders with more Chinese books at home tended to be better readers, even with the vocabulary knowledge and maternal education statistically controlled. Apart from these, children’s self-efficacy in reading was also a unique correlate of literacy skill. Self efficacy was basically a subscale of our questionnaire about reading that tapped how the child felt about himself or herself in relation to being good at reading and writing. As has been shown across many contexts, there is likely an interaction between feeling competent in something and actually doing it. As far as home environment goes, the more children experience a positive environment of reading, the better they seem to read. Overall, parents’ literacy beliefs also play an important underlying role in fostering children’s literacy development. Parents who engage in frequent literacy activities tend to create an enthusiastic environment for learning (Weigel, Martin, & Bennett, 2006). For example, children of mothers who were categorized in one study (Weigel, Martin, & Bennett, 2006) as either “facilitative” or “conventional” with regard to literacy beliefs showed different patterns of interest in reading development. “Facilitative” mothers were those who believed in the importance of parents providing their children with learning opportunities, such as spending time reading with their children at home. “Conventional” mothers were those who viewed schools as playing the most important role in teaching children, such that they themselves tended to spend less time engaging in shared literacy activities with their children. Interestingly, children whose mothers held facilitative beliefs were found to demonstrate greater interest in reading books and greater print knowledge than did those with conventional beliefs; this difference continued at least for a year. Thus, attitudes of both children and adults are important for children’s literacy development across cultures. One other aspect of environment that is important in Hong Kong, the home of the present authors, is exposure to languages at home. Most Hong Kong Chinese parents want their children to excel in learning English, but this is an arduous task for many families. Although English lessons are typical for Hong Kong Chinese children from age 3 onwards, the level of competence in English is not always high among Hong Kong Chinese students, perhaps in part because English is learned as a school subject divorced from any clear practical context. With this background, we examined the role of English exposure at home to vocabulary knowledge in kindergarten children in Hong Kong (Chan & McBride-Chang, 2005). We compared kindergartners (mean age = 5.7 years) from the same schools, where English was taught daily. All spoke Cantonese as their native tongue and had Cantonese-speaking parents. However, one group was cared for at least part time by domestic helpers from the Philippines, all of whom spoke English to the families for whom they worked, and the other group was cared for only by Cantonese-speaking caregivers (usually parents and grandparents). The group who lived and talked with Filipina helpers showed significantly better English vocabulary knowledge but significantly poorer Chinese vocabulary knowledge than did the group who spoke only Cantonese at home. Since vocabulary knowledge tends to be a strong correlate of early literacy (e.g., Wagner, Muse, & Tannenbaum,
100
C. McBride-Chang et al.
2007), these results further underscore the importance of home environment for beginning reading. Growth in language skills among children will vary depending upon the languages to which they are exposed at home, whether spoken by a native or second language speaker.
Parent–Child Literacy Although these correlational analyses are suggestive of some important aspects of home environment for children’s literacy development, they cannot get at causality. Our laboratory has completed three separate studies of parent–child reading in Hong Kong (Chow & McBride-Chang, 2003; Fung, Chow, & McBride-Chang, 2005; Chow, McBride-Chang, Cheung, & Chow, 2008) that suggest causality through use of an experimental design. Adopting the technique of Whitehurst and colleagues (e.g., Whitehurst, Arnold, Epstein, Angell, Smith, & Fischel, 1994; Whitehurst et al., 1988; Whitehurst, Zevenbergen, Crone, Schultz, Velting, & Fischel, 1999), we have used dialogical reading successfully on children between the ages of 5 and 9 in different investigations. Dialogic reading is a way for parents to share storybooks with their children such that children are encouraged to relate their own experiences and ideas to the books and eventually become more involved in book reading. Whereas traditional storybook sharing typically involves parents reading each page to their children and basically getting through the book, often minimizing discussion about the story, the goal of dialogic reading is to facilitate discussion. In adopting dialogic reading for the children in our studies, our overarching goal was to get children to talk more with their parents about several facets of the books. More conversations about the books tend to promote greater diversity of language and more discussions across varied topics. In Hong Kong particularly, with its strong emphasis on educational achievement even among the youngest children, who begin formal education at age 3, a lot of conversations between parents and children tend to focus on the “correctness” of the answer. Familiar questions in this category might include the following: “What color is the bird?” “How many gifts did she receive?” “Where is the treasure hidden now?” Based on the framework of the story, presumably each of these questions can be answered either correctly or incorrectly. Parents tend to be happy when children get the answers correct and to engage in more instruction when children’s responses are incorrect. In the philosophy of dialogic reading, such questions are in some ways preferable to forced choice questions, such as “Is the bird red?” “Does she have three gifts there?” or “Is the treasure buried under the sea?” because the technique emphasizes the superiority of open-ended answers, which make use of more vocabulary knowledge and may demand more syntactical skills than do simple yes/no answers. On the other hand, however, an even better approach to shared parent–child reading is a focus on broader experiences in relation to the story. Questions about these sorts of issues might center on evaluating or anticipating thoughts or feelings of the characters in the story (e.g., “How do you think the little bear feels now that his father is
7 Early Literacy at Home: General Environmental Factors and Specific Parent Input
101
back home? Why?”), relating the experience of the child to the situation in the book (e.g., “Tell me about a time when you had a really good dream. What was it about?”), or asking the child to anticipate the next move in the story (e.g., “What do you think the boy will do next? What would you do if you were in that situation?”). In all of our studies on dialogical reading, we trained parents in advance on the art of questioning. We gave them a list of different types of questions that could be asked (e.g., who, what, where, when, etc. – Whitehurst et al., 1988) or generally applied to any story and taught them to use these. During our research studies, we also tried to help parents to remember to use these techniques, so for every new book children received during our studies, parents found one or more suggested questions (strategically placed inside each book using small sticky papers) to help. However, parents were encouraged to think of their own questions during reading time and not to be limited by ours. We had a variety of questions to be addressed in our studies. To begin with, we wanted to understand the extent to which this technique would work for Hong Kong Chinese parents and children. Although the basic principle of the technique should be universally applicable across languages, we had to make a few adjustments. For example, all printed text is generally written in standard Chinese, or Mandarin, in Hong Kong. Mandarin is a different language from Cantonese. For example, it is pronounced using 4 tones, rather than the estimated 6–12 needed for Cantonese. It also has a somewhat different grammatical structure and different vocabulary from Cantonese. Thus, we had to decide what our focus was for the purpose of measuring success in dialogic reading, that is, whether we wanted to focus on Mandarin or Cantonese. Based on previous research (e.g., Whitehurst et al., 1988; Whitehurst et al., 1994), it seemed to make more sense to have parents use and discuss the stories using the vocabulary and structure of Cantonese. We also wanted to understand whether this technique would improve not only vocabulary but also aspects of print knowledge, such as character recognition. A final issue of interest in the present study was the extent to which dialogic reading could improve language and literacy even in hearing impaired children, so we modified our approach a bit for families with such children, primarily by using pictures as an aid for discussions in parent–child dyads. In all three of our studies, we consistently used the same approach to measuring the effectiveness of dialogic reading on children’s literacy development. All three of our studies yielded stronger vocabulary knowledge in the children in the group whose parents practiced the dialogic reading technique as compared to the group in which parents simply read as they normally would. Vocabulary knowledge in these studies involved word comprehension, that is, pointing to the picture that best demonstrated a given vocabulary word from among four choices. In all of these experiments, which lasted between 8 and 12 weeks (Chow & McBride-Chang, 2003; Fung, Chow, & McBride-Chang, 2005; Chow et al., 2008), families received one new book per week across groups (except for the control group, which received all books and some training on the dialogic reading technique only after the period of the study). Even within these short time periods, children of parents using the dialogic reading technique fared better than others, particularly in vocabulary knowledge-building. Parents also tended to be happy with the technique, to have perceived an increase in
102
C. McBride-Chang et al.
reading interest in their children when they used the technique, and to have applied this technique even in other readings they did with their children outside of the study. We were particularly pleased that these results held even for children with hearing impairments, whose hearing difficulties differed substantially from one another. Educators and parents in Hong Kong have been attracted to our website (http://www.psy.cuhk. edu.hk/~qefproject/) documenting these effects. We are delighted to report that as of this time, we continue to have approximately 1,000 hits on the site every month, a mark of practical success and interest in the community. However, results of these particular studies yielded inconsistent findings in relation to word recognition. That is, some children in the dialogic reading group tended to improve relative to their peers even in word reading over the time of the study whereas others did not. These results, in addition to the findings of Sénéchal and LeFevre (2002), who noted the importance of actually focusing on print in order to foster print awareness in children, a process that does not always occur during parent–child reading, and a visit from Iris Levin to our laboratories in Hong Kong in 2006, prompted us to become interested in shared parent–child writing as well. Aram and Levin (Aram, 2007; Aram & Levin, 2001, 2004) have pioneered in the area of exploration of mother–child joint writing. They have demonstrated strong links of their “Maternal Mediation” and “Print Mediation” scales to children’s concurrent and subsequent literacy development in Israel. Their demonstration that mothers’ strategies for promoting word-writing in their preschoolers are linked to subsequent reading and writing performance in children 2.5 years later, even controlling for previous writing skills (Aram & Levin, 2004), suggests that this specific focus on how parents scaffold children’s writing is essential for understanding early literacy development. Our laboratory learned a great deal about these ideas during a visit that Iris Levin made to Hong Kong in 2006. She convinced us that this approach would be particularly interesting to study in Chinese. We were also intrigued because, traditionally, Chinese parents and educators have emphasized rote learning of Chinese characters as the best way to learn to read and write (e.g., Packard et al., 2006; Wu, Li, & Anderson, 1999). Chinese characters seem to be complicated to learn, and there are thousands of them relative to the 27 letters of Hebrew. Chinese characters are each comprised of strokes, which form various shapes to make radicals. Indeed, 90% of modern Chinese characters are compound characters, consisting of both a semantic and a phonetic radical, which serve to cue the meaning and sound information of a given character. At the same time, however, there is relatively little reliability in the phonological information conveyed by the phonetic radicals in characters to generalize to new characters, making the task of learning all of these graphemes taxing, at least in relation to most alphabetic systems (e.g., Cheung & Ng, 2003). Hence, it seems that the emphasis by teachers and parents on memorization of Chinese by repeated writing seems to be a reasonable approach. We wondered, though, whether Chinese mothers would ask their children to do anything else apart from copying in order to learn to write new words in Chinese. Our findings (Lin et al., 2009) were fascinating in relation to this issue: We found that a focus on copying was most prevalent among mothers of the youngest children
7 Early Literacy at Home: General Environmental Factors and Specific Parent Input
103
and became less used with age in children ages 4–7 years. In addition, more analytic strategies, focusing on components of Chinese characters (i.e., explicitly telling children that the character should be divided into a phonetic radical and a semantic radical and explaining the sound and meaning of these, respectively) and the morphological and orthographic information contained within the Chinese words (e.g., pointing out that two words shared the same character and explaining the meaning of the character) increased with age. Perhaps most revealing for Chinese traditionalists might be the fact that, among these 67 children, with children’s ages, grade levels, nonverbal IQs, and mothers’ education levels statistically controlled, these two strategies were uniquely linked to Chinese word reading, in opposite directions. That is, more copying in these mother–child dyads was linked to poorer word reading, whereas more focus on the analytic properties of word reading was linked to better word reading. There are obvious pedagogical implications of this work. In particular, it may be that if we continue to observe similar future associations of these maternal strategies in relation to word-writing, additional training studies for parents might be useful. For example, training mothers to focus more on analytic aspects of word-writing and less on pure copying of words might help facilitate their children’s success in early literacy activities. The pioneering work by Levin and colleagues in writing and other literacy skills (e.g., Aram & Levin, 2001, 2004; Levin & Bus, 2003; Levin, Ravid, & Rapaport, 2001; Levin, Shatil-Carmon, & Asif-Rave, 2005) has encouraged educators and researchers around the world, our laboratory included, to explore a variety of cognitive factors related to literacy development. For example, beyond shared reading and writing experiences at home, our laboratory has also focused on individual cognitive skills that might contribute to reading development. These are reviewed below as a third component of the home environment that may contribute to early literacy success.
Metalinguistic Skills and Literacy Development We have focused particularly on two metalinguistic skills that may facilitate reading across cultures. These are phonological awareness and morphological awareness. Both are important for reading success, but their importance likely depends upon the language and orthography in question as well as the developmental period of the child. Because this chapter focuses on early literacy development in particular, we talk primarily about these effects for young children below. Phonological awareness is awareness of and access to a language’s sound system. Children who can successfully divide words by syllable, rime, or phoneme or, alternatively, who can synthesize these speech sounds to make words (e.g., /k/-/ae/-/t/ together makes “cat” in English) tend to have more success in reading at a young age. This success is particularly interesting because the ability to divide or synthesize sounds does not require print knowledge – only oral language skills. Phonological awareness appears to be quite important across a wide range of alphabetic orthographies, at least in very young children, because of the alphabetic principle, the idea
104
C. McBride-Chang et al.
that individual letters of the alphabet represent individual sounds. Thus, if children grasp the principle that each letter stands for a sound, it becomes much easier to read by synthesizing sounds represented by these letters together (for a review, see McBride-Chang, 2004). Moreover, training of phonological awareness, even in the absence of print, facilitates young children’s reading development (Bus & van IJzendoorn, 1999). Despite the importance of phonological awareness for reading across orthographies, including Chinese (e.g., Chow, McBride-Chang, & Burgess, 2005; Ho & Braynt, 1997; Siok & Fletcher, 2001), however, we have focused in the past couple of years particularly on the development of morphological awareness in relation to early reading. Morphological awareness involves awareness of and access to the meaning system in words, particularly sensitivity to morphemes, or single units of meaning within a word. Morphological awareness appears to be quite important for reading development in both alphabetic and Chinese orthographies, but the form in which it is measured differs substantially between the two. Morphological awareness appears to be associated with better reading even in young children because morphemes in words, primarily inflectional and derivational markers, provide clues to their meanings. Thus, if particular morphemes were spelled phonemically rather than morphemically, such clues would not exist (Balmuth, 1992). An example in English would be a comparison of the words “know” and “knowledge.” If these words were, alternatively, spelled “noe” and “nollij,” respectively, readers might not be able to grasp the overlap in these words’ meanings (Balmuth, 1992). In alphabetic scripts, morphological awareness typically involves using some meaningful word parts such as base roots, prefixes, and suffixes, to understand meanings in new words (Baumann et al., 2002). For example, we found that first grade Dutch children’s knowledge of inflectional morphology was a unique correlate of their word reading skills, apart from phonological awareness and math knowledge (Rispens, McBrideChang, & Reitsma, 2008). In Korean 4–5-year olds, we also found that morphological awareness was uniquely associated with word-reading, but only among irregular words (Cho, McBride-Chang, & Park, 2008). Because Korean is a very phonologically transparent orthography, regular words need not be accessed using morphological rules; children can easily grasp these via phonological processing only. However, for certain irregular words that cannot be decoded based on phonological knowledge alone, knowledge of the semantic content of the morpheme (similar to the English example above of the route word “know,” which is pronounced differently than it is spelled) is useful. Interestingly, this particular study focused on morphological awareness in the form of lexical compounding, unlike most other studies of morphological awareness in alphabetic languages. Lexical compounding is the compounding of two words to form a compound word. An example of this in English would be putting “cow” together with “boy” to form “cowboy.” Given the correlational evidence of the importance of morphological awareness in alphabetic orthographies, further studies have examined the extent to which training in morphological awareness facilitates reading, just as phonological awareness has been demonstrated to facilitate reading in previous work (e.g., Bus & van IJzendoorn, 1999). One study conducted in Norway focused on the effects of morphological and
7 Early Literacy at Home: General Environmental Factors and Specific Parent Input
105
phonological awareness training for kindergartners’ reading development. Consistent with previous findings (Elbro & Arnbak 1996; Lundberg, Frost, & Peterson, 1988), it was found that training in both metalinguistic skills was linked to reading achievement (Lyster, 2002). Interestingly, the effects of this training differed according to children’s metalinguistic skills: Children whose phonological awareness was particularly high tended to benefit more from morphological awareness training than from phonological awareness training. In contrast, for children with initially low levels of phonological awareness, morphological awareness training was not particularly beneficial. Although these findings are not unique for Norwegian only, these findings were likely attributable in part to how children are taught to read the Norwegian language. This alphabetic orthography tends to be relatively regular phonologically, such that many words can be decoded using a phonological route only. However, as morphological awareness becomes deeper, children’s knowledge of morphology can be an additional benefit in facilitating reading. These findings likely generalize across orthographies such that morphological awareness provides “added value” for reading words. In Chinese, early morphological awareness may focus less on inflections and derivations and more on lexical compounding, or compound word formations. Although lexical compounding likely occurs to a certain extent across languages, it is much more common in Chinese than in most other languages and forms one basis for organizing much of Chinese reading and writing (e.g., McBride-Chang, Shu , Zhou, Wat, & Wagner, 2003; Packard et al., 2006). Most Chinese words consist of two or more morphemes, and this composition helps to disambiguate each morpheme (e.g., McBride-Chang et al., 2005). Just as an English speaker would not know whether the word “sun” pronounced in isolation referred to the concept of “son” as in “grandson” or “sun” as in “Sunday,” in Chinese, compound words help to clarify the meanings of each morpheme. Chinese is also special in that there is a direct correspondence between syllable, morpheme, and character. All three are represented in the same unit. This is quite unlike English, for example, where morphemes can be long (e.g., tomato) or short (e.g., s in cats). Both because Chinese makes extensive use of lexical compounding and because it includes many homophones, we have pursued the importance of morphological awareness as a unique correlate to word reading in several studies, as a parallel to phonological awareness in alphabetic languages (e.g., Bus & van IJzendoorn, 1999). In three studies (McBride-Chang et al., 2003; McBride-Chang et al., 2005; Shu, McBride-Chang, Wu, & Liu, 2006), measures of lexical compounding, homophone sensitivity, or both were demonstrated to have unique associations to word reading in Chinese. All of these morphological awareness measures were administered orally, similar to the methods used to examine phonological awareness. For example, we would ask children to form a new compound word (e.g., “What would we call a game in which a ball is thrown into a bucket?” answer: bucket ball). Another task involved identifying the correct homophone, basically by asking children to point to or generate their own homophones. For example, given four pictures, for example, an envelope (信封 /seon3 fung1/), a petrol station (油站 /jau4 zaam6/), a post box (郵筒 /jau4 tung2/), and swimming (游水 /jau4 seoi2/), children were asked to select the one
106
C. McBride-Chang et al.
sharing the same target morpheme 郵 (/jau4/) of a stamp (郵票 /jau4 piu3/). At another level, we might ask children to generate a morpheme that had the same and different meaning as a given one (e.g., given “sun” in “Sunday,” a same morpheme word would be “suntan,” and a different morpheme word would be “grandson”) (for details, see McBride-Chang et al., 2003 or Shu et al., 2006). We reasoned that for children who are most clear about how morphemes in their native language fit together (e.g., in English, one could use the term “bucketball” to describe a game involving a ball, but not “ballbucket”), reading could be easier. This is because those children would have the linguistic knowledge to facilitate their own analytic learning of new words. For example, if they can learn a morpheme such as “ball” and how it is used in language, they can apply it to new contexts and to learning of new words, including eventually learning to recognize that morpheme in print. In addition, children who can distinguish homophones in language and in print will more successfully and quickly understand what they read (e.g., the meaning of “meatball” may be clearer if you understand the difference in meanings between “meat” and “meet”). These positive results in demonstrating lexical compounding and homophone knowledge as unique cognitive correlates of reading success in Chinese lead us to wonder whether training in morphological awareness could promote reading in Chinese, similar to the way in which morphological training in Norwegian was helpful in promoting reading development in Norway (Lyster, 2002). In a study using four separate groups of parents and children (Chow et al., 2008), we found that Hong Kong Chinese parents’ own training of their children in morphological awareness facilitated word reading over a 12-week training period. Our groups consisted of one in which parents were taught dialogical reading only (DR), a second in which parents were asked to read as they normally would with their children, that is, typical reading (TR), a third in which parents were trained both to read with their children using the dialogical reading technique and were also given training in how to teach their children about morphology (DR + MT), and a fourth control group. For all except the control group (who were only given training and books after the experiment was over), 12 books consisting of the hints for prompt questions were given to parents and children, and the parents were required to read each book twice per week for 20 min, either as they normally would (TR) or using the dialogical reading technique (DR). The DR + MT group was also taught how to sensitize children to homophones and lexical compounding through play. This study was unique in that all training was done at home by the child’s parent. Across the four groups, mean initial words read on a word-reading test ranged from 50 to 52. Twelve weeks later, the three groups that did not receive morphological training all achieved mean averages of approximately 60–61. However, the group with morphological training scored a mean of 67 words, a significant improvement over the other three groups. These were preliminary results, but we find them to be promising. Again, they suggest that early literacy activities on the part of parents can make a potentially large difference in children’s literacy performance over time. It is possible that even greater gains in word reading can be made when training in morphological awareness
7 Early Literacy at Home: General Environmental Factors and Specific Parent Input
107
is linked in some way to print knowledge, as has been demonstrated for phonological awareness training in alphabetic orthographies (Bus & van IJzendoorn, 1999). This is one important direction for future research in our own work.
Summary In this chapter, we have tried to demonstrate the importance of the home literacy environment for children’s early literacy development. The home literacy environment is multi-faceted and potentially multiplicative in its contributions to early reading and writing development. Some of these contributions are implicit. For example, parents’ educational levels, valuing of literacy, or attitudes about writing will likely be communicated in their styles of interacting with their children or the daily activities they model (e.g., McBride-Chang, 2004). Other contributions are clearly more explicit. For example, parents’ scaffolding of the writing process (e.g., Aram & Levin, 2001, 2004), style of shared book-reading with their children (e.g., Whitehurst et al., 1988; Chow & McBride-Chang, 2003), or training of metalinguistic skills (e.g., Chow et al., 2008) are all deliberate and effective attempts to communicate values in literacy to children to learn. Iris Levin has expanded our knowledge of literacy development, particularly in young children, greatly in the past 15 years or so. Her creative approaches to measuring and observing cognitive development have had important theoretical implications for the field of psychology and practical implications for educational policy in Israel and beyond. We particularly appreciate her broad cross-cultural perspective, which has allowed us and others (e.g., Levin & Bus, 2003) to explore universals and specifics of literacy development in very different cultural contexts. We look forward to even more collaborations and contacts with her and her colleagues in the field of literacy research as they continue to pioneer in the fascinating field of early literacy development.
References Aram, D. (2007). Sensitivity and consistency of maternal writing mediation to twin kindergartners. Early Education and Development, 18, 71–92. Aram, D., & Levin, I. (2001). Mother-child joint writing in low SES: sociocultural factors, maternal mediation, and emergent literacy. Cognitive Development, 16, 831–852. Aram, D., & Levin, I. (2004). The role of maternal mediation of writing to kindergartners in promoting literacy in school: a longitudinal perspective. Reading and Writing, 17, 387–409. Balmuth, M. (1992). The roots of phonics: a historical introduction. Baltimore, MD: York Press. Baumann, J. F., Edwards, E. C., Font, G., Tereshinski, C. A., Kame’enui, E. J., & Olejnik, S. (2002). Teaching morphemic and contextual analysis to fifth-grade students. Reading Research Quarterly, 37, 150–176. Bus, A. G., & van IJzendoorn, M. H. (1999). Phonological awareness and early reading: a metaanalysis of experimental training studies. Journal of Educational Psychology, 91, 403–414.
108
C. McBride-Chang et al.
Chan, T.Y., & McBride-Chang, C. (2005). Environment and bilingualism in Hong Kong kindergartners: The impact of foreign domestic helpers on early language-learning. Journal of Psychology in Chinese Societies 6, 179–193. Cheung, H., & Ng, L. (2003). Chinese reading development in some major Chinese societies: an introduction. In C. McBride-Chang & H.-C. Chen (Eds.), Reading development in Chinese children (pp. 3–17). Westport, CT: Praeger Press. Chiu, M. M., & McBride-Chang, C. (2006). Gender, Context, and Reading: a comparison of students in 41 countries. Scientific Studies of Reading, 10, 331–362. Cho, J.-R., McBride-Chang, C., & Park, S.-G. (2008). Phonological awareness and morphological awareness: Differential associations to regular and irregular word recognition in early Korean Hangul readers. Reading and Writing, 21, 255–274. Chow, B. W.-Y., & McBride-Chang, C. (2003). Promoting language and literacy development through parent-child reading in Hong Kong preschoolers. Early Education and Development, 14, 233–248. Chow, B. W.-Y., McBride-Chang, C., & Burgess, S. (2005). Phonological processing skills and early reading abilities in Hong Kong Chinese kindergarteners learning to read English as an L2. Journal of Educational Psychology, 97, 81–87. Chow, B. W.-Y., McBride-Chang, C., Cheung, H., & Choi, L. (2008). Dialogic reading and morphology training in Chinese children: effects on language and literacy. Developmental Psychology, 44, 233–244. Elbro, C., & Arnbak, E. (1996). The role of morpheme recognition and morphological awareness in dyslexia. Annals of Dyslexia, 46, 209–240. Fung, P.-C., Chow, B. W.-Y., & McBride-Chang, C. (2005). The impact of a dialogic reading program on deaf and hard-of-hearing kindergarten and early primary school-aged students in Hong Kong. Journal of Deaf Studies and Deaf Education, 10, 82–95. Ho, C. S.-H., & Braynt, P. (1997). Phonological skills are important in learning to read Chinese. Developmental Psychology, 33, 946–951. Korat, O., & Levin, I. (2001). Maternal beliefs and child development: comparison of text writing between two social groups. Journal of Applied Developmental Psychology, 22, 397–420. Lau, J. Y.-H., & McBride-Chang, C. (2005). Home literacy and Chinese reading in Hong Kong children. Early Education and Development, 16, 5–21. Leseman, P. P. M., & de Jong, P. F. (1998). Home literacy: opportunity, instruction, cooperation and social-emotional quality predicting early reading achievement. Reading Research Quarterly, 33, 294–318. Levin, I., & Bus, A. (2003). How is emergent writing based on drawing? Analyses of children’s products and their sorting by children and mothers. Developmental Psychology, 29, 891–905. Levin, I., Ravid, D., & Rapaport, S. (2001). Morphological and spelling among Hebrew-speaking children: from kindergarten to first grade. Journal of Child Language, 28, 741–772. Levin, I., Shatil-Carmon, S., & Asif-Rave, O. (2005). Learning of letter names and sounds and their contribution to word recognition. Journal of Experimental Child Psychology, 93, 139–165. Lin, D., McBride-Chang, C., Aram, D., Levin, I., Cheung, R.Y.-M., Chow, Y.Y.-Y., & Tolchinsky, L. (in press). Maternal mediation of writing in Chinese children. Language and Cognitive Processes, 24, 1286–1311. Lundberg, I., Frost, J., & Petersen, O. P. (1988). Effects of an extensive program for stimulating phonological awareness in preschool children. Reading Research Quarterly, 23, 263–284. Lyster, S. A. H. (2002). The effects of morphological versus phonological awareness training in kindergarten on reading development. Reading and Writing, 15, 261–294. McBride-Chang, C. (2004). Children’s literacy development (texts in developmental psychology series). London: Edward Arnold/Oxford Press. McBride-Chang, C., Cho, J.-R., Liu, H., Wagner, R. K., Shu, H., Zhou, A., et al. (2005). Changing models across cultures: Associations of phonological and morphological awareness to reading in Beijing, Hong Kong, Korea, and America. Journal of Experimental Child Psychology, 92(2), 140–160.
7 Early Literacy at Home: General Environmental Factors and Specific Parent Input
109
McBride-Chang, C., Shu, H., Zhou, A., Wat, C. P., & Wagner, R. K. (2003). Morphological awareness uniquely predicts young children’s Chinese Character Recognition. Journal of Educational Psychology, 95, 743–751. Packard, J. L., Chen, X., Li, W., Wu, X., Gaffney, J. S., Li, H., et al. (2006). Explicit instruction in orthographic structure and word morphology helps Chinese children learn to write characters. Reading and Writing, 19, 457–487. Rauh, V. A., Lamb-Parker, F., Garfinkel, R. S., Perry, J., & Andrews, H. F. (2003). Biological, social, and community influences on the third-grade reading levels of minority head start children: a multi-level approach. Journal of Community Psychology, 31(3), 255–278. Rispens, J., McBride-Chang, C., & Reitsma, P. (2008). Morphological awareness and early and advanced reading and spelling in Dutch: a cross-sectional study. Reading and Writing, 21, 587–607. Sénéchal, M., & LeFevre, J. (2002). Parental involvement in the developmental of children’s reading skill: a five-year longitudinal study. Child Development, 73(2), 445–460. Shu, H., McBride-Chang, C., Wu, S., & Liu, H. (2006). Understanding Chinese developmental dyslexia: morphological awareness as a core cognitive construct. Journal of Educational Psychology, 98, 122–133. Siok, W. T., & Fletcher, P. (2001). The role of phonological awareness and visual-orthographic skills in Chinese reading acquisition. Developmental Psychology, 37, 886–899. Sonnenschein, S., Baker, L., Serpell, R., Scher, D., Truitt, V. G., & Munsterman, K. (1997). Parental beliefs about ways to help children learn to read: the impact of an entertainment or a skills perspective. Early Child Development and Care, 127–128, 111–118. Tracey, D. H., & Young, J. W. (2002). Mothers’ helping behaviors during children’s at-home oral-reading practice: effects of children’s reading ability, children’s gender, and mothers’ educational level. Journal of Educational Psychology, 94, 729–737. UNICEF. (2001). The state of the world’s children 2001. Retrieved March 11, 2006, from http:// www.unicef.org/sowc01 Valenzuela, M. (1997). Maternal sensitivity in a developing society: the context of urban poverty and infant chronic undernutrition. Developmental Psychology, 33, 845–855. Vernon-Feagans, L., Hammer, C. S., Miccio, A., & Manolove, E. (2001). Early language and literacy skills in low-income African American and Hispanic children. In S. B. Neuman & D. K. Dickinson (Eds.), Handbook of early literacy research (pp. 192–210). New York: Guilford Press. Whitehurst, G. J., Epstein, J. N., Angell, A. L., Payne, A. C., Crone, D. A., & Fischel, J. E. (1994). Outcomes of an emergent literacy intervention in Head Start. Journal of Educational Psychology, 86, 542–555. Whitehurst, G. J., Falco, F. L., Lonigan, C., Fischel, J. E., DeBaryshe, B. D., Valdez-Menchaca, M. C., et al. (1988). Accelerating language development through picture-book reading. Developmental Psychology, 24, 552–558. Whitehurst, G. J., & Lonigan, C. J. (1998). The development and emergent literacy. Child Development, 69, 848–872. Whitehurst, G. J., Zevenbergen, A. A., Crone, D. A., Schultz, M. D., Velting, O. N., & Fischel, J. E. (1999). Outcomes of an emergent literacy intervention from Head Start through second grade. Journal of Educational Psychology, 91, 261–272. Wagner, R. K., Muse, A. E., & Tannenbaum, K. R. (2007). Vocabulary acquisition: implications for reading comprehension. New York, NY: The Guilford Press. Weigel, D. J., Martin, S. S., & Bennett, K. K. (2006). Contributions of the home literacy environment to preschool-aged children’s emerging literacy and language skills. Early Child Development & Care, 176(3/4), 357–378. Wu, X., Li, W., & Anderson, R. C. (1999). Reading instruction in China. Journal of Curriculum Studies, 31, 571–586.
Chapter 8
Reading Books to Young Children: What It Does and Does Not Do Monique Sénéchal
Abstract The present chapter is an overview of six studies that share a common theme: the contribution of shared reading to child outcomes. The first three studies are experimental in nature and show that the number of times as well as the manner in which adults read to children will affect children’s acquisition of comprehension and spoken vocabulary. The fourth study is an intervention with children who have poor vocabulary skills. The findings revealed that care givers can enhance children’s spoken vocabulary by reading books to them in an interactive manner, and that simply reading in their customary fashion may not promote vocabulary acquisition. The last two studies are correlated. They provide converging evidence that shared reading predicts children’s vocabulary, and that, children’s vocabulary is a robust predictor of reading comprehension. These studies also show the limits of shared reading because parent reports of shared reading did not predict children’s early literacy skills or word reading at the end of grade 1.
Shared Reading: What It Does Do Shared book reading is, first and foremost, a source of entertainment. At the same time, books can be a rich source of vocabulary learning. Novel words are often introduced in the text providing syntactic and semantic cues to their meaning. In addition, novel words are often illustrated, providing a picture of the word meaning. Moreover, the adult reader can facilitate word learning by demonstrating certain verbal behaviors. In fact, it is often recommended that adult readers actively involve young children during book reading in order to enhance the children’s learning. In this section, I describe research findings from my laboratory that show how young children can learn between 1 and 3 new words from shared book reading.
M. Sénéchal (*) Carleton University, Ottawa, ON, Canada e-mail:
[email protected] D. Aram and O. Korat (eds.), Literacy Development and Enhancement Across Orthographies and Cultures, Literacy Studies 101, DOI 10.1007/978-1-4419-0834-6_8, © Springer Science+Business Media, LLC 2010
111
112
M. Sénéchal
This description, taken from Sénéchal (2003) with permission from the publisher, includes three experimental studies with children between the ages of 3 and 5, as well as one intervention study with children with language delays. A short integrative summary of the findings is presented at the end of this chapter. The importance of early vocabulary development can be appreciated when we think that vocabulary measured early is one of the best predictors of eventual success in reading. Consider the following finding from a longitudinal study conducted in my laboratory (Sénéchal & LeFevre, 2002). In this study, we found that vocabulary measured at the beginning of kindergarten predicted reading comprehension at the end of grade 3 after controlling for parent education, non-verbal intelligence, phoneme awareness, and grade 1 reading, Given the predictive relation between kindergarten vocabulary and grade 3 reading, it becomes interesting to learn how we can promote early vocabulary development. Two types of vocabulary are considered in the present section: spoken and comprehension vocabulary. Spoken vocabulary refers to the words children can actually produce and is often labeled expressive vocabulary. In contrast, comprehension vocabulary refers to the words that children can understand, but not necessarily produce, and is often labeled receptive vocabulary. This distinction is useful because different factors might facilitate the acquisition of the two types of vocabulary. Research on spoken vocabulary suggests that young children can learn spoken vocabulary through imitation because imitation provides children with opportunities to practice retrieving the labels for newly acquired words. In contrast, repeated exposure to new words might be sufficient for the acquisition of comprehension vocabulary. In the research presented below, we considered whether imitation and repeated exposure influence spoken and comprehension vocabulary differently. In order to study the benefits of reading books, we developed a procedure that was used in our experimental studies. First, we selected picture books that would be attractive to young children. Second, we modified the text in the books to introduce novel words. To do so, we replaced words typically known to preschool children (e.g., baby) with rarer synonyms (e.g., infant). We modified the text to introduce 10–13 novel words. Third, we developed tests to measure comprehension and spoken vocabulary. In the comprehension vocabulary test, children were asked to choose the picture of a named item from a selection of four pictures. The illustrations for the novel words were different than those in the book. For example, the word angling was represented in the book by a man fishing, but in the test, by a girl doing the activity.. Therefore, the test measured children’s ability to learn new words and extend them to new examples of their meaning. Comprehension vocabulary was measured three times in most studies: once before the book reading, once immediately after, and once a week after the reading. Spoken vocabulary was measured by asking thechildren to label the words using the illustrations in the books because a pilot study revealed that children did not use the new words when labeling unfamiliar illustrations. Moreover, spoken vocabulary was not tested before the book reading because the pilot study also revealed that children did not use the novel words. After developing this procedure, we were equipped to test how children
8 Reading Books to Young Children: What It Does and Does Not Do
113
would learn from book reading events. For each study presented in this chapter, only the findings that were statistically reliable are described.
Study 1: Learning After Reading a Book Once In this study, we were interested in measuring whether children would learn new vocabulary after a single exposure to a book (Sénéchal & Cornell, 1993). We were also interested in assessing whether active involvement in the form of answering questions would enhance vocabulary more than simply listening to the story. One hundred and sixty children, who were 4- and 5-year olds, participated in the study. Each child was seen individually. We found that, after a single reading of the book, active involvement did not promote vocabulary learning more than simply listening to the book reading. Nonetheless, children did learn. Children, on average, could comprehend 1.6 new words as reflected by the difference between the number of words correctly identified before and after the book reading. Most interestingly, children could remember the new words learned 1 week after the book reading. We also found that 5-year olds learned more comprehension vocabulary than did the 4-year-old children. It is important to note, however, that children could not speak any new words after a single book reading.
Study 2: Learning After Reading Book Twice In this study, we were interested in testing whether the benefits of active involvement would appear after two readings of the books (Sénéchal, Thomas, & Monker, 1995). We read the books in one of three ways: (1) the reader read the book, repeating the novel words once and pointing to them as they were introduced in the text; (2) the reader asked the children to point to the novel words; or (3) the reader asked the children to label the novel words, providing the answer when children failed to do so. Hence, in each book reading situation, children were exposed to each novel word twice for a total of four times across the two book readings. Forty-eight 4-year-old children participated in the study and they were read to individually. The number of words learned for each book-reading situation is presented in Table 8.1. We found that children learned to comprehend new words in each situation, Table 8.1 Vocabulary acquisition after two readings of a book Vocabulary Reading condition Comprehension Spoken Listening to the story 1 0 Pointing to new words 2 1 Labeling the new words 2 2
114
M. Sénéchal
but that they learned more when they were actively involved in the book reading. Answering labeling or pointing questions was equally effective for the acquisition of comprehension vocabulary. The pattern was different for spoken vocabulary. Children did not learn to say the novel words after listening passively to two renditions of the book, but they did learn when they were actively involved. Moreover, children learned more when they had a chance to imitate the novel words by answering labeling questions as opposed to pointing to them. In this study as in the previous, we measured whether children would remember the words learned after 1 week. Again, we found no evidence of forgetting because children could comprehend and speak, on average, the same number of words as they did immediately after the second book reading. In addition to measuring book vocabulary, we measured children’s comprehension vocabulary with a standardized test, the Peabody Picture Vocabulary Test – Revised, to assess whether children who differ in vocabulary knowledge would also differ in how they learn from book reading events. We found that the manner of reading affected children with a greater or smaller vocabulary similarly, but that children with a smaller vocabulary learned less than did children with a greater vocabulary. These findings suggest that children with a poorer vocabulary benefit from book reading events, but that the amount they learn is attenuated.
Study 3: Learning After Reading the Book Three Times In this study, we assessed the amount of vocabulary learning after repeating the book reading three times (Sénéchal, 1997). Sixty 3- and 4-year-old children participated in one of three book reading events: (1) they listened to the book read once, (2) they listened to the book read three times, or (3) they answered labeling questions during three book reading events. As presented in Table 8.2, we found that children, on average, could not comprehend or speak any new words after a single reading of the book. However, children could comprehend and speak one new word after listening to the book three times, but most impressively, they could comprehend and speak three new words when they answered labeling questions during the three book reading events. In this study as in Study 1, we found that the younger children learned less, overall, than did the older children. It is possible that the younger children need a greater number of repetitions to learn at a level similar to that of the older children. Table 8.2 Vocabulary acquisition after three readings of a book Vocabulary Reading condition Comprehension Spoken Listening: 1 reading 0.4 0 Listening: 3 readings 1 1 Labeling: 3 readings 3 3
8 Reading Books to Young Children: What It Does and Does Not Do
115
Table 8.3 Percentage of times that children learned a new word given that they said it three times, twice, once or never during the book reading Vocabulary Child speaks the word Comprehension (%) Spoken (%) Three times 42 43 Twice 33 39 Once 44 26 Never 36 18
In this study, as in the others, we assumed that answering labeling questions would promote vocabulary acquisition because it provided occasions to imitate the new words. If this is true, then children, who answered the labeling questions more often, should learn more than children who were unable to answer the questions. Recall that children were read the book three times, and therefore, had three chances of answering the labeling questions, and that when they did not succeed the reader provided the new word. It is of interest to examine whether children learned more when they, and not the reader, provided the new words. I present, in Table 8.3, the percentage of times that children learned new words given the number of times they said the words during reading. Studying the results of Table 8.3 is instructive because the findings support the notion that comprehension and spoken vocabulary may not be sensitive to the same types of interactions. We see that children were just as likely to comprehend a new word whether they said the word twice during the book reading (33%) or whether the reader always provided the new word (36%). Also, children were just as likely to comprehend a new word whether they said it three times or once during the reading. In contrast, the results for the spoken vocabulary show a different pattern: There was a gradual decrease in the percentage of words learned as a function of the number of times the child, not the reader, spoke the words. Children learned 43% of the words they themselves spoke three times, but only learned 18% of the words they heard the reader speak during the reading. The findings in Table 8.3 support the idea that imitation enhances spoken vocabulary more than comprehension vocabulary, and this demonstration advances our understanding of vocabulary acquisition. Nonetheless, early childhood educators who want to promote vocabulary acquisition may want to use labeling questions because they do not hinder the acquisition of comprehension vocabulary, and, certainly, foster the acquisition of spoken vocabulary.
