Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Cultural variations in adult students' learning intentions for writing in English as a second language… Ostler-Howlett, Catherine E 1991

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1991_A8 O87.pdf [ 9.77MB ]
Metadata
JSON: 831-1.0078331.json
JSON-LD: 831-1.0078331-ld.json
RDF/XML (Pretty): 831-1.0078331-rdf.xml
RDF/JSON: 831-1.0078331-rdf.json
Turtle: 831-1.0078331-turtle.txt
N-Triples: 831-1.0078331-rdf-ntriples.txt
Original Record: 831-1.0078331-source.json
Full Text
831-1.0078331-fulltext.txt
Citation
831-1.0078331.ris

Full Text

C U L T U R A L VARIATIONS IN ADULT STUDENTS' LEARNING INTENTIONS FOR WRITING IN ENGLISH AS A SECOND LANGUAGE: A CASE STUDY by CATHERINE E. OSTLER-HOWLETT Bachelor of Arts (Honors), University of Victoria, 1985 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF LANGUAGE EDUCATION We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA June, 1991 © Catherine Ostler-Howlett, 1991 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of L c u A g i U C l g e h^ciuCod-Con The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT This stud)' describes the learning intentions of eleven adult students in the context of one English as a second language (ESL) writing class-5 Japanese, 3 Taiwanese, and 3 Europeans-and the extent to which their intentions for writing differed across the three cultural groups. Also described are the intentions of these students' teacher-an advocate of the process approach to teaching writing-and the extent to which the teacher's and students' intentions corresponded. Various data collection techniques were used, including classroom observations, self-reports of previous experiences studying and learning to write, verbal protocols while composing two writing tasks, and retrospective accounts after these writing tasks. The data were then tabulated and transcribed, and intention utterances from the transcripts coded for attention to aspects of writing, i.e., meaning and the expression of that meaning at the level of discourse organization and language use. Results indicated that these students attended mainly to discourse organization concerns at the "idea" level while performing ESL writing tasks. A narrative task elicited more attention to meaning (content organization), and an argument task elicited more attention to how to express that meaning (rhetorical organization) as well as significantly more intention statements than the narrative task. These intention statements proved to fit loosely with the teacher's objectives for each of the two writing tasks, as well as his attention to aspects of writing while reading over the students' compositions. Only slight differences emerged in what students of the different cultural groups attended to while they wrote, although these differences could not be assessed because of the small number of participants in the research. Overall, these students appeared to have accommodated, or socialized into, the teacher's priorities for their ESL writing class, displaying significant changes in their purposes for studying ESL writing over the duration of the course. The culture of the ESL classroom seemingly overrode the values for learning to write of the First language cultures of the individual participants. ii TABLE OF CONTENTS Abstract i i Table of Contents i i i L i s t of Tables v i L i s t of F igures v i i i Acknowledgements x i Chapter One: Introduction 1 1.1. Identification of the Problem 1 1.2. Purpose of the Present Study 3 1.3. Research Questions 4 Chapter Two: Review of Relevant Research 5 2.1. Intentions in E S L Wr i t ing Classes: Learn ing Content, Rhetorical and Composing Ski l ls, and Language 5 2.2. Cu l tu ra l Differences in Learn ing 9 2.3. The F i t Between Teachers ' Objectives and Students ' Learn ing Intentions in Wr i t i ng 13 2.4. Collecting Data 17 Chapter Three: Context and Methods 19 3.1. Part ic ipants 19 3.1.1. Selection of Participants 19 3.1.2. Profiles of the Student Part ic ipants 20 3.1.3. Educational Background of the Students 20 3.1.3.1. Self-rating of Learn ing Ab i l i t y 21 3.1.3.2. Self-rating of Wr i t ing Ab i l i t y 24 3.2. Setting 26 3.3. Da ta Collection 28 3.2.1. Questionnaires 29 3.2.2. Observations 29 3.2.3. Think-Aloud and Prompted Recall Protocols 30 3.2.3.1. Tasks 30 3.2.3.2. The Think-Aloud Procedure 31 3.2.3.3. The Prompted Recall Procedure 31 3.2.3.4. The Teacher 's Protocols 32 3.4. Coding of Da ta 32 3.4.1. The Students ' Think-aloud Protocols 32 3.4.2. The Students ' Retrospective Accounts 36 3.4.3. The Teacher 's Protocols 36 3.4.4. Rel iabi l ity Checks 37 3.5. Da ta Analys i s 38 i i i Table of Contents Chapter Four: Findings 40 4.1. Research Question 1: Overview of the Whole Group 40 4.1.1. Intentions in the Students ' Verba l Protocols 40 4.1.2. Intentions Stated in Retrospective Accounts 44 4.1.3. Summary for Research Question 1 46 4.2. Research Question 2: Cu l tura l Group Profiles 47 4.2.1. Research Question 2a: Intentions in the Students ' Verba l Protocols 47 4.2.1.1. Quantity of Intention Utterances 47 4.2.1.2. Intention Utterances in the Students ' Verba l Protocols for the Learn ing Record 50 4.2.1.3. Intentions as Stated in the Students ' Retrospective Accounts for the Learn ing Record 53 4.2.1.4. Intention Utterances in the Students ' Verba l Protocols for the Argument 53 4.2.1.5. Intentions as Stated in the Students ' Retrospective Accounts for the Argument 56 4.2.1.6. Summary for Research Question 2a 57 4.2.3. Research Question 2b: The Students ' Self-Reports of Background and Present Beliefs and Preferences 57 4.2.3.1. Activit ies in F i r s t Language Wr i t i ng Classes 58 4.2.3.2. Activit ies in Previous Engl ish Wr i t i ng Classes 61 4.2.3.3. Perceptions of the Present Class 64 4.2.3.4. Preferences 67 4.2.3.5. Relationship Between Experience and Preference 70 4.2.3.6. Summary for Research Question 2b 74 4.3. Research Question 3: The Teacher 's Perspective 74 4.3.1. Research Question 3a: The Teacher 's Objectives (Self-Reports) and Verba l Protocols 75 4.3.1.1. Objectives: Learn ing Record 75 4.3.1.2. Objectives: Argument 76 4.3.1.3. The Teacher 's Protocol Statements 77 4.3.1.4. The Teacher 's Protocol Statements by Cu l tura l Group: Learn ing Record 80 4.3.1.5. The Teacher 's Protocol Statements by Cu l tura l Group: Argument 83 4.3.1.6. Summary for Research Question 3a 85 4.3.2. Research Question 3b: Comparison of the Teacher 's and Students ' Intentions in Verba l Protocols 85 4.3.2.1. The Teacher 's Utterances Compared to the Students ': Whole Group Profile 85 4.3.2.2. The Teacher 's Utterances Compared to the Students': Cu l tura l Group Profiles 87 4.3.2.3. Summary for Research Question 3b 91 4.3.3. Research Question 3c: Change in the Students ' Purpose in Enrolling.... 92 4.3.3.1. Whole Group 92 4.3.3.2. Cu l tura l Group Profiles 95 4.3.3.3. Summary for Research Question 3c 100 iv Table of Contents Chapter F ive: Discussion and Implications 101 5.1. Discussion 101 5.1.1. Research Question 1: The Students ' Priorit ies in Wr i t ing 101 5.1.2. Research Question 2a: Cu l tura l Differences 103 5.1.3. Research Question 2b: The Students ' Perceptions of their Experience. 106 5.1.4. Research Question 3a: The Teacher 's Priorit ies 107 5.1.5. Research Question 3b: F i t Between the Teacher 's and Students ' Intentions 108 5.1.6. Research Question 3c: Change in the Students ' Global Intentions 109 5.2. L imitat ions and Assumptions 110 5.3. Implications 112 5.3.1. Implications for Learn ing and Teaching 112 5.3.2. Implications for Research 113 5.3.3. Suggestions for Further Study 114 References 116 Appendix A: Letters to Part ic ipants 120 Appendix B: Questionnaire 122 Appendix C: Instructions for Tasks 124 Appendix D: Instructions for Th ink ing Aloud 125 Appendix E: Questions for "Prompted Reca l l " 126 Appendix F: Instructions to the Teacher 127 Appendix G: Sample of a Student's Verbal Protocol 128 Appendix H: Sample of the Argument Task 131 Appendix I: Sample of the Teacher 's Verba l Protocol 132 v LIST OF TABLES Table 1 Self-Rating of Learn ing Abi l i ty by Student Part ic ipants Means of Cu l tura l Groups and Standard Deviations 22 Table 2 Self-Rating of Wr i t ing Ab i l i ty in Mother Tongue by Student Participants Means of Cu l tura l Groups and Standard Deviations 24 Table 3 Types of Intention Utterances in Verbal Protocols (as Percentages of Total Intention Utterances) Means of the Whole Group and Standard Deviations for Learn ing Record and Argument 42 Table 4 Proportions of Answers to Global and Local Questions in Retrospective Accounts Whole Group Profile for Learn ing Record and Argument 45 Table 5 Proportions of Types of Intentions in Retrospective Accounts Whole Group Profile for Learn ing Record and Argument 46 Table 6 Quant ity of Students ' Intention Utterances in Verba l Protocols (as Percentages of Total Utterances) Means of Cu l tura l Groups and Standard Deviations for Learn ing Record and Argument 48 Table 7 Types of Intention Utterances in Verba l Protocols (as Percentages of Total Intention Utterances) Means of Cu l tura l Groups and Standard Deviations for Learn ing Record 51 Table 8 Proportions of Types of Intentions in Retrospective Accounts Cu l tura l Group Profiles for Learn ing Record 53 Table 9 Types of Intention Utterances in Verba l Protocols (as Percentages of Total Intention Utterances) Means of Cu l tura l Groups and Standard Deviations for Argument 54 Table 10 Proportions of Types of Intentions in Retrospective Accounts Cu l tura l Group Profiles for Argument 56 v i List of Tables Table 11 K inds of Activit ies Students Remember Doing in F i r s t Language Wr i t ing Classes Means of Cu l tura l Groups and Standard Deviations 59 Table 12 K inds of Activit ies Students Remember Doing in Previous Engl i sh Classes Means of Cu l tura l Groups and Standard Deviations 62 Table 13 K inds of Activit ies Students Remember Doing in Their Present Engl ish Wr i t ing Class Means of Cu l tura l Groups and Standard Deviations 65 Table 14 K inds of Activit ies Students Would Prefer in Their Engl ish Wr i t ing Class Means of Cu l tura l Groups and Standard Deviations 68 Table 15 Types of Teacher 's Utterances (as Percentages of Total Utterances) Means of the Whole Group and Standard Deviations for Learn ing Record and Argument 78 Table 16 Types of Teacher 's Utterances (as Percentages of Total Utterances) Means of Cu l tura l Groups and Standard Deviations for Learn ing Record 81 Table 17 Types of Teacher 's Utterances (as Percentages of Total Utterances) Means of Cu l tura l Groups and Standard Deviations for Argument 83 Table 18 Students ' Purposes in Tak ing a Wr i t ing Class Means of the Whole Group (as Percentages) and Standard Deviations 93 Table 19 Students ' Purposes in Tak ing a Wr i t ing Class Means of Cu l tura l Groups (as Percentages) and Standard Deviations 96 vn LIST OF FIGURES Figure 1 Self-Rating of Learn ing Ab i l i t y by Student Part ic ipants Means of Cu l tura l Groups 23 Figure 2 Self-Rating of Wr i t ing Ab i l i ty in Mother Tongue by Student Part ic ipants Means of Cu l tura l Groups 25 Figure 3 Operational Definitions for Coding Categories 35 Figure 4 Types of Utterances in Verba l Protocols Means of the Whole Group for Learn ing Record and Argument 43 Figure 5 Quant ity of Students ' Intention Utterances in Verba l Protocols (as Percentages of Total Utterances) Means of Cu l tura l Groups for Learn ing Record and Argument 49 Figure 6 Types of Intention Utterances in Verba l Protocols (as Percentages of Total Intention Utterances) Means of Cu l tura l Groups for Learn ing Record 52 Figure 7 Types of Intention Utterances In Verba l Protocols (as Percentages of Total Intention Utterances) Means of Cu l tura l Groups for Argument 55 Figure 8 K inds of Activit ies Students Remember Doing in F i r s t Language Wr i t i ng Classes Means of Cu l tura l Groups 60 Figure 9 K inds of Activit ies Students Remember Doing in Previous Engl i sh Classes Means of Cu l tura l Groups 63 Figure 10 K inds of Activit ies Students Remember Doing in Their Present Eng l i sh Class Means of Cu l tura l Groups 66 Figure 11 K inds of Activit ies Students Would Prefer in Their Engl ish Wr i t i ng Class Means of Cu l tura l Groups 69 v i i i Figure 12 K inds of Wr i t ing Activit ies Students Remember and Prefer Means of the Japanese Group Figure 13 K inds of Wr i t ing Activit ies Students Remember and Prefer Means of the Taiwanese Group Figure 14 K inds of Wr i t ing Activit ies Students Remember and Prefer Means of the European Group Figure 15 Types of Teacher 's Utterances (as Percentages of Total Utterances) Means of the Whole Group for Learn ing Record and Argument. Figure 16 Types of Teacher 's Utterances (as Percentages of Total Utterances) Means of Cultura l Groups for Learn ing Record Figure 17 Types of Teacher 's Utterances (as Percentages of Total Utterances) Means of Cu l tura l Groups for Argument Figure 18 Comparison of Student and Teacher Profiles Means of the Whole Group Figure 19 Comparison of Student and Teacher Profiles Means of the Japanese Group Figure 20 Comparison of Student and Teacher Profiles Means of the Taiwanese Group Figure 21 Comparison of Student and Teacher Profiles Means of the European Group Figure 22 Change in Students ' Purposes in Tak ing a Wr i t ing Class Whole Group Profile Figure 23 Change in Students ' Purposes in Tak ing a Wr i t ing Class Cu l tura l Group Profiles for Want ing to Lea rn How to Write Good Compositions ix List of Figures Figure 24 Change in Students ' Purposes in Tak ing a Wr i t i ng Class Cu l tura l Group Profiles for Want ing to L e a r n How to Wr i te Correct English... 98 Figure 25 Change in Students ' Purposes in Tak ing a Wr i t i ng Class Cu l tura l Group Profiles for Want ing to Lea rn About Interesting Topics 99 x A C K N O W L E D G E M E N T S I owe a great debt of gratitude f irst and foremost to my advisor, Dr. Al i ster Cumming, for his guidance and encouragement throughout this project, and for sharing his expertise. T ru l y without h im this thesis would not have reached completion. Thanks are also due to my committee members, Dr. Mar ion Crowhurst and Dr. John Cooper; my colleagues at the Engl i sh Language Institute, especially Ve ra Wojna and Carole Trepanier; the teachers and students who participated in the study; Dean Mellow for assistance w i th statistics; Education Computing Services and Un ivers i ty Computing Services for my computer education; and for the four leaf clover and other encouragements, my friends and family. x i CHAPTER ONE: INTRODUCTION 1.1. IDENTIFICATION OF THE PROBLEM Exc i t ing claims have been made for wr i t ing as a means for learning and as a way of developing understanding (e.g., Emig, 1977; Langer and Applebee, 1987; Smith, 1982,). In an E S L (English as a second language) wr i t ing class, wr i t ing is a means for learning at least four things: the subject written about (content), discourse organization (rhetoric), wr i t ing as a ski l l (composing processes), and Engl i sh as a second language. Because of this quadruple agenda, learners and their teacher may have differing interpretations of the learning goals of their wr i t ing class. Students ' intentions may be based on such variables as previous educational experience, proficiency in Engl ish, and wr i t ing abil ity in their f irst language, as wel l as their immediate and long term learning or task goals. Teachers ' intentions may be based on their educational background, teacher training, ongoing professional development, and teaching experience, as wel l as their current curr iculum goals and lessons. In the context of the present study, where adult students are learning Engl i sh for Academic Purposes, professional debate has centered on whether teachers should focus instruction on wr i t ing "processes" or written "products." "Process wr i t i ng " teachers want students to "not s imply learn how to write, but to learn how to learn by wr i t i ng " (Liebman-Kleine, 1986, p. 787). They a im to "give [students] strategies, and then to help them, over and over, to figure out how to find a process that w i l l enable them to handle the current wr i t ing task and s ituation" (ibid.). Teachers doubtful about the efficacy of the "process approach" for academic wr i t ing stress that the academic wr i t ing teacher should a im to socialize students into the academic community by preparing them for the demands 1 Introduction of real academic situations, including essay examinations, which do not allow for the multiple drafts recommended by "process" advocates. In order to prepare students for how their university wr i t ing wi l l be evaluated, "producf ' -or iented teachers may convey to students the conventions, regularities, genres, and typical task types of academic wr it ing (Horowitz, 1986). Other "product" orientations may emphasize conventional grammar teaching, error correction, and rhetorical patterns. The adherents of the "process" approach, then, tend to focus on learners ' affective and cognitive-developmental needs, while those concerned about "product" tend to focus on learners ' instrumental needs in dealing wi th the constraints of academic life and the appropriate features of written discourse (Hamp-Lyons, 1986, p. 793). L i t t le is known about students ' positions in this controversy, because learners have few public forums to debate the val idity or usefulness of different instructional approaches. Wi th al l these factors forming people's intentions in E S L composition classes, a mismatch in intentions is a common diff iculty for teachers and learners, especially for those of different cultural and educational backgrounds (Kumaravadivelu, 1990). Teachers want ing to focus on the discovery of students ' wr i t ing processes may find that their students want to focus more on their wr i t ten language or even learning new content relevant to their fields of study. Second language students, and adults in particular, aware that they need to meet certain requirements for the conventions of academic writ ing, are often overly concerned with grammar, spelling and word choice at inappropriate stages in the composing process (Raimes, 1987). They may feel that the purpose of the E S L wr i t ing class is to address these issues in the manner to which they are accustomed through traditional instruction in their native countries, e.g., through practice sentences and constant correction of language forms by the teacher. On the other hand, teachers may 2 Introduction overemphasize mechanics and frustrate students who are struggling to create meaning while the3' compose. In any case, there may be a great divergence between the intentions of teachers and their students. There is a need for research to be able to identify the factors which form E S L students' intentions, as wel l as to discern how classroom teaching interacts with them in the context of wr i t ing instruction. 1.2. PURPOSE OF THE PRESENT STUDY The purpose of this study is (1) to describe the learning intentions of a small number of adult E S L wr i t ing students, (2) to establish whether these intentions vary wi th students' educational/cultural backgrounds, and (3) to assess the extent to which these students' intentions may correspond to their teacher's intentions, as demonstrated in his plans for instruction and responses to students' compositions. Because this research is exploratory, a case study approach was taken, focusing on a small number of students in one class performing only two tasks. B y selecting this one naturally-occurring group, it is hoped that the integrity and "ecology" of students ' classroom performance and student-teacher relations have been preserved. The present study thus takes into account the four "curr icu lum commonplaces": content, context, learning, and teaching. Specific hypotheses were not set because of the complexity of the factors involved in the "culture of the classroom" (Breen, 1987), a dearth of prior research or theories on this topic, as well as the smal l number of participants involved in the research. Nonetheless, six research questions (grouped into three related categories) were formulated to guide the study on the basis of exist ing research and theory (described in the next chapter of this thesis). 3 Introduction 1.3. RESEARCH QUESTIONS The specific questions to be addressed are: 1. What proportion of adult E S L students ' intention statements as reported in verbal protocols while composing, and retrospective accounts of their composing, relate to learning content organization, rhetorical organization, content language, or rhetorical language^ through their writing? To what extent are these two data sources consistent? 2a. To what extent do these intentions differ between students according to their educational/cultural background (i.e., Japanese vs. Taiwanese vs. European students) in quantity of total intention utterances and in the proportion of intention utterances relating to content organization, rhetorical organization, content language, or rhetorical language in verbal protocols and in retrospective accounts? 2b. To what extent do the perceptions of students from these cultural groups differ with respect to (i) their previous wr it ing classes in their f i rst languages and in Engl ish, (ii) the kinds of activities they do in their present class, and (iii) their preferences for an ideal class? 3a. What are the teacher's objectives for the tasks done by the students in the study? What proportion of his utterances while checking these tasks relate to content organization, rhetorical organization, content language, or rhetorical language? Is this consistent wi th his objectives? 3b. To what extent do the learning intentions of the students and the teaching objectives of their teacher correspond, as revealed in the verbal protocols? 3c. To what extent may the students' intentions for learning before enrolling in the wr i t ing course have changed to accommodate the teacher's intentions for the course? 1. F u l l operational definitions of these terms appear in Chapter 3, "Methods and Context," Section 3.4. Br ief ly, content organization and rhetorical organization refer to a writer ' s ideas, while content language and rhetorical language refer to the language used to express them. Content organization and content language refer to what the writer means and wants to say, and to invention and idea generation; rhetorical organization and rhetorical language refer to how the writer says what she or he wants to say, and is tied to text production processes. 4 CHAPTER TWO: REVIEW OF RELEVANT RESEARCH This chapter reviews previous research relevant to the three ma in research questions guiding this study (see Chapter One). The chapter considers research on (1) E S L students ' intentions in wr i t ing in an instructional context; (2) cultural learning differences, and (3) the fit between teachers ' objectives and their students ' learning intentions. Research relevant to the data collection methods used in the study is also reviewed. 