UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Teacher decision-making in the ESL classroom : the influence of theory, beliefs, perceptions and context Smith, Deborah Binnie 1992

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


831-ubc_1992_fall_smith_deborah_binnie.pdf [ 16.99MB ]
JSON: 831-1.0055028.json
JSON-LD: 831-1.0055028-ld.json
RDF/XML (Pretty): 831-1.0055028-rdf.xml
RDF/JSON: 831-1.0055028-rdf.json
Turtle: 831-1.0055028-turtle.txt
N-Triples: 831-1.0055028-rdf-ntriples.txt
Original Record: 831-1.0055028-source.json
Full Text

Full Text

TEACHER DECISION-MAKING IN THE ESL CLASSROOM: THE INFLUENCE OF THEORY, BELIEFS, PERCEPTIONS AND CONTEXT by DEBORAH BINNIE SMITH B.A., McMaster University, Hamilton, Ontario 1967 Dip.Ed. University of Toronto 1968 Dip.TESL, Royal Society of Arts, London, England , 1973 M.Ed. Acadia University, Wolfville, Nova Scotia, 1983 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY In THE FACULTY OF GRADUATE STUDIES Centre for the Study of Curriculum and Instruction Faculty of Education We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA Decemt)er 1991 @ Deborah Binnie Smith 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 The University of British Columbia Vancouver, Canada Date DE-6 (2/88) ABSTRACT This study is concerned with teacher decision-making in the English as a Second Language classroom. Specifically, the study examines the impact of teacher beliefs and perceptions, context factors and second language theory on planning and implementation decisions for the E S L instructional context. Nine E S L teachers in three post-secondary institutions participated in this qualitative study. Data were gathered through classroom observations, post-lesson conferences and interviews. These data were examined in terms of what instructional decisions teachers made and the factors that influenced these decisions from the individual teacher's perspective. Second, the data were analysed for internal consistency between stated beliefs and instructional decisions and external consistency between decisions and second language theory. In examining the role of the teacher in the E S L instructional context, this thesis contributes to both research and teaching theory in English as a Second Language. First, while regular classroom research has indicated that the role of the teacher and the ecology of the classroom are central to understanding the instructional context, E S L classroom studies have primarily focused on the learner, the learning process and language learning outcomes in this context. This thesis addresses this gap in the research by investigating the teacher's role in the E S L instructional setting and the factors that impact on teacher decision-making. Second, ESL classroom researchers have observed that theoretical ideas are implemented in various ways in the formal setting. While researchers have speculated on the reasons for teachers' eclectic use of theory in practice, there has been little exploratory research conducted to investigate this phenomenon. The findings from this present study indicate that teachers' instructional decisions are centrally influenced by both individually held beliefs about second language learning and teaching as well as experiential knowledge of the E S L classroom. These findings not only contribute to our understanding of the E S L instructional context from the teacher's perspective, but are also significant for the development of instructional theory. TABLE OF CONTENTS Abstract i i L i s t of Tables v i L i s t of Figures y i i Acknowledgements v i i i Chapter 1. BACKGROUND AND PURPOSE OF THE STUDY 1 A. Background 1 B. Purpose 10 C. Overview of the Thesis 11 Chapter I I . REVIEW OF THE LITERATURE: CURRICULUM AND INSTRUCTION IN ESL 13 A. An Overview Curriculum and Instruction ...14 B. Curriculum Design 17 C. Instructional Approach 22 Chapter I I I . REVIEW OF RESEARCH ON TEACHING: CLASSROOM PROCESSES 27 A. Teacher Behaviours and Pupil Learning Outcomes 30 1. Regular Classroom Studies 30 2. Second Language Classroom Studies 32 B. Research on Tasks and P a r t i c i p a t i o n Structures 35 1. Regular Classroom Studies 35 The Ecological Perspective 36 The Socio-Linguistic Perspective 38 The Co-operative Learning Models 44 2. Second Language Classroom Studies 48 Chapter V. THE RESEARCH DESIGN 75 A. Data C o l l e c t i o n and Analysis: A General Perspective 75 B. Data C o l l e c t i o n Procedures 81 1. Data Sample 81 2. Description of Sites and Teachers 84 Si t e A: English Bay Community College 84 Teacher 1: C l a i r e 84 Teacher 2: Mark 85 Teacher 3: Stan 85 Si t e B: P a c i f i c Studies Language Centre 85 Teacher 4: Helen 86 Teacher 5: Edie 86 Teacher 6: Fran 87 S i t e C; South Centre Community College 87 Teacher 7: Annie 88 Teacher 8: Sandy 88 Teacher 9: Pam 88 3. Program Selection 89 4. Description of S p e c i f i c Data C o l l e c t i o n Procedures 91 Observations 91 Conferences 92 Task Grid 93 Interviews 95 Documents 97 F i e l d Notes 97 C. Data Analysis Procedures 99 1. F i r s t Stage of Analysis: Lesson Tasks 99 2. Second Stage of Analysis 104 Chapter VI. INDIVIDUAL TEACHER FINDINGS 108 A. Pre-Active Factors 108 1. Teacher Experience and Academic Background 108 2. Context Factors I l l B. Sites and Individual Teachers 115 1. English Bay community College 115 C l a i r e 116 Stan 122 Mark 127 Concluding Comments 154 2. P a c i f i c Studies Language Centre 134 Helen 135 Edie 141 Fran 148 Concluding Comments 154 3. South Centre Community College 155 Annie 156 Sandy 162 Pam 167 Concluding Comments 72 C. Conclusion 174 Chapter VII. COMPARATIVE FINDINGS 180 A. The Central Role of Teacher B e l i e f s 180 1. Theoretical E c l e c t i c i s m 181 2. Consistency i n B e l i e f s & Practices....185 B. The Role of Context Factors 191 1. Administrative Factors 191 2. Collégial Factors 197 3. Student Factors 199 C. Levels and Phases of Decision-Making ....204 1. Curriculum Level 207 2. Lesson Level 211 3. Interactive Decisions 213 D. Simplifying the Decision-Making Process 215 1. Routines and Patterns 215 2. Inter-relationship of Decisions 219 E. Tasks and P a r t i c i p a t i o n Structures 222 1. Task Planning 223 2. Task Organization 224 3. Teacher Conceptions of Task 226 4. Task and P a r t i c i p a t i o n Structure ....230 5. Teacher Roles 236 F. Conclusion ..239 Chapter VIII. SUMMARY, CONCLUSIONS AND IMPLICATIONS 246 A. Summary of the Study 246 B. General Research Questions and Findings 248 C. Conclusions 253 D. Implications 256 REFERENCES 261 APPENDIX 1. Permission Forms 271 APPENDIX 2. Interview Schedule 274 APPENDIX 3. I n s t i t u t i o n a l Documents 283 APPENDIX 4. COLT Observation System 291 APPENDIX 5. Lesson Charts and Teacher Task Grids 293 LIST OF TABLES TABLE 5-1 Data C o l l e c t i o n 106 TABLE 5-2 Relationship Between Data C o l l e c t i o n and Analysis 106 TABLE 6-1 Teacher Ch a r a c t e r i s t i c s 109 TABLE 6-2 Contextual C h a r a c t e r i s t i c s of the Three Sites 112 TABLE 6-3 Lesson Summary: C l a i r e 121 TABLE 6-4 Lesson Summary:Stan 126 TABLE 6-5 Lesson Summary:Mark 132 TABLE 6-6 Lesson Summary;Helen 140 TABLE 6-7 Lesson Summary:Edie 147 TABLE 6-8 Lesson Summary:Fran 153 TABLE 6-9 Lesson Summary:Annie 161 TABLE 6-10 Lesson Summary : Sandy 166 TABLE 6-11 Lesson Summary:Pam 171 TABLE 7-1 Teacher Perceptions of I n s t i t u t i o n a l Goals....195 TABLE 7-2 Teacher Perceptions vs Implementation: P a r t i c i p a t i o n Structures 233 TABLE 7-3 Student-Centered P a r t i c i p a t i o n Structures: Teacher Perceptions vs Implementation 234 LIST OF FIGURES FIGURE 1. Theoretical Perspectives of Planning and Instruction i n ESL 2 FIGURE 2. Framework of Relationships 9 FIGURE 3. Model f o r the Study of Teaching (Dunkin & Biddle 1974) 28 FIGURE 4. Research on Teacher Decision-Making (Shavelson & Stern 1981) 60 FIGURE 5. Task Grid for Post-Observation Conferences 93 FIGURE 6. Task Analysis Chart 1:Interaction Segments — 1 0 0 FIGURE 7. Task Analysis Chart 2:Activity Level 103 FIGURE 8. Comparative Task Analysis 105 FIGURE 9. Teacher Perceptions & B e l i e f s + Task Implementation 105 FIGURE 10. Nunan's Theoretical Planning Framework 206 FIGURE 11. Planning Decisions 207 FIGURE 12. Lesson Scaffolding 217 FIGURE 13. Sequential & Hier a r c h i c a l Decision-Making ....222 FIGURE 14. Teacher Task Grid Analysis :Claire 228 FIGURE 15. Teacher Task Grid Analysis:Pam 229 FIGURE 16. Whole Class P a r t i c i p a t i o n Structure 234 ACKNOWLEDGEMENTS A number of people have provided support and encouragement throughout this research project. I would like to thank the members of my research committee: Dr. Don Fisher, my senior supervisor, for his constructive comments on the direction of the research and his positive words when I needed them most; and my committee members, Dr. Ian Housego, Dr. Bernie Mohan and Dr. Margaret Early for their efforts to improve the substantive content and presentation of the thesis and for their support throughout. Doctoral fellowships from The Centre for the Study of Teacher Education, UBC and the Province of British Columbia are also gratefully acknowledged. In particular, the time spent at the Centre as a doctoral fellow and working with new teachers in the Faculty of Education contributed significantly to my interest in examining teacher decision-making. I am most grateful to the nine teachers and their students who participated in this study. The teachers not only allowed me into their classrooms to observe lessons but also gave me a great deal of their time outside the classroom for conferences and interviews. Their professional attitudes towards their work, their thoughtful examination of planning and teaching decisions and their enthusiasm for teaching ESL added to my respect for the calibre and commitment of teachers in our profession. I would like to thank my colleagues at Douglas College for their unfailing support and for their efforts to make administrative arrangements to allow me to concentrate fully on the research at crucial points. I have appreciated the interest they have shown in the research and their helpful comments on a variety of issues. I am particularly thankful for the support my family has given me throughout this research project. My children, Kavan, Erin, Ceilidh and Liam have supplied me with endless cups of tea and words of encouragement. My husband, Paddy, has read numerous drafts of the thesis and provided invaluable input on the organization and clarity of the content. He now knows more about theory and practice in E S L than a political scientist would ever wish to know. Thankyou. Chapter 1 BACKGROUND AND PURPOSE OF THE STUDY A. Background of the Study This thesis is concerned with teacher decision-making in the English as a Second Language (ESL) classroom and the factors that impact on the types of decisions teachers make in this instructional context. It has been argued that teachers "... personalize and make practical - theoretical ideas" (Connelly & Diennes 1982:197) in order to meet the demands of a practical situation (Schwab 1969). The purpose of this study was to examine the decision-making process from the teacher's perspective and in particular how a group of E S L teachers approach instructional decision-making for the adult classroom and 'make practical' second language theory. More specifically, this declsron-making study focused on the impact of individual teacher beliefs, perceptions of the instructional task, experiential and theoretical knowledge and context factors on teachers' instructional decisions for the adult E S L classroom. In second language (L2) theory, various approaches to planning and teaching are advocated for the E S L teacher. Underlying each of these are theoretical assumptions about the nature of language and the nature of teaching and learning a second language. On a global level, two differing perspectives of language learning have dominated in E S L pedagogy; a product view and a process view. In theoretical models reflecting a product-oriented or "formalist" (Rivers) position, language is primarily viewed as a object to be mastered. Two mainstream approaches reflecting this perspective, the Audio-lingual Method (ALM) and the Cognitive Code Method both focus on learner mastery and accurate use of discrete language items through a transmission model of teaching. On the other hand, underlying the process-oriented models, there is the assumption that language is an activity not a product and emphasis is placed on the use of language for communicative purposes. This perspective underlies both the Communicative Language Teaching Approach (CLT) and more recently, Task-based Language Teaching (TBLT). These process-oriented teaching approaches emphasize student-centered classrooms in which teachers are facilitators rather than directors and activities whk)h focus on communication of meaning rather than accuracy and form. Eskey (1983) has suggested that the E S L field "has always suffered from a pendulum effect " in that there is a "melodramatic swinging back and forth between (these) two extreme positions" (p.315). Eskey's pendulum analogy is questionable given the consistent advocacy by theorists for a more process-oriented, student-centered approach to E S L instruction since the early 1970's. However, Eskey's characterization of the E S L field as one dominated by "extreme positions" is certainly evident when examining both the planning and teaching literature. For example, teachers are presented with the dichotomous planning models of language-based or task-based curricula design, and the division of classroom activities into two categories; form-focused or meaning-focused. In addition, classrooms are described as either student-centered with the teacher in the role of facilitator or teacher-centered with the teacher as the director and information-giver. Each of-these 'either-or' perspectives prescribed in the literature for the E S L context reflect underlying process or product assumptions about language instruction and can be grouped accordingly. Theoretical Perspectives of Planning and Instruction in E S L PRODUCT-ORIENTED PROCESS-ORIENTED planning focus: units of language tasks instructional model: transmission interactive classroom organization: teacher-centered student-centered teacher role: director facilitator instructional objective: mastery of language forms and accuracy of language use ability to use language fluently for communicative purposes lesson activities: drills and exercises designed to practise pre-selected language items small group/pair activities designed to provide opportunities to use language communicatively Figure 1 While in theory, L2 instructional