Current Issues in Reading, Writing and Visual Literacy : Research and Practice [1 ed.] 9781443885591, 9781443880305

Citation preview

Current Issues in Reading, Writing and Visual Literacy

Current Issues in Reading, Writing and Visual Literacy: Research and Practice Edited by

Christina Gitsaki Melanie Gobert Helene Demirci

Current Issues in Reading, Writing and Visual Literacy: Research and Practice Edited by Christina Gitsaki, Melanie Gobert and Helene Demirci This book first published 2015 Cambridge Scholars Publishing Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Copyright © 2015 by Christina Gitsaki, Melanie Gobert, Helene Demirci and contributors All rights for this book reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the copyright owner. ISBN (10): 1-4438-8030-2 ISBN (13): 978-1-4438-8030-5

TABLE OF CONTENTS

List of Tables ............................................................................................ viii List of Figures............................................................................................. xi List of Appendices ..................................................................................... xii List of Abbreviations ................................................................................ xiii Chapter One ................................................................................................. 1 Current Issues in Reading, Writing and Visual Literacy: An Introduction Christina Gitsaki, Melanie Gobert, and Helene Demirci Issues in Reading Literacy Chapter Two .............................................................................................. 10 Middle Grade Teachers’ Perceptions of Applied Reading Activity Structures and Strategies versus Observed Practice in Second Language Classrooms Nancy Allen Chapter Three ............................................................................................ 28 Qatari Middle School Students’ Perceptions of Strategies for L1 and L2 Reading Dudley W. Reynolds Chapter Four .............................................................................................. 46 How Do Arabic (L1) and English (L2) Reading Differ? Evidence from Think-Aloud Protocols Zohreh R. Eslami, Katherine L. Wright, and Sunni L. Sonnenburg-Winkler Chapter Five .............................................................................................. 68 Linguistic Differences between English and Science Texts for English Language Learners: A Quantitative Analysis Katherine Landau Wright and Zohreh R. Eslami

vi

Table of Contents

Chapter Six ................................................................................................ 86 Text-Talks as a Means to Scaffold Disciplinary Reading Literacy in Secondary Education Yvonne Hallesson and Pia Visén Chapter Seven.......................................................................................... 102 How Incidental is Incidental Vocabulary Learning? Marina Dodigovic Chapter Eight ........................................................................................... 117 A Content Analysis of the New Practical Chinese Reader I for Beginner Chinese Language Learners: Shortcomings and Recommendations Han Lin Chapter Nine............................................................................................ 134 A Study of the National High School English Wordlist in Taiwan Lee-Yen Wang Issues in Writing Literacy Chapter Ten ............................................................................................. 158 Collocations in Advanced L2 Writing Päivi Pietilä Chapter Eleven ........................................................................................ 175 Production of Formulaic Sequences in L2 Writing by Japanese Learners of English Natsumi Okuwaki Chapter Twelve ....................................................................................... 193 The Positioning of L2 Thesis Writers Emmaline Lear and Huifang (Lydia) Li Chapter Thirteen ...................................................................................... 209 Evaluating Chinese Learners’ Academic Writing Skills in Argumentation Essays Peiling Xing Chapter Fourteen ..................................................................................... 227 Written Corrective Feedback: An EFL Postgraduate Perspective Naif Althobaiti

Current Issues in Reading, Writing and Visual Literacy

vii

Issues in Visual Literacy Chapter Fifteen ........................................................................................ 246 Japanese Learners’ Higher-Order Inference-Oriented Schema Transfer in Interlanguage Comprehension Development Mariko Boku Chapter Sixteen ....................................................................................... 266 The Influence of Television on American Vocabulary Acquisition with North Queensland Children Danica Kelly, Gary Williams, and Susan Morrison Chapter Seventeen ................................................................................... 285 Digital Curation as a New Literacy: Students as Potential Curators of Information on the Web Nayara N. de Barros Contributors ............................................................................................. 298 Index ........................................................................................................ 304

LIST OF TABLES

Table 2-1: Table 2-2: Table 2-3: Table 2-4: Table 3-1: Table 3-2: Table 3-3: Table 3-4: Table 3-5: Table 4-1: Table 4-2: Table 4-3: Table 4-4: Table 4-5: Table 4-6: Table 4-7: Table 4-8: Table 4-9: Table 4-10: Table 5-1: Table 5-2: Table 5-3: Table 5-4: Table 5-5:

Demographics of survey sample. Organization of activity designs and strategies included in survey and observations. Comparison of self-reported and observed activity design. Comparison of self-reported and observed strategy (taught, modeled, or used). Instrument sections and forms. High performing readers’ strategy clusters for Arabic reading. Low performing readers’ strategy clusters for Arabic reading. High performing readers’ strategy clusters for English reading. Low performing readers’ strategy clusters for English reading. Summary of reading selections. Type of processing: Mean percentages and standard deviations. Statistically significant differences in the use of regulatory strategies. Statistically significant differences in the use of cognitive strategies. Use of content and language strategies in L1 and L2. Use of language-oriented strategies in L1 and L2. Domain of processing in L1 and L2. Regulatory strategy use. Cognitive strategy use. Cognitive-Iterative strategy use. Science and English texts’ readability scores. Measures of referential cohesion. Situation model indices. Syntactic complexity index. Standardized canonical discriminant function coefficients.

Current Issues in Reading, Writing and Visual Literacy

Table 6-1: Table 6-2: Table 7-1: Table 7-2: Table 7-3: Table 7-4: Table 9-1: Table 9-2: Table 9-3: Table 9-4: Table 9-5: Table 9-6: Table 9-7: Table 9-8: Table 9-9: Table 10-1: Table 10-2: Table 10-3: Table 11-1: Table 11-2: Table 11-3: Table 11-4: Table 11-5: Table 12-1: Table 12-2: Table 12-3: Table 12-4: Table 12-5: Table 13-1: Table 13-2:

ix

Analytical model for text movability: Text-based Dimensions. Analytical model for text movability: Associative and Interactive Dimensions. Vocabulary Size Test entry results. Vocabulary Size Test exit results. Vocabulary Size Test exit - entry comparison in Group A. Vocabulary Size Test exit - entry comparison in Group B. Words in GSL but absent in the ERWL (70 Words). Words in AWL but absent in the ERWL (52 Words). Words common to both AWL and ERWL listed by the CEEC Level. AWL words in CEEC Level 1. AWL words in CEEC Level 2 ranked by COCA (67 Words). AFDCE words absent in the ERWL. AFDCE words absent in ERWL in the 1,000-2,000 rank group. COCA and ERWL words Rank Comparison-I. COCA and ERWL words Rank Comparison-II. Lexical profiles of the writer groups. Number of collocations produced by the writer groups (per 1000 words). Distribution of collocation categories in the three subcorpora. Types of formulaic sequences (Source: Ohlrogge, 2009). Participants’ information. Descriptive statistics for length of Essay 1 and Essay 2. Number of FSs used in Essay 1 and Essay 2. Frequencies and percentages of FSs per essay type. Interactive features of metadiscourse. Interactional features of metadiscourse. Metadiscourse in results/discussion chapters (per 10,000 words). Interactive metadiscourse (per 10,000 words). Interactional metadiscourse (per 10,000 words). Methods of organization for argumentative essays (Source: Folse et al., 2003, p. 171). Participants’ perception of their writing skills.

x

Table 13-3: Table 13-4: Table 13-5: Table 13-6: Table 13-7: Table 13-8: Table 15-1: Table 15-2: Table 15-3: Table 15-4: Table 15-5: Table 15-6: Table 16-1: Table 16-2: Table 16-3: Table 16-4: Table 16-5:

Table 16-6:

List of Tables

The skills that the participants felt confident, less confident or not confident in using. The structural sentences used by students. Background information written by the students. The body paragraphs written by the students. The conclusion written by the students. References written by the students. Outline of the study. Procedure of tasks and tests 1-3. Analysis of Variance: Two tasks. Learners’ task scores and independent samples t-test of task scores. Correlation between tasks and groups: Movie A. Learners’ narrative summaries. Participant descriptive data. Sample vocabulary test items and expected responses. Children’s television viewing data. Frequency of television programs viewed. Comparison between individual vocabulary items, TV hours viewed and the proportion of American programs viewed. Comparison between TV hours viewed, proportion of American vocabulary used and proportion of American programs viewed.

LIST OF FIGURES

Figure 2-1: Figure 2-2: Figure 2-3: Figure 2-4: Figure 2-5: Figure 4-1: Figure 6-1: Figure 15-1: Figure 16-1: Figure 16-2: Figure 16-3: Figure 17-1:

A comparison of the reported use of activity structures with observed use of activity strategies. A comparison of the reported use of global strategies with the observed use of global strategies. A comparison of the reported use of support strategies with the observed use of support strategies. A comparison of the reported use of vocabulary strategies with the observed use of vocabulary strategies. A comparison of the reported use of compensation strategies with the observed use of compensation strategies. Flow chart for coding think-aloud utterances. Analytical model for lexical and conjunctive cohesion. Group comparison in comprehending high order meaning: Movie A. Language learning continuum. Language learning continuum for the study. Post-study language learning continuum. Story creation page © 2014 Storify. Used with permission.

LIST OF APPENDICES

Appendix 4-A: Analyses for all the strategies. Appendix 15-A: Tasks

LIST OF ABBREVIATIONS

ACL AEST AFDCE ALT Am AmE ANOVA APA ASL AST ASW AusE AV AWL BICS CAS CDS CEEC CELT CIDE COBUILD COCA COLT CPT DDA DRS EAP EFL EIL ELF ELL EMI ERWL ESL

Academic Collocation List Advanced English Subject Test A Frequency Dictionary of Contemporary English Assistant Language Teacher American American English Analysis of Variance American Psychological Association Average Sentence Length Advanced Subject Test Average Syllables per Word Australian English Audio-Visual Academic Word List Basic Interpersonal Communication Skills Complex Adaptive Systems Complex Dynamic Systems College Entrance Examination Center Comprehensive English Language Test Cambridge International Dictionary of English Collins Birmingham University International Language Database Corpus of Contemporary American English Corpus of London Teenage Language Constant Polarity Tag Descriptive Discriminant Analysis Discussion Results Section English for Academic Purposes English as a Foreign Language English as an International Language English as a Lingua Franca English Language Learners English Medium Instruction English Reference Word List English as a Second Language

xiv

ESP FSs G1 G2 G3 GSAT GSEAT GSL HAZ HDR HIOS IC ICC IELTS K-12 K1 K2 L1 L2 LLC LTC LTCI M MA MARSI MCD MEXT N NNS No. NPCR NPCR-I NS OECD PCK PhD PISA PLS-4 QtA RAs

List of Abbreviations

English for Specific Purposes Formulaic Sequences Group 1 Group 2 Group 3 General Scholastic Ability Test General Scholastic English Ability Test General Service List Heat Affected Zone Higher Degree by Research Higher-Order Inference-Oriented Schema Instructional Conversations Intercultural Communicative Competence International English Language Testing System Kindergarten through twelfth grade First 1000-word frequency band Second 1000-word frequency band First language Second Language London Lund Corpus Language Teacher Cognition Language Teacher Cognition Inventory Mean Master of Arts Metacognitive Awareness of Reading Strategies Inventory Macmillan Collocations Dictionary Ministry of Education, Culture, Sports, Science and Technology Number of Participants Non-Native Speaker Number New Practical Chinese Reader New Practical Chinese Reader I Native Speaker Organisation for Economic Co-operation and Development Pedagogical Content Knowledge Doctor of Philosophy Programme for International Student Assessment Preschool Language Scale 4th edition Questioning the Author Research Articles

Current Issues in Reading, Writing and Visual Literacy

RQM RT RT RTP SD Sig. SLA SORS SPIn TALIS TC TOEIC TPAInt TPBInt TV UC USA UWL VST WCF XJTLU ZPD

Retrodictive Qualitative Modelling Reciprocal Teaching Relevance Theory Reversed Polarity Tag Standard Deviation Significance level Second Language Acquisition Survey of Reading Strategies Student Participant Interview Teaching and Learning International Survey Teacher Cognition Test of English for International Communication Transcript of Participant A’s Interview Transcript of Participant B’s Interview Television University of Canberra United States of America University Word List Vocabulary Size Test Written Corrective Feedback Xi’an Jiaotong-Liverpool University Zone of Proximal Development

xv

CHAPTER ONE CURRENT ISSUES IN READING, WRITING AND VISUAL LITERACY: AN INTRODUCTION CHRISTINA GITSAKI, MELANIE GOBERT, AND HELENE DEMIRCI

Introduction Ancient civilizations such as the Babylonians, Sumerians, Confucian Chinese and the ancient Greeks were all literate societies although widespread literacy was not prolific. In the past, having access to texts in written form restricted the spread of literacy. Up until the 18th century, literate individuals were mostly those promoting religion, trade, state affairs, and members of nobility. By the mid-1800s literacy spread to the masses in Europe and North America when printed material became available albeit with some discrimination related to gender and class. Since this time schools have been the most influential institutions responsible for the spread of literacy in all societies (UNESCO, 2005). As schools are instrumental in developing literacy skills, it is necessary to understand the way educators believe, interpret and model the required skills which constitute literacy competence. The development of literacy competencies has been significant and they were recognized as a necessary right for every citizen because possessing literacy competencies is a prerequisite to playing a participatory role in society. Since the industrialization of education, reading, writing, and numeracy have been considered essential literacy skills. However, with the advent of the Internet and wireless technology, information, media, and technology skills have become essential components of literacy.

2

Chapter One

Given the advances in the availability of information and the moves towards enhancing 21st century skills and digital literacies, it is even more essential for educators to adopt effective methodologies which promote learner engagement in literacy competence development (Mirra, 2014). Lacking skills in these areas can impede an individual’s potential to participate in higher education and thus restrict their ability to find gainful employment upon graduation, which will prevent the individual from making a viable economical contribution to society. Developing literacy competencies can also contribute towards promoting life-long learners who have a thirst for knowledge (Mirra, 2014), which in turn will benefit society as a whole. As individuals engage in seeking knowledge in a digital age, developing information literacy skills becomes necessary. “Metacompetencies” (Lloyd, 2003, p. 87) are prerequisite skills required to enable individuals to access the information widely available. Information literacy is viewed as a key skill towards becoming a competent knowledge worker who cannot only access information, but is also skilled at integrating the information accessed into daily interactions in the workplace (Lloyd, 2003). Globalisation has also caused a trend towards second language (often English) literacy for those who want to have the best education and employment options. The development of literacy skills cannot be left to chance. With the onset of the digital age and the evolving nature of digital literacy, educators need to pay more attention by thinking critically about how this translates into the classroom (Brown, 2001). The classroom has also changed with the expansion of societies through immigration and refuge offered to those fleeing poor economic conditions and war. In some countries, over 15% of the public school population may be second language learners (Fast Facts, 2015). Many second language learners experience a myriad of problems mastering literacy in the second language. For some learners, these problems include coming from a different alphabetic writing system or having a language that does not traditionally have a written form, such as Hmong. Young learners especially must follow the same route to becoming literate in the second language as native speakers of the language. There are also no shortcuts to literacy in the second language for adolescent or young adult learners if the goal is higher educational studies through the medium of the second language. Yet, much second language teaching methodology assumes that second language learners are literate in their own language and must simply transfer this knowledge by meta-cognition to the second language. The reality in the classroom tells a different story with children

Current Issues in Reading, Writing and Visual Literacy: An Introduction

3

who speak a second language at home often being mandated as a special accommodation group in educational testing to set a fairer playing field. Bilingual children tend to lag up to two years behind their monolingual peers in literacy development. It has even been said that no one can be equally fluent in two languages, that one language will always be favored over another, and that mixing up the two languages in the developmental stage is proof of that. However, the literacy gap in the two languages spoken by bilingual children tends to disappear by mid-adolescence and certainly in the university years. Bi-literacy is poorly understood and studies have shown that learners tend to ‘think’ in the language in which they were taught a subject matter. Teaching new information literacy skills on top of the core literacy skills of reading, writing, and arithmetic poses new challenges for teaching. These students’ needs must be met by teachers in the classroom, if economic justice is to prevail. The contents of this book emphasize the need for educators to carefully consider the context within which they interact with learners, to carefully review the materials used and to be fully aware of the needs of the students. Given the literacy skills required in the 21st century and the proliferation of availability of information, learners at all levels need specialized skills which will enable them to participate to a required level from where they will be able to compete within the knowledge economy. There are three main parts in this volume. Part One comprises eight research papers addressing a range of issues in the development of reading literacy from a first language and a second language perspective. Part Two is a collection of five chapters investigating issues related to the development of writing literacy. Finally Part Three addresses issues in visual literacy.