2.1. INTENTIONS IN ESL WRITING CLASSES: LEARNING CONTENT, RHETORICAL AND COMPOSING SKILLS, AND LANGUAGE The f irst research question assumes that human intentions are central to learning, teaching, and writ ing. The idea of intentionality in learning is rooted in a cognitive perspective on education, which considers learning and teaching as problem-solving activities. The cognitive view presupposes that individuals ' mental activities guide and facilitate their learning, and that "people only solve the problems [goals] they define for themselves" (Flower and Hayes, 1981, p. 369). In cognitive wr i t ing theory, Hayes and Flower (1980) have pointed out that good writers pose and solve different problems than poor writers do, suggesting that poor writers could improve if they had a richer sense of what they were t ry ing to do, or i f they had more relevant goals for composing. Bereiter and Scardamal ia (1987) suggest a var iety of ways in which cognitively-oriented instruction might facilitate learning and wr i t ing by teachers or other mater ia l prompts assisting to structure learners ' goals and composing processes. If wr i t ing is a means for learning, then learning goals, or intentions, in an E S L wr i t ing class can be related to learning content, rhetorical patterns or stsdes and 5 Review of Relevant Research composing skil ls, or the code of the second language (L2). However, little information is available to describe what learners ' priorities are. The potential importance of each learning goal-content, rhetoric and composing, and language-is described below. Academic writ ing, be it for a teacher (as a "d i sp lay " of knowledge) or for a student's personal development, implies a learning of substantive content. Learn ing content through writ ing can be considered in the light of Bereiter and Scardamalia ' s (1987) "knowledge-transforming model," which they also call a model of " intentional learning": " i t involves the setting of goals to be achieved through the composing process and the purposeful pursuit of those goals" (p. 361). Langer and Applebee's (1987) studies of learning have examined the effects of various kinds of wr i t ing activities on learning content and reported that the most successful instruction occurred when students and a teacher had a shared understanding of the specific goals of an instructional activity (pp. 140-141). Also important for the wr i t ing student in an Academic Preparation class are rhetorical skil ls. For the purpose of this study, rhetorical skills as a learning goal include the organization and discourse form of a text being written, as wel l as its stylistic elements, and require certain composing skills, rhetorical knowledge, or wr i t ing processes to produce. Cognitively, this knowledge can be conceived as a "rhetor ical problem space" (Bereiter and Scardamalia, 1987). For second language learners, this rhetorical problem space probably also includes contrastive rhetoric concerns (Grabe and Kap lan , 1989), that is, distinctions between the rhetorical conventions of learners ' mother tongues and of the second language. Although contrastive rhetoric is a way of studying written language with greater attention to products than to processes, some of its observations or findings are relevant to composing processes. Grabe and K a p l a n point out, for example, that 6 Review of Relevant Research decisions involved in rhetorical organization are part of a learned process which results in a written product, and contrastively different products must have used contrastively different composing processes. Research into contrastive rhetoric reveals where some of the differences in students' priorities for learning wr i t ing may arise. While Engl ish wr i t ing instruction, especially by process wr i t ing advocates, stresses ideational or propositional content, wr i t ing in some languages may be more concerned with other rhetorical elements. Grabe and Kap lan (1989) cite examples of language registers which value the forms and sounds of the language more than content or understanding: Classical Arabic, as described by Ostler (1987) and formal Japanese, as described by Jenkins and Hinds (1987) and Wilkerson (1986, cited in Grabe and Kap lan , 1989). Jenkins and Hinds suggest that the emphasis on form over content in Japanese business letter wr i t ing " i s a result of a concern with mainta in ing or establishing a relationship rather than w i th orienting the letter toward one party or the other" (p. 341). The attitude toward audience thus may also differ considerably from culture to culture. Relative reader/writer responsibility is the basis of H inds ' (1987) exploratory typology of written languages. He suggests that Engl i sh is a writer-responsible language; French is reader-responsible; Chinese is a language in transit ion from reader-responsible to writer-responsible, and Japanese has a nonperson orientation, which means that it is reader-responsible, but more important than relative reader responsibility is social relational control. Reader-responsible languages assume a great quantity of shared knowledge, and thus require somewhat different kinds of rhetorical conventions. If transfer of f i rst language rhetorical patterns takes place in second language acquisition (SLA), then there are implications for teaching wr it ing: 7 Review of Relevant Research In addition to teaching students in E S L wr it ing classes that there are differences in rhetorical styles between Engl ish and their native language, it may be necessary to take a further step and teach a new way to conceptualize the wr i t ing process. It may be necessary to instruct students from certain countries, such as Japan, that the wr i t ing process in Engl i sh involves a different set of assumptions from the ones they are accustomed to working with. . . . Such non-native Engl ish writers wi l l have to learn that effective-written communication in Engl i sh is the sole provenience of the writer (Hinds, 1987, pp. 151-152). Mohan and Lo (1985), however, have observed that in Chinese, many rhetorical patterns are s imi lar to Engl ish; they suggest that interl ingual transfer would be positive, rather than interfering with L 2 composing. F ina l ly , students who are aware that they need to adhere to the conventions of the L 2 code in academic wr i t ing may have as a learning goal accurate language production. This goal of language accuracy may be quite distinct from a person's concerns for ideas or rhetoric while composing. The results of Cumming ' s (1989) study indicate that composing skills and language proficiency are two psychologically distinct components of overall E S L wr i t ing abil ity. Wr i t ing expertise is one of several cognitive skills which can be applied across languages, developing separately from proficiency in a second language. Process wr i t ing advocates c la im that students w i l l learn a language to the extent that they have something to communicate in wr i t ing and the opportunities to practise such expression (Zamel, 1982). Brannon and Knoblauch suggest that as student writers attempt to narrow the gap between intention and rhetorical effect, "the motive to solve technical problems is strengthened, in a context in which the writer ' s intentions matter more than the teacher's Ideal Tex t s " (1982, p. 165). In a s imilar way, for E S L writers, the goal of producing "comprehensible output" in Engl i sh "may be the trigger that forces the learner to pay attention to the means of expression needed in order to successfully convey his or her own intended meaning" (Swain, 1986, p. 133). Wr i t ing may thus be a 8 Review of Relevant Research means of learning the second language, and writ ing to learn language is one reason for using a "process approach" to teaching L 2 writ ing, especially for people who are literate and capable writers in their first language (Cumming, 1990). In the context of an academic preparation class, then, the different components of wr i t ing instruction-content, rhetoric and composing, and second language product ion-may be more or less important to students, depending on their previous experience w i th wr i t ing and wr i t ing instruction, among other factors. Since wr i t ing strategies and processes learned in the first language appear to cross over to wr i t ing in the second language, regardless of second language proficiency (Cumming, 1989; Jones and Tetroe, 1987; Raimes, 1987), students' previous experience may also affect the kinds of composing processes they use in English. 2.2. CULTURAL DIFFERENCES IN LEARNING The second set of questions addressed in this study concerns cultural variation in students ' learning intentions. Adu l t learners have learning styles strongly influenced by the values, expectations and goals of the educational systems they have participated in. This principle has strong implications for second language acquisition in classroom settings. Reid's (1987) survey of E S L students' learning style preferences shows that E S L students from different language/educational/cultural backgrounds often differ significantly in various ways from each other, as wel l as from native speakers of Engl ish, in their perceptual learning styles (p. 99). The difference in Chinese educational/learning style is observed by Burnaby and Sun (1989), who note that, aside from the many practical factors (such as class size) which make "Wes te rn " ideas of education difficult to apply to 9 Review of Relevant Research the Chinese context, the traditional relationship between teachers and students in the People's Republic of Ch ina strongly favours "teacher-centered methods and structured curr icu la " (p. 229). A s one of the teachers in their study remarked, there is a large "Cu l ture gap. Chinese don't think in the way most Westerners th ink " (p. 229). Differences between Chinese and Nor th Amer ican educational/learning styles have also been observed by Sampson (1984) and N g (1975), and in regards to Japanese styles by Robinson (1981, pp. 79-81). A central issue for second language teachers is whether to adapt their teaching style to their students' learning style preferences or not. The authors mentioned above recommend that teaching should be adapted to the students ' preferences. Sampson (1984) remarks, "Scientific findings that are acceptable and useful wi th in the context of one educational theory may be quite unacceptable within the context of another educational theory" (pp. 24-25). In the teaching of writ ing, M c K a y (1989) notes that topic development and discourse " i s largely a factor of cultural experience as wel l as social and educational pol icy" (p. 260), and she suggests placing students "w i th s imilar life experiences in the same wr i t ing classes," selecting "topics that are within the realm of those students ' experiences" and then str iv ing "to become aware of our own cultural expectations about the development of the topics we ass ign" (p. 262). Concern has also been raised by others about teachers imposing their own value systems on students whose education has not prepared them to share those value systems (e.g., Horowitz, 1986, p. 143; Politzer and McGroar ty , 1985, p. 114). A process orientation to E S L wr i t ing instruction may potentially be problematic in this regard, wi th its student-centered workshop format and ideals of wr i ters ' responsibility for their own texts. Louie (1984, cited in Br indley, 1986, p. 31) believes that the necessity for individuals to take control of their own destiny (as suggested in the philosophies of Ca r l 10 Review of Relevant Research Rogers, Paulo Fre i re, and Ivan Illich) is " a western middle-class notion which does not apply to second language learners in whose culture individuality is not valued to the same extent." In S i lva ' s view (1990) a teaching approach to wr it ing must be informed by a theory which includes a sensitivity "to the cultural, linguistic and experiential differences of individuals and societies" (p. 19). Other writers, however, seem to contradict these recommendations by advocating certain instructional methods regardless of learners ' previous instruction or preferences. Numerous studies involving Lat in -Amer ican students, for example, have shown positive results when process-oriented instruction in wr i t ing has been used in place of methods which put a heavy emphasis on editing and formal details of students' wr i t ten texts. Trueba (1987) found that Mexican-American students ' wr i t ing and th ink ing skills improved when their teachers used the knowledge and experience of the students as the content for pre-writ ing and wr i t ing activities. L ikewise, Mol l and D iaz (1987) organized a successful instructional intervention in which they attempted to change Mexican-American students ' motive for writ ing: wr i t ing "was to become an activity to communicate with someone else about something that mattered" (p. 308), i.e., a real issue in their own Lat ino community. L i teracy here is seen as a socializing force, an agent of change; therefore the methods used to teach it must be changed to implement that process. Betancourt and Phinney (1988) observed that different groups of bi l ingual (Spanish-English) writers experienced different sources of writer ' s block, depending on the language used and the amount of experience wr i t ing the individual students had. Less experienced writers, wr i t ing in Engl i sh (their L2), suffered the most apprehension with premature editing, and those wr i t ing in Spanish (their L I ) indicated negative attitudes about wr i t ing and evaluation. The researchers noted that for Puerto R ican students, 11 Review of Relevant Research traditional wr i t ing instruction places "undue emphasis on sentence-level g rammar " and on "editing-level evaluat ion" (p. 471). Betancourt and Phinney concluded that "the process-oriented model of teaching writ ing, with considerably lessened emphasis on static rules and sentence-level stylistics, would help to reduce the anxiety caused by premature wr i t ing and attitude in undergraduates" (p. 473). Reid (1987) cites a number of additional studies which show that adult learning styles can be modified to accommodate different educational settings and different instructors ' cognitive styles. Much other research on learning strategies in second languages also suggests that learners can be " t ra ined " to use specific learning strategies (Wenden and Rubin, 1987). Cohen and Cavalcant i (1990), in line with advocates of learner training, suggest that for learning writ ing, Learners ' expectations and preferences may derive from previous instructional experiences, experiences that may not necessarily be beneficial for the development of writ ing. Hence, i f the focus was on grammar in the students' courses, they may feel that focus should be on grammar in the teacher's feedback (Viv ian Zamel, p.c). In such cases, it may be v ita l for teachers not to cater to the students ' expectations but to shift those expectations according to what contributes most to the development of wr it ing skills (pp. 173-4). Radecki and Swales (1988) also note that revision is often perceived differently by students and teachers. While students may see revision as correcting surface level features of wr it ing, teachers may see it as a reassessment of meaning and reshaping of text. Radecki and Swales also believe in shifting students ' expectations to accommodate those of the teacher: 12 Review of Relevant Research As seen in our analysis, the majority of our respondents, 8 7 % . . . appreciate substantive comments that allow them to rethink a piece of writ ing, yet these very students also expect the instructor to correct a l l their surface errors. If this attitude is indicative of E S L students in general, then, E S L wr i t ing instructors are faced with a di lemma. If [ E SL wr i t ing instructors] do not surface-correct but respond to a writer ' s meaning, their credibility among their students can be impaired. Clearly, teachers must intervene and change student attitudes; one way for teachers to change their students is by sharing with them the research in wr i t ing (p. 364, my emphasis). In sum, the two schools of thought here are (1) adapt teaching to the students' values and preferences, and (2) teach according to the accumulated research and experience related to teaching wr i t ing to E S L students, regardless of students ' previous experience. These recommendations may be contradictory. Or, it may be that standards appropriate for one cultural group may not be appropriate for another. Perhaps the La t i n Amer ican and Nor th Amer ican students in the studies mentioned above are more cultural ly prepared to adapt to process wr i t ing methods than the Japanese and Chinese students described in other studies, even though both populations have been through educational systems which put emphasis on sentence-level wr it ing. 2.3. THE FIT BETWEEN TEACHERS' OBJECTIVES AND STUDENTS' LEARNING INTENTIONS IN WRITING El l i s (1985, p. 103) cites d iary studies in which learning style preferences of students and teachers of European/Western background differ greatly from each other. If this is the case, then one has to wonder how closely the preferences of students from non-Western backgrounds match those of their Western teachers. The third set of research questions in this thesis addresses students ' learning intentions from the teacher's perspective. 13 Review of Relevant Research In any wr i t ing class, teachers probably influence the goals that learning writers have. Ruggles Gere and Abbott (1985), for example, show that classroom assignments determine how students talk about their wr i t ing (i.e. their intentions for their work). The students in their study seemed to show considerable concern about their teachers ' opinions. Brannon and Knoblauch (1982) point out that such concern is a natural consequence of teachers determining the form and content of writ ing, and the criter ia which determines its success. Students easily end up compromising the authority they should have as writers, believing " tha t the teacher's agenda is more important than their own, and that what they wanted to say is less relevant than the teacher's impression of what they should have sa id " (p. 158). Brannon and Knoblauch call for teachers "to relinquish . . . control of student wr i t ing and return it to the writers: doing so wi l l not only improve student incentive to write, but wi l l also make [teachers'] responses to the wr i t ing more pertinent" (p. 161). In their view, "the teacher's role is to attract a writer ' s attention to the relationship between intention and effect, enabling a recognition of discrepancies between them, even suggesting ways to eliminate the discrepancies, but f inal ly leaving decisions about alternative choices to the writer, not the teacher" (p. 162). These authors thus present a radical view of the pr imacy of students ' intentions. The most prominent area of L 2 wr i t ing research which has analyzed the f it between teacher 's and students' intentions is that of teacher feedback to written composition. This research focuses on students ' wr itten products, but in the same way that contrastive rhetoric may be helpful in understanding contrastive composing processes, this research may be helpful in beginning to understand the interaction of the intentions of teachers and their students. 14 Review of Relevant Research Lek i (1990) writes, "the goals we set for our wr i t ing classes go a long way toward determining how we wi l l respond to our students' wr i t i ng " (1990, p. 59). A t the same time, her review of research shows that students ' preferences for error correction vary from situation to situation. In her own 1986 study, for example, Lek i found that students expressed a lack of interest in their teacher's reactions to the content of their papers: "Students reported that such commentary did not help their wr i t ing improve, whereas directives on development and organization and indications of errors did help their wr i t i ng " (1990, p. 62). Fa thman and Whal ley (1990) found s imi lar attitudes. Perhaps this finding is related to Zamel ' s observation (1985) that E S L teachers tend to address main ly issues of language form, "v iewing their students as language learners rather than developing wr i te r s " (p. 91). When E S L teachers address issues of content and organization they may not expect their students to take responsibility for doing anything about these problems. In contrast to the E S L students in the studies of Lek i (1990) and Fa thman and Whal ley (1990), Semke (1984) found that North Amer ican students learning German did not like their errors being pointed out to them, and preferred to receive supportive comments without indication of errors. This difference might appear to be based on cultural values. Radecki and Swales (1988), however, in a s imilar study, found that students va ry greatly in their response to feedback, but their study "d id not establish any correlations between response type and linguistic/cultural background" (p. 363). Cohen and Cavalcant i (1990) also conducted research based on the idea that in some cases there may be a misf it between what a teacher provides in the form of written feedback and what that teacher's students want. Cohen and Cavalcant i note, like Brannon and Knoblauch, that learners may make changes to their texts (and presumably, their composing processes) to suit their teacher's values, out of a belief that the teacher knows 15 Review of Relevant Research best. Their study analyzed three teachers ' self-reported feedback and actual feedback to written compositions, and their students' responses to this feedback. The research found that the f it between the teachers ' reported feedback and actual feedback varied in the three studies they conducted (p. 171), that the teachers ' feedback varied according to students ' proficiency level (p. 172), and that the fit between the teachers ' written feedback and the students ' interests also varied from study to study. The existing research on the congruence between what a teacher provides and what students may want is thus inconclusive. More may be revealed in the light of the ongoing context of a wr it ing class, the roles played by teachers and students, and the goals of the individuals involved. Brannon and Knoblauch (1981, cited in Lek i , 1990) would like to see research which looks at teachers ' responses to students' intermediate drafts, and at the ongoing dialogue between the teacher and student. Such research would be useful in throwing more light on the interaction between teachers ' and students ' intentions. A process wr i t ing class is an ideal setting in which to study intentions in learning. A s Bereiter and Scardamal ia (1989, p. 387) point out, strategy instruction alone probably cannot be sufficient to develop people into successful learners: opportunities for independent learning must be provided; relevant types of metaknowledge must be taught; higher-level parts of instructional processes must be progressively turned over to students; the setting of cognitive goals must be modeled; and the level of constructive effort to perform a task must be self-assessed. A l l of these conditions can be met in teaching the processes of writ ing. A composition class can thus provide an appropriate, natural ly-occurring context such as is needed for the study of the teaching-learning process among E S L learners and instructors. 