techniques can be identified as either process-oriented or product-oriented, whether this dichotomy also exists in practice has not been a focus in classroom-based research. What researchers have noted is that individual teachers blend theoretical ideas in different ways for classroom implementation. Researchers have also speculated on why this theory-practice gap occurs in the instructional context. Long (1989) for example, has suggested that teachers mix techniques from various instructional models because the demands of the instructional context may necessitate the use of a broader range of strategies than is offered by any single method. Spada (1984), has posited that teacher beliefs play a critical role in determining what occurs in practice. In Spada's conclusions from her 1984 study findings, she suggests that teachers adopt a particular approach to teaching ESL ; ... within the context of their own personal views as to how they think languages are learned best in the classroom. Even when a method is carefully prescribed and contains detailed techniques, lesson plans and materials to be used, teachers will incorporate such methodological procedures in different ways and to different degrees. (Spada 1984:149) While these speculations have suggested the critical role of the teacher and context in determining what occurs in this instructional setting, little research has examined the L2 setting from the teacher's perspective in order to ascertain how factors such as individual beliefs and context impact on their instructional decisions. Most 12 classroom studies have been concerned with the learner and relating classroom processes with language learning outcomes (product-oriented research) or with describing classroom interaction (social and linguistic) through the use of observation schemes (process-oriented research).i However, neither of these research traditions have focused on the teacher's perspective of the instructional context nor the process of decision-making that underlies classroom practices. On the other hand, researchers have been investigating the regular classroom from a teacher decision-making perspective over the past two decades. Central to this research tradition is the view that "any teaching act is the result of a decision ... that the teacher makes after the 1 See Allwright 1983; Gaies 1983; Long 1980,1984; Chaudron 1988 for reviews of classroOTa-based research. complex cognitive processing of available information" (Shavelson 1973). ^ The majority of these decision-making studies have examined this "basic teaching skill" (Shavelson 1973) in terms of teacher planning decisions; a smaller number have looked at decisions made in the process of teaching (viz., Clark & Peterson 1986; Shavelson & Stern 1981; Shavelson & Borko 1983). These decision-making studies have investigated not only how teachers approach the planning process but have also identified what factors have an impact on both pre-active and interactive decision-making. A consistent finding from these studies is that experienced teachers modify planning theory in practice; that is, rather than objectives guiding the planning process (as prescribed by the linear modeP ), teachers focus on activities and content. This findings parallels the theory-practice gap found to exist in classroom practices. More significantly this provides support for the position taken by practical knowledge theorists who argue that teachers modify theory in relation to individually held assumptions about the nature of teaching and learning. Researchers investigating the instructional context from this perspective have been concerned with the impact of teachers' personal theories and experiential knowledge of the classroom on the decision-making process (e.g., Elbaz 1983; Clandinin 1986; Connelly & Clandinin 1986). In the E S L context, a small number of researchers have started to address the decision-making research gap in L2 research by investigating how teachers approach curriculum planning (Cumming 1989, 1990) and the types of curriculum implementation decisions teachers make (Woods 1989). Findings from this research have indicated similarities to research findings on how regular classroom teachers approach decision-making. Most significantly. Woods found that teachers' planning decisions were not consistent with any one theoretical perspective but did reflect individually held beliefs about the nature of teaching and learning a second language. These 12 studies have primarily focused on teachers' thinking outside of the classroom (pre-active and post-active planning 2 in Clark & Peterson 1986:273 3 The linear model of planning (Tyler) involves first determining instructional objectives, then selecting classroom activities in order to achieve these objectives and finally deciding on evaluation procedures for assessing whether the objectives have been met or not. decisions) but not tiow tliese decisions were actually implemented in lessons. ^ In contrast, a major focus of the present study was to observe what teachers did in the classroom and to examine these practices in relation to teachers' stated beliefs about teaching and learning and their reflections on the factors that had influenced these lesson decisions. Such an action-reflection approach not only provides observable data on how theoretical ideas are made practical in the classroom, but also the individual teacher's perspective on the decision-making process that led to those actions. In this way, the present study provides a link between L2 research investigating pre-active decision-making and those studies focusing on classroom processes. The present study adopted task and participation structure as the basic unit of data collection and analysis. Researchers investigating the teaching-learning process in regular classrooms have Identified these as key constructs (e.g., Doyle 1979, 1983; Erickson 1986; Green & Harker 1982) in classroom interaction. Doyle, examining the classroom context from an ecological perspective, has concluded that the academic task structure is the central unit of lesson organization. Further, he has argued that this academic task structure is embedded in an evaluative framework which requires students to 'perform for grades' and adhere to a particular "behaviour ecology" (Doyle 1983:192). Other researchers, adopting a socio-linguistic approach to classroom research, have emphasized the inter-relationship of the academic task structure and social participation structure of lessons. These researchers have argued that in order for students to successfully accomplish lesson tasks, they must not only attend to the academic subject matter presented but also conform to the socio-linguistic norms for behaviour established by the teacher (viz., Mehan 1979; Erickson 1986). Task and social participation structure are also central concepts in process-oriented E S L theory and research (e.g., Littlewood 1981; Gandlin 1987; Long & Porter 1985; Pica & Doughty 1985, 1986; Long 1989; Nunan 1988). In both communicative and task-based theory, methodologists advocate student-centered classrooms in which learners are given opportunities to use the In Woods' study, some classroom observation data were collected; however, the main thrust of this research was directed at examining teachers' pre-active/post-active thinking about their planning/teaching practices. language for communicative purposes through meaning-focused tasks. In classroom-based research, task-based researchers have investigated the impact of both participation structure (teacher-centered and student-centered) and task type on learner language use in the instructional setting. Findings from these studies have provided empirical evidence for the benefits of student-centered tasks in terms of the increased quantity and quality of learner language use (input and output). However, in contrast to research investigating the task and participation structure in regular classrooms, L2 task-based studies have not addressed the impact of social and contextual factors on how tasks are accomplished in the E S L classroom. In addition, while research investigating teacher decision-making in the regular classroom has shown that teacher beliefs and perceptions play a central role in instructional decisions (e.g., Shavelson & Stern 1981; Clark & Peterson 1986), E S L task researchers have not investigated the impact of these teacher factors on task design and implementation. This thesis sought to address this gap in E S L classroom research by exploring this instructional setting from the teacher's perspective, and in particular the role that L2 theory, teacher beliefs and perceptions and contextual factors play in teacher decision-making. The following questions, unanswered by current classroom-based 12 research, served as sensitizing foci for the collection and analysis of the study data. Broad Research Questions: 1. How do ESL teachers approach instructional decision-making in the ESL context? 2. Is teacher decision-making, particularly decisions related to task and participation structure, influenced by individually held beliefs about L2 teaching/learning and perceptions of the instructional context and task? 3. How do context factors (institution, setting and students), impact on ESL teachers' planning and implementation decisions? 4. Are teachers' instructional decisions consistent with theoretical ideas about planning and instruction? Guiding the direction of the study was the following Framework of Relationships (Figure 2) which suggests the relationship between pre-active factors and decision-making in the E S L context. The framework emerged from the literature review of teacher decision-making research and regular/L2 classroom-based research and provided an initial direction to the emergent research design of this decision-making study. First, in terms of pre-active factors, decision-making research has indicated that both teacher characteristics and context factors impact on teacher planning and implementation decisions (viz., Shavelson & Stern 1981; Shavelson & Borko 1983). Teacher characteristics identified as particularly relevant to examine in this present study were individual beliefs about teaching and learning ESL, perceptions of the instructional task and context and L2 theoretical knowledge. It is important to note here that this study was concerned with stated beliefs and evidence of beliefs provided by observed classroom practices. In the case of the latter, researcher hypotheses concerning beliefs were validated with the teacher participant in the post-observation conferences and final interviews. Three groups of context factors which were selected as pertinent for the study included those related to the institution (e.g. administrative expectations), the setting (e.g., classrooms and audio-visual facilities) and student characteristics. The Research Framework (Figure 2) suggests that these pre-active factors impact on teachers' planning and implementation decisions for the E S L context. It should be noted that the examination of context factors involved focusing on two different types of data. First, information was gathered from both institutional documents and the participating teachers about the program and students. Second, teacher statements provided data on individual perceptions of the context and the impact these had on decision-making. In order to examine teachers' decisions, task and participation structure were selected as the central units of analysis. These constructs have been previously used as units of analysis in both regular and L2 classroom-based research. In the regular classroom, researchers have examined how task and participation structure serve to organize the academic and social structure of lessons (viz., Doyle 1983; Erickson 1986). Second, task and participation Structure are not only key concepts in E S L instructional theory but also a focus of considerable research activity in E S L classroom research (e.g., P ica & Doughty 1985a, 1985b; Pica et. al. 1990 ). The conceptualization of the present research and the development of the framework for guiding the study therefore draws from several areas in the research and theoretical literature; findings from decision-making research, classroom-based studies (in both the regular and E S L classroom) and E S L instructional theory. FRAMEWORK OF RELATIONSHIPS PRE-ACTIVE FACTORS Teacher Characteristics -beliefs about teaching & learning -perceptions of the instructional task and context -L2 theoretical knowledge Context Factors -institution -setting -students TEACHER DECISION-MAKING Plannii^ Decisions instructional decisions -cmriculum level -lesson level -task level organizational decisions -participation structures Implementation Decisions lesson tasks & participation structure -language learning focus (accuracy/fluency) -teacher role (director/facilitator) -timeframe (tasks) -% of time for teacher-centered vs student-centered tasks -task type Figure 2 This present study was concerned with the teacher's perspective of the E S L instructional task and context and accordingly it was conducted from a qualitative methodological perspective. In such an approach, a process of inductive reasoning is employed whereby "detailed descriptions of people's perceptions and social realities" are formed from many observations in the setting and then "from these descriptions an understanding or a theory ... (is formulated)... to explain the phenomenon" (McMillan & Schumacher 1974: 92). In an inductive approach, hypotheses are not established prior to entering the field but rather generated on the basis of the data. However, rather than beginning the research tabula rasa, it is useful to focus the study so that the process of data collection and analysis is a deliberative one and that sufficient relevant data is collected while still in the field (Erickson 1986). The four broad research questions stated previously and the Framework of Relationships provided this focus for the present study. B. Purpose of the Study The purpose of this study was to investigate the decision-making process for the adult E S L classroom from the teacher's perspective. More specifically, this study set out to examine the role that L2 theory, teacher beliefs about L2 instruction, perceptions of the instructional task and context played in the pedagogical decisions made by a group of adult-level E S L teachers. The goal of the study was not to test the effectiveness of a particular theory of instruction nor to evaluate the planning and implementation practices of the teachers involved. This was an exploratory study which focused on a relatively recent area of research interest in the adult ESL context -- teacher decision-making. In an exploratory study "phenomena which have not been studied previously" are examined (McMillan & Schumacher 1974:106). Although there have been a few studies on teaching decision-making (Cumming 1990; Woods 1989), the significance of this present research in relation to the ESL studies previously conducted is the extensive collection of classroom implementation data. The previous studies concentrated on teacher thinking about their instructional decisions but included little data on how these decisions were implemented in the classroom. The inclusion of lesson implementation data in the present study extends our understanding of E S L teacher decision-making in several ways. First it provides a comparison between teachers stated beliefs about teaching and learning E S L and how this translates into actions in the classroom. Secondly, teacher reflections on what occurred in the observed lessons allow for an analysis from the teacher's perspective of the impact of immediate contextual factors on decision-making in the process of teaching. Third, by observing both the task structure and related participant structures of actual lessons, we can begin to ground task-based theory in practice. Up to this point, classroom-based research on tasks has been of the experimental type. Researchers have designed the tasks in order to gather data on the impact of manipulating task variables and participation structure on learner language use. However, little attention has been paid to how teachers design and implement tasks and in particular, the factors that impact on these decisions. There were several stages in this study of E S L teacher decision-making: 1. Lesson observation data were collected in order to examine what teachers did in practice. 