Current Issues in Reading Literacy In Part One there are eight chapters describing research studies on issues related to the development of reading literacy. In Chapter Two, based on the proposition that strategic reading is important for comprehension, Nancy Allen examines teachers’ self-reported use of literacy strategies and compares the findings to data from classroom observations. The study was conducted in the Qatari educational context and findings indicated that teachers extensively over-reported their use of reading activity structures and strategies. This comparative study highlights the strengths, weaknesses, and areas of potential misconceptions regarding reading instruction.

4

Chapter One

In Chapter Three, Dudley Reynolds extends Allen’s study by looking at the use of reading strategies from the perspective of the student. Comparisons between the groupings of reading strategies selected by low and high performing students following a reading in their mother tongue (L1-Arabic) and a reading in their second language (L2-English) showed that the high and low performing students identified similar strategy groupings after completing a reading in their mother tongue, while the high performing students grouped strategies into smaller and more distinctive groupings after the English reading. The low performing students reported using fewer strategies. The findings not only indicate the importance of clear strategy constructs for L2 reading, but they also suggest that existing constructs for L1 reading may not be specific enough to be of use in the L2. In the following Chapter, Zohreh Eslami, Katherine Wright and Sunni Sonnenburg-Winkler utilized think-aloud protocols to investigate the mental processes of EFL students as they encountered difficulties in reading two texts, one in their mother tongue (Arabic) and one in English. Results showed that overall readers focused more on language and used higher proportions of language-oriented strategies, monitoring strategies, and above-clause strategies when reading texts in their L2 (English) than when reading in their L1 (Arabic). To further investigate the language needs of EFL students when reading science texts and the additional literacy support they may require, Katherine Landau Wright and Zohreh Eslami used Coh-Metrix, an automated online text analysis system, to compare the English-language texts used in English as a foreign language (EFL) and in science classes. It was found that the EFL texts tend to focus on basic communication skills, whereas the science texts require students to follow more complex text structures and vocabulary. Knowing how to scaffold the reading of such increasingly complex texts students encounter is of primary importance. In Chapter Six, Yvonne Hallesson and Pia Visén describe how texttalks may function as a scaffold for students’ disciplinary reading literacy. Through the use of the analytical concept of text movability and intertextual analyses of lexical and conjunctive cohesion between text and discussion, they uncovered various ways in which the students define concepts from the text, explore both text content and subject field, and objectively question content. The possibility of incidental vocabulary learning while reading and writing in intensive English medium instruction (EMI) university courses was investigated by Marina Dodigovic in Chapter Seven. A significant

Current Issues in Reading, Writing and Visual Literacy: An Introduction

5

vocabulary growth was found when bottom-up reading was paired with the use of tools and strategies conducive to vocabulary learning. Chapter Eight looks at a commercial reader for the teaching of Chinese as a foreign language. Han Lin identifies a number of weaknesses in terms of the logical arrangement of cultural content, the authenticity of expression, and the appropriateness of cultural reflections. By using dialogues that present a realistic image of contemporary Chinese culture and linking content to current language usage trends and habits in China, student understanding of Chinese culture at the beginner level can be enhanced. The final Chapter in this Part, Chapter Nine, provides an investigation of the national English Reference Word List (ERWL) in Taiwan. Through comparisons of the wordlist with the General Service List (GSL), the Academic Word List (AWL), and the Corpus of Contemporary American English (COCA), Lee-Yen Wang shows that while the ERWL is a reasonably defined list, it misses 52 academic words defined in the AWL, and 20 of them are with low or moderate ranks in the COCA’s 500K Frequency Word List. This study also shows that a vocabulary list is audience oriented and it can be greatly influenced by the teaching context.

Current Issues in Writing Literacy Part Two of this volume is a compilation of five papers on writing literacy. In Chapter Ten, Päivi Pietilä investigates the collocations used by advanced learners of English in their MA dissertations and compares that with a similar corpus by native English writers. Results showed that the English as a foreign language (EFL) writers used fewer collocations than the native English writers, but the distribution of the different collocation types was very similar across the groups, with collocations consisting of general vocabulary being the most frequent type. Erroneous collocations produced by the non-native writers were traceable to their L1. The use of formulaic sequences in the writings of Japanese EFL students was examined by Natsumi Okuwaki in Chapter Eleven. It was found that while there is a relationship between the use of formulaic expressions and essay quality, there was no association between formulaic language use and L2 proficiency suggesting that there is a threshold of L2 proficiency below which there is little relationship between the use of formulaic expressions and L2 proficiency. The use of interactive and interactional strategies in the results and discussion chapters of L2 doctoral writers was examined by Emmaline Lear and Huifang Li in Chapter Twelve. The results showed similarity in

6

Chapter One

the type and frequency of transitions in interactive metadiscourse, and boosters and hedges in interactional metadiscourse used by the L2 writers. These results will enable supervisors and academic advisors to better support L2 students develop their authoritative stance in thesis writing. The use of argumentative essay writing skills by university EFL students was the topic of Peiling Xing’s study in Chapter Thirteen. The study compared students’ perceptions about their use of argumentative essay writing skills with their actual essay writing performance. The results showed the students’ confidence levels of their argumentative essay writing skills use was substantially consistent with their actual performance. The last chapter in this Part of the volume, Chapter Fourteen, deals with teacher and student perceptions of written corrective feedback (WCF) at the postgraduate level. Naif Althobaiti demonstrates that while the teachers’ and students’ perceptions were congruent on the importance of WCF, they were at odds in terms of the amount of WCF used and the directness of the WCF.

Current Issues in Visual Literacy The third and final part of this volume, Part Three, comprises three chapters on visual literacy. In Chapter Fifteen, Mariko Boku compared how EFL learners infer the higher-order intention shown in reversed polarity tags (RPTs) in Audio-Visual (AV) materials, movie scripts and a short cartoon and whether Japanese learners’ higher-order inferenceoriented schema (HIOS) transfer occurs in the comprehension of RPTs in the second language. Results showed that learners tended to infer meaning by using AV information, which helped them build their schemas and was more helpful in the inference of higher-order intention from RPTs because it included the characters’ intonation, eye movement, and facial expression. In Chapter Sixteen, Danica Kelly, Gary Williams and Susan Morrison investigated the relationship between television viewing and the use of American vocabulary in Australian children under five years of age. The results indicated two-thirds of the children assessed produced American vocabulary, however, their use of American vocabulary was not significantly related to viewing increased amounts of television. The results of the study are presented in reference to current linguistic learning theories. The final chapter in the volume, Chapter Seventeen, is a position paper that discusses how digital curation can be understood as a new literacy in a

Current Issues in Reading, Writing and Visual Literacy: An Introduction

7

society that constantly produces exponentially increasing amounts of information. Nayara de Barros presents Storify, a curatorial platform which can help with the critical processing of information, and can also broaden perspectives on a great range of themes.

Summary and Conclusions While this is a small collection of studies on language literacy topics, the papers included in this volume investigate issues that are both typical of the field and also issues that have arisen from the recent developments in the field such as globalisation and the use of technology. Even though the list of topics represented in this volume is not by any means exhaustive, it does provide an insight into current research and the global trends in this field. It is hoped that the collection of research papers in this volume will be of use to researchers, language teachers, teacher educators, curriculum developers, and language materials designers. Policymakers and educators can use the findings to infuse research-based instruction into first language and second language literacy programs. Following the recommendations from the authors of each chapter, researchers can further enrich this knowledge base by focusing on the specific gaps in our knowledge and extending and validating these research studies in order to enable language education programs to better address classroom literacy issues and provide adequate support to language learners.

Acknowledgements This volume represents a collection of selected papers from the 17th World Congress of the International Association of Applied Linguistics (AILA2014) which was held in Brisbane, Australia. All papers included in this book underwent a rigorous double-blind review process. Initially 34 proposals were received of which 21 were selected for a double blind review process that involved a number of notable academics from different universities around the world. Through this process 16 papers were selected. These papers underwent further review and editing before being published in this book. Below is the list of academics (in alphabetical order) who were involved in the double blind review process.

8

Thomai Alexiou Sally Ali Tandy Bailey Ellie Cavalcanti Heidi Ciarlo Rania Jabr Liz Jones Emmaline Lear Daniel Mangrum Neil McBeath James McLaughlin Hala Nur Josephine O’Brien Denise Ozdeniz Daniel Perrin Ziad Rafhi Julie Riddleburger Ian Sinnot Helen Weston

Chapter One

Aristotle University United Arab Emirates University Higher Colleges of Technology Higher Colleges of Technology Higher Colleges of Technology American University of Cairo Zayed University University of Canberra Higher Colleges of Technology Sultan Qaboos University Higher Colleges of Technology University of Khartoum Zayed University Higher Colleges of Technology Zurich University of Applied Sciences Higher Colleges of Technology Khalifa University of Science and Technology Higher Colleges of Technology Higher Colleges of Technology

References Brown, J. S. (2001). Learning in the digital age. In The Internet and the university: Forum (pp. 71-72). Fast Facts. (2015). National Center for Education Statistics. Retrieved from https://nces.ed.gov/fastfacts/display.asp?id=96 Lloyd, A. (2003). Information Literacy The meta-competency of the knowledge economy? An exploratory paper. Journal of Librarianship and Information Science, 35(2), 87-92. Mirra, N. (2014). What is the purpose of literacy instruction in the 21stcentury classroom? Listening to the voices of high school English teachers. California English, 20(1), 9-11. UNESCO, (2005). Education for all. Literacy for life. Global Monitoring Report 2006. Retrieved from http://www.uis.unesco.org/Library/Documents/gmr06-en.pdf

ISSUES IN READING LITERACY

CHAPTER TWO MIDDLE GRADE TEACHERS’ PERCEPTIONS OF APPLIED READING ACTIVITY STRUCTURES AND STRATEGIES VERSUS OBSERVED PRACTICE IN SECOND LANGUAGE CLASSROOMS NANCY ALLEN

Abstract Since 2006, the language of instruction for science and mathematics in Qatar’s K-12 schools has changed twice; first, from Arabic to English and then, in 2012, from English to Arabic. Supporting English-based science literacy is thus a question of significant concern. This study, based on the proposition that strategic reading is important for comprehension, examines teachers’ self-reported use of literacy strategies and compares the findings to data from classroom observations. Ninety-eight teachers of English, science, and scientific English at the preparatory level completed a survey in which they indicated their use of reading activity structures and strategies on a Likert-type four-point scale. Observers noted the frequency of use of these same activity structures and strategies during 24 classroom visits. Findings indicated that teachers extensively overreported their use of reading activity structures and strategies. Comparing and contrasting the findings of these separate instruments highlights the strengths, weaknesses, and areas of potential misconceptions regarding reading instruction, especially as it relates to the content area of science. The findings have important implications for the professional development of teachers in this context.

Middle Grade Teachers’ Perceptions in Second Language Classrooms

11

Introduction In 2004, the small, oil-rich State of Qatar began a system-wide reform of its education system (Brewer, Augustine, Zellman, Ryan, Goldman, Stasz, & Constant, 2007). As part of this reform, science was taught in English from the elementary grades through college to enable students to fully participate in the greater scientific community. In 2012, policy makers reversed this decision, so that science would again be taught in Arabic. During 2013-2014, schools piloted a program to support Englishbased science literacy through three areas: science content in English courses, select scientific vocabulary in English in science class, and a new program in scientific literacy because the policy makers still felt that scientific literacy in English was important for full participation in the scientific community and wanted to make that opportunity available to all students. This research is part of a three-year funded project seeking to understand the ways in which reading was taught and practiced in these three venues with the goal of improving reading in science at the middle grades level in Qatar. As professional development for teachers is part of the project, this part of the study aimed to identify those areas of concern that should be addressed in the professional development program and also to identify areas for improvement in the current programs. This chapter describes our efforts to understand teachers’ self-perceptions of tasks related to reading and strategies to support literacy in science, how teachers incorporate reading tasks and classroom activities into their classes, what reading strategies they model and/or teach, and, finally, to understand the relationship between teachers’ self-perceptions (or at least self-reporting) and their practices in the classroom in relation to supporting students’ scientific literacy.

Background The theoretical perspective of the present study is that the effective use of reading strategies by second language (L2) learners increases comprehension (Cohen, 2011; Grabe, 2004; National Reading Panel, 2000; Pressley, Billman, Perry, Reffitt, & Reynolds, 2007). Traditionally, reading instruction was based on the assumption that the ability to read is determined by knowledge of vocabulary, morphology and the sentence structure of a language; however, more recent research (Taylor, Pearson, Peterson, & Rodriguez, 2003) has shown that strategic approaches to comprehension processes have a higher positive correlation to academic

12

Chapter Two

(reading and writing) growth and that the use of multiple strategies improves students’ comprehension (Gutherie, Wigfield, Barbosa, Perencevich, Taboada, Davis, Scafiddi, & Tonks, 2004; Guthrie, Wigfield, Humenick, Perencevich, Taboada, & Barbosa, 2006; Van Keer & Verhaeghe, 2005). Baker and Brown (1984a, 1984b) also suggested that comprehension increases, if readers strategically interact with text to construct meaning. Sweet and Snow (2003) added that comprehension is especially important in the later elementary grades as it forms the foundation for academic achievement in secondary school and preparation for the 21st century workforce. Research further supports the importance of teachers scaffolding students’ learning and use of reading strategies in the content areas (Finkbeiner, Knierim, Smasal, & Ludwig, 2012). The theoretical framework of this study, therefore, was that the explicit use and teaching of reading and comprehension strategies would enhance ESL learners’ comprehension of science. The L2 students in this study were being asked to read and comprehend science content. As Lee’s (2005) review of the research on science education for L2 English learners shows, this is a significant challenge. Science readings contain specialized forms of language and text (Chung & Berry, 2000; Luykx, Lee, & Edwards, 2008; Schleppegrell, 2004). These materials may require extensive use of inference (Best, Rowe, Ozuru, & McNamara, 2005) and the comprehension of technical vocabulary (Miller, 2009) and have unique genre patterns (Stoller, Jones, Costanza-Robinson, & Robinson, 2005). Better readers in both a first language (L1) and L2 explicitly monitor their comprehension while reading and can apply multiple, cognitive and meta-cognitive strategies before, during, and after reading to improve comprehension (Cohen, 2011; Grabe, 2004; McKeown, Beck, & Blake, 2009; Taylor, Stevens, & Asher, 2006). In a meta-analysis of 23 studies, Taylor, Stevens, and Asher (2006) determined that specifically teaching L2 students how to effectively use strategies increases their comprehension. Many researchers (Cervetti, Pearson, Barber, Hiebert, & Bravo, 2007; Cohen, 2010; Cohen & Gomez; 2008) thus advocate integrating the teaching of reading strategies appropriate for these texts with content area instruction. Another strand of research important to this study is the accuracy of self-reported data from teachers about their classroom practices. Although this area of study has been reported in the literature since the early 1970s (Weiss, 1973), there is neither an abundance of research nor agreement concerning the degree of accuracy of self-reported data (Ganey, 2010; Hook & Rosenshine, 1979). Kaufman and Junker (2011) did find, however, that although the degree of accuracy of self-reporting by teachers

Middle Grade Teachers’ Perceptions in Second Language Classrooms

13

varied from school to school, those schools with higher levels of accuracy also had higher levels of student achievement. This would suggest that understanding the relationship between self-reported data and classroom practices may have much to offer to professional development in education.