Study 4: Book Reading Intervention for Children with Language Delays In this study, we tested whether book reading would enhance the vocabulary of children with language delays (Hargrave & Sénéchal, 2000). The 36 children who participated were on average 4 years of age, but their spoken vocabulary was 13
116
M. Sénéchal
months behind their chronological age. The children attended 1 of 2 day-care centers in which the early childhood educators occasionally read books during circle time (i.e., with a group of eight children). In 1 day-care we asked teachers to read in the customary fashion, while in the other, we trained teachers to use dialogic reading. Dialogic reading (Whitehurst et al., 1994) is a technique to promote language acquisition during book reading by the use of questions, recasts of children’s verbalizations, praise, etc. Teachers were asked to read books daily during circle time for 4 weeks and to read each of the ten books we provided at least twice during this period. The ten books provided were selected because they (1) included illustrations of potential new words, (2) included short texts which ensured that teachers would have time to interact with the children, and (3) were borrowed from the neighborhood library to encourage teachers to make use of the library given that they had very limited budgets to buy books. We observed teachers book reading behaviors before the intervention and again during the intervention. We found that the teachers in the 2 day-cares did not differ in the manner in which they read to the children before the intervention, and that the regular-reading teachers did not change their reading behaviors during the intervention. As indicated in Table 8.4, teachers in the regular reading did not interact much with the children during the book reading. In contrast, teachers trained in dialogic reading changed their interactions dramatically. The behavior that increased the most sharply was the frequency with which they used WH-questions (what, who, why) during book reading. This first set of findings confirms that we were able to promote changes in the way that teachers read to the children. We found that children exposed to dialogic reading learned more spoken vocabulary than children exposed to regular reading. The spoken vocabulary gains for the dialogic-reading children corresponded to that expected in 4 months – that is, they gained in a single month, the amount of spoken vocabulary expected in 4 months as measured by a standardized test of spoken vocabulary (the Expressive One Word Picture Vocabulary Test). The children exposed to dialogic or regular reading, however, did not show any gains on a standardized measure of comprehension vocabulary (the Peabody Picture Vocabulary Test – Revised). Because standardized tests are general and do not test the specific vocabulary in the books, we also measured whether children were learning specific words introduced in the books by assessing their comprehension vocabulary before and after the intervention. We found that
Table 8.4 The average number of times that early childhood educators made various types of verbal interactions during book reading during the intervention Day-care Teacher behaviors Regular reading (%) Dialogic reading (%) WH-questions 2 37 Praise 0 7 Model 0 2 Repeat child utterances 0 7 Recast child utterances 2 14
8 Reading Books to Young Children: What It Does and Does Not Do
117
children in the two centers learned new book vocabulary: the dialogic-reading children comprehended two new words and the regular-reading children comprehended one new word. Taken together, the findings show that early childhood educators can promote vocabulary acquisition during circle time with up to eight children.
Shared Reading: What It Does Not Do In this section, I examine the limits of shared reading. To do so, I summarize two longitudinal studies from my laboratory in which we tested whether home literacy experiences such as shared reading and parent teaching would predict children’s vocabulary, phoneme awareness, and early literacy in kindergarten, as well as reading in grades 1, 3, and 4. Parents can make a significant contribution to their child’s reading achievement through informal and formal literacy experiences (Sénéchal, LeFevre, Thomas, & Daley, 1998). Informal literacy experiences refer to activities that involve print, but in which print per se is not the focus of the activities. For instance, parents do read the printed text during shared reading, but the focus of the activity is the story, not the print. In contrast to informal activities, formal literacy experiences are those in which print is the focus of the activity. For example, parents can tutor their child to learn the alphabet, print their names, or read and print other words. In the research presented in this section, informal and formal literacy experiences were assessed with measures of storybook exposure and parent teaching about literacy, respectively. In the previous section, we saw that shared reading can enhance young children’s vocabulary, and that early differences in vocabulary knowledge were a robust predictor of reading comprehension. Will it be the case that shared reading also predicts children’s early literacy skills? Past research has shown convincingly that early literacy skills provide the building blocks for the successful acquisition of literacy skills (see the reviews by Scarborough, 2001; Whitehurst & Lonigan, 2001).
Study 5: Home Literacy and Child Outcomes in English-Speaking Families We conducted a longitudinal study to test the long-term association among home literacy experiences in kindergarten and children’s early literacy skills as well as their eventual success in reading in grades 1 and 3 (Sénéchal & LeFevre, 2002). Figure 8.1 captures the design of the study and the main findings. In this study, parents completed questionnaires about home literacy experiences at the beginning of kindergarten, and children completed language and literacy measures in kindergarten, at the beginning of grade 1, and at the end of grades 1 and 3. The longitudinal sample included 110 families in kindergarten, 93 until the end of grade 1, and 66 until the end of grade 3. The results depicted in Fig. 8.1 are those for which
118
M. Sénéchal
stringent regression analyses revealed that the association between two variables was still present (and statistically significant) once we had controlled for the other variables. Each key finding is described in turn. The first finding of interest is that informal (shared reading) and formal (parent teaching) literacy experiences were unrelated to each other. The lack of relation suggests that some parents who read also tutor their child to learn early literacy skills, but some parents do not. This finding suggests that parents differ in the types of literacy activities that they include at home (Anderson, 1995; Teale, 1986). The remaining findings in Fig. 8.1 concern the association among home experiences and child outcomes. Interestingly, the two types of literacy experiences were differentially related to language, early literacy, and phoneme awareness in kindergarten. As would be expected from past research, shared reading predicted children’s oral language skills, whereas the frequency of parent teaching literacy was associated with the acquisition of early literacy skills. Shared reading and parent teaching, however, were not directly related to phoneme awareness. Specifically, the association between home literacy activities and phoneme awareness is mediated by children’s language and early literacy skills. The longitudinal relations between the home literacy activities measured prior to grade 1 and eventual reading outcomes are examined next. Figure 8.1 shows no direct or indirect link between informal literacy and reading in grade 1 (cf. de Jong & Leseman, 2001). It is not until more advanced reading skills were achieved that informal literacy experiences became indirectly associated to reading through their relation to early vocabulary skills. The pattern of findings was different for formal literacy activities. Figure 8.1 shows that parent teaching was directly related to early literacy, which, in turn, was associated to grade 1 reading, and grade 1 reading predicted more advanced reading skills. Taken together, these findings support the notion that informal (shared reading) and formal (parent teaching) home literacy experiences are differentially related to oral and written language.
Study 6: Home Literacy and Child Outcomes in French-Speaking Families A second longitudinal study was conducted in my laboratory to assess whether the previous findings would extend to French-speaking families (Sénéchal, 2006). In this correlational study, parents reported on home literacy experiences at the end of kindergarten, and children were assessed on measures of oral language, phoneme awareness, and early literacy in kindergarten, as well as reading in grades 1 and 4. The sample included 90 families until the end of grade 1, and we were able to follow 65 of these until the end of grade 4. As presented in Fig. 8.2, parent reports of shared reading predicted directly kindergarten vocabulary as well as the frequency with which children reported reading for pleasure in grade 4. In contrast, parent teaching literacy in kindergarten
8 Reading Books to Young Children: What It Does and Does Not Do
Shared Book Reading at Home
119
Vocabulary
Phonological Awareness
Parent Teaching about Reading and Writing
Home Experiences
Early Literacy
Beginning of Grade 1
Reading End of Grade 1
End of Grade 1
Reading End of Grade 3
End of Grade 3
Fig. 8.1 The relations among different home literacy experiences and child outcomes in an English sample (Sénéchal and LeFevre, 2002)
directly predicted kindergarten alphabet knowledge and grade 4 reading fluency. Moreover, parent teaching predicted indirectly reading in grades 1 and 4, whereas shared reading predicted grade 4 reading comprehension indirectly. These findings replicate and extend those of Sénéchal and LeFevre (2002). In sum, the two studies presented in this section capture a real distinction in the early literacy experiences of young children. The findings are consistent with the notion that informal literacy experiences (shared reading) enhance young children’s oral language skills whereas more formal literacy experiences (parent teaching) promote the acquisition of early literacy skills. The findings also show that early differences in vocabulary and early literacy are solid predictors of children’s eventual success in literacy.
Summary In this chapter, I provided evidence on the contribution and limits of shared reading. The findings can be summarized succinctly under six headings: 1. The benefits of reading the same book repeatedly. Young children comprehended and spoke more new words after listening to three repeated readings of picture books than after a single or two exposures to the books. In fact, children did not learn to say new words after listening to a single or two renditions of the books.
120
M. Sénéchal
Shared Book Reading at Home
Parent Teaching about Reading and Writing
Home Experiences
Vocabulary
Child Book Exposure
Phonological Awareness
Reading Comprehension
Early Literacy
Beginning of Grade 1
Word Reading
Reading Fluency
End of Grade 1
End of Grade 4
Fig. 8.2 The relations among different home literacy experiences and child outcomes in a French sample (Sénéchal, 2006)
2. The benefits of active involvement during repeated readings. Young children comprehended and spoke more new words when they were actively involved during the repeated book readings. Answering requests to label new words was particularly helpful. 3. Developmental and individual differences in vocabulary acquisition during shared reading. Older children learned more from book reading events than younger children. Five-year-old children learned more from book reading events than 4-year-old children, and 4-year olds learned more than 3-year-old children. In addition, children with a larger vocabulary learned more during book reading than children with a smaller vocabulary. Younger children and children with a smaller vocabulary may need more exposures to the same books to learn at the same level as other children. 4. The role of early childhood educators. Early childhood educators could implement repeated book readings during circle time with eight children or less. Most important, children with vocabulary delays, whose early childhood educators actively involved them during repeated readings, improved their spoken vocabulary more than children whose teachers involved them less during the book readings. 5. The limits of shared reading. In two large-scale longitudinal studies, shared reading at home, as reported by parents, was not a robust direct predictor of children early literacy skills, their phoneme awareness, or their eventual success in reading. These studies found that shared reading predicted children’s vocabulary, and that, in turn, children’s vocabulary predicted children’s phoneme awareness and reading comprehension. Most interestingly, parent reports of shared reading in kindergarten predicted children’s reports of reading for pleasure in grade 4.
8 Reading Books to Young Children: What It Does and Does Not Do
121
6. The contribution of parent teaching. In two large-scale longitudinal studies, parent reports of the frequency with which they taught their child to read and print words were robust predictors of early literacy skills. In turn, early literacy skills were robust predictors of grade 1 reading, and grade 1 reading predicted grade 4 reading. Most interestingly, parent teaching predicted directly children’s reading fluency in grade 4.
Conclusion Parents and early childhood educators are encouraged to use books as a source of fun, and learning for children. During shared reading, parents and early childhood educators can use simple techniques, such as asking labeling questions during repeated readings, to enhance the vocabulary of young children. Enhancing children’s vocabulary is important because of the important role vocabulary plays in reading comprehension. In the present chapter as in others (e.g., Jordan, Snow, & Porche, 2000), shared reading was not a robust predictor of early literacy or reading. Therefore, early childhood educators should be cautious about the claims they make on the benefits of shared reading.
References Anderson, J. (1995). How parents’ perceptions of literacy acquisition relate to their children’s emergent literacy knowledge. Reading Horizons, 35, 209–228. de Jong, P. F., & Leseman, P. P. M. (2001). Lasting effects of home literacy on reading achievement in school. Journal of School Psychology, 39(5), 389–414. Hargrave, A. C., & Sénéchal, M. (2000). Book reading interventions with language-delayed preschool children: the benefits of regular reading and dialogic reading. Early Childhood Research Quarterly, 15, 75–90. Jordan, G. E., Snow, C. E., & Porche, M. V. (2000). Project EASE: the effect of a family literacy project on kindergarten students’ early literacy skills. Reading Research Quarterly, 35, 524–546. Scarborough, H. S. (2001). Connecting early language and literacy to later reading (dis)abilities: evidence, theory, and practice. In S. B. Neuman & D. K. Dickinson (Eds.), Handbook of early literacy research (pp. 97–110). New York: Guilford Press. Sénéchal, M. (2006). Testing the home literacy model: parent involvement in kindergarten is differentially related to grade 4 reading comprehension, fluency, spelling, and reading for pleasure. Journal for the Scientific Study of Reading, 10(1), 59–87. Sénéchal, M. (2003). Read it again, Pam! On the importance of repeated reading for the development of language. In L. Girolametto & E. Weitzman (Eds.), Enhancing caregiver language facilitation in child care settings (pp. 6:1–6:9). Toronto: The Hanen Centre. Reproduced by the author with permission from The Hanen Centre. Sénéchal, M. (1997). The differential effect of storybook reading on preschooler’s expressive and receptive vocabulary acquisition. Journal of Child Language, 24, 123–138. Sénéchal, M., & Cornell, E. H. (1993). Vocabulary acquisition through shared reading experiences. Reading Research Quarterly, 28, 360–374. Sénéchal, M., & LeFevre, J. (2002). Parental involvement in the development of children’s reading skill: a 5-year longitudinal study. Child Development, 73, 445–460.
122
M. Sénéchal
Sénéchal, M., LeFevre, J., Thomas, E., & Daley, K. (1998). Differential effects of home literacy experiences on the development of oral and written language. Reading Research Quarterly, 32, 96–116. Sénéchal, M., Thomas, E., & Monker, J. (1995). Individual differences in 4-year-olds’ ability to learn new vocabulary. Journal of Educational Psychology, 87, 218–229. Teale, W. H. (1986). Home background and young children’s literacy development. In W. Teale & E. Sulzby (Eds.), Emergent literacy: writing and reading (pp. 173–206). Norwood, NJ: Ablex. Whitehurst, G. L., Arnold, D. S., Epstein, J. N., Angell, A. L., Smith, M., & Fischel, J. E. (1994). A picture book reading intervention in day-care and home for children from low-income families. Developmental Psychology, 30, 679–689. Whitehurst, G. J., & Lonigan, C. J. (2001). Emergent literacy: development from prereaders to readers. In S. B. Neuman & D. K. Dickinson (Eds.), Handbook of early literacy research (pp. 11–29). New York: Guilford Press.
Chapter 9
The Making of Literate Families: Considerations of Context and Misconceptions Linda M. Phillips
Abstract The case is made that the making of literate families is a daunting challenge. A challenge made even more so when pernicious claims are made, when uninformed views persist, and when low SES families are maligned in contexts beyond their control. First, the context of such claims is established. Next, I show that misconceptions have deleterious effects and persist despite substantial evidence to the contrary. Evidence presented includes the results of a longitudinal family literacy study of the Learning Together: Read and Write with Your Child program. Finally, I highlight that early literacy interventions are critically important in the making of literate families by pointing out that considerations of context and misconceptions of early language and literacy development require a common goal and mutual respect. The chapter concludes that there is much to gain from a respectful relationship between literacy in the home and literacy in the school. Such a relationship unlocks promising new courses of action in and expands the possibilities for the making of literate families. I met Iris Levin a long time before she met me. When I was a doctoral student in the early 1980s, I read her research with interest, admiration, and intrigue, and I continue to do so. We have met many times since my days as a student, and each occasion has been pleasant and intellectually rewarding. On this important occasion, the occasion of the retirement of Professor Iris Levin, I am privileged to contribute a chapter in her honor. Across the three decades of her stellar career, the unequivocal theme across all of her work has been the advancement of literacy and the sociocultural factors identified to shape and enhance progress. Two interconnected areas of study unite her research with that of mine in either direct or indirect ways: context and misconceptions. The making of literate families, it shall be shown, is a daunting challenge. It is especially daunting when pernicious claims are made, when uninformed views persist, and when low SES families are maligned in contexts beyond their control.
L.M. Phillips (*) University of Alberta, Edmonton, AB, Canada e-mail:
[email protected] D. Aram and O. Korat (eds.), Literacy Development and Enhancement Across Orthographies and Cultures, Literacy Studies 101, DOI 10.1007/978-1-4419-0834-6_9, © Springer Science+Business Media, LLC 2010
123
124
L.M. Phillips
Context Low-income families with a low-educational background have been the focus of my research for some two decades. Unfortunately, such families have been the target of generalizations associated with all that is bad about literacy development. All that is good about literacy development is portrayed eloquently by Mary Gordon (2003), “ … secure attachment, emotional and physical health, a long attention span, a love of storybooks, a large vocabulary, school readiness, academic and financial success, and even ‘happiness’” (p. 19).This portrayal is an idealization that rarely represents life in any family. Nonetheless, when children from low-income families with low-educational background present at school and do not know the basic print concepts, cannot write their names, and are deemed “at-risk” of school failure, the home is immediately perceived to be the problem. More precisely, the mothers are seen to be less than ideal (Hulbert, 2005), and factors such as poverty, lack of information, or second language considerations are not necessarily considered (Burns, Griffin, & Snow, 1999). However, if children from middle- and high-class backgrounds present with difficulties, then the home is not perceived to be the problem; rather, attention is turned to more etiological considerations. The assumption of the making of literate families has thus been one-sided for a long time, in favor of the middle and upper classes. Writing in the mid-1800s, Harriet Martineau noted that there was little expectation that literacy was a part of the lives of pauper families. Moreover, the absence of literacy was taken to be the legitimization of their poverty. More than a hundred years on, ethnographers, such as Shirley Brice-Heath (1983) and Denny Taylor (1983), were the first to document rich and varied forms of literacy in the homes of the poor. Notwithstanding the evidence, the perception among those providing mainstream advice on family literacy was that “literacy practices in low-income homes either did not exist or were counterproductive to the needs of schooling” (Smythe, 2006, p. 281). The precise nature of the differences in the school achievements of children from varying backgrounds warrants clarification. In 1997 Trevor Cairney noted that explanations for such differential school performance often fall into two extreme positions. He described “deficit driven explanations” which place the blame on families seen as incapable of creating an environment of support necessary for school success and “educational inadequacy explanations” which view the discrepancies as resulting from the failure of schools to develop student strengths and abilities (p. 62). Cairney, however, rejected these two explanations as inadequate because of their failure to take into account the fact that differences between people based on class, race, and culture are associated with different relationships with the curriculum in schools. It is his view, and I concur, that the different relationships the families have with the curriculum of the schools result in a gap between the homes and the school, which is, in part, responsible for differential school performance. Thus, it is critical to forge informed collaborations between homes and schools for the advancement of family literacy (Sample Gosse & Phillips, 2007). Such collaborations depend on an appeal to the research evidence to provide an accurate and objective portrayal of family literacy.
9 The Making of Literate Families: Considerations of Context and Misconceptions
125
In a recent longitudinal quasi-experimental, control group, mixed methods study of 5 years’ duration with 156 low-income and low-educational background parents and their children ages 3+ years (Phillips, Hayden, & Norris, 2006; Phillips & Sample, 2005; Sample Gosse & Phillips 2007), we identified ways to work collaboratively with families to address school literacy skills while respecting their backgrounds and home literacies. We interviewed the families individually and asked each what they wanted to learn. Their requests were a focal aspect of the Learning Together: Read and Write with Your Child intervention program. Families were unanimous in their request to learn how to improve their children’s reading and writing. Parents reported many literacy practices in the home such as searching for supermarket coupons, reading the Bible, and the television guide. However, these reports were at odds with their children’s teachers. Teachers experienced their children starting school without the print concepts considered essential for success with reading. To bridge the gap, we encouraged teachers to enquire about and to build on home literacy practices in their classrooms, and we encouraged parents to learn the expectations for their children when they arrive at school so that they can better prepare at home for successful school experiences. Consider the case of Lori. She was 29 with three children and grew up in several foster homes in a large prairie city in Western Canada. She left school after she became involved in drugs and was pregnant at an early age. Lori made it clear to us that she wanted to know how to teach her children their ABCs. She confided that she would say the alphabet and sing the song but her children “never could tell which one from which one.” The Learning Together staff respected and accommodated Lori’s interest and together they created a learning focus meaningful to everyone involved (Phillips & Sample, 2005, p. 91). Lori’s example raises an important point about negotiating the gap between home and school and exemplifies Cairney’s deficit-driven explanation. Parental views of literacy were frequently perceived by teachers to be narrow with a focus on learning the alphabet and identifying words. Teachers were encouraged to accept the parents’ view as a starting point while building a broader view of reading as the construction of meaning. Another strategy was to work with teachers to assist parents to develop a range of strategic behaviors for sharing text with their children. Teachers were encouraged to see how parents and children would respond to help and guidance rather than reminders of what they could not do. Researchers and educators must be open to acknowledging the universal parental dream that their children enjoy successful school experiences even when families do not present themselves in traditional, middle-class ways. When invited to elaborate and to provide examples, parents are often able to clearly articulate what they believe they need to learn in order to help their children and these beliefs are personally powerful because they are based on the parents’ own experiences (Phillips & Sample, 2005). Families and teachers each must accept responsibility for developing a positive and mutually respectful relationship that keeps the child at the center (Churchill, 2003). Treating parents as a collective group obscures differences that must be considered if productive collaborations are to be developed. We have learned that in addition to what is “typical” of families with certain cultural, linguistic, and social backgrounds, there is always variability within the groups in terms of beliefs about children’s learning.
126
L.M. Phillips
Maternal beliefs, mother–child interaction, and child’s literacy were studied across high and low SES groups by Korat and Levin (2001). Contrary to conventional beliefs at the time that low SES families read less to their children, initiate fewer questions, read passively, and do not engage children with the print (McCormick & Mason, 1986; Ninio, 1980), Korat and Levin found counterevidence for writing. In fact in their study of the collaborative writing of 40 mothers and their second-grade children, they found no differences in maternal pedagogical beliefs such as granting the child autonomy and adjudicating their child’s ability between the two social classes of families. They did find that low SES mothers were less able to translate their beliefs into practice even though they had similar belief systems to those of high SES mothers. The research literature typically compares low SES to middle or high SES groups. As expected, the low SES groups are not usually identified to be adequate but rather deficient. In a companion study, Aram and Levin (2001) examined mother–child joint writing in low SES families only. Their interest was in researching whether all low SES families are homogenous in order to learn whether differences were relevant to literacy development. They found differences in all of the sociocultural background factors including maternal literacy, children’s literacy tools, activities at home, and maternal mediation in joint writing with 5–6-year olds. Aram and Levin identified a vicious circle (p. 20). By this they meant that because some mothers did not understand their child’s reading, writing, and cognitive abilities then they expected less of their children, and the mothers who recognized their children’s letter knowledge and other skills capitalized upon those skills and mediated the children’s writing to a higher level. In other words, those who knew more learned more. Teaching parents when and how to mediate their children’s literacy experiences will enhance the chances of low SES mothers to mediate their children’s literacy. Parental mediation of children’s learning is advocated as an effective route to reading and writing (Rogoff, 1990). Forty-one low SES mothers and their kindergarten children participated in a study of the relation of mother–child joint writing and storybook reading to emergent literacy (Aram & Levin, 2002). They compared mother– child joint writing to storybook reading as predictors of emergent literacy. The effects of parental mediation for joint writing (ranged from 20% to 36%) was significantly higher than for storybook reading (8%). These researchers further found differences in the quality of parental mediation among the highly similar low SES families, thus challenging the persistent view of homogeneity associated with low SES families. Guiding parents to engage in and to mediate joint writing with children is a printbased and active experience for the children whereas storybook reading is an aural and passive experience for the children with the adults doing the reading. The latter is recommended in the popular media to families as the answer to reading development although the former is more effective (Phillips, Norris & Anderson, 2008). It is imperative that we teach parents how best to develop their children’s literacy. The work of Iris Levin and her former students, now her colleagues, goes a long way in answering and explaining the differential performance, if any, in children’s literacy, challenges persistent misconceptions, and makes repeat calls for teaching parents how best to help their children regardless of SES.
9 The Making of Literate Families: Considerations of Context and Misconceptions
127
Misconceptions Misconceptions may have deleterious effects and persist despite substantial evidence to the contrary. We have seen in the previous section that there is an abiding misconception that differences exist only between social classes in how families interact with their children rather than within and across social class. Indeed, there are a number of mistaken assumptions by many educators, social workers, and politicians that family literacy programs will solve educational and social ills despite issues of linguistic variations, race, poverty, and gender. Some mistaken assumptions include pervasive beliefs that there is an absence of literacy practices, a lack of interest in their children, and bad parenting in the homes of the poor and uneducated. Programs built on assumptions that imply families are responsible for their educational and economic circumstances would undoubtedly lead to questionable results. Those who work with families as they struggle for a life of dignity for their children and then for themselves know the fallacy of these assumptions especially at a time when the definition of family is changing. The definition of family and the context in which families live is not as narrow as in the past. Presently, we must recognize that as Gadsen (2000) says, “… there is no appropriate, prototypical model of family structure” and that social, cultural, and linguistic variability predominates (Sample Gosse & Phillips, 2006). The number of one-parent families has been increasing. Many of these families are under significant financial, emotional stress (Brodkin, 2002; Kaye, 2003) and face second language challenges. Consequently, they have less time to spend with and fewer resources available for their children, and are at the same time expected to participate in mainstream home–school and extracurricular activities. These families are frequently shunned when they do not participate and their non-participation interpreted as disinterested, irresponsible, and noncaring. Parents from non-mainstream cultures may hold perceptions of children’s learning and the roles of parents and teachers in facilitating that learning that are different from those of the mainstream schools (Purcell-Gates, 2000). Parents who speak languages other than the one used in their children’s schools may find it difficult to support their children and participate in their formal education. In addition, there are differences between parental attitudes based on social class that impact on the relationship between home and school. These include class-related differences in parents’ sense of self-efficacy, their perception of their children’s abilities, their view of their role in their children’s education, and their own attitudes toward school (Eccles & Harold, 1996). Finally, all families regardless of cultural, linguistic, and social background are impacted by modern lifestyle demands. Many adults now spend the largest portion of their waking day doing paid work, and unpaid work takes up much of the remaining time left in a day (Williams, 2002). Families are increasingly rushed with the resulting time squeeze meaning less and less time for exclusive interaction with children and support for their learning. In a recently completed study, mentioned previously in more detail, (Phillips, Hayden, & Norris, 2006), there were a total of 156 rural and urban families of low-income
128
L.M. Phillips
and low-educational backgrounds with children aged 3+ years considered to be at risk for future school failure. Ninety-eight percent of the families include mothers ranging in age from 21 to 26 years with 50% being in their 20s. Forty-seven percent of all family literacy participants are single, and 69% are Caucasian, 13% are Aboriginal, and the rest are Chinese, Kurdish, Jamaican, Spanish, and South African families. The number of children per family ranges from one to six and the average number is 2.5 children. A very high percentage (53%) of all the mothers reported pregnancy as either the cause of their early school leaving or a consequence shortly thereafter. Some are from arranged marriages, some from war-torn countries where girls are not permitted an education, some are from families with alcohol and drug addiction problems, others are from poor and broken families, and some just quit school. The many constellations of families and ethnic groups within our study add diversity and interest. Despite the diversity, we have found that most families are extraordinarily univocal in what they bring to family literacy programs. They bring a reflective, informed, and determined perspective that clusters around five points that are not mutually exclusive. These points include a complicated longing and disabling regret that they did not complete high school; despite unfortunate life circumstances, they are ready to move on in the interest of their children; despite a lack of capital in either money, land, or acquired knowledge they are not deterred; they know there is a “book world” beyond the world they see every day and they want it for their children; and, they want the opportunity for their children and grandchildren “to be able to read and write” and to be “better than me” (Phillips & Sample, 2005, p. 98). My colleagues and I have found large funds of knowledge in low-educational and low-socioeconomic families (Phillips, Hayden, & Norris, 2006; Sample Gosse & Phillips, 2007). We have met families who care deeply and profoundly about their children, and families who hold beliefs about how best to help their children to be successful in school. Our research corroborates the work of Ybarra (1999) on Latino life in the United States: “Despite our diversity, we are nonetheless bound by a common search for a better life for ourselves and our children” (p. 84). Families in the Phillips and Sample (2005) study, which was part of the larger longitudinal study (Phillips, Hayden, & Norris, 2006), astutely recognized the lack of magic in mere reading to their children. However, they directed the problem at themselves and not the schools as is evident in their comments such as: “I must not know how to read to my children”; “I must not be doing it right”; “My boys want to learn their ABCs, but they’re not and I read to them everyday”; and “I am depressed,” “cause I read to Suzie all the time, but she’s not learning to read, if she is, I don’t see it.” Nevertheless, they implied in their expressions of concern a belief that the answers to their children’s reading problems lay in reading to them. Some in the literacy field argue that the reason these children fail in school is because their language and literacy is different from that of the language patterns and literacy events privileged by the schools and that the schools should change to honor the literacy of the home (Auerbach, 1995). The families in our study at no time gave any hint whatsoever that they felt their language and literacy was either deprived or undervalued as demonstrated by their comments in the preceding paragraph.
9 The Making of Literate Families: Considerations of Context and Misconceptions
129
These families saw a need to build on what they were doing at home in order to provide alternative experiences and opportunities for their children to acquire basic knowledge about language and literacy. We also found in our longitudinal and multidimensional study addressed previously, that the families articulated a clear point of view of what they believe they need for their children, provided justifications for, and offered supportive examples of their experiences to demonstrate those needs (Phillips, Hayden, & Norris, 2006). The design and sample included five treatment sites, three urban and two rural, and 158 children ages 3+, and 156 parents. The control families were matched on several factors including age, ethnicity, and sex of the children and received no intervention. The Learning Together: Read and Write with Your Child Program is a three-part family literacy preschool program: adult, child, and joint adult–child daily sessions. The program includes eight units of study for 90 hours across 12 weeks of instruction. The treatment families were taught the Learning Together program prior to formal schooling for the children and all families were tested and interviewed at five points in time: pretest, posttest after 12 weeks, first-year follow-up, second-year follow-up, and third-year follow-up. The families in our study expressed concern in one-on-one interviews that they wanted to move beyond what they were doing at home for their children. In accord with the work of Sénéchal, Lefevre, Thomas, and Daley (1998), the families in our study expressed a desire to learn more about the ways to engage in formal literacy experiences wherein the goal is to focus more on the print so that their children would be better prepared to understand and engage in what Catherine Snow (1983) calls decontextualized print and Shirley Brice Heath (1991) calls the sense of being literate. The families are the insiders in the family literacy debate and they have their beliefs about what they need to help their children. Academic debates by outsiders who claim that family literacy programs privilege school-based literacy rather than home-based literacy are suppressing the very wishes and needs of the insiders. Successful, schooling experiences for their children is what most concerns these families. Family literacy research and programs based on a familial perspective is clearly an important alternative route against the dissenting point of view about families’ perceived literacy needs. Dissension over whether and how families interact with their children can have far reaching and long-term consequences. Underpinning parents’ wishes lie certain expectations for learning and they recognize that they need more than their previous experiences and current knowledge and parents want help sooner rather than later. Regrettably, low-income families bear the brunt of inconsiderate and unfair remarks about their misuse of time. Time is argued to be a vital element in building and fortifying family relationships; however, because of poverty of resources, lack of discretionary funds, unrestrained television watching, and emotional burdens the use of family time is thought to be idiosyncratic and chaotic among the poor and less-educated, and those relying on welfare (Tubbs, Roy, & Burton, 2005). Authors of literacy advice texts expect mothers to manage home space and time in order to do all that is institutionally expected of them and imply that if parents really care about their children they will “make time” to read to their children, engage in family activities, and create
130
L.M. Phillips
meaningful activities (Trelease, 1995). These expectations are perceived to be possible within middle and higher class families, and thus lower class families unable for multiple reasons to conform are dismissed as irresponsible and non-caring (Lasky, 2000). Folklore and urban myths take hold and often make it difficult to dissuade parents of beliefs about what and when to teach their children. There has been a mantra for decades, for example, that teaching children letter names interferes with their learning to read (Feitelson, 1965). Research by others, however, documented the highly positive and significant relationship between knowledge of letter names and reading and spelling acquisition (e.g., Blatchford & Plewis, 1990; Scarborough, 1990; Share, Jorm, Maclean, & Matthews, 1984). These studies were conducted with English-speaking children only until Levin, Patel, Margalit, and Barad (2002) analyzed the role of letter names in connecting speech and print in Hebrew in a set of two interconnected studies. They studied 80 kindergartners from upper-middle class native Hebrew-speaking families, ten were removed because the children knew fewer than ten letters. These children were retested in Grade one (66 in total); of which three were retained in kindergarten and three others either moved or did not consent. The first study with kindergartners showed significant and positive effects of mapping letter names onto oral words, letter saying, and written spelling. In the second study, 54 kindergarten children also upper-middle class native Hebrew speakers, Levin, Patel, Margalit, and Barad (2002) found letter names to have a positive and significant effect on word recognition as well as on oral and written spelling. These results are consistent with those of English-speaking children and prompted a call for early teaching of the letter names. Note in study one that 14% of the children from upper middle-class families knew fewer than ten letters and experienced sufficient difficulties in kindergarten to be retained. These children performed more like children from lower-class families. Regardless of class, the time to teach letter names is sooner rather than later. Knowing the letter names can be likened to having money in the bank, because as Levin and her colleagues suggest knowing the letter names can also have a socio-cognitive function, “It is easier for adults to support children in printing a word or in searching for a written word when the children have letter knowledge that affords common literate discourse” (p. 29). One way that very young children gain from letter knowledge is from experiences with their own names. In a sample of 243 children ranging in age from 2 years and 11 months to 4 years and 11 months and all from a poor neighborhood on the outskirts of Tel Aviv, Levin took the lead on another interesting and larger study on children’s writing (I report here on only one aspect). The children were asked to write their own names as well as four dictated words with varying phonological length from one to four syllables (carrot, cucumber, chicken, fish). They found that more children scored higher on the writing scale and illustrations (graphic, writing like writing, symbolic) on their name than on dictated words and this was true for children across the age range and SES levels. They found the gap between name and word writing was significantly larger for the children in the older group than for the younger children. Enriching children’s understanding of their name by “naming the letters or linking the sounds to the letters, children become aware of letter names and letter sounds, and eventually of the alphabetic principle itself” (Levin, Both-De
9 The Making of Literate Families: Considerations of Context and Misconceptions
131
Vries, Aram, & Bus, 2005, p. 476). Thus far we have seen from Professor Levin’s research that children can learn both more and sooner than normally thought. Misconceptions about children and their families have been shown to lead to unexpected and negative responses by many, including teachers. In a telling study, Wesley and Buysse (2003) conducted focus groups with parents, prekindergarten teachers, kindergarten teachers, and elementary school principals to examine beliefs and expectations about school readiness among parents and professionals. While parents and school professionals shared a common concern with the overall development of children, principals and kindergarten teachers often referred to great contrasts in the richness of children’s experiences prior to school, referring to “haves” and “have-nots” (p. 362). The school professionals reported a need to educate parents about how to teach their children certain skills prior to school entry. The kindergarten teachers held negative views about parents, who they saw as failing to do their job as children’s first teachers (see also the work of Cairney, 1997; Sample Gosse & Phillips, 2007). Returning to the study by Korat and Levin (2001) with 20 low SES and 20 high SES mothers, they confirmed that low SES mothers’ beliefs about their second grade children’s writing level were realistic but that more local writing features such as spelling were of concern rather than the more global writing features such as genre. High SES mothers, on the other hand, considered spelling to be developmental and focused more on content. But, fundamentally the high SES and low SES mothers both wanted their children to develop independence in their writing, ownership of their writing, and were aware of their children’s’ developmental constraints. Yet, the prevalent view is that the parents from these two SES groups have different pedagogical beliefs with those from the high SES valuing independence and ownership and those from the low SES not even interacting with their children. It seems neither the low nor high SES mothers were represented in the prevalent view. Lasky (2000) found that teachers reported feeling negative emotions when parents were viewed as uncaring or irresponsible; not supportive of teachers’ efforts; or not respectful of teachers’ professional judgment and these emotions made teachers less willing to extend themselves to parents (p. 849). The teachers in the Lasky study expected parents to conform to institutional norms of appropriate parent behavior based on middle-class mothers in two-parent homes (p. 853). Three years on, Hauser-Cram, Sirin, and Stipek (2003) found that when teachers believed the education-related values of parents differed from their own values, they rated children as less competent academically and held lower expectations for their future success. These lowered ratings continued to be evident even when children’s socioeconomic status and academic skills were controlled. The results of these studies suggest that today’s teachers may have difficulty collaborating with diverse families unless they can be challenged personally and professionally to change their views of the families. Moreover, parenting practices and schooling outcomes are not strongly related to SES or family structure. When parents were categorized according to their style (authoritative, authoritarian, permissive, irresponsible), the groups were not homogenous with respect to SES. Moreover, a full range of variables describing family structure (single, married, teenage parent) and SES accounted for only 2–6% of the variation in parents’ practices (rational, responsive, firm approach, reasons
132
L.M. Phillips
with child), (Chao & Willms, 2002, p. 164). Low SES children are vulnerable, and especially so when schools hold lower expectations for their literacy success. As Graue and Brown (2003) suggested – “mismatches in beliefs and practices represent missed opportunities by school people as much as they represent parenting that is limited in its outcomes” (p. 732). Children are the ones to lose the most from missed opportunities. An example is the different expectations that teachers and parents hold about when children will learn to read. In an attempt to encourage parents to read to and to have their children read to them, teachers easily overlook the fact that parents often do not have professional knowledge of reading. Parents frequently do not understand the distinction between a book that is meant to be read to children and one that is to be read by the child. For instance, children can listen to and comprehend a story well beyond their independent level of reading – many parents do not know this fact. What would happen if families were taught to mediate their preschool children’s literacy development? Evidence from a recent longitudinal quasi-experimental, control group, mixedmethods study extending across 5 years with 156 parents and their preschool children 3+ years from low-income and low-educational backgrounds (Phillips, Hayden, & Norris, 2006) revealed that age was not the determining factor in children’s literacy development but rather what children did not know about the essential print concepts to be successful in reading. The Learning Together program worked for children regardless of gender and no matter their beginning reading age between 36 and 60 months. In essence, there was no time better than another to intervene if the children did not know the concepts of print necessary to advance in their reading development. Time was not the factor; rather, what the children knew or more specifically what they did not know was the factor. This study showed that when low-income and low-educational background families are taught specific skills and strategies to read and write, their children learn. The Learning Together program continued to have a positive influence for children in the lowest 70–80% of scores at the pretest until we ceased follow-up 3 years after the program ended. When parents know how to teach and support their children’s reading and writing, they will try to so, and when children know what is expected of them, they actively engage. These children and their families demonstrated that letter knowledge, phoneme awareness, word recognition, and story comprehension can be learned with direct instruction, explicit expectations, and active engagement in print that matters. Thus, the work of Iris Levin and others has served as a catalyst to tell a more positive story of the capabilities of families regardless of socio-cultural factors than has been the case for most of the last two centuries.
Concluding Remarks Literacy has historically been the purview of educators. Currently, we are witnessing an expansion that includes families. Unfortunately, many educators have misconceptions that low-educational and low-income families do not care about their children. Rather, it is often the case that these families do not understand when and how best
9 The Making of Literate Families: Considerations of Context and Misconceptions
133
to help the development of their children’s language and literacy. In this chapter, I have attempted to show that early literacy interventions are critically important in the making of literate families by pointing out that considerations of context and popular and prevalent misconceptions of early language and literacy development require a common goal and mutual respect. Capitalizing upon this view as a foundation, I attempted to demonstrate the critical necessity for obtaining increased understanding of the views maintained, challenges faced, and dreams held by families. There is much to gain from a respectful relationship between literacy in the home and literacy in the school. Such a relationship, I conclude, unlocks promising new courses of action in and expands the possibilities for the making of literate families.
References Aram, D., & Levin, I. (2001). Mother-child joint writing in low SES: sociocultural factors, maternal mediation, and emergent literacy. Cognitive Development, 16, 831–852. Aram, D., & Levin, I. (2002). Mother-child joint writing and storybook reading: relations with literacy among low SES kindergartners. Merrill Palmer Quarterly, 48, 202–224. Auerbach, E. R. (1995). Which way for family literacy: intervention or empowerment? In L. Morrow (Ed.), Family literacy connections in schools and communities (pp. 11–28). Newark, DE: International Reading Association. Blatchford, P., & Plewis, I. (1990). Preschool reading related skills and later reading achievement: further evidence. British Journal of Educational Psychology, 16, 425–428. Brodkin, A. M. (2002). I’m at daddy’s this week. Early Childhood Today, 17(3), 20–22. Burns, M. S., Griffin, P., & Snow, C. E. (1999). Starting out right: a guide to promoting children’s reading success. Washington: National Academy Press. Cairney, T. H. (1997). Acknowledging diversity in home literacy practices: moving towards partnership with parents. Early Childhood Development and Care, 127, 61–73. Chao, R. K., & Willms, J. D. (2002). The effects of parenting practices on children’s outcomes. In J. D. Willms (Ed.), Vulnerable Children: Findings from Canada’s National Longitudinal Survey of Children and Youth. Edmonton, AB: The University of Alberta Press. Churchill, S. L. (2003). Goodness-of-fit in early childhood settings. Early Childhood Education Journal, 31, 113–118. Eccles, J. S., & Harold, R. D. (1996). Family involvement in children’s and adolescents’ schooling. In A. Booth & J. F. Dunn (Eds.), Family school links (pp. 3–34). Mahwah, NJ: Lawrence Erlbaum Associates. Feitelson, D. (1965). Structuring the teaching of reading according to major features of the language and its script. Elementary English, 42, 870–877. Gadsen, V. L. (2000). Intergenerational literacy within families. In M. L. Kamil, P. B. Mosenthal, P. D. Pearson & R. Barr (Eds.), Handbook of reading research (Vol. III, pp. 871–887). Mahwah, NJ: Lawrence Erlbaum Associates. Gordon, M. (2003). The early years: the precious years. Vancouver, BC: Roots of Empathy. Graue, E., & Brown, C. P. (2003). Preservice teachers’ notions of families and schooling. Teaching and Teacher Education, 19, 719–735. Hauser-Cram, P., Sirin, S. R., & Stipek, D. (2003). When teachers’ and parents’ values differ: teachers’ ratings of academic competence in children from low income families. Journal of Educational Psychology, 95(4), 813–820. Heath, S. B. (1983). Ways with words. Cambridge: Cambridge University Press. Heath, S. B. (1991). The sense of being literate: historical and cross-cultural features. In R. Barr, M. L. Kamil, P. B. Mosenthal & P. D. Pearson (Eds.), Handbook of reading research (Vol. II, pp. 3–25). New York: Longman.