2. Post-observation conferences following each observed lesson and final interviews provided data on why individual teachers implemented particular techniques and strategies in their lessons. These data included teachers' reflections on their decisions (pre-active and inter-active) for the lessons observed, and planning/teaching in general, and how these related to individual beliefs about teaching/learning E S L and perceptions of the instructional task and context. 3. These lesson, conference and interview data were then examined for congruency between stated teacher beliefs and classroom actions and for consistencies/inconsistencies with theoretical perspectives on teaching and learning E S L C. Overview of the Thesis Chapter 1 has provided the baol<ground to the study and the purpose of the research. Chapter II contains a review of the literature on second language curriculum design and methodology. This chapter provides an overview of the mainstream planning and instructional approaches prescribed in the E S L literature and the theoretical assumptions concerning E S L teaching and learning underlying these. Since a major objective of this study was to examine how teachers 'make practical' theoretical ideas, it is critical to first explore the dominant theoretical perspectives and instructional methods based on these, presented in the E S L literature. A second essential area of the literature review conducted for this study focused on relevant research from both regular and 12 classroom studies: Chapter III examines classroom-based studies with particular attention paid to research investigating tasks and participation structures; Chapter IV is concerned with L1 and 12 decision-making research. This chapter reviews the findings from studies examining how teachers approach the decision-making process at the curriculum and lesson levels, and the factors that have been found to impact on teachers' instructional decisions. Chapter V discusses the research design of this qualitative study and the data collection/analysis procedures employed. The study findings are reported in Chapter VI (Individual Teacher Findings) and Chapter VII (Comparative Findings). Chapter VII first provides the summary and conclusions of the study and then discusses the implications of the findings for theory, practice and further research. CHAPTER \\ REVIEW OF THE LITERATURE: CURRICULUM AND INSTRUCTION IN TEACHING ENGLISH AS A SECOND LANGUAGE Practical Knowledge researchers argue that teachers modify theoretical ideas in order to meet the needs of the classroom situation (e.g. Connelly & Diennes 1982). This modification has been observed in ESL classrooms (e.g., Long & Sato 1983; Spada 1984; Long 1989), but little research has been conducted on how individual teachers view L2 theory, how they might adapt theoretical ideas for the classroom or what factors might influence these decisions. A particular focus of this present study of E S L teacher decision-making, was to examine how these teacher-based issues related to the implementation of theory into practice. In order to explore the role of theory in teacher decision-making, an essential first step is to review the mainstream approaches to L2 curriculum design and classroom instruction presented in the L2 literature. In Chapter I of this thesis, it was pointed out that underlying these theoretical approaches are either process-oriented or product-oriented assumptions about second language learning and teaching. Each of these theoretical perspectives involves differing views concerning the goals of second language instruction and related activities for achieving these, the organization of curriculum content and the classroom, and the role of the teacher in the classroom (see Table 1-1). While Eskey (1983) has suggested that E S L instructional theory constantly swings back and forth between these two extremes of process and product assumptions, a review of the literature shows that this pendulum analogy of the E S L field is a questionable one. Since the early 1970's, methodologists have been assigning an increasingly important role to the classroom use of student-centered tasks which focus on language for communicative purposes. In other words, there has been a consistent trend in the E S L literature towards teaching language as a process of communication not as a product to be dissected and memorized. Although there has been this consistency in theory, the question for the present study concerns the extent to which teachers have wholly adopted this perspective in practice. Researcher observations in the second language classroom suggest that they have not since it has been noted that teacher practices reflect a 'blend' of theoretical ideas (e.g. Spada 1984; Long 1989). This thesis set out to investigate further this theory-practice gap observed by previous researchers but from the teacher's perspective. More specifically, this study was guided by the following broad questions concerning the relationship between teacher beliefs and perceptions, 12 theory and pedagogical decisions: 1. To what extent are individual teachers' beliefs and personal theories about second language teaching and learning consistent/inconsistent with current L2 theory ; 2. How do teachers beliefs impact on their instructional decision-making particularly in relation to: i) their instructional goals (e,g, accuracy/fluency development); ii) the organization of the classroom for activities (student-centered/teacher-centered participation structures); iii) their role in the classroom (director/facilitator). This review of the literature first discusses the theoretical assumptions underlying both the product and process approaches to second language teaching and why there was a redirection in E S L pedagogy in the 1970's. This general overview is followed by a more detailed discussion of approaches to curriculum design and classroom methodology presented in the L2 literature. A. An Overview of E S L Curriculum and Instruction Before 1970, curriculum and instruction in E S L was primarily influenced by structural linguistics and behavioural models of teaching and learning. Accordingly, curriculum content consisted of grammatical structures hierarchically sequenced from 'simple' to 'complex', and instructional techniques led learners through carefully sequenced steps of teacher-directed presentation, drill and practise. The goal of the language class was for students to be able to "manipulate tlie structures of tiie language easily and without error" (Johnson 1981:2). However, in practice, it was found that although second language learners could mimic the isolated elements of language in the classroom, they had difficulty in synthesizing these memorized language items for 'real-life' communication outside of the formal setting; that is, there was an "apparent inability ...to transfer knowledge of the target language to situations outside of language classrooms" (Long 1976:137). This phenomenon of structurally competent but communicatively incompetent second language learners emerging from the E S L classroom was a situation that L2 methodologists had to address if learners' language learning needs were to be appropriately met (Johnson 1981:1). The work of linguists Hymes and Halliday and research in the field of socio-linguistics provided a basis for a redirection in E S L curriculum design and classroom pedagogy. Hymes (1972) argued that to be communicatively competent required not only knowledge of the rules of grammar (grammatical competence) but also an understanding of the socio-cultural appropriateness of utterances (socio-Iinguistic competence). In relating the concept of communicative competence to the 12 context. Stern argued that the learner needs to gain the intuitive mastery the native speaker possesses to use and interpret language appropriately in the process of interaction and in relation to a social context. (Stem 1983:229) The concept of communicative competence was further elaborated by Canale & Swain (1980) who argued that the component of 'strategic competence' should be included in such a model: no communicative competence theorists have emphasized communication strategies used to deal with communication breakdowns (such as) false starts, hesitations, how to avoid grammatical forms not mastered fully. (Canale & Swain 1980:25) In addition to these three components, theorists included a fourth --discourse competence. This relates to the need for learners to be able to connect "a series of utterances into a meaningful whole" (Savignon 1983:38). In other words, communicative competence also involves the ability to deal with 'chunks' of language input and to participate actively in extended discourse situations. Savignon (1983) suggested that in learning a second language, these four communicative competence components, -- grammatical, socio-linguistic, strategic and discourse, develop not in sequence, "as one strings pearls in a necklace" (Savignon 1983:45) but as an interactive whole. While at the lower levels of proficiency, when learners have limited knowledge of and competence in the use of the language, there is more dependence on strategic and socio-linguistic strategies for communication. However, as language proficiency increases, linguistic and discourse competencies play the more prominent roles (Savignon 1983). The L2 instructional approach which emerged from this theoretical model of communicative competence and the perspective of language as a communicative process, was the Communicative Language Teaching (CLT) approach. Savignon (1983) proposed the following as the guiding tenets of this second language instructional approach: 1. Language use is creative. Learners use whatever knowledge they have of a language system to express their meaning in a variety of ways. 2. Language use consists of many abilities in a broad communicative framework. The nature of the particular abilities needed is dependent on the roles of the participants, the situation and the goal of the interaction. 3. L2 learning, like L I learning, begins with the needs and interests of the learner. 4. A n analysis of learner needs and interests provides the most effective basis for materials development. 5. The basic unit of practice should always be a text or chunk of discourse. Production should begin with the conveyance of meaning. Formal accuracy in the beginning stages should neither be required or expected. 6. The teacher assumes a variety of roles to permit learner participation in a wide range of communicative situations. (Savignon 1983:24) This process perspective of the nature of language learning and teaching, stimulated a fundamental shift in E S L instructional theory. The product view that the most effective way to teach a second language was to ensure learner mastery of discrete language items was replaced by the perspective that the classroom should provide opportunities for the learner to develop skills in using language as a tool of communication. In the literature, teachers were presented with a new approach to curriculum design and specific techniques and strategies for classroom instruction based on this process perspective of second language teaching and learning. This review now turns to a more detailed examination of each of these components of E S L instructional theory. B. Curriculum Design In planning L2 curricula, the approach taken reflects underlying assumptions about language and language teaching/learning -Whatever course designers or teachers think language learning involves will influence the elements of the target language they present to their students - words, structures, notions, etc. - and how they should be presented - in isolation, in context, accompanied by rules, as naturally occurring parts of authentic excerpts, etc. (Long & Crookes 1989:2) In traditional structure-based curriculum design, structures were pre-selected and sequenced from simple to complex according to linguistics analysis. On the other hand, in the communicative model, the starting point is the learner and his/her communicative needs. Taylor points out: the selection of appropriate linguistic information to be taught is chosen on the basis of its contribution to the acquisition of skills or performance of specific tasks which are communicatively relevant to students' individual needs. (Taylor 1983:72) In developing curricula for this instructional model, teachers (planners) conduct a needs analysis of their learners and from this, select the appropriate communicative language functions to include in the course. In determining the functional content of a course, the teacher (or planner) can consult the list of language functions identified as appropriate for the adult learner by the Council of Europe (1977). This list provides the teacher with an extensive inventory of communicative functions such as expressing likes and dislikes, asking for