16 Review of Relevant Research 2.4. COLLECTING DATA In the interests of preserving the authentic "cu l ture " or "ecology" of the group of participants, the present research determined to study only one class in depth, using a variety of methods of data collection to reveal, describe and compare learner and teacher intentions and points of view (Breen, 1987). A review of methods of collecting learning strategies data (Cohen, 1987) reveals that self-report data is very useful to this end. Wi th classroom observation alone, a researcher cannot tel l what subjects are thinking, and quiet students may reveal little relevant information. Fo r wr i t ing performance, it is v i r tual ly impossible to understand individual thinking or performance without some form of verbal report (Bereiter and Scardamalia, 1987). In the present study, self-report data included "thinking-aloud" protocols, "prompted reca l l " protocols, and questionnaires. The thinking-aloud technique is "beneficial in that it provides direct evidence about processes that are otherwise invisible, yields r ich data and thus promotes exploration of cognitive processes" (Cohen, 1987, p. 38). In wr i t ing process research, concurrent thinking-aloud protocols have been found to be useful because of the challenging, long term nature of wr i t ing tasks: "people rapidly forget many of their own local work ing goals once those goals have been satisf ied" (Flower and Hayes, 1981, p. 377), mak ing retrospective reports or interviews after wr i t ing unreliable because of memory constraints. The val idity of think-aloud protocols to describe wr it ing processes has been controversial (e.g., Cooper and Holzman, 1983; Voss, 1983). However, Er icsson and Simon's (1984) extensive study of protocol analysis in a variety of task types shows that thinking aloud does not significantly alter the sequence of cognitive processes, that verbal concurrent reports provide a nearly complete record of the sequence of information that is 17 Review of Relevant Research heeded during task performance, and that overall, verbal ly reported information is as regular and val id as other types of behavioral data. A prompted recall of a task involves reporting on thinking processes after the task has been completed, and supplements thinking aloud data. Recall data are useful for revealing the general structure of thinking processes, because retrospective retrieval uses higher-level organizational cues, like subgoals, leaving out most of the detailed information such as appears in concurrent verbal report during task performance (Ericsson and Simon, 1984, p. 379). Interviews, questionnaires and classroom observations were also used in the present study to supplement data on thinking processes. In language learning strategy research, Na iman , Frohl ich, Stern and Todesco (1978, p. 100) found that a carefully constructed interview could reveal useful information about students ' language learning goals, supplementing classroom observations in a productive way. L i t t le interview or observational research has been done in regards to learning to write in a second language (Cumming, 1991); almost al l second language learning strategy research has focused on oral language (see studies in Wenden and Rubin, 1987). 18 CHAPTER THREE: CONTEXT AND METHODS The present stucfy set out to address the research questions posed in the introductory chapter by collecting information in case study fashion from eleven students and two teachers of one adult E S L wr i t ing class wi th a process orientation to instruction. In order to obtain as full a view as possible of the students' intentions, a number of data collection techniques were used, including observations, self-reports, verbal protocols, and retrospective accounts. The data were then analyzed with the research questions in mind. Fo rma l hypotheses were not set because of the complexity of factors involved-age, gender, and length of time in Canada, for example, could not easily be controlled-and the lack of information in the smal l data sample to address them al l adequately, a paucity of theory to guide hypothesis construction, and very little previous research on the present topic. The study is by nature, then, exploratory and descriptive and is l imited in its interpretations to the one pedagogical context. Details of the participants, setting, data collection, coding of data, and data analysis follow. 3.1. PARTICIPANTS 3.1.1. Selection of Participants The study recruited two teachers of an E S L writ ing class and 11 of their students of three distinct cultural groups (5 Japanese, 3 Taiwanese, and 3 Europeans) at a univers ity in western Canada. The teachers responded to a letter about the study circulated at the university institute for E S L instruction asking for a teacher of writ ing (see Appendix A). The students then responded to a letter circulated in the teachers ' class (see Appendix A). After an enthusiastic endorsement of the study by the teachers 19 Context and Methods involved, eighteen of the twenty-eight students in the class volunteered to participate. To achieve the research aims of comparing cultural groups, several potential participants from diverse cultural backgrounds had to be excluded (one Indonesian, one Chinese from the People's Republic of China, and one Israeli). The most predominant cultural groupings among the volunteers were init ia l ly 7 Japanese students, 5 Taiwanese, and 3 Europeans. Two Japanese students withdrew from the study after completing the questionnaire, and two Taiwanese students withdrew after completing the f irst task, leaving the total as 11 participants. Da ta from the students who withdrew were not included in the analyses. 3.1.2. Profiles of the Student Participants The participating students ranged in age from nineteen to the mid twenties, with one European student in her late thirties. A l l five of the Japanese students who remained in the study to the end were female, while only one of the Taiwanese students was female and the other two were male. Two of the Europeans were female and one was male. F i r s t languages were Japanese and Mandar in, and in the European group, Ital ian, Swiss-German, and French. A l l of the participants had been in Canada from 2 to 9 months except for a Taiwanese student who had been here 14 months. 3.1.3. Educational Background of the Students One student in each group had only completed high school. In the Japanese group, the other four had between two and four years of college or university tra in ing in Japan (in Engl ish l iterature and Japanese literature). One of the Taiwanese students had two years of univers ity (majoring in Business) and the other person had 5 years of college (in Mechanics) and one year of university (in Business). Neither of the two other European 20 Context and Methods students had advanced academic training; one had done three years of apprenticeship in a vocational program, and the other, the oldest of the group, had been in a vocational program but had left a year early, at seventeen. The European students claimed to have never studied Engl i sh composition before coming to study in Canada; only one Taiwanese student had, for four months; and three of the Japanese students said they had, for 8 to 9 years. For most of the participants, this was their f i rst wr i t ing class in Canada, except for the Japanese students, who appeared to have studied Engl i sh composition for the same length of time as their stay in Canada. 3.1.3.1. Self-rating of Learning Ability Appendix B shows a self-rating scale administered to the students prior to their participating in the research. In a l l three categories of learning abil ity—learning language, writ ing, and aspects of one's field of work or s tudy-the European students rated themselves higher than the Taiwanese, who rated themselves higher than the Japanese (see Table 1 and Figure 1). A l l three groups rated themselves highest in learning aspects of their study or work, followed by learning language, and last ly by learning writ ing. Statistical tests of these differences were not carried out, however, because of the smal l number of people with in each subgroup. A l l of the participants rated themselves as "poor" or " f a i r " (rather than "good" or "excellent") at learning writ ing. Only the Europeans considered themselves "good" at learning languages or at studying or learning their work. 21 Context and Methods Table 1: Self-Rating of Learning Ability by Student Participants: Means of Cultural Groups and Standard Deviations Japan Ta iwan Europe (n = 5) (n = 3) (n = 3) 1 = poor; 2 = fair; 3 = good; 4 = excellent A s a learner of: Language X 1.6 2.3 3.0 SD (0.49) (0.47) (0.82) Wr i t i ng X 1.4 1.7 2.3 SD (0.49) (0.94) (0.47) Study or work X 2.2 2.7 3.3 SD (0.75) (0.47) (0.47) 22 Context and Methods Figure 1: Self-Rating of Learning Ability by Student Participants: Means of Cultural Groups excellent 4 T 23 Context and Methods 3.1.3.2. Self-rating of Writing Ability The student participants also self-rated their wr i t ing abilities in their mother tongues, using a scale adapted from Cumming, 1989 (see Appendix B). The responses of the students in the present study were s imilar to Cumming ' s "bas ic " or " inexpert " writer groups for a Quebecois(e) population. In validation studies of this scale across Japanese, Quebecois(e), and Mex ican adult E S L students, "expert " writers rated themselves as 5 and "average" writers scored 4 (Alister Cumming, personal communication). Un l ike their ratings of learning abilities, the students' confidence in their f irst language wr it ing abil ity was roughly the same for a l l three groups (see Table 2 and Figure 2). Scores for al l three categories—writing letters to people you don't know, wr it ing reports at school, and wr i t ing reports at work-hovered around the "sometimes easy, sometimes diff icult" mark. Table 2: Self-Rating of Writing Ability in Mother Tongue by Student Participants: Means of Cultural Groups and Standard Deviations Japan Ta iwan Europe (n = 5) (n = 3) (n = 3) 1 = N O , too difficult; 2 = N O T V E R Y W E L L , always difficult; 3 = Y E S , B U T sometimes easy, sometimes difficult; 4 = Y E S , usual ly easy; 5 = Y E S , V E R Y W E L L , a lways easy Could you write: to people you X 3.2 3.3 3.0 don't know SD (0.75) (0.47) (0.82) reports at X 3.0 2.7 3.0 school SD (1.10) (0.47) (0.00) reports at X 2.8 2.7 3.3 work SD (1.17) (0.47) (1.25) 24 Context and Methods Figure 2: Self-Rating of Writing Ability in Mother Tongue by Student Participants: Means of Cultural Groups YES, 4 j usually easy I 25 Context and Methods 3.2. SETTING The wr i t ing class under observation was part of a non-credit Academic Preparat ion Program in Engl i sh as a Second Language for students planning to enter Canadian or Amer ican universities. The students were not formal ly graded, and there was thus little external pressure to "pass " or obtain a certain grade. The present research was conducted just past the midpoint of the university term (weeks 8 to 11 of a twelve-week term), a point which it was anticipated could reveal a tension between the teachers ' and students ' goals, a point at which the students would be either accommodating the teachers ' values or resisting them. The two instructors team-teaching the class under observation were extremely committed to teaching wr it ing processes rather than focusing on wr itten products. This meant that the instructors managed the class much like a wr i t ing workshop, in which students worked on specific aspects of whatever wr i t ing work was in progress (see Cumming, 1991, for a case study description of one of the teachers ' classroom routines). The class met for an hour and forty minutes four days a week. The classroom was a large area otherwise used as a recreation room, and the 28 students were assigned to groups of three and spread around the room, where they worked together in their groups to apply the heuristics and strategies they had been instructed to use. The typical procedure for any given class was for the teachers to model a wr i t ing operation. Then the students were given a clear, l imited set of questions or operations (never more than four items) to apply to their own or a group member 's paper. Often a time l imit was set for the task. For example, in an "Eva luat ion Derby" , the students had ten minutes to answer the following questions about the content of the papers by the people on their immediate right: 26 Context and Methods 1. Do you like it? (1 - 5) 2. F ind one or two things you like about it. 3. A re the divisions interesting? ( 1 - 5 ) 4. A re the points interesting? ( 1 - 5 ) Rhetorical issues were addressed in a s imilar way, for example in a task to assess a draft of an argument paper: Count the opposing points. A re there at least three? Language concerns were also treated in this way: Edit Draf t 3 of the f irst argument: a) Subject-Verb agreement b) Sentence Types c) Sentence Balance In summary, students were expected to work through different processes of wr it ing in the context of the class, where the teachers could help them with the specific tasks of the day. The class situation was a little unusual in that there were two teachers, as a result of bringing what were originally two classes together to function as a process-writing workshop. Because the teachers worked as a team, I observed both of them during the classes preceding and during the data collection on wr i t ing tasks, but only one teacher was asked to comment on the student tasks for the research study. It is difficult to generalize about any concrete differences in the students ' behavior in class. Generally, al l the participants in this study were well motivated and applied themselves conscientiously to the tasks they had been given. One observable difference in groups, however, was the Europeans ' tendency to lead. They seem to have understood the teachers ' instructions more clearly than the others, which put them in a position to take a " tu to r " role within their groups. For example, the Ital ian student had to prompt his partner what to do during a kind of brainstorming task, which required each partner to take a turn tell ing the other his or her opinions. S imi lar ly , during a debate the French 27 Context and Methods student took the lead: " You have to say the reason, and we have to write what they say." In another case, while other groups became distracted with language concerns, the Swiss student made some especially accurate, on-task observations compared to the other members of her group: "I have some problems w i th the divisions." A n d later, " Y o u have to make it more clear." The tasks at hand on the days I observed had main ly to do with content and rhetorical organization, and although students did stay mainly on task, they also tended to digress to content language or rhetoric language concerns (see definitions of these terms in section 3.4 below), regardless of their cultural background. Given instructions to write a f irst draft of an argument fast, for example, one student digressed from the task when she discussed with two members of her group whether "absent" is a noun or an adjective. I observed a Japanese student and a Taiwanese student on different occasions t ry ing to get definitive answers about matters of style from an outside authority, i.e., a teacher or teaching assistant: "Should I write a type one or a type two argument?" "Wh ich one is better?" "Wh i ch type of comparison essay is better?" The European students did not ask this kind of question. 3.3. DATA COLLECTION In order to collect data on students' intentions from various perspectives, the study included (a) student questionnaires, (b) observations of classes, and (c) concurrent "think-aloud" and "prompted reca l l " protocols. 28 Context and Methods 3.2.1. Questionnaires Questionnaires provided background information about students ' previous education or tra in ing in writ ing, and general goals for the course (see Appendix B). These questionnaires were administered at an in it ia l meeting of al l participants. 3.2.2. Observations The purpose of the classroom observations was to document the context of the study and to see what the students' intentions or priorities appeared to be in this situation. Two aspects of student performance were documented in this setting: the kinds of questions they asked, and other indicators of focus of attention, such as what they took notes on. A profile for each student was then established from these data. I also documented the teachers ' explanations of tasks and principles behind them, and what they emphasized as being important. These observations took place during three of the classes which lead up to the two class assignments used as tasks in the study (for a total of six classes). On the days I observed, the students appeared to hardly notice my presence because they were so involved with their writ ing. In addition, they were used to having assistants in the class and thought nothing of asking me questions about their work in progress. I functioned, therefore, as a "part ic ipant observer" in the classroom context, so any "observer effect" on students ' behavior can be assumed to be min imal . 29 Context and Methods 3.2.3. Think-Aloud and Prompted Recall Protocols 3.2.3.1. Tasks The purpose of the study was to determine what students' intentions are in the context of one academic wr i t ing class. To that end, in consultation with the teachers, two academic wr i t ing tasks, involving some complexity of thought (and therefore requiring students to think intentionally), were chosen for the study. These assignments were an integral part of the course and were not imposed art if ic ial ly for the purposes of the research. The f irst assignment was a "Learn ing Record" in which the students were to describe the wr i t ing course, give their opinions of the most and least useful activities they had done, and to give suggestions for improving the course in the future. The instructions for the assignment indicated explicitly that the students were not to "wor ry about language form," and that the purpose of the assignment was "to give their opinions" (see Appendix C). The second assignment was a draft of an argument essay. When the data were collected the students were at various stages of their drafting, so some were working on a second and third draft of their f i rst argument, and others were working on a first draft of a second argument (see Appendix C). Both assignments could normal ly be completed in less than an hour, according to the teachers ' estimation of time for task completion. The participants were asked to verbally report as much as they could of what they paid attention to while doing these assignments. 30 Context and Methods 3.2.3.2. The Think-Aloud Procedure A t the in it ia l meeting of al l participants I explained how to think aloud, then demonstrated this technique by modelling aloud my own thinking processes as I did an algebra problem. I then asked the participants to think aloud while practis ing other problems (see Appendix D). These procedures follow Cumming (1989) and Ericsson and Simon (1984). The students then met with me individually for data collection and to compose the relevant tasks. They were urged to speak in their f irst language i f they were more comfortable doing so. A lmost al l the students claimed they found it easier to think in Engl ish when they were wr it ing in Engl ish, and refused to use their f irst language. One Japanese student made short comments, such as exclamations, or translations of words, in Japanese. Only the French student actually "thought" in her first language for extended parts of the verbal protocol. I speak French, so coded and analyzed her protocol in exactly the same way as the Engl ish protocols. The students ' thinking-aloud protocols for the writ ing tasks were audio tape-recorded. The French student mentioned above found the thinking-aloud process generally difficult and said she was bothered by the presence of the tape-recorder. The other participants, who were ten to fifteen years younger than she, seemed barely to notice the machine. 3.2.3.3. The Prompted Recall Procedure Immediately following completion of each task -or , for a few participants, when an hour had passed~I asked the students what their intentions for the wr i t ing assignment were at the beginning of their work, and at the end. Together we looked at the written 31 Context and Methods text and I asked the students to recall their intentions for each paragraph, and what they had been thinking when they had made major changes in their texts (see Appendix E). A l l of the students' protocols for each assignment were completed within a week of each other, except in the case of two students who were i l l during the week of the second assignment and who did the second task early the following week. 3.2.3.4. The Teacher's Protocols One of the two teachers was then asked what his specific objectives for the assignment were (see Appendix F) and to give his impressions of the students ' completed assignments by producing concurrent verbal reports of what he paid attention to while checking these. This teacher was chosen to provide the verbal protocols because he was the one who tended to be more dominant in the planning for this class. A s the instructor was fami l iar wi th thinking aloud methods, demonstrations and tra in ing of this technique were not considered necessary. This data collection aimed as much as possible to document the process the teacher would actually perform in responding to the students ' compositions, in spite of my being present and tape-recording what he said. 3.4. CODING OF DATA 3.4.1. T h e S t u d e n t s ' T h i n k - a l o u d P r o t o c o l s Verbal izations of decision-making in the students ' tape-recorded think-aloud protocols were transcribed. Verbalizations of text production or reading aloud of written text were excluded from the transcripts (after Cumming, 1989) in order to isolate decision statements or instances of ongoing reflection about writ ing, rather than lower-level 32 Context and Methods processes of text generation. The transcripts were segmented into " idea units," considering Chafe 's operational definitions: (1) [An idea unit] is spoken with a single coherent intonation contour, ending in what is perceived as a clause f inal intonation; (2) it is preceded and followed by some kind of hesitation, ranging from a momentary break in t iming to a filled or unfilled pause lasting several seconds; (3) it is a c lause-that is, it contains one verb phrase along with whatever noun phrases, prepositional phrases, adverbs and so on are appropriate; and (4) it is about seven words long and takes about two seconds to produce. Idea units do not always conform to this prototype, but on the whole they are clearly identifiable elements of spoken language, and deviations from the prototype are usual ly explainable in interesting ways (Chafe, 1985, p. 106). S imi lar definitions are given by Er icsson and Simon (1984, p. 205), who say that "appropriate cues for segmentation [of verbal protocols] are pauses, intonation contours, etc., as well as syntactical markers for complete phrases and sentences" because these "units of articulation wi l l correspond to integrated cognitive structures" and "pauses and hesitations w i l l be good predictors of shifts in processing of cognitive structures" (1984, p. 225). In practice, two out of three of the following cr iter ia were used to divide the think-aloud protocols into idea units, or utterances: (1) one intonation contour (2) pauses of three seconds or more (3) a single semantic unit (a clause or a phrase) If in doubt (as for