134
L.M. Phillips
Hulbert, A. (2005). The real myth of motherhood: reconsidering the maternal memoir-cum-manifesto. Slate. Retrieved October 17, 2007, from http://www.slate.com/id/2114498/nav/ais/. Kaye, M. (2003). Divorced? Protect your kids. Homemakers, 48–55. Korat, O., & Levin, I. (2001). Maternal beliefs, mother-child interaction, and child’s literacy: comparison of independent and collaborative text writing between two social groups. Journal of Applied Developmental Psychology, 22, 397–420. Lasky, S. (2000). The cultural and emotional politics of teacher-parent interactions. Teaching and Teacher Education, 16, 843–860. Levin, I., Both-De Vries, A., Aram, D., & Bus, A. (2005). Writing starts with own name writing: from scribbling to conventional spelling in Israeli and Dutch children. Applied Psycholinguistics, 26, 463–477. Levin, I., Patel, S., Margalit, T., & Barad, N. (2002). Letter-names: effect on letter retrieval, spelling and word recognition in Hebrew. Applied Psycholinguistics, 23, 269–300. McCormick, C. E., & Mason, J. M. (1986). Intervention procedures for increasing preschool children’s interest in and knowledge about reading. In W. H. Teale & E. S. Sulzby (Eds.), Emergent literacy: writing and reading (pp. 90–115). Norwood, NJ: Ablex Publishing Corporation. Ninio, A. (1980). The naïve theory of the infant and other maternal attitudes in two subgroups in Israel. Child Development, 50, 976–980. Phillips, L. M., Hayden, R., & Norris, S. P. (2006). Family literacy matters: a longitudinal parentchild literacy intervention study. Calgary, AB: Detselig. Phillips, L. M., Norris, S. P., & Anderson, J. (2008). Unlocking the door: is parents reading to children the key to early literacy development? Canadian Psychology, 49, 82–88. Phillips, L. M., & Sample, H. (2005). Family literacy: listen to what families have to say. In J. Anderson, M. Kendrick, T. Rogers & S. Smythe (Eds.), Critical issues in family, community, and school literacies: intersections and tensions (pp. 91–107). Mahwah, NJ: Erlbaum. Purcell-Gates, V. (2000). Family literacy. In M. L. Kamil, P. B. Mosenthal, P. D. Pearson & R. Barr (Eds.), Handbook of reading research (Vol. III, pp. 853–870). Mahwah, NJ: Lawrence Erlbaum Associates. Rogoff, B. (1990). Apprenticeship in thinking: cognitive development in social context. New York: Oxford University Press. Sample Gosse, H. L., & Phillips, L. M. (2006). Family literacy in Canada: foundation to literate society. In A. McKeough, L. Phillips, V. Timmons & J. Lupart (Eds.), Understanding literacy development: a global view (pp. 113–135). Mahwah, NJ: Erlbaum. Sample Gosse, H. L., & Phillips, L. M. (2007). No gain in blame: fostering collaborations between home and school. In Y. Goodman & P. Martens (Eds.), Critical issues in early literacy: research and pedagogy (pp. 191–201). Mahwah, NJ: Lawrence Erlbaum Associates. Scarborough, H. S. (1990). Very early language deficits in dyslexic children. Child Development, 61, 1728–1743. Sénéchal, M., LeFevre, J., Thomas, E., & Daley, K. (1998). Differential effects on home literacy experiences on the development of oral and written language. Reading Research Quarterly, 33(1), 96–116. Share, D. L., Jorm, A. F., Maclean, R., & Matthews, R. (1984). Sources of individual differences in reading acquisition. Journal of Educational Psychology, 76, 1309–1324. Smythe, S. (2006). The good mother: a critical discourse analysis of literacy advice to mothers in the 20th century. Unpublished doctoral thesis. University of British Columbia, Vancouver, British Columbia, Canada. Snow, C. (1983). Literacy and language: relationships during the preschool years. Harvard Educational Review, 53, 165–189. Taylor, D. (1983). Family literacy: young children learning to read and write. Exeter: Heinemann. Trelease, J. (1995). The read-aloud handbook (Third revised edition ed.). New York: Penguin. Tubbs, C. Y., Roy, K. M., & Burton, L. M. (2005). Family ties: constructing family time in lowincome families. Family Process, 44(1), 77–91. Wesley, P. W., & Buysse, V. (2003). Making meaning of school readiness in schools and communities. Early Childhood Research Quarterly, 18(3), 351–375.
9 The Making of Literate Families: Considerations of Context and Misconceptions
135
Williams, C. (2002). Time or money? How high and low income Canadians spend their time. Canadian Social Trends, Summer 2002, No. 65. Ottawa, ON: Statistics Canada. Ybarra, L. (1999). The family/la familia. In E. J. Olmos, L. Ybarra & M. Monterrery (Eds.), Americanos: Latino life in the United States/La vida Latina en los Estados Unidos (p. 84). New York: Little, Brown & Company.
Chapter 10
Spellings of Words: A Neglected Facilitator of Vocabulary Learning Linnea C. Ehri and Julie Rosenthal
Abstract Vocabulary learning is central to reading ability and academic achievement. Vocabulary researchers and educators have viewed its essence as a process of associating the pronunciations and meanings of words in memory, and they have neglected the contribution that spellings might make to vocabulary learning. We review theory and evidence that shows this to be a serious oversight. Once children become literate, they retain the spellings of words bonded to their pronunciations and meanings in memory. Several studies show that spellings of words are retained in memory and influence phonemic and syllabic segmentation of words. They enhance memory for pseudowords, and they impact the detection of oral rhyming words. Two studies show that exposing second and fifth graders to the spellings of new vocabulary words enhances their memory for pronunciations and meanings of the words. Students with better developed orthographic knowledge benefit more from spellings in learning vocabulary words than students with weaker knowledge. In fact, the detection of a Matthew effect suggests that differences in orthographic knowledge create a difference in vocabulary size that grows increasingly large over time. These findings carry implications for enhancing vocabulary learning and instruction. Teachers need to show the spellings of new vocabulary words when they discuss their meanings. Students need to stop and pronounce unfamiliar words rather than skip them during independent reading. Researchers need to incorporate orthography into their theories explaining vocabulary acquisition, specifically, phonological working memory theories, and they need to attend to its influence in the studies they conduct. The heart of language development is vocabulary learning. Although acquisition centers on learning individual words and their meanings, it extends much beyond this (Nagy & Scott, 2000). As words are learned, they become embedded in a network of semantic connections to other words. Not only verbal definitions and interconnections, L.C. Ehri (*), and J. Rosenthal City University of New York Graduate Center, New York, NY, USA e-mail:
[email protected] This article is reprinted with minor adaptations from the article by L. Ehri and J. Rosenthal, (2007), “Viewpoints: Spellings of words: A neglected facilitator of vocabulary learning” that appeared in Journal of Literacy Research, 39(4), 389–409. Reprinted by permission of Taylor & Francis, http://www.informaworld.com. D. Aram and O. Korat (eds.), Literacy Development and Enhancement Across Orthographies and Cultures, Literacy Studies 101, DOI 10.1007/978-1-4419-0834-6_10, © Springer Science+Business Media, LLC 2010
137
138
L.C. Ehri and J. Rosenthal
but also visual images contribute to the representations of words in memory (Sadoski, 2005). The correlation between students’ vocabulary knowledge and their reading comprehension is extremely high and positive (rs ranging from .63 to .73), almost as high as the correlation between different measures of reading comprehension (rs ranging from .64 to .79; Cutting & Scarborough, 2006). The heavy involvement of vocabulary in reading ability underscores its importance for academic achievement in school. Thus, it is essential to advance our understanding about how students’ vocabularies grow and how instruction can support this growth. The purpose of this chapter is to consider theory and evidence regarding the contribution and importance of word spellings for vocabulary learning and instruction. First, we review background studies revealing the way written words are stored in memory and how they influence speech. This research has revealed that spellings exert a powerful influence, so it is surprising that the contribution of orthography to vocabulary learning has been neglected by vocabulary researchers. Next, we present evidence for the effect of written words on vocabulary learning by summarizing the results of two experiments. Finally, we discuss the implications of this work for theory and practice.
Possible Ways to Read Words Consider the following three sets of words and the processes that readers might use to read each set: Set 1: rume, rane, taik, gote, yung, pillgrum, kartune, selafain Set 2: said, was, one, tongue, sugar, ocean, iron, yacht Set 3: faster, step, grass, hunger, elbow, interesting, excellent, contribution The first set consists of single- and multi-syllabic pseudoword spellings that are unfamiliar to readers. The second set consists of irregularly spelled real words, including three high frequency words from the Dolch list and five common long words. The third set consists of common, regularly spelled words. Ehri (1998) has distinguished four different ways to read words. Words might be decoded by applying knowledge of grapheme–phoneme relations or larger syllabic units to convert spellings to pronunciations whose identities are then recognized by accessing their meanings in memory. This process would work to read the pseudo-words in Set 1. Their spellings are unfamiliar but when decoded, they are recognized as real words. A second way to read words is by analogy to words whose spellings are already known, for example, reading rane by analogy to plane. A third way is by prediction using context cues, partial letters, or a combination, for example, “doctors and n ….” The foregoing strategies are useful mainly for reading unfamiliar words, including new vocabulary words. The fourth way portrays how words that have been read previously are processed: by memory or sight. The sight of such words activates their pronunciations and meanings in memory. When words are read from memory, they are accessed as whole
10 Spellings of Words: A Neglected Facilitator of Vocabulary Learning
139
units, without prior decoding or analogizing or prediction. Reading is immediate and automatic. Because the words in Set 2 are irregularly spelled, a decoding strategy will fall short, so they must be remembered and read by sight. But what about the words in Set 3? They are regularly spelled, so they could be decoded. But they have been read before, so they could also be read by sight. Notice that you recognized the meanings of Set 3 words differently from Set 1 words. Set 3 pronunciations and meanings were recognized immediately from their written forms without any decoding or lexical search. Ehri’s (1980, 1992, 2005) research and that of others indicates that not just high frequency or irregularly spelled words but all written words when practiced become familiar and are read from memory. This would include new vocabulary words once their pronunciations and meanings are learned.
Reading Words from Memory If all familiar words are read from memory, then explaining how this capability is acquired becomes an important part of explaining how reading skill develops. People used to think that sight words were read by rote memorizing the shapes or visual forms of words. However, such a view cannot explain the facts: how thousands of words are so well remembered, how visually similar words are so easily distinguished, and how learning happens so quickly. To explain word memory, a more powerful mnemonic system is needed. Such a mnemonic is provided by the alphabetic writing system. According to Ehri (1992), memory for written words involves a process of forming connections between spellings, pronunciations, and meanings in memory. Knowledge of the grapheme–phoneme system provides the glue that connects graphemes in spellings to phonemes in spoken words. This information is stored as amalgams representing individual words in memory, as diagramed in Fig. 10.1.
Forming Word Amalgams in Memory • Process of forming connections Spelling Meaning glue
Pronunciation Knowledge of the grapheme-phoneme system provides the glue connecting spellings to pronunciations in memory
Fig. 10.1 Diagram of connection-forming process to secure the spellings of words to pronunciations and meanings in memory
140
L.C. Ehri and J. Rosenthal
To secure regularly spelled words in memory, connections are formed linking each grapheme to its phoneme. For example, each letter in step is connected to a phoneme. In the word grass containing four letters but three phonemes, the final two letters (a digraph) are connected to one phoneme /s/. The word check has five letters forming three graphemes (CH–E–CK), each connected to one of the three phonemes. Connections apply not only to regularly spelled words but also to irregularly spelled words to store them in memory. It turns out that most letters in irregularly spelled words can be connected to phonemes in their pronunciations, for example, all but the S in island, all but the W in sword, all but the UE in tongue. From these examples, it becomes clear what knowledge is needed for readers to store words in memory. Phonemic segmentation skill is needed to analyze phonemes in pronunciations of words. Knowledge of grapheme–phoneme correspondences provides the glue. Grapho-phonemic mapping skill is needed to connect graphemes to phonemes within specific words. When students apply a decoding strategy to read unfamiliar words, this activates mapping relations. Share (2004) has referred to the use of decoding as a self-teaching strategy that establishes written words in memory. As readers retain blends of graphemes and phonemes in memory, larger sequences of letters become familiar units useful for securing words in memory. These sequences may be spellings of words, or parts of words such as rime spellings (e.g., –AME in name, same, came), or syllables. Multi-syllabic words may be remembered by forming connections between these larger written units and syllables in pronunciations, for example, the three units in SUB–STI–TUTE.
Specific Letters Are Remembered Various lines of research have provided evidence for the connectionist theory of sight word learning. Studies have shown that when unfamiliar words are read, specific letters seen in the words are retained in memory. Ehri (1980) had second graders practice reading eight identically pronounced pairs of made-up words naming animals. For example, some practiced weeple while others practiced wheople. After a delay, students wrote the words from memory. She found that if they had read WH, they always wrote WH, never just W. Likewise, if they had read W, they always wrote W, never WH. This was true not only when they spelled words correctly but also when they misspelled other parts of the words. This showed that students did not just decode print to speech, store only pronunciations in memory, and later devise spellings of the words. Rather they stored the specific letters for those words in memory. Reitsma (1983) gave first graders varying numbers of practice trials to learn to read alternative written versions of several identically pronounced word pairs. He found that four trials were the minimum needed for students to show by their spellings that they were reading the words from memory. Share (2004) had third graders practice reading words spelled in one or another way in text. He found that students remembered letters unique to the words they had read even when they had seen the
10 Spellings of Words: A Neglected Facilitator of Vocabulary Learning
141
words only once, and moreover they remembered them up to a month later. These findings suggest that retaining spellings of words in memory is easy and happens quickly once students acquire knowledge of the writing system.
Spellings of Words Influence Speech Studies have provided evidence that spellings are bonded to pronunciations of words in memory. Ehri and Wilce (1979) gave first and second graders several practice trials to learn a set of spoken pseudowords, for example, mav, rel, kip, guz. In one condition, spellings accompanied the spoken words during study periods. In the control condition, the words were pronounced several times but spellings were not present. Findings showed that students remembered the spoken words better when they had seen spellings than when they had not. Spellings were not present when children recalled pronunciations of the pseudowords, so the boost had to come from the presence of spellings secured to pronunciations in memory. This effect is commonly experienced when an unfamiliar personal name is more easily remembered when its spelling is seen. In studies using a phonemic segmentation task, it was reasoned that if graphemes are glued to phonemes in specific words stored in memory, then graphemes should influence how readers analyze phonemes in the words when there is ambiguity, that is, when there is more than one way to analyze phonemes in the words and when spellings can differ on this point. Ehri and Wilce (1980) compared how readers segmented words such as pitch and rich into phonemes. Although the rimes of these words are pronounced identically, it is ambiguous whether or not these two words contain a medial /t/ phoneme. If one pronounces the words, it is possible to distinguish a /t/ in articulation (i.e., tongue touching the roof of the mouth). But the presence of /t/ is flagged in only one of the spellings. Findings revealed that fourth graders segmented pitch into four phonemes (P–I–T–CH) but rich into only three phonemes, and they never detected a /t/ sound in rich. Spellings were not shown so students’ segmentations were influenced by their memory for spellings. Students were not just segmenting letters because the two letters in digraphs did not prompt separate segments. These findings were verified in a controlled experiment that taught students to read unfamiliar words with ambiguous spellings and then assessed how they segmented the words. In another study, Ehri and Wilce (1986) examined what sound readers thought was in the middle of words like ladder and letter. In American English, the sound is the same in both words, an intervocalic alveolar flap closer to /d/ than /t/. Results of an experiment in which readers practiced words with these flaps, either by reading them or by speaking them, showed that exposure to the spellings influenced their conceptualizations. Students who saw words with flaps spelled T were much more likely to regard them as containing the /t/ phoneme than students who only practiced saying these words. Not only grapheme–phoneme units but also grapho-syllabic units influence the sounds that readers conceptualize in words. Ehri (1984) examined good and poor
142
L.C. Ehri and J. Rosenthal
spellers’ ability to segment ambiguous multisyllabic words into syllables. She found that good spellers tended to segment the words into syllables suggested by spellings whereas poor spellers’ segmentations reflected conflated pronunciations, for example, interesting segmented as in-ter-es-ting by good spellers versus in-tres-ting by poorer spellers, and comfortable segmented as com-for-ta-ble by good spellers versus comf-ter-ble by poor spellers. Seidenberg and Tanenhaus (1979) showed that spellings stored in memory influenced performance in an orally conducted rhyming task in which words were heard but not seen. When the rime spellings of word pairs that rhymed were consistent, judgments were facilitated whereas when the rime spellings were consistent but words did not rhyme, judgments were impeded. For example, students were faster judging that clue and glue rhymed than judging that clue and shoe rhymed. Students took longer to judge that bomb and tomb did not rhyme than to judge that bomb and room did not rhyme. These findings support the idea that the connections linking written words to pronunciations in memory include larger units such as subsyllabic rime spellings.
Vocabulary Learning The essence of vocabulary learning is regarded as associating the meanings of new words with their pronunciations. Although written words may be included during instruction, their presence has not been regarded as making much difference in what students learn. The commonly held view seems to be that when new words are learned through reading, the written forms are converted to pronunciations and it is the pronunciation that is associated with the meaning and stored in memory. The spelling disappears from the picture. In making recommendations about how to teach vocabulary more effectively, experts have said little about the spellings of words. For example, Blachowicz and Fisher (2004) recommended several steps to improve teachers’ success in expanding students’ vocabularies: reading text aloud to their classes, stopping at difficult words and providing definitions, rereading text to cement new word meanings, and having students act out the meanings of new words. However, they did not suggest displaying the spellings of new words. Although Beck and McKeown (2002) advised teachers to post the spellings of vocabulary words on a word wall after they taught the words from trade books that were read aloud, they did not clarify why this might be an important step.
Two Experiments There is good reason to expect that spellings will help students learn new vocabulary words. According to the connectionist theory of word learning, spellings help to secure pronunciations of words in memory by connecting graphemes to phonemes.
10 Spellings of Words: A Neglected Facilitator of Vocabulary Learning
143
So exposure to spellings should speed up learning the pronunciations of new vocabulary words. If pronunciations are stored in memory earlier, then meanings also may be learned earlier as well. We conducted two laboratory experiments to examine whether spellings improve the learning of new vocabulary words (Rosenthal & Ehri, 2008). We worked with students individually. We gave them an explicit learning task that involved teaching sets of words on flash cards. The words selected were low-frequency nouns that students were not expected to know, as affirmed by their teachers. Students rehearsed the pronunciations and meanings of the words over several trials. During the initial study trial, the words were introduced. All subsequent trials tested their recall of the words. After each recall attempt, correct responses were provided. To teach the meanings of words, we used pictures, definitions, and multiple sentences containing the words and clarifying their meanings and use. Trials continued until students reached a criterion or a maximum number of trials. Each student learned two sets of vocabulary words. In the treatment condition, they learned spoken words that were accompanied by spellings during study periods (i.e., when the words were introduced and after each recall attempt). In the control condition, students learned spoken words without spellings. We measured how quickly students learned the words as the learning trials progressed, and how well they remembered them after a delay of 1 day. It is important to note that when recall of pronunciations of the words was tested, spellings were not present, so any benefit had to come from memory for the spellings. Also it is important to note that when spellings of the words were shown, no attention was drawn to their presence, and students had no need to decode them because as soon as the words were shown, the experimenter pronounced them. Thus, retention of spellings in memory happened incidentally. The hypothesis tested was that students will learn the pronunciations and meanings of new words better when they see spellings of the words during study periods than when they do not. Our explanation was that grapheme–phoneme connections would be activated by spellings and as a result would secure the pronunciations and meanings of words in memory earlier during learning.
Experiment with Second Graders In the first experiment, the participants were 20 second graders, mean age 7 years, 7 months, enrolled in an urban school with a large minority population. On average, students were reading at the second grade level but their vocabulary level was below average. Each student was taught the pronunciations and meanings of two sets of six concrete nouns. Spellings were shown as students learned one set. Spellings were not shown as students learned the other set. The particular word set and the order that students completed the conditions were counterbalanced across students. Examples of the words taught are: Gam, a family of whales; Cur, a homeless dog; Sod, wet, grassy ground; Yag, fake jewelry; Keg, a barrel. During learning, the words were not
144
L.C. Ehri and J. Rosenthal
only defined but also embedded in different sentences to clarify meanings. For example, sentences taught for the word keg were: A keg can hold many kinds of liquids. A keg is big and can hold a lot of water. A keg is usually made out of wood or plastic. A keg is big and round. Some people keep pickles in a keg. In the spelling present treatment condition, the following steps were employed to teach vocabulary words. An initial study trial occurred first. The six words, their spellings, and their meanings were introduced. For each, a card was displayed with a drawing of the object named by the noun and a spelling printed beneath the picture. The experimenter pronounced the word and its definition and the student repeated them. The remaining trials tested students’ recall of the words’ pronunciations and meanings. All six words were tested on each trial. Pronunciation recall trials were interleaved with definition recall trials. Students were given a minimum of six and a maximum of nine trials to learn pronunciations and meanings to a criterion of three perfect consecutive trials. On pronunciation recall trials, the student saw each drawing with no spelling present and tried to recall its pronunciation. Then the experimenter gave the correct answer by pronouncing the word, showing its spelling, giving its definition, and embedding the word in a clarifying sentence. The student repeated the word and sentence. Different clarifying sentences were given on different test trials. On definition recall trials, the student heard each word pronounced, saw its spelling, and tried to recall its meaning. Then the experimenter supplied the correct answer by giving the word’s definition followed by a clarifying sentence. The student repeated the word and its meaning. In the spelling absent control condition, the procedures were the same except that the spellings of words were never shown. Students pronounced the words extra times, actually several more times than students in the spelling present condition, to make sure that practice was not a factor explaining any difference in memory for the words. Figure 10.2 displays students’ performance as word learning progressed over the test trials. It was easier to recall meanings than pronunciations of the words. Recall was superior when spellings were seen than when they were not seen. This advantage held for the recall of meanings as well as the recall of pronunciations. In recalling pronunciations, the benefit of seeing spellings grew larger over trials. It was not possible to observe a parallel effect for the recall of meanings because ceiling effects limited improvement over trials. All of the differences were statistically significant. The day after each set of words was taught, posttests were given to assess delayed memory for the pronunciations, spellings, and meanings of the words. Results are shown in Fig. 10.3. In the pronunciation recall test, students heard each definition and recalled the word. Findings showed that after 1 day, students still recalled pronunciations better when they had seen spellings than when they had not, indicating that the impact of spellings lasted beyond the end of training. In the spelling recall test, students spelled words they had seen more accurately than words they had not, indicating that the spellings they saw were stored in memory. In the meaning recognition test, students matched the words to clarifying sentences they had heard during training. Their performance was almost perfect with no
10 Spellings of Words: A Neglected Facilitator of Vocabulary Learning
145
Mean number correct by trial Number of words/definitions
6 5 4
WD spellings WD no spellings
3
PW spellings PW no spellings
2 1 0 1
2
3 Trial number
4
5
Fig. 10.2 Mean number of spoken words recalled (PW) and definitions recalled (WD) over trials as a function of whether spellings were present or absent during learning in Experiment 1
Mean Percent Correct on Posttests 100% 80% 60%
Spell Seen Spell Not Seen
40% 20% 0%
Pronunc.
Spellings
Meanings
Posttest Measures Fig. 10.3 Mean percentage correct in recalling pronunciations, spellings, and meanings of vocabulary words on the delayed posttests when spellings had accompanied vocabulary learning (in gray) and when spellings had not accompanied learning (in white) in Experiment 1
difference between conditions, indicating that students had mastered the meanings of both sets of words. From this study we conclude that second graders learned the pronunciations and meanings of vocabulary words better when they were exposed to spellings of the words than when they only practiced speaking the words. Some might consider these findings quite remarkable because the spellings were simply exposed without any attention drawn to them by the experimenter. In addition, as soon as they were shown, the experimenter pronounced the words, so children had neither time nor need to decode the words. Our explanation is that when spellings were seen, heard, and repeated by the child, grapho-phonemic connections were spontaneously activated
146
L.C. Ehri and J. Rosenthal
to secure the pronunciations in memory. Better secured pronunciations provided a stronger base for attaching meanings earlier during the course of learning.
Experiment with Fifth Graders We performed a second experiment to see if findings would hold for older students. We reasoned that because older students have greater knowledge of the orthographic system than younger students, their vocabulary learning might benefit even more by seeing spellings of the words. Alternatively, their larger spoken vocabularies may make it easier to learn new words without spellings, so little advantage might result. Also having greater decoding proficiency, they may anticipate the spellings of new words spontaneously even when the words are only heard (Stuart & Coltheart, 1988). The participants were fifth graders from the same lower SES school as the second graders. Students were divided into two groups that differed in their orthographic knowledge: 14 higher level readers, and 18 lower level readers. The higher group read real words at a 7.3 grade equivalent (GE) level and nonwords at a 4.8 GE level. The lower group read words at a 4.6 GE level and nonwords at a 2.2 GE level. The groups also differed in their spelling ability. The same procedures were followed as before with a few exceptions. The words were low-frequency nouns that were longer, consisting of two and three syllables. Ten words were taught in each condition. Students were given a minimum of five and a maximum of eight trials to learn pronunciations and meanings of the words to a criterion of three perfect successive trials. Examples of the words that were taught are: Barrow, a small hill; Tandem, a horse-drawn carriage; Fribble, a foolish shallow person; Tamarack, a big tree found all over America; Proboscis, a really big nose. Meanings were taught with pictures, defining sentences, and clarifying sentences as before. Two hypotheses were tested: (1) Spellings will help fifth graders learn the pronunciations and meanings of new vocabulary words better than no spellings; (2) Students with stronger orthographic knowledge (higher level readers) will benefit more from spellings than students with weaker orthographic knowledge (lower level readers). Figure 10.4 shows the mean performance of higher and lower level readers as they recalled pronunciations of the words during the first five trials. It is apparent that higher readers outperformed lower readers. Recall of pronunciations was better when words were learned with spelling aids than without spelling aids. This was true for both groups. Statistical tests confirmed these differences. The advantage provided by spellings became apparent on the first recall trial for the higher readers but not until the second trial for the lower readers. The latter performance matched that of second graders in Experiment 1 where the advantage became evident on Trial 2 as well (see Fig. 10.2). These findings support our expectation that spellings would help to establish pronunciations in memory earlier during the course of learning.
10 Spellings of Words: A Neglected Facilitator of Vocabulary Learning
147
Spoken Words Recalled (max. 10) 10
Number of Spoken Words
9 8 higher level spellings present
7 6
lower level spellings present
5
higher level spellings absent
4
lower level spellings absent
3 2 1 0
1
2
3 Trials
4
5
Fig. 10.4 Mean number of spoken words recalled over trials by higher level and lower level readers as a function of whether spellings were present or absent during learning in Experiment 2
The size of the advantage of spellings was much larger for higher than for lower readers, and remarkably it grew larger and larger over the first three trials for the higher readers. After that, ceiling effects suppressed further detection of this pattern. In contrast, the size of the advantage over trials was more similar for lower readers after the first trial. One reason for the difference might be that the higher readers had better knowledge not only of grapho-phonemic units but also of larger syllabic spelling units than lower readers, and this gave the higher readers an advantage in forming connections to store multisyllabic words in memory. These findings suggest a Matthew effect of the rich getting richer (Stanovich, 1986). Specifically, those with rich orthographic knowledge acquire richer and richer vocabularies over time compared to those with poorer orthographic knowledge. Figure 10.5 shows the mean performance of higher and lower readers recalling definitions of the words over trials. It is apparent that higher readers outperformed lower readers. Spellings benefited recall for both reader groups during the first three trials, but after that, performance was close to ceiling. Statistical tests confirmed these differences. These findings show that fifth graders learned vocabulary words better when they saw spellings of the words than when they only spoke the words. Comparison of performance in Figs. 10.4 and 10.5 shows that fifth graders remembered definitions more easily than pronunciations of the words. This was true for second graders as well (see Fig. 10.2). This shows that spellings helped readers with the harder part of vocabulary learning, remembering pronunciations of words. Why meanings were especially easy to remember in this study may be explained by several factors known to facilitate vocabulary learning (Sadoski,
148
L.C. Ehri and J. Rosenthal
Definitions Recalled (max. 10)
Number of definitions
10
9
higher level spellings present lower level spellings present
8
higher level spellings absent
7
lower level spellings absent
6
5 1
2
3 Trials
4
5
Fig. 10.5 Mean number of definitions recalled over trials by higher level and lower level readers as a function of whether spellings were present or absent during learning in Experiment 2
2005). The words were all concrete nouns. Use of pictures enabled the formation of visual images in memory. Embedding words in multiple defining and clarifying sentences helped connect them to other known words and concepts. Three posttests were given 1 day after students learned the words. They were shown training pictures and recalled pronunciations of the words. The words were dictated and students wrote their spellings. They inserted the words into cloze sentences that clarified meanings but were different from training sentences. Mean performance by higher and lower readers in the spelling present and spelling absent conditions is displayed in Fig. 10.6. In all cases, performance when spellings had been seen during learning was better than performance without spellings. Thus, effects of the spelling treatment persisted, at least 1 day later. Students were able to spell many more of the words they had seen than not seen, indicating they had stored the spellings in memory. The advantage of seeing spellings was apparent for both higher readers and lower readers. However, higher readers showed larger benefits than lower readers (i.e., compare the difference in heights of the paired columns in Fig. 10.6). Nevertheless, even though lower readers had weaker orthographic knowledge, they still had enough knowledge to benefit from spellings in learning vocabulary words. All results were statistically significant. Children’s comments were noted as they attempted to recall pronunciations of words on the posttest. Some children named letters before they spoke the words. One child, after mispronouncing a word and then seeing its spelling, exclaimed, “Oh, I misspelled it!” Another child, when trying to recall hicatee (a kind of turtle), sighed, “I know there are two E’s at the end.” These observations bolster the claim that spellings themselves were stored and accessed to enhance memory for pronunciations.
10 Spellings of Words: A Neglected Facilitator of Vocabulary Learning
Sp
149
Mean Percent Correct on Posttests
90%
Sp
80%
Sp
70%
Sp
60%
Sp
50% 40%
Sp
30% 20% 10% 0%
Hi Rs
Lo Rs
Recall Pronunciations
Hi Rs
Lo Rs
Write Spellings
Hi Rs
Lo Rs
Fill Cloze Sentences
Fig. 10.6 Mean percentage correct in recalling pronunciations, writing spellings, and filling vocabulary words in cloze sentences on the delayed posttests when spellings had accompanied vocabulary learning (in black) and when spellings had not accompanied learning (in white) in Experiment 2
Findings of this study confirmed our hypotheses and supported the following conclusions. Fifth graders learned the pronunciations and meanings of new vocabulary words better when they were exposed to their spellings than when they only spoke the words. Students with stronger orthographic knowledge benefited more from seeing spellings than students with weaker orthographic knowledge. Matthew effects were apparent, suggesting that the gap in vocabulary size distinguishing those with richer from those with poorer orthographic knowledge may grow increasingly wide over time. Our explanation for findings is that when students are exposed to the spellings of new vocabulary words, grapho-phonemic connections are activated. This better secures pronunciations of words in memory. One reason is that it may be hard to distinguish all the phonemes in spoken words. If so, letters in spellings help to clarify their identities. Another reason is that spellings themselves become bonded to pronunciations in memory and secure pronunciations earlier during the course of learning. Better secured pronunciations provide a stronger base for learning meanings. The fact that the effect of spellings on memory was incidental suggests that grapho-phonemic mapping relations are activated automatically to secure new words in memory. Spellings helped both second and fifth graders, indicating that the effect of spellings is not limited developmentally to the period of beginning reading or to more advanced levels but extends over all levels of reading, at least during the elementary grades.
150
L.C. Ehri and J. Rosenthal
Implications for Vocabulary Learning and Instruction We can suggest several implications of our findings. Teachers need to become aware of the importance of spellings for vocabulary learning so they do not slight them in their teaching. When teachers encounter, pronounce, and explain new vocabulary words to their students, they should take time to display the spellings of the words, for example, when they are reading a story aloud to the whole class. Our findings are important for experts who advise teachers about how to strengthen vocabulary instruction. They need to recognize the contribution that exposure to spellings can make and to include this step as one of their recommendations along with an explanation of its contribution. Findings carry implications for students. Independent reading is thought to be the primary way that students beyond third grade acquire new vocabulary words (Chall, Jacobs, & Baldwin, 1990). This type of reading is typically conducted silently. It is not uncommon for students reading independently to skip words they do not know and fill in the meaning by guessing. However, this guessing strategy does little for vocabulary learning. Our findings suggest that a better strategy is for students to stop and say the words aloud as they figure out their meanings. Spellings of words are especially valuable for students who are learning a second language. Spellings help to clarify phonemes in pronunciations when students listen to spoken words while inspecting their spellings, thereby securing more precise representations in memory. Also spellings clarify differences between similar sounding words. Hatch and Brown (1995) give as an example a Spanish speaker who defined happened as felices, meaning happy. They suggest that such confusions can be combated by helping second language learners retain the spellings of words in memory. Many second language learners report heavy reliance on orthography for building their vocabularies and their competence with spoken language. In a study of high school students learning a foreign language, Sparks and Artzer (1997) found that word decoding skill was the best predictor of year-end foreign language oral proficiency, better even than students’ grades in their foreign language class the previous year. The authors interpreted their findings to show the utility of print for building representations of spoken words in memory. Our findings carry implications for reading instruction by underscoring the importance of developing students’ orthographic knowledge. Experiment 2 showed that fifth graders who possessed stronger word reading, decoding, and spelling skills benefited much more from spellings in learning new vocabulary words than students with poorer orthographic skills. In Experiment 1, the correlation between second graders’ word reading skill and their memory for pronunciations of vocabulary words learned with spellings was very high, r = .67, p < .01. The way to maximize chances that students will acquire full knowledge of the writing system is to provide systematic phonics, word analysis, and spelling instruction during the elementary grades (Ehri, Nunes, Stahl, & Willows, 2001). Beginning-level instruction should help children master the three skills needed to establish sight words in memory,
10 Spellings of Words: A Neglected Facilitator of Vocabulary Learning
151
specifically, phonemic segmentation, grapheme–phoneme relations, and graphophonemic mapping/decoding skill. Our findings carry implications for vocabulary researchers as well. Current theories about the storage of vocabulary words in memory need to be revised to take account of orthography. Gathercole (2006) has proposed a theory to explain how phonological representations of words are established in memory. A key process is phonological short-term memory which maintains a new word in working memory until a more permanent phonological representation is stored. Phonological short-term memory is assessed with a nonword repetition task. Good readers have been shown to have superior phonological working memories for words compared to poor readers, and this difference is thought to explain why good readers have larger vocabularies. However, the theory focuses exclusively on phonological processes and pays no attention to orthography. This is clearly an oversight as our findings show that orthography impacts memory for the phonological representation of new vocabulary words. Evidence bearing on this issue can be drawn from Fig. 10.4 showing that the difference favoring good over poor readers has more to do with superior orthographic knowledge rather than superior phonological memory. If we compare higher and lower readers, we see that they differed very little in remembering the pronunciations of words without spellings. This indicates that their phonological memories were not that different. However, if we compare higher and lower readers on memory for pronunciations when spellings had been stored in memory, we see a huge difference. This indicates that orthographic memory explains more of the variance than phonological memory in vocabulary learning. Interestingly, Gathercole and Baddeley (1989) provided evidence consistent with this possibility. They found that the effect of short-term working memory on vocabulary learning is substantial in children below 6 years of age but is diminished in children above age 6. Their explanation is that children above 6 have accumulated more spoken words in memory and have practiced repeating more words than children under 6 (Gathercole, 2006). However, our findings suggest another explanation that diminished effects result from children above 6 acquiring literacy skills, specifically, orthographic knowledge that they use to link the spellings of words to their phonological forms in memory. This issue awaits further research. In sum, our view is that the field of vocabulary learning and instruction has neglected an important skill that facilitates vocabulary growth once children learn to read and spell words. Learning the spellings of words provides a big boost in learning their pronunciations and meanings. Teachers, students, educators, and researchers need to become aware of these findings and incorporate them into their practice.
References Beck, I., & McKeown, M. (2002). Bringing words to life: robust vocabulary instruction. New York: Guilford. Blachowicz, C., & Fisher, P. (2004). Vocabulary lessons. Educational Leadership, 61, 66–69.
152
L.C. Ehri and J. Rosenthal
Chall, J., Jacobs, V., & Baldwin, L. (1990). The reading crisis. Cambridge, MA: Harvard University Press. Cutting, L., & Scarborough, H. (2006). Prediction of reading comprehension: relative contributions of word recognition, language proficiency, and other cognitive skills can depend on how comprehension is measured. Scientific Studies of Reading, 10, 277–299. Ehri, L. (1980). The development of orthographic images. In U. Frith (Ed.), Cognitive processes in spelling (pp. 311–338). London, UK: Academic. Ehri, L. (1984). How orthography alters spoken language competencies in children learning to read and spell. In J. Downing & R. Valtin (Eds.), Language awareness and learning to read (pp. 119–147). New York: Springer. Ehri, L. (1992). Reconceptualizing the development of sight word reading and its relationship to recoding. In P. Gough, L. Ehri & R. Treiman (Eds.), Reading acquisition (pp. 107–143). Hillsdale, NJ: Erlbaum. Ehri, L. (1998). Grapheme-phoneme knowledge is essential for learning to read words in English. In J. Metsala & L. Ehri (Eds.), Word recognition in beginning literacy (pp. 3–40). Mahwah, NJ: Erlbaum. Ehri, L. (2005). Development of sight word reading: phases and findings. In M. Snowling & C. Hulme (Eds.), The science of reading, a handbook (pp. 135–154). UK: Blackwell. Ehri, L., Nunes, S., Stahl, S., & Willows, D. (2001). Systematic phonics instruction helps students learn to read: evidence from the National Reading Panel’s meta-analysis. Review of Educational Research, 71, 393–447. Ehri, L., & Wilce, L. (1979). The mnemonic value of orthography among beginning readers. Journal of Educational Psychology, 71, 26–40. Ehri, L., & Wilce, L. (1980). The influence of orthography on readers’ conceptualization of the phonemic structure of words. Applied Psycholinguistics, 1, 371–385. Ehri, L., & Wilce, L. (1986). The influence of spellings on speech: are alveolar flaps /d/ or /t/? In D. Yaden & S. Templeton (Eds.), Metalinguistic awareness and beginning literacy (pp. 101– 114). Portsmouth, NH: Heinemann. Gathercole, S. E. (2006). Nonword repetition and word learning: the nature of the relationship. Applied Psycholinguistics, 27, 513–543. Gathercole, S., & Baddeley, A. (1989). Evaluation of the role of phonological STM in the development of vocabulary in children: a longitudinal study. Journal of Memory and Language, 28, 200–213. Hatch, E., & Brown, C. (1995). Vocabulary, semantics, and language education. New York: Cambridge University Press. Nagy, W., & Scott, J. (2000). Vocabulary processes. In M. Kamil, P. Mosenthal, P. Pearson, & R. Barr (Eds.), Handbook of reading research: Volume III (pp. 269–284). Mahwah, NJ: Erlbaum. Reitsma, P. (1983). Printed word learning in beginning readers. Journal of Experimental Child Psychology, 75, 321–339. Rosenthal, J., & Ehri, L. (2008). The mnemonic value of orthography for vocabulary learning. Journal of Educational Psychology, 100, 175–191. Sadoski, M. (2005). A dual coding view of vocabulary learning. Reading and Writing Quarterly, 21, 221–238. Seidenberg, M., & Tanenhaus, M. (1979). Orthographic effects on rhyme monitoring. Journal of Experimental Psychology: Human Learning and Memory, 5, 546–554. Share, D. (2004). Orthographic learning at a glance: on the time course and developmental onset of self-teaching. Journal of Experimental Child Psychology, 87, 267–298. Sparks, R., & Artzer, M. (1997). Prediction of foreign language proficiency. Journal of Educational Psychology, 89, 549–562. Stanovich, K. (1986). Matthew effects in reading: some consequences of individual differences in the acquisition of literacy. Reading Research Quarterly, 21, 360–407. Stuart, M., & Coltheart, M. (1988). Does reading develop in a sequence of stages? Cognition, 30, 139–181.
Part 3
Bilingualism – Acquiring a Second Spoken and Written Language
Chapter 11
Early Schooling and Growth of Literacy in the Transition to School Frederick J. Morrison, Carol MacDonald Connor, and Annemarie Hindman
Abstract The chapter describes an ongoing project aimed at understanding the impact of instruction on children’s early literacy growth. Derived from direct observations of classroom instruction, findings revealed that the most effective instruction depends on the initial skill levels of the child. These child X instruction interactions imply that efforts to improve children’s literacy skills will be greatly enhanced by a focus on tailoring instruction to each child’s profile of strengths and weaknesses. An intervention is described that attempts to provide such indivi dualized instruction combining a small-group focus with computer support that specifies the amounts and types of instruction each child needs to advance. Findings from the intervention demonstrated greater reading growth for the individualized instruction group over a control group. Further, the more the instruction matched recommended amounts and types for each child, the more children grew. Theoretical and practical implications are discussed. In the ongoing effort to understand and improve the literacy skills of American children, several important insights have begun to focus on and shape theoretical and empirical work. First, it is becoming increasingly evident that meaningful individual differences in important language, cognition, literacy, and social skills emerge before children begin formal schooling in kindergarten or first grade (Morrison, Bachman, & Connor, 2005; Shonkoff & Phillips, 2000). Second, this early variability is influenced by a number of interacting factors in the child, the family, the preschool, and the larger sociocultural context (ECCRN, 2004).Third, longitudinal work has revealed the lasting effects of early experiences on reading acquisition, performance grades, and drop-out rates (Entwisle, Alexander, & Olson, 2005; Juel & MindenCupp, 2000). Finally, recent evidence reveals that the early elementary schooling experiences of children are highly variable, in some cases exacerbating the individual
F.J. Morrison (*), C.M. Connor, and A. Hindman University of Michigan, Ann Arbor, MI, USA e-mail:
[email protected] D. Aram and O. Korat (eds.), Literacy Development and Enhancement Across Orthographies and Cultures, Literacy Studies 101, DOI 10.1007/978-1-4419-0834-6_11, © Springer Science+Business Media, LLC 2010
153
154
F.J. Morrison et al.
differences that emerge among children prior to school entry (Pianta, La Paro, Payne, Cox, & Bradley, 2002). The cumulative impact of these ongoing trends has directed attention to school transition (roughly the period from 3 years of age until third grade) as a unique and important milestone in the academic development of children and as a foundational experience for early school success. In this chapter, we will first review literature demonstrating the impact of early experiences (including in preschool) on the cognitive, literacy, and social development in children. Next, we focus on the nature of the effects of early schooling on literacy growth, emphasizing that the most effective instruction must take into account the child’s skill levels and other characteristics. Finally, we present the results of two series of experiments, one demonstrating the importance of child-by-instruction interactions in literacy growth, the other presenting findings on an intervention aimed directly at individualizing instruction based on children’s assessed level of functioning.