information, and accepting/refusing invitations. Socio-cultural contexts, relevant to specific learner needs (e.g., talking to a teacher on parents night), can then be selected for embedding the language in meaningful situations. In addressing the language needs of students in the academic setting, research investigating language used across the curriculum areas identified discourse functions particular to that context. These included functions such as sequencing, classifying, comparing, hypothesizing and evaluating (Mohan 1986). In planning courses designed to prepare students for the academic setting (EAP courses), these discourse functions can be used as the course core. While in both the communicative and academic functional designs, language functions constitute the central core of the curriculum plan, grammatical structures are still included. However, these are not the primary organizational focus of planning as in the more traditional curriculum design approach. Only those grammatical items needed for the selected functions are included. In addition, these structures are not organized sequentially to be learned one at a time but are included in spiral fashion around the functional core of the curriculum (Widdowson & Brumfit 1981:209). As a result learners are exposed to the pre-selected structural items a number of times but in relation to different functions throughout the course. Planning curricula with a functional core was considered more congruent with the process view of second language^ teaching and more conducive to ensuring that the learner's communicative needs would be met (e.g., Widdowson & Brumfit 1981). At the same time, both grammar-based and function-based curricula ('Synthetic, Type A' designs have been criticised as language-centered rather than learner-centered models since course content (structures or functions) is still primarily selected and sequenced according to linguistic criteria not the learners (Stern 1981; Johnson 1982; Richards 1984; Long & Crookes 1989). On the other hand, in a learner-centered model or Type B design,^ course content evolves in response to learner needs and input as the course proceeds. This model is reflected in alternative L2 instructional methods such as the Silent Way (Gattegno 1972, 1976), Community Counselling Learning (Curran 1976) and Suggestopaedia (Lozanov 1979). These methods, while differing in specific instructional strategies employed, stress the central and ongoing role that the learner plays in determining lesson content. More specific to curriculum design, the Negotiated syllabus has been proposed as one which meets a learner-based, process orientation to language (Clarke 1991). A negotiated syllabus does not exist prior to a particular course, but is 'negotiated' during the course between the learners and the teacher. Clarke states: learners... involve themselves in fundamental decisions concerning content, materials, methodology, testing and evaluation (both of their own performance and of the evolving syllabus) so that design becomes a process concept, a dynamic and flexible dimension of the learning experience. (Clarke 1991:14) ^ Long & Crookes (1989) however, view both structural and functional curricum design as product-oriented and argue that to reflect the process orientation of instruction, 'task' should be the basic unit of planning. * A synthcdc curriculum design organizes language content into discrete items (structures, functions) to be presented one at a time; the learner is expected to synthesize these language items for 'real life' use. A Type A syllabus design (White 1988), involves the pre-selection fo language to be taught; that is course content is determined prior to meeting the specific learners (Long & Crookes 1989:4-6). A Type B syllabus (White 1988) is a 'negotiated' one in that selection of course content is determined by both students and the teacher during the course (Long & Crookes 1989). Clarke (1991) points out that a "Negotiated syllabus could end up being any one of the various syllabus types" (P. 15) --structural, functional, task - but this will depend on "early negotiations" between the learners and the teacher rather than determined before the course begins. The emphasis here is therefore on learner involvement in curriculum decisions, not assumptions about the nature of language and language teaching/learning (process or product) which underlie the functional and structural design approaches. Task theorists, (Long & Crookes 1989; Nunan 1988) regard both structural and functional designs as product-based and question the validity of such 'synthetic' approaches which organize language content into discrete elements. They argue that these designs are not only inconsistent with the current process orientation of classroom instruction but also research findings which indicate that second languages are not acquired in sequential units or in isolation from actual language use (e.g., Long & Crookes 1989). By adopting 'task' as the basic unit of planning, these theorists contend that the necessary link between curriculum and instruction is established. Nunan (1989) states: ... the distinction between syllabus design and methodology becomes difficult to sustain: one needs not only to specify both the content (or ends of learning) and the tasks (or means to learning) but also integrate them. (Nunan 1988:15) Three task-based syllabus types (considered analytic syllabus design « ) have been developed: procedural (Prahbu 1987), process (e.g., Breen & Candlin 1980; Candlin & Murphy 1987), and task (e.g.. Long 1989; Long & Crookes 1989). While these three syllabus types adopt 'task' as the basic unit of planning, they differ in how content is determined and in certain implementation features. ^ An analytic syllabus design contrasts with a synthetic type in that language is presented in "whole chunks" as opposed to discrete items and the learner is expected to "induce the rules and perceive the irregularities" in the language system (Long & Crookes 1989:4). First, while in both procedural and task syllabus types tasks are pre-selected, in the latter, these classroom (or 'pedagogical') tasks are determined by a needs analysis of 'real life' task needs. In contrast to the pre-course selection of tasks, the content of the process syllabus is 'negotiated' between teacher and students during the course itself. A second fundamental difference lies in classroom implementation. In the procedural approach, teacher-centered format is emphasized and small group work avoided on the basis that student-interaction results in "degenerate input" and " promote(s) fossilization" of errors (Long & Crookes 1989:34). In addition, the pedagogical focus is placed on task completion not on the language used in the task process. In the process design, with its negotiated approach to course development, learners are provided with a number of options concerning what they want to learn (from a range of problem-solving tasks provided), how they want to learn it (in small groups, in pairs etc.) and when. In this approach, the emphasis is placed on the process of learning not language or language learning processes (Long & Crookes 1989:38). In the classroom implementation of the task-based design, emphasis is placed on student-interaction in small groups/pairs and the language used, particularly for negotiation of meaning, in the process of completing tasks. Long & Crookes (1989) stress that the task-based approach they advocate is based on psycholinguistic rationale and classroom research of the relationship of task design and language learning processes. The findings from this research (discussed in detail in Chapter III) , have significantly contributed to our understanding of how task design variables (e.g., task types, participant roles, expected outcomes) have an impact on student language use and particularly on learner use of negotiation of meaning strategies (viz., Long 1989;Pica et al. 1990). In summary, various approaches to curriculum design have been developed with each reflecting different theoretical views about the nature of second language learning and teaching. In this present study on decision-making, a focus of particular interest was on how individual practitioners approach curriculum and lesson planning, and the extent to which this is consistent with one particular theoretical perspective or reflects the "blending" approach found to exist in classroom practices. C. Instructional Approach In Communicative Language Teaching theory, dramatic changes were advocated for instructional objectives {communicative competence goals rather than mastery of discrete linguistic items), lesson activities (interactive tasks not teacher-centered drills) and roles for the learner (active participant) and the teacher (facilitator and native speaker resource). Communicative methodologist, Keith Morrow (1981), also argued that classroom practices could not be just a "collection of techniques and activities" but part of a "consistent methodology" based on an underlying set of principles in the light of which specific procedures or activities or techniques can be evaluated, related and applied. (Morrow 1981:59) Morrow (1981) suggested five central pedagogical principles of a communicative teaching model: 1) Classroom procedures and techniques must be selected according to the overall objective of promoting communicative competence. 2) There needs to be an emphasis on helping students to deal with whole chunks of language if they are to respond to the "instantaneous" aspect of real communication. 3) Three fundamental components of 'real' communication must be addressed in the language classroom activities; information gap, choice and feedback. ^ ^ Morrow argued that these three components of 'real interaction' must be built into L2 classroom activities so that learners can cope with these in communicative situations outside the classroom. i) an information gap occurs when one participant holds information that the other does not; the participants try to 'bridge' this gap through interaction. ii) choice occurs in 'real interaction' since the speaker has options in both the topic discussed and the language used. This sets up an unpredictability factor for the listener because s/he doesn't know what 4) Students need to be actively involved in using language to communicate meaningfully; otherwise they are not provided with adequate practise opportunities for developing communication skills. 5) A communicative perspective requires a flexible approach to learner errors. Only grammatical and phonological errors which impede meaning should be addressed. The main emphasis is on how successfully learners can communicate meaning; if this goal is achieved, structural errors should be ignored. (Morrow 1981:60-65) While Morrow's fifth principle suggests that learning language structures and developing accuracy of language use are not considered to be very important in communicative teaching theory, this is not the case. i ° However, there are two differing perspectives on this issue presented in the literature: the separationist position and the unificationist position (Johnson 1982), The separationist view is that structures and functions should be presented prior to communicative practice; this position is reflected in Rivers' (1983) "skill-getting to skill-using' continuum. Alternatively, the unificationist view is that structural/ functional items should be provided to the learner only if the need for this becomes evident in the process of communicating (e.g. Brumfit 1980; Savignon 1983). In order to implement the communicative perspective of language instruction, the organization of the classroom (participant structure) and student/teacher roles were considered fundamental concerns. In contrast to the dominant role of the teacher in the "lock-step" (Long 1976) instructional format of traditional methods (e.g., audio-lingual), there was a shift in emphasis to the use of small group, student-centered activities which focused on communciative language use. Methodologists argued that in the less inhibiting situation of working in peer groups, learners are more likely to experiment and take risks with the language as they collaborate on a given task (e.g. Long 1977; Cummins 1984). Bejaranov (1987) for example, concluded from his study that wil l be said by the speaker. iii) feedback indicates whether the listener has understood the speaker's intended meaning. When there is non-comprehension, die speaker can use clarification, rephrasing or reptition of the previous utterance in order to promote listener comprehension. '•^  In the communicative competence model, linguistic competence is one of the four components (Canale & Swain 1980;Savignon 1983). "in groups, students experience less inhibition and tension since discourse serve(s) communicative needs rather than demands for public recitation" (p. 495). In contrasting the type of discourse that takes place in whole class format with small group organization, Long noted: (teacher-led instruction) tends to be if not a drill per se, then 'drill-like' with a pedagogic focus on formal correctness; i.e. on how something is said rather than what is said. (However), working face-to-face with ... peers, relieved of the need for grammatical accuracy in everything they say, students are more likely to experiment (with the language). (Long 1977:287) In considering how language classrooms are organized for achieving different purposes, Kramsch (1987) suggests a useful continuum for considering the relationship between student/teacher roles and language learning tasks. At one end of the continuum is instructional discourse. In this interaction type, student/teacher roles are fixed "with expected and predictable behaviour patterns acquired through years of schooling " (p. 18). Tasks in instructional discourse are concerned with conveying information with an emphasis on the accurate use of language and language skill acquisition. At the other extreme, natural discourse, roles and tasks are more varied and open to negotiation. Tasks focus on the exchange of information and negotiation of meaning with others. Here the emphasis is directed to the process of interaction and the 'negotiation of meaning' (Long 1977) required to complete the task successfully. These tasks emphasize fluency of language use and the development of interactive skills (Kramsch 1987:19). Kramsch states that "neither extreme ever exists in its pure form exclusively" (p. 18) and proposes that there is a middle position on the continuum which she labels "convivial discourse". However, although she suggests that this involves the negotiation of social roles between the