fast talkers), the taped speech was divided as a sentence would be punctuated in its written form: " . . . and dividing, yeah./" "The least useful? / Wel l , at the m o m e n t . . . " 33 Context and Methods The " idea units " which reflected verbalizations of intentions were extracted f rom the tape-recorded data. Each intention utterance was judged to represent goals and future states of the participant or his or her piece of wr it ing. Pre l iminary analyses indicated that such intention utterances could be recognized by: - auxi l iary verbs indicating possibility, such as "might, " "could," "may, " " can, " "would, " be able to," " w i l l , " "would l ike." - auxi l iary verbs indicating predictabil ity, such as " w i l l , " " sha l l , " "be going to," "be V + ing " (with future time indicator) - auxi l iary verbs indicating necessity, such as "should," "ought," "must, " "have to," "have got to," "had better." -- cognitive psychological verbs, where the propositional content of the sentence is about the subject or his or her piece of wr it ing and some action in the future is implied, or a decision is implied which may change the future state of what is under discussion, e.g., "I don't know if " r eca l l " is r ight." - forward-pointing expressions like " let ' s see what I can put here," "wha t else." - commands to self to carry out a process or attend to a topic, e.g., "Next , body." " L e t me read." These intention utterances were then coded for reference to content, rhetoric, or language use, in four categories: content/discourse organization (CO), rhetoric/discourse organization (RO), content/language code (CL), and rhetoricAanguage code (RL). The f irst two categories (CO and RO) refer to a writers ' ideas; the latter (CL and RL ) to the local (lexical) or more general (discourse level) aspects of language used to express them. C O and C L refer to the substantive, ideational, or conceptual aspects, i.e., what the wr i ter means and wants to say, and to invention and idea generation; RO and R L to how the writer says what she or he wants to say, and is tied to text production. The operational definitions for these categories are given below in F igure 3: 34 Context and Methods Figure 3: Operational Definitions for Coding Categories CONTENT — what one means and wants to say - invention; idea generation CO ( C O N T E N T / D I S C O U R S E O R G A N I Z A T I O N ) : intention utterances referring to the working out of problems of belief and knowledge, e.g., opinions, mora l decisions, inferences about matters of fact, formulations of causal explanations (Bereiter & Scardamalia, 1987). ~ includes retr ieval of information and relevant experience, analysis of ordering principles of that information and knowledge, and the formulation of the problems which give rise to inquiry into beliefs and knowledge (Freedman & Pringle, 1980). e.g., " A general statement, what could be a gen—a general statement could be about, about exams in general." e.g., " Le t ' s see what we have done here in this course." CL ( C O N T E N T / L A N G U A G E CODE) : intention utterances referring to semantics and word choice; matching choice of words to standard usage. e.g., "I should check this, "responsibil ity, responsibil ity." e.g., "Oh, what ' s the word in Engl i sh. " RHETORIC -How one says what one wants to say - tied to text production RO ( RHETOR IC/D I SCOURSE O R G A N I Z A T I O N ) : intention utterances referring to the working out of problems of achieving goals of the composition — includes organization of the composition (attention to rhetorical structures such as introduction, body, and conclusion, examples, coherence). e.g., " Le t ' s see where I can introduce this." e.g., "I have to ask uh, what he means by organize." — intention utterances referring to interaction between writer and audience e.g., "So now I'm thinking how, how can I make my sentence clear." — intention utterances referring to carry ing out mechanical procedures. e.g., " I ' l l write it down" [ideas], e.g., "I can say [writes]." RL ( R H E T O R I C / L A N G U A G E CODE) : intention utterances referring to arranging and rearranging grammar, spelling and mechanics for accurate expression, considering the impact of the language on an audience. Typ ica l actions wi l l be deletion, addition, changes of spelling, combining. e.g., "I don't need 'also. ' " e.g., "I w i l l , I w i l l combine two sentences." e.g., " I ' l l write [this word] down." 35 Context and Methods 3.4.2. The Students' Retrospective Accounts The prompted recall protocols were also transcribed and coded using the same categories as above. In an attempt to be representative of the whole process of wr i t ing each assignment, a subset of the data was taken consisting of the gist of the responses to the questions described above (see Appendix E). For the responses to the questions asking about intentions for each paragraph and about local changes, where there may have been several questions and responses, the most frequently coded response was tallied. Where two coded responses tied for frequency, both were tallied. 3.4.3. The Teacher's Protocols The teacher's verbal reports while checking the assignments were also transcribed, divided into " idea units " as described above, then coded for attention devoted to content, rhetoric, and language use, as above. The teacher's comments were coded for these categories only as they pertained to the student text at hand. Examples from the teacher's protocol were: Content Organization: "Oh, she decided apparently to write about unwriting as a useful act iv i ty " " B u t it ' s just about utterly lacking in content" Content Language: "Oh, she doesn't know what 'couple' means " " ' Some shopper,' meaning 'shopkeeper ' " Rhetorical Organization: "I think she's indicating that that ' s part of her introduction" "[the suggestions] are not real ly c lear " "Oh , he's going back to his notes there then " Rhetorical Language: " A n d she has a reasonable variety of sentence structure" " H e spent a lot of time editing, look at this!" 36 Context and Methods A fifth category, Pedagogical Remarks (P), was added for the coding of the teacher's protocol. These utterances revealed efforts to convey an instructional purpose ("and so I, anyway, perceive a lot of what we were doing was teaching writ ing, not teaching Engl ish") or evaluation ("I think she's done a real ly good job of what we asked her to do"). In addition, the pedagogical remarks included comments about the general abil ity of the students ("because she's a fa ir ly diligent person"), personal reactions ("I 'm curious"), and self-direction ("Well, let 's go back"). The students ' and teacher's protocols represent slightly different kinds of data: while only the students ' intention statements were coded, al l of the teacher's statements were coded. " Intent ion" statements as such could not be collected from the teacher since reading is a different task than writ ing, and the teacher's role differs from that of the student. Presumably, what the teacher commented on while checking the assignments was what he considered important and what most l ikely formed his framework of intentions for the assignment. 3.4.4. Reliability Checks Random samples of 10% of the data f rom the teacher and students were checked for intracoder rel iabil ity on the idea unit divisions (94.6% agreement on 334 decisions in the student protocols; 89.4% agreement on 104 decisions in the teacher protocol). Intercoder rel iabi l ity was checked on another 1 0 % of the data for coding decisions wi th another experienced E S L teacher (85.1% agreement on 47 decisions in the student protocols) and an experienced E S L wr i t ing teacher (85.9% agreement on 71 decisions in the teacher protocol). Da ta from two student participants who made less than five 37 Context and Methods utterances during the think-aloud procedure were excluded from the analyses of the verbal protocols. 3.5. DATA ANALYSIS The data from the observations and questionnaires were reviewed and tabulated. The proportion of intention utterances to total utterances was calculated for the Learn ing Record task and the Argument task for the whole group and for each cultural group, and a t-test performed to assess differences between means on the whole group. T-tests were not possible to do meaningfully across each of the cultural groups because of the very smal l numbers in each group. The proportions of the students ' coded intention statements in the think-aloud protocols and their responses in the retrospective accounts relating to content organization, rhetorical organization, content language, and rhetorical language were calculated and ranked, and group profiles were established for the Learn ing Record and Argument tasks. The results of the two data sources-think-aloud protocols and retrospective accounts-were analyzed for consistency using the Mann-Whitney two-sample rank test, a non-parametric statistical test which assesses the statistical probability of whether two groups are drawn from the same population, using the rank of each case. This test was used as a rough indicator of similarities across the cultural groups. The profiles were then analyzed impressionistically for variations between the groups. The proportions of the teacher's protocol statements relat ing to content organization, rhetorical organization, content language, and rhetorical language were also calculated and ranked. These were matched with his interview statements about his objectives for the tasks to ascertain i f these data corroborate one another. The ranked data were compared to the students' profiles to determine the relative emphasis on content 38 Context and Methods organization, rhetorical organization, content language, and rhetorical language, and analyzed with the Mann-Whitney two-sample rank test. In addition, the change in students' intentions over the time they had been in the class was assessed statistically with t-tests for the whole group. Comparisons between the different kinds of data-questionnaire, observation, concurrent think-aloud and prompted recall—were done impressionistically to assess the rel iabil ity of each data source and to verify interpretations. 39 CHAPTER FOUR: FINDINGS The findings of the study are presented in this chapter as they address the research questions posed in Chapter One. The first section of this chapter describes the students' intentions for E S L wr i t ing as revealed by their concurrent verbal protocols and retrospective accounts of the two course assignments. The second section of this chapter describes the extent to which these intentions var ied for the three cultural groups; these patterns are supplemented by background information about students' personal histories of wr i t ing instruction, gleaned from the questionnaires. The final section describes the teacher's perspective, specifically (1) the extent to which the teacher focused on the same aspects of composing as the students did and (2) the extent to which the teacher's pedagogical philosophy and methods may have influenced the students' learning intentions. 4.1. RESEARCH QUESTION 1: OVERVIEW OF THE WHOLE GROUP The first research question was: 1. What proportion of adult E S L students ' intention statements as reported in verbal protocols while composing, and retrospective accounts of their composing, relate to learning content organization, rhetorical organization, content language, or rhetorical language through their writ ing? To what extent are these two data sources consistent? In order to answer this question, the intentions of the whole group of students were considered irrespective of their cultural backgrounds. 4.1.1. Intentions in the Students' Verbal Protocols In both wr i t ing tasks, students' attention, as revealed by their verbal protocols while they wrote, focused mostly on content organization and rhetorical organization rather than language concerns (see Table 3 and Figure 4). However, their intentions 40 Findings showed different patterns for the Learn ing Record and for the Argument (see Appendices G and H for samples of a student's verbal protocol and written task). In the Learn ing Record, students devoted an almost equal amount of attention to content organization and rhetorical organization (36.1% and 34.9% respectively) and much less on content language and rhetorical language (16.8% and 8.9%). In the Argument, students focused overwhelmingly on rhetorical organization (57.1%), and much less on content organization (11.6%) and content language and rhetorical language (6.6% and 11.3% respectively). A lmost al l these categories showed considerable var iat ion from person to person (as indicated by high standard deviations from group means), except in the case of the rhetorical organization category for the Argument task where the variat ion was proportionally smaller. 41 Findings Table 3: Types of Intention Utterances in Verbal Protocols (as Percentages of Total Intention Utterances): Means of the Whole Group and Standard Deviations for Learning Record and Argument Task Learn ing Record Argument X SD X SD Content Organization 36 .1% (27.0%) 11.6% (15.3%) Rhetorical Organization 34.9% (27.0%) 57 .1% (23.5%) Content Language 16.8% (29.5%) 6.6% (10.4%) Rhetorical Language 8.9% (13.8%) 11.3% (14.1%) Unclassif ied 3.3% (10.0%) 2.3% (3.6%)! 1. Means were calculated from students ' overall proportions of coded utterances, so percentage figures do not add to 100%. 42 Findings Figure 4: Types of Utterances in Verbal Protocols: Means of the Whole Group for Learning Record and Argument 66.8 x • Learning Record B flrgunent 43 Findings 4.1.2. Intentions Stated in Retrospective Accounts The questions asked in the retrospective interviews were intended to represent two aspects of the whole process of the wr i t ing tasks, and were thus divided into two types. Global, or general, questions asked about students' intentions when they began and ended their wr i t ten drafts, and their intentions for each paragraph with in those drafts. Local, or specific, questions asked students about aspects of their drafts they had indicated particular attention to while composing by circling, underlining, or changing something. These local questions also asked students to describe what they planned to do next wi th the piece of writ ing. General ly, students responded to the global questions wi th answers related to content organization (see Table 4). References to rhetorical organization occurred in responses to both the global and local questions, wi th the greater proportion appearing in the local domain of the Learn ing Record and the global domain of the Argument. A l l language concerns in the Learn ing Record (100%) were in response to local questions. This category of response also formed the greater proportion of answers to the local questions for the Argument. 44 Findings Table 4: Proportions of Answers to Global and Local Questions in Retrospective Accounts: Whole Group Profile for Learning Record and Argument Task Learn ing Record Argument global local global local Content Organization 93.5% 6.5% 78.3% 21.7% Rhetorical Organization 27 .3% 72.7% 57.7% 42 .3% Content Language 0.0% 100.0% 40.0% 60.0% Rhetorical Language 0.0% 100.0% 28.6% 71.4% Fo r their answers to questions about the Learn ing Record task, the rank order of students ' references to the four categories was slightly different to those in the verbal protocols (see Table 5). Content organization received the most attention (42.2%), wi th a high percentage of attention to rhetorical language (31.3%) followed by rhetorical organization (17.2%) and content language (9.4%). The rank order of proportions of students ' answers to questions about the Argument task was exactly the same as that of their intention utterances in the verbal protocols. The Mann-Whitney two-sample rank test showed no statistical differences between the proportions of the students' intention utterances in the verbal protocols and the proportions of their answers in the retrospective accounts (for the Learn ing Record, W = 18.0, p = n.s.; for the Argument, W = 16.0, p = n.s.). This statistical evidence indicates that these two sets of data tend not to differ significantly, and are therefore consistent. This f inding holds for the two sources of data across cultural groups as wel l (for the Japanese and European groups, for both tasks, W = 17.0, p = n.s.; for the Taiwanese 45 Findings group, W = 16.0, p = n.s. for the Learn ing Record, and W = 15.0, p = n.s. for the Argument). These results should be treated with caution, however, as the smal l number of participants in the study and the considerable variation within groups may have made statistical significance nearly impossible to reach in this analysis. Table 5: Proportions of Types of Intentions in Retrospective Accounts: Whole Group Profile for Learning Record and Argument Task Learn ing Record Argument Total Total Content Organization 42.2% 31.7% Rhetorical Organization 17.2% 41.7% Content Language 9.4% 8.3% Rhetorical Language 31 .3% 18.3% 4.1.3. Summary for Research Question 1 In sum, students focused ma in ly on content organization and rhetorical organization for the Learn ing Record task, and mainly on rhetorical organization for the Argument task. Trends in the students ' verbal protocols while composing two assignments, and their retrospective accounts of their intentions for those assignments were shown to be consistent (or at least, not inconsistent) according to the Mann-Whitney two-sample rank test. 46 Findings 4.2. RESEARCH QUESTION 2: CULTURAL GROUP PROFILES 4.2.1. Research Question 2a: Intentions in the Students' Verbal Protocols The second research question was: 2a. To what extent do these intentions differ between students according to their educational/cultural background (i.e., Japanese vs. Taiwanese vs. European students) in quantity of total intention utterances and in the proportion of intention utterances relating to content organization, rhetorical organization, content language, or rhetorical language in verbal protocols and in retrospective accounts? 4.2.1.1. Quantity of Intention Utterances The Japanese and European groups made a much larger percentage of intention utterances while doing the Argument task than when doing the Learn ing Record (an increase from 8.0% to 34.4% and from 21.0% to 41 .3% respectively; see Table 6 and Figure 5). However, the Taiwanese group made a smaller proportion of intention utterances in the Argument task than in the Learn ing Record task (a decrease from 8.5% to 4.8%). The Taiwanese group's decrease in intention utterances may have been due to the fact that the person who made the most intention utterances in the Learn ing Record made less than five utterances during the Argument assignment and was thus excluded from the data analysis for the Argument task. T-tests were not performed on these results because the groups were too small to assess statistical significance. However, for the whole group, a paired, two-tailed, two-group t-test showed that the Argument task prompted significantly more intention utterances (t = -3.10; p< 0.01). The Europeans tended not only to be more verbal, producing roughly twice as many utterances as the Japanese and Taiwanese groups, but they also made a much 47 Findings higher percentage of intention utterances: their mean was 21.0% for the Learn ing Record compared to 8.0% and 8.5% for the Japanese and Taiwanese respectively, and 41 .3% compared to 34.4% and 4.8% for the Argument task. Table 6: Quantity of Students' Intention Utterances in Verbal Protocols (as Percentages of Total Utterances): Means of Cultural Groups and Standard Deviations for Learning Record and Argument Task Learn ing Record Argument X SD X SD Japan (n = 4) 8.0% (3.3%) 34.4% (10.9%) Ta iwan (n = 3 for L.R.) 8.5% (5.8%) 4.8% (4.8%) (n = 2 for Argument) Europe (n = 3) 21.0% (10.4%) 41 .3% (26.9%) 48 Findings Figure 5: Quantity of Students' Intention Utterances in Verbal Protocols (as Percentages of Total Utterances): Means of Cultural Groups for Learning Record and Argument 56.6 - r -Japan Taiwan Europe • Learning Record B Argunent 49 Findings 4.2.1.2. Intention Utterances in the Students' Verbal Protocols for the Learning Record Seemingly because of the smal l numbers of intention utterances they produced, the Japanese and Taiwanese groups displayed a great deal of variation within their groups as to the types of intention utterances they made (see Table 7 and Figure 6). The Europeans were the most consistent in this behavior as a group (as indicated by very low standard deviations from the group mean). The European group's intentions focused on each of the categories in decreasing order: First, content organization (34.3%), followed by rhetorical organization (30.2%), then rhetorical language (21.3%) and content language (14.2%). The Taiwanese students ' intentions followed a s imi lar order, but with a very high percentage of content organization (58.3%), and slightly less emphasis on rhetorical language than the Europeans (25.0%), and little emphasis on content language and rhetorical language (8.3% each). The Japanese priorities averaged out differently, with the greatest emphasis on rhetorical organization (45.8%) followed by content language (25.0%) and content organization (20.8%). None of the Japanese students' intentions for the Learn ing Record referred to rhetorical language concerns. 50 Findings Table 7: Types of Intention Utterances in Verbal Protocols (as Percentages of Total Intention Utterances): Means of Cultural Groups and Standard Deviations for Learning Record Content Organization Rhetorical Organization Content Language Rhetorical Language Unclassif ied J apan (n = X SD 4) 20.8% (21.7%) 45.8% (36.1%) 25.0% (43.3%) 0.0% (0.0%) 8.3% (14.4%) Ta iwan (n X SD = 3) 58.3% (31.2%) 25.0% (20.4%) 8.3% (11.8%) 8.3% (11.8%) 0.0% (0.0%) Europe (n X SD = 3) 34.3% (4.1%) 30.2% (1.7%) 14.2% (10.1%) 21.3% (15.27%) 0.0% (0 .0%) 2 2. Means were calculated from students ' overall proportions of coded utterances, so percentage figures do not add to 100%. 51 Findings Figure 6: Types of Intention Utterances in Verbal Protocols (as Percentages of Total Intention Utterances): Means of Cultural Groups for Learning Record Japan Taiwan Europe • CO H RO • CL DD RL 5 Unclassified 52 Findings 4.2.1.3. Intentions as Stated in the Students' Retrospective Accounts for the Learning Record A l l three groups gave a large proportion of answers in the content organization category (45.5%, 38.9% and 50.0% for Japan, Ta iwan and Europe respectively; see Table 8). The Taiwanese group gave an equal proportion of answers referr ing to rhetorical language, while the other two groups gave the second greatest proportion of their answers in this category. The Japanese and Taiwanese groups gave the least proportion of answers pertaining to content language (9.1% and 5.6%), but the Europeans gave the least pertaining to rhetorical organization (5.6%). Table 8: Proportions of Types of Intentions in Retrospective Accounts: Cultural Group Profiles for Learning Record Japan Ta iwan Europe (n = 5) (n = 3) (n = 3) Content Organization 45.5% 38.9% 50.0% Rhetorical Organization 21.2% 16.7% 5.6% Content Language 9 .1% 5.6% 11.1% Rhetorical Language 21.2% 38.9% 33.3% 4.2.1.4. Intention Utterances in the Students' Verbal Protocols for the Argument There was more consistency within groups w i th respect to intentions for the Argument, especially in attention to rhetorical organization, as indicated by lower standard deviations (see Table 9 and Figure 7). A l l three groups' intentions were main ly concerned w i th rhetorical organization (62.6% of the Europeans ' intention utterances, 59.8% of the Japanese, and 43.5% of the Taiwanese). The Europeans made the greatest percentage of 53 Findings references to content organization (27.7%) while the Japanese and the Taiwanese made very few (3.1% and 4.4% respectively). The Japanese students also made intention utterances referring to rhetorical language and content language (14.4% and 20.1%) while the Taiwanese made none, and the Europeans very few (0.6% and 7.1%). Table 9: Types of Intention Utterances in Verbal Protocols (as Percentages of Total Intention Utterances): Means of Cultural Groups and Standard Deviations for Argument Content Organization Rhetorical Organization Content Language Rhetorical Language Unclassif ied Japan (n = X SD 4) 3.1% (5.4%) 59.8% (10.2%) 14.4% (11.6%) 20 .1% (15.8%) 2.7% (4.6%) Ta iwan (n X SD = 2) 4.4% (4.4%) 43.5% (43.5%) 0.0% (0.0%) 0.0% (0.0%) 2.2% (2.2%) Europe (n X SD = 3) 27.7% (16.2%) 62.6% (9.6%) 0.6% (0.8%) 7.1% (7.6%) 2.0% (2 .8%) 3 3. Means were calculated from students' overal l proportions of coded utterances, so percentage figures do not add to 100%. 