The Influence of Preschool Experience on Emergent Literacy Accumulating evidence reveals that high-quality preschool experiences lead to stronger academic outcomes, including literacy skills, especially for children at risk for academic underachievement (Graue, Clements, Reynolds, & Niles, 2004; Nelson, Benner, & Gonzalez, 2003; Reynolds, Ou, & Topitzes, 2004). However, there is substantial variability in classroom practices and the resulting child outcomes across programs and studies (Barnett, 1995; Nelson et al., 2003), raising questions about how specific types of preschool experiences relate to literacy outcomes. Mounting evidence confirms that teachers’ sensitivity and responsiveness (NICHD-ECCRN, 2002), their use of cognitively challenging talk and rare words (e.g., Dickinson & Tabors, 2001), and their styles of book reading including Dialogic Reading (Whitehurst, Arnold, et al., 1994), as well as child-initiated practices (Schweinhart & Weikart, 1988), contribute to positive literacy outcomes (NICHD-ECCRN, 2002). Until recently, there was not enough evidence to prove that a primary focus on academic skills provides consistently stronger child outcomes (Graue et al., 2004; Stipek, Feiler, Daniels, & Milburn, 1995).Yet, understanding the content of specific preschool literacy activities and the amount of time spent in such activities is crucial, especially with the inauguration of new universal pre-kindergarten initiatives and the recent findings regarding limited Head Start efficacy (Pianta, 2007; Stipek, 2006).
Short- and Long-Term Benefits of Preschool Intervention Preschool intervention research has revealed short- and long-term social and cognitive gains for students at risk for academic underachievement as a result of poverty, disabilities, and other factors. Documented advantages include reduced referral to
11 Early Schooling and Growth of Literacy in the Transition to School
155
special education, reduced grade retention, higher rates of high school graduation, and reduced levels of juvenile delinquency (Barnett, 1995; Barnett, Young, & Schweinhart, 1998; Dickinson & Smith, 1994; Durlak, 2003; Lazar, Darlington, Murray, Rocye, & Snipper, 1982; see also Nelson et al., 2003). Perhaps the most compelling long-term positive outcomes have resulted from research on model programs such as the High Scope/Perry Preschool project, the Abecedarian project, and the Chicago Child-Parent Centers (or Chicago Title 1 Project). For example, the Chicago Child-Parent Centers calculated that for every dollar spent on the preschool intervention, society realized a $7.14 return in saved education and societal costs from reduced referral to special education, reduced grade retention, and reduced juvenile delinquency (Reynolds et al., 2004). Still, in light of society’s substantial investment in public preschool programs such as Head Start, Early Reading First, Title 1, and universal preschool, “… greater attention to how the long-term effects come about is needed to inform program improvement and expansion efforts” (Reynolds, Temple, & Ou, 2003, p. 1301). Studies have clearly revealed that cognitively challenging talk (Dickinson & Tabors, 2001; Smith & Dickinson, 1994); exposure to rare words (Beals & Tabors, 1995; Huttenlocher, Vasilyeva, Cymerman, & Levine, 2002); shared book reading using dialogic reading techniques (Cook-Gumperz, 1986; Crain-Thoreson & Dale, 1999; Dickinson & Tabors, 2001; Lonigan & Whitehurst, 1998; Whitehurst, Arnold, et al., 1994), play (Pelligrini & Galda, 1993), and use of playful activities to stimulate learning (Bowman, Donovan, & Burns, 2001; Howes & Smith, 1995), all enhanced children’s language development, later reading skills and overall academic success.
Teacher-Directed Versus Child-Centered Instruction in Preschool: A False Dichotomy In the preschool literature a controversy has existed about the most effective method of instruction for children’s early language and emergent literacy skills. The debate centers on the appropriateness and effectiveness for pre-kindergartners of direct, explicit instruction of academic skills, usually provided by the teacher, versus an approach that emphasizes discovery learning, especially through play-like activities, that are more under the control of the child. Those espousing the latter view tend to view preschool children as naturally curious and active learners for whom explicit instruction is “developmentally inappropriate” and potentially emotionally harmful. Those recommending some explicit teaching of basic skills point to the gaping chasm in skill levels among American children that opens up during the preschool period and grows in magnitude over the elementary school years as necessitating more directed early intervention to close the gap (Morrison et al., 2005). Research on this controversy has not been entirely consistent, but recent evidence suggests that a strict dichotomy between explicit, teacher-directed instruction and child-centered approaches may be false, akin to the phonics versus whole-language debate in early reading (Connor, Morrison, & Katch, 2004; Stipek, 2006). These studies
156
F.J. Morrison et al.
(Stipek et al., 1995) tended to pit one kind of curriculum against the other, which may oversimplify the complex nature of classroom experiences and their effects on students’ outcomes. New evidence, using teacher report, suggests that curricula high in teacher-directed instruction with specific content and simultaneously high in child initiated/teacher responsive approaches yielded more positive short- and long-term educational and social outcomes for preschoolers than did programs that emphasized one over the other or neither (Graue et al., 2004). In other words, using a more balanced approach yielded stronger student growth (Pianta, 2007).
From Balanced to Individualized Instruction Yet promoting balanced practices raises the distinct possibility that the most effective balance may be different for different children. Mounting evidence, including our own, demonstrates that the efficacy of instructional practices depends on the skill level of the student. For example, in a recent study (Connor, Morrison, Fishman, Schatshneider, & Underwood, 2007) preschool children with below-average alphabet and letter–word identification skills at the beginning of the year showed more growth when they spent greater amounts of time in code-focused activities, whereas the opposite was true for children entering with higher skills, for whom more time in meaning-focused activities yielded the strongest growth. Similar patterns have been uncovered across the school transition period (Foorman, Francis, Fletcher, Schatschneider, & Mehta, 1998; Juel & Minden-Cupp, 2000).
Dimensions of Instruction One challenge facing researchers is to develop conceptualizations of the complexities of preschoolers’ classroom experiences while also permitting statistical analysis of the effects of these experiences on child outcomes. Much of preschool research has compared the relative effectiveness of different curriculum types based on different theories of learning. However, during any given school day, teachers may provide (or students may experience) a variety of activities that incorporate several major curriculum types. The teacher may use direct instruction for teaching letters, a cognitive curriculum approach during sharing time, and more traditional approaches as (s)he sets up the dramatic play center. Comparing one curriculum with another may overlook the complexity of learning activities preschoolers actually experience. In one study (Connor, Morrison, & Slominski, 2006), we organized the content of classroom literacy activities along three central dimensions: (1) teacher- versus teacher–childversus peer- versus child-managed, (2) code-focused versus meaning-focused, (3) student- versus classroom-level instruction. The dimension teacher-, teacher–childversus child-managed considers who is focusing the child’s attention – the teacher or the student or whether the attention is jointly focused (teacher and students interacting). A teacher reading a book to students without discussion, even if the children
11 Early Schooling and Growth of Literacy in the Transition to School
157
selected the book, would be considered teacher-managed (TM) because the teacher is focusing the children’s attention and is doing most of the talking. Sharing, scaffolding, or interactive read-alouds are considered to be teacher–child-managed (TCM) within this framework because the teacher is actively involved with and responsive to the child. In contrast, activities where the child is working independently or with peers, such as completing worksheets, “pretend” reading alone or with a friend in the library corner, are considered child-managed (CM) since the student is directing his or her own attention without the support of the teacher. The code- versus meaningfocused dimension captures the content focus of language and emergent literacy activities. While clearly, preschoolers are not taught to read as such, there is evidence that appropriate preschool activities can enhance their emergent literacy – including letter knowledge and phonological awareness. These kinds of activities may be considered emergent code-focused activities. These would include teaching children how to name and write letters, to rhyme words, to relate letters to the sounds they make, and to sound-out words (phonological decoding). In contrast, activities designed to help students understand words and passages, comprehend what is read to them or what they are reading, and enhance receptive and expressive language skills including listening comprehension, which support emerging reading comprehension may be considered meaning-focused activities. Such activities include explaining the meaning of a word (i.e., vocabulary activities), reading aloud to children, discussion and sharing, and children’s emergent reading and writing activities. The dimension classroom- versus student-level considers the extent to which instruction is the same or different for each student in the classroom. Language and literacy activities may be classroom level, such as when the teacher is reading aloud during circle time. All of the students are doing substantially the same thing at the same time. Even if the students are working in small groups or individually, if they are all doing substantially the same thing (e.g., play), then that is classroom-level instruction. In contrast, for student-level instruction, students in one classroom are engaged in substantially different activities at the same time. Teachers may provide student-level instruction in small groups or they may work with children individually (e.g., centers with different activities, tutoring one child while the rest do other activities).
Child Factors It has become increasingly clear that factors in the child play independent and interactive roles in shaping literacy trajectories over the school transition period. Clearly the level of a child’s skill at any point in time reflects the impact of previous experiences in and out of the classroom context. But at each time point, other things being equal, the child’s skill level will predict their achievement and learning. Hence, knowing the concurrent level of a child’s functioning facilitates understanding of their performance as well as the experiences that will maximize their growth. Accumulating evidence reveals that a number of literacy skills in preschool and kindergarten consistently predict children’s later reading and
158
F.J. Morrison et al.
a cademic success, including alphabet knowledge, phonological awareness, letter–word recognition, and phonological decoding (NICHD-ECCRN, 2005; Rayner, Foorman, Perfetti, Pesetsky, & Seidenberg, 2001; Snow, Burns, & Griffin, 1998). Language skills, particularly vocabulary, are also consistent predictors of later reading success, especially as comprehending what is read becomes important Scarborough, 1998, 2001; Snow et al., 1998; Storch & Whitehurst, 2002).Rich, meaning-focused literacy experiences at home (Beals & DeTemple, 1993; Senechal, LeFevre, Thomas, & Daley, 1998), during storybook reading (Robbins & Ehri, 1994) and notably, in preschool (Dickinson, Anastasopolous, McCabe, Peisner-Feinberg, & Poe, 2003; Dickinson & Tabors, 2001) can contribute to young children’s language growth.
Child by Instruction Interactions In recent work, we have examined child by instruction interactions in first and second grade (Connor, Morrison, & Underwood, 2007), third grade (Connor, Morrison & Petrella, 2004), and most recently in preschool (Connor et al., 2006). In the latter study, 156 typically developing 3- and 4-year olds were assessed in fall and spring of the school year on alphabet and letter–word decoding and expressive vocabulary, while classrooms were videotaped mid-year for half a day. Videotaped classroom observations revealed substantial variability in amount and types of language and emergent literacy activities, across classrooms and for individual children within classrooms. Hierarchical linear modeling revealed that, generally, more time in emergent code-focused activities (e.g., playing letter and rhyming games) was associated with preschoolers’ alphabet and letter–word recognition growth while more time in meaning-focused activities (e.g., storybook-reading) was related to vocabulary growth. Only teacher- and teacher–child-managed activities, including structured whole- and small-group work were associated with alphabet and letter–word growth whereas child-managed meaning-focused experiences, including play, were associated with vocabulary growth. Overall, the effect size for child-level code-focused instruction (small-group) was about ten times greater than was its classroom-level (whole-class) counterpart. Most noteworthy, several child by instruction interactions revealed that the impact of different activities varied with preschoolers’ incoming vocabulary and emergent literacy skills. For example, whereas more time spent in teacher–child-managed code-focused activities yielded generally greater growth in alphabet and letter–word recognition, this effect was much stronger for students with relatively low (<25th percentile) initial alphabet and letter–word decoding scores. In contrast, teacher–child meaning-focused activities benefited exclusively children with relatively high initial scores. Moreover, one striking finding was that more minutes of play significantly enhanced vocabulary growth, but primarily for children with comparatively low starting vocabularies. The pattern of findings from this study sheds light on the ongoing controversy about the most effective method of enhancing preschoolers’ emergent literacy development (Bowman et al., 2001; Graue et al., 2004; Stipek et al., 1995). Our results,
11 Early Schooling and Growth of Literacy in the Transition to School
159
along with other preschool research (Graue et al., 2004; Whitehurst, Epstein, et al., 1994), support the idea that explicitly teaching letters, letter sounds, phonological decoding, and phonological awareness, in conjunction with rich meaning-focused experiences, including play, may yield stronger outcomes than programs that focus solely on one to the exclusion of the other. Explicit-code-focused activities, where teachers and children work together, may be especially important for preschoolers with weaker literacy skills at the beginning of the year. Based on our observations of the classrooms in this study, these explicit-code-focused activities were consistently delivered in interactive ways. No drilling of skills or time spent completing worksheets was observed. Rather, alphabet and letter-sound activities, for example, were embedded in songs, craft activities, and games. Opportunities to practice these skills were also provided in dramatic play centers (e.g., reading menus or writing a prescription when pretending to be a doctor), although our results suggest that, relative to emergent reading development, the greater potential benefit to children provided through play is in vocabulary development. Explicit-meaning-focused activities contributed significantly to emergent decoding and vocabulary growth as well. Thus reading to children, sharing, discussion, and other meaning-focused activities found in high-quality preschool programs should not be abandoned in favor of an exclusive focus on code-related activities. Our results reveal that the impact of selected patterns of instruction will vary significantly depending on preschoolers’ entering vocabulary and emergent reading skills.
An Intervention to Individualize Instruction in First Grade These and related findings (Foorman et al, 1998; Juel & Minden-Cupp, 2000) imply that the most effective pattern of instruction (amount, type, and change over time) from preschool to third grade depends fundamentally on initial language and literacy skill levels of the child. Thus, educational efforts to individualize (or personalize or differentiate) instruction for each child could prove highly effective. Toward that end we are currently engaged in an intervention study (funded jointly by IES and NICHD, Carol Connor, PI) aimed at assessing the impact (via randomized control trial design) of individualizing instruction on first graders’ early reading growth. There are four key elements of the design: 1. Assessment. At the beginning of the school year, children’s vocabulary and letter–word reading skills are assessed via standardized tests (Woodcock–Johnson-III Tests of Achievement; (Mather & Woodcock, 2001). They are also assessed in the winter and spring of the year. 2. Assessment-to-Instruction software (A2i). Based on the child’s initial score, a computer program created for this project utilizes algorithms from our research to specify for each child the amounts and types of each kind of instruction he or she should receive on a daily or weekly basis (i.e., teacher-managed codefocused, teacher-managed meaning-focused, child-managed code-focused and child-managed meaning-focused) in order to reach grade level if starting below or attain 1 year’s growth if starting above grade level.
160
F.J. Morrison et al.
3. Small groups. The A2i software also recommends small group assignments for children based on their initial score profiles. Teachers may modify the groupings based on other factors (e.g., student compatibility) and can also organize their daily lessons via the software. Finally, they select children’s daily activities by clicking a list of activities from a lesson planner that organizes activities by amount and type of instruction, which can be indexed to any core curriculum. 4. Professional development. Teachers received two full day development sessions focusing on the background and rationale for the study, gaining experience with the software and individualizing instruction via small groups. In addition, coaches (experienced classroom teachers) visited each experimental teacher on a regular basis to help them use A2i and to mentor them in implementing individualized instruction. Results revealed four major findings to date (Connor et al., 2007; Connor et al., 2009). First, children in the intervention group showed greater growth in word reading and reading comprehension skills (analyzed using HLM, controlling for fall scores) than did children in the control group. In addition, greater fidelity of implementation, assessed via observer rating scale, yielded stronger growth. Second, the more time teachers spent logged onto A2i, the stronger their students’ reading growth and this effect was greater for students with weaker fall vocabulary skills (see Fig. 11.1). Third, the closer the instruction received by the child matched the software recommended amounts and types, the more children’s reading skill grew over the year (see Fig. 11.2). Take together these finding reveal that child by instruction interactions are most likely causally related to improvements, or lack thereof, in students’ language and literacy skill development. If so, this suggests that there is no one “best” way to teach all children how to read proficiently. Also
Spring Comprehension W Score
472 470
Fall vocabulary about 1 year above age expectations
468 466
Fall vocabulary about 1 year below age expectations
464 462 460 Control
90 min
180 min
270 min
360 min
Amount of time using A2i
Fig. 11.1 The effect of increasing using of Assessment to Instruction (A2i) software from fall to spring on students’ WJ passage comprehension
11 Early Schooling and Growth of Literacy in the Transition to School
161
120 115
Standard Score
110 105 100 95 90 85
10
20
30
DFR (minutes) TM-CF predicting PC
CM-MF predicting PC
Fig. 11.2 The effect of distance from recommended (DFR), which is the absolute value of the observed amount minus the A2i recommended amount of instruction, on students WJ passage comprehension standard scores (mean = 100, SD = 15)
important is using assessment results to guide and evaluate the efficacy of instruction. Additionally, the teachers that reached higher levels of fidelity and, hence, whose students showed greater literacy skill growth, used an evidence-based core reading curriculum but provided instruction in flexible homogeneous skill-based groups that changed in response to children improving literacy skills. Finally, however, we observed substantial variability among teachers in the fidelity of implementation of individualizing instruction. The reasons for this variability appear to be related to teachers’ basic classroom organization skills, familiarity with technology, their responsiveness to the professional development provided, and their willingness or ability to change their literacy practice. Overall, the findings to date are very encouraging regarding the efficacy of individualizing instruction but also reveal that full implementation is challenging for some teachers.
Conclusions and Implications Research over the past three decades had firmly established the pivotal role school transition plays in shaping children’s academic trajectories. The unique contribution of early schooling to successful transition is gradually becoming clearer, as we learn more about the nature of the instructional experiences children receive and how those interact with the incoming skills children bring to
162
F.J. Morrison et al.
the classroom. Our observational work has emphasized the importance of looking at child by instruction interactions for understanding how and why children do or do not progress academically from one grade to the next. Our intervention work has demonstrated the power of individualizing instruction for improving children’s literacy growth. A number of implications flow from these conclusions. First, the notion of what constitutes quality instruction will need re-examination. Clearly, from our findings, what is considered high-quality instruction for one child could be low quality for another. Second, providing effective instruction for each child clearly requires early and ongoing assessment of the strengths and weaknesses in a child’s profile so that appropriate instruction can be implemented. Third, a focus on individualizing emphasizes the learning of all children and not just closing the gap between groups of children. Finally, in contrast to present practices, teachers should know the skills and prior experiences of the students who are entering their classes in the fall so that effective instruction can be programmed from the start and the continuity of children’s development is finally taken seriously. Striving to individualize instruction may be a major challenge for upcoming generations of teachers, but the effort could reap enormous benefits in improving the literacy skills of all our children.
References Barnett, S. W. (1995). Long-term effects of early childhood programs on cognitive and school outcomes. Future of Children, 5(3), 25–50. Barnett, W. S., Young, J. W., & Schweinhart, L. J. (1998). How preschool education influences long-term cognitive development and school success. In W. S. Barnett & S. S. Boocock (Eds.), Early care and education for children in poverty: promises, programs, and long-term results. Albany: State University of New York Press. Beals, D. E., & DeTemple, J. M. (1993). Home contributions to early language and literacy development. Yearbook of the National Reading Conference, 42, 207–215. Beals, D. E., & Tabors, P. O. (1995). Arboretum, bureaucratic and carbohydrates: preschoolers’ exposure to rare vocabulary at home. First Language, 15, 57–76. Bowman, B. T., Donovan, S., & Burns, M. S. (2001). Eager to learn: educating our preschoolers. Washington DC: National Academy Press. Connor, C. M., Morrison, F. J., Fishman, B. J., Schatschneider, C., & Underwood, P. (2007). THE EARLY YEARS: algorithm-guided individualized reading instruction. Science, 315(5811), 464–465. Connor, C. M., Morrison, F. J., & Katch, L. E. (2004). Beyond the reading wars: exploring the effect of child-instruction interactions on growth in early reading. Scientific Studies of Reading, 8, 305–336. Connor, C. M., Morrison, F. J., & Petrella, J. N. (2004). Effective reading comprehension instruction: examining child instruction interactions. Journal of Educational Psychology, 96, 682–698. Connor, C. M., Morrison, F. J., & Slominski, L. (2006). Preschool instruction and children’s literacy skill growth. Journal of Educational Psychology, 98(4), 665–689. Connor, C. M., Morrison, F. J., & Underwood, P. S. (2007). A second chance in second grade: the independent and cumulative impact of first- and second-grade reading instruction and students’ letter-word reading skill growth. Scientific Studies of Reading, 11(3), 199–233.
11 Early Schooling and Growth of Literacy in the Transition to School
163
Connor, C. M., Piasta, S. B., Glasney, S., Schatschneider, C., Fishman, B., Underwood, P., et al. (2009). Individualizing student instruction precisely: effects of child by instruction interactions on first graders’ literacy development. Child Development, 80, 77–100. Cook-Gumperz, J. (1986). Caught in a web of words: some considerations on language socialization and language acquisition. In J. Cook-Gumperz, W. A. Cosaro & J. Streeck (Eds.), Children’s worlds and children’s language (pp. 37–68). New York: New Babylon. Crain-Thoreson, C., & Dale, P. S. (1999). Enhancing linguistic performance: parents and teachers as book reading partners for children with language delays. Topics in Early Childhood Special Education, 19(1), 28–39. Dickinson, D. K., Anastasopolous, L., McCabe, A., Peisner-Feinberg, E. S., & Poe, M. D. (2003). The comprehensive language approach to early literacy: the interrelationships among vocabulary, phonological sensitivity, and print knowledge among preschool-aged children. Journal of Educational Psychology, 93(3), 465–481. Dickinson, D. K., & Smith, M. W. (1994). Long-term effects of preschool teachers’ book readings on low-income children’s vocabulary and story comprehension. Reading Research Quarterly, 29(2), 105–122. Dickinson, D. K., & Tabors, P. O. (2001). Beginning literacy with language. Baltimore: Paul H. Brookes. Durlak, J. A. (2003). The long-term impact of preschool prevention programs: a commentary. Prevention and Treatment, 6(32), 1–5. ECCRN. (2004). Multiple pathways to early literacy achievement. Harvard Educational Review, 74, 1–29. Entwisle, D. R., Alexander, K. L., & Olson, L. S. (2005). First grade and educational attainment by age 22: a new story. American Journal of Sociology, 110, 1458–1502. Foorman, B. R., Francis, D. J., Fletcher, J. M., Schatschneider, C., & Mehta, P. (1998). The role of instruction in learning to read: preventing reading failure in at-risk children. Journal of Educational Psychology, 90, 37–55. Graue, E., Clements, M. A., Reynolds, A. J., & Niles, M. D. (2004). More than teacher directed or child initiated: preschool curriculum type, parent involvement, and children’s outcomes in the Child-Parent Centers [Electronic Version]. Education Policy Analysis Archives, 12. Retrieved February 5, 2005 from http://epaa.asu.edu/epaa/v12n72/ Howes, C., & Smith, E. (1995). Relations among child care quality, teacher behavior, children’s play activities, emotional security, and cognitive activity in child care. Early Childhood Research Quarterly, 10(4), 381–404. Huttenlocher, J., Vasilyeva, M., Cymerman, E., & Levine, S. (2002). Language input and syntax. Cognitive Psychology, 45, 337–374. Juel, C., & Minden-Cupp, C. (2000). Learning to read words: linguistic units and instructional strategies. Reading Research Quarterly, 35, 458–492. Lazar, I., Darlington, R., Murray, H., Royce, J., & Snipper, A. (1982). Lasting effects of early education: a report from the consortium for longitudinal studies. Monographs of the Society for Research in Child Development, 47(2–3). Lonigan, C. J., & Whitehurst, G. J. (1998). Relative efficacy of parent and teacher involvement in a shared book-reading intervention for preschool children from low income backgrounds. Early Childhood Research Quarterly, 13(2), 263–290. Mather, N., & Woodcock, R. W. (2001). Woodcock Johnson III tests of achievement: Examiner’s manual. Itasca, IL: Riverside. Morrison, F. J., Bachman, H. J., & Connor, C. M. (2005). Improving literacy in America: guidelines from research. New Haven, CT: Yale University Press. Nelson, J. R., Benner, G. J., & Gonzalez, J. (2003). Learner characteristics that influence treatment effectiveness of early literacy interventions: a meta-analytic review. Learning Disabilities Research and Practice, 18(4), 255–267. NICHD-ECCRN. (2002). Child-care structure – process – outcome: direct and indirect effects of child-care quality on young children’s development. Psychological Science, 13(2), 199–206.
164
F.J. Morrison et al.
NICHD-ECCRN. (2005). Pathways to reading: the role of oral language in the transition to reading. Developmental Psychology, 41(2), 428–442. Pelligrini, A. D., & Galda, L. (1993). Ten years after: a reexamination of symbolic play and literate research. Reading Research Quarterly, 28(2), 162–175. Pianta, R. C. (2007). Editorial. Journal of School Psychology, 45(6), 587–588. Pianta, R. C., La Paro, K. M., Payne, C., Cox, M. J., & Bradley, R. H. (2002). The relation of kindergarten classroom environment to teacher, family, and school characteristics and child outcomes. Elementary School Journal, 102(3), 225–238. Rayner, K., Foorman, B. R., Perfetti, C. A., Pesetsky, D., & Seidenberg, M. S. (2001). How psychological science informs the teaching of reading. Psychological Science in the Public Interest, 2, 31–74. Reynolds, A. J., Ou, S.-R., & Topitzes, J. W. (2004). Paths of effects of early childhood intervention on educational attainment and delinquency: a confirmatory analysis of the Chicago child-parent centers. Child Development, 75(5), 1299–1328. Reynolds, A. J., Temple, J. A., & Ou, S. R. (2003). School-based early intervention and child well-being in the Chicago longitudinal study. Child Welfare, 82, 633–656. Robbins, C., & Ehri, L. C. (1994). Reading storybooks to kindergartners helps them learn new vocabulary words. Journal of Educational Psychology, 86, 54–64. Scarborough, H. S. (1998). Early identification of children at risk for reading disabilities: phonological awareness and some other promising predictors. In B. K. Shapiro, P. J. Accardo & A. J. Capute (Eds.), Specific reading disability: a view of the spectrum (pp. 75–119). Timonium MD: York Press, Inc. Scarborough, H. S. (2001). Connecting early language and literacy to later reading (dis)abilities: evidence, theory, and practice. In S. B. Neuman & D. K. Dickinson (Eds.), Handbook of early literacy research (pp. 97–110). New York: Guilford Press. Schweinhart, L. J., & Weikart, D. P. (1988). Education for young children living in poverty: childinitiated learning or teacher-directed instruction. Elementary School Journal, 89(2), 213–225. Senechal, M., LeFevre, J.-A., Thomas, E. M., & Daley, K. E. (1998). Differential effects of home literacy experiences on the development of oral and written language. Reading Research Quarterly, 33(1), 96–116. Shonkoff, J. P., & Phillips, D. A. (eds). (2000). From neurons to neighborhoods: the science of early childhood development. Washington, DC: National Academy Press. Smith, M. W., & Dickinson, D. K. (1994). Describing oral language opportunities and environments in Head Start and other preschool classrooms. Early Childhood Research Quarterly, 9(3–4), 345–366. Snow, C. E., Burns, M. S., & Griffin, P. (eds). (1998). Preventing reading difficulties in young children. Washington, DC: National Academy Press. Stipek, D. (2006). No child left behind comes to preschool. The Elementary School Journal, 106(5), 455–465. Stipek, D., Feiler, R., Daniels, D., & Milburn, S. (1995). Effects of different instructional approaches on young children’s achievement and motivation. Child Development, 66(1), 209–223. Storch, S. A., & Whitehurst, G. J. (2002). Oral language and code-related precursors to reading: evidence from a longitudinal structural model. Developmental Psychology, 38(6), 934–947. Whitehurst, G. J., Arnold, D. S., Epstein, J. N., Angell, A. L., Smith, M., & Fischel, J. E. (1994). A picture book reading intervention in day care and home for children from low-income families. Developmental Psychology, 30(5), 679–689. Whitehurst, G. J., Epstein, J. N., Angell, A. L., Payne, A. C., Crone, D. A., & Fischel, J. E. (1994). Outcomes of emergent literacy intervention in Head Start. Journal of Educational Psychology, 86, 542–555.
Chapter 12
Role of Orthography in Literacy Acquisition and Literacy Problems Among Monolinguals and Bilinguals R. Malatesha Joshi
Abstract There may be various reasons, such as the number of books available at home, genetics, type of instruction etc., as to why a child has difficulty learning to read and spell. Additionally, the type of writing system of a language can also affect the literacy acquisition. According to “Orthographic Depth Hypothesis (ODH),” the degree of correspondence between orthography and phonology is an important factor of the way literacy skills are acquired. Seymour (Theoretical framework for beginning reading in different orthographies. In R.M. Joshi & P.G. Aaron (Eds.), Handbook of orthography and literacy. Lawrence Erlbaum Associates) examined the speed and accuracy of familiar word reading and nonword reading in 15 different European writing systems and concluded that “the establishment of an effective sight vocabulary and decoding needs about 2 years of reading experience in English as against 1-year in many European languages.” The questions that arise, then, are how does the orthography of one language influence the acquisition of literacy skills in another language? If a person is dyslexic in one language, would he/she be dyslexic in other languages as well? Can reading models, such as the Componential Model, that have proven beneficial for the assessment of reading problems among English-speaking children be useful to other orthographies such as Spanish orthography? This chapter presents results from our studies that show that orthographic differences do make a contribution in the acquisition of literacy skills, and that certain orthographies can slow down literacy acquisition in beginning readers.
Introduction Failure to master the reading and spelling skills has serious implications both at the individual as well as the societal level, and hence, it has attracted the attention of researchers from diverse fields, such as cognitive psychology, neurosciences,
R.M. Joshi (*) College of Education and Human Development, Texas A & M University, College Station, TX, USA e-mail:
[email protected] D. Aram and O. Korat (eds.), Literacy Development and Enhancement Across Orthographies and Cultures, Literacy Studies 101, DOI 10.1007/978-1-4419-0834-6_12, © Springer Science+Business Media, LLC 2010
167
168
R.M. Joshi
linguistics, speech and hearing sciences, and education. Various reasons have been proposed for the reading problems: poor oral language development (Hart & Risley, 1995; Moats, 2001), the number of books available at home (Chiu & McBrideChang, 2006), genetics (Pennington & Olson, 2005), and poor classroom instruction, especially at the early primary grade levels (Foorman, Francis, Shaywitz, Shaywitz, & Fletcher, 1997; Torgesen, 2005; Vellutino, Scanlon, & Jaccard, 2003). In addition to these factors, the structure of the writing system of a language or the orthography of a language can also influence literacy acquisition. According to Henderson (1984), orthography is defined as the “graphemic patterns of a written language and their mapping onto phonology, morphology, and meaning” (p. 1). Seymour (2007) concluded, after comparing different European orthographies, that linguistic differences, especially aspects of syllable structure and variations in orthographic depth, may be crucial in determining the amount of learning necessary for successful progression through each phase. Further, Seymour, Aro and Erskine (2003) concluded, after examining the speed and accuracy of familiar word reading and nonword reading in eight writing systems (Finish, Spanish, Italian, Greek, Portuguese, French, Danish, and English), that the establishment of an effective sight vocabulary and decoding needs about 2 years of reading experience in English as against 1 year in many European languages. The potential influence of orthographies on reading is explained by “Orthographic Depth Hypothesis” (ODH), proffered by Frost, Katz and Bentin (1987), according to which, “... lexical word recognition in shallow orthographies is mediated primarily by phonemic cues generated prelexically by grapheme-to-phoneme translation. In contrast, lexical access for word recognition in a deep orthography relies strongly on orthographic cues, whereas phonology is derived from internal lexicon.” (p. 113). The later formulations of this hypothesis are less stringent in labeling the core processes “orthographic” or “phonological” according to the depth of the orthography, since phonological processing has also been shown to be involved in word recognition in deep orthographies (Lukatela & Turvey, 1999; p. 1060). According to the Orthographic Depth Hypotheses, the degree of correspondence between orthography and the phonology of the written word is one of the many factors that has an influence on the rate in which literacy skills are acquired. One of the influential models to explain reading difficulties in English language is the Componential Model of Reading (Aaron & Joshi, 1992; Joshi & Aaron, 2000). According to this model, the literacy acquisition of children in the classroom is affected not only by cognitive factors but also by environmental and psychological factors (cf., Berninger, Dunn, Lin, & Shimada, 2004; Dudley-Marling, 2004). The Component model of reading is broadly conceptualized and takes this fact into account. Components that have an influence on the acquisition of literacy skills are organized into three domains and constitute the component model of reading. The three domains of the component model are (1) cognitive domain, (2) psychological domain, and (3) ecological domain. The cognitive domain of the model has two components: word recognition and comprehension. The psychological domain includes components such as motivation and interest, locus of control, learned helplessness, learning styles, teacher expectation, and gender differences. The ecological domain includes the following components:
12 Role of Orthography in Literacy Acquisition and Literacy Problems
169
home environment and culture, parental involvement, classroom environment, and dialect. It has to be added that the components of the cognitive domain can satisfy the condition of independence fairly well whereas the components of the psychological and ecological domains do not satisfy this requirement nearly as well. Nevertheless, the component model provides a framework for teachers and psychologists for navigating their course through the various assessment formats and determining remedial strategies for use in the classroom. The importance of these psychological and environmental factors have been recognized by educators for a long time and has been also empirically documented (Berninger, et al., 2004; Dudley-Marling, 2004). The inspiration for the component model of reading comes from a report by Gough and Tunmer (1986) who presented a “Simple view of reading” by noting that the two most important constituents of reading are the ability to decode the word and the ability to comprehend text. Gough and Tunmer expressed this proposition in the form of the formula: R = D × L, wherein R is reading comprehension, D is decoding, and L is linguistic comprehension, as assessed by a test of listening comprehension. They set the value of each variable to range from 0 to 1. It follows then, if D is 0, then R is 0; if L is 0, then also R is 0. The validity of the formula was tested by Hoover and Gough (1990) by tracking and assessing 254 English–Spanish bilingual children from grades 1 to 4. The investigators found that a substantial proportion of the variance in reading comprehension was accounted for by the product of decoding and listening comprehension (grade 1, r = 0.71; grade 2, r = 0.72; grade 3, r = 0.83; grade 4, r = 0.82). Catts, Hogan, and Fey (2003), in a longitudinal study of children in early elementary grades, reported a low correlation of .16 between word recognition and listening comprehension indicating their relative independence. The independent nature of word recognition and comprehension is also reported by Carver (1998), Catts and Kamhi (1999), and de Jong and van der Leij (2002). Other investigators have described word recognition skill and comprehension skill as lower level processing and higher level processing skills, respectively (Hannon & Daneman, 2001; Pressley, 2000). The “verbal efficiency theory” (Perfetti, 1988) captures the essence of the bi-componential nature of reading by stressing the independent role of verbal efficiency and text comprehension. Verbal efficiency theory states that even though word recognition accuracy is a necessary but not sufficient for fluent reading comprehension. However, a high level of efficiency and automaticity is required so that the reader can devote more attention to meaning and comprehension. Frith and Snowling (1983) report that some children with autism can read aloud much better than they can comprehend whereas children with dyslexia can comprehend sentences better than they can decode nonwords. Dyslexic children and hyperlexic children also show that comprehension and decoding skills are dissociable (Aaron, Franz, & Manges, 1990; Healey, 1982). It is also reported that a substantial number of poor readers are deficient in decoding skills but have better comprehension skills, as determined by their performance on tests of listening comprehension (Crain, 1989; Shankweiler et al., 1995). The existence of children who can decode written words fairly well but cannot comprehend what they have read is less well publicized even though educators have recognized the existence of this type of poor readers for
170
R.M. Joshi
a long time and described them “word callers.” Research indicates that about 10% of the poor readers fall into this category (Stothard, 1994; Yuill & Oakhill, 1991). According to Mirak, Scarborough, and Rescorla (2003) some children who had normal word-level processing skills in earlier grades turn out to be deficient in reading comprehension when they reached fourth and fifth grades probably because the ability to comprehend what is read emerges later. This study further indicates that the two components of cognitive component of reading show different developmental trajectories and, therefore, it is important to diagnose which of the two components is lagging behind so that appropriate remedial instruction can be provided. This brief survey of studies indicates that word recognition and comprehension are separable component skills of reading of the cognitive domain. The utility of the Componential Model has been empirically supported. The independent nature of components, therefore, carries with it the potential for resulting in poor readers of more than one kind. In a study involving approximately 200 children in grades 3–6, it was found that approximately 8% of the children exhibited only decoding problems (dyslexia type), another 8% of the children exhibited only comprehension problems (hyperlexia type), yet another 8% of the children exhibited both decoding and comprehension problems (mixed type) (Aaron, Joshi, & Williams, 1999). Further, when children were identified as having decoding problems only and then provided them with systematic decoding instruction, there was an improvement in decoding as well as in comprehension; similarly, when children were identified with having comprehension problems only, and then provided with systematic instruction in comprehension, the comprehension scores improved significantly. A comparative group of children with reading problems but were not given differentiated instruction did not show improvement in reading comprehension (See Aaron, Joshi, Boulware-Gooden, & Bentum, 2008). In this chapter, we will present two studies that utilized the cognitive component of Componential Model in other writing systems. In the first study, cognitive domain was applied to Spanish orthography and in the second study dyslexia and hyperlexia were studied in bilingual subjects whose two languages had different writing systems – one language was alphabetic and the other language was syllabic or sometimes considered as alpha-syllabic.
Study I As mentioned above, Componential Model has been applied successfully with English-speaking children. The purpose of this study was to apply Componential Model to Spanish-speaking children. Spanish orthography can be considered more transparent than English orthography. In English, there are 26 letters, 44 phonemes, and approximately 250 graphemes while in Spanish spoken in Mexico, there are 28 (29, if the letter W is included) letters, 22 phonemes, and 29 graphemes. As it is apparent, there is a much closer correspondence between phonemes and graphemes in Spanish than English orthography. Thirty-eight participants from grade 2 and 42 participants from grade 3 who spoke Spanish at home and whose classroom instruction
12 Role of Orthography in Literacy Acquisition and Literacy Problems Table 12.1 Percentage of contribution by decoding (D) and reading comprehension among English and Spanish children Grade English 2 (n = 49) LC & D = 47% 3 (n = 54) LC & D = 48% 4 (n = 55) LC & D = 50% 2 LC = 33%; D = 35% 3 LC = 37%; D = 35% 4 LC = 41%; D = 14%
171
listening comprehension (LC) to Spanish (n = 38) LC & D = 57% (n = 42) LC & D = 60% LC = 45%; D = 25% LC = 47%; D = 15%
was in Spanish participated in the study. A comparative group of 49 children from grade 2, 54 children from grade 3, and 55 children from grade 4, whose home language and instruction was in English were selected for the study. Both groups were administered the decoding, listening comprehension, and reading comprehension subtests from the respective versions of Woodcock Battery (Woodcock-Johnson III Diagnostic Reading Battery, Woodcock, Mather, & Schrank, 1999; Bateria III, Woodcock, Muñoz-Sandoval, McGrew, & Mather, 2005). Multiple regression analyses were computed to see the contributions by decoding and listening comprehension to reading comprehension at the text level. Such analyses showed that approximately 50% of the variance in reading comprehension can be explained by decoding and comprehension among English speakers in grades 2–4. However, among Spanish speakers about 60% of the variance in reading comprehension can be explained by decoding and listening comprehension. A breakdown in the contribution of individual components (Table 12.1) showed that the performance of third grade Spanish-speaking children resembled the performance of fourth grade English-speaking children which supports the Orthographic Depth hypothesis.