The Study A descriptive methodology was employed in this study. The purpose was to provide as true a picture as possible of the phenomena under study (Shenton, 2004). Data for this study consisted of two main sources. The first was an online survey that we created in Arabic that was sent to science, scientific English, and English teachers in preparatory (grades seven and eight) schools in Qatar. The survey had been developed and then revised in reference to feedback from individuals experienced in teacher education, English as a second language (ESL), K-12 teaching, and professional development. The survey collected demographic data about the teachers and then asked them to rate the frequency with which they engaged in specific practices related to the organization of reading. (e.g., how the reading activity was organized) and to the teaching, modeling and/or use of specific strategies for comprehension. A Likerttype four-point scale was used (1=I don’t do this, 2=I rarely do this, 3=I sometimes do this, and 4=I frequently do this). To help gain a better understanding of the nature of reading tasks in the curriculum, there were seven descriptive phrases related to activity structure targeted in both the survey and observations. These dealt with whether students read aloud and/or silently, whether the teacher read aloud, whether there were activities that required reading, whether topics associated with a reading were discussed before or after reading, and whether readings other than the text were included. There followed 28 strategies for reading and reading instruction. The strategies were divided into categories according to purpose: global understanding, support for reading, vocabulary learning, and compensation strategies. Although there is no universally accepted way to categorize strategies (Hsiao & Oxford, 2002; Oxford & Cohen, 1992), we drew from a number of scholars to form our categories (Abbott, 2006; Anderson, 1991; Block, 1992; Chamot & El-Dinary, 1999; Cohen, 2011; Hosenfeld, 1977; Mokhtari & Sheorey, 2007). The seven activity structures (descriptive terms used to identify the form of a reading activity) and 28 strategies were also listed in the classroom observation form that was used to collect data during the observations. The online

14

Chapter Two

survey had three vocabulary learning strategies that were not on the observation form, specifically: teach students to use mnemonics, use vocabulary flash cards, and have students write sentences with vocabulary words in them. These were excluded from the observation form as it was considered unlikely that they would be demonstrated in randomly selected classrooms. The sample was stratified, consisting of teachers within 12 Qatari preparatory schools that were targeted for data collection during the baseline investigations. Government schools in Qatar are differentiated by the year in which they became part of the education reform process (Brewer et al., 2006), which also equates to the number of years they have taught mathematics and science in English. The schools that entered the reform during the same year are called a cohort. Schools were chosen from different cohorts, with four schools from early cohorts, four from mid-reform cohorts, and four from recent cohorts. In Qatar, schools are differentiated according to gender; for this reason within each group of four schools from similar cohorts there were two boys’ and two girls’ schools. The sample was therefore stratified according to grade level, cohort, and gender. The online survey was made available to all teachers of science, English, and scientific English in the 12 schools. Complete responses were received from 98 teachers, which represented a return rate of approximately 80%. The demographics of the sample are representative of the teaching population in Qatar. The males in the sample, all of whom teach in the boys’ schools, are almost exclusively expatriates. In girls’ schools, there are some Qatari female teachers, but again the majority are female expatriates who have been educated in their home countries and often have taught there as well. For the overwhelming majority (96%), their highest qualification was a bachelor’s degree. Most of the teachers in the sample were experienced; more than a third or the survey respondents had 12 or more years of teaching experience; only 12% had less than that or had three years (see Table 2-1). In addition to the data obtained from the survey, we observed classes in the 12 target schools, using an observation instrument with the same activity structures and strategies as the survey (see Table 2-2). For practical reasons and to minimize disruption in the schools, the observations were limited to science and scientific English classes. In each school, the administrators selected the classes to be observed.

Middle Grade Teachers’ Perceptions in Second Language Classrooms

15

Table 2-1: Demographics of Survey Sample. Question Highest degree earned Highest degree earned outside of Qatar Previously taught in other countries

Years of experience

Options Bachelor’s Degree Master’s Degree Yes No Science teacher English teacher Scientific English 0-3 4-7 8-11 12+

% 96 4 90 10 62 85 71 12 24 31 34

Table 2-2: Organization of activity designs and strategies included in survey and observations.

Sample

Activity Design

Reading Strategies

Online Survey N=98 46 science/13 scientific English/39 English) Activity Design (7) descriptive phrases Student reads aloud Student reads silently Teacher reads aloud Activities that required reading Topics that require reading discussed prior to reading Topics that require reading discussed after reading Outside readings are included

Observations N=24 12 science/12 scientific English

Global understanding (8) Support (8) Vocabulary (9) Compensation (3)

Global understanding (8) Support (8) Vocabulary (6) Compensation (3)

16

Chapter Two

Data were entered into frequency tables and analyzed through traditional mathematical procedures (calculating means, modes, range, standard deviation, etc.).

Results and Discussion On the surveys, the teachers reported using all the activity structures and strategies at least sometimes or frequently, with the mean for 32 of the 35 items (activity structures plus strategies) on the survey at or above 3.0 on a four-point scale, with a range of 2.7 to 3.7 (see Tables 2-3 and 2-4). The survey responses would thus indicate that there is frequent attention to strategy instruction by teachers across the curriculum, but the observational data present a different picture. We observed an average of five classroom activities during a 50-minute science class, with an average of three of the five classroom activities involving some kind of reading. In the scientific English classes, there were more classroom activities, an average of seven, with most of them involving reading (an average of six). We observed students reading aloud during at least one activity in every observation but one. Silent reading was also common. Interestingly, in 10/24 observations (42%) the teacher also read aloud, although it was one of the lower scoring items on the survey (2.9/4.0). The observers saw only two examples of teachers engaging students in discussion about the reading before engaging in reading and having students read from materials other than the text. Table 2-3: Comparison of self-reported and observed activity design. Activity Design

Students read aloud from text Students read silently Students do exercises Class reads then talks Teacher reads aloud Class talks then reads Students read, not from text

Survey mean (N=98) 3.2

Descriptive Range

% Observed (N=24)

Sometimes

96

3.4 3.7 3.2 2.9 3.1 2.7

Sometimes Frequently Sometimes Sometimes Sometimes Sometimes

79 63 50 42 8 8

Middle Grade Teachers’ Perceptions in Second Language Classrooms

17

Table 2-4: Comparison of self-reported and observed strategy (taught, modeled, or used). Strategy

Global Strategies Read all before answering Predict what the reading is about Access prior knowledge Talk about the organization Provide a reason for the reading Predict what’s next Evaluate predictions Form mental pictures Support Strategies Learn from pictures Summarize during reading Paraphrase Underline the main idea Underline words defined in text Create mental pictures Make notes in margin Read a whole sentence first Vocabulary Strategies Translate Explain difficult words first Give definition Break words into parts Guess meaning based on content Students use dictionaries Things words sound like Write words on cards Use words in writing sentences Compensation Strategies Force finish Soft reading aloud Re-reading

Survey mean (N=98)

Descriptive Range

% Observed (N=24)

3.4 3.0 3.5 3.0 3.2 3.1 3.2 3.2

Sometimes Sometimes Frequently Sometimes Sometimes Sometimes Sometimes Sometimes

4 4 0 0 0 0 0 0

3.6 2.9 3.5 3.4 3.4 3.2 3.2 3.3

Sometimes Sometimes Sometimes Sometimes Sometimes Sometimes Sometimes Sometimes

33 8 8 4 4 0 0 0

3.0 3.4 3.3 3.3 3.4 3.0 3.4 3.4 3.2

Sometimes Sometimes Sometimes Sometimes Sometimes Sometimes Sometimes Sometimes Sometimes

54 33 13 13 8 0 0 0 0

3.0 3.0 3.4

Sometimes Sometimes Sometimes

4 0 0

18

Chapter Two

With respect to specifically teaching or modeling strategies, observational data differed even more from reported data. Six of eight global understanding strategies listed were not observed in any classrooms. While students were observed reading, they were primarily reading for the purpose of completing activities and worksheets, for example reading vocabulary words and finding the answers to specific questions, but not engaging in extended text reading where passage gist or main ideas were important. The only global strategies that were observed (each observed in one classroom only) were the teacher reminding the class to read everything before trying to answer questions about it (read all before answering) and the teacher asking the students what might come next in the text (predict what’s next). These findings are consistent with research (Nahmias, 2010; Ness, 2010; Nichols, Zellner, Rupley, Willson, Kim, Mergen, & Young, 2006; Stoller, Jones, Costanza-Robinson, & Robinson, 2005) that indicates that teachers in the content area may feel restricted in teaching reading strategies by established policies and a pressure to cover content standards. The focus in the observed classrooms seemed to be to find the right answer and complete the assigned activities, that is, to fulfill their content area responsibilities. A larger percentage of the listed support strategies were observed. Two of the eight were observed in two classrooms (summarizing and paraphrasing), and underlining main ideas and words defined in the text were each observed in one class. Learning from pictures, however, was observed in 33% of the classrooms. Obviously this was a strategy that teachers were comfortable using. With respect to modeling, the most commonly observed strategies related to vocabulary. This is not surprising given the emphasis in the scientific English courses on vocabulary learning. Translation was observed in 54% of the classes, and explaining difficult words in 33% of the observations. What is surprising, however, is that even the vocabulary strategies were not encouraged more frequently. There were seven additional strategies from our target list that were either not observed or observed in three or fewer classes. Of the three compensation strategies, only telling students they may need to force themselves to finish reading (forced finish) was observed at all. Figures 2-1 to 2-5 graphically show the comparison of teachers’ selfreported use of literacy strategies and the observations of strategy teaching, modeling, or use within the science, English, and scientific English classes. To make the comparison, we calculated the percentage of respondents who answered that they sometimes or frequently engage in a practice (i.e., response of 3 or 4) on the survey and compared that to the

Middlee Grade Teacheers’ Perceptionss in Second Lannguage Classroo oms

19

percentage oof classes wheere a strategy was modeledd or explicitly taught at least once. A Across all straategy types, a high h percentagge of teacherss reported that they enggaged in a praactice, but we did not obserrve them doing g it. Only in activity sttructure are thhe observed fiigures high ennough to be in n anyway comparable.. Only one of these activiity structures,, students rea ad aloud, was the obseerved occurrennce higher thaan the reportedd occurrence. ͳͲͲ ͻͲ ͺͲ ͹Ͳ ͸Ͳ ͷͲ ͶͲ ͵Ͳ ʹͲ ͳͲ Ͳ ‡’‘”–‡† •‡ „•‡”˜‡† •‡

Figure 2-1:: A comparisson of the reeported use oof activity sttructures with observved use of acttivity strategiies. As Figurre 2-1 shows, although alll activity struuctures were observed, o they were alll, with the exception of stu udents read alo loud, observed d at lower levels than reported. Thhis discrepanccy was most pronounced with the strategies off discussing a text prior to reading and rreading from materials m other than thhe texts.

Chapter Two

20

ͳͲͲ ͻͲ ͺͲ ͹Ͳ ͸Ͳ ͷͲ ͶͲ ͵Ͳ ʹͲ ͳͲ Ͳ

Reported Use U Observed Use U

Figure 2-2: A comparisoon of the reported use of global strategies with the observeed use of glob bal strategies. As Figuure 2-2 showss, the use of global strateggies was greaatly overreported annd under-usedd. This led us to questtion whether teachers understood how to teach or model thee strategy. Preevious researcch in this area has inddicated that teachers in the content area tend to avoid d teaching reading strattegies, as theyy may not feell qualified to ddo so (Nahmiias, 2010; Nichols, Zeellner, Rupleey, Willson, Kim, Mergeen, & Young g, 2006). Teachers seeemed to valuue the use off these strateggies, as they reported using them at a high leveel, but did nott demonstratee this in the cllassroom. We also connsidered that teachers had misconceptioons about the nature of the strategy and what it meant m to use it..

Middlee Grade Teacheers’ Perceptionss in Second Lannguage Classroo oms

21

ͳʹͲ ͳͲͲ ͺͲ ͸Ͳ ͶͲ ʹͲ Ͳ

Reported d Use Observed d Use

Figure 2-3:: A comparisson of the reeported use of support strategies s with the observed use off support stra ategies. As notedd previously, strategies to support vocaabulary build ding were used to a ggreater extentt than those in i any other category. Th his is not surprising. IIn Nahmias’s study (2010), content teachhers indicated that they found learning vocabularry to have th he greatest im mpact among potential reading straategies on conntent area ach hievement, annd thus they tended t to focus their ssupport of reaading in this area. a This apppears to be th he case in this study ass well. As Figurre 2-5 showss, only the strrategy describbed as telling g students that sometim mes they have to force theemselves to ccontinue readiing, even though they think it is boring, forced finish, fi was em mployed. Why the other two strategiees, clearly reqquiring no speecialized experrtise to teach, were not employed iss not clear from m these data.

Chapter Two

22

ͳͲͲ ͻͲ ͺͲ ͹Ͳ ͸Ͳ ͷͲ ͶͲ ͵Ͳ ʹͲ ͳͲ Ͳ

Reported Use U Observed Use

Figure 2-4: A comparisoon of the reported use of vvocabulary strategies s with the observed use off vocabulary strategies. ͳͲͲ ͺͲ ͸Ͳ ͶͲ ʹͲ Ͳ

Reporteed Ues Observeed Use

Force finish

Soft reading aloud

Re-readingg

Figure 2-5: A comparrison of thee reported u use of comp pensation strategies w with the obserrved use of co ompensation strategies.

Middle Grade Teachers’ Perceptions in Second Language Classrooms

23

Conclusion The findings from this study were helpful in structuring the professional development that followed. They indicated that teachers were aware of, and, at least at some level, valued, most of the investigated strategies for developing and supporting students’ strategic reading abilities, as they reported using these strategies at high frequency levels. For some reason, however, teachers were not, in fact, doing so. Drawing from the research and from our own inferences, we concluded that teachers may either not recognize the impact these strategies have on achieving content standards and expectations or do not feel competent in teaching and/or modeling these strategies. Teachers may also have misconceptions about the nature of the strategy and/or how to use it during reading. Our decision was to ensure that these elements were strongly embedded in the professional development of these science, English, and scientific English teachers. Further research may shed light on the validity of our conclusions and the effectiveness of the professional development designed on them.

Acknowledgement This article was made possible by NPRP grant # 4-1172-5-172 from the Qatar National Research Fund (a member of the Qatar Foundation). The statements made herein are solely the responsibility of the author.

References Abbott, M. L. (2006). ESL reading strategies: Differences in Arabic and Mandarin speaker test performance. Language Learning, 56(4), 633– 670. doi:10.1111/j.1467-9922.2006.00391.x Anderson, N. J. (1991). Individual differences in strategy use in second language reading and testing. The Modern Language Journal, 75(4), 460–472. Barna, J. (1974). Invasion of privacy as a function of test set and anonymity. Perceptional and Motor Skills, 38(3c), 1028–1030. Best, R. M., Rowe, M. P., Ozuru, Y., & McNamara, D. S. (2005). Deeplevel comprehension of science texts: The role of the reader and the text. Topics in Language Disorders, 25(1), 65–83. Block, E. L. (1992). See how they read: Comprehension monitoring of L1 and L2 readers. TESOL Quarterly, 26(2), 319–343. doi:10.2307/3587008

24

Chapter Two

Brewer, D., Augustine, C., Zellman, G., Ryan, G., Goldman, C., Stasz, C., & Constant, L. (2007). Education for a new era: Design and implementation of K–12 education reform in Qatar. MG-548-QATAR, Santa Monica, Calif.: RAND Corporation. Retrieved from http://www.rand.org/pubs/monographs/MG548/ Chamot, A. U., & El-Dinary, P. B. (1999). Children’s learning strategies in language immersion classrooms. The Modern Language Journal, 83(3), 319–338. Chung, T., & Berry, V. (2000). The influence of subject knowledge and second language proficiency on the reading comprehension of scientific and technical discourse. Hong Kong Journal of Applied Linguistics, 5(1), 187-225. Cohen, A. (2011). Strategies in learning and using a second language. Upper Saddle River, NJ: Prentice Hall. Finkbeiner, C., Knierim, M., Smasal, M., & Ludwig, P. H. (2012). Selfregulated cooperative EFL reading tasks: Students' strategy use and teachers' support. Language Awareness, 21(1-2), 57-83. Fuller, C. 1974. Effect of anonymity on return rate and response bias in mail surveys. Journal of Applied Psychology, 59(3), 292–6. Fulner, S. M., & Frijters, J. C. (2009). A review of self-report and alternative approaches in the measurement of student motivation. Educational Psychology Review, 21(3), 219-246. Goh, J., Lee, O., & Salleh, H. (2010). Self-rating and respondent anonymity. Educational Research, 52(3), 229-245. doi: 10.1080/00131881.2010.504060. Good, T., & Brophy, J. (2007). Looking in classrooms (10th ed.). Upper Saddle River, NJ: Allyn & Bacon. Grabe, W. (2004). Research on teaching reading. Annual Review of Applied Linguistics, 24(1), 44–69. Guba, E. G. (1981). Criteria for assessing the trustworthiness of naturalistic inquiries. Educational Resources Information Center Annual Review Paper, 29(2), 75-91. Guthrie, J., Wigfield, A., Barbosa, P., Perencevich, K., Taboada, A., Davis, M., Scafiddi, N., & Tonks, S. (2004). Increasing reading comprehension and engagement through concept-oriented reading instruction. Journal of Educational Psychology, 96(3), 403-423. (ERIC Document Reproduction Service No. EJ685000). Guthrie, J., Wigfield, A., Humenick, N., Perencevich, K., Taboada, A., & Barbosa, P. (2006). Influences of stimulating tasks on reading motivation and comprehension. Journal of Educational Research,

Middle Grade Teachers’ Perceptions in Second Language Classrooms

25

99(4), 232-245. (ERIC Document Reproduction Service No. EJ744233). Holgraves, T. (2004). Social desirability and self-reports: Testing models of socially desirable responding. Personality and Social Psychology Bulletin, 30(2), 161-172. Hook, C. M., & Rosenshine, B. V. (1979). Accuracy of teacher reports of their classroom behavior. Review of Educational Research, 49(1), 111. Hosenfeld, C. (1977). A preliminary investigation of the reading strategies of successful and non-successful second language learners. System, 5(2), 110–123. Hsiao, T., & Oxford, R. L. (2002). Comparing theories of language learning strategies: A confirmatory factor analysis. The Modern Language Journal, 86(3), 368–383. doi:10.1111/1540-4781.00155 Kidder, C., & Brinkman, P. (1971). When directness is the better part of valour. Journal of Personality and Social Psychology, 18(2), 238–46. Lee, O. (2005). Science education with second language learners: Synthesis and research agenda. Review of Educational Research, 75(4), 491-530. Luykx, A., Lee, O., & Edwards, U. (2008). Lost in translation: Negotiating meaning in a beginning ESOL science classroom. Educational Policy, 22(5), 640-674. McKeown, M. G., Beck, I. L., & Blake, R. G. K. (2009). Rethinking reading comprehension instruction: A comparison of instruction for strategies and content approaches. Reading Research Quarterly, 44(3), 218-252. Miller, L. (2009). Engineering lectures in a second language: What factors facilitate students’ listening comprehension? Asian EFL Quarterly Journal, 11(2), 8-30. Mokhtari, K., & Sheorey, R. (Eds.). (2007). Reading strategies of firstand second-language learners: See how they read. Norwood, MA: Christopher-Gordon Publishers, Inc. Nahmias, C. (2010). Content teachers’ perceptions of the factors that promote or inhibit infusion of content area reading strategies into instruction. Doctoral Dissertation, Florida State University, USA. Retrieved from http://diginole.lib.fsu.edu/cgi/viewcontent.cgi?article=3409&context=etd National Institute of Child Health and Human Development. (2000). Report of the National Reading Panel. Teaching children to read: An evidence-based assessment of the scientific research literature on reading and its implications for reading instruction (NIH Publication No. 00-4769). Washington, DC: U.S. Government Printing Office.