54 Findings Figure 7: Types of Intention Utterances In Verbal Protocols (as Percentages of Total Intention Utterances): Means of Cultural Groups for Argument 79 -r Japan Taiwan Europe • CO •RO • CL CD RL 5 Unclassified 55 Findings 4.2.1.5. Intentions as Stated in the Students' Retrospective Accounts for the Argument In their retrospective accounts of their wr i t ing the Argument, a l l three groups followed the same general trend, seen in the whole group profile, of attending most to rhetorical organization, followed closely by content organization (see Table 10). A s in the verbal protocols for the Argument, the greatest proportion of al l students ' responses in post-task interviews was in the rhetorical organization category. In the retrospective reports, however, unlike in the verbal protocols, the Europeans placed the least emphasis of the three groups on the idea level (25.0% and 35.0% on content and rhetorical organization respectively) and the greatest on language (20.0% for each of content and rhetorical language). The Taiwanese group, meanwhile, placed 26.7% of their answers in the content language category and none in rhetorical language, and the Japanese group put the least emphasis on rhetorical language (17.6%) and only a little on content language (2.9%). The groups were too smal l in number to assess the significance of these differences meaningfully. Table 10: Proportions of Types of Intentions in Retrospective Accounts: Cultural Group Profiles for Argument Japan (n = 5) Ta iwan (n = 3) Europe (n = 3) Content Organization 38.2% 33 .3% 25.0% Rhetorical Organization 41.2% 40.0% 35.0% Content Language 2.9% 0.0% 20.0% Rhetorical Language 17.6% 26.7% 20.0% 56 Findings 4.2.1.6. Summary for Research Question 2a Generally, in the verbal protocol data, the European group made twice as many intention utterances as the Japanese and Taiwanese groups. In the Learn ing Record the European and Taiwanese groups displayed a similar pattern of proportions of intention utterances in the verbal protocols, with the most emphasis on content organization, while the Japanese group attended to rhetorical organization most. The responses of al l three groups in the retrospective accounts were main ly related to content organization. In the verbal protocols for the Argument, rhetorical organization was the main focus for all three groups, wi th the European group also making a large proportion of utterances relating to content organization, and the Japanese and Taiwanese making hardly any. The retrospective accounts followed the same trend although the European group put the least emphasis in this task on the idea level and the most emphasis on language. 4.2.3. Research Question 2b: The Students' Self-Reports of Background and Present Beliefs and Preferences Because previous experiences in wr i t ing classes in their f i rst languages and Engl ish are presumed to influence their present behaviors and beliefs, and in order to explore more ful ly potential cultural differences in students ' intentions, a secondary research question was formulated in relation to research question two: 2b. To what extent do the perceptions of students from these cultural groups differ wi th respect to (i) their previous wr i t ing classes in their f i rst languages and in Engl ish, (ii) the kinds of activities they do in their present class, and (iii) their preferences for an ideal class? 57 Findings The self-report data reported below came from answers to the questionnaire item, "Wha t kinds of activities do you remember doing in wr i t ing classes?" (see Appendix B). 4.2.3.1. Activities in First Language Writing Classes A l l the students claimed to have done some or a lot of memorizing texts and reading l iterature in their first language wr it ing classes, and a little or no journal wr i t ing (see Table 11 and Figure 8). The Japanese students also said they had done some or a lot of g rammar and vocabulary study, and a little or some of a l l other pedagogical activities listed. The Taiwanese students had done a little or no study of g rammar and wr i t ing of multiple drafts, but a little or some of a l l other activities listed. The Europeans said they had done some or a lot of g rammar study. The students did not report any emphasis on any one of the categories of rhetorical or content organization or language. The greatest var iat ion, indicated by large standard deviations from means with in al l three groups, occurred in the category of studying types of composition (for example, argument, cause and effect, comparison/contrast) and studying the organization of compositions. Considerable variat ion occurred with in cultural groups for most other categories of language or composition study, as indicated by standard deviations equalling or exceeding group means. Apparent ly, the kinds of language and composition study done by these individuals in their native education systems were not very consistent, vary ing nearly as much within cultural groups as across them. 58 Table 11: Kinds of Activities Students Remember Doing in First Language Writing Classes: Means of Cultural Groups and Standard Deviations 0 = none; 1 = a little; 2 = some; 3 = a lot Japan Ta iwan Europe (n = 5) (n = 3) (n = 3) Rhetorical style-based activities studying composition types X 1.80 1.33 1.00 SD (1.17) (1.25) (1.41) memoriz ing texts X 2.20 2.67 2.00 SD (0.75) (0.47) (0.82) studying organization X 1.80 1.00 1.00 SD (0.98) (0.82) (1.41) Language-based activities studying g rammar X 2.40 0.67 2.67 SD (0.49) (0.94) (0.47) studying vocabulary X 2.20 1.33 1.00 SD (0.40) (0.94) (1.41) studying mechanics X 1.40 2.00 1.00 SD (1.02) (0.82) (0.82) Content or process-based activities reading l iterature X 2.60 2.33 2.33 SD (0.49) (0.47) (0.94) wr i t ing multiple drafts X 1.80 0.67 1.67 SD (0.75) (0.47) (1.25) keeping a journal X 0.60 1.00 0.67 SD (0.80) (0.82) (0.94) 59 Figure 8: Kinds of Activities Students Remember Doing in First Language Writing Classes: Means of Cultural Groups 3.eT Legend T = studying about composition types (e.g., argument, cause and effect, etc.) M l = memorizing texts by expert writers O = studying about organization (paragraphs, introductions, conclusions, etc.) G = studying g rammar V = studying vocabulary M2 = studying mechanics (punctuation, capitalization, etc.) R = reading books or studying l iterature D = writ ing multiple drafts J = keeping a journa l 60 Findings 4.2.3.2. Activities in Previous English Writing Classes General ly, students perceived that their Engl i sh classes before this one had focused more on language aspects than on the organization of written discourse (see Table 12 and Figure 9). Two of the Europeans claimed they had never studied Engl i sh wr it ing before; the other European had only studied language aspects of Engl ish writ ing. The Japanese students had focused more on language activities, especially g rammar and vocabulary, than on organizational activities, except for reading literature. L i ke the Japanese, the Taiwanese students had focused more on language activities than the others, but less so than the Japanese. None had studied composition types in Engl i sh wr i t ing classes. The statistical significance of these differences, however, could not be assessed because of the smal l number of participants, as in other analyses across the cultural groupings. There was more consistency within groups in the reports of the kinds of activities done in previous Eng l i sh classes than in the reports of f irst language wr i t ing experiences. 61 Findings Table 12: Kinds of Activities Students Remember Doing in Previous English Classes: Means of Cultural Groups and Standard Deviations 0 = none; 1 = a little; 2 = some; 3 = a lot J apan Ta iwan Europe (n = 5) (n = 3) (n = 3) Rhetorical style-based activities studying composition types X 0.60 0.00 0.00 SD (0.80) (0.00) (0.00) memorizing texts X 1.20 0.67 0.00 SD (0.98) (0.47) (0.00) studying organization X 1.00 0.67 0.00 SD (1.10) (0.94) (0.00) Language-based activities X studying g rammar 2.80 1.33 1.00 SD (0.40) (0.94) (1.41) studying vocabulary X 2.40 1.67 1.00 SD (0.49) (1.25) (1.41) studying mechanics X 1.40 1.33 0.67 SD (1.02) (1.25) (0.94) Content or process-based activities reading l iterature X 1.60 0.33 0.00 SD (1.02) (0.47) (0.00) wr i t ing multiple drafts X 0.60 1.00 0.00 SD (0.49) (1.41) (0.00) keeping a journal X 0.40 1.00 0.00 SD (0.49) (0.82) (0.00) 62 Figure 9: Kinds of Activities Students Remember Doing in Previous English Classes: Means of Cultural Groups 3.e T • Japan • Tainan • Europe Legend T = studying about composition types (e.g., argument, cause and effect, etc.) M l = memorizing texts by expert writers O = studying about organization (paragraphs, introductions, conclusions, etc.) G = studying grammar V = studying vocabulary M2 = studying mechanics (punctuation, capitalization, etc.) R = reading books or studying l iterature D = wr i t ing multiple drafts J = keeping a journal 63 Findings 4.2.3.3. Perceptions of the Present Class In regards to the present E S L composition class, al l the students were fa ir ly certain that they were studying some or a lot of composition types and organization, and not memoriz ing texts, and that they were doing little or no reading of l iterature and some or a lot of multiple drafts (see Table 13 and Figure 10). The Japanese and Europeans var ied in their responses about keeping a journal, although the Taiwanese all agreed they did not do this. The greatest general var iat ion in responses with in al l groups was in the language category, except that the Japanese group agreed they were doing some or a lot of g rammar study, and the Taiwanese group agreed they were doing some or a little study of mechanics. The Japanese group perceived that in this class they were doing some or a lot of g rammar and vocabulary study, as well as study of composition types and organization, and the wr i t ing of multiple drafts. The Japanese group thus put more emphasis on grammar, vocabulary, composition types and organization than the other two groups. The Europeans were very close to the Japanese group in their perception of the amount of study of composition types and organization, but the Europeans thought they were only doing a little or some work on language aspects. The Europeans were also more certain than the others that they were doing a lot of multiple drafts of their compositions (there was no var iat ion among the three students). The Taiwanese group perceived the least emphasis on grammar and vocabulary, reporting they were doing these activities a little. They also perceived the least amount of study of composition types (a little or some) and organization (some) and wr i t ing of multiple drafts (a little or some). 64 Findings Table 13: Kinds of Activities Students Remember Doing in Their Present English Writing Class: Means of Cultural Groups and Standard Deviations 0 - none; 1 = a little; 2 = some; 3 = a lot Japan Ta iwan Europe (n = 5) (n = 3) (n = 3) Rhetorical style-based activities studying composition types X 2.60 1.33 2.33 SD (0.49) (0.94) (0.94) memoriz ing texts X 0.60 0.00 0.00 SD (0.49) (0.00) (0.00) studying organization X 2.80 2.00 2.67 SD (0.40) (0.00) (0.47) Language-based activities studying g rammar X 2.20 1.00 1.67 SD (0.40) (0.82) (1.25) studying vocabulary X 2.40 1.00 1.33 SD (0.80) (1.41) (0.94) studying mechanics X 1.60 1.33 1.67 SD (0.80) (1.25) (0.47) Content or process-based activities reading l iterature X 1.20 0.33 0.33 SD (0.75) (0.47) (0.47) wr i t ing multiple drafts X 2.60 1.33 3.00 SD (0.49) (0.94) (0.00) keeping a journal X 1.40 0.00 1.33 SD (1.20) (0.00) (1.25) 65 Findings Figure 10: Kinds of Activities Students Remember Doing in Their Present English Class: Means of Cultural Groups 3 . e T Legend T = studying about composition types (e.g., argument, cause and effect, etc.) Ml = memorizing texts by expert writers O = studying about organization (paragraphs, introductions, conclusions, etc.) G = studying g rammar V = studying vocabulary M2 = studying mechanics (punctuation, capitalization, etc.) R = reading books or studying literature D = writ ing multiple drafts J = keeping a journa l 66 Findings 4.2.3.4. Preferences There was surprisingly little var iat ion in students' preferences within groups for types of wr i t ing or language activities (see Table 14 and Figure 11). The widest variation with in groups occurred in the Europeans ' preferences for amount of g rammar and vocabulary activities. Although statistical comparisons were not possible, the most distinct difference across groups appeared in the Taiwanese stating they would prefer some or a lot of memoriz ing of texts by expert writers while the European group would prefer none, and the Japanese group none or a little. The Taiwanese would also prefer some or a lot of study of mechanics, while the Europeans and Japanese would prefer only a little or some. The Europeans placed little value on reading l iterature, wanting little or none of it, while the Japanese would prefer some or a lot and the Taiwanese a little or some. The Europeans, on the other hand, placed great value on the wr it ing of multiple drafts, with al l three students saying they would prefer to do this a lot. In contrast, the Japanese placed slightly less value on multiple drafts of compositions, wanting some or a lot, and the Taiwanese even less (a little or some). The Japanese would prefer to do some or a lot of journal writ ing, while the Taiwanese would like to do a little or some, and the Europeans a little or none. The other categories found the whole group more or less in agreement: a l l the students wanted some or a lot of study of composition types and organization. The groups were also fa ir ly close in agreement on grammar and vocabulary, wi th the Europeans wanting the most g rammar (some) and the Japanese and Taiwanese wanting a little or some. The Japanese wanted the most vocabulary work (some or a lot), while the Taiwanese and Europeans wanted only a little or some. 67 Table 14: Kinds of Activities Students Would Prefer in Their English Writing Class: Means of Cultural Groups and Standard Deviations 0 = none; 1 = a little; 2 - some; 3 = a lot Japan Ta iwan Europe (n = 5) (n = 3) (n = 3) Rhetorical style-based activities studying composition types X 2.40 2.00 2.33 SD (0.80) (0.00) (0.94) memorizing texts X 0.80 2.33 0.00 SD (0.75) (0.94) (0.00) studying organization X 2.60 2.67 3.00 SD (0.49) (0.47) (0.00) Language-based activities studying g rammar X 1.80 1.67 2.00 SD (0.98) (0.47) (1.41) studying vocabulary X 2.40 2.00 1.67 SD (0.80) (0.82) (1.25) studying mechanics X 1.20 2.33 1.67 SD (0.75) (0.47) (0.94) Content or process-based activities reading l iterature X 2.40 1.67 0.33 SD (0.49) (0.47) (0.47) writ ing multiple drafts X 2.60 1.67 3.00 SD (0.80) (0.47) (0.00) keeping a journa l X 2.40 1.33 0.67 SD (0.49) (0.47) (0.94) 68 Figure 11: Kinds of Activities Students Would Prefer in Their English Writing Class: Means of Cultural Groups Legend T = studying about composition types (e.g., argument, cause and effect, etc.) M l = memorizing texts by expert writers O = studying about organization (paragraphs, introductions, conclusions, etc.) G — studying grammar V = studying vocabulary M2 = studying mechanics (punctuation, capitalization, etc.) R = reading books or studying literature D = writing multiple drafts J = keeping a journal 69 Findings 4.2.3.5. Relationship Between Experience and Preference The limited sample size in this study prohibits statistical correlations between the students' experience and their present preferences for E S L wr it ing instruction. However, an impressionistic comparison can be made by observing which of the responses to the experience categories-f irst language wr i t ing classes, Engl i sh wr it ing classes, and the present c las s -are most s imilar to the corresponding preference categories (see Figures 12, 13 and 14). Such an impressionistic comparison reveals that the Japanese group's preferences for seven categories are closest to their present experience with only the exception of reading l iterature (closest to f irst language wr i t ing classes) and studying mechanics (closest to f i rst language and previous Engl i sh wr i t ing classes). L ikewise, the European group's preferences are closest to their present experience for seven categories, excepting the study of g rammar and journal wr i t ing (closest to f irst language wr i t ing classes). The Taiwanese group's preferences are more s imi lar to their past experiences in f irst language and Engl ish wr i t ing classes, with the exception of the study of composition types, organization and the wr i t ing of multiple d ra f t s - a l l rhetorical and process aspects of composing-which were closest to their experience in the present class. 70 Findings Figure 12: Kinds of Writing Activities Students Remember and Prefer: Means of the Japanese Group 3.8 T • Now ( t h i s class) ED Preference Legend T = studying about composition types (e.g., argument, cause and effect, etc.) Ml = memoriz ing texts by expert writers O = studying about organization (paragraphs, introductions, conclusions, etc.) G = studying g rammar V = studying vocabulary M2 = studying mechanics (punctuation, capital ization, etc.) R = reading books or studying literature D = wr i t ing multiple drafts J = keeping a journa l 71 Figure 13: Kinds of Writing Activities Students Remember and Prefer: Means of the Taiwanese Group 3.e T Legend T = studying about composition types (e.g., argument, cause and effect, etc.) Ml = memorizing texts by expert writers O = studying about organization (paragraphs, introductions, conclusions, etc.) G = studying g rammar V = studying vocabulary M2 = studying mechanics (punctuation, capitalization, etc.) R = reading books or studying l iterature D = writ ing multiple drafts J = keeping a journal 72 Figure 14: Kinds of Writing Activities Students Remember and Prefer: Means of the European Group Legend T = studying about composition types (e.g., argument, cause and effect, etc.) M l = memoriz ing texts by expert writers O = studying about organization (paragraphs, introductions, conclusions, etc.) G = studying g rammar V = studying vocabulary M2 = studying mechanics (punctuation, capital ization, etc.) R = reading books or studying literature D = wr i t ing multiple drafts J = keeping a journal 73 Findings 4.2.3.6. Summary for Research Question 2b In summary, students did not appear to remember any special emphasis on content, language or rhetoric or composing processes in their first language wr it ing classes, but language aspects were evidently strong in their previous Engl ish wr i t ing classes. Responses about the present class revealed a surpris ing amount of variat ion considering al l the participants were in the same class. The Japanese students, for example, perceived that they were doing more study of grammar, vocabulary and reading l iterature than the European and Taiwanese students did. The students of the three groups appeared to differ with respect to preferences for kinds of pedagogical activities, especially in the categories of memoriz ing texts, study of mechanics, reading of l iterature, and wr i t ing of multiple drafts and journals. The Japanese and European groups' preferences for activities in an ideal class were more s imilar to their experiences in the present class than to their previous wr i t ing classes, while the Taiwanese group's preferences were more s imilar to their previous classes, wi th the exception of studying rhetorical structure and writ ing multiple drafts. 4.3. RESEARCH QUESTION 3: THE TEACHER'S PERSPECTIVE The third set of research questions concerned the extent to which the learning intentions of the students and the teaching objectives of their teacher corresponded. This section first describes the teacher's stated objectives for the tasks used in the study, then the results of analyses of the teacher's verbal protocols while reading and checking over the students ' assignments. Nex t the results from the teacher's data are compared with the whole group profile and to the cultural group profiles. F ina l ly , the change in students' purposes for taking this wr i t ing class is documented. 74 Findings 4.3.1. Research Question 3a: The Teacher's Objectives (Self-Reports) and Verbal Protocols The first part of the third set of research questions was: 3 a. What are the teacher's objectives for the tasks done by the students in the study? What proportion of his utterances while checking these tasks relate to content organization, rhetorical organization, content language, or rhetorical language? Is this consistent wi th his objectives? Information on the teacher's objectives as reported below was elicited during interviews prior to his think-aloud sessions. 4.3.1.1. Objectives: Learning Record The teacher wanted from the students " a picture of the whole course from [the] students' point of view." He stated that the objective for the Learn ing Record "wasn ' t to get a piece of wr i t ing that we would assess, it was more to get them reflecting on the course and let us know how they were feeling about i t . " He made it clear that this objective was not language-related, pointing out the instructions given to the students: "Don ' t worry about language and form, the purpose is to give us your opinions." Before looking over what the students had produced for the think-aloud protocols he said, " I 'm quite sure that al l I ' ll be looking at w i l l be to what extent they are, in fact, commenting on what we did in the course, why, what was useful, what wasn 't, the objective as implied in the assignment." In terms of the coding scheme used on the verbal protocols and the retrospective accounts, his p r imary objective for the Learn ing Record was related to content organization. 