Study II The purpose of this study was to explore the nature of dyslexia and hyperlexia in bilingual subjects. As mentioned earlier, the two important psychological components of reading are decoding and comprehension and it was found in one of our earlier study that approximately 8% of the children in elementary grades exhibited dyslexia-type syndrome and another 8% exhibited hyperlexia-type syndrome. That is dyslexics have poor decoding skills but good linguistic comprehension skills while hyperlexics have good decoding skills but poor comprehension skills. However, there are not many published reports on bilingual dyslexia and bilingual hyperlexia despite the fact that the majority of world’s population is bilingual. Dyslexia, according to the International Dyslexia Association (2003), “is characterized by difficulties with accurate and/or fluent word recognition and by poor spelling and decoding abilities. These difficulties are often unexpected in relation to other cognitive abilities.” American Hyperlexia Association (2003) defines, hyperlexia is characterized by “a level of word identification skills far above those of
172
R.M. Joshi
reading comprehension (both at the level of single words and connected text) and of general cognitive functioning of an individual.” Much of the information on reading problems in bilinguals has come from the study of bilingual alexics – fluent readers who have lost the ability to read. As early as 1895, James Hinshelwood, an ophthalmologist from Glasgow, reported the case of a 58-year-old teacher of French and German who suddenly lost his ability to read printed and written materials in all the languages he knew. This description is almost identical to the same provided by Dejerine 3 years earlier. Hinshelwood wrote that the patient “spoke as fluently as ever … his mental powers were as vigorous as ever, nor was there any defect of memory apart from the loss of memory for the visual symbols of language” (1895, p. 1565). Obler (1984), after reviewing the literature on bilingual alexia, concluded that “unfortunately, there is virtually no literature on childhood dyslexia in bilinguals” (p. 493). In fact, Hinshelwood (1902) also had wondered, “how is it that there are so few recorded cases of these partial forms of word-blindness, that is, cases of dissociation in polyglots? I think the reason is simply that the patient is not thoroughly examined by testing his power of reading all the characters and all the languages with which he is familiar … It has been taken for granted that if a patient was unable to read one language, he would therefore be unable to read any language, and that there was no necessity for applying further tests.” Majority of the published studies have reported problems in both the languages. For instance, Critchley (1970) mentions three cases of bilingual dyslexia, two were Arabic–English dyslexics and one Hebrew–English dyslexic. Even though the subjects made analogous errors in both languages, the errors in Arabic could be considered more as visual errors than the English words. However, these were not detailed analysis. Based on the limited evidence available on bilingual dyslexia in 1984, Obler, suspected that “if a child is dyslexic in learning to read one language, then he or she is as a rule dyslexic in learning to read the next” (p. 493). However, not everyone agrees with this. For instance, Wydell and Butterworth (1999) studied a developmental bilingual dyslexic, who was dyslexic only in English but not in Japanese, even though he had studied Japanese later. So far, most of the cases reported on hyperlexia are English-speaking monolingual subjects. However, there is one case of trilingual hyperlexic reported by Lebrun, Van Endert, and Szliwowski (1988). A microcephalic girl, Isabelle, could read in three languages, Dutch, French, and English at the age of 10 years. Dutch was her mother tongue and she could speak and read aloud fluently. French was her second language and her spoken command of it was rather limited but her reading aloud was as fluent as it was in Dutch. She could use English only to read but not to speak it. Her IQ on the Terman–Merrill was 55 at 4 years and 9 months which put her mental age at 2 years and 8 months. When tested Stanford–Binet Intelligence Scale at 7 years and 9 months, her mental age was between 4 and 5 years. Since Isabelle was hyperactive, Lebrun et al. could not determine whether her poor comprehension in spoken and written language was because of lack of ability or because of attentional problems. Nevertheless, Isabelle’s comprehension of both spoken and written language was poor in all the three languages albeit to a different degree. Her comprehension was best in Dutch and worst in English. There are more
12 Role of Orthography in Literacy Acquisition and Literacy Problems
173
research studies on dyslexia than on hyperlexia. For instance, an examination of the ISI Web of Science revealed that during the last 10 years (1994–2004), there were 2,573 papers published on the topic of dyslexia while there were only 28 papers published on hyperlexia. We conducted a detailed study of one bilingual dyslexic and one bilingual hyperlexic. The performance of MS, a bilingual hyperlexic, and VN, a bilingual dyslexic, was compared with a normally achieving subject, AN. Both MS and VN were about 16 years at the time of testing and AN was 15 years of age. All three came from socio-economically middle-class families and were fluent in two languages: English and Kannada. Kannada is one of the major Dravidian languages of South India. It is the state language of Karnataka, spoken by about 50 million people and has a long history of about 1,500 years. Kannada script presents a mixture of syllabic and alphabetic principles. There is almost one-to-one graphophonological equivalence expressed in syllable structure, regular signs of vowels being attached to the basic consonant forms. Each of the vowels, in addition to its syllabic (primary) form, has an intrasyllable (secondary) form which is used in writing a CV syllable and each syllable form can be analyzed into its consonant and vowel components similar to Korean Hangul. A detailed analysis of Kannada orthography can be found in Karanth (2003, 2007) and Prakash & Joshi (1995). All three participants were administered the Raven’s progressive Matrices (Raven, 1984), a non-verbal measure of mental ability; and decoding, listening comprehension at the word level as well as at the passage level; reading comprehension through cloze format as well as through answering questions after reading two passages, speed of processing at the letter level as well as at the word level, and spelling were measured both in Kannada and in English. Their performance showed that there were qualitative and quantitative differences in the decoding and comprehension performances of MS (hyperlexic) and VN (dyslexic) in both languages. While the few decoding and spelling errors of MS were due to regularization, the large number of decoding and spelling errors of VN appeared to be the result of random guesses. VN committed very few regularizations and word substitution errors. Both reading comprehension and listening comprehension skills of the hyperlexic subject are low. Speed of processing words and letters was related to decoding but not to reading comprehension. Regarding reading problems among bilinguals, a review of the literature shows that virtually all the studies have shown that bilinguals have exhibited reading problems in both the languages except for the Wydell and Butterworth’s (1999) study where they found that their subject was dyslexic only in English but not in Japanese. Perhaps, this could be explained by the relative distance between two orthographies. Most of the studies compared two alphabetic languages and hence reading problems were exhibited in both the languages. However, in the case study of Wydell and Butterworth, the two languages – English and Japanese – are farther apart than, for instance, English and French. English uses an alphabetic writing system while Japanese uses a combination of morphemic and syllabic writing system. As another great Israeli researcher, Dina Feitelson explained as early as in 1966, “among alphabetic languages, even languages that have entirely different
174
R.M. Joshi
origins or orthographies may share characteristics that translate into similar learning problems. With respect to learning to decode, the more characteristics two orthographies have in common, the more it seems possible to make use of strategies that have been shown to be effective in one orthography also in the second.” It appears that the Literacy problem is a universal phenomenon and is of global concern. However, these reading problems are influenced by the unique features of a culture and its language. Considering the fact that the majority of world’s population is bilingual/multilingual, more research on reading disability in bilingual population is needed. I have presented two studies utilizing the “Orthographic Depth Hypothesis” and the “Componential Model” that could be valuable in exploring literacy acquisition and literacy breakdown among speakers of languages other than English as well as among bilinguals. Acknowledgments I wish to thank the editors for inviting me to contribute a chapter in a volume dedicated to Dr. Iris Levin. Iris has contributed widely to various aspects of literacy and orthography. In this chapter, I have tried to combine these two areas in honoring her.
References Aaron, P. G., Franz, S., & Manges, A. (1990). Dissociation between pronunciation and comprehension in reading disabilities. Reading & Writing: An Interdisciplinary Journal, 3, 1–22. Aaron, P. G., & Joshi, R. M. (1992). Reading problems: remediation and consultation. New York: Guilford Press. Aaron, P. G., Joshi, R. M., Boulware-Gooden, R., & Bentum, K. (2008). Diagnosis and treatment of reading disabilities based on the component model of reading: an alternative to the discrepancy model of learning disabilities. Journal of Learning Disabilities, 41, 67–84. Aaron, P. G., Joshi, R. M., & Williams, K. A. (1999). Not all reading disabilities are alike. Journal of Learning Disabilities, 32, 120–127. American Hyperlexia Association. (2002). What is hyperlexia? Retrieved December 3, 2007. http://www.hyperlexia.org/aha_what_is.html. Berninger, V. W., Dunn, A., Lin, S. C., & Shimada, S. (2004). School evolution: scientist-practitioner educators creating optimal learning environments for all students. Journal of Learning Disabilities, 37, 500–508. Carver, R. P. (1998). Predicting reading level in Grades 1 to 6 from listening level and decoding level: testing theory relevant to simple view of reading. Reading and Writing: An Interdisciplinary Journal, 10, 121–154. Catts, H. W., Hogan, T. P., & Fey, M. E. (2003). Subgrouping poor readers on the basis of individual differences in reading-related abilities. Journal of Learning Disabilities, 36, 151–165. Catts, H. W., & Kamhi, A. G. (eds). (1999). Language and reading disabilities. Needham Heights, MA: Allyn & Bacon. Chiu, M. M., & McBride-Chang, C. (2006). Gender, context, and reading: a comparison of students in 43 countries. Scientific Studies of Reading, 10, 331–362. Crain, S. (1989). Why poor readers misunderstand spoken sentences. In D. Shankweiler & I. Y. Liberman (Eds.), Phonology and reading disability (pp. 30–42). Ann Arbor, MI: The University of Michigan Press. Critchley, M. (1970). The dyslexic child. Springfield, IL: Charles C. Thomas.
12 Role of Orthography in Literacy Acquisition and Literacy Problems
175
de Jong, P. F., & van der Leij, A. (2002). Effects of phonological abilities and linguistic comprehension on the development of reading. Scientific Studies of Reading, 6, 51–77. Dudley-Marling, C. (2004). The social construction of learning disabilities. Journal of Learning Disabilities, 37, 482–489. Feitelson, D. (1966). The alphabetic principle in Hebrew and German contrasted with the alphabetic principle in English. In P. Tyler (Ed.), Linguistics and reading. Newark, DE: International Reading Association. Foorman, B. R., Francis, D. J., Shaywitz, S. E., Shaywitz, B. A., & Fletcher, J. M. (1997). The case for early reading interventions. In B. Blachman (Ed.), Foundations of reading acquisition and dyslexia: implications for early intervention (pp. 243–264). Mahwah, NJ: Lawrence Erlbaum Associates. Frith, U., & Snowling, M. (1983). Reading for meaning and reading for sound in autistic and dyslexic children. British Journal of Developmental Psychology, 1, 320–342. Frost, R., Katz, L., & Bentin, S. (1987). Strategies for visual word recognition and orthographical depth: a multilingual comparison. Journal of Experimental Psychology: Human Perception and Performance, 13, 104–115. Gough, P., & Tunmer, W. (1986). Decoding, reading, and reading disability. Remedial & Special Education, 7, 6–10. Hannon, B., & Daneman, M. (2001). A new tool for measuring and understanding individual differences in reading comprehension. Journal of Educational Psychology, 93, 103–128. Hart, B., & Risley, T. R. (1995). Meaningful differences in the everyday experience of young American children. Baltimore, MD: Paul H. Brookes. Healey, J. (1982). The enigma of hyperlexia. Reading Research Quarterly, 17, 319–338. Henderson, L. (1984). Orthographies and reading. Hillsdale, NJ: Lawrence Erlbaum Associates. Hinshelwood, J. (1895). Word-blindness and visual memory. The Lancet, 21, 1564–1570. Hinshelwood, J. (1902). Four cases of word-blindness. The Lancet, 1, 358–363. Hoover, W. A., & Gough, P. B. (1990). The simple view of reading. Reading and Writing: an Interdisciplinary Journal, 2, 127–160. International Dyslexia Association. (2003). What is dyslexia? Retrieved December 3, 2007. http:///www.interdys.org/FAQWhatis.htm. Joshi, R. M., & Aaron, P. G. (2000). The component model of reading: simple view of reading made a little more complex. Reading Psychology, 21, 85–97. Karanth, P. (2003). Cross-linguistic study of acquired reading disorders. Dordrecht, the Netherlands: Kluwer Academic Publishers. Karanth, P. (2007). The Kagunita of Kannada – Learning to read and write an Indian alphasyllabary. In R. M. Joshi & P. G. Aaron (Eds.), Handbook of orthography and literacy (pp. 389– 404). Mahwah, NJ: Lawrence Erlbaum Associates. Lebrun, Y., Van Endert, C., & Szliwowski, H. (1988). Trilingual hyperlexia. In L. K. Obler & D. Fein (Eds.), The exceptional brain (pp. 253–264). New York: Guilford Press. Lukatela, G., & Turvey, M. T. (1999). Reading in two alphabets. American Psychologist, 53, 1057–1072. Mirak, L. J., Scarborough, H. S., & Rescorla, L. (2003). Late-emerging disabilities. Journal of Educational Psychology, 95, 211–224. Moats, L. C. (2001). Overcoming the language gap. American Educator, 29, 4–9. Obler, L. (1984). Dyslexia in bilinguals. In R. N. Malatesha & H. A. Whitaker (Eds.), Dyslexia: a global issue (pp. 477–496). The Hague: Martinus Nijhoff. Pennington, B. F., & Olson, R. K. (2005). Genetics of dyslexia. In M. Snowling & C. Hulme (Eds.), The science of reading: a handbook (pp. 453–472). Oxford: Blackwell. Perfetti, C. A. (1988). Verbal efficiency theory in reading ability. In M. Daneman, G. E. MacKinnnon & T. G. Waller (Eds.), Reading research: advances in theory and practice (pp. 109–143). New York: Academic Press. Prakash, P., & Joshi, R. M. (1995). Orthography and reading in Kannada: a Dravidian language. In I. Taylor & D. R. Olson (Eds.), Scripts and literacy (pp. 95–108). Dordrecht, the Netherlands: Kluwer Academic Publishers.
176
R.M. Joshi
Pressley, M. (2000). What should comprehension instruction be the instruction of? In M. L. Kamil, P. B. Mosenthal, D. Pearson & R. Barr (Eds.), Handbook of reading research (Vol. 3, pp. 545–561). Mahwah, NJ: Lawrence Erlbaum Associates. Raven, J. C. (1984). Raven’s progressive matrices. London, UK: Lewis. Seymour, P. H. K. (2007). Theoretical framework for beginning reading in different orthographies. In R. M. Joshi & P. G. Aaron (Eds.), Handbook of orthography and literacy (pp. 441–462). Mahwah, NJ: Lawrence Erlbaum Associates. Seymour, P. H. K., Aro, M., & Erskine, J. M. (2003). Foundation literacy acquisition in European orthographies. British Journal of Psychology, 94, 143–174. Shankweiler, D., Crain, S., Katz, L., Fowler, A. E., Liberman, A., Brady, S., et al. (1995). Cognitive profiles of reading-disabled children: comparison of language skills in phonology, morphology, and syntax. Psychological Science, 6, 149–156. Stothard, S. (1994). The nature and treatment of reading comprehension difficulties. In C. Hulme & M. Snowling (Eds.), Reading development and dyslexia (pp. 85–102). London: Whurr. Torgesen, J. K. (2005). Recent discoveries on remedial intervention for children with dyslexia. In M. J. Snowling & C. Hulme (Eds.), The science of reading: a handbook (pp. 521–537). Oxford, UK: Blackwell. Vellutino, F., Scanlon, D., & Jaccard, J. (2003). Toward distinguishing between cognitive and experiential deficits as primary sources of difficulty in learning to read: a two-year follow-up of difficult to remediate and readily remediated poor readers. In B. R. Foorman (Ed.), Preventing and remediating reading difficulties: bringing science to scale (pp. 73–120). Baltimore: York Press. Woodcock, R. C., Mather, N., & Schrank, F. A. (1999). Woodcock-Johnson III Diagnostic Reading Battery. Itasca, IL: Riverside Publishing Co. Woodcock, R. C., Muñoz-Sandoval, A. F., McGrew, K. S., & Mather, N. (2005). Bateria III. Itasca, IL: Riverside Publishing Co. Wydell, T. N., & Butterworth, B. L. (1999). A case study of an English-Japanese bilingual with monolingual dyslexia. Cognition, 19, 491–514. Yuill, N., & Oakhill, J. (1991). Children’s problems in text comprehension: an experimental investigation. New York: Cambridge University Press.
Chapter 13
Becoming Literate in a Foreign Language Liliana Tolchinsky
Abstract Mastering the language(s) of a literate community implies learning its written and spoken modalities and second-language learners must cope with the two modalities simultaneously. The study explores Chinese and Moroccan 5 to 8 years-old awareness to morphological processes and the relationship that they establish between the spoken and the written representation of these processes. Children were at their initial stages of learning Catalan, a language and a orthographic system that differ typologically from their own. Moreover, Chinese and Moroccan children earlier experience with spoken and written language differs to a large extent form the one under way. Our findings show that there is a subtle influence of learners’ first-languages in their mastery of critical aspects of nominal morphology of the target language. However, there are some critical matters that seem to reappear in all studied groups, irrespective of their first-languages. As for their conceptualization of the speech-writing relationship, many children were still unaware of the basic representational feature of alphabetic writing; namely, that whenever a spoken utterance gets a written representation, the changes this utterance suffers should be reflected in its written representation. The domain of literacy has undergone a turning point over the last two decades. From being considered as a school matter, an object of instruction, literacy has become a domain of knowledge that is worth studying from a developmental point of view. The work of Iris Levin made a notable contribution in this sense. Her studies are an obligatory reference in the study of early literacy throughout the world, and they inaugurated the developmental approach to literacy in Hebrew. In 1984, Iris and I sent our first common paper to Applied Psycholinguistics to be considered for publication.The goal of that paper was to investigate the development in preschoolers’ conceptualization of the writing system from its graphic rendering and its mapping into meaning, by analyzing children’s writing and their reading of their own writing (Tolchinsky-Landsmann & Levin, 1985). Together with L. Tolchinsky (*) University of Barcelona, Barcelona, Spain e-mail:
[email protected] D. Aram and O. Korat (eds.), Literacy Development and Enhancement Across Orthographies and Cultures, Literacy Studies 101, DOI 10.1007/978-1-4419-0834-6_13, © Springer Science+Business Media, LLC 2010
177
178
L. Tolchinsky
the usual requests for revision, the editor’s letter included a sentence that pointed to the fact that never before had the journal published anything on this topic. In the early 1980s, approaching literacy from a psycholinguistic developmental perspective was indeed a novelty. It was a time when both researchers and educators believed that what children know about reading and writing is what they are being taught at school. Our paper showed that there is an interesting development going on before schooling. Of course this is not a spontaneously emerging development; rather, it is an indicator of children’s sensitivity to the characteristics of writing. It shows that writing is an attractive object for children from very early on. In this first publication, it was not only the developmental approach that was innovative, but also the fact that the study of literacy was addressed through writing – and not merely through reading. Children were asked to write and to interpret what they had written, and this was carried out at a time when researchers considered that asking children to write at such an early age was detrimental for their development (Auzias, Casati, Cellier, Delaye, & Verleure, 1977). Nowadays, it is such an ordinary request that it is hard to grasp how innovative it was 25 years ago. Of course there were relevant antecedents to this approach to literacy. In the late 1920s, Luria and Vigotsky advanced the need to explore “the prehistory of written language development” that, according to the authors, starts “long before a teacher puts a pencil in a child’s hand” (Luria, 1929/1978; p. 147). However, this approach and its consequences for understanding literacy remained quite absent from psycholinguistic inquiry until the groundbreaking work of Ferreiro (1988), whose goal was to trace the evolution of writing in the framework of a psychogenetic approach to development. This line of research was particularly influential in our own work on literacy. Back in 1970, Charles Read had produced his pioneering study of young children’s categorization of speech sounds, as evidenced by the invented spelling found in their compositions (Read, 1971). This work stimulated subsequent studies on spelling (e.g., Henderson & Beers, 1980) and Carol Chomsky’s courageous proposal in 1971, “Write first, Read later.” But the perspective taken in our first study – in line with Ferreiro’s and her collaborators’ – differs, however, from the one on invented spelling in that we attempted to understand children’s ways with writing before spelling. We focused on the development of writing before children realize that writing represents the sounds in a word and look for correspondences between spoken segments and written elements. Parallel to Ferreiro’s work in Latin-America and to that of Charles Read, a more ethnographic, socio-cultural approach to early literacy can be traced in the work of Clay (1975), and Teale and Sulzby (1986). They point out that children’s interest in the written word does not emerge “naturally,” and describe the circumstances which might trigger, propitiate, enhance, or provoke reading and writing practices at home or in school settings. Theirs and many other related studies have demonstrated the extent to which family literacy practices can be of fundamental importance when it comes to succeeding at school (e.g., Heath, 1983; Scollon & Scollon, 1981). The third and most important novelty of the Applied Psycholinguistics study is that it was the first work ever published in English on the development of writing in Hebrew. Of course, many other studies had been published on literacy in Hebrew before that (see Feitelson, 1988), but they did not include developmental informa-
13 Becoming Literate in a Foreign Language
179
tion about reading or writing. That initial study and others that came out at that time (e.g., Levin & Tolchinsky-Landsmann, 1990; Tolchinsky-Landsman & Levin, 1987) paved the way for quite an important number of studies by Iris and her students, exploring the ways in which children become literate in Hebrew. There is no doubt that the work of Iris Levin, and her collaborators, is crucial for understanding the process of becoming literate in that language. Studies in Hebrew were carried out with native speakers or with speakers proficient in using L2 Hebrew, despite being new immigrants. Such was the rule in most studies on emergent literacy. The study I am going to present was, in contrast, carried out with children at their initial stages of learning a new language and a new writing system.
Becoming Literate in a Foreign Language The present study shares the main tenets of the developmental approach to literacy concerning the active role of the learner in the process of becoming literate, the recognition of children’s sensitivity to a printed environment and the uses of writing, and the importance of this early knowledge for further literacy learning. We assume that children are doing critical cognitive work with the input provided by the environment, as well as by other adults and children using printed materials; moreover, we believe that the notions children develop through their interactions with print and other readers and writers are the building blocks of further literacy learning. In our view, these basic tenets are valid both for children acquiring the written modality of their first language (WL1) and for children acquiring the spoken and written varieties of an L2. We also support the idea that it is useful for teachers to be aware of the notions children develop about the written language because, on the one hand, a teaching practice that builds on learners’ previous knowledge is more efficient and, on the other, because learners’ previous knowledge may interfere with the instructional approach. Therefore, it is useful for teachers to have information about the notions children develop in learning and L2 and a WL2. The latter is particularly relevant in contexts in which the number of foreign students attending regular classes is in constant increase. During the late 1990s there was an abrupt and massive growth of immigration in Catalonia thus increasing the need to understand the learning processes of Catalan as an L2. Second-language learners in a literate community, however, are immersed in the spoken and written varieties of the language(s) of the community and are exposed to printed input of different kinds; by necessity, they must cope with the two modalities simultaneously. Thus, we focused on children who are learning both an L2 and a WL2. Eighty-four children from 5 to 8 years of age participated in the study to be reported here. Twenty-six were from Morocco, twenty-eight from China, and thirty Catalan controls. Moroccan and Chinese children have settled in Barcelona and have attended Catalan schools for approximately 18 months. Children were divided into two age groups, the youngest group (5–6-year old) was attending preschool in
180
L. Tolchinsky
Barcelona and the eldest one (7–8-year old) was in second or third grade of primary school and has been attending school in their home countries for about a year. Barcelona is a Spanish/Catalan bilingual city, but Catalan is the only language taught at school until third grade of primary school. Children knew enough Catalan to understand the interviewer’s instructions, though they were far from being proficient speakers of Catalan. The study was designed to address one of the fundamental questions in L2 acquisition, namely, to what extent does the L1 influence the process of learning an L2. Two hypotheses were in the background: on the one hand, there is the hypothesis that L1 knowledge has a strong influence on L2 learning (Liceras, 1997; Smith, & Tsimpli, 1995; Tsimpli, & Roussou, 1991). Accordingly, it is expected that the kind of mistakes and difficulties learners will find in the L2 will differ as a function of their L1s. An alternative hypothesis (Purdue, 1993) suggests that, whatever L2 is being acquired, learners go through similar stages of acquisition that are not so strongly related to a learner’s L1. Many studies have been carried out to explore the achievement and/or rate of learning to read and write by non-native learners (see for a review Verhoeven, 2005). They have found that individual variation is much larger among non-native learners than among native learners. Unfortunately, the influence of learners’ L1 on their learning outcomes is far from clear. Studies that have focus on the extent to which the structural distance between the languages and writing systems in which the children are proficient affect their literacy learning (Bialystok, Luk, & Kwan, 2005) or have explored the cognitive effects of bilingualism on literacy (Byalistock, 2002) showed that structural proximity between writing systems is more beneficial to literacy learning than distance between them. The present study serves to shed some light as to the influence of L1 on learning an L2 because participants’ L1s strongly differ in their degree of morphological richness – the most important typological property of a language (Dressler, 2004). Chinese participants’ L1 was Qina tian hua, which, as any Chinese language, has a very poor morphology. In Qina tian hua almost every word consists of a single morpheme. Children of Moroccan origin, by contrast, spoke Darija, which has a much richer morphology. Catalan, the target language, as any Romance language, is also very rich in inflectional morphology. As for the possible effect of distance between the learner’s WS1, no difference should be expected between the performance of Chinese and Moroccan children because the two writing systems strongly differ from the target. Our interest centered in how these children conceive the relation between the spoken and the written modality in the first stages of learning an L2 and a WL2. This question is particularly relevant for Chinese and Moroccan children because their earlier experience with spoken and written language differs to a large extent form the one under way. Both for Chinese and Moroccan children, the relationship between spoken and written language in their home countries is very different from the relationship between the two modalities in the target language. In their home countries, the language that is spoken is not written and, conversely, the language which is written is not spoken. Becoming literate in China or in Morocco implies gaining command of a different language: Mandarin in China;
13 Becoming Literate in a Foreign Language
181
Standard Modern Arabic (SMA) in Morocco. In Catalonia, there is not such diglossia: becoming literate means having access to the written modality of the spoken language. The goal was, thus, to explore the effect of diglossia in children’s conceptualization of the links between speech and writing in Catalan. Given the difference in the morphological systems, we focused on children’s sensitivity of two morphological processes in the target language: number inflection, and derivation of locative nominals. In Catalan number assumes singular and plural values, and number changes are marked by means of suffix /-s/ and its allophonic variants, affecting both the noun and the determiner. That is to say, number agreement is compulsory. In Darija, the feature of number assumes singular and plural values, as well, and number morphemes include suffixes or infixes that are added to the singular form of the noun, but there is no number agreement in the nominal system. In contrast, Chinese languages lack declensions or morphemes for marking number distinctions, and use particles for marking the grammatical feature of number; there is no number agreement in the nominal system. Therefore, we aimed at determining whether children are aware that in the language they are learning the indefinite nominal group used for describing a picture with a singular object (e.g., un cotxe “det-sg car”) should be modified to describe a picture showing a plurality of objects. And, in addition, if they know that the modification should affect both the determiner and the noun. Similar differences among the three languages are noted for locatives. In Catalan locative nouns are obtained by adding suffix /-ə’riə/ to the base form of the noun, whereas in Qina tian hua locatives are N + N compounds. In SMA derivation of locatives is obtained by typical processes of Semitic, concatenated morphology in which a prefix and a vocalic pattern are inserted in a tri-consonantal root. For example, from [k t b] maktabun “office” and maktabatun “library.” The kind of vocalic pattern and the phonetic realization of the prefix depend on features of the consonantal root. In Darija, however, this is not a productive mechanism and the locatives that are formed in that way – such as mahjara “butchery” – have been inherited from Classical Arabic. The current way for locative formation is by paraphrases of the form muta’a d’sabato “place for shoe.” We aimed at determining whether children modify the nominal they use to describe a picture with a single object when they are asked to describe a picture showing the place at which this object is sold. Most importantly, we wanted to determine whether children realize that a written representation produced for a certain spoken utterance should be modified to reflect any modification on such spoken utterance. We assumed that if L2 learning is influenced by the characteristics of learners’ L1, we should find a difference in the way Chinese and Moroccan children learn the functioning of Catalan inflectional and derivational morphology. Moreover, if experience with a richly inflected language enhances learners’ sensitivity to L2 morphology – as seems to be the case for L1 acquisition (Dressler, 2007), we should find that Moroccan children learn better and faster the morphological marking of number, and the formation of locative nominals than the Chinese children. We further assumed that, due to the similar diglossic situation the two groups of children were part of, we should find a similar approach to the links between speech and writing in both language groups.
182
L. Tolchinsky
As we shall see in the following lines, in spite of the typological differences between the languages included here, there are similar processes that are underway in the two experimental groups, and even in the control group. We will also see that despite children’s early sensitivity to print, they do not necessarily become aware of the fact that if a certain written representation is produced for a certain spoken utterance, any changes in such spoken utterance should be reflected in writing.
Spoken Representation of Number Inflection All the children participated in a number of individual, semi-structured tasks designed to evaluate spoken and written production. The interview was conducted in a separate room of the school by two interviewers. One of the tasks focused on numberinflection. Children were presented with an initial picture of an object with a caption below (e.g., for a picture showing one car the caption read
“det-sg car”). They were required to say what was there in the picture, without any indication on the part of the interviewer. Afterwards, regardless of the kind of description produced by the child, the interviewer said to him/her what was written on the caption (e.g., Aquí hi posa ‘un cotxe’; ‘Here, it says ‘a car’). Children were then presented with a target picture with the same object, but differing in number (plurality of objects, if the initial picture had been a singular presentation, or one object, if the initial picture had been a plural presentation). The target picture contained an empty label below the object or objects. Children were then asked to name the target picture (e.g., for a picture with eight identical cars they were expected to say uns cotxes “detpl cars” and to write a caption for it on the empty label). The list of nominal groups used in the inflectional task was the following: un cotxe (“a car”), una casa (“a house”), un quadre (“a picture”), una poma (“an apple”), un llibre (“a book”), una taula (“a table”), un arbre (“a tree”), una porta (“a door”), and their plural counterparts. All the selected words were concrete, countable nouns with rather simple disyllabic structure – consonant clusters of a maximum of two elements – stressed in the penultimate syllable – the commonest stress pattern for Catalan. The utterances children used for naming the initial picture and the target picture, as well as the written representation of the target picture, were analyzed (for details see Tolchinsky, Salas, & Perera, 2007; Tolchinsky & Salas, 2009). If children are already sensitive to the way in which Catalan represents number changes in indefinite NGs, they will modify both the noun and in the article. Children reacted very differently when they had to produce a singular, indefinite nominal group, compared to when they had to produce a plural, indefinite nominal group. Almost half of the utterances produced by Chinese and Moroccan children reflected the change from plural to singular in the determiner and the noun respecting number agreement within the nominal group. This behavior tended to increase with age. Utterances that did not reflect this change from plural to singular were more frequently produced among the Chinese than among the Moroccan children, suggesting that the task itself was more difficult, in general, for the Chinese children. A completely different picture is obtained for the production of plural, indefi-
13 Becoming Literate in a Foreign Language
183
nite NGs. Not a single utterance produced by the youngest Chinese children, and only three of those produced by the Moroccan children reflected the change from singular to plural in the determiner and the noun, showing internal nominal group agreement. When children were able to show such changes, they did it by means of a numeral or a quantifier. However, and most interestingly, also for the Catalan controls the plural condition was much more difficult than the singular one. Only 6% of the youngest Catalans’ responses were indefinite, plural NGs, while 73% of them reflected plurality using numerals or quantifiers. Although the use of indefinite forms increased with age, it accounted for just 29% of utterances. In around half of their utterances, there was still a preference for numerals or other quantifiers over the indefinite plural article. In other words, both Chinese, Moroccan, and, to some extent, Catalan children, were reluctant to produce plural indefinite-article NGs, while being prone to produce singular ones, when required. The cardinal article un “a” is considered essential for describing singular objects and, as such, is included in children’s utterances when they are asked to reflect a change from plural to singular. This element, as any cardinal number, has lexical meaning, its use is compulsory, and there are no alternatives to it. To reflect a change from singular to plural the situation is very different. Although un (“a”) alternates with uns (“a-pl”), the cardinal function of uns can be fulfilled by any numeral or quantifiers (e.g., molts “many”). Moreover, uns has no independent lexical meaning its use is not compulsory and, even in the absence of any modifier the plural morpheme suffixed to the noun accomplishes the representation of plurality. These reasons converge to explain children’s preference for the use of the indefinite/cardinal article + N for describing singular objects, and just the N(pl) or a quantifier/cardinal + N(pl), for a plurality of objects. This characteristic of the L2 had a similar effect on both Chinese and Moroccan children, and even to some extent on the Catalan controls, leading to a very similar behavior in their use of the singular, indefinite article, and a mere quantitative – though not qualitative – difference in the controls. This is not to say that there were no indicators of L1 influence on children’s learning process. We found such influence, not so much in the correct or more adequate responses, but in the kind of deviated responses children produced. For example, although both Chinese and Moroccan children have problems with representing the plural by means of an indefinite nominal group, Chinese children tended to represent plurality by changing the initial article (for a numeral, for instance), leaving the N unmodified; whereas Moroccan children preferred to produce modifications in the noun, deleting the indefinite article or leaving it unmodified. Below is an example of a 6-year old from Moroccan origin (ex. 1). When asked to describe a plural presentation (target picture with many tables) after having described a singular presentation he deleted the article and modified the noun. In contrast (ex. 2), for the same pair of pictures, a Chinese child of similar age uses a quantifier and leaves the noun unmodified. Ex. 1 (Initial picture displaying a table, with a caption) Interviewer: i això què és? “and what is this?” Child: taula “table” Interviewer: una taula “a table” Child: una taula “a table”
184
L. Tolchinsky
Interviewer: molt bé hi posa una taula “very good, it says a table” (Target picture displaying many tables, without a caption) Interviewer: i aquí? “and here” Child: taules “tables” [deleting the article and modifying the name] Ex. 2 (Initial picture displaying a table, with a caption) Interviewer: i això què és? “and what is this?” Child: una taula “a table” (Target picture displaying many tables, without a caption) Interviewer: i aquí? “and here?” Child: *molts taule * “many table” These and other related results show that learners’ L1 only slightly affects the degree of successful performance in the L2. Children from Moroccan origin perform better than children from Chinese origin but not significantly so. However, the influence is evident in the kind of strategy or the kind of mistakes children produce with respect to the particular features of the target language. Additional evidence in this same direction is provided by the way Chinese and Moroccan children solve the derivation of locatives.
Spoken Representation of Locative Derivatives Another task in the study focused on the derivation of locative nouns. It had an identical structure to the inflection task: children were presented with an initial picture with an object (e.g. sabata “shoe”) – base noun condition– or with a picture of the store were the object can be bought at (e.g., sabateria “shoe-locative”; “shoestore”) – derivative condition. Again, a total of eight words – four for each condition – were used. If children are already sensitive to the way in which Catalan forms locative nouns from common base nouns, they should add a suffix to the noun to form the locative and, conversely, delete a suffix in order to obtain the base noun. In general, the derivation task was more difficult than the number inflection task. And, as in the previous task, the scenario that emerged for the derivation of base nouns from locatives was quite dissimilar to that of the formation of locative derivatives. Parallel to the differing performance in children’s production of a singular, indefinite nominal group as compared to the utterance of a plural nominal group, production of base nouns was more successful than production of locatives. Just to give a hint, in the youngest group of Chinese children, 20% of the productions were regular base nouns, after having been presented with the corresponding locative; however, only 2.5% of this group’s productions contained a regular locative after the presentation of the base noun. In spite of the fact that children’s performance improves with age, this difference between the production of base nouns and locative derivatives was maintained. It should be noted that, like in the number inflection task, the Moroccan children performed better than the Chinese: they got a higher percentage of productions consisting of regular base nouns and
13 Becoming Literate in a Foreign Language
185
locatives although the difference did not attain significance. The effect of the L1 that was subtle in the degree of success to complete the task is noticeable in the differing strategies children employ throughout the derivation task. The youngest Chinese tended to repeat what they themselves or the interviewer had said to describe the initial picture when asked to describe the target one. Moroccan children, on the other hand, explored other ways of solving the situation. The next example (ex. 3) shows a 5-year-old Moroccan child resorting to a paraphrase instead of deriving a locative by adding a suffix to the base noun. Ex. 3 (Initial picture displaying a shoe, with a caption) Interviewer: Com es diu això? “How is this called?” Aquí hi posa SABATA ‘Here (pointing at the caption) it says “shoe.” Child: sabata “shoe.” (Target picture displaying a shoe-store, without a caption) Interviewer: El lloc on es venen/compren sabates es diu... “The place where shoes are sold is called …?” Child: di per comprar sabatas “say_3rd.sg1. for selling shoes” The same strategy was applied by Ibrahim each time he was asked to produce a locative derivative. In so doing, the child was using the mechanism for generating locatives that is productive in his L1.
Links Between Spoken Modifications and Written Representations The participants in the study were asked to fulfill a sociolinguistic questionnaire with the help of the interviewer, to establish the language(s) spoken at home, their time of residence in Catalonia, and the socio-educational background of their parents or caretakers (profession, composition of the family, etc.). The interviewer also asked the child about his/her contact with books, newspapers, if he/she used to visit libraries, and so on. In sum, we inquired about the overall presence of written material in the child’s closest environment. Most children came from rural, rather deprived environments. Those from Morocco have attended M’sid or Koranic school usually found next to the masque in all villages of rural Morocco. In addition to that, and in order to get a fuller picture of children’s literacy experience, the younger children had to participate in a semi-structured interview, designed to determine (1) their familiarity with children’s books; (2) the extent to which they are aware of the function of socially-meaningful texts (such as supermarket tickets, calendars, and timetables); and (3) the vocabulary they have for naming these items. Results showed that there were no language-related differences
1
The first verb used by the child (to say) is in third person singular.
186
L. Tolchinsky
in participants’ familiarity with children’s books in any age group or in participants’ grasping of the function of everyday written material. The only item in which a significant difference noted was in participants’ knowledge of the name of some socially meaningful texts, with Chinese children showing a better performance than Moroccan Arabs. This indicates that the general literacy level, as measured by this instrument, is rather similar in both experimental groups. Nevertheless, a few Chinese and Moroccan children were not quite clear about the speech/writing relations in the WL2. To explore children’s conception of the links between speech and writing, we paid special attention to how the child used the caption on the initial picture in order to write the caption for the target one. We should recall that the initial picture with the caption remained all the time in front of the child, so that he/she could use it at his/her convenience. If the child modified the description of the initial picture to describe the target one and moreover modified the original caption to write the new caption, we regarded this as an indication of the child being aware that changes in oral descriptions should be reflected in writing. The following example (ex. 4) illustrates a correspondence between spoken and written representation. This 6-year-old child from Moroccan origin has described correctly the singular presentation. When asked to write it down, he doubted, like many other children, whether to include or not the article he has used in the spoken description. The interviewer let him decide. He eventually decides not to use it and only writes down the noun without the plural suffix. Ex. 4 (Target picture displaying a framed picture, without caption) Interviewer: I aquí, què hi ha? “And, here, what is it here?” (pointing at the target picture) Child: un quadre “a picture” Interviewer: Escriu-ho (el que hi ha aquí) “write, write what it’s here” Child: pu puso un quadre o nomes más quadre * “put a picture or just more pictures” [should he put a picture or just more pictures] Interviewer: el que vulguis, el que tu creguis “as you wish, as you think” Interviewer: Escriu-ho (el que hi ha aquí) “write, write what it’s here” In this example, the child’s written caption reflects exactly the modification he
did in the spoken description. However, if the child produced a spoken description of the target picture that differed from the one produced for the initial picture but, nevertheless, decides to copy from the original caption; this would be an indicator that he/she does not consider necessary to show in writing the changes he/she has produced in the spoken description.
13 Becoming Literate in a Foreign Language
187
Similarly, if the child repeated verbatim the description he or the interviewer had provided for the initial picture to describe the target one and, in spite of that, writes a caption very different from the original, it was taken, as well, as an indicator of the child neglecting the necessary link between spoken and written representations. In almost half of the cases in which the youngest Chinese children had to produce a written representation for a spoken utterance they behaved as though it were not compulsory for writing to show changes made in spoken utterances: they either replicated the original caption, having produced a change in the spoken description of the target, or repeated what they had uttered for the initial picture, but then did not copy from the original caption. In the following example (ex. 5), the child did produce an utterance that reflected a change from plural to singular, but nevertheless he copied from the written caption. Ex. 5 (Target picture displaying a many tables, without a caption) Interviewer: I aquí, què hi ha? “And, here, what is it here?” (pointing at the target picture) Child: taula...taulas un, dos, tres, cuatra, sinc, sis, set “table...tables one, two, three, four, five, six, seven” Interviewer: Escriu-ho (el que hi ha aquí) “write, write what it’s here” (i.e., copies from the caption on the initial picture)
[CHY135161] A similar trend was found among the Moroccan children. In almost half of the cases they behaved as though the changes they have performed in the spoken utterances should not be necessary reflected in writing. Although this kind of performance tended to decrease slightly with age, it still remained a popular strategy among the older Moroccan children. Such responses were nonexistent in the control group, who massively changed both the spoken utterance and its written representation.