teacher and learners to accomplish language tasks, she does not elaborate on the organizational format (whole class vs small groups) nor the language learning focus (accuracy or fluency). It Is significant that Kramsch suggests that a middle position exists in second language classrooms since most L2 literature tends to imply that only two pedagogical formats occur: 1) whole class instruction with the teacher in control and the language learning focus on structure and accuracy or 2) small group/pair activities with the students in control and the language learning focus on communication of meaning. Kramsch makes another interesting point about teacher use of student-centered/whole class format when she suggests that teachers perhaps avoid small group work for certain activities. Many teachers are still reluctant to have students do in small groups what they feel they can do more conveniently, quickly, and accurately in a teacher-controlled, whole-class situation. (Kramsch 1987:24) This raises a question about the extent to which teachers actually use small group or pair organization and for what purposes in a given lesson. In addition, even if teachers adopt a predominantly student-centered instructional approach, do they adopt a directive role when they want to "more conveniently, quickly and accurately" (Kramsch) deal with certain structural points? This would be inconsistent with the theoretical position that teachers should assume a facilitative/resource role (vs the directive role associated with whole class format) in student-centered activities. Rivers, for example, states: real interaction in the classroom requires the teacher to step out of the limelight to cede a full role to student in developing and carrying through activities, to accept all kinds of opinions and be tolerant of errors the student makes while attempting to communicate. (Rivers 1987:9) In addition to specifying a student-centered participation structure and a facilitative teacher role, methodologists argued that the task must 'compel' students to interact with each other (e.g., Rivers 1987). The topic needs to be relevant to student needs and interests, and focused on communication of meaning, not on correct form. Rivers (1987) states: Students achieve facility in using a language when their attention is focused on conveying and receiving authentic messages, (that is, messages that contain information of interest to speaker and listener in a situation of importance to them both), (p.4) From the perspective of the communicative methodologist therefore, student-centered activities are a central part of instruction in the E S L classroom. Further, these activities should be 1) focused on communication of meaning not accuracy of language use; and 2) concerned with topics that are relevant to students' needs and interests. In addition, teachers should assume the role of facilitator and native speaker resource during these activities. In conclusion, the socio-linguistic view of language as a communicative process has had a powerful impact on both E S L curriculum design and classroom practices. A fundamental change has been the redirection in L2 instructional theory from a concern for learner mastery of discrete grammatical items to a focus on the development of communicative competence. This has stimulated new approaches to both curriculum design and instructional techniques being prescribed for the practitioner in which the emphasis is placed on the development of communicative skills within an interactive instructional model. Key concepts in this process approach to second language teaching are lesson tasks which focus on communication of meaning and the predominant use of a student-centered participation structure. Task and participation structure have also been Identified as central units of analysis in regular and L2 classroom-based research. The thesis now turns to a review of this research in Chapter 111. CHAPTER III REVIEW OF RESEARCH ON TEACHING: CLASSROOM PROCESSES In designing research on teaching, the researcher begins with certain assumptions about the nature of the phenomenon of interest (teaching) and the influence of that phenomenon on another (learning). These assumptions, taken together, constitute a conceptual framework or a paradigm or model. The researcher's choice of such a ... paradigm is extremely important because it influences and constrains the subsequent decisions about what is to be studied ... and how it is to be studied.... (Ryan 1984:1) This chapter reviews the research in both regular and L2 instructional contexts which have informed the present study. Although there are obvious differences between these two teaching contexts, the research paradigms guiding the various studies reviewed here, the purposes for which these studies were undertaken and the methodologies used have a great deal in common. In addition, both fields of inquiry have developed and broadened along similar paths moving from an achievement-based research perspective, to a more contextually-based focus on the teaching-learning process itself. For example, while regular classroom researchers were concerned with identifying generic teacher behaviours which correlated with student achievement test gain, second language researchers, through comparative studies, sought to identify the most effective methodological approach for language learning. Secondly, while regular classroom researchers, critical of the achievement-based approach, moved into a more contextually-based paradigm, second language researchers also became more concerned with the interplay of the many variables (teacher/student/context) in the instructional context. Other similarities between the two fields of research include the central role given to classroom observation for examining specific teacher and student behaviours in lessons, and the use of classification systems to study classroom discourse patterns. A particularly useful framework for examining classroom-based research in these two educational contexts is Dunkin and Biddle's (1974) model for the research on teaching. This conceptual framework identified four main categories of variables which were important to consider when examining the teaching process: i) presage variables: teacher beliefs, experiences, training and personal qualities; ii) context variables: characteristics of the community, school, classroom setting and the pupils; iii) process variables: the observable behaviours of teachers and pupils in the classroom; and iv) product variables: the immediate and long-term effects of teaching on pupils. Dunkin & Biddle's model inter-related these variables in the following way: Presage Variables teacher characteristics in the classroom Context Variables characteristics of community, school, classroom & pupils P r o c e s s Variables observable teacher <-> student behaviours Product Var iables learning outcomes Model for the Study of Teaching Dunkin and Biddle 1974 Figure 3. The present study, and the review of the research literature (Chapters III and IV), is concerned with three of these categories of variables: context, presage and process. This chapter reviews regular and L2 classroom-based research; Chapter IV examines teacher decision-making research. The first part of this chapter. Teacher Behaviours and Pupil Learning Outcomes, is an overview of process-product research which has been conducted in regular and second language classrooms. These studies have been primarily concerned with identifying particular teaching behaviours (regular classroom studies) or language teaching methodologies (L2 studies) that correlate with positive student learning outcomes (as measured by proficiency tests). A major objective of this research has been to find the most effective teaching behaviours which could then be incorporated into teacher preparation programmes and/or evaluation programmes. This research approach has been criticised for neglecting to account for the impact of context variables and teacher characteristics (beliefs, training and experience) on the teaching process. Given that the present study seeks to address this research gap in L2 research in particular, product-based studies and more significantly, the criticisms raised about this research paradigm, have provided considerable input into the conceptual framework of this thesis. The second part of the chapter reviews regular and 12 classroom-based research investigating task and participation structure. In the regular classroom context, researchers adopting a process-oriented research model, have examined these constructs from an ecological perspective (Doyle 1977,1983,1986) and a socio-Iinguistic perspective (Erickson 1981,1982,1986; Green 1981,1982,1983). Task and participation structure have also been a focus of considerable research in the E S L classroom. However, in contrast to the more contextually-focused research in the regular classroom, these studies have been conducted within an experimental research framework which has controlled for contextual and teacher factors. Tasks have been researcher-designed and implemented as isolated entities in E S L classes with the primary objective being to correlate variables such as task type and participation structure with language use outcomes. This task-based research has therefore paid little attention to the role that the individual teacher or context factors play in the design and implementation of tasks in the 'real' classroom setting. The research for this thesis has attempted to address this gap in l_2 research and in particular examine how individual teacher beliefs, theoretical knowledge and context factors impact on teacher decisions about tasks and related participation structures for the E S L classroom. A. TEACHER BEHAVIOURS AND PUPIL LEARNING OUTCOMES 1. Regular Classroom Studies Prior to 1970, teaching research was concerned with identifying teacher personality characteristics which correlated with effective teaching practices (Ryan 1984:2). In order to identify these characteristics, researchers asked students and administrators to list and rank order the characteristics of those teachers they believed to be most effective in the classroom. However, since these descriptions were not grounded in classroom observations, no correlation between descriptions of teachers and classroom teaching behaviours or learning outcomes was attempted. Thus there was no empirical evidence gathered to show that -teachers possessing these characteristics were actually more effective in helping students achieve any of the educational goals of education than teachers who lacked them (Medley 1977:13). Dunkin & Biddle's model for research on teaching (1974) provided an impetus for research linking process and product variables. This process-product paradigm stimulated "the most vigorous and productive ... programs of research on teaching for a decade" (Shulman 1986:9). Guiding this research model was the belief that by isolating those teaching behaviours which correlated with end-of-year student achievement test gain, effective teaching practices could be identified and subsequently prescribed in both in-service and pre-service teacher education programmes (Shulman 1986:10). Process-product study data were gathered primarily in elementary classrooms and correlated with reading and mathematics achievement test scores. Observation schemes with pre-determined teacher behaviour categories (e.g., question types) were used to gather empirical evidence of the frequency with which teachers used these behaviours. These frequency counts were tallied and then correlated with student scores on achievement tests. In the majority of these studies, the focus was on identifying those effective teacher behaviours related to classroom management (e.g., allocating turn-taking) and/or instructional behaviours (e.g., feedback strategies). The overall objective was to find stable, effective teaching behaviours which could be generalized across all settings, teachers and students. Clusters of teaching behaviours, labelled 'direct instruction strategies' (Rosenshine 1979) which were found to correlate consistently with achievement test gain included the following: task-oriented teachers, large-group instruction, teacher-organized and directed activities, low level/factual questions, business-like pace and minimum student choice in the selection of materials and activities (Rosenshine 1979). Various criticisms of this research were raised. First, the paradigm implicitly assumed that the frequency of certain teaching behaviours was the most important factor for student achievement. This perspective thus ignored contextual factors (e.g., student characteristics, materials and classroom variables). Doyle (1979) argued that rather than 'stimulus behaviours', researchers should be studying the behaviours which respond to momentary/spontaneous classroom events. Similarly, Ryan (1984) posited that identifying teacher behaviours in isolation from context was of questionable use. She argued that the process-product studies were guided by "laundry lists" of variables rather than a conceptual framework concerning how these fit into the larger context of the teaching-learning process. It has also been pointed out that teacher perceptions of student needs and abilities influence teacher practices. For example. Green ( 1983:186) states: Teachers may choose different strategies for higher, middle and low students due to differing theories about what each group of students needs. These approaches are chosen on the basis of the teacher's perceptions of the students and the situations and performance expectations for those particular students. Other criticisms raised about the process-product studies included: 1) its conception of teaching was too simplistic and technological (Zumwalt 1982); 2) it neglected to account for the role that the student plays in the learning environment and on outcomes (Doyle 1979); 3) it promoted only narrow education goals as defined by achievement tests (Peterson 1979); and 4) it ignored wtiy teachers used certain behaviours over others in particular situations (Ryan 1984). Researchers critical of the process-product model began exploring the classroom context on a more explanatory level. One research approach was to try and identify variables which were thought to 'mediate' between teaching behaviours and student performance. However, although these process-mediating-product studies used a wider lens to capture a more contextualized view by including variables such as materials, time allocation and student characteristics, they still continued to rely on achievement test gain as the "ultimate criterion for effectiveness" (Shulman 1986:13). 