26

Chapter Two

Nichols, W. D., Zellner, L. J., Rupley, W. H., Willson, V. L., Kim, Y., Mergen, S., & Young, C. A. (2006). What affects instructional choice?: Profiles of K-2 teachers' use of reading instructional strategies and methods. Journal of Literacy Research, 37(4), 437-458. Oxford, R. L., & Cohen, A. D. (1992). Language learning strategies: Crucial issues of concept and classification. Applied Language Learning, 3(1-2), 1-35. Pressley, M., Billman, A, Perry, K., Reffitt, K., & Reynolds, J. (Eds). (2007). Shaping literacy achievement: Research we have, research we need. New York: Guilford Press. Sampson, M., Linek, W., Raine, I., & Szabo, S. (2013). The influence of prior knowledge, university coursework, and field experience on primary pre-service teachers use of reading comprehension strategies in a year-long, field-based teacher education program. Literacy Research and Instruction, 52(4), 281-311. Available from: ERIC, Ipswich, MA. Schleppegrell, M. J. (2004). The language of schooling: A functional linguistics perspective. Mahwah, NJ: Erlbaum. Shenton, A. (2004). Strategies for ensuring trustworthiness in qualitative research projects. Education for Information, 22(2), 63-75. Stoller, F. L., Jones, J. K., Costanza-Robinson, M. S. & Robinson, M. S. (2005). Demystifying disciplinary writing: A case study in the writing of chemistry. Across the Disciplines, A Journal of Language, Learning, and Academic Writing, 2. Retrieved from http://wac.colostate.edu/atd/Ids/stoller.cfm Sweet, A., & Snow, C. (2003). Rethinking reading comprehension. Solving problems in the teaching of literacy. New York: The Guilford Press. Taylor, A. M., Stevens, J., & Asher, J. W. (2006). The effects of explicit reading strategy training on L2 reading comprehension: A metaanalysis. In J. Norris, & L. Ortega, (Eds.), Research on language learning and teaching (pp. 231-344). Philadelphia, PA: John Benjamins. Van Keer, H., & Verhaeghe, J. (2005). Effects of explicit reading strategies instruction and peer tutoring on second and fifth graders’ reading comprehension and self-efficacy perceptions. Journal of Experimental Education, 73(4), 291. ERIC Document Reproduction Service No. EJ726370. Weiss, J. (1973). Validating and improving instruments for describing openness of school programs. Toronto: Ontario Institute for Studies in Education.

Middle Grade Teachers’ Perceptions in Second Language Classrooms

27

Wildman, R. C. (1977). Effects of anonymity and social setting on survey responses. The Public Opinion Quarterly, 41(1), 74–9.

CHAPTER THREE QATARI MIDDLE SCHOOL STUDENTS’ PERCEPTIONS OF STRATEGIES FOR L1 AND L2 READING DUDLEY W. REYNOLDS

Abstract This chapter uses a hierarchical cluster analysis to identify latent groupings of reading strategies on questionnaires completed by Qatari middle school students (N = 414) completing a scientific reading in Arabic (L1) and then completing a related reading in English (L2). Students were divided into high and low performing groups based on a selected response comprehension measure. Findings indicate that the high and low performing students identified similar strategy groupings following the Arabic reading. With respect to the English reading, however, the high performing students grouped strategies into smaller and more distinctive groupings than they had for the Arabic text. The low performing students had more groupings for the English reading as well, but they were not as focused; they also reported using fewer strategies. The findings indicate the importance for L2 reading of clear strategy constructs. They also suggest that constructs existing for L1 reading may not be specific enough to be of use in the L2.

Introduction Differences in the use of explicit cognitive and metacognitive strategies have long been held as one of the key factors that distinguishes more effective from less effective readers in both first language (L1) (Paris, Lipson, & Wixson, 1983; Pressley & Afflerbach, 1995) and second language (L2) (Anderson, 1991; Block, 1992; Cohen, 2011; Hosenfeld, 1977; Koda, 2005; Mokhtari & Sheorey, 2007; Rubin, 1975). As a result

Qatari Middle School Students’ Perceptions

29

considerable attention has been paid to schemes for reading strategy instruction in both L1 (Hilden & Pressley, 2007; Pressley & Harris, 2006) and L2 (Carrell, Pharis, & Liberto, 1989; Chamot, 2005; Chamot, Anstrom, Bartoshesky, Belanger, Delett, Karwan, & Keatley, 2004); Hosenfeld, Arnold, Kirchofer, Laciura, & Wilson, 1981). One of the conclusions reached by a number of researchers is that effective instruction focuses on the use of multiple strategies as part of what might be termed strategic behavior rather than the application of individual strategies (Cohen, 2007; Grabe, 2008; Macaro, 2001). According to Anderson (1991, pp. 468–9) “the reader must know how to use a strategy successfully and orchestrate its use with other strategies”. This paper investigates how more- and less-successful Qatari middle school students grouped reading strategies when asked to report on their usage immediately after reading a text in their L1 (Arabic) and then again after a L2 (English) reading.

Background Qatari Educational System The study reported on in this chapter was conducted as part of the first stage of an educational design research project (Cobb, Confrey, Lehrer, Schauble, et al., 2003) aimed at improving Qatari middle school students’ comprehension of science texts. Since 2006 Qatar has participated in the Programme for International Student Assessment (PISA), which measures 15-year-olds’ abilities in reading, mathematics, and science using a test translated into the students’ first language (see www.oecd.org/pisa). Qatar has consistently ranked among the poorest performing countries. On the 2012 administration, Qatar ranked 62 of 64 countries with respect to science literacy (see gpseducation.oecd.org). This is problematic for a country that recognizes that its “future economic success will increasingly depend on the ability of the Qatari people to deal with a new international order that is knowledge-based and extremely competitive” (General Secretariat for Development Planning, 2008, p. 13). In the Qatar school curriculum, students begin encountering more advanced science readings and topics in middle school (Grades 7-9), and so this was deemed the best age to target with an intervention. Key to designing the intervention, however, was a need to understand how students at this age read science texts in both Arabic and English. Prior to the 2012-13 school year, science instruction for the majority of Qatari students was delivered through English (Brewer, Augustine,

30

Chapter Three

Zellman, Ryan, Goldman, Stasz, & Constant, 2007). The language of instruction was switched to Arabic in fall 2012 with the recognition that students still needed English scientific literacy. Science teachers were encouraged to incorporate occasional English readings; a one-hour-perweek course on ‘scientific English’ was added to the curriculum; and the Arabic language science textbook added English glosses for important terms. The need for both Arabic and English scientific literacy is further demonstrated by searching for an Arabic scientific term on local internet search engines such as google.com.qa. In a small scale research study by the author, a review of the first page of hits from searches for 15 chapter titles from a Grade 8 science textbooks indicated that 69% of the hits directed towards a site in Arabic, 29% to a site in English, with the rest of the sites being bilingual, Arabic and English, sites. In short, Qatari students and adults need to be able to read and integrate knowledge from science texts in both Arabic and English.

Reading Strategies Cohen (2011) defines language learning and use strategies as: “thoughts and actions, consciously chosen and operationalized by language learners, to assist them in carrying out a multiplicity of tasks from the very onset of learning to the most advanced levels of target-language performance.” ( p. 7)

He goes on to distinguish strategies from the skills they facilitate saying “a skill [is] the ability to do something such as looking up a word in a dictionary, whereas strategies are the consciously selected means used to operationalize this skill” ( p. 19). The examples in the literature of mental and motor actions that language users can engage in when reading strategically are extensive, however there is no clear consensus on how to categorize such actions (Hsiao & Oxford, 2002; Oxford & Cohen, 1992). Anderson (1991) identified 29 strategies divided into four categories (supervising, support, paraphrase, and establishing coherence) in his analysis of think-aloud protocols produced by college students during a reading test. Mokhtari and Sheorey (2007) in turn developed 30-item strategy awareness inventories for first and second language readers, the Metacognitive Awareness of Reading Strategies Inventory (MARSI) and the Survey of Reading Strategies (SORS) respectively, whose items they group into global reading strategies, problem solving strategies, and support strategies. Other authors (Abbott, 2006; Block, 1992; Chamot & El-Dinary, 1999; Hosenfeld, 1977) influenced by processing theories of

Qatari Middle School Students’ Perceptions

31

reading tend to dichotomize strategies into whether they address comprehension of extended sections of text, which Abbott (2006) refers to as ‘top-down’ strategies, or comprehension of smaller linguistic units, which she refers to as ‘bottom-up’ strategies. Research into strategy use and effectiveness has demonstrated considerable variation related to task and reader characteristics. In the context of L2 readers, specific questions have arisen about whether a certain level of L2 linguistic proficiency is necessary before students can or will use strategies they already use in their L1 (Alderson & Urquhart, 1984; Bernhardt & Kamil, 1995; Clarke, 1979; Cummins, 1979; Goodman, 1970; Koda & Zehler, 2007; Malcolm, 2012; Segalowitz, 2000) and whether differences related to age (Alsheikh, 2014; Alsheikh & Mokhtari, 2011) and cultural learning styles (Abbott, 2006; Erler & Finkbeiner, 2007; Fender, 2008) exist. Collectively these streams of research indicate the need to base any pedagogical intervention on baseline research into patterns of usage in the target population and context, and in particular ways in which the target readers cluster strategies. The present study considers the need to distinguish between the perceptions of more and less effective readers. Because of the multilingual nature of the Qatari context, it also considers whether these perceptions vary cross-linguistically. The research question therefore is: 1. Are there differences in the ways that high and low performing Qatari middle school students cluster strategies when reading science texts in their L1 (Arabic) and L2 (English)?

The Study The study involved 12 middle schools operated by Qatar’s Supreme Education Council. The sample was stratified to include 6 girls’ and 6 boys’ schools. Additionally, because the curricular reforms that introduced the instruction of science in English were implemented in different years at different schools, it was assumed that students’ English proficiency might vary in relation to the length of time their school used Englishmedium instruction. A second level of stratification was added therefore with four schools being early adopters of English, four mid, and four late. Each school was asked to administer the research instrument in two 7th Grade classes and two 8th Grade classes. Slightly more responses were received from female (N = 573) than male (N = 462) students, while the

32

Chapter Three

responses from each grade level were more balanced (Grade 7, N = 506; Grade 8, N = 529).

Instrument An instrument was created to gauge students’ performance and strategy usage when reading in Arabic and English (see Table 3-1). Different forms were created for the Grade 7 and Grade 8 participants based on curricular topics. Additionally, because it was not possible to access students for the purpose of piloting the instruments, two forms for each grade level were created as a way of checking whether a particular reading selection might be influencing what students reported about strategy usage. Table 3-1: Instrument sections and forms. Grade and Topic

Reading 1 (Arabic)

Grade 7 Genetics 550 Words

Reading Strategy Inventory 1 Reading 2 (English)

Grade 7 Steam Engine 423 Words

Grade 8 Smoking 549 Words

Grade 8 Solar Energy 432 Words

20 Items 582 Words

574 Words

651 Words

306 Words

Reading Strategy Inventory 2

21 Items

Comprehension Questions

3 Selected Response Questions based on Arabic Reading and 3 on English Reading; 1 Open Response Inference Item

Test Taking Strategy Inventory

7 Items

The instrument comprised three performance tasks: read an Arabic text, read an English text, then answer six selected and one open response items about the texts. After each of these tasks, students completed a questionnaire asking whether they definitely did not (1), maybe (2), or definitely did (3) use a particular strategy while performing the task. The decision to use a three-level scale followed Macaro and Erler’s (2008)

Qatari Middle School Students’ Perceptions

33

strategy report questionnaire for 7th Grade French as a foreign language learners. Analysis focused on the strategies used on the two reading tasks, and gauged performance level based on the number of selected response items answered correctly. Items for the strategy questionnaires were selected following a review of instruments and analyses of think-aloud protocol data reported in the literature (Anderson, 1991; Chamot et al., 2004; Cohen & Weaver, 2006; Grabe, 2008; Macaro & Erler, 2008; Mokhtari & Sheorey, 2007). Three general classes of strategies relevant for both the Arabic and English readings were identified: strategies focusing on general or global understanding such as previewing the text before reading (six items); strategies such as underlining main ideas that focus on supporting specific comprehension tasks (eight items); and strategies addressing comprehension difficulties such as reading difficult parts again (three items). The final class of items targeted vocabulary strategies. For the Arabic L1 reading, three items targeting the identification of difficult words, use of word structure for decoding, and guessing based on context were included. For the English L2 reading, two items targeting use of translation were included along with the items on word structure and guessing from context. In total there were 18 identically worded items asked on both questionnaires, 2 unique items asked about the Arabic text, and 3 unique items asked about the English. Since the project as a whole assumes that Qatari students need to be able to integrate knowledge from Arabic and English texts, the instrument combined three selected response items based on the Arabic text and three based on the English as part of a single exercise. Questions were not marked for which passage students should refer to. Within each set of three items, one was designed to be answered by searching through the passage for a factual detail, one required more general comprehension of the passage’s main ideas, and the final required respondents to make an inference.

Analysis As might be expected based on the lack of piloting, a number of difficulties emerged during the administration. First, the instrument’s length meant that many participants did not complete all items. For the present study, only participants who answered all of the reading strategy items about both the Arabic and English texts as well as all 6 selected response items were included. This reduced the sample size to a total of 414 participants (Grade 7 = 185, Grade 8 = 229; Female = 252, Male =

34

Chapter Three

162). However, the distributions by grade level, gender, and form appear to be roughly the same as the distributions among the respondents as a whole. Second, the comprehension questions proved to be difficult on all four forms with none of the participants who completed the six selected responses items getting them all right. The mean number of correct responses on the four forms ranged from 1.7 (SD = 1.0) on the Grade 7 steam engine readings to 2.4 (SD = 1.2) on the Grade 8 readings on smoking. With respect to gender, female participants (M = 2.1, SD = 1.2) scored higher than male participants (M = 1.8, SD = 1.1). A univariate analysis of variance indicated that the differences related to both form and gender were significant. To reduce the importance of these differences, reading performance was converted to a two-level scale with ‘more effective readers’ being defined as those who got four or five of the selected response items correct (N = 42) and ‘less effective readers’ being those who got from zero to three of the items correct (N = 372). With this scale, differences in reading performance related to form and gender were not significant in a univariate analysis of variance, and it was felt that this would be a more useful measure of reading performance. To identify latent groupings of reading strategy items within the two sets of participants (high performing and low performing), hierarchical cluster analysis was used. Independent analyses were conducted for the items that participants responded to following the Arabic reading and the items that participants responded to following the English reading. Hierarchical cluster analysis initially identifies pairs of variables which are the most similar and then proceeds to cluster them with other variables such that the clusters are maximally similar to each other and maximally dissimilar from members of other clusters. The process proceeds until all variables are part of a single cluster. A between-groups method, which maximizes the average distance between all pairs of objects in the different clusters, was used, with distance being measured by the squared Euclidian distance. The following section reports clusters identified through visual inspection of a tree diagram known as a dendogram that is produced by the cluster analysis procedure. The dendogram indicates the initial variable groupings but then also the ‘stretch’ required to include additional variables as members of the cluster. The grouping of two or more variables is indicated by a node that joins lines from the variables. With each successive node of the diagram, more variables are included in the cluster. If a greater stretch is required to add a member, this is represented by a greater distance between the nodes. While most successive nodes are

Qatari Middle School Students’ Perceptions

35

formed by merging previous nodes, it is also possible that some variables will not join in a cluster until very high in the diagram. These variables are typically interpreted as outliers that do not really cluster with any of the other variables. Identification of potentially meaningful clusters is done by visual inspection and identifying points where the distance between the nodes preceding the point is less than the distance required to go to the next level. It should be kept in mind that the ultimate usefulness of this procedure depends upon whether it reveals interpretable patterns in the data that would not have been otherwise apparent.