Discussion Mastering the language(s) of a literate community implies learning its written and spoken modalities. Children who, because of migration, start this process after age 5 profit from spoken and printed information of different kinds provided by the environment. We assumed that this information is actively assimilated by the learner:
188
L. Tolchinsky
it is selected and, probably, transformed, as a function of children’s knowledge and ways of understanding. In the experimental situation that we designed for the study, the interviewer provided both spoken and written information – when reading the captions that describe the initial picture – that the child could use in order to complete in the task. The two main topics of our study were addressed by observing how children exploit this information. Examining the way in which children modified or repeated initial utterances to describe the target pictures, we were able to appreciate their sensitivity to the morphological mechanisms of the L2. An assessment of the role of copying enabled us to infer children’s degree of awareness of the links between speech and writing in the host linguistic community. Note that the first topic – morphological processes – dwells on a fundamental feature of Catalan as a Romance language: sensitivity to morphological processes is key in the process of becoming a Catalan speaker. The second topic – the links between speech and writing – is also crucial in the literate development within an alphabetic system. We were particularly interested in determining the extent to which typological differences between learners’ L1s, as well as the situations of diglossia from which they proceeded, affect their sensitivity to morphology and their awareness of the links between the WL2 and the L2. Our findings show that there is a subtle influence of learners’ L1 in their mastery of critical aspects of nominal morphology of the L2. In line with Dressler’s (2007) hypothesis, growing up in a language with rich morphology like Darija seems to enhance sensitivity to the morphological mechanisms of Catalan – also morphologically rich – more than growing up in a language of poor morphology, such as the language spoken by our Chinese subjects, but only slightly. Children from Moroccan origin find it easier to reflect changes in number through inflection, though not significantly so. The linguistic features of the L1s, however, seem to explain some of the mistakes learners committed in their attempts to describe verbally changes in number or derivation of a locative noun from a base noun. However, there are some critical matters that seem to reappear in all studied groups, irrespective of their L1. The cardinal interpretation of indefinite articles that explains both children’s use of the singular, indefinite article and their reluctance to use the plural, indefinite one in the three languages involved is an example of a core issue that prevails over typological differences. The overall difficulty of noun derivation over number inflection is another example of cross linguistic generality. The evaluation of children’s functional knowledge of literacy showed that they were quite sensitive to the contexts of use and to the function of social texts, found themselves at ease with children’s books, and were able to discriminate numbers, letters, and other signs. Nevertheless, many children were still unaware of the basic representational feature of alphabetic writing; namely, that whenever a spoken utterance gets a written representation, the changes this utterance suffers should be reflected in its written representation. Certainly, alphabetic writing – and, for that matter, any writing system – cannot reflect all the changes that take place in spoken utterances like, say, changes in the intentionality of the speaker (Olson, 1994). However, changes which imply addition or deletion of words, morphemes, or pho-
13 Becoming Literate in a Foreign Language
189
nemes are generally reflected in writing. From the way children related to the written caption in relation to the spoken utterances they had produced we conclude that many children were not clear about this principle of writing. They copied the written caption while changing the verbal description or, alternatively, they produced a different caption while repeating the verbalization. It could be argued that children’s way of dealing with the written/spoken relation only reflects children’s ignorance of letter/sound correspondences in the WL2 or that it is just a strategy applied by them to solve the situation. Against this interpretation we should take into account that most of the participants in the study have acquired already the conventional letter to sound correspondences in Catalan. Moreover, the tasks we have used in the study were beyond children’s knowledge of the L2, thus, everyone looked for ways to solve the situation; the interesting point here, the point that is worthwhile analyzing, is why they resort to that particular strategy. Although much more work is required in understanding the acquisition of derivational morphology, a domain that has not been too much studied to date (Ravid, 2004), it seems rather difficult for L2 learners. The study of how children find their way in a new language and a new writing system is an emerging field of inquiry, as it was once the study of early literacy from a developmental point of view; our hope is that it will be as fruitful and stimulating as the other has been. Acknowledgments The study reported in the chapter was partially supported by an I+D grant from the Spanish Ministry of Education BSO2003-06020.
References Auzias, M., Casati, I., Cellier, C., Delaye, R., & Verleure, F. (1977). Écrire à cinq ans?. Paris: Press Universitaire de France. [Writing at five?]. Byalistock, E. (2002). Acquisition of literacy in bilingual children: a framework for research. Language Learning, 52(1), 159–199. Bialystok, E., Luk, G., & Kwan, E. (2005). Bilingualism, biliteracy, and learning to read: interactions among languages and writing systems. Scientific Studies of Reading, 9, 43–61. Chomsky, C. (1971). Write first, read later. Childhood Education, 47(6), 296–299. Clay, M. (1975). What did I write? Beginning writing behaviour. NH: Heinemann Educational Books. Dressler, W. U. (2004). Degrees of grammatical productivity in inflectional morphology. Italian Journal of Linguistics, 15, 31–62. Dressler, W. U. (2007). Introduction. In S. Laaha & S. Gillis (Eds.) Typological perspectives on the acquisition of noun and verb morphology. Antwerp Papers in Linguistics, 112 Feitelson, D. (1988). Facts and fads in beginning reading a cross-language perspective. Norwood, NJ: Ablex. Ferreiro, E. (1988). L’écriture avant la lettre. In H. Sinclair (Ed.), La production de notations chez le jeune enfant: langage, nombre, rythmes et melodies (pp. 17–70). Paris: Press Universitaire de France. [The production of notations in young children: language, number, rhytms, and melodies]. Heath, S. B. (1983). Ways with words: language, life and work in communities and classrooms. Cambridge, MA: Harvard University Press.
190
L. Tolchinsky
Henderson, E., & Beers, J. (eds). (1980). Developmental and cognitive aspects of learning to spell. Newark, DE: International Reading Association. Levin, I., & Tolchinsky- Landsmann, L. (1990). Becoming literate: referential and phonetic strategies in early reading and writing. European Journal of Behavioural Development, 12, 369–384. Liceras, J. M. (1997). La evolución del concepto de parámetro en la teoría lingüística y las consecuencias para la gramática comparada. In P. En Fernández Nistal & J. M. Bravo (Eds.), Aproximaciones a los estudios de traducción (pp. 71–113). Valladolid: SAE, University of Valladolid. [The evolution of the concept of parameter in linguist theory and its consequences for descriptive grammar. In Fernández Nistal, P. & Bravo, J.M. (eds.) Approximations to the study of Translation]. Luria, A. R. (1978). The development of Writing in the child. In M. Cole (Ed.), The selected writings of A. R. Luria (pp. 146–194). New York: M.E. Sharpe Inc.. (Original work published 1929). Olson, D. (1994). The World on Paper. Cambridge University Press Purdue, C. (1993). Adult language acquisition. Crosslinguistic perspectives (2 volumes). Cambridge: Cambridge University Press. Ravid, D. (2004). Derivational morphology revisited: later lexical development in Hebrew. In R. Berman (Ed.), Language development across childhood and adolescence (pp. 53–81). TILAR Amsterdam: John Benjamin. Read, Ch. (1971). Pre-school children’s knowledge of English phonology. Harvard Educational Review, 41, 1–34. Scollon, R., & Scollon, S. B. K. (1981). Narrative, literacy and face in interethnic communication. Norwood, NJ: Ablex. Smith, N., & Tsimpli, I. M. (1995). The mind of a savant: language learning and modularity. Oxford: Blackwell. Teale, W., & Sulzby, E. (eds). (1986). Emergent literacy: writing and reading. Norwood, NJ: Ablex. Tolchinsky-Landsmann, L., & Levin, I. (1985). Writing in preschoolers: an age related analysis. Applied Psycholinguistics, 6, 319–339. Tolchinsky-Landsman, L., & Levin, I. (1987). Writing in four to six years old: representation of semantic and phonetic similarities and differences. Journal of Child Language, 14, 127–144. Tolchinsky, L., Salas, N., & Perera, J. (2007). Spoken and written representation of number in L2 Catalan indefinite DPs. Catalan Review, XXI, 321–350. Tolchinsky, L., & Salas, N. (2009). Written representations of nominal morphology by Chinese and Moroccan children learning a Romance language. In Ch. Bazerman, R. Krut, K. Lunsford, S. McLeod, S. Null, P. Rogers and A. Stansell (Eds.), Traditions of Writing Research. New York and London: Routledge Tsimpli, I.M. & Roussou, A. (1991). Parameter-resetting in L2? In UCL Working Papers in Linguistics 3, 149–169 Verhoeven, L. (2005). Literacy development across language boundaries. In D. Ravid & H. Baat Zeev Shydkrot (Eds.), Perspectives on language and language development (pp. 437–452). Dordrecht: Kluwer.
Chapter 14
Rudiments of Inflectional Morphology Skills in Emergent English–Hebrew Biliterates Esther Geva and Dana Shafman
Abstract The present study investigated the development of inflectional morphological skills in primary school children learning Hebrew as a foreign language. Participants were grade 1 and 2 children whose home language is English, attending a bilingual English-Hebrew day school in Canada. Side by side with a growing repertoire of Hebrew vocabulary, grammatical skills and sentence comprehension, rudiments of various inflectional skills are already part of the linguistic skills of primary school children who have had one to two years of exposure to Hebrew in the classroom. Hebrew as L2 children who had better syntactic skills were more successful on more analytic morphological inflection tasks as well. Familiarity with specific lexical items was more important for the successful inflections of specific words than general vocabulary knowledge. On the whole, just like monolingual children, Hebrew as L2 children who have better developed vocabulary and morphological skills are better able to comprehend sentences, and vice versa.
Morphological Awareness, Vocabulary, and Literacy Skills Morphology provides one of the organizing principles of the mental lexicon (Aitchinson, 2003). Processes of morphological analysis underlie lexical expansion in school children, and a substantial proportion of the words children know are acquired through morphological form-to-meaning mappings (Anglin, 1993). The use of word parts, including morphological awareness (MA), to identify and retain meanings is an area that has been receiving increasing attention among researchers and educators. Morphological awareness entails the ability to reflect on and manipulate morphemes, the smallest meaningful units in words. It involves the ability to reflect on complex words in a way that may assist with the derivation of their meaning. According to Carlisle (2003) “… morphological learning is such that phonologic E. Geva (*) University of Toronto, Toronto, ON, Canada e-mail: [email protected]
D. Aram and O. Korat (eds.), Literacy Development and Enhancement Across Orthographies and Cultures, Literacy Studies 101, DOI 10.1007/978-1-4419-0834-6_5, © Springer Science+Business Media, LLC 2010
191
192
E. Geva and D. Shafman
and orthographic representations of morphemes in written language cannot be understood as a function of phonological awareness alone without regard for morphology or for morphological awareness alone without regard for phonology” (p. 307). MA is a complex construct that involves phonological, semantic, syntactic, and orthographic knowledge (Ravid & Malenky, 2001).
MA and Literacy Skills The role of lexical and morphological knowledge in enhancing literacy development has been established in studies involving typically developing and language-impaired children (Carlisle, 2000), and difficulties in identifying morphological relationships are related to poor reading comprehension (Mahony, Singson, & Mann, 2000). Research involving monolingual English as a first language (L1) and Hebrew as L1 children has shown that morphemic awareness and the ability to segment and manipulate morphemes within complex words continue to develop through their school years. In Hebrew and English alike, this ability makes independent contributions to reading and spelling over and above phonemic awareness and decoding skills (Aram, 2005; Fowler & Liberman, 1995; Levin, Ravid, & Rapaport, 1999; Ravid, 2001; Singson, Mahony, & Mann, 2000). Performance on MA tasks tends to correlate with various linguistic and reading tasks, including phonemic awareness (Muter, Hulme, Snowling, & Stevenson, 2004; Singson, Mahony, & Mann, 2000), pseudoword decoding (Deacon & Kirby, 2004; Nagy, et al., 2006; Singson, et al., 2000), various measures of vocabulary knowledge (Carlisle, 2000; Deacon & Kirby, 2004; Ku & Anderson, 2003; Nagy, Berninger & Abbot, 2006; Singson et al., 2000), word reading tasks (Deacon & Kirby, 2004; Muter et al., 2004; Singson, et al., 2000), reading of morphologically complex words (Carlisle, 2000; Saiegh-Haddad & Geva, 2007), and reading comprehension (Carlisle, 2000; Deacon & Kirby, 2004; Ku & Anderson, 2003; Muter et al., 2004). Ku and Anderson (2003), who reviewed the literature, conclude that, across several languages, children develop inflectional morphology skills in their home or first language (L1) before skills involving derivational and compound morphology, with the latter two continuing to develop throughout the elementary school years.
Morphological Processes Alphabetic orthographies preserve morphological relatedness in the spellings of words (Verhoven & Perfetti, 2003), but they vary in morphological transparency, or the degree to which the sound and the meaning of a complex word can be recovered from its internal morphological structure (Elbro & Arnbak, 1996). Concatenative languages, like English, utilize linear morphological processes (prefixing or suffixing) to generate new words from free stems (McCarthy, 1981). In English, complex words typically retain the unique phonological and orthographic identity of the stem, thus preserving morphological transparency (e.g., consider care–carelessness). English stems may undergo phonological and/or orthographic shifts (e.g., complete–completion).
14 Rudiments of Inflectional Morphology Skills
193
Word derivation in non-concatenative languages, like Hebrew, is non-linear and complex (Ravid & Malenky, 2001). Word formation involves the simultaneous affixation of two linguistic units: a consonantal root (e.g., L-M-D) that signals the core meaning or the semantic family of the word, and a word-pattern (“mishkal”), which is a fixed and primarily vocalic template or pattern that instantiates the root as a unique lexical item. Both the root and the word-pattern are bound morphemes that cannot stand on their own as independent words. The concurrent affixation of the consonantal root within fixed slots in the word-pattern often leads to discontinued phonological and/or orthographic representations of the root. For instance, applying the consonantal root L-M-D onto the word pattern CiCuC (where C represents the slots intended for the insertion of the root consonants) results in a disrupted representation of the root L-M-D in the word limud (studying), and the vocalic pattern CaCCan, yields lamdan (scholar). This example illustrates the resultant morphological opacity in Hebrew.
The Development of Hebrew Morphological Skills in Monolingual Hebrew Speakers Hebrew can be characterized as a synthetic language that is rich in morphological structures. Ravid (1995) and Shimron (2006) stress the complexity of Modern Hebrew and emphasize that experience with and exposure to oral Hebrew is essential in order for MA to develop. Recent research evidence suggests that from a young age children whose home language (L1) is Hebrew are influenced by their language typology and use its characteristics, including information about the root, as a core morphological entity when they read and spell in Hebrew (e.g., Ben-Dror, Bentin, & Frost, 1995; Gillis & Ravid, 2006; Ravid & Bar-On, 2005). MA is directly related to children’s reading and writing skills in Hebrew (Levin, et al., 1999; Ravid & Schiff, 2004). Of particular relevance to the current chapter is Ravid’s (2001) summary suggesting that in the development of morphology, Hebrew as L1 speaking children first mark inflections (such as gender and number) around the age of 2 years, and that derivations appear between the ages of 3 and 6 years. Further mastery in Hebrew morphology however begins with formal reading and writing instruction (Ravid, 1995). The complex relations between oral language, writing, and morphological skills were explored by Levin et al. (1999). Their findings suggest a “bootstrapping” model, according to which writing skills enhance oral morphology skills and vice versa. They explain that becoming aware of common spelling features of semantically related words contributes to an awareness of the morphemic connection between these words. Awareness of the orthographic code provides children with clues to the morphological infrastructure of Hebrew. Given the complexity of Hebrew morphology it is perhaps not surprising that difficulties with morphological skills are especially detrimental to Hebrew readers (Ben-Dror et al, 1995). Likewise, Ravid, Levie, and Ben-Zvi (2003) report that Hebrew-speaking children who are weak readers have poorer MA skills than their stronger reading peers.
194
E. Geva and D. Shafman
MA: L1 and L2 Relations It is reasonable to assume that conclusions based on research with monolingual children are also relevant to children who develop their vocabulary skills in a second language (L2). Recently, researchers have begun to examine MA in bilingual learners, including English–French learners (e.g., Deacon, Wade-Woolley, & Kirby, 2006), Chinese-English (Wang, Chen, & Chen, 2006), English–Arabic (SaieghHaddad & Geva), Russian–Hebrew (Schwartz, Geva, Share, & Leikin, 2007), Hebrew–English (Kahn-Horwitz, Shimron, & Sparks, 2005; Schiff, & Calif, 2007), and English–Hebrew (Bindman, 2004). Deacon et al. (2006) conclude that MA can be applied cross-linguistically. However, the results of Bindman’s (2004) cross-linguistic study suggest that the nature of the L1–L2 relationships may depend on children’s proficiency in the L2. Saiegh-Haddad and Geva (2007) report that while there were significant correlations between English and Arabic phonological awareness the relationships between morphological skills in L1 and L2 may not be universal. Relatedly, Schiff and Calif (2007) who studied Hebrew–English bilinguals also conclude that the extent of L1–L2 morphological relations may depend on the “language proximity” of the particular language features under study. The differences among studies may be related to methodology, the age of the participants, typological differences, and the extent to which children had an opportunity to develop language proficiency and literacy skills in their respective languages. Unlike phonological awareness, specific aspects of MA may be more sensitive to language-specific skills. In addition, the extent and nature of the relationships among complex L1 and L2 language skills may be mitigated by L1 and L2 language proficiency to varying degrees, and below a certain language proficiency threshold it may not be possible to note cross-linguistic relationships (Saiegh-Haddad & Geva, 2007). Whether the development of morphological skills in children learning Hebrew as a foreign language (HFL) follows the same route as native Hebrew speaking children is an important question both theoretically and practically. It is reasonable to expect that Hebrew morphological skills would be slower to develop in HFL children, and that MA development would be dependent on contextual factors such as exposure, and intra-individual cognitive-linguistic skills. It is also reasonable to expect that the Hebrew oral and spoken language components are interrelated as they are in children whose L1 is Hebrew. From a young age native speakers of Hebrew are influenced by the morphological features of their language and use its characteristics when acquiring reading and spelling skills. This development depends on other linguistic achievements, including phonological awareness, vocabulary learning, and learning to read and spell. In turn, morphological awareness can facilitate higher levels of reading and writing in Hebrew. What about HFL learners who begin to acquire their Hebrew language and literacy skills concurrently at school, in the absence of the same level of exposure that native speakers receive at home and in their community since infancy? In this chapter we begin to explore early steps in the emergence of morphological skills in HFL children, and the extent to which these skills are bootstrapped to other linguistic skills. The present study sought to investigate the development of inflectional morphological skills in primary school children learning
14 Rudiments of Inflectional Morphology Skills
195
Hebrew as a foreign language. Two aspects were of interest: (1) whether different parts of inflectional morphology skills developed sequentially or in parallel, and (2) the extent to which these skills depended on vocabulary and syntactic skills, or whether rudiments of inflectional morphological skills could be observed regardless of Hebrew vocabulary and syntactic knowledge. In particular, we wanted to examine the role of vocabulary and syntactic skills in understanding individual differences in inflectional morphological awareness of beginning HFL learners.
Method Participants Participants were drawn from a private Hebrew day school in Toronto, Canada. A total of 50 children in Grades 1 (N = 19) and 2 (N = 31) participated. The Grade 1 children were drawn from two classes (N = 8 and 11, respectively) totaling 9 males and 10 females, with a mean age of 83.11 months (SD = 2.51). The Grade 2 children were drawn from three classes (N = 14, 7, and 10, respectively) totaling 15 males and 16 females with a mean age of 94.23 months (SD = 3.36). All participants spoke English as their first and home language, and were learning Hebrew as L2 in a bilingual day-school program, in which the required English curriculum is taught for half of a day, and the other half is devoted to the Hebrew language, literacy, and cultural components. Children were exposed to approximately 2½ h of Hebrew per day, 5 days a week (total 12–13 h per week). Most of the exposure to Hebrew took place within and surrounding working with Hebrew text. None of the children used Hebrew as a means of communication at home.
Procedure The data presented here are part of a larger, longitudinal study, currently in progress, and are part of the Time 1 findings. Measures were administered one-on-one in the spring. All instructions for the Hebrew measures were given in English to ensure the child understood what was required. Further, no measure was administered unless it was clear that the child understood the task at hand. Only those with returned parental consent were tested (62% of the Grade 1 students and 47% of the Grade 2 students).
Measures Nonverbal Ability Nonverbal ability was measured using the Matrix Analogies Test-Expanded Form (MAT; Naglieri, 1985). This is a standardized measure that requires the child to point to the missing piece of a design from a group of possible fillers.
196
E. Geva and D. Shafman
Inflectional Morphology Real-word inflections. This task was based on Shatil (1997), and included 14 real, high-frequency Hebrew word pairs. For example, the child would hear: etz, etzim (tree, trees), and would be asked in English “which word says there are many? etz or etzim?” Nonword inflections. This task was adapted from Shatil (1997). It included 14 nonword Hebrew pairs that follow Hebrew inflectional conventions. For example, the child would hear: kaxesh, kaxsha – and then asked “Which word is for a girl? kaxesh or kaxsha” (correct answer: kaxsha). Analogies. An experimental, 18-item, expressive word analogy task, developed by the authors, measured the child’s ability to produce the inflected version of a word, in a manner analogous to a pair presented first. Puppets were used to clarify the task and make it more interesting. For example: (1) adom – adomim; [red(singular) – red (plural)]; yarok -??? [green (singular)]; expected response – yerukim [green (plural)]. (2) rakdanit-rakdan [dancer, (feminine, singular) – dancer (masculine, singular)] paxdanit -??? [someone who is scared (feminine, singular)] – expected response: paxdan [someone who is scared (masculine, singular)]. Analogy items assessed a range of inflectional aspects such as singular–plural nouns, masculine–feminine adjectival inflections, and simple tenses.
Hebrew Language Proficiency Receptive language. A Hebrew adaptation of the standardized Test for Reception of Grammar (TROG) developed by Geva, Wade-Woolley, and Shany (1997) measured comprehension of grammatical contrasts. In this 20-item task, the child hears a sentence, for example, hatzvi ratz (the deer is running) and is asked to point to one of four pictures that best corresponds to the sentence heard. Oral comprehension. A Hebrew oral cloze task developed by Geva and Siegel (2000) measured sentence comprehension, by filling in the missing word in each sentence, e.g., be’xanukah madleekim…? (On xanukah we light…?), with the appropriate answer being either nerot (candles) or xanukiah (menorah). There were a total of 20 items and all were administered. Receptive vocabulary. A standardized Hebrew adaptation of the Peabody Picture Vocabulary Test – Third Edition (Solberg & Nevo, 1979) was used. The child hears a word and then chooses one of four pictures that best corresponds to the word. As the test was designed for native speakers of Hebrew, only the first 39 items were administered to all children. Morphological vocabulary was a receptive vocabulary choice task that measured children’s familiarity with the 14 Hebrew base words in the Real-word inflections task described above. For example, “Does etz mean tree or flower?” Five of the same vocabulary items appeared in the analogies task as well.
14 Rudiments of Inflectional Morphology Skills
197
Results The Development of Morphological Skills Descriptive statistics of the language and morphological tasks are presented separately for Grades 1 and 2, in Table 14.1. The means are slightly higher in Grade 2 showing that there were some gains between Grade 1 and Grade 2 on all aspects of Hebrew as L2 development. A multivariate analysis of variance (MANOVA) was run to determine an overall grade effect. The Wilk’s of .69 only approached significance, F(9, 40) = 2.03, p = .06. Univariate analysis revealed that there was a significant grade effect on the Analogies task, F(1, 48) = 9.573, p < .01. A MANCOVA covarying age in months was not significant either (Wilk’s of .74; F(9, 39) = 1.56, p = .16). Relatedly, the significant univariate grade effect held up for the Analogies task, F(2, 47) = 4.69, p < .01. Regardless of age and grade, there were no significant differences on the other measures, even though consistently, Grade 2 means were somewhat higher than the respective Grade 1 means.
Correlates of Morphological Skills Merging data Correlations among variables, controlling for age, were calculated within each grade. Identical patterns of significant relationships among the variables were displayed, and all variables were significantly intercorrelated (p < .05). The only exception, in both grade levels, was the non-significant correlation Table 14.1 Descriptive statistics for all measures in Grades 1 (N = 19) and 2 (N = 31) Measure Items Grade Mean SD t Receptive language 20 1 14.21 3.14 .06 2 14.16 2.63 Oral comprehension 20 1 3.68 3.61 .97 2 4.71 3.66 Receptive vocabulary 39 1 19.26 6.49 1.16 2 21.39 6.14 Morphological 14 1 12.84 2.29 1.79 vocabulary 2 13.81 1.52 Real-word inflections 14 1 10.79 1.72 1.09 2 11.39 1.98 Nonword inflections 14 1 10.95 1.31 .23 2 11.06 1.97 1 6.84 3.52 Analogies 18 3.09** 2 9.45 2.45 **p < .01
198
E. Geva and D. Shafman
between Real-Word Inflections and Receptive Language. Given this pattern and since the groups did not differ on non-verbal intelligence, and there were no significant differences between grades on most measures, the data of the two grades were amalgamated into one group (N = 50, mean age = 90 months, SD = 6.24 months). This increased the statistical power. Descriptive statistics for the new amalgamated group are presented in Table 14.2, and the correlations appear in Table 14.3. Overall, one should note that all the morphological awareness tasks are highly correlated with each other and with measures of oral language. Again, the only exception is the lack of a significant correlation between the Real-Word Inflections task and the Receptive Language task. Paired samples t-tests examined whether there was a significant difference between performance on the productive, expressive Oral Comprehension measure, and the Receptive Language, both of which had 20 items. Children performed significantly better on the Receptive Language task than on the Oral Comprehension production task, t(49) = 21.96, p < .001. As for the tasks focusing on inflectional morphology, an examination of the percentages in Table 14.2 indicates that the mean scores on the more demanding Analogies task are much lower than the tasks that are more receptive in nature. Paired t-tests revealed the differences between Real-Word Inflections and the Analogies task, and between Nonword Inflections and the Analogies task were significant, t(49) = 12.14, p < .001 and t(49) = 13.57, p < .001, respectively.
Table 14.2 Descriptive statistics for all measures (N = 50) Measure Mean SD Receptive language 14.18 2.80 Oral comprehension 4.32 3.63 Receptive vocabulary 20.58 6.29 Morphological vocabulary 13.44 1.88 Real-word inflections 11.16 1.88 Nonword inflections 11.02 1.73 Analogies 8.46 3.13
Table 14.3 Pearson inter-correlations across all measures (N = 50) Measure 1 2 3 4 1. Receptive language 1 2. Oral comprehension .54** 1 3. Receptive vocabulary .58** .60** 1 4. Morphological vocabulary .53** .29* .46** 1 5. Real-word inflections .16 .49** .32* .38** 6. Nonword inflections .34* .54** .55** .34* 7. Analogies .49** .64** .62** .48** *p < .05 **p < .01
Percentage correct 71 22 53 96 79 78 47
5
6
7
1 .50** .29*
1 .45**
1
14 Rudiments of Inflectional Morphology Skills
199
Predictors of Inflectional Morphology In order to find out whether performance on inflectional morphological tasks can be understood by considering vocabulary and grammatical skills, hierarchical linear regression analyses (Stepwise) were run. For each of the three inflectional morphology tasks, three oral language proficiency tasks (Oral Comprehension, Receptive Vocabulary, and Morphological Vocabulary) were rotated on individual steps as independent variables.1 Regression analyses are presented in Table 14.4. Predictors of Real-Word Inflections. When Morphological Vocabulary was entered first, the Receptive Vocabulary task did not account for unique variance above that provided by the Morphological Vocabulary measure (3%, ns). However, the Morphological Vocabulary measure did account for unique variance above that provided by the Receptive Vocabulary task (6%, p < .05). Oral Comprehension accounted for unique significant variance regardless of the step it was entered on (Step 1 = 24%, p < .001; Step 3 = 14%, p < .05), with only the Morphological Vocabulary measure explaining significant unique variance over and above that explained by Oral Comprehension (6%, p < .05). In other words, Real-Word Inflection was explained uniquely by Oral Comprehension and by Morphological Vocabulary. Predictors of Nonword Inflections. Within these models, the Oral Comprehension task and Receptive Vocabulary measure accounted for significant unique variance, regardless of the step on which they were entered (variance accounted for ranged from 6–29%, p < .05–.01, and 6–31%, p < .05–.01, respectively). The Morphological Vocabulary measure only accounted for significant variance on step 1 in these particular regression models presented, and lost its unique contribution when entered after Receptive Vocabulary (1%, ns) and after Oral Comprehension (3%, ns). Given that the words in this measure are unrelated to the nonwords tested in the measure, this is not surprising. Predictors of the Analogies task. Similar to the Real-Word Inflections task, both the Oral Comprehension and Morphological Vocabulary measures accounted for significant unique variance, regardless of what step they were entered on (variance accounted for ranged from 12 to 41%, p < .01, and 5 to 23%, p < .05–.01, respectively). While the Receptive Vocabulary measure contributed unique significant variance when entered on Step 2 (8–19%, p < .01 respectively), when entered last, it did not account for significant variance above and beyond both Oral Comprehension and Morphological Vocabulary. Given that five of the vocabulary items in this measure appeared in the Morphological Vocabulary measure, this finding highlights the importance of knowing the word being inflected.
Receptive language is not presented within these regression models as it was not significant beyond step 1 and did not change the overall model.
1
*p < .05 **p < .0
3
2 3 1 2
3 1
2
.46
.18 .46 .32 .29
.09 .34
.25
.14
.03 .14 .10 .07
.01 .14
.06
.24 .001 .06
8.91**
1.45 8.91** 5.29* 3.80
.30 7.86**
3.92*
15.03** .05 4.12*
.32
.51 .32 .55 .10
.32 .34
.19
.54 .36 .10
.06
.20 .06 .31 .01
.06 .11
.03
.29 .08 .01
4.75*
13.86** 4.75* 21.17** .56
4.17* 6.05*
2.37
19.40** 6.24** .52
.43
.49 .43 .61 .26
.24 .48
.32
.64 .35 .25
.49 .03 .28
1 2 3
Oral comprehension Receptive vocabulary Morphological vocabulary Morphological vocabulary Receptive vocabulary Morphological vocabulary Receptive vocabulary Oral comprehension Receptive vocabulary Morphological vocabulary Oral comprehension
b
Table 14.4 Predictors of inflectional morphology tasks – multiple regression summary (stepwise) table (N = 50) Real-word inflections Nonword inflections Step Predictor b b R2 change F change R2 change F change
.12
.19 .12 .37 .05
.03 .23
.09
.41 .08 .05
11.46**
15.32** 11.46** 28.07** 4.25*
3.12 14.55**
8.98**
33.48** 7.11** 4.81*
Analogies R2 change F change
200 E. Geva and D. Shafman
14 Rudiments of Inflectional Morphology Skills
201
Discussion Studying morphological skills in beginning L2 learners opens an interesting window into how linguistic skills develop, the relations among them, and the emergence of metalinguistic skills. In this chapter, we began to explore the relations between oral language and aspects of morphological skills in HFL. Clearly, morphological skills, like other language skills, develop as a function of exposure to language and literacy. It is noteworthy that the HFL children who were in Grade 2 had somewhat better vocabulary, morphological and syntactic skills than their counterparts in Grade 1. However, these skills take time to develop, and thus the differences between Grade 1 and Grade 2 did not reach statistical significance on any of the measures except for one of the inflectional morphology tasks. The picture that emerges from the analysis we have carried out so far is that side by side with a growing repertoire of Hebrew vocabulary, grammatical skills and sentence comprehension, rudiments of various inflectional skills are already part of the linguistic skills of primary school children who have had 1–2 years of exposure to Hebrew within a formal classroom context. We also found, as might be expected, that children were more successful on receptive morphology tasks that required children to attend to the inflectional elements and respond to guided questions targeting the functions of those inflections. They had more difficulty however on an expressive task that required them to utilize their morphological awareness and generate a new answer. Children’s performance demonstrated an increasing awareness of some morphological rules, and a growing ability to recognize and isolate morphemes that mark Hebrew inflections.2 Independent of familiarity with the meaning of the stem, children demonstrated an increasing ability to utilize analogies involving cardinal rules for inflections of nouns and verbs, based on gender and number, and apply that knowledge to generate morphologically correct inflections. The errors that children make are highly informative of what they have already learned. For example, children heard the noun-pair beitzim-beitza [eggs, (feminine, plural), egg (feminine, singular)], followed by zeitim- [olives (masculine, plural)]. In this case children were supposed to produce the singular version zayit (olive). Note that the correct response is different from the noun pattern beitza provided in the analogy. A common error was for children to rely on the beitza analogy and produce zayta. In other words, they noted the noun pattern of the example which is typical of feminine nouns in Hebrew that often end in /ah/ and applied it to zeitim. As another illustration, children heard the high frequency verb-pair rotzeh-rotzah [want (masculine, present tense, singular) – want (feminine, present tense, singular)], followed by oxel [eat (masculine, present tense, singular)]. In this case the correct response is oxelet [eat (feminine, present tense, singular)]. Again one of the frequently noted errors was for children to mimic the pattern presented in the first pair, thus offering oxlah.
It should be noted that item analysis based on various morpheme types is currently under way.
2
202
E. Geva and D. Shafman
Typically, in Hebrew, the inflection of masculine nouns from singular to plural involves the addition of /im/, for example, yeled – yeladim (child–children). However, a special pattern applies to nouns that often come in pairs. In this case, / ayim/ is added to the stem, for example, ayin (eye)-eynayim (eyes). When children were given the noun pair ayin-eynayim, followed by /yad/ (hand) many responded with yadim instead of yadayim (hands). It appears that many of these beginning HFL children have already acquired the basic default concatenating rule of adding /im/ to the noun, but they over-generalized this rule to the special case of the “zugiyim” (i.e., the plural noun pairs). As children are working their way towards correctly inflecting words, they tend to apply high frequency, salient, simple inflectional patterns such as those involving male–female nouns, as well as the cardinal manner of marking plurality. As for predictors of emerging inflectional morphology skills, this exploratory study showed that sentence comprehension accounts for unique significant variance in children’s ability to correctly apply rudiments of inflectional morphology.3 This observation is valid whether the task requires children to focus on inflections of real words or nonwords. They are increasingly aware of inflection patterns related to aspects such as male/female nouns, singular/plural nouns (male/female), plural noun pairs, possessive (mine, him, us), and verb inflections (male/female, singular/plural). Children who are better able to comprehend sentences and respond correctly when asked to provide a missing word in sentences, are also more successful on the more analytic morphological inflection tasks targeted in this chapter. Vocabulary knowledge also aids children. Clearly, general vocabulary knowledge is related to various aspects of inflectional morphology. At the same time, as the analysis involving the Real-Word Inflections task indicated, being familiar with the lexical items being tested proved to be more important for the successful inflections of specific words than general vocabulary knowledge. The Nonword Inflections task can be thought of as a metalinguistic application of inflectional rules to unfamiliar words. Findings indicate that general Receptive Vocabulary skills contribute significantly to children’s ability to apply simple inflectional rules, even when the students are not familiar with the meaning of the specific words. General vocabulary skills and general emerging syntactic skills play a pivotal role in the emergence of basic morphological principles in HFL children, though familiarity with specific vocabulary items appears to provide an additional important scaffold for novice Hebrew L2 learners. Jointly these two sets of language skills (syntactic skills and vocabulary) explain between 30 and 40% of the variance on morphological receptive skills, and over 50% of the variance on the expressive morphological Analogies task. It is important to caution, however, that no causal claims can be made, and indeed it is reasonable to hypothesize that over time one would find mutually enhancing relationship between morphological skills, syntactic skills, and vocabulary, such that
We plan to analyze items by category types (e.g., plural, plural pairs, past tense) once data collection is complete.
3
14 Rudiments of Inflectional Morphology Skills
203
those children who have better developed vocabulary and morphological skills are likely better able to comprehend sentences, and vice versa. The current study shows that this observation is applicable to young HFLs just the way it is to young monolingual learners. While it remains to be seen how individual differences on these linguistic measures is related, and perhaps bootstrapped, to the emergence of reading and spelling skills, it is nevertheless noteworthy that elements of inflectional skills are already part of the linguistic repertoire of primary school children who have had 1–2 years of exposure to Hebrew within a formal classroom context. Acknowledgments We wish to thank the staff and children at Associated Hebrew Day School (Neptune Branch) in Toronto for their support and patience. We are also grateful to the University of Toronto for enabling access to undergraduate interns.
References Aitchinson, J. (2003). Review of language change: progress or decay? Journal of Language and Social Psychology, 22, 132–139. Anglin, J. M. (1993). Vocabulary development: a morphological analysis. Monographs of the Society for Research in Child Development, 58, v–165. Aram, D. (2005). Continuity in children’s literacy achievements: a longitudinal perspective from kindergarten to school. First Language, 25, 259–289. Ben-Dror, I., Bentin, S., & Frost, R. (1995). Semantic, phonologic, and morphologic skills in reading disabled and normal children: evidence from perception and production of spoken Hebrew. Reading Research Quarterly, 30, 876–893. Bindman, M. (2004). Grammatical awareness across languages and the role of social context: evidence from English and Hebrew. In T. Nunes & P. Bryant (Eds.), Handbook of children’s literacy. Great Britain: Kluwer Academic Publishers. Carlisle, J. F. (2000). Awareness of the structure and meaning of morphologically complex words: impact on reading. Reading and Writing, 12, 169–190. Carlisle, J. F. (2003). Morphology matters in learning to read: a commentary. Reading Psychology, 24, 291–322. Deacon, S. H., & Kirby, J. R. (2004). Morphological awareness: just “more phonological?” The roles of morphological and phonological awareness in reading development. Applied Psycholinguistics, 25, 223–238. Deacon, S. H., Wade-Woolley, L., & Kirby, J. (2006). Crossover: the role of morphological awareness in French Immersion children’s reading. Developmental Psychology, 43, 723–746. Elbro, C., & Arnbak, E. (1996). The role of morpheme recognition and morphological awareness in dyslexia. Annals of Dyslexia, 46, 209–240. Fowler, A. E., & Liberman, I. Y. (1995). The role of phonology and orthography in morphological awareness. In L. B. Feldman (Ed.), Morphological aspects of language processing. England: Lawrence Erlbaum Associates, Inc. Geva, E., & Siegel, L. S. (2000). Orthographic and cognitive factors in the concurrent development of basic reading skills in two languages. Reading and Writing, 12, 1–30. Geva, E., Wade-Woolley, L., & Shany, M. (1997). Development of reading efficiency in first and second language. Scientific Studies of Reading, 1, 119–144. Gillis, S., & Ravid, D. (2006). Typological effects on spelling development: a crosslinguistic study of Hebrew and Dutch. Journal of Child Language, 33, 621–659. Kahn-Horwitz, J., Shimron, J., & Sparks, R. L. (2005). Predicting foreign language reading achievement in elementary school students. Reading and Writing, 18, 527–558.
204
E. Geva and D. Shafman
Ku, Y., & Anderson, R. C. (2003). Development of morphological awareness in Chinese and English. Reading and Writing: An Interdisciplinary Journal, 16, 399–422. Levin, I., Ravid, D., & Rapaport, S. (1999). Developing morphological awareness and learning to write: a two-way street. In T. Nunes (Ed.), Integrating literacy, research and practice. Dordrecht: Kluwer. Mahony, D., Singson, M., & Mann, V. (2000). Reading ability and sensitivity to morphological relations. Reading and Writing, 12, 191–218. McCarthy, J. (1981). A prosodic theory of non-concatenative morphology. Linguistic Inquiry, 12, 373–418. Muter, V., Hulme, C., Snowling, M. J., & Stevenson, J. (2004). Phonemes, rimes, vocabulary, and grammatical skills as foundations of early reading development: evidence from a longitudinal study. Developmental Psychology, 40, 665–681. Naglieri, J. (1985). Matrix analogies test. New York: Psychological Corporation. Nagy, W., Berninger, V., & Abbot, R. (2006). Contributions of morphology beyond phonology to literacy outcomes of upper elementary and middle school students. Journal of Educational Psychology, 98, 134–147. Ravid, D. (1995). Language change in child and adult Hebrew: a psycholinguistic perspective. Oxford, England: Oxford University Press. Ravid, D. (2001). Learning to spell in Hebrew: phonological and morphological factors. Reading and Writing, 14, 459–485. Ravid, D., & Bar-On, A. (2005). Manipulating written Hebrew roots across development: the interface of semantic, phonological and orthographic factors. Reading and Writing, 18, 231–256. Ravid, D., Levie, R., & Ben-Zvi, G. A. (2003). The role of language typology in linguistic development: implications for the study of language disorders. In Y. Levy & J. Schaeffer (Eds.), Language competence across populations: toward a definition of specific language impairment (pp. 171–196). Mahwah, NJ: Erlbaum. Ravid, D., & Malenky, D. (2001). Awareness of linear and nonlinear morphology in Hebrew: a developmental study. First Language, 21, 25–56. Ravid, D., & Schiff, R. (2004). Learning to represent vowels in written Hebrew: different factors across development. First Language, 24, 185–208. Saiegh-Haddad, E., & Geva, E. (2007). Morphological awareness, phonological awareness, and reading in English–Arabic bilingual children. Reading and Writing, Published online: June 30, 2007. Schiff, R., & Calif, S. (2007). Role of phonological and morphological awareness in L2 oral word reading. Language Learning, 57, 271–298. Schwartz, M., Geva, E., Share, D. L., & Leikin, M. (2007). Learning to read in English as third language: the cross-linguistic transfer of phonological processing skills. Written Language & Literacy, 10, 25–52. Shatil, E. (1997). Predicting reading ability: evidence for cognitive modularity. Unpublished doctoral dissertation. University of Haifa. Shimron, J. (2006). Reading Hebrew: the language and psychology of reading it. Mahwah, NJ: Erlbaum. Singson, M., Mahony, D., & Mann, V. (2000). The relation between reading ability and morphological skills: evidence from derivation suffixes. Reading and Writing, 12, 219–252. Solberg, S., & Nevo, B. (1979). Preliminary steps towards an Israeli standardization of the Peabody Test. Megamoth, 3, 407–413. [Hebrew]. Verhoven, L., & Perfetti, C. (2003). Introduction to this special issue: the role of morphology in learning to read. Scientific Studies of Reading, 7, 209–217. Wang, M., Chen, C., & Chen, S. (2006). Contribution of morphological awareness to ChineseEnglish biliteracy acquisition. Journal of Educational Psychology, 98, 542–553.
Part 4 Teaching and Learning
Chapter 15
Early Identification and Intervention to Prevent Reading Difficulties Orly Lipka and Linda S. Siegel
Abstract This study presents a longitudinal examination of the development of reading and reading-related skills of native English (L1) and non-native English speakers. Reading and related cognitive abilities were examined in children kindergarten and in Grade 5. The analyses were conducted to investigate the influence of the balanced literacy program that was implemented in kindergarten and Grade 1 for the English as a Second Language (ESL) and L1 students in Grade 5. Another aim of the study was to investigate the reading patterns of ESL children in kindergarten and in grade 5 compared to those of their native English-speaking classmates from kindergarten to Grade 5. Finally, we examined how children at risk for reading disabilities (RD) could be identified. The findings provided support for a model of early identification and intervention for all children at risk for reading failure. Furthermore, the results showed that learning English as a second language is not an impediment to successful decoding. In addition, there is a constant need to assess the reading skills of students in the classroom over the elementary years in order to detect difficulties that may emerge when the reading demands change over the years, and different reading strategies are required. Finally, the results demonstrate the heterogeneity of the RD group and call for further longitudinal examination of different RD subgroups.