2. Second Language Classroom Studies Similar to the research focus in the regular classroom, second language classroom researchers attempted to correlate certain classroom processes with gains in student language development. However, as opposed to investigating "laundry lists" of single teaching variables, these researchers were concerned with identifying the 'best' method for teaching a second language; that is, they were interested in a 'package' of teacher behaviours prescribed by the specific methods under investigation. Large-scale studies (e.g. Scherer-Wertheimer (1964); Smith 1969)»' investigated the effectiveness of such L2 methods as the Audio-lingual (a behavioural-based approach) and Grammar-translation (a cognitively-based approach). However, results from these comparative studies were inconclusive as far as identifying the 'one best method'. Two primary reasons have been suggested for this. First, no systematic observational data ' ' For further discussion of these large-scale method studies, see Chaudron, C. 1988, Second Language Classrooms: Research on teaching and learning (Cambridge: Cambridge University Press). were gathered in and across the classes studied; second, researchers assumed that the participating teachers were faithfully adhering to the strategies prescribed by the methods they were supposedly using (Allwright 1989; Spada 1984). As a result, "... even though groups of learners were found to perform differently on certain measures, there is no way of knowing to what to attribute this differential performance" (Spada 1984:8). Other studies investigated the comparative effectiveness of specific techniques such as teaching grammar explicitly and implicitly (Carlsson 1969) and strategies teachers used for implementing language drills (Politzer 1970). However, the results of these studies were also inconclusive. These disappointing findings from comparative studies led researchers to doubt the ability of the process-product model to provide an understanding of the variety of factors that influence the second language teaching setting. In addition, criticism of adopting a comparative research approach was directed at the questionable underlying premise that there indeed was a 'one best' method or set of techniques. Based on examination of the observation data, it was evident that actual classroom practices implemented across the different methods did not differ a great deal (Long 1989). This suggested that despite the fact that different methods may exist in theory, there is a "blurring of distinctions" between theory and classroom practice (Long 1989). Long observed that this may be due to the demands of the instructional context which necessitate implementing a wider range of strategies than any single method encompasses (Long 1989). A different perspective was offered by Spada (1984) who speculated that differing teacher beliefs played a critical role in how the teachers in her study implemented the Communicative Language Teaching approach; "teachers will tend to implement a particular approach to L2 instruction within the context of their own personal views as to how they think languages are learned best in the classroom (Spada 1984:149). Given the inconclusive results of these comparative method studies and the realization that such a global approach was unable to capture the complexity of the instructional context: The time was ripe for an alternative approach ... that would no longer see the language teaching world in terms of major rival 'methods' and one that would be more respectful of the complexities of the language teacher's task. (Allwright 1989:10) In conclusion, criticisms of the process-product model in both regular and second language classroom research emphasized two major concerns which significantly informed the present research. First, the research goal of identifying teaching behaviours which could be prescribed for âJi teaching situations meant that specific contextual factors which impacted on the teaching process were not taken into account. This raised a number of questions concerning the influence of context factors on decision-making. A second set of questions emerged from the theory-practice discrepancies that researchers found in classroom studies. Despite researcher observations that such discrepancies existed, why teachers used certain strategies over others and particularly, the factors which might account for this theory-practice gap, were not explored. Relevant questions emerging from the literature review, therefore, related first to the role of context, and second to the impact of teacher beliefs and L2 theory on teacher decision-making. On context factors 1. What institutional, setting and student characteristic factors do teachers consider as enhancing and/or limiting their planning and teaching decisions? 2. How do context factors differ from one institutional setting to the next? 3. What impact do contextual factors have on teacher decision-making? On L2 theonz/teacher beliefs 1. How do individual teacher beliefs about the nature of language learning and teaching impact on planning and teaching decisions? 2. To what extent are teachers' beliefs about L2 teaching/learning and related planning and teaching practices consistent/inconsistent with L2 theory. The following section examines studies employing a process-process research model and specifically those concerned with the task and participation structures of classrooms. B. RESEARCH ON TASKS AND PARTICIPATION STRUCTURES 1. Regular Classroom Research In regular classroom research, two research perspectives -- ecological and socio-Iinguistic - offer alternative, more contextuatly focused models to the process-product approach. These perspectives regard the classroom as a "communicative context" (Green 1983) in which both the teacher and students work together to achieve educational goals. Guiding this more descriptive research focus are several distinctive features. First, rather than using observation scales with pre-determined variables to observe, researchers depend on "richly detailed, finely grained descriptions formulated on the basis of extensive, unstructured experience in classrooms" (Doyle 1977:188). Therefore descriptive categories are determined as a result of observation; they do not direct the observation. Second, analysis is based on the view that the actions of teachers and students are powerfully influenced by the demands of the particular setting. Thus contextual factors play a major role in analyzing what occurs in the classrooms observed. This feature is particularly significant when comparing this 'ecological perspective with the 'experimental approach' to task research taken in L2 research. In the latter studies, instructional setting factors and more importantly, the impact these might have on task implementation, are not taken into account. The third feature of the ecological/socio-linguistic studies, and one which has been significant in developing the conceptual framework for the present research, is its focus on why certain classroom behaviours occur rather than just what occurs. The following section first reviews Doyle's research on the ecology of the classroom and his concept of lesson task structures: this is followed by a discussion of the socio-Iinguistic perspective of task and participation structure in classrooms. The ecological perspective on academic task structure. Doyle's research has essentially focused on how the various actions and events of the teaching-learning context are managed by both teachers and students (Doyle 1979) and how the task structure in classrooms serves to integrate the "managerial and academic dimensions of classroom life" (Doyle 1990:130). Doyle argues that a crucial task of teachers is to organise and manage students and materials effectively. Failure to successfully accomplish this "can lead to a breakdown of academic work with predictable consequences for student achievement" (Doyle 1983:179). The difficulty involved in establishing and maintaining an effective managerial/ organisational system is acknowledged by Doyle and evident through his frequent references to the "multidimensionality" and unpredictability" of classroom life. In terms of the academic focus in instruction, Doyle refers to classroom lessons as a "system of overlapping task structures ...each of which consists of a goal, a set of operations to achieve that goal...and a behaviour ecology" (Doyle 1983:192). In addition, Doyle views the curriculum as a collection of academic tasks which require certain activities to be accomplished. As such, he includes this subject matter component as a classroom process variable. A similar perspective of task as the means of integrating curriculum and instruction is put forward by L2 task-based theorists (Long& Crookes 1989; Nunan 1989). However, unlike the L2 task-based research approach, Doyle approaches the study of academic tasks from a contextual point of view; that is, he sees them as embedded in the organisational and evaluative framework of the instructional context. On the organizational level, academic tasks are carried out within an activity system. Activities, which Doyle views as the basic unit of classroom instructional organization, are defined by several features: time boundaries (usually 10-20 minutes), behavioural expectations (e.g., silence during seatwork), student grouping (e.g., pairs or small groups) and topic focus (Doyle 1986). The evaluative framework of classrooms links academic tasks to an accountability system by which students are rewarded for successful performance - a "performance-grade exchange" (Doyle 1983:181). This performance relates not only to formal assessments such as tests and assignments, but to student classroom performance in such activities as discussions and question-answer sessions. Two contextual factors impacting on student performance on tasks are the result of this evaluative system. These factors are risk and ambiguity. Risk refers to the "stringency of the evaluation criteria" (p.183) for tasks and the possibility of actually meeting those criteria. For example, opinion-giving tasks are relatively low in risk since there is a wide range of acceptable outcomes. The second factor, ambiguity, is related to how clear the routes to successfully accomplishing the task are and to what extent the product can be predicted. Tasks involving memorization, for example, are low in ambiguity since both the process and the product are well defined. Doyle (1983) views an academic task as consisting of three components: i) a product that students are expected to produce; ii) processes students use to create the product; and iii) the resources given to accomplish the task product. Further, he suggests a typology based on the cognitive processing required for accomplishing tasks. According to this typology there are four main task types. The first type are memory tasks. These require students to identify or reproduce information they have been exposed to before; for example, a spelling test. Second, there are procedural or routine tasks which involve applying routine steps to produce answers; for example, mathematical problems. Doyle identifies the third type as comprehension tasks. These ask students to apply knowledge previously learned to a new situation; for example, writing a descriptive paragraph using previously learned steps. Finally, there are opinion tasks in which students state preferences and opinions (Doyle 1983:162-3). In relating task types to risk and ambiguity factors, Doyle views memory and routine tasks as low in both ambiguity and risk since these usually involve predictable answers and explicit evaluative criteria. However, comprehension tasks are high both in ambiguity and risk. This is due to the fact that such tasks (for example, writing a descriptive paragraph) require creativity as well as more student uncertainty regarding the criteria for evaluation. Finally, opinion tasks are high in ambiguity since a variety of answers is possible, but low in risk since more than one answer can be correct. This task classification is a useful heuristic device for task planning in that it provides consideration of the cognitive processes involved for task completion. Moreover, by pointing out the risk and ambiguity factors inherent in each task type, implications for the ease or difficulty students may experience in accomplishing these tasks (beyond the cognitive skills required by the task itself), can be considered. This has particular relevance for the L2 context where it is argued (e.g., Krashen's 'affective filter hypothesis, 1982), that students require a 'safe' affective environment for language acquisition to occur. In applying Doyle's risk/ambiguity scale to the L2 context, tasks which require the memorization or manipulation of grammar items would involve low risk and ambiguity. This may explain why many students express a desire to have more of these types of activities even though from a theoretical perspective they are not considered particulary useful for promoting communicative competence. Of particular concern is the effect that a teacher's 'mixed message' regarding expectations for a task may have on student performance. Doyle points out that students look for cues in teacher feedback as to what the 'real' expectations for task performance are (despite what the teacher may state they are). For example, (in the L2 classroom) if a teacher states that the focus for a particular interactive task is on communication of meaning but in monitoring the task, emphasises correction of grammar, will students respond by being more concerned with the accuracy than the content of their utterances? In addition, do these 'mixed messages' increase the level of risk and ambiguity for task accomplishment? Of interest to the present study was the relationship between teachers' stated language objectives (accuracy/fluency) for tasks and the type of feedback they provided to students during task implementation. Also of interest, was how teachers responded (in comments made during post-observation conferences in the study) to any observed inconsistencies between their stated language objectives for tasks and feedback during implementation. The socio-linguistic perspective on tasks and participation structures. Socio- linguistic research has also focused on task structure in classrooms but from a discourse perspective. This research has been concerned with "how language in the form of interactions between teacher and students and among peers ... functions in support of the acquisition and development of knowledge" (Green 1983:168). Guiding this research perspective are three fundamental premises. 