Results and Discussion The results of the study are discussed first in terms of the strategy clusters with the Arabic (L1) readings and then in terms of the strategy clusters with the English (L2) readings.

Strategy Clusters with the Arabic Reading When reading the Arabic text in their L1, the high and low performing students cluster strategy items from the questionnaire in similar ways. For each group, analysis of the respective dendogram suggested three main strategy clusters plus several outliers. Table 3-2 identifies the clusters for the high performing readers and Table 3-3 presents the clusters of the low performing readers. Cronbach’s alpha was used as a measure of scale reliability for each cluster, with all the identified clusters demonstrating acceptable to good reliability. The mean response for the items in each cluster was also computed as an indicator of whether the students perceived the group of strategies to be something they definitely did or did not use with higher values indicating greater usage. Interestingly, the three clusters seem to correspond to low, medium, and high usage categories. For both the low and high performing groups the least used cluster comprised three items related to physically marking the test paper (i.e., underlining or writing notes). The cluster has been labeled ‘Kinesthetic’ in the tables. Interestingly these three items emerged as a cluster on the English readings as well, and in each analysis they were the least used. The instrument’s instructions specifically indicated that students could write on the test booklet. Visual inspection of the booklets indicated, however, that students in fact made few marks on the test paper. Perhaps they were used to being told not to write on booklets for other tests.

36

Chapter Three

Table 3-2: High performing readers’ strategy clusters for Arabic reading. Cluster Kinesthetic Cronbach’s alpha = .84 Mean = 1.48

Bottom Up Cronbach’s alpha = .83 Mean = 2.07

Top Down Cronbach’s alpha = .66 Mean = 2.40

Note: N = 42

Reading Strategy Items 7. I underlined or circled the main ideas of paragraphs. 8. I underlined or circled words that were defined in the text. 9. I wrote notes on the paper about why a part of the text was important. 03. I noticed that the text was divided into sections, and I thought about the order of these sections. 04. I predicted what would come next in the text. 12. I stopped while reading and summarized the part I just read before continuing. 14. I repeated in my mind what parts of the text meant. 15. I thought about the meaning of difficult words in the text. 16. I thought about the roots of difficult words so that I could figure out their meaning. 17. I guessed the meaning of words or phrases by looking at other parts of the text. 19. I read difficult parts again. 01. I looked quickly at the whole text to understand what it was about before I started reading. 02. I thought about what I already know about chickens, diseases, and genes while I was reading. 05. I thought about whether my predictions were right or wrong. 06. I asked questions in my mind while I was reading about things I would like to know more about. 10. I made pictures in my mind to help me understand the text. 11. I tried to read a whole sentence before stopping to figure out what individual words meant. 13. I used the pictures to help me understand what parts of the text were about.

Qatari Middle School Students’ Perceptions

37

Table 3-3: Low performing readers’ strategy clusters for Arabic reading. Cluster Kinesthetic Cronbach’s alpha = .71 Mean = 1.57 Bottom Up Cronbach’s alpha = .61 Mean = 2.03

Top Down Cronbach’s alpha = .64 Mean = 2.29

Reading Strategy Items 7. I underlined or circled the main ideas of paragraphs. 8. I underlined or circled words that were defined in the text. 9. I wrote notes on the paper about why a part of the text was important. 04. I predicted what would come next in the text. 12. I stopped while reading and summarized the part I just read before continuing. 15. I thought about the meaning of difficult words in the text. 16. I thought about the roots of difficult words so that I could figure out their meaning. 17. I guessed the meaning of words or phrases by looking at other parts of the text. 01. I looked quickly at the whole text to understand what it was about before I started reading. 02. I thought about what I already know about chickens, diseases, and genes while I was reading. 05. I thought about whether my predictions were right or wrong. 06. I asked questions in my mind while I was reading about things I would like to know more about. 10. I made pictures in my mind to help me understand the text. 11. I tried to read a whole sentence before stopping to figure out what individual words meant. 14. I repeated in my mind what parts of the text meant.

Note: N = 372

The second cluster of strategies perceived by both the high and low performing readers in relation to the Arabic reading involved strategies that facilitate making sense of individual words or sections of text. Following the literature on reading strategies (Abbott, 2006), this cluster has been labeled as ‘Bottom up’ strategies. Finally, the most commonly used group of strategies for both groups invokes a more global orientation to processing focused on applying outside knowledge, making predictions, and trying to understand either sentences or the text as a whole. Again

38

Chapter Three

following Abbott (2006), these have been labeled as ‘Top down’ strategies. That these are the strategies Qatari students’ report using the most echoes Abbott’s (2006) finding that when adult Arab learners of English were compared with Mandarin learners of English, Arab learners had a higher preference for ‘top down’ strategies.

Strategy Clusters with the English Reading For the English reading, the high and low performing groups proved more distinct. Analysis of the dendogram for the high performing group suggested five smaller clusters all with acceptable to high reliability (see Table 3-4). In contrast to the results for the Arabic reading, the clusters seem to have a more specialized focus and do not necessarily differ based on the items’ mean. After the Kinesthetic cluster, a second cluster (Recursion) emerged comprised of two items indicating a willingness to revisit glossed keywords presented at the beginning of the English reading as well as difficult parts of the text. Next was a cluster of items that reference stopping, starting, and looking at other parts of the text. These have been labeled ‘Process Management.’ The fourth cluster (Translation) involves strategies that invoke alternate representational systems (i.e., Arabic, mental pictures, text graphics) as a tool for understanding. The final cluster involves connecting with previous knowledge, thinking about sequence, making predictions, and asking mental questions. These items all indicate a willingness to go beyond figuring out the surface meaning, and have been labeled ‘Extension’ strategies. It should be noted that it is not really clear why item 37 about breaking a word into parts would cluster with the strategies in this group. Of the five clusters, the Translation, Recursion, and Extension strategies were most commonly used. The process strategies, which some may have perceived as interrupting fluent reading since they involve stopping and starting, were less common. Table 3-5 presents the results for the low performing readers. Since the performance score was based on comprehension of both the Arabic and English texts, it is likely that the low performing group had particular difficulty with the English text.

Qatari Middle School Students’ Perceptions

39

Table 3-4: High performing readers’ strategy clusters for English reading. Cluster Kinesthetic Cronbach’s alpha = .91 Mean = 1.56 Recursion Cronbach’s alpha = .69 Mean = 2.26 Process Management Cronbach’s alpha = .75 Mean = 2.09

Translation Cronbach’s alpha = .82 Mean = 2.32

Extension Cronbach’s alpha = .79 Mean = 2.21

Note: N = 42

Reading Strategy Items 27. I underlined or circled the main ideas of paragraphs. 28. I underlined or circled words that were defined in the text. 29. I wrote notes on the paper about why a part of the text was important. 36. I went back to the translated vocabulary words at the beginning of the article while reading. 40. I read difficult parts again. 32. I stopped while reading and summarized the part I just read before continuing. 38. I guessed the meaning of words or phrases by looking at other parts of the text. 41. I forced myself to keep reading if I started to get bored or think about other things. 30. I made pictures in my mind to help me understand the text. 31. I tried to read a whole sentence before stopping to figure out what individual words meant. 33. I used the pictures to help me understand what parts of the text were about. 34. I repeated in my mind what parts of the text meant. 35. I translated important words into Arabic. 22. I thought about what I already know about genes and inherited characteristics while I was reading. 23. I noticed that the text was divided into sections, and I thought about the order of these sections. 25. I thought about whether my predictions were right or wrong. 26. I asked questions in my mind while I was reading about things I would like to know more about. 37. I tried to break difficult words into parts so that I could figure out their meaning.

40

Chapter Three

Table 3-5: Low performing readers’ strategy clusters for English reading. Cluster Kinesthetic Cronbach’s alpha = .70 Mean = 1.69 Difficulty Cronbach’s alpha = .43 Mean = 2.15 Translation Cronbach’s alpha = .74 Mean = 2.16

Looking Backwards and Forewords Cronbach’s alpha = .62 Mean = 2.06

Reading Strategy Items 27. I underlined or circled the main ideas of paragraphs. 28. I underlined or circled words that were defined in the text. 29. I wrote notes on the paper about why a part of the text was important. 37. I tried to break difficult words into parts so that I could figure out their meaning. 40. I read difficult parts again. 26. I asked questions in my mind while I was reading about things I would like to know more about. 30. I made pictures in my mind to help me understand the text. 31. I tried to read a whole sentence before stopping to figure out what individual words meant. 33. I used the pictures to help me understand what parts of the text were about. 34. I repeated in my mind what parts of the text meant. 35. I translated important words into Arabic. 36. I went back to the translated vocabulary words at the beginning of the article while reading. 38. I guessed the meaning of words or phrases by looking at other parts of the text. 22. I thought about what I already know about genes and inherited characteristics while I was reading. 23. I noticed that the text was divided into sections, and I thought about the order of these sections. 24. I predicted what would come next in the text. 25. I thought about whether my predictions were right or wrong. 32. I stopped while reading and summarized the part I just read before continuing.

Note: N = 372

Analysis of the dendogram for this group suggested four clusters rather than the five present in the high performing group. While generally still within an acceptable range, the alphas for the clusters also indicate lower

Qatari Middle School Students’ Perceptions

41

reliability than the scales for the high performing students. As with the high performing students, the second cluster has only two items, one of which again involves reading difficult parts over. The other item, however, is breaking difficult words into parts. This cluster therefore seems to involve strategies for dealing with difficulty and may reflect the lower performing students’ perception of the task. The third cluster has again been labeled ‘Translation’ because it includes all the items from that cluster in the high performing students’ results. However, it also includes items such as asking mental questions and guessing from context, which seem to only marginally relate. The final cluster brings together strategies that seem to involve looking backwards at earlier sections of the text and previous knowledge or forwards in the form of predictions. Reviewing and predicting are generally perceived as useful for building a global model of a text’s meaning, and this cluster may reveal how the lower performing students conceive of top-down sense making. One of the most interesting findings with respect to the lower performing students’ reported usage, however, is that the cluster with the highest mean score is the Translation cluster (M = 2.16) which is lower than the means for the higher performing students’ Translation, Recursion, and Extension clusters. Thus, the lower students seem to be reporting less overall usage of strategies on the English passage in contrast to the Arabic passage where the cluster means for the two groups were roughly parallel.

Conclusion Analysis of the clusters for the high and low performing students indicate clear differences in the way that Qatari students perceive their options when reading Arabic versus English texts. With the Arabic text, both groups reported using top-down strategies more frequently than bottom-up strategies. In summarizing the literature on reading strategies, Grabe (2008) concludes that more and less effective readers often report using the same strategies but what distinguishes them is the way they use the strategies. One of the limitations of this study was that it focused more on the students’ strategy constructs than how they used the constructs to aid comprehension (or not). With the English text, the students’ clusters seem more nuanced. The high performing students distinguish between strategies that invoke alternative representations such as pictures or L1 representations, strategies that manage the reading process, and strategies that prompt the reader to extend the text and make connections. The lower performing students also make more distinctions with respect to the English text than

42

Chapter Three

they did for the Arabic text, but the constructs seem fuzzier or harder to interpret than those of the higher performing students. Most importantly, the lower performing students also report using strategies less. The finding that the higher performing students used strategies more on the English text and had clearer constructs underscores the importance for L2 readers of being able to consciously invoke strategies to address specific needs while reading. Furthermore, because L1 reading processes may be more automated, they may not be as aware of the strategies they already use. Knowing that a reading can be approached from both topdown and bottom-up perspectives is probably not sufficient when they encounter the more difficult task presented by L2 reading. As a basis for pedagogy this suggests a clear need for consciousness-raising activities such as think-alouds which expose the cognitive and metacognitive aspects of reading. Teachers can read aloud a class text modeling their thought process as they do so. To encourage active attention, students can be given an inventory to check when they hear a strategy being modeled. This activity can be followed by having students perform a similar thinkaloud in small groups while the other students take notes or mark strategies and then have to provide feedback. The goal of such activities should always be to build students’ explicit awareness of what can be done, why, and how it helps.

Acknowledgement This article was made possible by NPRP grant # 4-1172-5-172 from the Qatar National Research Fund (a member of the Qatar Foundation). The statements made herein are solely the responsibility of the author.

References Abbott, M. L. (2006). ESL reading strategies: Differences in Arabic and Mandarin speaker test performance. Language Learning, 56(4), 633– 670. doi: 10.1111/j.1467-9922.2006.00391.x Alderson, J. C., & Urquhart, A. H. (Eds.). (1984). Reading in a foreign language: A reading problem or a language problem. London: Longman. Alsheikh, N. O. (2014). The perceived and actual use of metacognitive reading strategies by the UAE high school students. Journal of ELT and Applied Linguistics (JELTAL), 2(1), 140–153. Alsheikh, N. O., & Mokhtari, K. (2011). An examination of the metacognitive reading strategies used by native speakers of Arabic

Qatari Middle School Students’ Perceptions

43

when reading in English and Arabic. English Language Teaching, 4(2), 151–160. doi: 10.5539/elt.v4n2p151 Anderson, N. J. (1991). Individual differences in strategy use in second language reading and testing. The Modern Language Journal, 75(4), 460–472. doi: 10.1111/j.1540-4781.1991.tb05384.x Bernhardt, E. B., & Kamil, M. L. (1995). Interpreting relationships between L1 and L2 reading: Consolidating the linguistic threshold and the linguistic interdependence hypotheses. Applied Linguistics, 16(1), 15–34. doi: 10.1093/applin/16.1.15 Block, E. L. (1992). See how they read: Comprehension monitoring of L1 and L2 readers. TESOL Quarterly, 26(2), 319–343. doi: 10.2307/3587008 Brewer, D. J., Augustine, C. H., Zellman, G. L., Ryan, G. W., Goldman, C. A., Stasz, C., & Constant, L. (2007). Education for a new era: Design and implementation of K-12 education reform in Qatar (Monograph No. RAND_MG548). Santa Monica, CA: RAND Corporation. Retrieved from: http://www.rand.org/pubs/monographs/MG548/ Carrell, P. L., Pharis, B. G., & Liberto, J. C. (1989). Metacognitive strategy training for ESL reading. TESOL Quarterly, 23(4), 647–678. doi: 10.2307/3587536 Chamot, A. U. (2005). Language learning strategy instruction: Current issues and research. Annual Review of Applied Linguistics, 25, 112– 130. doi: 10.1017/S0267190505000061 Chamot, A. U., Anstrom, K., Bartoshesky, A., Belanger, A., Delett, J., Karwan, V., & Keatley, C. (2004). The elementary immersion learning strategies resource guide (2nd ed.). Washington, D.C.: National Capital Language Resource Center. Retrieved from http://www.nclrc.org/eils/ Chamot, A. U., & El-Dinary, P. B. (1999). Children’s learning strategies in language immersion classrooms. The Modern Language Journal, 83(3), 319–338. doi: 10.1111/0026-7902.00025 Clarke, M. A. (1979). Reading in Spanish and English: Evidence from adult ESL students. Language Learning, 29(1), 121–150. doi: 10.1111/j.1467-1770.1979.tb01055.x Cobb, P., Confrey, J., Lehrer, R., Schauble, L., & others. (2003). Design experiments in educational research. Educational Researcher, 32(1), 9–13. doi: 10.3102/0013189X032001009 Cohen, A. D. (2007). Coming to terms with language learner strategies: Surveying the experts. In A. D. Cohen & E. Macaro (Eds.), Language