Early Identification and Intervention to Prevent Reading Difficulties Much progress has been made over the past several decades with regard to understanding and treating students with reading disability (RD). Students with RD suffer mostly from deficits in phonological processing and decoding, and many respond to some form of literacy intervention (such as phonological awareness O. Lipka (*), and L.S. Siegel University of British Columbia, Vancouver, BC, Canada e-mail: [email protected]
D. Aram and O. Korat (eds.), Literacy Development and Enhancement Across Orthographies and Cultures, Literacy Studies 101, DOI 10.1007/978-1-4419-0834-6_15, © Springer Science+Business Media, LLC 2010
205
206
O. Lipka and L.S. Siegel
training). It is crucial to identify children with reading difficulties as early as possible (e.g., Snow, Burns, & Griffin, 1998) because reading failure has a negative impact on achievement in all academic areas as well as in extracurricular activities and peer relations (e.g., Stanovich, 1986). In addition to academic problems, such as grade retention (e.g., McLeskey & Grizzle, 1992) and dropout (e.g., Lichtenstein & Zantol-Wiener, 1988), Adolescents with RD are at an increased risk of developing social problems (e.g., Sabornie, 1994), and emotional difficulties, such as depression (e.g., Gregg, Hoy, King, Moreland, & Jagota, 1992). Also, this population is at risk for problems with self-concept (e.g., Boetsch, Green, & Pennington, 1996), juvenile delinquency, and substance use and abuse (Beitchman, Wilson, Douglas, Young, & Adlaf, 2001). A high prevalence of RD has been identified among adolescent homeless youth and adolescents who have committed suicide (Barwick & Siegel, 1996; McBride & Siegel, 1997). Therefore, providing early identification and intervention is crucial. As a result of the increasing number of immigrants to Canada, there are many children who speak ESL. Specifically, the 2001 Canadian Census data showed that the proportion of Canada’s population born outside the country had reached its highest level in 70 years (Statistics Canada, 2001). ESL students are defined as those whose first language learned and spoken at home with their parents, siblings, and grandparents was not English. The result is an increase in students entering the Canadian school system with limited English-language skills. A number of the component skills necessary for reading and spelling acquisition, both in children with ESL and native English speakers, were investigated in the present study. Specifically, this study focused on phonological processing, syntactic awareness, and working memory – skills that have been demonstrated to increase significantly during the acquisition of reading (e.g., Siegel, 1993). Phonological awareness has been found to be one of the critical skills for learning to read and for fluent reading for native English speaking students from early preschool to university (e.g., Calfee, Lindamood, & Lindamood, 1973). Furthermore, phonological awareness tests have also been reported as good predictors of reading abilities for L1 students (e.g., Gilbertson & Bramlett, 1998; Gottardo, Stanovich, & Siegel, 1996). Phonological awareness skills are essential for ESL students from different native language backgrounds to reach an adequate reading level (e.g., Durgunoglu, Nagy, & Hancin-Bhatt, 1993). In addition to phonological awareness, two other processes, working memory and syntactic awareness, have been linked to the development of reading skills in the English language, and these are disrupted in children with RD (for a review, see Siegel, 1993). Working memory involves the retention of information in short-term storage, while processing involves incoming information and retrieving information from long-term storage (Baddeley, 1983). Working memory is relevant to reading because the reader must decode and/or recognize words while remembering what has already been read and concurrently retrieving information, such as grapheme–phoneme conversion rules (e.g., Baddeley, 1983; Daneman & Carpenter, 1980; Siegel, 1993, 1994; Siegel & Ryan, 1988). In addition, the working memory has a limited capacity, so when there is a greater demand on the executive system,
15 Early Identification and Intervention to Prevent Reading Difficulties
207
there is less processing space and cognitive energy available for subsidiary systems. The literature on learning disabilities provides evidence for the significant role of working memory in reading processes (e.g., Siegel & Ryan, 1989). Very few studies have investigated the working memory of children with ESL. Gholamain and Geva (1999) found that working memory in first and second languages contributed significantly to single-word recognition and pseudoword reading skills in L1 and second language (Gholamain & Geva, 1999). Similar results were found by Da Fontoura and Siegel (1995). In this study, they administered English and Portuguese tasks to Portuguese children who were learning ESL. They found that the working memory task discriminated between readers with RD and typical readers. Syntactic awareness refers to the ability to understand the basic grammatical structure of a language. This skill is critical for fluent and efficient reading of text, which requires word prediction for understanding and is defined as the ability to understand the grammatical structure of a language. A number of studies have found significant difficulties with this skill in poor readers, such as, Siegel and Ryan (1988), who found that students with reading disabilities performed significantly more poorly on a syntactic awareness measure than typical readers. Waltzman and Cairns (2000) demonstrated that poor readers in the third-grade had more trouble with the interpretation of pronouns in some sentence contexts when compared to children with normal reading abilities. Joanisse, Manis, Keating, and Seidenberg (2000) found that dyslexic children of around 8 years old made more errors in inflecting verbs for the past tense than control subjects. Syntactic awareness skill differentiates between language groups. For example, Da Fontoura and Siegel (1995) compared the reading development of Portuguese– Canadian ESL children to L1 students, and no differences were found except on a syntactic awareness task. The Portuguese–Canadian ESL group had significantly lower scores on the English Oral Cloze task than the monolingual L1 students (Da Fontoura & Siegel, 1995). In this task, the examiner read sentences to the students with a missing word in it, and the students were asked to provide the missing word in each sentence. The target words were part of syntactic categories such as past tense, comparative and superlative, conjunctions, prepositions, pronouns, and past participles. Overall, there are few longitudinal studies that have examined reading development among ESL students. There is a special importance in using a longitudinal design for research with ESL students to develop a model that will provide developmental benchmarks (McCardle, Mele-McCarthy, & Leos, 2005). Therefore, the overall goal of this study was to examine the trajectories of reading development of children in Grade 5 who had been followed since kindergarten. The children had different linguistic backgrounds (L1 and ESL), but they were all part of a balanced literacy program that was implemented in their school district in kindergarten and in Grade 1. Specifically, this study was guided by the following three research questions: (1)What is the influence of the balanced literacy program that was implemented in kindergarten and Grade 1 on the ESL and L1 students in Grade 5? (2) How are the reading patterns of ESL children different from those of their native English speaking classmates from kindergarten to Grade 5? (3) How can children at risk for reading difficulties be identified?
208
O. Lipka and L.S. Siegel
Method Participants All the children in this study were tested in the fall of their kindergarten year (when they were 5 years old) and in spring of Grade 5 (when they were 10). Overall, there were 756 children: 635 L1 and 121 ESL Children. Within the ESL group, the mean age in kindergarten was 63.82 months (SD = 3.35 months), and the standard deviation was 3.35. The mean age in Grade 5 was 130.23 months (SD = 3.29 months) and the standard deviation was 3.29 months. Within the L1 group, the mean age in kindergarten was 64.28 months (SD = 3.42 months) and the standard deviation was 3.42, and in Grade 5 the mean age was 130.66 months (SD = 3.70 months) with a standard deviation of 3.70 months. There were 32 languages spoken by the children in this study, and the largest linguistic subgroups were Chinese (Cantonese and Mandarin) and Farsi, followed by Slavic (Russian, Bulgarian, Serbian, Polish, Croatian), Japanese and Korean, and Pilipino and Tagalog. Other languages were Spanish, Hindi, Gujarati, Turkish, Finnish, Afrikaans, Tamil, Italian, French, Dutch, German, Greek, Kurdish, Norwegian, Punjabi, Romanian, and Swedish. The sample included all the schools from one school district in Canada and represented a wide range of socioeconomic backgrounds in the province of British Columbia. Therefore, having a diverse socioeconomic status (SES) group reduced the possibility that the performance of the ESL children as a group was related to a specific SES status. In addition, the correlation of reading skills and SES declined significantly from kindergarten to Grade 3, indicating a positive influence of good schooling in this sample (for further discussion about the SES and reading ability in this district, see D’Angiulli & Siegel, 2004; D’Angiulli, Siegel, & Maggi, 2005).
Kindergarten and Grade 5 Measures We administered a battery of tasks to assess early literacy, phonological processing, spelling, grammatical sensitivity, lexical access, and memory skills to the children in kindergarten. In Grade 5, students were retested on the WRAT Reading subtest, and additional measures were incorporated into the battery to assess cognitive and reading processes.
Reading On the WRAT Reading subtest (blue form: Wilkinson, 1993) each child was asked to name capital letters and to read some simple words. This task was assessed in kindergarten and in Grade 5.
15 Early Identification and Intervention to Prevent Reading Difficulties
209
Letter identification. Each child was presented with 26 written lowercase letters in a random order and was asked to name the letter. This task has a maximum score of 26 and was assessed only in kindergarten. On the Word Identification subtest (Woodcock Reading Mastery Tests-Revised; Woodcock, 1987, W-J word identification, each child in Grade 5 was asked to read aloud a list of words of increasing difficulty. On the Word Attack subtest ((Woodcock Reading Mastery Tests-Revised; Woodcock, 1987), children in Grade 5 were required to decode as many pseudowords as possible from the list. The task administration was discontinued when all items in a given level were failed. On the One-Minute Word Reading test each child in Grade five was presented was presented with a list of words of increasing difficulty and was asked to read as many words as possible within a 1-minute time period. The tan form of the WRAT3: Reading subtest was used to develop the word list.
Reading Comprehension On the Stanford Diagnostic Reading Test (SDRT: Karlsen & Gardner, 1994), each child in Grade 5 was required to read the short passages within a booklet and to provide responses to multiple-choice questions within a prescribed time limit. The test measured students’ reading skills in the areas of vocabulary, comprehension, and scanning for information. This test was group administered.
Spelling In the spelling task in Kindergarten, the children were asked to print their names and five words: mom, no, I, dad, cat. On the WRAT3 Spelling subtest (Wilkinson, 1993), the children in Grade 5 were presented orally with words of increasing difficulty and were required to generate the correct spelling. This subtest was group administered.
Phonological Processing On the Sound Mimicry (Goldman, Fristoe, & Woodcock, 1974), each child in kindergarten was asked to repeat pseudowords of increasing difficulty that had been read to him/her by the examiner. Pseudowords ranged in difficulty from vowel-consonant syllables to polysyllabic pseudowords. On the Rhyme Detection task (Muter, Hulme, & Snowling, 1997) each child in kindergarten was shown four pictures. A picture of the target word appeared above
210
O. Lipka and L.S. Siegel
three pictures. Each child was asked which of the words rhymed with the target word. For example, “What rhymes with boat? Foot, bike, or coat?” There were three demonstration items and ten test items. On the Syllable Identification and Phoneme Identification tasks (Muter et al., 1997) each child in kindergarten was required to complete words. In the syllable identification section, the examiner presented a picture (i.e., table) to the child. The examiner said the first part of the word (i.e., “ta”) and asked the child to complete the word (i.e., “ble”). In the phoneme identification task, the examiner presented a picture (e.g., fish) and said the first part of the word (i.e., “fi”) and asked the child to complete the word (i.e., “sh”). The task consisted of eight syllable identification items, eight phoneme identification items, and two demonstration items for each section. In the Phoneme Deletion task (Muter et al., 1997) the examiner presented the child in kindergarten with a picture of a word, and the child was asked to delete a phoneme (initial or final) from the word. The task consisted of eight initial phoneme deletion items, eight final phoneme deletion items, and four demonstration items for each section. On the Phonological Awareness task, the child in Grade 5 was asked to say a pseudoword and then to say the pseudoword again either without one of its phonemes (e.g., “Say sisp,” “Now say sisp without the /p/ sound”), or without one of its syllables (e.g., “Say conpadly,” “Now say conpadly without the /pad/”). Phonemes and syllables were deleted from the initial, middle, and final positions of pseudowords. There were 30 items arranged in order of difficulty. Administration was discontinued when five consecutive errors were made. Phonological Decoding Fluency in Grade 5 was assessed by the one-minute pseudoword reading test. Each child was presented with a list of pseudowords from the Word Attack subtest of the WRMT-R (Form H; Woodcock, 1987) and was asked to read as many words as possible within a 1-minute time period.
Lexical Access The Rapid Automatized Naming test was used to assess phonological recoding in lexical access or word retrieval in kindergarten children (RAN: Denckla & Rudel, 1976). Each child was asked to name 40 items on a page consisting of line drawings of five different items (i.e., tree, chair, bird, pear, car) repeated in random order eight times. A practice trial of the five items was presented before the presentation of the 40 items to ensure the child knew the target words. The score was the time taken (in seconds) to name the 40 items. The Rapid Automatozed Naming (RAN) test was used in the Grade 5 to assess efficiency in lexical retrieval. The children were requested to name individual numbers (1–9) presented in a random order in a 5 × 5 array. Each child’s performance was timed in seconds.
15 Early Identification and Intervention to Prevent Reading Difficulties
211
Syntactic Awareness The Oral Cloze task was used to assess syntactic awareness skills (Siegel & Ryan, 1988; Willows & Ryan, 1981). In the Kindergarten year, each child was required to listen to the examiner read 12 sentences, each with a missing word, and then to provide a word that created a semantically and syntactically well-formed sentence. The class of the missing word varied as nouns, adjectives, prepositions, and verbs. An example of a sentence was the following: “Dad ________ Bobby a letter several weeks ago.” In Grade 5, 20 sentences were read to the child and the child was asked to provide the missing word in each sentence. Sample sentences include: “Betty ___ a hole with her shovel.”
Memory On the Stanford-Binet Working Memory for Sentences Test (Thorndike, Hagen, & Sattler, 1986) each child in kindergarten was asked to repeat sentences ranging from simple two-word sentences to complex sentences. The task was discontinued when the child failed at least three out of four items in two consecutive levels. Verbal working memory in Grade 5 was measured with the Working Memory for Words task by Siegel and Ryan (1989). The child was orally presented with a set of sentences missing the final words and was asked to provide the missing word for each sentence. The children were then asked to repeat the word that they had provided for the end of each sentence. The number of sentences in each set increased and the total score was 12. Sample sentences include: “Running is fast, walking is ___. At the library people read ___. An apple is red, a banana is ___.” Administration was discontinued when all items on a given level were failed.
District Reading Program In the North Vancouver school district, all children received phonological awareness instruction in kindergarten. The phonological awareness program, “Firm Foundations” (North Vancouver School District No. 44, 2001), was a classroombased program for both L1 and ESL students. The students that were identified as being at risk for reading problems received additional phonological awareness training provided by the classroom and resource teachers in small groups and on an individual basis. This phonological awareness training was based on the prototype of the program, Launch into Reading Success (Bennett & Ottley, 2000). In addition, the “Firm Foundations” program consisted of early literacy skills development, letter–sound relationship, and language development. For instance, small groups and individuals were provided with different activities in a play format such as
212
O. Lipka and L.S. Siegel
rhymes, sound–symbol, early writing activity (journals), and letter identification activities (baking letter-shaped cookies). Overall, the intervention was provided three to four times a week for 20 minutes. The intervention occurred in the context of developing a language and literacy-rich environment with story reading and retelling, journals, and reading children’s books of different levels. In Grades 2–5 the district implemented the Reading 44 program (North Vancouver School District No. 44, 2001), a classroom program that was written by the teachers of North Vancouver. The program included the “Daily Dozen” or 12 reading strategies and instructional activities and graphic organizers for classroom use to encourage students to learn these strategies.
Results In order to investigate the influence of the balanced literacy program that was implemented in kindergarten and Grade 1 on the ESL and L1 students in Grade 5, we conducted several analyses. First, we examined the proportions of ESL and L1 students that were classified as at risk in kindergarten and in Grade 5, then we compared the performance of the two language groups by a series of analysis of variance in kindergarten and in Grade 5. The definition of RD or below-average reading achievement varies across studies; some researchers have used the 20th percentile as a cutoff (e.g., Lovett, Steinbach, & Frijters, 2000), whereas others used the 25th percentile (e.g., Fletcher et al., 1998; Lesaux & Siegel, 2003). Stanovich (1999) suggested a more stringent criterion, such as the 15th percentile, or even the 10th, on nationally standardized measures of reading. We have determined a child to be at risk in kindergarten if their standard score on the WRAT Reading subtest was equal to or less than 85. Scores in the borderline range were standard WRAT Reading subtest between 86 to 94. Children whose score of the WRAT Reading subtest was above 94 were identified as not at risk for reading difficulties. In Grade 5, children were classified as having poor reading skills if their standard score was equal to or less than 85 on either the W–J Word Attack subtest or the WRAT Reading subtest. Students with typical reading skills were defined as students who obtained standard scores equal to or greater than 95 on both measures. Students classified as borderline obtained standard scores between 86 to 94 on either the W–J Word Attack subtest or the WRAT Reading subtest. In kindergarten, within the L1 group, 16.37% of the children were classified as at risk for reading failure, 12.59% were classified as borderline, and 68.64% were classified as not at risk. Similar frequencies were within the ESL group: 20.6% were classified as at risk for reading failure, 7.43% were classified as borderline, and 71.9% were classified as not at risk for reading failure. In Grade 5, 2.99% of the L1 children were identified as reading disabled, 10.70% were classified as borderline, and 86.29% of the L1 children were identified as typical readers. Of the ESL children in Grade 5, 1.65% were identified as reading disabled, 7.43% were identified as borderline, and 90.9% were identified as typical readers. In both
15 Early Identification and Intervention to Prevent Reading Difficulties
213
groups, there were low levels of students with poor reading skills, probably due to the balanced intervention program. Next, we examined the differences between the two language groups: ESL and L1 students in kindergarten by reading status. A series of analysis of variance (ANOVA) for the not at risk children were conducted, with language status (ESL, L1) as a fixed factor and all the kindergarten tasks as dependent measures. In kindergarten, not at risk children with ESL had significantly lower scores than the not at risk L1 children on measures of literacy, phonological processing, syntactic awareness, and memory. The ESL group had significantly lower scores than the L1 group on the Letter Identification task, F(1, 519) = 8.93, p < .01, h2 = .008 and simple spelling measure, F(1, 519) = 5.96, p < .05, h2 = .01, on two of the phonological processing tasks, the Sound Mimicry measure, F(1, 519) = 6.55, p < .05, h2 = .01, and the Rhyme Detection task: F(1, 519) = 13.28, p < .001, h2 = .01, on the Oral Cloze task: F(1, 519) = 10.30, p < .02, h2 = .017, and on the Memory for Sentences task, F(1, 519) = 37.89, p < .001, h2 = .06. We examined the differences between the at risk groups in Kindergarten. A series of ANOVA were conducted, with language status (ESL, L1) as fixed factors and all the kindergarten tasks as dependent measures. The at risk children in kindergarten from both language groups showed very similar performance on the literacy, phonological processing, lexical access, syntactic awareness, and memory measures. In Grade 5 we examined the reading and cognitive skills of the L1 and ESL children. Since there were only 21 RD students (2 ESL and 19 L1), separate analyses were not conducted for the poor readers, and only the typical readers were included in the analysis. The ESL typical readers group in Grade 5 performed in a similar manner to the L1 English typical readers group on most of the measures such as reading, phonological processing, verbal working memory, and lexical access. Specifically, the L1 and the ESL groups did not have significantly different scores on the WRAT Reading subtest, F(1, 657)=.40, ns, h2 = .00, the W–J Word Identification subtest, F(1, 657) = .17, ns, h2 = .00, the W–J Word Attack subtest, F(1, 657) = .02, ns, h2 = .00, and the Word Reading Fluency test, F(1, 657) = .01, ns, h2 = .00. Also on the phonological processing tasks there were no differences between the L1 and the ESL typical achieving readers on the pseudoword deletion tasks, F(1, 657) = 3.08, ns, h2 = .00 and on the Phonological Decoding Fluency task, F(1, 657) = .01, ns, h2 = .00. In addition, there were no statistically significant differences between the groups on the Rapid Automatized Naming task, F(1, 657) = .71, ns, h2 = .00, and on the Working Memory for Words task, F(1,657) = .1.46, ns, h2 = .00. There were main differences between the groups on three measures. The ESL group had significantly lower scores than the L1 English group on two tasks, the Oral Cloze task, F(1, 657) = 9.28, p < .01, h2 = .01 and on the SDRT reading comprehension task, F(1, 657) = 7.03, p < .001, h2 = .01. However, the ESL group had significantly higher scores than the L1 group on the WRAT Spelling subtest, F(1, 657) = 4.64, p < .05, h2 = .00.
214
O. Lipka and L.S. Siegel
In order to investigate how children at risk for reading difficulties can be identified, we examined the various trajectories (i.e., children who began at risk and remained at risk, children whose risk status had fluctuated over the years) of word reading development within the RD group in Grade 5. We found that there were three distinct subgroups within the Grade 5 RD group. The first subgroup consisted of 12 students who demonstrated poor performance (i.e., below the standard score of 85) in kindergarten on the WRAT Reading subtest as well as poor performance either on the WRAT Reading subtest or the W–J Word Attack in Grade 5; we refer to this subgroup as poor readers (PR). The second group consisted of three students who were borderline on the WRAT Reading subtest in kindergarten or the W–J Word Attack subtest. We referred red to this group as borderline RD. The third subgroup consisted of six students who started as not at risk on the WRAT Reading subtest in kindergarten; but in Grade 5, were classified as RD on either the WRAT Reading subtest or the W–J Word Attack subtest. We refer to this subgroup as not at risk RD. In order to examine the profiles of the three RD subgroups, we examined their mean scores in kindergarten and in Grade 5. These profiles were of particular interest, since these students unlike their peers, did not benefit from the balanced literacy program as the other students in the district did, and probably needed more support in other areas. For this analysis, scores that were at least one below 1 standard deviation (SD) below the not at risk group were described as marked difference. In kindergarten, the poor reader group performed 2 SD below the typical L1 group on three measures: percentile, the WRAT Reading, and letter identification. This group performed more than 1 SD below the typical L1 group in the memory for sentences test. The Borderline group performed more than 1 SD below the typical L1 group on 4 measures: WRAT Reading, Letter Identification, percentile, and RAN time. The not at risk RD group performed more than 1 SD below the typical L1 group on two measures in kindergarten: percentile and memory for words, demonstrating average word reading and letter identification skills in kindergarten. In Grade 5, the poor readers group performed more than 3 SD below the typical L1 group on WRAT Reading, and pseudoword reading fluency. This group performed more than 2.5 SD below the typical L1 group on W–J Word Identification and Word Attack, and on the reading comprehension test. They performed more than 2 SD below the typical L1 group on the word reading fluency, and spelling, and more than 1 SD below the typical group on the RAN, pseudoword deletion, working memory and the syntactic awareness tasks. The profile of this group demonstrated severe difficulty with most of the reading and cognitive skills in Grade 5. The borderline-RD group performed more than 3 SD below the typical L1 group on the WRAT Reading, and below 2.5 SD on W–J Word Identification, spelling, and word reading fluency. This group performed more than 2 SD below the typical L1 group on W–J Word Attack, word reading fluency, lexical access, and reading comprehension. They performed more than 1 SD below the typical L1 group on oral cloze and 1.5 SD below the typical L1 group on pseudoword deletion. The not at risk RD group has scores that were 2.5 SD below the typical L1 group on the WRAT Reading, W–J Word Attack, and the pseudoword fluency. They performed more than 2 SD below the typical L1 group on the W–J Word Identification and
15 Early Identification and Intervention to Prevent Reading Difficulties
215
word reading fluency. The not at risk RD performed 1.5 SD below the typical L1 group on the WRAT Spelling and more than 1 SD on lexical access, syntactic awareness, and reading comprehension. This group performed within the average range on the working memory measure.
Discussion The first goal of the study was to examine the influence of the balanced literacy program that was implemented in kindergarten and Grade 1 on the performance of ESL and L1 students now in Grade 5. There was a significant decrease in the number of children with reading difficulties from kindergarten to Grade 5. For the majority of children who experienced early reading difficulties in kindergarten, their difficulties were likely remediated through a balanced early reading program that included small group phonological awareness instruction for all children, as well as phonics instruction in Grade 1 and reading comprehension strategies. Many of the at risk children received targeted, direct phonological awareness instruction in small groups in Grade 1 as well. This model provided considerable support for the benefits of small group instruction in kindergarten, and a balanced approach to literacy activities in order to reduce the incidence of reading failure, for both native English speakers and students with ESL. Another objective of this study was to investigate the reading patterns of ESL children compared to their native English-speaking classmates from kindergarten to Grade 5. Not surprisingly, in kindergarten, the children with ESL did not perform as well as L1 speakers on most of the measures. Specifically, the children with ESL performed more poorly than the L1 English children on some of the phonological processing measures and in the areas of, syntactic awareness, memory for sentences and spelling. Since all these tasks involved language proficiency. These tasks required the children to manipulate and remember English, and proved difficult for all ESL-speaking children as compared to their native English-speaking peers. By Grade 5, however, the students with ESL were performing at similar levels to their L1 English-speaking peers. In spite of their later exposure to the phonological structures of the English language, ESL speakers were as successful as the L1 speakers at manipulating the sounds of the English language (i.e., phonological awareness) and at applying grapheme–phoneme mapping rules to read unfamiliar words with ease and speed (i.e., phonological decoding fluency). However, the ESL speakers still lagged behind L1 speakers in understanding the grammatical structures of oral language, even at the upper-elementary grades. Furthermore, the L1 speakers responded correctly to a greater number of questions than the ESL speakers on the SDRT reading comprehension test. This finding is consistent with Verhoeven’s (1990) study of bilingual Turkish-Dutch children. Even after 20 months of literacy instruction, the reading comprehension performance of the Turkish-Dutch children was lower than the native Dutch-speaking children despite similar word reading skills. The author suggested that the finding was likely due to
216
O. Lipka and L.S. Siegel
syntactic ability and oral language proficiency. Similarly with the present study, the ESL speakers showed comparable performance on word reading, lower performance on SDRT reading comprehension, and lower performance on the Oral Cloze task. It is important to note that the SDRT mean scores for both groups were well within the average range, suggesting that ESL speakers were not at a disadvantage according to the normative criterion of the test. The children with ESL performed better than L1 English children on the WRAT Spelling subtest that required accurate spelling of real words. The spelling process for real words refers to words that have usually been seen in print in the past and thus have an orthographic representation in a sight word vocabulary. A suggested explanation is that children with ESL have metalinguistic abilities due to their exposure to different language systems, and this metalinguistic ability assisted them when learning to spell new language units. Overall, we found that ESL students can perform in a similar way to L1 students on measures of reading and cognitive abilities after 6 years in English schooling, utilizing a balanced literacy program. We also found that ESL students had an advantage in spelling; future research can examine in more details whether ESL students in general or only specific ESL samples have this advantage. ESL students in Grade 5 performed poorly on syntactic awareness skills, and reading comprehension skills as compared to their peers, although their reading comprehension ability on average fell within the normal range. Future studies should examine the reading comprehension skills of ESL students in upper grades in relation to different comprehension abilities, as well as different types of texts. Another aim of this study was to examine the different trajectories of the students that were identified as RD in Grade 5. We found three different profiles of students with RD: Poor readers, Borderline RD, and not at risk RD. In kindergarten, five measures differed between the RD groups: The phonological awareness measure, the word/letter reading measure: WRAT Reading and letter identification, the memory for sentence measure, and the lexical access measure, RAN. The poor reader group performed more poorly than the not at risk readers on the phonological awareness, word reading, memory for sentence and letter identification. The borderline group performed more poorly than the not at risk readers on word reading, memory for sentences and letter identification measures, phonological awareness, and lexical access. The third group that was not identified as being at risk in kindergarten based on the WRAT Reading, but ended up as RD in Grade 5, performed more poorly than the not at risk children in kindergarten on two measures: phonological awareness and memory for words measure. These results indicate that there is a need to assess students in kindergarten with these measures and to take into consideration the students that perform at least 1 SD below the typical L1 on at least two measures. In Grade 5, the three RD groups performed more poorly than the typical readers on most of the measures, indicating the ability of the measures to identify students with difficulties, but also the severity of the RD. The poor reader group had the poorer performances, on most of the measures, followed by the borderline RD group and the not at risk RD group.
15 Early Identification and Intervention to Prevent Reading Difficulties
217
It is of note that in our sample, we had two ESL students who were classified as RD in Grade 5. The percentage of the students with ESL in the overall RD group did not allow us to examine different trajectories or RD between language groups. One implication of our study concerns early intervention programs. The school district was committed to a balanced literacy program, again demonstrating the critical role that phonological processing plays in reading acquisition for both native and non-native speakers. The findings indicate the importance of phonological skills to facilitate reading development. The phoneme awareness-training program in kindergarten and Grade 1 seems to have had a benefit as it helped to bridge the gap in the reading skills of children with ESL entering kindergarten with limited English or no English exposure. If detection and remediation are available for children with ESL during the early years of school, their reading development can be similar to their native English-speaking peers in later grades. However, it is necessary to provide direct instruction for reading comprehension and of syntactic skills for the ESL children. These results indicate several issues that pertain to early identification and reading development: First, there is a constant need to assess the reading skills of students in the classroom over the elementary years in order to detect difficulties that may emerge over the years when the reading demands change and different reading strategies are required. Second, early identification and monitoring should include measures that assess phonological processing, word reading, lexical access, and memory. And finally, the results demonstrate the heterogeneity of the RD group, and call for further longitudinal examination of different RD subgroups. Future research on the development of English language skills in ESL students from different language backgrounds should include a focus on transfer between the first and second languages, the special characteristics of each language system, and the interplay between them. In addition, future research should consider such variables as the age of first exposure to English, literacy instructional methods, the proportion of ESL students in the classroom in which the child is being educated, and the specific characteristics of the first language of the student. Whenever possible, it is important to consider language and reading skills in the first language. Acknowledgments This research was supported by grants from the Natural Sciences and Engineering Research Council of Canada and the Canadian Language and Literacy Research Network to Linda S. Siegel. The authors wish to thank the students, principals, teachers, parents, and administrators in the North Vancouver school district for their invaluable contributions and support.
References Baddeley, A. D. (1983). Working memory. Philosophical Transactions of the Royal Society of London, 302, 311–324. Barwick, M., & Siegel, L. S. (1996). Learning difficulties in adolescent clients of a shelter for runaway and homeless street youth. Journal of Research on Adolescence, 6, 649–670.
218
O. Lipka and L.S. Siegel
Beitchman, J. H., Wilson, B., Douglas, L., Young, A., & Adlaf, E. (2001). Substance use disorders in young adults with and without LD: predictive and concurrent relationships. Journal of Learning Disabilities, 34, 317–332. Bennett, L., & Ottley, P. (2000). Launch into reading success through phonological awareness training. Austin, TX: Pro-Ed. Boetsch, E. A., Green, P. A., & Pennington, B. F. (1996). Psychosocial correlates of dyslexia across the life span. Development and Psychopathology, 8, 536–539. Calfee, R., Lindamood, C., & Lindamood, P. (1973). Acoustic–phonetic skills and reading-kindergarten through twelfth grade. Journal of Educational Psychology, 64, 293–298. Da Fontoura, H. A., & Siegel, L. S. (1995). Reading, syntactic and working memory skills of bilingual Portuguese–English Canadian children. Reading and Writing: An Interdisciplinary Journal, 7, 139–153. D’Angiulli, A., & Siegel, L. S. (2004). Early literacy instruction, SES, and reading development in English language learners and children with English as a first language. Learning Disabilities Research & Practice, 19, 202–213. D’Angiulli, A., Siegel, L. S., & Maggi, S. (2005). Literacy instruction, SES, and word-reading achievement in English-language learners and children with English as a first language: a longitudinal study. Learning Disabilities Research & Practice, 19, 202–213. Daneman, M., & Carpenter, P. A. (1980). Individual differences in working memory and reading. Journal of Verbal Learning and Verbal Behavior, 19, 450–466. Denckla, M., & Rudel, R. G. (1976). Rapid “automatized” naming (R.A.N.): dyslexia differentiated from other learning disabilities. Neuropsychologia, 14, 471–479. Durgunoglu, A. Y., Nagy, W. E., & Hancin-Bhatt, B. J. (1993). Cross-language transfer of phonological awareness. Journal of Educational Psychology, 85, 453–465. Fletcher, J. M., Francis, D. J., Shaywitz, S. E., Lyon, G. R., Foorman, B. R., Stuebing, K. K., et al. (1998). Intelligent testing and the discrepancy model for children with learning disabilities. Learning Disabilities Research & Practice, 13, 186–203. Gholamain, M., & Geva, E. (1999). Orthographic and cognitive factors in the concurrent development of basic reading skills in English and Persian. Language Learning, 49, 183–217. Gilbertson, M., & Bramlett, R. K. (1998). Phonological awareness screening to identify at risk readers: implications for practitioners. Language, Speech, and Hearing Services in Schools, 29, 109–116. Goldman, R., Fristoe, M., & Woodcock, R. (1974). Goldman–Fristoe–Woodcock Auditory Skills Test Battery. Circle Pines, MN: American Guidance Service. Gottardo, A., Stanovich, K. E., & Siegel, L. S. (1996). The relationship between phonological sensitivity, syntactic processing, and verbal working memory in reading performance of third grade children. Journal of Experimental Child Psychology, 63, 563–582. Gregg, N., Hoy, C., King, M., Moreland, C., & Jagota, M. (1992). The MMPI-2 profiles of adults with learning disabilities in university and rehabilitation settings. Journal of Learning Disabilities, 25, 386–395. Joanisse, M. F., Manis, F. R., Keating, P., & Seidenberg, M. S. (2000). Language deficits in dyslexic children: speech perception, phonology, and morphology. Journal of Experimental Child Psychology, 77, 30–60. Karlsen, B., & Gardner, E. (1994). Stanford diagnostic test. San Francisco: Harcourt Brace. Lesaux, N. K., & Siegel, L. S. (2003). The development of reading in children who speak English as a second language. Developmental Psychology, 39, 1005–1019. Lichtenstein, S., & Zantol-Wiener, K. (1988). Special education dropouts (ERIC Digest No. 451). Reston, VA: ERIC Clearinghouse on Handicapped and Gifted Children. (ERIC Document Reproduction Service No. ED 295 395). Lovett, M. W., Steinbach, K. A., & Frijters, J. C. (2000). Remediating the core deficits of developmental reading disability: a double-deficit perspective. Journal of Learning Disabilities, 33, 334–358. McBride, H., & Siegel, L. S. (1997). Learning disabilities and adolescent suicide. Journal of Learning Disabilities, 30, 652–659.
15 Early Identification and Intervention to Prevent Reading Difficulties
219
McCardle, P., Mele-McCarthy, J., & Leos, K. (2005). English language learners and learning disabilities: research agenda and implications for practice. Learning Disability Research &Practice, 20, 68–78. McLeskey, J., & Grizzle, K. L. (1992). Grade retention rates among students with learning disabilities. Exceptional Children, 58, 548–554. Muter, V., Hulme, C., & Snowling, M. (1997). The phonological abilities test. London: Psychological Corporation. North Vancouver School District No. 44. (2001). Firm foundations: early literacy teaching and training. North Vancouver, BC, Canada: Author. Sabornie, E. J. (1994). Social-affective characteristics in early adolescents identified as learning disabled and nondisabled. Learning Disability Quarterly, 17, 268–279. Siegel, L. S. (1993). The development of reading. In H. W. Reese (Ed.), Advances in child development and behavior (Vol. 24, pp. 63–97). San Diego, CA: Academic. Siegel, L. S. (1994). Phonological processing deficits as the basis of dyslexia: implications for remediation. In M. J. Riddoch & G. W. Humphreys (Eds.), Cognitive neuropsychology and cognitive rehabilitation (pp. 379–400). Hove, UK: Lawrence Erlbaum Associates, Inc. Siegel, L. S., & Ryan, E. B. (1988). Development of grammatical-sensitivity, phonological and short-term memory skills in normally achieving and subtypes of learning disabled children. Developmental Psychology, 24, 28–37. Siegel, L. S., & Ryan, E. B. (1989). The development of working memory in normally achieving and subtypes of learning disabled children. Child Development, 60, 973–980. Snow, C., Burns, M., & Griffin, P. (eds). (1998). Preventing reading difficulties in young children. Washington, DC: National Academy Press. Stanovich, K. E. (1986). Matthew effects in reading: some consequences of individual differences in the acquisition of literacy. Reading Research Quarterly, 21, 360–407. Stanovich, K. E. (1999). The sociopsychometrics of learning disabilities. Journal of Learning Disabilities, 32, 350–361. Statistics Canada. (2001). 2001 census. Ottawa, ON, Canada: Author. Thorndike, R. L., Hagen, R. P., & Sattler, J. M. (1986). Technical manual: Stanford Binet Intelligence Scale (4th ed.). Chicago: Riverside. Verhoeven, L. (1990). Acquisition of reading in a second language. Reading Research Quarterly, 25, 90–114. Waltzman, D., & Cairns, H. (2000). Grammatical knowledge of third grade good and poor readers. Applied Psycholinguistics, 21, 263–284. Wilkinson, G. S. (1993). The wide range achievement test-3. Wilmington, DE: Jastak Associates. Willows, D. M., & Ryan, E. B. (1981). Differential utilization of syntactic and semantic information by skilled and less skilled readers in the intermediate grades. Journal of Educational Psychology, 73, 607–615. Woodcock, R. (1987). Woodcock reading mastery tests – revised. Circle Pines, MN: American Guidance Service. Woodcock, R., & Johnson, M. B. (1989). Woodcock-Johnson psycho-educational battery-revised. Itasca IL: Riverside Publishing.
Chapter 16
Whatever Happened to Pedagogical Theory? David R. Olson
Abstract This chapter makes the controvertial claim that the skills that distinguish readers from non-readers provide an unreliable guide as to what should be taught. Furthermore, such research has inadvertently encouraged an extremely teacher centered pedagogy in many Western countries especially the United States and Britain. It traces this development to Thorndike’s behavioristic assumptions about teaching and learning. A more general theory of pedagogy makes a sharp distinction between teaching and learning and explores the options available to educators as they attempt to reconcile the agency of the learner, that is, what the learner is trying to do and willing to do, with the goals and standards of the teacher, that is, what the teacher hopes to help the learners achieve. It is argued that through the negotiation of shared goals and standards learners can come to take some responsibility for their own learning. The teacher’s responsibility is to help them do so. Pedagogy, the enterprise of turning teaching into a psychological theory, died early in the twentieth century when theories of teaching gave way to the theories of learning, particularly in the hands of Thorndike (1913). While Thorndike’s learning theory has long since been abandoned, his tests and measurement approach to the study of both teaching and learning, remain the backbone of educational research. My purpose in this study is to resurrect the theory of teaching by stepping back from the century long concern with learning (see also Shulman, 1990). Episodes of teaching are easy to recognize, but difficult to define, let alone assess. Barzun (1945, p. 23), an astute observer, described teaching in this way: The teaching impulse goes something like this: a fellow human being is puzzled or stymied. He wants to open a door or spell ‘accommodate.’ The would be helper has two choices. He can open the door, spell the word; or he can show the pupil how to do it for himself. The second way is harder and takes more time, but a strong instinct in the ‘born teacher’ makes him prefer it (Barzun, 1945, p. 23).
D.R. Olson (*) University of Toronto, Toronto, ON, Canada e-mail: [email protected]
D. Aram and O. Korat (eds.), Literacy Development and Enhancement Across Orthographies and Cultures, Literacy Studies 101, DOI 10.1007/978-1-4419-0834-6_16, © Springer Science+Business Media, LLC 2010
223
224
D.R. Olson
In fact teaching is a relatively straightforward matter when the goal originates with the child: “He wants to spell ‘accommodate’” for then the teacher can help. But when the goal originates with the teacher, that is, when it is the teacher who wants the child to learn to spell “accommodate,” the traditional dilemmas of teaching arise. The learner may not want to spell “accommodate,” he or she may be willing to accept alternative spellings, and may be uninterested in the general orthographic spelling rules that raise teaching to Barzun’s standard. Then the teacher’s job is both to create the desire or intention, and help the learner to realize it. Once the desire or intention, however modified, is in place, Barzun’s account of teaching is more or less complete and in accordance with the philosophical analysis of the concept. Teaching involves intention on the part of the teacher to have the learner’s behavior meet a goal or standard. However, on its own this principle is insufficient, in that teaching would then be no different from animal training in which behavior is shaped through reinforcement to meet a standard. Training may be seen as a kind of teaching but fails to capture its essence. A second level of intention is therefore involved, namely, the intention of the learner to meet the goal or standard. This would require that the learner have some notion of the goal or standard. This, in turn, would allow the learner to monitor his or her own learning to see that it meets the goal or standard, the correct spelling, to advert to Barzun’s example. Meeting the standard entitles the learner to new opportunities as well as earning some autonomy from the controlling effects of the teacher. And finally, as Barzun noted, teaching does not merely impose the correct solution but has some depth, involving a principle, rule or set of reasons appropriate to setting, achieving and monitoring the standard. This is what distinguishes teaching from indoctrination. As an aside, it is worth noting that the principle or rule is adduced from the specific examples and is not directly taught and applied, as has become common in reading pedagogy. Hirst (1973, p. 171) summarized the concept of teaching in a way congruent with this focus on intention: A teaching activity is the activity of a person, A (the teacher), the intention of which is to bring about an activity (learning), by a person, B (the pupil), the intention of which is to achieve some end state (e.g., knowing, appreciating) whose object is X (e.g., a belief, attitude, skill). We could formalize the account of teaching in this way: 1. Outcome – goal or standard 2. Process – involves learner’s intention to bring about 1 3. Process – involves teacher’s intention to bring about 2 The scheme seems sufficiently flexible to allow for both teacher-initiated and student-initiated activity, as well as for the pursuit of a common goal that has been described as involving joint intentions (Bruner, 1966; Olson, 2007). At the same time, it leaves completely open how the intention in (2) and (3) is to be carried out in practice. These practices constitute what are commonly thought of as “methods,” options available to the teacher to bring about the relevant intentions and goals.