75 Findings 4.3.1.2. Objectives: Argument Because the students were working on different drafts of their wr it ing, the teacher commented on his objectives for each draft. Second draft of the f irst argument The teacher's objectives for the second draft of the f irst argument task included having the students (a) " f ind or put in at least three opposing arguments," (b) "check their papers to see that they had a balance of different sentence types," (c) "focus on their transitions to see that they had enough transitions," and (d) "check whether or not their connecting words were being used accurately." In summary, he said, "so we were really working on an idea level, or a content level, in presenting the opposing arguments, or a rhetorical level, and the other, at a more mechanical language level." Th i rd draft of the f irst argument In this draft, the teacher wanted students to pay particular attention to (a) the introduction, which should have included a general statement, a narrowing of that statement, and some kind of organizer; and (b) the conclusion, which should have been some kind of reference or connection to the f inal paragraph of the body, including a reference to the introduction, a general statement broadening from a fa i r ly narrow statement, and some kind of prediction or recommendation. 76 Findings F i r s t draft of the second argument The teacher said he ordinari ly would not look at draft one, but i f he did, for this task he would look for some evidence of opposing argument, an introduction, and a conclusion. Summary of objectives for the Argument Rhetorical organization was the teacher's p r imary focus for the Argument task, regardless of the stage of the students ' drafting, wi th some attention to rhetorical language in the second draft of the f irst argument. 4.3.1.3. The Teacher's Protocol Statements For the Learn ing Record, the greatest proportion of the teacher's utterances overall were pedagogical remarks (56.9%), followed by a large proportion of utterances referring to content organization (36.0%; see Table 15 and F igure 15). The teacher did not pay much attention to rhetorical organization (4.3%), and even less to content language (1.5%) and rhetorical language (1.3%). The teacher's utterances while responding to the Argument task were more evenly distributed among the rhetorical organization, pedagogical remarks and content organization categories (34.5%, 32.3%, and 22.6% respectively), but he sti l l placed little emphasis on language (see Appendix I for a sample of the teacher's verbal protocols). 77 Findings Table 15: Types of Teacher's Utterances (as Percentages of Total Utterances): Means of the Whole Group and Standard Deviations for Learning Record and Argument Task Learn ing Record Argument X SD X SD Content Organization 36.0% (18.2%) 22.6% (11.6%) Rhetorical Organization 4 .3% (5.1%) 34 .5% (15.3%) Content Language 1.5% (2.9%) 5.6% (11.6%) Rhetorical Language 1.3% (4.0%) 5.0% (6.3%) Pedagogical Remarks 56.9% (14.4%) 32 .3% (17.3%) 4 4. Means were calculated from overal l proportions of coded utterances, so percentage figures do not add to 100%. 78 Findings Figure 15: Types of Teacher's Utterances (as Percentages of Total Utterances): Means of the Whole Group for Learning Record and Argument 79 Findings Aside from making pedagogical remarks, the teacher did indeed focus the greatest proportion of his attention on the content organization of the students' Learn ing Records, and for the Argument, he focused the greatest part of his attention on rhetorical organization, followed by content organization. The teacher's stated objectives for the tasks were thus consistent wi th his performance while looking over the students' written efforts. 4.3.1.4. The Teacher's Protocol Statements by Cultural Group: Learning Record The proportions of the teacher's utterances for the Learn ing Record followed a very s imilar pattern of distribution for each cultural group (see Table 16 and Figure 16). One slight difference appeared in less attention to content organization for the Taiwanese students (26.6%) compared to the Japanese (41.7%) and Europeans (44.4%), and correspondingly higher proportions of pedagogical remarks for the Taiwanese group. 80 Findings Table 16: Types of Teacher's Utterances (as Percentages of Total Utterances): Means of Cultural Groups and Standard Deviations for Learning Record Content Organization Rhetorical Organization Content Language Rhetorical Language Pedagogical Remarks Japan (n = X SD = 5) 41.7% (19.4%) 5.1% (5.1%) 0.9% (1.8%) 0.0% (0.0%) 52.3% (15.9%) Ta iwan (n X SD = 3) 26.6% (15.5%) 6.6% (5.4%) 0.0% (0.0%) 4.4% (6.3%) 62.3% (15.1%) Europe (n X SD = 3) 44.4% (15.6%) 1.7% (2.4%) 3.6% (3.9%) 0.0% (0.0%) 50.3% (12.7%) 5 5. Means were calculated from overall proportions of coded utterances, so percentage figures do not add to 100%. 81 Findings 82 Findings 4.3.1.5. The Teacher's Protocol Statements by Cultural Group: Argument Again, as for the Learn ing Record, the teacher's utterances about the Argument fell into a very s imilar pattern for each cultural group (see Table 17 and Figure 17), although the Taiwanese students received slightly more pedagogical remarks, more attention to their rhetorical language, and less attention to their content organization and rhetorical organization. Table 17: Types of Teacher's Utterances (as Percentages of Total Utterances): Means of Cultural Groups and Standard Deviations for Argument Content Organization Rhetorical Organization Content Language Rhetorical Language Pedagogical Remarks J apan (n = X SD = 5) 26 .1% (4.3%) 35.4% (12.9%) 12.0% (13.5%) 2.2% (1.5%) 24.4% (4.9%) Ta iwan (n X SD = 3) 15.0% (11.4%) 26.7% (3.6%) 2.5% (2.0%) 13.1% (6.2%) 42.7% (8.9%) Europe (n X SD = 3) 19.6% (14.7%) 40.8% (17.9%) 0.0% (0.0%) 2.3% (2.1%) 37.3% (24 .6%) 6 6. Means were calculated from overall proportions of coded utterances, so percentage figures do not add to 100%. 83 Findings Figure 17: Types of Teacher's Utterances (as Percentages of Total Utterances): Means of Cultural Groups for Argument 5« X Japan Taiwan Europe • CO B BO B CL GD RL B P 84 Findings 4.3.1.6. Summary for Research Question 3a Overal l , the teacher's objectives for the Learn ing Record mainly concerned content organization, and for the Argument, mainly rhetorical organization with some emphasis on rhetorical language. Aside from his pedagogical remarks, the verbal protocols showed the teacher's responses to the students' tasks to be consistent with his objectives, although for the Argument a fa i r ly large proportion of his utterances were also related to content organization. The teacher's responses to the students ' wr i t ing were s imilar across cultural groups, although his responses to the Taiwanese students ' wr i t ing differed slightly from those for the Japanese and European students. 4.3.2. Research Question 3b: Comparison of the Teacher's and Students' Intentions in Verbal Protocols The second part of the third set of research questions was: 3b. To what extent do the learning intentions of the students and the teaching objectives of their teacher correspond, as revealed in the verbal protocols? For the sake of comparing the teacher and student profiles, the teacher's pedagogical remarks were discarded from this analysis. The sum of the remaining four categories was calculated as 100% of the teacher's utterances analyzed in this section. 4.3.2.1. The Teacher's Utterances Compared to the Students': Whole Group Profile Overal l , the teacher's and students' attention to aspects of their wr i t ing were quite s imi lar except that the teacher paid considerably more attention to content organization than the students did (although some attention to this category on the students' part may not have been coded as not reflecting intentions). The students paid more attention to 85 Findings rhetorical organization and the language categories, except for content language in the Argument (see Figure 18). Figure 18: Comparison of Student and Teacher Profiles: Means of the Whole Group 96 -r 86 Findings 4.3.2.2. The Teacher's Utterances Compared to the Students': Cultural Group Profiles For the Learn ing Record, the trends for al l three groups were the same as that for the whole group, wi th the Taiwanese students being the closest to the teacher in proportion of attention to content organization (see Figures 19, 20 and 21). For the Argument task, again the teacher paid much more attention to content organization than the Japanese and Taiwanese students and slightly less to rhetorical organization. However, for these categories in this task the European proportions were closest to the teacher's, even reversing the trend seen in the Japanese and Taiwanese group profiles in that the Europeans attended more to content organization than the teacher did. The extra emphasis on the content language category in the Japanese profile (12.0%) reflects the teacher's attempts to make sense of the vocabulary used in one almost incomprehensible paper. L ikewise, the extra emphasis on the rhetorical language category in the Taiwanese profile reflects the teacher's difficulty in following the logic in one paper with especially convoluted sentence construction. 87 Findings Figure 19: Comparison of Student and Teacher Profiles: Means of the Japanese Group 88 Findings Figure 20: Comparison of Student and Teacher Profiles: Means of the Taiwanese Group ee-r LRCO LRRO ISO. LRRL ARCO ARM) ARCL ARRL • Students B Teacher 89 Findings Figure 21: Comparison of Student and Teacher Profiles: Means of the European Group 96 T 90 Findings As seen in the above comparisons, the proportions of the teacher's and the students' attention to content organization, rhetorical organization, content language, and rhetorical language appear to correspond quite closely. The Mann-Whitney two-sample rank test gives statistical evidence that the teacher's and students' verbal protocols tend not to differ for the whole group profile (for the Learn ing Record, W = 21.0, p = n.s.; for the Argument, W = 25.0, p — n.s.), nor for each of the cultural groups (for the Learn ing Record and Argument respectively, W = 20.5, 19.0, p = n.s. for the Japanese group; W = 23.0, 15.0, p = n.s. for the Taiwanese group; W = 22.0, 20.0, p = n.s. for the European group). Technically, the two sets of data appear to come from the same population, according to this empirical analysis. However, because the teacher was only one participant in this comparison, there may have been insufficient var iat ion in this analysis to have logically reached statistical significance. Thus this finding of no significant differences should be interpreted with caution. 4.3.2.3. Summary for Research Question 3b General ly, the teacher focused more of his attention on content organization than the students, who focused a greater proportion of their intentions on rhetorical organization, content language, and rhetorical language. There were some slight differences in the correspondence of the intentions of the different cultural groups to the utterances of the teacher, but the results for the Mann-Whitney two-sample rank test indicated that the students' intentions and teacher's utterances did not differ significantly. 91 Findings 4.3.3. Research Question 3c: Change in the Students' Purpose in Enrolling The last part of the third set of research questions was: 3c. To what extent may the students' intentions for learning before enrolling in the wr i t ing course have changed to accommodate the teacher's intentions for the course? 4.3.3.1. Whole Group When asked to state their in it ia l purpose in enrolling in an E S L wr i t ing class, students put most emphasis on learning to write correct Engl i sh (45.9%), followed by learning how to write good compositions (33.2%), with the least emphasis on learning about topics through wr i t ing (20.9%; see Table 18 and Figure 22). Their purpose shifted, though, after eight weeks in this process wr i t ing class: learning to write good compositions became the most important item (43.2%) and learning to write correct Engl i sh became less important (30.5%). Learn ing about topics through wr it ing became more important (26.4%). Two-group, paired, two-tailed t-tests showed that the decrease in interest in learning correct Engl ish was statistically significant (t = 3.29, p< 0.008). The increase in interest in learning wr it ing approached but did not reach statistical significance (t = -1.88, p < 0.09), while the change in interest in learning about topics was also not statistically significant (t — -1.57, p — n.s.). 92 Table 18: Students' Purposes in Taking a Writing Class: Means of the Whole Group (as Percentages) and Standard Deviations Time of the term Week 0 Week 8 X S.D. X S.D. Purpose: To learn Wr i t ing 33.2% (12.7%) 43.2% (14.5%) Correct Engl i sh 45.9% (18.7%) 30.5% (13.6%) About topics 20.9% (10.6%) 26.4% (13.7%) 93 Figure 22: Change in Students' Purposes in Taking a Writing Class: Whole Group Profile Findings Findings 4.3.3.2. Cultural Group Profiles The European group's attitude toward the purpose of an E S L wr i t ing class appeared to change the most from the beginning to near the end of the term, especially with respect to learning how to write compositions (increasing from 28.3% to 53.3% priority) and learning how to write correct Engl i sh (decreasing from 56.7% to 26.7% priority; see Table 19 and Figures 23, 24 and 25). The Japanese and Taiwanese groups reported a less dramatic pattern of change. However, one Taiwanese student did report that his in it ia l purpose was 8 5 % learning correct Engl i sh and 5% learning how to write good compositions, which changed to 6 0 % learning correct Engl i sh and 3 0 % learning to write good compositions. The Japanese and European groups' interest in learning about topics increased slightly (10% and 5% respectively), but in this area the Taiwanese interest decreased slightly. 95 Findings Table 19: Students' Purposes in Taking a Writing Class: Means of Cultural Groups (as Percentages) and Standard Deviations Time of the term Week 0 Week 8 X S.D. X S.D. Japan Purpose: To learn Wr i t ing 40.0% (8.9%) 41.0% (13.6%) Correct Engl i sh 38.0% (13.3%) 27.0% (10.8%) About topics 22.0% (7.5%) 32.0% (11.7%) Ta iwan Purpose: To learn Wr i t ing 22.2% (15.4%) 36.7% (4.7%) Correct Engl i sh 39.6% (28.0%) ' 4 0 . 0% (14.1%) About topics 20.6% (13.0%) 23.3% (9.4%) Europe Purpose: To learn Wr i t ing 25 .1% (9.2%) 53.3% (17.0%) Correct Engl i sh 49.5% (13.0%) 26.7% (12.5%) About topics 14.5% (9.4%) 20.0% (16.3%) 96 Findings Figure 23: Change in Students' Purposes in Taking a Writing Class: Cultural Group Profiles for Wanting to Learn How to Write Good Compositions 66 T 28 - -18 - -Ueek « week 8 • Japan Taiwan Europe 97 Findings Figure 24: Change in Students' Purposes in Taking a Writing Class: Cultural Group Profiles for Wanting to Learn How to Write Correct English Ucck 9 Heck B -•-Japan -•• Taiwan Europe 98 Figure 25: Change in Students' Purposes in Taking a Writing Class: Cultural Group Profiles for Wanting to Learn About Interesting Topics 99 Findings 4.3.3.3. Summary for Research Question 3c Overal l , the students' emphasis on learning correct Engl i sh decreased to a statistically significant extent while their interest in learning how to write good compositions increased (but not to a statistically significant extent) over the duration of eight weeks in the process-writing course. Learn ing about topics through wr i t ing appeared to become only slightly more important for these students. The Europeans appeared to change their attitude the most markedly, compared to the Japanese and Taiwanese students. 100 CHAPTER FIVE: DISCUSSION AND IMPLICATIONS The research described in this thesis forms a case study which made systematic analyses and comparisons among groups of students and their teachers ' approaches to two E S L wr i t ing tasks in one classroom environment. The overall approach was exploratory, descriptive, and naturalistic. This chapter stands back from the detailed analyses of the study (1) to discuss the salient findings of the research in reference to the broader context of other research on the teaching and learning of wr i t ing in Engl i sh as a second language, (2) to identify l imitations of the research, and (3) to suggest implications the present study may have for learners, teachers, and researchers. 5.1. DISCUSSION 5.1.1. Research Question 1: The Students' Priorities in Writing The adult E S L students in the present study attended most to aspects of discourse organization, that is, to the " i dea " level of composition. Cumming (1989), Raimes (1987), and Zamel (1983) also found that meaning was more important than language accuracy for the adult E S L students in their studies of E S L composing processes. In the present study, students apparently intuit ively used wr i t ing to develop their understanding of their topics (through focus on the meaning level of composition), whether they could have verbalized this as a conscious goal or not. Whether students focused more on content organization or rhetorical organization appeared to depend largely on the difference in task type, that is, they attended more to content organization while doing the Learn ing Record task and more to rhetorical organization while doing the Argument task (although these differences were not tested for 101 Discussion and Implications significance). The Argument task was assumed to be a more cognitively demanding wr i t ing task than the narrat ive Learn ing Record task, perhaps demanding more "knowledge-transforming" than "knowledge-tell ing" and hence resolving demands between content and rhetorical cognitive problem spaces (Bereiter and Scardamal ia, 1987). Although the Learn ing Record asked for students' opinions, the fact that it was based on fami l iar experiences probably made it less demanding on the learners ' wr i t ing abilities than the Argument task. In addition, their teachers ' instructions for the Learn ing Record explicitly requested students not to worry about language form. Raimes (1987) used two s imilar task types to the ones in the present study, one a letter with explicit purpose and audience and about personal experience, the other an argument about a complex issue, and found differences in students' behaviors. Cumming (1989) also contrasted adult E S L students ' performance on a less cognitively demanding task (a letter) wi th two more cognitively demanding wr i t ing tasks (argument and summary tasks) and found that they produced significantly different behaviors. The results Cumming obtained are also s imilar to the findings of this study with respect to attention to aspects of wr it ing for basic and average intermediate writers. Rhetorical and language concerns seem to be the aspects of wr i t ing that these students were most aware of as learning goals. The kinds of questions asked in class reflected this phenomenon. For example, some students wanted to know if this or that kind of rhetorical pattern was better, or else they had specific questions about language use, but they did not ask questions about the topic they were wr i t ing on or about how to compose. L ikewise, in the retrospective accounts of their intentions for the Learn ing Record, students devoted a greater percentage of attention to rhetorical language than while 102 Discussion and Implications actually doing the task (although this was not tested for statistical significance), i.e., the3' did not attend to it as much as they apparently thought they should. Perhaps the retrospective responses have their source in old " scr ipts " from previous educational t ra in ing which values the accuracy of expression over content. 5.1.2. Research Question 2a: Cultural Differences Differences between cultural groups or between the two wr i t ing tasks for cultural groups could not be tested for significant differences because the smal l number of people in each group made it impossible to do so meaningfully. In most analyses, moreover, differences across cultural groups appeared but were accompanied by considerable variat ion with in these groups. Thorough assessment of between group differences on intentions in E S L wr i t ing tasks w i l l require a larger scale, future study. Nonetheless, the students in the present study had roughly the same proficiency in Engl i sh and skills in wr it ing in their mother tongues, so cultural/educational differences may have accounted for differences which did occur in the findings. S imi lar ly , in it ia l differences in the groups' beliefs in their learning abilities may have affected their performance in this class and this research. The European group's greater confidence in their learning abil ity may be related to the fact that their philosophy of education (Western) may be closer to that of their teachers. Not only were the European students more confident themselves about their learning abil ity, but the teacher considered each one of them competent writers--at least by the standards of the c la s s - in spite of the fact that they had less formal education than most of the Japanese and Taiwanese students, and also, like the other students, rated themselves as basic writers (see Chapter 3, Section 3.1.3.2). 