44

Chapter Three

learner strategies: 30 years of research and practice (pp. 29–45). Oxford, UK: Oxford University Press. —. (2011). Strategies in learning and using a second language (2nd ed). New York: Routledge. Cohen, A. D., & Weaver, S. J. (2006). Styles and strategies-based instruction: A teachers’ guide. Minneapolis, MN: Center for Advanced Research on Language Acquisition, University of Minnesota. Cummins, J. (1979). Linguistic interdependence and the educational development of bilingual children. Review of Educational Research, 49(2), 222–251. doi: 10.3102/00346543049002222 Erler, L., & Finkbeiner, C. (2007). A review of reading strategies: Focus on the impact of first language. In A. D. Cohen & E. Macaro (Eds.), Language learner strategies (pp. 187–206). New York: Oxford University Press. Fender, M. (2008). Arabic literacy development and cross-linguistic effects in subsequent L2 literacy development. In K. Koda & A. M. Zehler (Eds.), Learning to read across languages: Cross-linguistic relationships in first-and second-language literacy development (pp. 101–124). New York: Routledge. General Secretariat for Development Planning. (2008). Qatar national vision 2030. Retrieved from: http://www.gsdp.gov.qa/portal/page/portal/gsdp_en/qatar_national_visi on/qnv_2030_document Goodman, K. S. (1970). Psycholinguistic universals in the reading process. Journal of Typographic Research, 4(2), 103–110. Grabe, W. (2008). Reading in a second language: Moving from theory to practice. New York: Cambridge University Press. Hilden, K. R., & Pressley, M. (2007). Self-Regulation through transactional strategies instruction. Reading & Writing Quarterly, 23(1), 51–75. doi: 10.1080/10573560600837651 Hosenfeld, C. (1977). A preliminary investigation of the reading strategies of successful and non-successful second language learners. System, 5(2), 110–123. doi: 10.1016/0346-251X(77)90087-2 Hosenfeld, C., Arnold, V., Kirchofer, J., Laciura, J., & Wilson, L. (1981). Second language reading: A curricular sequence for teaching reading strategies. Foreign Language Annals, 14(5), 415–422. doi: 10.1111/j.1944-9720.1981.tb01661.x Hsiao, T., & Oxford, R. L. (2002). Comparing theories of language learning strategies: A confirmatory factor analysis. The Modern Language Journal, 86(3), 368–383. doi: 10.1111/1540-4781.00155

Qatari Middle School Students’ Perceptions

45

Koda, K. (2005). Insights into second language reading: A cross-linguistic approach. New York: Cambridge University Press. Koda, K., & Zehler, A. M. (Eds.). (2007). Learning to read across languages: Cross-linguistic relationships in first- and second-language literacy development. New York: Routledge. Macaro, E. (2001). Learning strategies in foreign and second language classrooms. London: Continuum. Macaro, E., & Erler, L. (2008). Raising the achievement of youngbeginner readers of French through strategy instruction. Applied Linguistics, 29(1), 90 –119. doi: 10.1093/applin/amm023 Malcolm, D. (2012). Changes in awareness of academic reading strategies among Arab medical students. Arab World English Journal, 3(2), 4– 30. Mokhtari, K., & Sheorey, R. (Eds.). (2007). Reading strategies of firstand second-language learners: See how they read. Norwood, MA: Christopher-Gordon Publishers, Inc. Oxford, R. L., & Cohen, A. D. (1992). Language learning strategies: Crucial issues of concept and classification. Applied Language Learning, 3(1-2), 1-35. Paris, S. G., Lipson, M. Y., & Wixson, K. K. (1983). Becoming a strategic reader. Contemporary Educational Psychology, 8(3), 293–316. doi: 10.1016/0361-476X(83)90018-8 Pressley, M., & Afflerbach, P. (1995). Verbal protocols of reading: The nature of constructively responsive reading. Mahwah, NJ: Lawrence Erlbaum Associates. Pressley, M., & Harris, K. R. (2006). Cognitive strategies instruction: From basic research to classroom instruction. In P. A. Alexander & P. H. Winne (Eds.), Handbook of Educational Psychology (2nd ed., Vol. 2, pp. 265–286). Mahwah, NJ: Lawrence Erlbaum. Rubin, J. (1975). What the “good language learner” can teach us. TESOL Quarterly, 9(1), 41–51. Segalowitz, N. (2000). Automaticity and attentional skill in fluent performance. In H. Riggenbach & R. Schmidt (Eds.), Perspectives on fluency (pp. 200–219). Ann Arbor, MI: University of Michigan Press.

CHAPTER FOUR HOW DO ARABIC (L1) AND ENGLISH (L2) READING DIFFER? EVIDENCE FROM THINK-ALOUD PROTOCOLS ZOHREH R. ESLAMI, KATHERINE L. WRIGHT, AND SUNNI L. SONNENBURG-WINKLER

Abstract This study explores the first language (L1) and second language (L2) reading strategies used by middle school English learners in Qatar when presented with science texts in Arabic and English followed by questions to determine comprehension. Think-aloud protocols were utilized to investigate the on-line mental processes of these students as they encountered difficulties in reading the two texts. We used a threedimensional categorization scheme set forth by Stevenson, Schoonen, and de Glopper (2003) to compare participants’ L1 and L2 reading strategies. Results show that readers use higher proportions of language-oriented strategies, monitoring strategies, and above-clause strategies when reading texts in their L2 (English) than when reading in their L1 (Arabic). Overall, our readers focused more on language when reading texts in their L2. However, our findings do not show that this inhibits them from focusing on top-down global strategies.

Introduction Reading is critical to academic success and, as more countries are moving to include English in their K-12 curriculum, reading in English is becoming a particularly valuable skill. Yet second language (L2) readers of English who are still early in the language-learning process often

How Do Arabic (L1) and English (L2) Reading Differ?

47

experience difficulty, particularly when reading content-specific texts such as those frequently used in science classrooms. An important issue for researchers in the field of L2 reading is whether the reading process is fundamentally different in the first language (L1) as opposed to L2. Research in L1 and L2 reading has shown that good readers use strategies that are not used by poor readers (Grabe & Stoller, 2002; Jiménez, García, & Pearson, 1996). The study of reading strategies can provide researchers insight into what readers do while reading, as these are thought to reflect their on-line reading processes (Cavalcanti, 1987; Pressley & Afflerbach, 1995). To determine strategies used during reading, think-aloud protocols have been commonly employed in both L1 and L2 reading research to develop a taxonomy of reading strategies (e.g., Anderson, 1994), to compare L1 and L2 reading (e.g., Sarig, 1987), and to identify reading strategies of stronger and weaker readers (e.g., Block, 1986; Cohen, 2011a). This study explores the L1 and L2 reading strategies used by middle school English learners in Qatar when asked to think aloud while reading two science texts followed by questions to determine comprehension both in their L1 (Arabic) and L2 (English).

Reading Strategies in First and Second Language Research has shown that effective use of reading strategies increases comprehension (e.g., Cohen, 2011a, 2011b; Grabe, 2004). Better readers in both L1 and L2 explicitly monitor their comprehension while reading and can apply multiple cognitive and metacognitive strategies before, during, and after reading to improve comprehension (Cohen, 2011a, 2011b; Grabe, 2004; McKeown, Beck, & Blake, 2009). Yet there is no clear agreement on how to categorize these strategies and strategic actions (Oxford & Cohen, 1992). Cohen (2011b) offers a broad description, defining strategies as thoughts and actions, consciously chosen and operationalized by language learners, to assist them in carrying out a multiplicity of tasks from the very onset of learning to the most advanced levels of target-language performance. Others refer to reading strategies as ‘a bag of tricks’ that does not actually build reading skills but, instead, indirectly improves comprehension (Willingham, 2006, p. 39). These differing conceptions of the core value and purpose of reading strategies highlight the diverse ways in which reading strategies are discussed and addressed in the literature. Though the amount of research in this area is certainly vast, numbering over 500 studies in the last three decades (Willingham, 2006), there is no consensus on the terms and

48

Chapter Four

categories involved. This is underlined by the diverse aspects of strategies addressed by different studies; some have compared how groups of students differ in their strategy use (e.g., Griva, Alevriadou, & Semoglou, 2012; Yau, 2009), some focused more specifically on one overarching type of categorization such as direct versus indirect strategies (e.g., Rubin, 1987), and others have attempted to create a taxonomy of reading strategies for different types and ages of learners (e.g., Gürsoy, 2010). Such variety in reading strategy research has led to multiple interpretations of strategy definitions, classifications, and categorizations. In addition, there is some debate concerning the extent to which readers in L1 and L2 use similar or different strategies. Most studies have found that although there seems to be a large degree of similarity in the kinds of reading strategies used in L1 and L2 (e.g., Hardin, 2001; Jiménez, García, & Pearson, 1995), there are substantial differences in the frequency of particular types of strategies. Fitzgerald (1995) found that differences in L1 and L2 reading strategies were more quantitative than qualitative. In other words, learners tend to use different amounts of strategies in their L1 and L2 in addition to using different types of strategies.

Background Two competing hypotheses have been offered to account for reading strategy use in L1 versus L2. The first one is Clarke’s (1980) short circuit hypothesis, which states that low proficiency in a learner’s L2 results in a ‘short circuit’ of effective strategies. When strong L1 readers are confronted with a difficult L2 text, the top-down processing used in L1 reading changes to bottom-up processing in L2. The second hypothesis, proposed by Lee (1991), states that L2 readers simultaneously combine both bottom-up and top-down knowledge, and that successful L2 readers are oriented neither from the bottom or the top; they use both types of strategies as needed (bi-orientation hypothesis). Some researchers, such as Horiba (1990), suggest that even good L1 readers focus more on language problems when reading in their L2 (due to linguistic knowledge limits) and, thus, are inhibited in the use of global strategies; Walczyk (2000) terms this inhibitory processing. In other words, L2 readers – even those who are proficient L1 readers – may not use known reading strategies during L2 reading because they are focused on determining linguistic meaning (Horiba, 1990; Stevenson, Schoonen, & de Glopper, 2003). Conversely, others have noted that learners may use compensatory strategies when reading in their L2, meaning that they use

How Do Arabic (L1) and English (L2) Reading Differ?

49

more and different strategies in order to compensate for their lack of linguistic knowledge (Walczyk, 2000). To gain insight into this issue of compensatory versus inhibitory processing in L2 reading, it is important to compare participants’ input processing in both their first and second language. Stevenson et al. (2003) classified three dimensions of L2 reading strategy use: Orientation of Processing (focused on content or language), Type of Processing (monitoring/regulating the reading process, processing the text meaning itself, or re-reading), and Domain of Processing (focused on the linguistic elements at, above, or below the clause level). This three-dimensional scheme allows comparison between participants’ L1 and L2 reading strategies as well as examination of differences between high- and lowachieving readers. When considering that L2 readers may be inhibited by their linguistic constraints, it is important to determine whether they are equipped with appropriate reading strategies that are focused on language. Several studies differentiate between those strategies that are Language-Oriented and those that are Content-Oriented (Stevenson et al., 2003; Walczyk, 2000). Language-Oriented strategies are those intended specifically for processing the linguistic elements of a text while Content-Oriented strategies look to determine the meaning of the overall text itself. Language-Oriented strategies allow L2 readers to determine the immediate text, but these can be time-consuming and require a shift in readers’ attention away from the overarching meaning of the whole text, which may require their understanding of the full text and content to suffer, which aligns with Clarke’s (1980) short circuit hypothesis. Proficient readers, according to Walczyk (2000), carry out language processing automatically so that they are better able to focus on the main ideas presented in a text without getting bogged down by linguistic intricacies. Such readers tend to use more Content-Oriented strategies as they attend to things such as determining the gist of a text and activating their own schema in order to utilize prior knowledge. Due to the linguistic demands of L2 reading, these readers may tend to focus on language problems and use more Language-Oriented strategies.

Context of the Study The international spread of English has left an indelible mark on many countries, particularly their education systems. One such affected nation is the Middle Eastern country of Qatar, which has recently reorganized its educational structure. Qatar exhibits educational hybridity, which

50

Chapter Four

combines foreign systems and models with local traditions of education and cultural heritage. Educational institutions in Qatar have attempted to balance between preserving the country’s Arabic heritage and preparing its youth to participate in the global economy. In 2004, the State of Qatar began a system-wide reform of its education structure (Brewer, Augustine, Zellman, Ryan, Goldman, Stasz, & Constant, 2007). As part of this reform, science was taught in English from the elementary grades through college to enable students’ participation in the greater scientific community. In 2012, policymakers reversed this decision so that science would again be taught in Arabic. However, support for scientific English continues to be offered through three avenues: English vocabulary in science courses, science content in English courses, and a scientific literacy program. To assess how prepared students are for this intersection of science and English, this study examines the L1 and L2 reading strategies used by middle school English learners when presented with informational science texts.

The Study The present study combines the use of L1 and L2 reading strategies in order to explore similarities and differences exhibited by participants as they read science texts in both languages. The research questions are: 1. Do Qatari middle school students use the same proportions of regulatory, cognitive, and cognitive-iterative strategies in reading L1 vs. L2 texts? 2. Do Qatari middle school students use the same proportions of content-oriented and language-oriented strategies in reading Arabic (L1) and English (L2) texts? 3. Do Qatari middle school students use the same proportions of above-clause, clause, and below-clause strategies in reading L1 vs. L2 texts?

Participants The participants in this study were seventh and eighth grade students (N = 12 and N = 11, respectively) enrolled in independent schools in Qatar. As Qatari schools are not co-educational, we selected both girls’ and boys’ schools to achieve a gender balance. Furthermore, since the curricular reforms aiming to teach science in English were implemented in different years at different schools, it is reasonable to assume that students’

How Do Arabic (L1) and English (L2) Reading Differ?

51

English proficiency varied due to the length of time that their school used English-medium instruction. Efforts were made to have a balanced number of participants from different schools (early, mid-, and late adopters). Permission slips were sent home to parents, and teachers identified strong readers out of those for whom parental permission was granted.

Procedure To gain further insight into how L2 readers use strategies when they encounter difficulties in the text, we included think-aloud protocols in our data collection. According to Leow, Grey, Marijuan, and Moorman (2014), think-aloud protocols are widely used in researching and teaching second language acquisition as a means of accessing participants’ L2 cognitive processing, particularly in showing how input was processed or why participants handled L2 input in a certain manner. Compared to methods of eliciting input processing information after participants’ interactions, think-alouds allow researchers to collect richer data that “offers insights into the processes and strategies employed by learners as they interact with L2 data during the input and intake processing stages” (Leow et al., 2014, p. 114). Think-alouds have been used with both L1 and L2 readers as a way for teachers to know what reading strategies students are using and to help them use effective strategies in their reading (e.g., Richards & Anderson, 2003). Whereas most previous studies on L2 strategy use have used survey data, the present study utilizes think-alouds in addition to comprehension questions to examine Qatari middle school students’ strategy use in both their L1 and their L2. In the present study, students met one-on-one with a teacher-researcher during the normal school day. The students were told that they were going to read passages and answer questions at the end. They were assured that their performance on this task would not impact their grades. Students were asked to read aloud and explain their thinking in either Arabic or English. If a student was not pausing to voice their thoughts, the teacherresearcher would prompt the student to verbalize their thought processes (e.g., ‘What are you thinking about?’ or ‘What is going on in your head?’). This procedure is similar to other research using think-aloud protocols (e.g., Bowles, 2010). Students were presented with two informational scientific texts–one written in Arabic, one in English. The topics of these passages varied; however, each pair of readings focused on one subject (see Table 4-1). The second passage was not a direct translation of the first. At the end of both

52

Chapter Four

readings, students answered six multiple-choice questions (three questions for each reading) and one short answer question asking students to synthesize information from the readings. Within each set of three items, one item was designed to be answered by searching through the passage for a factual detail, one required more general comprehension of the passage’s main ideas, and the final required respondents to make an inference. Students could respond in Arabic or in English. Table 4-1: Summary of reading selections. Subject Area Mechanical

Life Sciences Life Sciences Physical Science

Title of Arabic Passage Manufacturing the Smallest Steam Engine in the World Revealing the Chicken’s Genome Smoking in the Car Breaks Toxic Limit Producing Clean Electricity at a Low Cost

Title of English Passage Stirling Engines

Genomics and the Human Genome Project What Is Nicotine? Photovoltaic Energy

Grade Level 7

N

7

7

8

5

8

6

5

All think-aloud sessions were audio recorded and transcribed. The transcriptions were then translated by the teacher-researcher into English, which allowed for notes to be made identifying when a child was reading, misreading, or thinking-aloud.

Coding We constructed a coding scheme following the work of Stevenson et al. (2003), which broadly categorized students’ utterances as either: regulatory, or those thoughts which reflected a regulation of the reading and learning process; cognitive, or thoughts reflecting a processing of the text or information; or cognitive-iterative, the rereading of text. Each of these categories was further subdivided to reflect whether the students were focused on language or content and how they were processing this information (see Figure 4-1).

How Do Arabic (L1) and English (L2) Reading Differ?

53

Planning Content

Monitoring Evaluating

Regulatory Planning Language

Monitoring Evaluating

Using the text behind Content

Using the Text Ahead Using Extratextual Knowledge

Above-Clause

Rest Paraphrasing

Cognitive

ThinkAloud Utterance

Translating Rest Language

Clause

Paraphrasing Translating Rest

Below-Clause

Paraphrasing Translating

Unsuccessful Above-Clause Successful Unsuccessful Cognitive-Iterative

Clause Successful Unsuccessful Below-Clause Successful

Figure 4-1: Flow chart for coding think-aloud utterances.