1) Face-to-face interaction, basic to classroom discourse, is governed by culturally determined norms. 2) Students need to possess a 'classroom competence' in order to be successful in this context. This includes both linguistic and social competence so that they know "with whom, when and where they can speak and act...(and can provide) the speech and behaviour that are appropriate for a given classroom situation (N/lehan 1979:133). 3) The classroom context is a unique communicative setting in that verbal interactions are evaluated and more restricted than in other settings (Erickson 1986). Socio-Iinguistic researchers view classroom discourse as both 'rule-governed' and 'constructivist' (Erickson 1986). It is rule-governed in the sense that students need to know and competently use the social and communicative strategies that the teacher signals as expected. At the same time, classroom discourse is constructivist because the teacher and students do not "rotely follow scripts" but both "actively shape the lesson" by responding to "exigencies of the moment" (Erickson 1986:191). This constructivist element is most evident when a 'breakdown' occurs. Such a breakdown may be caused by student use of inappropriate social or communicative strategies; for example, several students responding to a question posed to one particular student. On the other hand, breakdown may occur because of a student's non-comprehension of the academic content. In the former situation, the teacher might modify the communicative structure of the lesson by allowing students to respond to questions without being individually solicited. In terms of the latter, the teacher might momentarily modify the academic task structure focus in order to provide an explanation of a concept to the student who is having difficulty. Student factors (particularly misbehaviour and non-comprehension of subject matter), have been found to be the primary influence for teacher modification to planned or routine paths through lessons in classroom-based decision-making studies '2 This view can be related to the discourse interaction continuum suggested by Kramsch (1985,1987) for the E S L classroom in that teacher/student roles are governed by the particular discourse interaction type implemented. (viz., Clarke & Peterson 1986). Interactive decisions and the factors influencing these in the second language classroom were also of interest in the present research. As indicated above, the socio-linguistic perspective of the classroom suggests that two structures operate simultaneously: a social participation structure and an academic task structure. While the former is concerned with the "communicative rights and obligations of individuals" (Erickson 1986:155), the latter focuses on the presentation and learning of academic content. Erickson argues that for successful participation in the instructional setting, students need to be competent not only in attending to classroom discourse norms, but also in grasping subject matter concepts. Teachers signal expectations (verbally and non-verbally) for student performance in both the academic task structure (subject matter) as well as the social participation structure (the social and communicative norms to which the students are expected to adhere). In addressing the issue of participation structure in terms of student grouping, the literature states that there are three main alternatives: whole class, small groups (including pairs) and individual (Philips 1972). Research findings have shown that in both regular and L2 instruction, the whole class, teacher-fronted format tends to be the predominant participation structure (viz., Edwards & Furlong 1978; Long & Sato 1983). Further, in relating this social organization to classroom discourse, it has been found that teachers centrally direct and control classroom talk; that is, they act as the 'gatekeepers' (Jackson 1968) of communication. In this role, a teacher either 'monopolizes' the speaker position (as in giving directions or explanations) or interacts with one student at a time while the rest are expected to listen attentively. Student participation may be voluntary or compulsory depending on whether the teacher solicits specific students for responses or comments. Edwards (1982) concluded from his review of a number of classroom interaction studies that: the typical patterns of interaction seem to transcend differences in the age and ability of students, in the prior intentions of teachers, and between areas of the curriculum. There is a persistence of whole-class teaching in which the teacher can only manage so many potential participants by allocating turns, monopolizing key turns and asking closed questions so as to retain the right to evaluate the ansvi^ers and to regularly regain the initiative. In the lessons proper, teachers lecture a great deal, ask questions to test pupil's knovt'ledge of facts already provided or to lead them by frequent use of cues and clues to answers already determined, and insist on putting into words the authorized versions of what is being said and has been learned. (Edwards 1982:517) Classroom researchers have attempted to explain why, despite the "presence of 30 or more communicators, prolonged periods of time are spent listening to only one" (Edwards & Furlong 1976:10). Several interpretations related to management issues, have been put fonward. LeCompte (1981), for example, concludes that this centralized teacher control is a management technique and a product of the 'work ethic' orientation of classrooms. For her, the framework of the instructional setting she observed was based on five 'work norms': i) time as a valued commodity; ii) academic achievement; iii) conformance to authority; iv) the value of orderliness; and v) the importance of keeping schedules (LeCompte 1981:179). LeCompte's speculation that this whole class participant structure was a feature of the 'management core' of classrooms is shared by a number of others (e.g., Jackson 1968; Good & Brophy 1978; Doyle 1979) in that teacher management strategies are a way of coping with the "unpredictability and multidimensionality" (Doyle 1983; 1986) of classroom life. This perspective of the classroom context as unpredictable and multidimensional promotes the often used analogy of the teacher as a conductor coping with simultaneous contextual and academic demands. The teacher's role in orchestrating these demands involves not only presenting academic content but at the same time dealing with the immediate demands of the classroom. This aspect of teaching is a significant issue to consider when examining the teaching process. Lack of attention to the 'unpredictable' nature of the teaching context, and more significantly how an individual teacher responds, was a major criticism of the product-based research paradigm. It might also be a criticism directed at current task-based research in the L2 classroom since in these studies, context factors and the impact these might have on task implementation are not considered. Certain instructional strategies have been suggested as ways in which teachers maintain centralized control in the classroom. For example, teachers predominant use o1 low-level, factual questions is seen by Doyle(1983) as a "complexity-reduction" technique. By using such 'display of knowledge' type questions as opposed to more open-ended ones, teachers are able to pace the content flow of the lesson and reduce the unpredictability of student responses. At the same time, the number of students contributing to the lesson is maximized since closed, factual questions tend to elicit short, 'to the point' responses. Moreover, the teacher regains the floor quickly and can "continue ...allocation of turns at speaking as well as shape the meaning of what is said in the desired direction" (Edwards & Furlong 1976:17). In terms of the effects of teacher-fronted instruction on learning outcomes, direct instruction advocates (e.g., Rosenshine) would argue that this format has positive impact in terms of achievement test gain. Interestingly, Wong-Fillmore (1985) also concluded from her study of elementary bilingual classrooms that teacher-fronted instruction had positive effects for the 12 students in these classes. This is surprising given the current advocacy for small group work in order to promote language learning (Long & Porter 1985;Long 1989; Pica,Young & Doughty 1987). Wong-Fillmore's bilingual classroom study (1985), in which she investigated the effects of instructional language and participation structure on the L2 students' language development, found that the more 'successful' teachers predominantly used whole class instruction. She concluded that this was a positive instructional format in that there was consistency in "how the lessons were organized, in the activities that were undertaken during each phase of the lesson, and in the language that was used in its conduct" (p. 29). Wong-Fillmore suggested that this instructional organization provided a "kind of scaffold for the interpretation of new materials" (p.30). In other words, the L2 students were able to focus their attention on the content of the lesson since they were familiar with the instructional delivery format and language, In contrast, the 'less successful' teachers in the study had not established '3 Interestingly, Wong-Fillmore noted that in the 'successful' classrooms, lessons were almost "scripted" in terms of the language teachers used; this contradicts the constructivist view of classroom discourse put forward by socio-linguistic researchers (e.g., Erickson 1986). a regular, daily instructional routine. These teachers used a variety of approaches for introducing and practising academic work. Wong-Fillmore observed that these teachers had to spend time explaining the 'format' of each new learning activity since students were frequently unsure of what they were expected to do. Although L2 language methodologists might argue that these additional explanations provide valuable language input for l_2 learners (e.g.,Krashen), those theorists promoting direct instruction techniques in the regular classroom, would suggest that constantly having to explain activities would adversely affect the momentum of the lesson, the amount of time spent on subject content and the actual time-on-task for students. All of these features have been found to be crucial elements for effective classrooms (Rosenshine 1979). Although Wong-Fillmore concludes that the teacher-fronted approach was the j<ey component in the 'successful' classrooms, a different interpretation would suggest that it was not this format per se but rather the well-established management and organizational routines that were key factors (viz., Doyle 1983,1986; Good & Brophy 1978). This becomes more evident in Wong-Fillmore's description of the 'less successful' classrooms. In these there seemed to be a more 'laissez-faire' management and organizational model operating as indicated by periods of time during which "students talked freely with each other" (p.22) rather than on given tasks monitored by the teacher. Wong-Fillmore's positive conclusions about the benefits of teacher-fronted instruction also seems contradictory given the concerns she expresses about the limited language practice she observed the 12 students had in this format. For example, Wong-Fillmore comments on the teachers' extensive use of low-level questions which were frequently answered by the teacher (pseudo-questions) and secondly, the use of inappropriate turn-allocation strategies which resulted in the more linguistically competent (LI) students contributing most of the responses. The instructional language used in teacher-fronted instruction appears to be problematic both from the perspective of developing higher-order thinking skills and in addition for L2 learners, the practise of language in communicative situations. In commenting on the teacher's dominant role in classroom discourse, Edwards states that the classroom is a context in which students are rarely able to; challenge the information being presented to them, to explore alternatives or test out their own solutions to problems. When hypotheses are formulated, conclusions drawn and generalizations constructed, these complex activities are likely to be the responsibility of the teacher, (p. 517) A concern about the minimal involvement that students in whole-class instruction have in the use of the decision-making skills that Edwards alludes to, has provided impetus for research on small group work in the regular classroom. It is argued that this organizational approach offers several advantages (Slavin 1989). First, group tasks enable students to adopt the roles usually reserved for teachers in the teacher-fronted format such as topic initiator, question-poser, and encourager. In addition, group tasks provide a way in which students can develop problem-solving and decision-making skills. Research on co-operative learning in small groups has indicated that there are significant learning benefits resulting from this approach provided that the following two critical features are built into the task structure: i) There needs to be a group goal which serves to motivate students to participate actively in the group task; ii) There must be individual accountability which provides a 'vested interest' for each member of the group to contribute and help each other achieve the group goal (Slavin 1989). According to Slavin (1989), traditional small group approaches, which usually lack these two critical features, "have been found to have small or non-existent effects on student learning" because students are encouraged to work together but given little structure and few incentives to do so ..." (p. 149). Slavin suggests that a group goal may be in the form of a report which is evaluated by the teacher and presented to the class. Secondly, individual accountability can be operationalized by having the grade for the whole group based on the average of individual scores on a quiz following task completion. Co-operative group learning models. Various co-operative group learning models have been developed. Each one differs slightly in terms of the use of six basic clnaracteristics. Tl ie first two, group goals and individual accountability, have been found to be critical for learning benefits and are central to all the models. The other four features are found in some of the models but not all. These include equal opportunities for success, team competition, task specialization and adaption to individual needs. Equal opportunities for success are provided in some models by awarding points to individual students on the basis of academic