103 Discussion and Implications The cultural affinity with the teacher, the greater confidence in learning abil ity, or the teacher's perception of their greater wr i t ing competence, could al l be related to the finding that the European group uttered a much higher proportion of intention utterances than the other two groups. This finding may indicate that the Europeans were more intentional in their approach to writ ing. Other differences between cultural groups in the present study may be related to the findings of some contrastive rhetoric studies about what is important or valued in the wr i t ing of different cultures, and about argument structure. In the verbal protocols for both the Learn ing Record and the Argument tasks, the Japanese attended least of al l three groups to content organization. This finding lends support to the contrastive rhetoric studies which find that in Japanese wr i t ing form is valued over content (Wilkerson, 1986, cited in Grabe and Kap lan , 1989; Jenkins and Hinds, 1987). There was, however, a discrepancy between what the Japanese students said and what they did: in their retrospective accounts for the Learn ing Record and the Argument they considered content organization more than they did in their think-aloud protocols. Perhaps they verbalized in the retrospective interview what they had been trained in class to think of. The European group, however, talked less about content organization in the retrospective interview than they had in the verbal protocols. What students choose to talk about in the retrospective interviews may reflect an awareness of their weaknesses in writ ing, those aspects of wr i t ing they believe they should have focused on more. On the other hand, students may not need to attend to processes in the verbal protocols that are automatic for them, but rather to those aspects which are more difficult and thus require greater attention or control (Bereiter and Scardamalia, 1987). This interpretation may suggest that the Japanese students were less competent writers than the other participants since they 104 Discussion and Implications required more attention to rhetorical aspects of composing; for the other students these rhetorical aspects may have been more automatic. A s one European student remarked in his think-aloud protocol, Yeah, one, one thing is, yeah, is I, at the beginning I, I find ["unwriting"] useful, and then after, I don't know what happened, it 's I, I use it less and less and less. And, yeah probably because it became automatic for me, like to use it. L ike meantime I was writ ing, I divided my ideas at the same time (Learning Record task). Cu l tura l differences in argument structure may also be related to some differences in behavior across groups. Although Connor (1987) found that cross-culturally (England, F in land, Germany, and the U.S.), the argument structure of the best essays in her study was the same, she did not include non-Western essays or cr iter ia for mark ing (her method of analysis of these texts made reader-based assumptions and were rated by Amer ican Ph.D. students). Rhetorical differences are the most l ikely cause of the energy that at least three students—two Japanese and one Ta iwanese-put into balk ing at certain aspects of argumentation, that is, expecting disagreement and stating the opposing arguments in the paper. One Japanese student asked the teacher, an observer, and me at separate times during the same class why one should present opposing arguments—would that not weaken one's point? The teacher had anticipated this k ind of resistance, and had explained at the outset that the Nor th Amer ican academic community values persuasive wr i t ing which shows explicitly that both sides of the argument had been considered before the writer came to a conclusion. Nevertheless, in practise, these students st i l l resisted the inclusion of opposing arguments. None of the European students resisted in this way. 105 Discussion and Implications 5.1.3. Research Question 2b: The Students' Perceptions of their Experience Participants in this research reported a variety of perceptions of the present c l a s s -in which presumably students should have had the same experience-suggesting that their intentions and learning may have taken very different forms within the one context. The greatest difference in perceptions of the present class occurred with respect to language categories. The Japanese students perceived that they were doing more study of grammar, vocabulary and composition types than the European and Taiwanese students did (although this distinction could not be assessed for statistical significance). This difference may be due to the process-oriented nature of the class; perhaps some students imposed their own priorities on the process. Studies of contrastive rhetoric would support the Japanese emphasis on the language code (Jenkins and Hinds, 1987; Wilkerson, 1986, cited in Grabe and Kap lan , 1989). However, there was more consistency, as well as a clearer pattern of a focus on language, in reports about experiences in Engl i sh wr i t ing classes than in f irst language wr i t ing classes. The variat ion with in cultural groups in the reports of experiences in f irst language wr i t ing classes could be due to a number of factors, including memory and the influence of students ' present knowledge. The variat ion with in cultural groups that did emerge from reports about experiences in Engl i sh wr i t ing classes may be part ly a result of some students recall ing experiences in general Engl i sh classes and others recall ing experiences in Engl i sh writing classes. The Japanese and European students ' preferences for an ideal class were closest to their perceptions of the present class. This tendency would indicate either that these two groups adapted their preferences to those of the teacher, or that the methods of the present class were appropriate for them. The studies cited by Reid (1987) showing that 106 Discussion and Implications adult students adapt to instructor styles would support the former interpretation; process-wr i t ing advocates (Betancourt and Phinney, 1988; Mo l l and Diaz, 1987; Trueba, 1987; Zamel, 1982) might support the latter. Perhaps the more students adhere to their previous tradit ional tra in ing in wr i t ing or language study, the more difficulty they might have with a process approach to learning writ ing. 5.1.4. Research Question 3a: The Teacher's Priorities The teacher in the present research devoted most of his attention to pedagogical remarks while reading over his students' compositions and secondarily to the organization of the content of the compositions. In both tasks, the teacher adhered to his pedagogical functions as wel l as to his stated objectives for the tasks, attending to content concerns in the Learn ing Record task and shifting to proportionally more concerns for rhetorical organization (as per his stated teaching objectives) for the Argument task. The reason the teacher attended less to content organization for the Taiwanese group and made more pedagogical remarks than for the other two groups is l ikely because two of the Taiwanese students did not follow the instructions for the Learn ing Record, and for the Argument, one students ' wr i t ing was especially difficult to comprehend. A l lowing for those factors, the teacher applied his cr iter ia consistently across the three groups. The teacher's overwhelming attention to content belies those studies accusing E S L teachers of only focusing on language concerns, as indicated by marg ina l comments and error corrections in students' texts (e.g., Zamel, 1985). However, the present students were not being graded by any outside criteria, such as a standardized test, but only by the teacher's cr iter ia as outlined in his instructional objectives. In any case, the lesser focus on language on the teacher's part does not mean that he thought language accuracy 107 Discussion and Implications unimportant; he f i rmly stated his belief that students learn language through the process of wr i t ing and expressing meaning, regardless of whether they share this belief (see Cumming, 1990; Zamel, 1983). 5.1.5. Research Question 3b: Fit Between the Teacher's and Students' Intentions The main difference between the teacher's and students' profiles in the verbal protocols was that the teacher focused more on content organization than the students did, for both tasks. This finding may be a result of the beliefs the teacher holds about the purpose of wr i t ing (to express meaning), or of the objectives he set for the tasks and his adherence to them as cr iter ia for commenting on the students ' written efforts; or his focus on content may simply reflect priorities in the reader-writer relationship (i.e., to receive a message " sent " by the writer, and only secondarily to "decode" where necessary). In spite of this extra attention to content on the teacher's part, the rank order of attention to aspects of wr it ing by the teacher and students was the same for the whole group, so apparently there was a good " f i t " between teacher and student intentions. The present results show a greater congruence than was obtained by Cohen and Cavalcant i (1990) in a s imi lar comparison between L 2 students ' and teachers ' priorities in composition writ ing. In other words, the teacher and students shared an understanding of the goals of their assignments and by this criterion, the instructional approach appeared relevant (Langer and Applebee, 1987). This shared understanding was probably possible because the teacher gave very clear, discrete, and limited written instructions for the tasks. These very clear instructions possibly provided the context to shape the intentions of the students documented in this research, suggesting that this case study might present a situation in which the students 108 Discussion and Implications let the teacher's " Ideal Text " matter more than their own. This possibility may have been especially important for the Argument task, in which a number of organizational aspects were emphasized in the teacher's instructions. A s Ruggles Gere and Abbott (1985) found, a teacher's assignments appear to determine how students talk about their work. The students in the present study did not appear to be too concerned about their teacher's opinions of their wr i t ing (they were not being graded, after all), but they did generally focus on aspects of wr it ing that the teacher wanted them to focus on. On the other hand, the students may have mostly agreed with the benefits of using the teacher's instructions as a checklist for their assignment; most of them referred to these instructions as they wrote. Perhaps these students would have obeyed the teacher in a product-oriented class too. Another explanation for the good " f i t " between teacher and student intentions is that the students accommodated to the teacher's values as they socialized into the classroom environment. A s Breen (1987, p. 145) points out, the cultures of second language classrooms tend to be highly normative and to invite conformity rather than divergence. 5.1.6. Research Question 3c: Change in the Students' Global Intentions The teacher wanted above al l to teach wr i t ing processes, and secondarily rhetorical structures. Language and content learning were to be by-products of the wr it ing process. Thus the factor students and teacher were most l ikely to disagree on (since the students' more overt priorities were rhetorical structures and language) was the place of language teaching in the wr i t ing class (cf. Rothschild, 1991). However, the students ' priority for 109 Discussion and Implications learning language did lessen significantly over the duration of their being in this present class. Analyses of questionnaire responses about students ' in i t ia l purposes in enrolling in the course and their present priorities in the course showed all students changing their attitudes as to the purpose of a writ ing class: a l l came increasingly to accommodate the teacher's values, but the most dramatic change occurred with the European group. This difference could be a result of a cultural aff inity wi th the teacher. In the North Amer ican context, students without such a cultural aff inity may adapt more easily to "Wes te rn " methods than they might have in their own country if a Western teacher had tried to implement a process approach to writ ing. Rothschild (1991) found students to be aware and accepting of "wha t is l ikely a novel approach" (p. 72) and concluded that they recognized the advantages outweighed the disadvantages. Perhaps the shift in students' attitudes reflects a new belief that they use wr i t ing to develop their understanding of a l l aspects of writ ing. 5.2. LIMITATIONS AND ASSUMPTIONS Given the current lack of knowledge of the issues explored in this research, it was impossible to formulate clear hypotheses about cultural differences in E S L wr i t ing classes or to set up a large scale study to assess them. The results are focused, then, not on hypotheses but on open questions and descriptive analyses. A related l imitation is that since a very smal l number of participants, from only three cultural groups, in one intact class at a certain point part way through one course were used, the sample is not representative of any part icular population and thus findings cannot be generalized to other situations or populations. These E S L learners appeared to have relatively low 110 Discussion and Implications writ ing skills in their mother tongues, and may not represent the ful l range of wr it ing skills possible to assess. Moreover, al l participants for the research were volunteers who were aware of the general goals of the study, a factor which may have biased their wr i t ing performance or self-reports. However, the advantage of describing an intact group of participants is that the natura l "cu l ture " of one class was preserved and not art if ic ial ly manipulated to conform to experimental constraints. A potential l imitation of the research method of the " th ink-aloud" protocol is that all but one of the participants did not use their f i rst languages in the verbal reports, possibly constraining their ability to " th ink " aloud. Although they refused the researcher's request to think aloud in their first languages, they might have used their first languages to represent their thinking processes had a bil ingual researcher been present. F ina l l y , in the data analysis, although certain aspects of the data were analyzed statistically, the many "non-significant" results were difficult to interpret as they indicated the absence, rather than confirmation, of significant differences, and probably arose as a result of the smal l number of participants in the research or within-group variation in the analyses. Such results should therefore be treated with caution. Some general logical assumptions were made in this study which may be open to question. F i r s t , the research assumed that human intentions are central to learning (see Bereiter and Scardamal ia, 1989), although no assessments were attempted to establish whether learning did or did not occur in the present context. Second, it was assumed that the student participants were generally motivated and interested in "w r i t i n g " and in contributing to this research, a fact attested to by the large number of in it ia l applicants for the research study (18). Thirdly, the students ' verbalizations of their composing processes 111 Discussion and Implications were assumed to represent their intentions, although this correspondence between verbalization and cognition is open to debate (Cooper and Holzman, 1983; Voss, 1983). 5.3. IMPLICATIONS 5.3.1. Implications for Learning and Teaching The present study suggests several implications for curr iculum design and teaching practice. This study confirms previous research concluding that more cognitively demanding wr it ing tasks involve different kinds of behaviors than less cognitively demanding tasks, suggesting this principle may be important when designing curr icula or assessments based on E S L wr i t ing (Cumming, 1989). Individual student differences also must be expected by E S L teachers, as both common sense as well as the findings of this study indicate. This study cannot make generalizations based on ethnicity, nor is it desirable to do so because of the possible negative effects of stereotyping; nevertheless a teacher can observe whether students are underconfident learners, for example, or whether the rhetoric of their native language has potentially different values or structures than that of Engl ish. That students appeared to have adapted to or socialized into this part icular class may indicate that in their own context (i.e., North Amer ican, in this case) teachers can implement methods which seem intuit ively right, or theoretically justified, even i f the characteristics of these methods are typical ly "Western " , and could be unacceptable in students ' native countries. Rothschild (1991) comes to a s imilar conclusion, at least for a workshop or process approach to teaching writ ing. Probably i t would be difficult for 112 Discussion and Implications teachers to use these same methods in Ta iwan or Japan , as Sampson (1984) and Burnaby and Sun (1989) suggest. 5.3.2. Implications for Research One clear implication for research is the value of multiple approaches to data collection on E S L writ ing instruction. For example, concurrent verbal reports, retrospective interviews, questionnaires, and classroom observations proved to corroborate many findings in the research. Although the purpose of observations in the present study was to t ry to see what students' intentions appeared to be during the class, it was extremely difficult to use observational methods for more than a supplementary approach to the more consistent data of the think-aloud and prompted recall protocols and questionnaires. Na iman, Frohl ich, Stern and Todesco (1978) came to a s imi lar conclusion in analyz ing oral strategies for second language learning. However, such a supplement was valuable, putting the think-alouds, retrospective interviews, and questionnaires into context and facil itating some of the insights in this chapter. The observations showed that the think-aloud behavior was congruent with the class activity. For instance, having seen the class in action, I knew that in the think-aloud sessions, the students were consciously applying what they had learned: "O.K., f i rst I have to write an introduction" (Argument task). "I want to say objections to my statement" (Argument task). In addition, some participants proved easier to study using think-aloud methods than others. The European group, for example, made almost as many utterances as the two other groups combined, thereby providing considerably more data for analyses. For 113 Discussion and Implications some participants, extra practise in thinking aloud with the tape recorder could have been helpful. This study also demonstrates the value of considering E S L wr i t ing holistically within a classroom context which accounts for the perspectives of a teacher and various subgroups of students. Whereas many previous studies have assessed the performance of either E S L student groups or of teachers alone on experimental tasks, the present research was able to demonstrate a close relation between a teacher's and students ' intentions alike, capturing relations which may more closely resemble the actual processes and interactions of classroom situations than experimental task analyses would permit. 5.3.3. Suggestions for Further Study In exploring the central themes of this thes i s -cu l tura l var iat ion and learning intentions—two distinct kinds of culture emerged: the first language culture of the individual participants, and the classroom culture, jointly constructed by these individual participants (Breen, 1987). It would appear that in the present study, the classroom culture overrode the individual cultures represented, and the data elicited therefore may not represent Japanese, Taiwanese, and European cultures as much as i t represents the unique culture of the particular class group observed. This issue should be taken into account in future research, and could be assessed in more controlled, larger scale studies. In order to determine the effects of first language culture on learning intentions, whole classroom groups of students sharing a first language culture could be compared, perhaps contrasting foreign language and second language settings (for example, Japanese learners of Engl i sh in Japan vs. Japanese learners of Engl i sh in Canada). Such large group comparisons would permit significant differences to 114 Discussion and Implications be tested for between groups while accounting for contextual variation. Longitudinal studies would also be very revealing. To what extent would the students ' behavior differ from their present behavior in another three months? S ix months? The observation in the present study that different wr it ing tasks elicit different behaviors could also provide another line of inquiry in future research. Larger scale comparisons across tasks, and across wr i t ing abil ity levels would be worthwhile. F ina l l y , with regards to the fit between teacher's and students' intentions, and the extent to which students may s imply do what a teacher tells them to do, i t would be instructive to assess the performance of a much larger number of teachers. In addition, other kinds of classroom cultures, such as those of more product-oriented classes, would provide worthwhile comparisons. A n increased number of studies of classroom cultures would deepen our understanding of how a priori differences between individual participants, for example those ar is ing from their native society and its values, may be modified in the classroom context. 115 REFERENCES Bereiter, C , & Scardamalia, M . (1987). The psychology of written composition. Hi l lsdale, New Jersey: Lawrence E r lbaum Associates. Bereiter, C. and Scardamalia, M . (1989). Intentional learning as a goal of instruction. In L.B. Resnick (Ed.), Knowing, learning, and instruction (pp. 361-392). Hi l lsdale, N J : Lawrence E r lbaum Associates. Betancourt, F., & Phinney, M . (1988). Sources of wr i t ing block in bil ingual writers. Written Communication, 5, 461-478. Brannon, L., and Knoblauch, C. H. (1982). On students ' rights to their own texts: A model of teacher response. College Composition and Communication, 33, 157-166. Breen, M . P. (1987). The social context for language learning: A neglected situation? Studies in Second Language Acquisition, 7, 135-158. Brindley, G. (1986). Some current issues in SLT . Prospect, 1, 11-44. Burnaby, B., & Sun, Y . (1989). Chinese teachers ' views of Western language teaching: Context informs paradigms. TESOL Quarterly, 23, 219-238. Chafe, W. L. (1985). Differences between speaking and writ ing. In D. R. Olson, N. Torrance, and A. H i ldyard (Eds.), Literacy, language and learning: The nature of consequences of reading and writing (pp. 105-123). New York: Cambridge Univers i ty Press. Cohen, A., and Cavalcanti , M . C. (1990). Feedback on compositions: Teacher and student verbal reports. In B. K r o l l (Ed.), Second language writing: Research insights for the classroom (pp. 155-177). N e w York: Cambridge Univers i ty Press. Cohen, A. (1987). Studying learner strategies: How we get the information. In A. Wenden and J . Rubin (Eds.), Learner strategies in language learning (pp. 31-40) Toronto: Prentice Ha l l . Connor, U. (1987). Argumentat ive patterns in student essays: Cross-cultural differences. In U. Connor & R. B. Kap l an (Eds.), Writing across languages: Analysis ofL2 text. Reading, Mass: Addison-Wesley. Cooper, M., & Holzman, M . (1983). Ta lk ing about protocols. College Composition and Communication, 34, 284-293. S Cumming, A. (1989). Wr i t ing expertise and second-language proficiency. Language I Learning, 39, 81-141. 116 References Cumming, A. (1990). Metalinguistic and ideational thinking in second language composing. Written Communication, 7, 482-511. Cumming, A. (1991, in press). Instructional routines in E S L composition teaching. Journal of Second Language Writing, 1. El l i s , R. (1985). Understanding Second Language Acquisition. Oxford: Oxford Univers i ty Press. Emig , J . (1977). Wr i t ing as a mode of learning. College Composition and Communication, 28, 122-128. Ericsson, K. A., & Simon, H. A. (1984). Protocol analysis: Verbal reports as data. Cambridge, Mass: M.I.T. Press. Fa thman, A. K., & Whal ley, E. (1990). Teacher response to student writ ing: Focus on form versus content. In B. K r o l l (Ed.), Second language writing: Research insights for the classroom (pp. 178-190). New York: Cambridge Un ivers i ty Press. Flower, L., & Hayes, J . R. (1981). A cognitive process theory of writ ing. College Composition and Communication, 32, 365-387. Freedman, A., & Pringle, I. (1980). Reinventing the rhetorical tradition. In A. Freedman & I. Pringle (Eds.), Reinventing the rhetorical tradition (pp. 173-185). Conway, Arkansas: L & S Books, Un ivers i ty of Centra l Arkansas for C C T E . Grabe, W., & Kap lan, R. B. (1989). Wr i t i ng in a second language: Contrastive rhetoric. In D. M . Johnson & D. H. Roen (Eds.), Richness in writing: Empowering ESL students (pp. 263-283). N e w York: Longman. Hamp-Lyons, L. (1986). No new lamps for old yet, please. TESOL Quarterly, 20, 790-796. Hayes, J . R., & Flower, L. S. (1980). Identifying the organization of wr i t ing processes. In L. W. Gregg and E. R. Steinberg (Eds.), Cognitive processes in writing (pp. 3-30). Hil lsdale, New Jersey: Lawrence E r lbaum Associates. Hinds, J . (1987). Reader versus writer responsibility: A new typology. In U. Connor & R. B. Kap lan (Eds.), Writing across languages: Analysis ofL2 texts (pp. 141-152). Reading, M A : Addison-Wesley. Horowitz, D. (1986). Process, not product: less than meets the eye. TESOL Quarterly, 20, 141-144. Jenkins, S., & Hinds, J . (1987). Business letter writ ing: Engl ish, French, and Japanese. TESOL Quarterly, 21, 327-349. 