Results We first ran one-way ANOVAs to ensure there was no statistically significant difference between grade levels or reading tasks. The level of statistical significance was set at .05. We then compared the instances of

54

Chapter Four

regulatory, cognitive, and cognitive-iterative think-aloud utterances while reading in Arabic and English to identify how students approached these tasks differently. While there were no statistically significant differences in the total number of instances in each strategy category, there were differences in the uses within each strategy. All uses of strategies are reported in the number of instances each student demonstrated the strategy. While only statistically different strategies are presented in the tables here, complete tables, including the analyses for all of the strategies, are available in Appendix 4-A.

Type of Processing Table 4-2 shows the use of main strategies as far as the type of processing is concerned. As shown in both L1 (Arabic) and L2 (English), students used cognitive strategies the most (Arabic: 8.61; English: 13.17) followed by regulatory strategies (Arabic: 3.70; English: 3.56). Cognitiveiterative strategies were used the least (Arabic: 1.64; English: 1.83). Furthermore, cognitive strategies were used more while reading L2 texts (13.17) than when reading L1 texts (8.61). The proportion of other strategy types was similar when reading L1 and L2 texts. Table 4-2: Type of processing: Mean percentages and standard deviations.

Regulatory Cognitive CognitiveIterative

Arabic English Arabic English Arabic English

Mean 3.70 3.56 8.61 13.17 1.56 1.83

Standard Dev. 3.64 3.27 5.48 15.69 2.17 2.15

Regulatory The total use of regulatory strategies (see Appendix 4-A) was similar in Arabic and English (Arabic: 3.70 and English: 3.56). Monitoring was the most frequently used strategy among the regulatory strategies. As shown above in Figure 4-1, regulatory strategies can be language-oriented or content-oriented. Students were far more likely to use regulatory strategies to monitor their understanding of the content when reading in

How Do Arabic (L1) and English (L2) Reading Differ?

55

Arabic than in English (see Table 4-3). Specifically, students engaged in more monitoring and evaluation of the content when reading in Arabic. However, when reading in English, students were more likely to pause in order to evaluate their understanding of the language. The greater attention to language-related monitoring strategies while reading in English (L2) can be described in terms of the fact that readers recognized their lack of understanding of language in the text more frequently. Table 4-3: Statistically significant differences in the use of regulatory strategies.

Content Monitoring Content Evaluating Language Evaluating

Arabic English Arabic English Arabic English

Mean 2.04 .78 1.00 .09 .00 .35

Standard Deviation 2.10 1.56 2.15 .29 .00 1.03

Std. Error Mean .44 .33 .45 .06 .00 .21

Sig. (2Tailed) .002 .045 .005

Cognitive Students used cognitive processes differently when reading in Arabic and English (see Table 4-4). They were far more likely to use textual and extratextual knowledge to process the content when reading in Arabic. However, most language cognitive processes did not demonstrate a statistically significant difference between the English and Arabic readings. The only difference demonstrated was in translation, which was frequently demonstrated in English but not in Arabic as this was the students’ first language. Translation above-clause level had the highest mean compared to the other two levels (clause and word level).

Cognitive-Iterative Cognitive-iteration, demonstrated when students reread passages, was coded as either above-clause (rereading more than a clause), at clause, or below-clause (rereading less than a clause). There were no statistically significant differences between students’ uses of cognitive-iteration at any level between their English and Arabic readings.

Chapter Four

56

Table 4-4: Statistically significant differences in the use of cognitive strategies.

Mean

Standard Deviation

Std. Error Mean

Content Using Text Behind

Arabic

2.91

2.81

.59

English

.70

1.36

.28

Content Using Extratextual Knowledge

Arabic

1.04

1.97

.41

English

.30

.70

.15

Language Above-Clause Translating

Arabic

.04

.21

.04

English

4.04

6.74

1.41

Language Clause Translating

Arabic

.04

.21

.04

English

2.83

4.15

.87

Language Below-Clause Translating

Arabic

.00

.00

.00

English

.7

1.58

.33

Sig. (2Tailed) .003

.047

.010

.004

.046

Orientation of Processing While students tended to use more content-oriented strategies than language-oriented strategies when reading in Arabic (M = 7.33 vs. M = 4.96), this difference was not statistically significant at the .05 level (see Table 4-5). Conversely, in English the frequency of language-oriented strategies was much higher than the frequency of content-oriented strategies (15.88 vs. 2.39; p = .002). Furthermore, the difference between the mean frequencies of content-oriented strategy use between the two languages was also significant (p < .000).

How Do Arabic (L1) and English (L2) Reading Differ?

57

Table 4-5: Use of content and language strategies in L1 and L2.

Mean Arabic

English

Content Strategies Language Strategies Content Strategies Language Strategies

Standard Deviation

Std. Error Mean

7.35

4.16

.87

4.96

4.97

1.04

2.39

3.22

.67

15.87

17.26

3.6

Sig. (2Tailed) .125

.002

Content and language-oriented strategies (orientation of processing) can be either regulatory or cognitive. As shown in Table 4-6, the majority of Language-Oriented strategies in both languages were cognitive (English: M = 13.61; Arabic: M = 4.61). Among language-oriented subcategories, paraphrasing (Arabic: M = 4.35; English: M = 5.65) and translating (Arabic: M = .08; English: M = 6.77) were the most common cognitive languageoriented subcategories. Increase in cognitive language-oriented strategies in English can be described in terms of the frequent use of translation as a strategy in English. Inspecting the content-oriented subcategories indicates that extratextual (Arabic: M = 2.91; English M = .70) and using textbehind strategies (Arabic: M = 1.04; English: M = .30) were much more frequent in Arabic than in English. However, the frequency of text-ahead strategies is slightly higher in English (M = .09) than in Arabic (M = .04). Table 4-6: Use of language-oriented strategies in L1 and L2.

Arabic

English

Regulatory Language Cognitive Language Regulatory Language Cognitive Language

Mean

Standard Deviation

Std. Error Mean

.35

.83

.17

4.61

4.60

.96

2.26

2.43

.51

13.61

17.28

3.6

Sig. (2Tailed) < .000

.005

Chapter Four

58

Domain of Processing In both their L1 and L2, students used at-clause strategies most frequently, followed by above-clause strategies, then below-clause strategies (see Table 4-7). Below-clause strategies were used least in both languages (Arabic: M = .08; English: M = .83). In contrast, at-clause strategies were used most in both the students’ L1 and L2 (Arabic: M = 4.00; English: M = 7.88). Above-clause strategies and at-clause strategies were used noticeably more in English than in Arabic (English: M = 6.73 and M = 7.88 respectively; Arabic: M = 2.08 and M = 4.00). Paired sample t-tests revealed that students used a significantly higher proportion of these strategies at all levels in English than in Arabic. Table 4-7: Domain of processing in L1 and L2.

AboveClause AtClause BelowClause

Arabic English Arabic English Arabic English

Mean 5.13 9.60 .96 5.00 .09 .83

Standard Deviation 5.59 12.25 1.52 6.19 .29 1.72

Std.Error Mean 1.16 2.55 .32 1.29 .06 .36

Sig. (2Tailed) .040 .005 .054

Results indicate that, in general, there were more cognitive languageoriented strategies (e.g., paraphrasing and translations in English) that were directed toward above-clause and at-clause levels (higher domains) in English than in Arabic (above-clause: Arabic, M = 2.08; English, M = 6.73; at-clause: Arabic: M = 4.00; English: M = 7.88). There was a slightly higher level of cognitive-iterative strategies in English (M = 1.83) compared to Arabic (M = 1.56). However, the successful cognitive-iterative strategies were slightly higher in Arabic (M = 1.43) than in English (M = 1.17). The proportion of use was slightly more at above-clause level for Arabic (M = 1.31) than for English (M = 1.13) and conversely higher for at-clause level in English (M = .64) than in Arabic (M = .17). The proportion for below-clause level cognitive-iterative strategies was twice as much in Arabic (M = .08) than in English (M = .04). This means that readers reread both larger and smaller chunks of texts and slightly more often in Arabic than in English. Among language-oriented subcategories, paraphrasing (Arabic: M = 4.35; English: M = 5.65) and translating (Arabic: M = .08; English: M =

How Do Arabic (L1) and English (L2) Reading Differ?

59

6.77) were the most common cognitive language-oriented subcategories. Increase in cognitive language-oriented strategies in English can be described in terms of the frequent use of translation as a strategy in English.

Discussion The results of our study reveal that although there appears to be a similarity in the kinds of strategies used in L1 and L2 reading, there are considerable differences in the use of certain strategies. Readers appear to compensate for comprehension insufficiencies by shifting their attention to strategies that directly tackle specific comprehension problems. When reading texts in L2, the readers focused more attention on processing language-related issues than when reading in their L1, supporting the findings of previous research (e.g., Horiba, 1996, 2000; Zwaan & Brown, 1996). We did not find evidence that these readers compensated for their L2 deficiencies by using global strategies (Walczyk, 2000). Furthermore, content-oriented strategies were used less when reading in L2 than in L1. A higher proportion of language-oriented strategies and a lower proportion of content-oriented strategies were used when reading L2 texts (M = 15.88 vs. M = 2.39) than when reading L1 texts (M = 4.96 vs. M = 7.33). This was mainly due to translating and paraphrasing. Nevertheless, proportionally less of the content-related strategies does not necessarily indicate that readers were inhibited in processing the text content since readers used a notable amount of language-related strategies such as translation and paraphrasing in their L2. Therefore, our findings do not support the claim made in the literature that attention to comprehension deficits inhibits the amount of content processing (Clarke, 1980; Horiba, 1990). As far as type of processing is concerned, our findings show that readers in this study used primarily cognitive strategies in both their L1 and L2 followed by regulatory strategies then cognitive-iterative strategies. Cognitive strategies were used significantly more when reading L2 texts. The use of other strategy types was similar in both languages; students used similar amounts of regulatory strategies in L1 and L2 reading. This supports previous findings that readers are able to make use of regulatory (i.e., metacognitive) strategies in their L2 (Davis & Bistodeau, 1993; Yamashita, 2002). It is possible that lower language proficiency in L2 compared to L1 leads students to use a greater number of languageoriented regulatory (monitoring) strategies when reading L2 texts than when reading L1 texts.

60

Chapter Four

As it relates to the domain of processing, readers in both their L1 and L2 used clause strategies most frequently, followed by above-clause strategies then below-clause strategies (i.e., below-clause strategies were used least in both languages). Readers used a significantly (p = .05) higher proportion of above-clause and clause level strategies in English than in Arabic. This was mainly due to the use of more above-clause and clausal translation and paraphrasing in English than in Arabic. Contrary to the claim made by L2 researchers (e.g., Clarke, 1980) that L2 reading is typified by focusing on small chunks of text, and similar to Stevenson et al.’s (2003) study findings, we found that readers focus on larger chunks of text when reading in L2 than in L1. The readers mainly reread, translated, and paraphrased larger chunks of text in their L2 to compensate for comprehension problems.

Conclusion As in other studies, this study has its limitations. There were only twenty-three participants in this study. Future studies are suggested to use a larger sample size to be able to make stronger, more generalizable claims in the use of different strategies in L1 and L2 reading. Our readers were middle-school Qatari students; students from other backgrounds, of varying ages, and with different levels of language proficiency might yield alternative findings in strategy use compared with this study. However, the use of the multidimensional model proposed by Stevenson et al. (2003) has provided more precision, clarity, and depth in comparing students’ reading strategies in L1 and L2. Research has shown that knowledge about reading strategies makes a significant contribution to reading comprehension in both L1 and L2 (Janzen, 2008). It is our hope that the findings of this study will provide educators with insights and information about the uses of reading strategies and which strategies might be more effective to teach to middleschool Arabic-English bilingual readers.

Acknowledgement This article was made possible by NPRP grant # 4-1172-5-172 from the Qatar National Research Fund (a member of the Qatar Foundation). The statements made herein are solely the responsibility of the authors.

How Do Arabic (L1) and English (L2) Reading Differ?

61

References Anderson, R.C. (1994). Role of the reader’s schema in comprehension, learning, and memory. In R.B. Ruddell, M.R. Ruddell & H. Singer (Eds.), Theoretical models and processes of reading (pp. 469–482). Newark, Delaware: International Reading Association. Block, E. (1986). The comprehension strategies of second language readers. TESOL Quarterly, 20(3), 463-494. Bowles, M. A. (2010). The think-aloud controversy in second language research. New York: Routledge. Brewer, D. J., Augustine, C. H., Zellman, G. L., Ryan, G. W., Goldman, C. A., Stasz, C., & Constant, L. (2007). Education for a new era: Design and implementation of K-12 education reform in Qatar (Monograph No. RAND_MG548). Santa Monica, CA: RAND Corporation. Retrieved from http://www.rand.org/pubs/monographs/MG548/ Cavalcanti, M. C. (1987). Investigating FL reading performance through pause protocols. In C. Faerch & G. Kasper (Eds.), Introspection in second language research (pp. 230-250). Clevedon, UK: Multilingual Matters. Cohen, A. D. (2011a). L2 learner strategies. In E. Hinkel (Ed.), Handbook of research in second language teaching and learning (pp. 681–698). Abingdon, U.K.: Routledge. —. (2011b). Strategies in learning and using a second language (2nd ed.). Harlow, England: Longman Applied Linguistics/Pearson Education. Clarke, M. A. (1980). The “short circuit” hypothesis of ESL reading – Or when language competence interferes with reading performance. Modern Language Journal, 64(2), 203-209. Davis, J. N., & Bistodeau, L. (1993). How do L1 and L2 reading differ? Evidence from think aloud protocols. Modern Language Journal, 77(4), 459–471. Fitzgerald, J. (1995). English-as-a-second-language learners’ cognitive reading processes: A review of research in the United States. Review of Educational Research, 65(2), 145–190. Grabe, W. (2004). Research on teaching reading. Annual Review of Applied Linguistics, 24, 44-69. Grabe, W., & Stoller, F. (2002). Teaching and researching reading. New York: Longman. Griva, E., Alevriadou, A., & Semoglou, K. (2012). Reading preferences and strategies employed by primary school students: Gender, socio-

62

Chapter Four

cognitive and citizenship issues. International Education Studies, 5(2), 24-34. Gürsoy, E. (2010). Investigating language learning strategies of EFL children for the development of a taxonomy. English Language Teaching, 3(3), 164-175. Hardin, V. (2001). Transfer and variation in cognitive reading strategies of Latino fourth-grade students in a late-exit bilingual program. Bilingual Research Journal, 25(4), 539-561. Horiba, Y. (1990). Narrative comprehension processes: A study of native and non-native readers of Japanese. Modern Language Journal, 74(2), 188–202. —. (1996). Comprehension processes in L2 reading: Language competence, textual coherence, and inferences. Studies in Second Language Acquisition, 18(4), 433-473. —. (2000). Reader control in reading: Effects of language competence, text type, and task. Discourse Processes, 29(3), 223-267. Janzen, J. (2008). Teaching English language learners in the content areas. Review of Educational Research, 78(4). 1010-1038. Jiménez, R. T., García, G., & Pearson, P. (1996). The reading strategies of bilingual Latina/o students who are successful English readers: Opportunities and obstacles. Reading Research Quarterly, 31(1), 90112. Lee, J. F. (1991). On the dual nature of the second language reading proficiency of beginning language learners. In R. V. Teschner (Ed.), Assessing foreign language proficiency of undergraduates (pp. 198215). Boston: Heinle. Leow, R. P., Grey, S., Marijuan, S., & Moorman, C. (2014). Concurrent data elicitation procedures, processes, and the early stages of L2 learning: A critical overview. Second Language Research, 30(2), 111127. McKeown, M. G., Beck, I. L., & Blake, R. G. (2009). Rethinking reading comprehension instruction: A comparison of instruction for strategies and content approaches. Reading Research Quarterly, 44(3), 218-253. Oxford, R. L., & Cohen, A. D. (1992). Language learning strategies: Crucial issues of concept and classification. Applied Language Learning, 3(1-2), 1-35. Pressley, M., & Afflerbach, P. (1995). Verbal protocols of reading: The nature of constructively responsive reading. Hillsdale, NJ: Erlbaum. Richards, J., & Anderson, N. A. (2003). How do you know? A strategy to help emergent readers make inferences. The Reading Teacher, 57(3), 290-293.

How Do Arabic (L1) and English (L2) Reading Differ?