16 Whatever Happened to Pedagogical Theory?
225
The inability of philosophers to set out just how any of these intentions were to be fulfilled, that is to provide any guidance as to how teachers are to achieve their goals, has left psychologists and educators free to follow Thorndike’s lead and to abandon altogether the analysis of teaching, to focus rather on learning. Indeed, in its most extreme form, behaviorists claimed that if there was no learning there was no teaching, a mantra sometimes taken over by educators themselves. For these and other reasons, the topic of teaching has yielded pride of place to the topic of learning. Learning is both desirable, subject to experimental analysis, and assessable by tests and is therefore the topic of choice in an era of accountability.
Method as “What Works” As noted, philosophers have been less helpful in developing a theory of method, leaving that rather to the empirical efforts of psychologists and educators. Philosophers have by and large retreated to the task of discussing the goals of education, leaving the concern with method to the empirical sciences. The very concept of “method” assumes that method is independent of the goals to be achieved; goals and methods have come to be separate domains of enquiry. The contemporary attempt by legislators and some professional educators to improve the degree of student learning has fueled the search for the best method for delivering the goals specified by the authorities. The range of these goals, whether to include both facts and theories, certainties and opinions, sciences or humanities, is left to the planners. The job of the school, and the psychologists who study it, is seen as one of determining optimal means for achieving the mandated goals, to find out what works or at least works best. The attempt to empirically determine “what works” is subject to the criticism that it perhaps makes an overly sharp distinction between goals and methods. And second, it assumes a generalizability that may not be warranted; what works in one context, with one set of students, one set of teachers, one set of parents may not work in another. And third, it leaves the choice of method to completely ad hoc trial and error rather than being guided by a theory. Yet, a theory is necessary if we are to move beyond mere trial and error, that is, anything that on average works better than its copyrighted competitors, and into the actual experimentation necessary to advance a science. The problem then arises as to whether or not one could construct a theory that could provide some guidance to experimentation and ultimately to practice. It should be acknowledged as obvious that some teaching practices are more successful than others at least in some contexts for some learners for some types of goals. This is the commonsense knowledge of all practicing educators. The problem, however, is that there is no clear conception of what the options are, among which teachers must choose. Candidates are so numerous as to be overwhelming: drill and practice vs. explanations; reading and study vs. explicit teaching; discovery vs. didacticism; tutorial vs. group lecture; note taking vs. engaged listening; teaching principles vs. teaching clear cases; rigid planning vs. seizing “teachable
226
D.R. Olson
moments,” and so on. Yet when subjected to well-controlled experimental study, such variables have yielded disappointingly little. Results may be summarized by saying that the differences are negligible, inconsistent, but, as Gage and Berliner (1975) wryly noted, the group labeled “experimental” tended to outperform the group labeled “control.” This is not to say that no attempts have been made to advance a theory of method. I will set out three: one from the nineteenth century and two from the twentieth. The most famous nineteenth-century attempt was that of Herbart (Dunkel, 1970) who argued that pedagogy could be turned into a science by setting out the critical episodes in teaching in four stages, first, clarification, that is, turning experiences into clear concepts; second, associating these new concepts with old ones; third, organizing these concepts into systems; and fourth, methodizing the concepts into useful practices. Hence, the component processes in pedagogy consisted of clarification, association, systematization, and methodization. Although Herbart advised against too rigid an imposition of these steps, within a generation it became as rigid and mechanical as the classical methods it had sought to replace and would later become the object of much of Dewey’s criticism. Yet, something like Herbart’s method is involved in all explicit teaching and, indeed, finds a reprise in the lesson plans of most teachers. What is clear is that the theory has more to do with the traditional goals of conveying or transmitting knowledge than in increasing the reflectiveness and autonomy of the learners, what were to become the overriding concern of twentieth-century reformers, the Progressives led by Dewey and later the Cognitivists led by Bruner. Dewey (1976) disparaged “method,” leaving choices of the actions of the teacher to general considerations of the aims of education with due attention to local opportunities for action and discourse. Followers favored the Project method in which students with the help of the teacher would take on a project such as growing a garden or constructing a series of maps depicting the known world or experimenting with shadows or with gravity. But as Lagemann (2000) has shown Dewey’s proposals were largely set aside in favor of Thorndike’s behaviorism. As mentioned it was Thorndike’s behaviorism that led to the replacement of a theory of teaching by the theory of learning with learning being assessed by standardized tests. Bruner (1966) took the step of reviving pedagogical theory in terms of a theory of instruction. He set out four principles of instruction that he had found to be particularly useful in designing a curricular project Man: a course of study. These principles were seen as applicable to any discipline or domain. What was to be explained was how teaching takes or fails to take its effect. The first of these principles spelled out the importance of isolating and activating the predispositions of the child that are relevant to the particular learning task. Second, the theory would specify how a body of knowledge should be organized so that it could be readily grasped by a learner. This body of knowledge was to be thought of as a “structure” or “generic coding system” or general theory that gave meaning to the particular facts involved and that would permit generalization. Such generic codes would include such things as tropisms in biology, forces and causes in physics, intentions in the social sciences, functions in mathematics, and so on. Third, the theory would set out appropriate sequencing to allow the grasp of “structure” and
16 Whatever Happened to Pedagogical Theory?
227
invite inference to further facts. And finally, it would spell out the rewards or entitlements earned by mastering the domain. The heart of the instructional process, the method most appropriate to meeting the requirements of the theory of instruction, Bruner proposed, was the process of discovery. He described discovery in several ways such as teaching in the hypothetical rather than the declarative mode or as “teaching the hypothetical.” “It is far more interesting to learn the facts after one has tried to figure them out for oneself” (Bruner, 1966, pp. 94–95). This is the art of getting and using information to test out one’s hunches and conjectures, what Bruner described as cognitive strategies that permitted one to “go beyond the information given.” Dewey’s Project method and Bruner’s Discovery method have found a limited place in the school, but they remain secondary to the more traditional methods of didactic, expository teaching supplemented by question and answer routines, discussion, reading, and homework and standardized tests. Indeed, these options are so large and vaguely defined that they may be more appropriately thought of as techniques or strategies contingently applied than as stipulated pedagogical methods. If a method denotes a configuration of variables around a common principle or theme, then traditional expository methods may usefully contrast with more studentcentered methods, and this distinction has invited some interesting research. One large-scale empirical study of method that did have some initial plausibility was the so-called “Eight year study” conducted in the 1930s by a team led by Ralph Tyler (Cremin, 1961, p. 255; Lagemann, 2000, p. 141) that was designed to compare classical, didactic methods with the then more revolutionary methods of Progressivism. Thirty public and private alternative high schools implemented a progressivist curriculum over an 8-year period. Subject areas were integrated and an emphasis was placed on self-initiated projects, on the arts, and on community involvement. Performance of the graduates of these programs at college entry was then compared to that of matched samples of students graduating from regular high schools. Extensive analysis of the data showed that performance differences were small with some marginal advantage to the “experimental” groups, who, in addition, were found to be more engaged in social and artistic activities. A decade after the end of the experiment, little trace of these reforms remained (Tyack & Cuban, 1995, p. 100). Ravitch (1985, p. 22), in her analysis of the history of educational reform, found that attempts to make schooling more relevant and child centered through adding values education and vocational education, as well as through dissolving basic disciplines such as history and geography into “social studies,” contributed little to achieving higher levels of knowledge and character development and may have even interfered with the achievement of the school’s more intellectual goals. If we may summarize these bands of research in terms of methods, we may conclude that even such generic methods as “traditional” versus “progressive” are far from discrete in conception or clear in outcome. With such discouraging results, it is somewhat understandable that researchers have simply opted for an unprincipled search for “what works.” If clear alternatives cannot be formulated, why not, the argument goes, simply try anything that comes to mind and see what works best. And especially if these ad hoc methods are at least indirectly informed by recent developments in cognitive developmental theory, the
228
D.R. Olson
chance of hitting on a successful method would seem to be enhanced. In my earlier review of the “search for method,” I concluded that no theory of method exists: “The search for the method for arriving at truths, as with Bacon, or for salvation, as with Wesley, or for school learning, as with Thorndike, and his followers, is neither appropriate nor productive. There is no way to ignore the goals, beliefs, and intentions of the teachers and learners; there are no shortcuts to learning or to gaining an education” (Olson, 2003, pp. 208–210, italics added). It is perhaps appropriate, therefore, to revisit the concept of pedagogy in a way that does not so clearly distinguish the philosopher’s concern with intention and the psychologist/educator’s concern with method, with “what works.”
A Theory of Pedagogy To this point we have established that teaching involves the intentions not only of the teachers but also the learners, that methods and their effects on learning have been difficult if not impossible to define, and that the relations between goals and means, remains relatively unexplored. The question remains as to whether or not a pedagogical theory linking goals and means may be formulated. Do different pedagogies lead to the same goal or do they imply different goals? Would such a theory set out alternative pedagogies or do all pedagogies have the same basic form? It may be argued that it is premature to attempt to formulate a theory of pedagogy before we have a better sampling of how teaching proceeds in different species, in different societies, on different tasks, with different types of teachers and learner, and so on. Indeed, there is a need for comparative and cultural research that would help to put some boundaries on the social practices we generally regard as teaching. These varying practices have been explored in a preliminary way elsewhere (Premack & Premack, 1996; Kruger & Tomasello, 1996; for a summary see Olson, 2003, 2007). Yet teaching is so pervasive in modern societies as well as advancing rapidly in developing countries that we may at least make some tentative steps in setting out some well-known forms of pedagogy, their uses and their presumed, and some cases demonstrated, effects. All pedagogy involves the attempt by an adult to alter the beliefs, intentions, and actions of learners; in other words all pedagogy assumes an asymmetry in knowledge, power, and authority between the teacher and the taught. Teaching attempts fall readily into two classes: individual as opposed to group teaching. Whether through sanctions, demonstrations, or telling, the teacher (whether parent, teacher, master, or older sibling) attempts to alter the behavior of the learner to meet some goal or standard. One-to-one teaching provides an optimal setting for arriving at mutual understandings in the service of jointly held goals. Group methods aspire to such common behavior and understandings but leave much more responsibility to the learners to regulate their own behavior and to make sense of the instruction. Group methods also allow learners to get some notion of what their contemporaries think and know and thereby to set some standard of expectation as
16 Whatever Happened to Pedagogical Theory?
229
to what they may realistically aspire to. Tutorial methods have been usefully described by both Dewey and Bruner as a form of scaffolding (see Olson, 2003) in which the master offloads some of the more difficult parts of a task to allow novices to succeed with the simpler components while progressively withdrawing support as the learner becomes capable of taking on the more difficult components. The method is also characteristic of apprenticeships and skill learning and is common in remedial educational contexts. Group methods, on the other hand, have been common since antiquity in the education of elites but took the form of graded classes and whole group instruction with the beginnings of mass education (Chartier & Hebrard, 2001). Second, within group methods we have two alternative forms of instruction that go back to classical antiquity of Greece and Rome, namely, the lecture and the seminar. The lecture is more expository, setting out the facts and theories in a domain, the seminar allowing more scope for the learners to bring their own experience and prior understandings to bear on a topic. These methods were seen as deriving from diverging conceptions of knowledge (Stace, 1920).Very roughly Socrates and Plato saw knowledge as essentially deductive with conclusions necessarily following from reasons. If one knew the reasons the conclusions could not be otherwise. Aristotelians and other Empiricists, on the other hand, thought of knowledge as empirical and available to the senses rather than exclusively to reason. The distinction, known to Plato (Stace, 1920) but more explicitly set out by Kant, is between necessary and contingent truths. Some things may be known as necessary truths without any teaching. This is the heart of the “Socratic method.” Socrates showed that even a slave boy could arrive at such necessary truths by means of questions and answers alone with no provision of information or instruction. Typical of such truths are those that make up geometrical and mathematical knowledge. If something is “known,” it is necessarily true; if one is a bachelor, he is necessarily unmarried. So too, the sum of the interior angles of a triangle have to add up to a straight angle. There is no need to measure hundreds of angles in triangles to confirm that this is the case. It is a provable, necessary rather than empirical truth and learners may be led to construct such truths for themselves by discourse and reflection. Plato argued that all real knowledge was of this form; consequently, he disparaged the importance of first-hand experience and he discounted the effects of direct teaching. Empiricist critics, on the other hand, insisted that there are contingent truths that result from observations, facts, experiments, and theories which just happen to be true and which one would know only by having first-hand experience or having been shown or told. This was the domain for the lecture - to teach the truths that a learner could not deduce for him or her self. Only in the nineteenth century, in the hands of Immanual Kant did the pedagogical distinction, which I have described as between the lecture and the seminar/tutorial, come to be seen as dependent not on the more general conception of knowledge but on a distinction between types of knowledge. Kant (1966) distinguished necessary from contingent truths; these provided a basis for his critiques of “pure” as opposed to “practical” reason. Necessary truths are best illustrated by appeals to mathematics and formal logic; contingent truths are those empirical
230
D.R. Olson
truths that could have been otherwise. Some modern philosophers have disputed the validity of the Kantian distinction (Quine and Putnam, as I recall). Nonetheless, it may be argued that these classical alternative methods, the lecture and seminar, are not alternative means to the same goal but appropriate to somewhat different goals. Whether those goals are those implied by alternative conceptions of knowledge as they were for Plato and Aristotle, or to different domains of knowledge, the necessary and the contingent as argued by Kant, it is clear that these contrasting methods are not distinguishable from the goals they serve. These two grand pedagogies, it seems, are particularly appropriate to different domains of knowledge. Both the lecture and the seminar as generalized forms may be further subdivided. Thus, the lecture method may be subdivided into expository methods in which the teacher sets out the knowledge in a systematic fashion and more interactive methods in which the same content is conveyed through question and answer techniques, the familiar question, response, correction, format common to many classroom lessons. Both have the same concern for so-called coverage as did the lecture, but they add a new concern for standards of correctness whether embodied in tests or in question–answer routines. The seminar method, too, may be further subdivided on the basis of teacher involvement and control. The teacher-led seminars assume that the teacher at the end of the day knows the most appropriate outcomes and is present to keep the discussion on track. Student-led seminars are based on the assumption that through discussion even novices will arrive at warranted understandings and agreement. Indeed, some recent evidence seems to bear this out. I have been told, although the original source is lost, that student groups given a topic and only two rules: first, everyone has to have a turn, and, second, comments have to stay on topic, yielded quite impressive results. Even fourth graders seem capable of honoring such rules at the same time as expressing and evaluating their own and others offerings. In traditional pedagogy, the seminar was always seen as secondary to the lecture and often followed the lecture. The seminar provided an occasion for seeing that the content had been understood. However, as traditional pedagogy gave way to a focus on learning, methods became, as noted, essentially ad hoc; any method would do so long as it met assessment criteria. Repeatedly through the twentieth century critics objected to the narrowness of the criteria being assessed by achievement tests. Something important was lost when education was defined exclusively in terms of the learning as measured through achievement tests. The two critical turns mentioned earlier, those led by Dewey and by Bruner, deserve attention.
Dewey and Progressivism Progressivism was the pedagogical side of American Pragmatism. Just as knowledge was to be seen as the solution to problems rather than the somewhat starry-eyed search for objective truth, so too, education was to be seen as acquiring the compe-
16 Whatever Happened to Pedagogical Theory?
231
tence to define problems and work to achieve their solution. This pedagogy may be seen as a radical outgrowth of the student-led seminar in that it came to rely heavily on what the learner brought to the learning both in terms of prior knowledge as well as the goals and standards to be used in appraising learning. Goal setting, planning and assessment of performance in the light of standards were all assigned to the learner to an unprecedented degree. Critics were quick to criticize Pragmatism for its somewhat loose and vague conception of truth, that is, the correspondence between theory and fact, and its consequent downplaying of assessment in terms of fixed standards. As we saw earlier, empirically the effects of pragmatism did little to advance educational standards, but on the other hand it should be noted it did little to reduce them either and had the secondary bonus of making schooling if not exactly enjoyable, at least bearable.
Brunerian Constructivism Bruner (1960) sees himself completely within the American Pragmatic tradition and his views have been shown to bear a strong affinity to those of Dewey (Olson 2007). For both, knowledge is seen as constructed by the learner in the attempt to solve problems and advance understanding. Both theorists rely heavily on the learner’s prior knowledge and intuition, on the ability of learners to revise their knowledge on the basis of new evidence, and on the importance of conversation and collaboration in the revision of belief and the advance of knowledge. However, Bruner adds one new dimension, at best implicit in Dewey, to the understanding of how children construct their knowledge. This is Bruner’s notion of “structure.” Whether described as coding systems, models, or theories, structure is what permits one to “go beyond the information given.” Bruner was concerned to show how knowledge could be constructed with the aid of abstract coding systems so that knowledge could be inferred by the child rather than merely learned and remembered. In one of his more famous examples, sixth-grade children were given the task of figuring out where the great cities of the American Northeast would be most likely to grow, given a map containing only information about physical features, natural resources, agricultural land, bodies of water and waterways, and the like, but not the actual location of cities. Class discussion produced several plausible theories. On what could be described as the “water transportation” theory, they located Chicago where the three great lakes met; on the basis of the “resource” theory they put it at the head of Lake Superior, near iron ore deposits and so on. Combining their theories they made plausible, and evaluable, theories that in some cases turned out to be true; their predictions corresponded largely to the actual location of real cities. Not only did the topic create great interest, the knowledge the children had acquired was generative, it could give rise to new knowledge. Even admitting massive overlap between the categories, we may yet summarize our discussion of these options by means of a taxonomic classification (see Fig. 16.1).
232
D.R. Olson Teaching //
\\
Individual // Directive/Didactic
\\ Tutorial
Group //
\\
Lecture
Seminar //
Student led //
Teacher led
\\
Project Method Discovery method
Fig. 16.1 Pedagogical options
How Are Pedagogies Related? Our final task is to ask how these alternatives are related. Are they variations on a theme or do they constitute real differences? And are the differences observed alternative means to the same ends or means particularly suited to different ends? As I mentioned at the outset, educators have been led to assume that methods are to be seen as alternative means to a given end, their task being to find the method that best works. Yet, methods are also often seen as having more than one goal. The reason teachers ask their students to read books is not simply because reading is the best means for obtaining particular information but rather, or at least in addition, to teach them how to get information from books. There is no other means for teaching the latter, so in this sense, alternative means are sometimes the means to new and different goals. Hence, we must consider the possibility that each distinctive method is an optimal means for achieving a somewhat different goal. This may be no less true for Dewey’s projects and Bruner’s discovery method that it is for reading books; the method entails a somewhat distinctive set of goals. If true, it is inappropriate to evaluate two methods with the same metric. Rather it would be important to choose methods on the basis of the scope of the goals the method addresses. In fact, this was the rationale behind Dewey’s advocacy of the so-called Project method. In creating a garden one not only learned the relevant biological knowledge one also learned to plan, cooperate, and take responsibility for one’s own learning. The rationale behind Bruner’s Discovery method was that children not only acquired the requisite knowledge but also learned how to go about testing hypotheses and constructing their knowledge, again becoming an agent of one’s own learning. Although one may distinguish a number of apparently distinctive pedagogies, they may all be seen as variations on the classical forms of lecture and seminar, the first relying heavily on the special knowledge and authority of the teacher, the latter on the intellectual resourcefulness of the learner. Rather than being alternative
16 Whatever Happened to Pedagogical Theory?
233
methods leading to the same goals, each of these methods, I have shown, is tied to a particular conception of knowledge. If knowledge is thought of a systematic and deductive, as for example in mathematics, one is led to implement more Socraticstyled seminars; if knowledge is thought of as empirical and factual then a lecture would be favored. Method follows one’s epistemological assumptions. However, as mentioned, the distinction took on a new clarity with Kant’s distinction between necessary and contingent knowledge. Socratic seminars would be seen as particularly appropriate to the exploration of necessary, deductive relations between concepts; knowledge may be created by mere reflection. Lectures, on the other hand, would be seen as particularly appropriate to contingent knowledge, facts and relations that are simply empirical and that could be otherwise. In fact, modern pedagogies embedded in curricula tend to choose content in such a way as to encourage both fact learning and theory building. Progressivism’s contribution was to insist that the two traditional methods could be reduced to one through the taking on of real problems and carrying out of real projects. In any real project one must extrapolate from both local facts and general theories to design a usable plan. In addition, learners would learn how to take charge of their own learning including the setting and monitoring of standards of correctness. They would not only have to meet the teacher’s standards, they would have to set and meet their own. The explicit teaching of bodies of knowledge would be rendered unnecessary because the students could look up, discuss, or otherwise construct the necessary understandings while at the same learning how to learn. Bruner’s constructivism may be seen as not only offering and justifying a new pedagogy, discovery learning, but also providing a way of reconciling the two traditional pedagogies, the seminar and the lecture. Recall that the seminar is premised on the assumption that acquiring knowledge is grasping the necessary relations between beliefs; geometry was seen as the clearest example of such knowledge. The lecture is premised on the assumption that most natural knowledge is empirical and contingent, things could have turned out differently. Bruner’s contribution, perhaps his most important contribution, was to show that even contingent knowledge, the knowledge delivered by the senses and embodied in the sciences, is learned and remembered primarily in terms of models, codes, coding systems or structures that permit one to generalize, that is to go beyond the given facts. These structures permit contingent, presumably arbitrary facts, to be seen as part of a general model. Hence the gap between necessary and contingent breaks down; even the contingent is deductable from the more general model. It follows that teaching should be premised on the construction and use of these “generic coding systems.” It is these generic models that allow for conjecture, hypotheses, and ultimately debate that Bruner sought through the method of discovery and through what he later described as teaching the hypothetical. Such models not only allow deduction, they aid the retention of the specific facts that are so valued by some educators. This resolution of the relations between lecture and seminar, between the teacher centered and the learner centered is, in itself, a contribution to pedagogical theory in that it frees the educator from the confines of any traditional or modern “method” and to tune teaching to the contextual resources and situational demands relevant at the time.
234
D.R. Olson
My own moral perspective on pedagogy (Olson, 2003) embeds Bruner’s conception of pedagogy into a framework of intentionality, responsibility, accountability, and the earning of entitlements. What one knows, understands, and infers is subordinated to one’s ability to take on and successfully meet one’s obligations. The process of managing the necessary information and other resources in successful meeting of one’s responsibility would entitle the learner to the credentials offered by the school and offer access to higher levels of education. As such, this moral perspective, moral in that it addresses teaching and learning in terms of meeting one’s obligations, adopts the concerns of Dewey and Bruner for children’s agency and intentionality, and furthers, one hopes, the development of the fledgling science of pedagogy.
References Barzun, J. (1945). Teacher in America. Indianapolis: Liberty Press. Bruner, J. S. (1960). The process of education. Cambridge, MA: Harvard. Bruner, J. S. (1966). Toward a theory of instruction. Cambridge, MA: Harvard. Chartier, A. M., & Hebrard, J. (2001). Literacy and schooling from the cultural historian’s point of view. In T. S. Popkewitz, B. M. Franklin & M. Pereyra (Eds.), Cultural history and education: critical essays of knowledge and schooling (pp. 263–288). New York: Routledge. Cremin, L. A. (1961). The transformation of the school: progressivism in American education 1876–1957. New York: Knopf. Dewey, J. (1976). The child and the curriculum. In J. A. Boydston (Ed.), John Dewey: the middle works, volume 2: 1902–1903 (pp. 271–291). Carbondale: Southern Illinois University Press. Dunkel, H. B. (1970). Herbart and Herbartianism: an educational ghost story. Chicago: University of Chicago Press. Gage, N., & Berliner, D. (1975). Educational psychology. Chicago: Rand McNally. Hirst, P. H. (1973). What is teaching? In R. S. Peters (Ed.), The philosophy of education (pp. 163–177). Oxford: Oxford University Press. Lagemann, E. C. (2000). An elusive science: the troubling history of educational research. Chicago: University of Chicago Press. Kant, I. (1966). Critique of pure reason (F. Max Muller, trans.). Garden City, NJ: Anchor Books. Originally published in 1881. Kruger, A., & Tomasello, M. (1996). Cultural learning and learning culture. In D. Olson (Ed.), Handbook of education and human development: new models of teaching, learning, and schooling (pp. 169–187). Oxford, UK: Blackwell. Olson, D. R. (2003). Psychological theory and educational reform: how school remakes mind and society. Cambridge, UK: Cambridge University Press. Olson, D. R. (2007). Jerome Bruner. New York: Continuum. Premack, D., & Premack, A. J. (1996). Why animals lack pedagogy and some cultures have more of it than others. In D. R. Olson & N. Torrance (Eds.), The handbook of education and human development (pp. 302–323). Cambridge, MA: Blackwell. Ravitch, D. (1985). The schools we deserve: reflections on the educational crises of our times. New York: Basic Books. Shulman, L. (Ed.) (1990) Research in teaching and learning: a project of American educational research association. New York: Macmillan. Stace, W. T. (1920). (Ed.) A critical history of Greek philosophy. London: Macmillan. Thorndike, E. L. (1913). Educational psychology: volume II: the psychology of learning. New York: Teachers College Columbia University. Tyack, D., & Cuban, L. (1995). Tinkering toward utopic: a century of public school reform. Cambridge, MA: Harvard University Press.
Index
A Acrophonic letter, 7 Alphabetic knowledge, children’s acquisition, 3 Alphabetic orthographies, 55, 192 Alphabetic–phonetic principle, 7, 9 Alphabetic–phonetic spelling, 5 Alphabetic–phonetic writing, 12–13 Alphabetic scripts, morphological awareness in, 104 Alphabetic spellings, 36 syllabic spellings as partial, 38–39 Alphabetic writing, 189 Analogies, for measuring child’s ability to produce inflected version of word, 196 Analysis of variance (ANOVA), 64, 70, 73, 213 Anderson, A., 90 Anderson, J., 90 ANOVA. See Analysis of variance (ANOVA) Aram, D., 102, 126 Aro, M., 168 Artzer, M., 150 Assessment-to-Instruction software, 159 effect of increasing use of, 160 B Baddeley, A., 151 Barad, N., 130 Barzun, J., 223 Batista, A., 32, 34, 38 Beck, I., 142 Becoming Literate in the City, 86 Beginning-level instruction, 150 Bentin, S., 168 Ben-Zvi, G. A., 193 Berliner, D., 225 Bialystok, E., 24
Bilingual alexia, 172 Bilingual day-school program, 195 Bilingual dyslexic, 173 Bilingual hyperlexic, 173 Blachowicz, C., 142 Bonferroni tests, 64, 74 Book reading, 115–117 Bowman, M., 56 Brazilian Portuguese spelling development alternative views, 33–35 letter name knowledge, 32–33 longitudinal study of, 35 Brice-Heath, S., 124 Brown, C. P., 132 Brunerian constructivism, 231 Bruner, J. S., 226, 231 Bus, A. G., 48 Butterworth, B. L., 172, 173 Buysse, V., 131 C Cairney, T., 124 Cairns, H., 207 Calif, S., 194 Canning, A., 46 Cardoso-Martins, C., 32, 34, 38, 56 Carlisle, J. F., 191 Carver, R. P., 169 Catts, H. W., 169 Child-by-instruction interactions, 154 Childhood educators, 116 Children alphabetic–phonetic writing promotion, 12–13 alphabetic sequence memorization, 24 understanding, 7
235
236 Children (cont.) attention during shared reading, 45–47 book reading intervention, 115–117 environmental and psychological factors impact, 168 generalization of phonetic writing, 7–9 home literacy and outcomes English-speaking families, 117–119 French-speaking families, 119–120 income level impact, 98 knowledge of surface features letters, 22 print, 18–20 word, 20–22 with language delays, 115–117 letter name knowledge, 11 letters use, random and ambiguous, 8 literacy competencies assessment, 87, 93 literacy development, 85, 94 name writing influence, 9 parental coaching behavior influence, 48–49 parental mediation of learning, 126 phonetic sensitivity, 11–12 phonological analysis skills, 93 print–speech correspondences, 35 spelling–sound relations understanding, 17 spelling stages development, 33 Children’s emerging competencies, 94 Child’s home culture engagement with literacy-relevant activities, 91–93 parents’ beliefs and children’s activities, 88–91 Chiu, M. M., 98 Classroom literacy, dimensions of, 156 Clay, M., 178 Code Coaxers, 49 Code-focused vs. meaning-focused instruction, 156 Cognitive-linguistic skills, intra-individual, 194 Componential Model of Reading, 168, 170, 174 application to dyslexic and hyperlexic children, 171–174 Spanish-speaking children, 170–171 Comprehension vocabulary, 112, 114 Concatenative languages, 192 Conlon, A., 46 Connection-forming process, 139 Corrêa, M. F., 33 Critchley, M., 172 Cross-linguistic relationships, 194
Index Cue length effect on letter reporting and spelling, 70 method of analyzing, 72–73 phonemic isolation, 68–71 Cues, position of, 65–67 D Da Fontoura, H. A., 218 Daley, K., 129 Darija, 180, 181, 188 de Jong, P. F., 169 Dewey and progressivism, 230–231 Dewey, J., 226 Dialogic reading, 100, 116 Disyllabic letter names, 74 Dyslexia-type syndrome, 171 Dyslexic children comprehension and decoding skills, 169 inflecting verbs errors, 207 E The Early Childhood Project, 85, 86 methods used in, 87–88 Early Reading First, 155 Ehri, L., 34, 56, 138, 139, 140 Ehri’s phase model, spelling skills, 37–38 Emergent literacy, preschool experience, 154 English as a second Language (ESL), 205, 206, 208, 212, 213 English Oral Cloze task, 207 English-speaking families, home literacy, 117–119 Erskine, J. M., 168 ESL. See English as a second Language (ESL) Evans, M. A., 45, 46, 47, 49 Expressive One Word Picture Vocabulary Test, 116 Ezell, H. K., 51 F Fagan, W. T., 48 Feitelson, D., 173 Feng, G., 46 Ferreiro, E., 33, 34, 178 Ferreiro’s stage model, spelling skills, 36–37 Fey, M. E., 169 “Firm Foundations,” 211 Fisher, P., 142 Foreign language, literacy in, 179–182
Index French-speaking families, home literacy, 119–120 Frith, U., 169 Frost, R., 168 G Gage, N., 225 Gallimore, R., 89 Gathercole, S. E., 151 Generic coding system, 226 Geva, E., 194, 196, 207 GFW Sound Mimicry subtest, 209 Gholamain, M., 207 Goldenberg, C., 89 Gough, P., 169 Gough, P. B., 169 Grapheme–phoneme system, 139 Grapho-phonemic mapping, 140, 149 Grapho-syllabic units, influence on sounds, 141 Graue, E., 132 Guo, J., 46 H Handbook of orthography and literacy, 167 Hauser-Cram, P, 131 Hayden, H. M. R., 48 Head Start, 155 Heath, S. B., 93, 129 Hebrew alphabet composition, 56 Hebrew as a foreign language (HFL), 194 Hebrew–English bilinguals, 194 Hebrew language proficiency, 196 Hebrew letter names effectiveness of cue’s length, 74 incongruent partial, 59 role in early literacy, 55 Hebrew morphological skills, 194 development in monolingual Hebrew, 193 Hebrew morphology, 193 Henderson, L., 168 HFL. See Hebrew as a foreign language (HFL) Hierarchical linear modeling, 158 Hierarchical regression analysis, phonetic writing, 12 Hinshelwood, J., 172 Hirst, P. H., 224 Hogan, T. P., 169 Home literacy and child outcomes in English-speaking families, 117–119 in French-speaking families, 119–120
237 Home literacy environment characteristics of, 98–100 children’s literacy development, 100 Homophone sensitivity, associations to Chinese, 105 Hoover, W. A., 169 Hyperlexia-type syndrome, 171 Hyperlexic children, comprehension and decoding skills, 169 I Individualized instruction, 156 elements in design, 159 Inflectional morphology, 104, 196 analogies, 196 nonword inflections, 196, 202 predictors of, 199–200 real-word inflections, 196, 198, 202 Instruction code-focused vs. meaning-focused, 156 dimensions of, 156–157 child-by-instruction interactions, 158–159 child factors, 157 intervention to individualize instruction, 159–161 individualized, 156 Intra-individual cognitive-linguistic skills, 194 J Joanisse, M. F., 207 Joint writing, mother–child, 102, 126 Justice, L. M., 46, 51 K Kamhi, A. G., 169 Katz, L., 168 Keating, P., 207 Korat, O., 98, 125, 126 Kruskal Wallis test, 8 L Lagemann, E. C., 226 Language deeper features of, 22–26 surface features of children’s knowledge of letters, 22 children’s knowledge of print, 18–20 children’s knowledge of words, 20–22 Lankford, C., 46
238 Lasky, S., 131 Launch into Reading Success program, 211 Learning after reading book once, 113 thrice, 114–115 twice, 113–114 Learning Together: Read and Write with Your Child program, 125, 129, 132 Lebrun, Y., 172 LeFevre, J., 102, 119, 129 Letter identification subtests, 209 Letter name children’s phonological awareness, 57 effect on children’s ability to connect print and speech, 58 phoneme isolation, 75 evidence from non-literate children and adults, 32–33 incongruent partial, 59 individual differences in use, 60 knowledge in children, 11 method for analyzing impact, 60–64 spelling skills in Brazilian Portuguese, 32–33 Letter reporting and spelling, 67–68 Letters acrophonic, 7 phonetic use of, 5, 7 random vs. ambiguous use of, 4–8 visual forms of, 6 Letter–sound relationship, 4 Letter–word recognition, 158 Levie, R., 193 Levin, I., 31, 57, 97, 102, 123, 126, 193 Lexical compounding, 105, 106 Linn, R., 50 Literacy competencies, children, 87, 93 Literacy development in children, 85, 94 children and adults attitudes impact, 99 cognitive factors related to, 103 contribution of parent teaching, 120–121 in foreign language, 179–182 in low-income families with poor educational, 123 metalinguistic skills and, 103–107 nature of parents’ beliefs and, 87 Literacy-related interactions between children and parents, 87 cognitive–linguistic aspects of, 88 Literacy skills, morphological awareness, 192 Lynch, J., 90
Index M Man: a course of study, 226 Manis, F. R., 207 MANOVA. See Multivariate analysis of variance (MANOVA) Mansell, J., 49 Margalit, T., 130 Martineau, H., 124 Martin, L. E., 48 “Maternal Mediation” scales, 102 Matrix Analogies Test-Expanded Form, 195 McBride-Chang, C., 98 McEvoy, M. A., 48 McKeown, M., 142 Metalinguistic skills, and literacy development, 103–107 Meyer, L. A., 50 Mirak, L. J., 170 Morphemes, phonologic and orthographic representations, 191–192 Morphological awareness (MA), 191–192 and literacy skills, 192 for reading, 104 Morphological processes, 192–193 Morphological skills correlates of, 197–198 development of, 197 Morphological vocabulary, 196 Mother–child dyads, 103 Multi-syllabic words, 140 Multivariate analysis of variance (MANOVA), 197 N Name series spellings, 56 Name writing, influence onalphabetic– phonetic knowledge, 9 Ninio, A., 44 Non-verbal intelligence, 112, 198 Nonword inflections, 196, 202 predictors of, 199 Number inflections, spoken representation of, 182–184 O Obler, L., 172 One-Minute Word Reading Test, 209 Oral Cloze task, 211, 213 Oral comprehension, 196, 198 Oral language development, 168 Orthographic Depth Hypothesis (ODH), 167, 168, 171, 174
Index Orthographic knowledge, 146 Orthographic memory, 151 Orthography definition, 168 P Parental coaching behavior, 48–49 Parental mediation, of children’s learning, 126 Parent–child dyads, 101 Parent–child interactions with books, 45 in shared reading, 48 Parent–child literacy, 100–103 Partial alphabetic spellings, 37 syllabic spellings as, 38–39 Partial letter name cues, 75 Patel, S., 130 Peabody Picture Vocabulary Test–Revised, 114, 116 Peabody Picture Vocabulary Test–Third Edition, 196 Pearson inter-correlations, 198 Pedagogy theory, 228–230 Phillips, L. M., 128 Phoneme deletion task, 210 Phoneme identification tasks, 210 Phonemic segmentation, 141 Phonetic sensitivity, in children, 11–12 Phonetic writing, 4 prediction of, 12 Phonological analysis skills, of children, 93 Phonological awareness training for kindergartners’, 105 Phonological decoding fluency task, 210, 213 Polysyllabic pseudowords, 209 Poor readers (PR), 214 Pre-alphabetic spellings, 37 Preschool intervention, short-and long-term benefits, 154–155 “Print Mediation” scales, 102 Print–speech correspondences in children, 35 Pronunciation recall trials, experiment with fifth graders, 146–149 second graders, 143–146 Pseudoword reading skills, 207 Pseudoword spellings, 138 Public preschool programs, 155 Purcell-Gates, V., 90 Pursoo, T., 46 Q Qina tian hua, 180, 181
239 R Rapid automatized naming (RAN), 210, 213 Raven’s progressive Matrices, 173 Ravid, D., 74, 193 Ravitch, D., 227 Read, C., 178 Reading difficulties. See Reading disabilities (RD) Reading disabilities (RD), 212 early identification and intervention, 205–207 Real-word inflections, 196, 198, 202 predictors of, 199 Receptive language, 198 Receptive vocabulary, 196, 202 Reese, L., 89 Reitsma, P., 140 Rescorla, L., 170 Rhyme detection task, 209 Richmond-Welty, E. D., 58 Rodrigues, L. A., 33 Ryan, E. B., 207, 211 S Saiegh-Haddad, E., 194 Saint-Aubin, J., 45, 47 Sample, H., 128 Scarborough, H. S., 170 Schiff, R., 194 School-based literacy, 129 SDRT. See Stanford Diagnostic Reading Test (SDRT) Seidenberg, M., 142 Seidenberg, M. S., 207 Sénéchal, M., 102, 112, 119, 129 Seymour, P. H. K., 168 Shany, M., 196 Shapiro, J., 90 Share, D., 140 Shared reading children’s attention to, 45–47 contribution to child outcomes, 111 evolving nature and value of, 44–45 limits of, 117 parent–child interactions in, 48 as source of entertainment, 111 vocabulary learning and, 111 Shatil, E., 196 Shaw, D., 49 Shimron, J., 193 Siegel, L. S., 196, 207, 211 Single-word recognition, 207 Sirin, S. R., 131
240 Skibbe, L., 46 Smolkin, L. B., 46 Snow, C., 129 Snow, C. E., 91 Snowling, M., 169 Socioeconomic status (SES), 208 Sociolinguistic questionnaire, 185 Sound series spellings, 56 Sparks, R., 150 Speech, influence of word spellings on, 141–142 Spelling development in Brazilian Portuguese, 33–35 Ehri’s phase model, 37–38 Ferreiro’s stage model, 37 longitudinal study, 35 stages of, 33 Spelling skills of children, 4 development in children, 31 Ehri’s phase model, 37–38 Ferreiro’s stage model, 36–37 letter name knowledge and, 32–33 Spelling–sound relations, understanding in children, 17, 26 Spoken representation of locative derivatives, 184–185 of number inflection, 182–184 Spoken vocabulary, 112, 115. See also Vocabulary Stadler, M. A., 48 Stahl, S. A., 50 Standard Modern Arabic (SMA), 181 Stanford Achievement Test Total Reading Score, 3 Stanford–Binet Intelligence Scale, 172 Stanford–Binet Memory for Sentences test, 211 Stanford Diagnostic Reading Test (SDRT), 209, 215, 216 Stanovich, K. E., 212 Stipek, D., 131 Stroop test, 43 Sulzby, E., 178 Syllable identification tasks, 210 Symbolic function, of writing, 3 Szliwowski, H., 172 T Tanenhaus, M., 142 Taylor, D., 124 Teacher-directed vs. child-centered instruction, 155
Index Teaching, one-to-one, 228 “Teaching the hypothetical,” 226 Teale, W., 178 Teale, W. H., 89 Test for Reception of Grammar (TROG), 196 Thomas, E., 129 Thorndike, E. L., 223 Thorndike’s learning theory, 223 Tincoff, R., 58 Title 1 preschool program, 155 Tolchinsky, L., 20, 56 Treiman, R., 4, 20, 31, 56, 58, 74 Tunmer, W., 169 Tyler, R., 227 V van der Leij, A., 169 Van Endert, C., 172 van IJzendoorn, M. H., 48 van Kleeck, A., 48 Verbal efficiency theory, 169 Visual-graphic symbols of phonemes, 59 Vocabulary acquisition after three readings of book, 114–115 two readings of book, 113 of children with language delays, 115 development, 112, 194 knowledge in children, 101 learning and instruction, 142 implications for, 150–151 learning and shared reading, 111, 142 literacy skills and, 191–192 types of, 112 Vowel-consonant syllables, 209 W Wade-Woolley, L., 196 Waltzman, D., 207 Wardrop, J. L., 50 Wesley, P. W., 131 Wilce, L., 141 Wilcoxon test, 7 Williamson, K., 46 W–J Word Attack subtest, 212 Woodcock Johnson Reading Mastery Test, 209 Woodcock Johnson Tests of AchievementRevised, 93 Word derivation, non-concatenative languages, 193 Word identification subtest, 209
Index Word learning, theory of, 140 Word reading different ways for, 138 from memory, 139–140 Word spellings, influence on speech, 141–142 Word Suppliers, 49 Word-writing, mothers’ strategies for promoting, 102 Working Memory for Words task, 211 WRAT Reading subtest, 208, 212–214
241 WRAT3 Spelling subtest, 209 Written language, characteristics of, 19 Wydell, T. N., 172, 173 Y Yaden, D. B., 46 Yberra, L., 128 Yopp, H. K., 46 Yopp, R. H., 46