117 References Jones, S., & Tetroe, J . (1987). Composing in a second language. In A. Matsuhash i (Ed.), Writing in real time: Modeling production processes (pp. 34-57). Norwood, New Jersey: Ablex. Kumaravad ive lu , B. (1990). Classroom observation: A neglected situation. TESOL Newsletter, 24, pp. 5, 32, 34. Langer, J . A., & Applebee, A. N. (1987). How writing shapes thinking: A study of teaching and learning. N C T E Research Report No. 22. Lek i , I. (1990). Coaching from the margins: Issues in written response. In B. K ro l l (Ed.), Second language writing: Research insights for the classroom (pp. 57-68). New York: Cambridge Univers i ty Press. L iebman-Kleine, J . (1986). In defense of teaching process in E S L composition. TESOL Quarterly, 20, 783-788. M c K a y , S. L. (1989). Topic development and written discourse accent. In D. M . Johnson & D. H. Roen (Eds.), Richness in writing: Empowering ESL students (pp. 253-262). New York: Longman. Mohan, B., & Lo, W. A. -U. (1985). Academic wr i t ing and Chinese students: Transfer and developmental factors. TESOL Quarterly, 19, 515-534. Mo l l , L. C , & Diaz, S. (1987). Change as the goal of educational research. Anthropology and Education Quarterly, 18, 300-311. Na iman , N., Frohl ich, M., Stern, H. H., & Todesco, A. (1978). The good language learner. Toronto: OISE. Ng, G. A. (1975). The educational background of the adult Chinese student. TESL Talk, 6, 36-40. Ostler, S. (1987). Engl ish in parallels: A comparison of Engl i sh and Arab ic prose. In U. Connor & R. B. Kap l an (Eds.), Writing across languages: Analysis of L2 texts (pp. 169-185). Reading, M A : Addison-Wesley. Politzer, R. L., & McGroarty , M . (1985). A n exploratory study of learning behaviors and their relationship to gains in linguistic and communicative competence. TESOL Quarterly, 19, 103-124. Radecki, P. M., & Swales, J . M . (1988). E S L student reaction to written comments on their wr itten work. System, 16, 355-365. Raimes, A . (1987). Language proficiency, wr i t ing abil ity, and composing strategies: A study of E S L college student writers. Language Learning, 37, 439-467. 118 References Reid, J . (1987). The learning style preferences of E S L students. TESOL Quarterly, 21, 87-111. Robinson, G. L. (1981). Issues in second language and cross-cultural education. Boston, Mass.: Heinle and Heinle. Rothschild, D. T. (1991). The Adu l t Engl ish as a second language writer and the wr i t ing workshop approach: Performance, biodemographic variables, and attitudes. Master ' s Thesis, Un ivers i ty of Br i t i sh Columbia, Vancouver, Canada. Ruggles Gere, A., & Abbot, R. D. (1985). Ta lk ing about writ ing: The language of wr i t ing groups. Research in the Teaching of English, 19, 362-385. Sampson, G. P. (1984). Export ing language teaching methods from Canada to China. TESL Canada Journal, 1, 19-31. Semke, H. D. (1984). The effects of the red pen. Foreign Language Annals, 17, 195-202. S i lva, T. (1990). Second language composition instruction: Developments, issues, and directions in E S L . In B. K r o l l (Ed.), Second language writing: Research insights for the classroom (pp. 11-23). New York: Cambridge Univers i ty Press. Smith, F. (1982). Writing and the writer. Toronto: Holt, Rinehart and Winston. Swain, M . (1986). Communicative competence: Some roles of comprehensible input and comprehensible output in its development. In J . Cummins and M . Swain (Eds.), Bilingualism in education (pp. 116-137). Trueba, H. T. (1987). Organizing classroom instruction in specific sociocultural contexts: Teaching Mex ican youth to write in Engl ish. In S. R. Goldman and H. T. Trueba (Eds.), Becoming literate in English as a second language (pp. 235-252). Voss, R. F. (1983). Janet Emig ' s the composing processes of twelfth graders: A reassessment. College Composition and Communication, 34, 278-283. Wenden, A., & Rubin, J . , Eds. (1987). Learner strategies in language learning. Toronto: Prentice-Hal l. Zamel, V . (1982). Writ ing: The process of discovering meaning. TESOL Quarterly, 16, 195-209. <—^ Zamel, V. (1983). The composing processes of advanced E S L students: S ix case studies. TESOL Quarterly, 17, 165-187. Zamel, V. (1985). Responding to student wr it ing. TESOL Quarterly, 19, 195-202. 119 APPENDIX A: LETTERS TO PARTICIPANTS February , 1990 Dear colleague: I am writing to ask if you would be willing to participate in a study of the intentions of students in an ESL writing class. The study, called "Cultural Variations in Adult Students' Learning Intentions for Writing in ESL," aims to develop a better understanding of how students learn to write in their second language. Participation in the study will require about two hours of your time, in addition to several hours during which I would like to observe the participating students' behaviour in class. You will be asked some questions about your goals for the course and specifically, for two learning tasks you have given the students. You will be asked to give your impressions of the participating students' completed assignments by giving verbal reports of what you pay attention to while marking them. (A demonstration of how to do this will be given before the session begins.) These verbal reports will be tape-recorded and analyzed as data for the study. Your participation in this research will be completely confidential. Your name will be concealed by use of a pseudonym in analyzing and reporting the results. You are not, of course, under any obligation to participate. You may withdraw from the study at any time for any reason. Refusal to participate will not jeopardize your employment status. If you would like to participate, or you would like further information about this project, please telephone me at . or complete the attached consent form and return it to me at my mail box at I will contact you to arrange a convenient time and place to meet to further explain the research. Thank you for your interest. Sincerely, Catherine Ostler Howlett Dear Catherine: I would be interested in participating in the research project described in your letter of February , 1990. I understand that I will be able to withdraw from this project at any time. I have retained a copy of the letter for my reference. name . signature telephone date 120 Appendix A: Letters to Participants March 5, 1990 Dear student: I am an instructor at and a graduate student in the Department of I am writing to ask if you would like to participate in a research project called "Cultural Variations in Adult Students' Learning Intentions for Writing in ESL." You will be asked some questions about your experience of writing and writing classes, and then you will do two writing tasks (homework from your class) in English. These tasks will be at different times, a couple of weeks apart. Each task will take about an hour to complete, taking about 2-3 hours in total, including filling in the brief questionnaire about your writing experience. In exchange for your time, I will offer one hour of tutorial or English conversation. The purpose of the study is to see how people learn to write in a second language. While you are doing the writing tasks for the study, you will be asked to say aloud what you are thinking about (in your native language or in English). This will be tape-recorded and analyzed later. Techniques to do this will be demonstrated and practised at our first meeting. This research is not connected in any way to courses at The names of people who participate in this study will be strictly confidential. If you decide to participate in the project, you will be free to drop out of it at any time. Refusal to participate or withdrawal from the project will not influence your academic status or your performance evaluation in this course. If you wish to participate, or you would like further information about this project, could you please telephone me at or complete the attached consent form and return it to me at the (I have a mail box in the main office), or give it to your teacher. I will contact you to arrange a convenient time and place to meet to explain the research further. Thank you for your interest. Sincerely, Catherine Ostler Howlett Dear Catherine: I agree to participate in the research project described in your letter of March 5, 1990. I understand that I will receive one hour of tutorial or English conversation in exchange for my time. I will be able to withdraw from this project at any time, if I so choose. My withdrawal from the project will not influence my performance evaluation in this class. I have kept the letter for my reference. name " signature telephone date 121 APPENDIX B: QUESTIONNAIRE Name: \ , Country in which you received your education: Language of education: Major subjects at school/university: (Humanities, Sciences, Vocational) Length of time in Canada: years months Length of time studying English composition in your own country: years months Length of time studying English composition in Canada or other English-speaking place: years months How would you rate yourself as a language learner? excellent good fair poor How would you rate yourself as a learner of written composition? excellent good fair poor How would you rate yourself as a learner in your field of study or work? excellent good fair poor Can you, or could you, in your mother tongue, ( P I e a s e c i r c I e ^ n u m b e r ) a) write letters to people you do not know? 1 2 3 4 5 b) write reports at school? 1 2 3 4 5 c) write reports at work? 1 2 3 4 5 122 Appendix B: Questionnaire Was/Is your purpose in enrolling in an ESL writing class to (Enter approximate percentages, totalling 100%) At the beginning of this term At this time —learn how to write good compositions —learn how to write correct English —learn about topics you are interested in writing about What kinds of activities do you remember doing in writing classes? Enter 0 (none); 1 (a little); 2 (some); or 3 (a lot) Primary/El emenlary/S LI econdary English College/1 LI Ini versity English. Now English Which are, or would be, helpful to you now? (English) . reading books or studying literature studying grammar writing multiple drafts of an essay studying vocabulary studying about composition types (e.g., argument, cause and effect, etc.) memorizing texts by expert writers studying punctuation, capitalization, etc. keeping a journal studying about the organization of written composition other (please explain) 123 APPENDIX C: INSTRUCTIONS FOR TASKS Learning Record for Week 9 Write a description of this wr i t ing course (approximately 300 words). In your opinion what is the most useful activity we've done? What is the least useful? Do you have suggestions for improving this course in the future? Don ' t worry about language form. The purpose is to give us your opinions. We wi l l not mark your paper. Argument (The instructions for each draft were written on the blackboard.) D ra f t #2 of F i r s t Argument -- 3+ opposing - Balance - Transit ions - Connecting Words Dra f t #3 of F i r s t Argument - Intro -- Body ~ Conclusion F i r s t Draf t of Second Argument Opinion: New Dra f t 124 EAPPENDIX D: INSTRUCTIONS FOR THINKING ALOUD (after Ericsson and Simon, 1984, p. 378) In my research, I am interested in what you think about, part icular ly your intentions, when you are composing. I am going to ask you to T H I N K A L O U D as you do a wr i t ing assignment. I don't want you to t ry to plan out what you say or t ry to explain to me what you are saying. Jus t act as i f you are alone in the room speaking to yourself. It is most important that you keep talking. If you are silent for any long period of time I w i l l ask you to talk. Do you understand what I want you to do? Good. I ' ll demonstrate wi th an algebra problem [the simplification of 3(x-4)(x-5) + 3x - 2x]. A re there any questions? Good. Now we wi l l begin wi th some practice problems. F i r s t , I want you to mult ip ly these two numbers in your head and tell me what you are thinking as you get an answer. " W h a t is the result of mult ip ly ing 24 x 36?" Good. Now I wi l l give you two more practice problems before we proceed with the ma in experiment. I want you to do the same thing for each of these problems. I want you to think aloud as before as you think about the question. A n y questions? Here is your next problem [What is 1082 - 395?]. Good, now here is another practice problem. Please think aloud as you try to answer it. "How many windows are there in your parents ' house?" 125 APPENDIX E: QUESTIONS FOR "PROMPTED RECALL' Global 1. What were your intentions/goals/objectives for this composition when you started to write? (Please explain). 2. What were your intentions/goals/objectives for this paragraph? [each paragraph]. 3. What were your intentions/goals for this composition when you finished writing? (Please explain). Local 4. W h y did you change/underline/circle this? [for three areas of attention]. 5. What wi l l you do to this piece of wr i t ing before you hand it in/are finished with it? 126 APPENDIX F: INSTRUCTIONS TO THE TEACHER Global What were your objectives for this assignment? What are you looking for when you check over this assignment? Local [the same tra in ing and "warm-up " as for students ' think-aloud protocols] W i l l you please think aloud as you go over each assignment? 127 APPENDIX G: SAMPLE OF A STUDENT'S VERBAL PROTOCOL [The numbers in the protocol correspond to those on this student's piece of writ ing (Appendix H).] 1. So I have to make an introduction, and I have a general statement, but I don't know about narrow, or organizing, opinion. Have to put opinion. [Reads "Should [the institute] have certain claim to restrain adult students?"] I don't think so, so [writes] 2. Next is I have to make a body. [That] is the reasons opposing argument. We have an opposite opinion: [shock?], miss lectures, disturb class, [enter?] teacher general policy. U h , I understand the, the student who absent from class have a shock, it 's uh, O.K. One of the most important things is the, [reenter?] teacher,and disturb class, and miss lecture. This number two, and [reads]. I'd like to s t ren- I 'd like to make opinion more strong. So that I have to, to, to connect the no side and yes side. Let ' s see [writes, but not on this draft-notes? " Adu l t students must know how teachers feel when they are absent from their classes."] U m , mature . . . and my, my former draft said "most of the [institute] students are over 18 years old . . . to control themselves." A n d my f irst draft doesn't have an opposite opinion, so that I have to, to have—make an opposite opinion, and after I have to put my opinion to make my sentence stronger. [Reads "The adult students must know how teachers feel when they are absent from their classes."] I have to, to make a sentence using the transit ion or connecting words, let 's see. [Reads "[on] the other hand, so"] I think the " so " is the better. [Reads over the sentence with "so."] It 's very difficult to make opposite opinion. Why I'd like to strengthen my opinion is, uh [reads "most of the [institute] students are over 18 years old. That is they are mature enough to control themselves. They must understand what is important or useful to use much money on abroad"]. O.K. That ' s one paragraph. And, this paragraph said over 18 years old students are mature enough to control themselves. That means mature students must know how teachers feel. So, they should know, they should know. O.K. I'd like to change from "mus t " to "should." Adu l t students should—should know how teachers feel when they are absent from their classes. [Writes, but not on this draft-notes? "So they should, they must know which classes which is important or useful to use much money."] And I wrote, I think this is not good words. I'd like to change "to use much money in fore ign" - I don't know how to spell th i s - " fore ign country." [Reads " Adu l t students should know how teachers feel when they are absent from their classes. So they must know what is important or useful to use much money in foreign country."] No, it doesn't make sense. 128 Appendix G: Sample of a Student's Verbal Protocol M m , these are no t - i t doesn't-these sentences don't connect [erases them]. The f irst I'd l i ke - I should make an introduction. [Reads "There are quite strict attendance policies at [the inst itute]" I think there is my opinion. [Reads " In my opinion the [institute] should not care of students absences."] O.K. Bu t I think this, this introduction is weak. But it ' s O.K. now. And , O.K., next, body. M m , body must have reasons opposite [inaudible] argument. U h , why do I think students not, not students can absent from schools? [Reads over reasons.] M m , O.K. [writes] 3. I don't know how to spell "control. " 4. [Reads "The reasons why I do--"] I think I don't need "do." ["why I think so are that most of the students"] that is, are f i rst of all—I think I should put the " f i r s t " because I have a few reasons. ["The reasons why I think so are f irst of al l t h a t - " ] that, I don't know which i s - a f te r " t ha t " or before " that . " [Reads "I think so that," "The reasons why I think so are that," " f i r s t of a l l " I think after " that . " [Reads "most of the students are over eighteen years old, that is, they are mature enough to control themselves."] [writes] 5. I didn 't put any opposite, opposite, mm, opposite opinions so I have to, O.K. U m , [reads, "some students have own jobs which they have to do, at same time as classes. Also, every student might get an accident, visitors from their countries or personal problems. The [inaudible] should not interfere wi th business. Only students miss the classes and lose money. These, my opinions are w e a k - a little bit weak. M m , I'm confused. [Reads "I understand that absent students disturb classes. Bu t they might have jobs or accidents or visitors from their countries or personal problems."] Yeah, that 's, that 's, that is good. "A lthough, " this is another sentence. 6. Oh, I wrote the same sentence before, um, [reads "The reasons why I think so are that most of the students are over 18 years . . . that is, they are mature enough to control themselves"]. H m . O.K. [Reads "A l though . . when they are absent,"] [writes] 7. [Wrote in " a t same time as classes."] No, their jobs they have to do. I think I have to erase this. O r "They might have own jobs which they have to do" or mm, accident, or get an accident, accident or, no, I don't like " o r " in here, [writes] 8. I don't know that word, anyway, " in jured" 129 Appendix G: Sample of a Student's Verbal Protocol 9. [Reads the last paragraph she has written] (or just lazy.) Ju s t from . . . lazy, laziness. I don't know there is this kind of word. Lazy , laziness. I ' ll put the~this one. Laziness. U h , next paragraph is [reads from previous draft or notes: " If teachers or classmates are disturbed by other students' absence, they have obvious reasons which are blamed by others. Example, we upset a little when teacher explain the matter which was supposed to be done before to some students. We waste our time doing it. In addition to, they are mean to teachers who have to come and teach us even terrible ra iny day or when they are in bad mood. We have duty to response. That is a conscientious decision. Adu l t students should understand the feelings of others"]. M m . This paragraph is " no " side. I can ' t write that. I have to make my paragraph stronger, so this is not good. M y opinion or reasons is they have a job, they are mature enough, they have [inaudible] responsibility, and they can decide, or they have respect. H m . I think these reasons and, and this [inaudible] opinion, are connected each other, so I can't, I think I can 't make two or three paragraphs. This paragraph says [inaudible]. It means they have -a re mature enough to control themselves. The second paragraph says they have own business, O.K. A n d the th ird one is um, um, [This side] is to refer to paragraph 1. So, I use work, I use mature, uh, I have to use responsibility and, and respect. M m . O.K. I think [that's nice?] and the draft one and third paragraph said "some people feel bored and sleepy—or sleepy in class. In my opinion, they can be excused to attend, attend, because they bother other students." O.K. [writes] 10. Ah , not " the, " I think " a " is better. " A person, student who does n o t - " H m , my letter is not good. 11. "also," no, not "also." 12. O.K. But I think I don't st i l l have, have uh, have any opposite opinions. I think much better than before. I think this is done. Finished! 130 APPENDIX H: SAMPLE OF THE ARGUMENT TASK [Numbers indicate the point during the composing process at which the student verbalized her thoughts; letters indicate the point during the teacher's reading at which the teacher verbalized his thoughts. The numbers correspond to those in the student's verbal protocol (Appendix G) and the letters correspond to those in the teacher's verbal protocol (Appendix I).] There are quite strict attendance policies at [the institute]. Few students have been left because of them so f a r . ^ Sould [the institute] have certain cal im to restrain adult students?-^ In my opinion, [the institute] should not care of students absences. 2 ^ f .Vst o f a l l The reasons why I think so are that A most of the students are over 18 years old, that is, they are mature enough to con t ro l 3 themselves. 4 The must understand what is important or useful to use money in [ ] country.^ Although everybody understands the students^ disturb the classes when they are absent, ^  they might have own jobs-^ which they have to do,^ get an accident, visiters from their countries, or personal problem, which is sickness,-^ [ ],^ or just from [ ].9 The teachers also have responsibilites to make classes interesting. M a n y students feel bored or sleepy in classes. Those people might bother other students. For example, a person ^ who does not have homework, his partners H cannot work wi th h im, it means, the partners weist their time for his [ ]. So, I prefer they do not come to classes. 131 APPENDIX I: SAMPLE OF THE TEACHER'S VERBAL PROTOCOL [The letters in the protocol correspond to those on the student's piece of wr i t ing (Appendix H).] A. Have left, have been kicked out B. Yukiko? I 'm surprised. C. O.K., they 're responsible, she says. D. O.K., she's presented an opposing point there. E. Yeah, so she's pointing out exceptions. She agrees that absenteeism is disruptive, but there are exceptions. F. Wel l , that ' s an interesting point of view. If they 're not prepared, they should be absent. General ly speaking, she's done a fa i r ly good job of coherence, wi th the transitions, and uh, a number of them, for example, " i n my opinion," "the reasons why, " " f i r s t of a l l , " "a lthough," "teachers also have responsibilities," " for example." General ly speaking she's attempted to make it fa i r ly coherent. I don't think she's developed the topic as well as i t should have been at this stage, for al l the work that we put into it. She's really said we need rules, but there are exceptions, then she lists the exceptions. A n d she has a reasonable var iety of sentence structure. 132 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0078331/manifest

Comment

Related Items