63

Rubin, J. (1987). Learner strategies: Theoretical assumptions, research history, and typology. In A. Wenden & J. Rubin (Eds.), Learner strategies in language learning (pp. 1-30). Englewood Cliffs, NJ: Prentice-Hall. Sarig, G. (1987). High-level reading in the first and foreign language: Some comparative process data. In J. Devine, P. L. Carrell, & D. Eskey (Eds.), Research in reading in English as a second language (pp. 108– 120). Washington: TESOL. Stevenson, M., Schoonen, R., & de Glopper, K. (2003). Inhibition or compensation? A multidimensional comparison of reading processes in Dutch and English. Language Learning, 53(4), 765-815. Walczyk, J. J. (2000). The interplay between automatic and control processes in reading. Reading Research Quarterly, 35(4), 554–566. Willingham, D. T. (2006). The usefulness of brief instruction in reading comprehension strategies. American Educator, 30(4), 39-50. Yamashita, J. (2002). Reading strategies in L1 and L2: Comparison of four groups of readers with different reading ability in L1 and L2. International Review of Applied Linguistics, 136, 1–35. Yau, J. (2009). Reading characteristics of Chinese-English adolescents: Knowledge and application of strategic reading. Metacognition & Learning, 4(3), 217-235. Zwaan, R. A., & Brown, C. M. (1996). The influence of language proficiency and comprehension skill on situationǦmodel construction. Discourse Processes, 21(3), 289-327.

Chapter Four

64

Appendix 4-A. Analyses for all the strategies Table 4-8: Regulatory strategy use.

Mean

Standard Deviation

Std. Error Mean

Arabic

.30

.93

.19

English

.43

1.47

.31

Arabic

2.04

2.10

.44

English

.78

1.56

.33

Arabic

1.00

2.15

.45

English

.09

.288

.06

Arabic

.00

.00

.00

English

.13

.46

.10

Arabic

.35

.83

.17

English

1.78

2.15

.45

Arabic

.00

.00

.00

English

.35

1.03

.21

Arabic

3.70

3.64

.76

English

3.56

3.27

.68

Content

Planning

.45

Monitoring

< .00

Evaluating

.04

Language

Planning

.19

Monitoring

< .00

Evaluating Total uses of Regulatory Strategies

Sig. (2Tailed)

.12

.82

How Do Arabic (L1) and English (L2) Reading Differ?

65

Language At-Clause

Language Above-Clause

Content

Table 4-9: Cognitive strategy use.

Mean

Standard Deviation

Std. Error Mean

Using the text behind

Arabic

2.91

2.81

.59

English

.70

1.36

.28

Using the text ahead

Arabic

.04

.21

.04

English

.09

.42

.09

Using extratextual knowledge

Arabic

1.04

1.96

.41

English

.30

.70

.15

Arabic

.09

.29

.06

English

.04

.21

.04

Arabic

.65

.98

.21

English

1.52

2.25

.47

Arabic

.04

.21

.04

English

4.04

6.74

1.41

Arabic

.09

.29

.06

English

.00

.00

.00

Arabic

3.70

4.35

.91

English

4.40

5.78

1.21

Arabic

.04

.21

.04

English

2.83

4.15

.87

Sig. (2Tailed) .00

.67

.05

Rest

.58

Paraphrasing

.09

Translating

.01

Rest

.16

Paraphrasing

.40

Translating

< .00

Chapter Four

Language Below-Clause

66

Arabic

.00

.00

.00

English

.00

.00

.00

Arabic

.00

.00

.00

English

.09

.29

.06

Arabic

.00

.00

.00

English

.70

1.58

.33

Arabic

8.61

5.48

1.14

English

13.17

15.69

3.27

Rest

N/A

Paraphrasing

.16

Translating

Total uses of Cognitive Strategies

.05

.09

How Do Arabic (L1) and English (L2) Reading Differ?

67

At-Clause

Above-Clause

Table 4-10: Cognitive-Iterative strategy use.

Mean

Standard Deviation

Std. Error Mean

Arabic

.09

.29

.06

English

.52

1.88

.39

Arabic

1.22

2.15

.45

English

.61

1.08

.22

Arabic

.00

.00

.00

English

.13

.46

.10

Arabic

.17

.49

.10

English

.52

.85

.18

Arabic

.04

.21

.04

English

.00

.00

.00

Arabic

.04

.21

.04

English

.04

.21

.04

Arabic

1.56

2.17

.45

English

1.83

2.15

.45

Unsuccessful

.29

Successful

.20

Unsuccessful

.19

Below-Clause

Successful

.10

Unsuccessful

.33

Successful

Total uses of Cognitive-Iterative Strategies

Sig. (2Tailed)

1.0

.68

CHAPTER FIVE LINGUISTIC DIFFERENCES BETWEEN ENGLISH AND SCIENCE TEXTS FOR ENGLISH LANGUAGE LEARNERS: A QUANTITATIVE ANALYSIS KATHERINE LANDAU WRIGHT AND ZOHREH R. ESLAMI

Abstract This study offers findings from an analysis of Qatari English-language school texts which highlight the needs of English Language Learners (ELLs) in content-area classes. Recent Qatari Education Reforms aim to simultaneously develop students’ content literacy in their first and second languages. Using Coh-Metrix, an automated online text analysis system, we make comparisons between the English-language texts used in English as a foreign language (EFL) and in science classes. These comparisons allow us to identify areas in science where students may need additional literacy support. It was found that the EFL texts tend to focus on basic communication skills, whereas the science texts require students to follow more complex text structures and vocabulary. By providing direct instruction in these recommended areas, students’ second language (L2) reading comprehension and content knowledge may improve.

Introduction Language is the medium through which concepts and skills are learned and deeper and more complex disciplinary understandings are constructed over time (DiCerbo, Anstrom, Baker, & Rivera, 2014; Nagy & Scott, 2000; Stahl, 1999; Wendt, 2013). The norms and patterns of language in different academic disciplines needs to be made explicit to students. As Christie (1985) argues, success in school is largely a matter of learning the

Linguistic Differences Between English and Science Texts

69

patterns of discourse and discipline specific language features through which academic concepts and skills are developed. Academic English is an important factor in students’ academic success, however, the distinctive characteristics of academic language in different subject areas can be difficult to learn. It is important to investigate the distinguishing characteristics of academic language in different subject areas to gain a better understanding of the challenges English language learners (ELLS) and other students face in content areas. All academic disciplines have a unique language and way of communication that must be directly taught to students (Warren, 2013). As scientists read texts, their goal is to gain a full understanding of the content, which requires thinking “about the phenomenon being presented in prose, to visualize it, and to manipulate it in formulas and equations” (Shanahan & Shanahan, 2008, p. 51). While this may be challenging for many students learning science, ELLs carry the additional burden of completing the task in a second language. Therefore, science teachers with second language students must identify how the texts they assign differ from those students are exposed to in language classes and understand how to support both students’ content-knowledge and language acquisition. The current study seeks to support this goal by conducting a quantitative linguistic analysis of two sets of texts used as part of Qatar’s bilingual education program. In this model, science is first taught in the students’ native language (L1) and later reinforced in English (L2), theoretically allowing students to develop both English and Arabic content-area knowledge and language skills. However, students need content-specific support in order to learn how to read and comprehend scientific texts written in a second language. The current study examines the linguistic demands of English language textbooks and science readings (written in English), to understand how these second language students’ content-literacy skills are being supported in English language class and where additional scaffolding is needed.

Background Many researchers note the centrality of language in content-area teaching and its importance for ELLs (e.g., Janzen, 2008). The distinguishing features and meanings of the academic language are different to the language used in informal social contexts. Moreover, academic language can also vary from one content-area to another. The academic uses of discipline-specific language as well as the meaning of individual words and genre structures need to be explicitly taught for

70

Chapter Five

students to succeed in school and to develop their content knowledge. To be able to instruct their students, teachers must understand how the syntactic and lexical features of their discipline’s language relate to effective comprehension (Janzen, 2008). Reading scientific material requires not only specialised language skills, such as specific academic vocabulary (Fang, 2005), but also higherorder thinking skills and metalinguistic knowledge (Carnine & Carnine, 2010). In addition, all science students must negotiate between their home language (common discourse used with friends and family), instructional language (the phrasing used by teachers to explain concepts), and science language (the sophisticated language used by textbooks and field experts) (Yore & Treagust, 2006). Teaching second language students requires an understanding that, in addition to learning content, students use language to learn and express their knowledge. Teachers must support L1 development and use researchbased strategies to build students’ content-area literacy skills (Lee & Buxton, 2013), which has been shown to improve ELLs academic performance (Carlo, August, McLaughlin, Snow, Dressler, Lippman, Lively, & White 2004). To provide students with the language support they require, content-area teachers in turn need support and guidance from literacy experts to ensure student success in second-language academic programs (Scott, 2013).

The Language of Science The vocabulary used in scientific literature provides a hurdle for ELLs. Shanahan and Shanahan’s (2008) think-aloud study demonstrated that scientific vocabulary is particularly challenging as many words have both general language and domain specific meanings. Science also has a habit of nominalization, or changing verbs to nouns, such as describing a “process of dissolution” rather than saying something dissolved (Shanahan & Shanahan, 2008, p. 52). These factors make reading and writing in science especially difficult for English language learners. According to Kintsch (1986), texts can be represented as either a textbase or situation model. The former is focused on the semantic content of the text, and a strong textbase representation reflects a general coherence of the text. The situation model, by contrast, is the mental representation that develops as a result of reading a text (Kintsch, 1986). Creating a situation model is especially critical in science education, where texts describe experiments and cause/effect relationships essential to understanding content (Shanahan & Shanahan, 2008). Kintsch’s (1986)

Linguistic Differences Between English and Science Texts

71

research established that just because a reader understands a text does not mean that he or she has created a situation model. For example, most of us can read and comprehend the instruction manual for a car, but that does not mean we have created a mental representation of the inner workings of an engine and would be able to describe it in our own words. The cognitive demands of reading and writing in a second language make it even more difficult for language learners to create an effective situation model. Therefore, ELL students need language classes in order to acquire sufficient vocabulary and rhetorical skills that will enable them to understand texts, as well as support in science classes in order to understand relationships and develop a situation model of the scientific texts. In order to address these concerns, the present study examined two sets of texts used with ELLs in Qatar to answer the following questions: 1. What are the linguistic differences between science and English texts used to teach EFL students? 2. What are the implications of these linguistic differences on students’ ability to read and comprehend texts for science class?

The Study Textbook Sample We examined four sets of texts used in Qatari independent schools; two English as a foreign language (EFL) textbooks and two sets of English-language science readings. The first EFL textbook, To the Top 2, was created by MM Publications from the United Kingdom. This textbook is the second in a series of English-language texts designed to promote young teenagers from beginning to intermediate levels of English language proficiency. The textbook and accompanying materials aim to combine grammar structures, language functions, pronunciation, and skill work (MM Publications, 2012). The second English-language text was Upstream, produced by Express Publications also from the United Kingdom. Much like To the Top, this text is part of a larger course aimed at developing all areas of language communication. This series also reflects the guidelines of the Council of Europe Framework (Express Publishing, 2012). Additionally, this book contains reading passages called ‘Culture Clips’, which included descriptions of holidays and authentic literature excerpts.

72

Chapter Five

We also analyzed two English language Science texts from Qatari independent schools, both locally authored. The first series, referred to as Science Related Readings, was designed by Qatar’s Curriculum Standards Office and aligns with Qatar’s science curriculum standards (Science Curriculum Standards Office, 2013a). Each text consists of three to five pages of reading followed by a few pages of review activities. The lessons in the second group of science texts, referred to here as Scientific English, begins with an image of an interactive Smartboard and graphics corresponding to the lesson. It is followed by a casual conversation between a teacher and student presenting the content. Each semester’s book contains 15 one-page lessons with two to three pages of structured activities.

Analysis Tool: Coh-Metrix The automated online text analysis system Coh-Metrix (Grasser, McNamara, Louwerse, & Cai, 2004) was used to gauge text cohesion. The system is available online for free and evaluates text for over 100 measures of cohesion, language, and readability with the goal of providing an instant gauge of text difficulty (McNamara, Graesser, McCarthy, & Cai, 2014). Developed by researchers at the University of Memphis, this tool has been used in a variety of studies including text analysis, discourse evaluation, and assessing student achievement (Duran, McCarthy, Graesser, & McNamara, 2007). Coh-Metrix has primarily been used to evaluate text and trade books (for example, see Smolkin, McTigue, & Yeh, 2013), allowing for a more holistic text analysis. Because CohMetrix focuses on overall text cohesion, we included passages containing a minimum of 20 words or two complete sentences. As Coh-Metrix produces 108 different variables for each analysis, it was necessary to narrow our focus to make the data manageable. To begin, we consulted Smolkin et al. (2013) who used Coh-Metrix to analyze the explanations in science trade books. These researchers focused on seven variables from Coh-Metrix (i.e., causal content, causal particles, causal verbs, causal cohesion, positive causal connectives, argument overlap, and stem overlap) to analyze the quality of explanations provided in science textbooks. To sort through the remaining variables, we referred to the Coh-Metrix indices descriptions provided online by the developers. We selected indices which directly related to our study and its purpose. The included indices fell into six broad categories: Readability, Text Easability, Referential Cohesion, Situation Model, Syntactic Complexity, and Word Information.

Linguistic Differences Between English and Science Texts

73

Readability. These first three indices examined were the Flesch Reading Ease, Flesch-Kincaid Grade Level, and the Coh-Metrix L2 Readability estimations. The Flesch and Flesch-Kincaid formulas have been widely used by both researchers and practitioners. These formulas consider the length of words and sentences but cannot account for content and context. The Coh-Metrix L2 Readability scale considers word overlap, syntax, and word frequency and has been shown to highly correlate with L2 readers’ performance on cloze tasks (McNamara et al., 2014). While not the primary focus of the current study, these measures have been validated through repeated use in research and educational settings and provide external validity for our claim that statistically significant differences existed between these two sets of texts. Text Easability. The objective of including these indices in Coh-Metrix is to “transcend traditional measures of readability that focus on surface characteristics of text” (McNamara, et al., 2014, p. 83). The Referential Cohesion index, which measures how explicit the text is in making connections between ideas, was included as we hypothesized that these explicit connections would aid second language reading comprehension. This variable is one of five that generally accounts for over half the variance in overall easability scores across writing samples (McNamara et al., 2014). Referential Cohesion. While the referential cohesion index included under text easability provides an overview of a text’s cohesion, the indices in this category delve deeper into the elements which make a text more or less cohesive. This category measures the overlap, or coreference, of content words between sentences. According to the developers, coreferences provide linguistic cues to aid readers in making inferences and connections between ideas in a text (McNamara et al., 2014). These measures have been used in prior research examining scientific literature, and results have demonstrated that higher values would make texts easier to comprehend (Smolkin et al., 2013). With second language students, overlapping content words would be less taxing on working memory. Situation Model. The situation model, or the mental representation that develops as a result of reading a text (Kintsch, 1986), is especially critical to understanding science content (Shanahan & Shanahan, 2008). The CohMetrix indices within the Situation Model category quantify the extent to which the text uses language that would facilitate the development of such mental representations (McNamara et al., 2014). We used two indices that measure causal verbs and particles as these are prevalent in high-quality science texts (Smolkin et al., 2013).

74

Chapter Five

Syntactic Complexity. Sentences become more difficult, or syntactically complex, when they are longer, have more words before the main verb, and more words per noun phrase (McNamara et al., 2014). The “Left Embeddedness” variable in the Coh-Metrix indices was included because a higher number of words before the main verb of the main clause would tax working memory for ELL students.

Analysis Following the work of Crossley and McNamara (2009), we conducted a descriptive discriminant analysis (DDA) to identify salient differences in the two types of text. DDA uses categorical group membership (i.e., the genre of a text) to describe continuous outcome variables (i.e., Coh-Metrix scores) (Huberty, 1994). As DDA is usually accomplished with fewer than 10 variables (Huberty, 1994), we needed to reduce the number of variables produced by Coh-Metrix. To select the variables for inclusion, we analyzed mean differences. As these samples did not meet the required homogeneity of variance assumption, t-tests rather than ANOVAs were used throughout. We identified Coh-Metrix items whose mean difference was statistically significant, and included these in the DDA.

Results Before delving into specific Coh-Metrix indices, the readability scores for the English and Science texts were considered. We conducted independent samples t-tests to determine if there were statistically significant differences between these two sets of texts and found that the difference was significant at the Į = .05 level for all three measures of readability. These results are detailed in Table 5-1. As a significant difference was found on these three established measures, we determined there was sufficient external validity to continue the Coh-Metrix analysis. Independent sample t-tests were conducted for the remaining indices and those with statistically significant differences are described in the following sections.

Linguistic Differences Between English and Science Texts

75

Table 5-1: Science and English texts’ readability scores.

Coh-Metrix Variable

Flesch Reading Ease

FleschKincaid Grade level

Coh-Metrix L2 Readability

Description Widely used reading difficulty formula. Defined as: 206.832 – (1.015 x ASL) – (84.6 x ASW). Range from 0 to 100 Widely used formula for determining text grade level. Ranges from 0 to 12. Defined as: (.39 x ASL) + (11.8 x ASW) – 15.59 Second language readability score developed by Coh-Metrix researchers

Science Text Mean (SD)

English Text Mean (SD)

70.36 (11.7)

82.64 (8.10)