improvement. Team competition, a feature of the student learning team models (Slavin & colleagues 1978, 1986), has been found to promote motivation for completing the task well enough to compete with other learning teams in a final tournament. This component is also a positive factor in encouraging co-operation within teams as they are preparing for the competition. Task specialization is a feature common to most of the models and involves each member of a group choosing or being given a particular sub-task which contributes to the final group product. The last characteristic, adaptation to individual needs, is not found in most of the models since the focus is usually on group-paced learning; however, several models do take individual needs into account. Four different co-operative learning models have been developed: Jigsaw (Aronson 1978), Learning Together (Johnson & Johnson 1975), Group Investigation (Sharan & Sharan 1976) and Student Learning Teams, which include several different approaches (Slavin & colleagues 1978 & 1986). In Aranson's (1978) Jigsaw model (an approach often referred to in the small group task L2 literature) students are assigned to group teams. Each member of the team is responsible for researching one section of a given topic. Once this research has been completed, group members share their findings. According to Slavin (1989), a shortcoming of this model is the lack of group product to be produced. This reduces the incentive to help each other with the research and to actively participate in the sharing phase of the task (Slavin 1989:150). The second model. Group Investigation (Sharan & Sharan 1976), also has individual group members researching specific aspects of a topic; however, in contrast to the Jigsaw approach, it incorporates group goals and individual accountability features by requiring a group report to be produced and evaluated. In Slavin's (1989) view, the group investigation model is more successful than the jigsaw model for achieving learning objectives because of the existence of these features. The Student Learning Team model (Slavin & colleagues 1978,1986), which encompasses several different sub-models, emphasises three central features: team rewards (through evaluation procedures); individual accountability (members tutor each other); and equal opportunity for success (individual improvement over past performance contributes to the team's final grade). The fourth model. Learning Togetlier (Johnson & Johnson 1975), incorporates an interesting additional emphasis on developing collaborative group skills. Students are assigned particular roles for group work such as recorder of information, summarizer and observer of group interaction. In addition, students are explicitly taught language expressions useful for these roles. Because Johnson & Johnson believe that co-operative skills don't happen automatically, they stress that students need to learn and practise behaviours which encourage co-operation in small group settings. Equally important is the teacher's role in monitoring co-operative group behaviour and intervening when inappropriate behaviour occurs. These co-operative learning models developed for the regular classroom provide an important link to L2 classroom research on tasks and participation structure. Both the co-operative learning advocates and L2 task-based researchers argue that group-based tasks provide more opportunities for student interaction and active involvement in learning than large-group instruction (Slavin 1989; Long 1989; Pica.Kanagy & Falodun 1990). In addition, both identify individual accountability and group goals as critical features of tasks for learning benefits to occur (Slavin 1989; Pica,Kanagy & Falodun 1990). It is significant to note, however, that while in the co-operative learning models a distinction is made between the jigsaw approach and the group investigation approach (in terms of the features of individual accountability and evaluated group goals), this is not the case in L2 task-based literature. 'Group investigation' is not a label used to identify an 12 task type in second language theory. However, since in the L2 literature, 'jigsaw' is a label applied to tasks which involve a two-way information exchange (similar to the notion of individual accountability) and a single outcome, this task type appears to be more similar in design to the group investigation model in co-operative learning . Apart from the differences in terminology used, both the co-operative learning researchers and the 12 task-based researchers agree that the features of individual accountability and group goals are critical task components. In conclusion, this section of the literature review has examined two key constructs: academic task structure and social participation structure in the instructional setting. Research findings from regular classrooms indicate that large-group, teacher-fronted instruction is the major participant structure used by teachers. Further, these studies show that instructional language in this format is controlled and dominated by the teacher and that student contributions tend to be limited to a display of knowledge level. Secondly, both ecological and socio-linguistic studies have identified task structure as a valuable unit of analysis for the study of teaching. However, these two research models perceive task structure from different vantage points. Doyle (through the ecological perspective), views the task structure as embedded in the organisational and evaluative systems operating in classrooms. Accordingly, he stresses not only the importance of the teacher's task in maintaining effective management but also the impact of the evaluative system on how academic tasks are performed. In addition, Doyle's task typology is based on the cognitive processes required for the accomplishment of tasks. Socio-linguistic researchers, however, approach task structure from a discourse perspective and have focused on how language is used by both teachers and students during lessons. They argue that in studying the teaching-learning context, task structure cannot be separated from the communicative and social norms which operate and guide behaviour in the classroom. Finally, the notion of co-operative group learning was briefly discussed. This approach has been advocated as an alternative to whole-class instruction because it not only promotes social collaborative skills but engages students in decision-making and problem-solving activities not usually found in teacher-fronted instruction. This small group, task approach has also been found to be of significant benefit for promoting second language acquisition processes in the l_2 classroom (Long 1989; Pica, Kanagy & Falodun 1990). 2. Second language Classroom Studies In second language pedagogy, task has been a prominent feature of instruction since the advent of a more communicative approach in the 1970's. The concept of task in the L2 classroom includes two central features. First, tasks are usually associated with small group work. This participation structure is a central element of the communicative approach and continues to be advocated based on findings in current task-based research. Second, small group tasks are considered an optimal context for promoting second language acquisition processes. Given the importance of these features to the conceptualization of tasks in second language literature, these concepts will be addressed first. In communicative language teaching, 'task' is a label attached to those classroom activities designed to promote a learner's overall communicative competence. These communicative tasks, implemented in small groups or pairs, involve students in practising language used to express particular functions (e.g., asking for information) within a socio-cultural context relevant to the student's needs in the target (or second) language setting (e.g., asking for directions to a particular place in the local community). For such communicative tasks, emphasis is placed on 'getting one's meaning across' rather than on linguistic accuracy. In this way, group communicative tasks provide opportunities in the instructional context for learners to perceive, comprehend, and ultimately internalize L2 words, forms and structures ...(in activities designed)... not for the sake of producing language as an end in itself, but as a means of sharing ideas and opinions.... (Pica,Kanagy & Falodun 1990:2) Interaction tasks have been promoted by methodologists from two vantage points: small group work enables students to use language in the process of communication and second, group work changes the evaluative and managerial structure of the teacher-fronted classroom to one in which the students assume control of the communication process (Johnson and Morrow 1978; Littlewood 1981; Long 1989). In relation to whole-class instruction, Long (1989) summarizes the pedagogical rationale for the use of group tasks in the second language classroom. To begin with, in teacher-fronted instruction, individual students have minimal opportunity to practise the language. It has been well documented in both L2 and regular classrooms, that student contributions to discourse are limited to short (often one word) answers and the majority of teacher questions are display of knowledge type (e.g., Barnes 1975; Long & Sato 1983). In addition, teacher feedback to student responses focuses on the accuracy of the content or language (the latter in language classrooms). This question-response-feedback cycle dominating teacher-fronted instruction is maintained at a brisk pace in order to retain the attention of all students and to keep the lesson moving forward (Good & Brophy 1978). As previously discussed, researchers have suggested that these features are a result of the managerial and evaluative framework of the classroom context in which the teacher and students are engaged in a "performance-grade exchange" process (Doyle 1983). In contrast, second language methodologists increasingly argue that group work allows not only for an increase in the quantity of language practise students have but also enhances the quality of that practise. This perspective is shared by Barnes (1977) who suggested that group work encourages "exploratory talk" or "talking to learn" which he felt was critical for learning but unlikely to occur in the inhibiting whole class format with the "audience effect" coupled with a teacher concern for "short, polished products" (Long & Porter 1985:211). Groups provide a more secure, less inhibiting context thereby lowering the "affective filter" (Krashen 1978, 1980, 1981) which is considered a critical component for language acquisition. Further rationale for group work involves the possibility for individualizing instruction; that is, teachers can give groups different language learning tasks based on group interests, needs and proficiency level. The use of lesson tasks is not new in L2 methodology. Nor is the view that small group tasks are most useful for achieving fluency objectives. Small group tasks are an integral part of the communicative language teaching approach. However, more recent 12 classroom research has contributed a psycholinguistic basis for the use of this pedagogical technique (Long 1989; Long & Crookes 1989). Task-based L2 researchers believe it is critical that the use and design of tasks is grounded in findings from two L2 research fields: second language acquisition research and research investigating the processes involved in L2 learning in the formal context. Current classroom-based research on tasks has provided empirical evidence that group interaction tasks do promote second language acquisition processes in the formal setting (viz.. Long & Porter 1985; Pica & Doughty 1985,1986; Long 1989). In these studies an important underlying construct is the comprehensible input hypothesis (Krashen 1978, 1980) and its relationship to language learning. Comprehensible input refers to the idea that language addressed to non-native speakers is often simplified in various ways in order to aid comprehension. This 'foreigner talk' has been compared to the language that caretakers use in communicating with young children (motherese) learning their first language. Research on foreigner talk has shown that when speaking to second language speakers, native speakers modify their language in the following ways: they speak more slowly, articulate more carefully, and use less complex linguistic structures, higher frequency vocabulary and fewer idioms (Long & Porter 1985:213). Similarly, 12 classroom research investigating teacher talk has shown that E S L teachers employ foreigner talk strategies when interacting with learners in large-group instruction format (e.g., Brulhart 1985; Early 1987). An added dimension to comprehensible input in terms of the instructional context is Krashen's input hypothesis. Krashen (1981) argues that in the instructional context a learner must be exposed to language input that is slightly beyond (i + 1) the learner's current acquisition level (or interlanguage). Moreover, Krashen contends that if comprehensible input is available in the instructional setting, it will automatically include the appropriate level (i + 1) of input for acquisition to occur. Although there are unfamiliar linguistic elements present in tlie input, this is "compensated for by hearing them used in a situation and embedded in other language which they do understand" (Long & Porter 1985:214). While Krashen maintains that learner 'output' is not a necessary condition for the process of comprehensible input + acquisition to occur. Swain (1985) argues that learner output is indeed a crucial factor. Based on findings from a study of a French immersion programme in which the students upon completing the course lacked native-speaker productive competence. Swain concluded this was not due to the lack of comprehensible input but lack of student comprehensible output. Swain suggests that student output during the course had been limited for two reasons. First, there had been limited opportunity for students to practise the language in the classroom setting. Secondly, and more significantly, they had not been 'pushed' to produce at a level beyond a functional adequacy. Because they had been understood by teachers and fellow students (despite errors) and had not receive "negative input" for these errors, the students had not proceeded to a more competent interlanguage stage (Swain 1985:249). Current interlanguage research has been attempting to document what makes input comprehensible for second language learners in the instructional setting and how output is affected. Study findings suggest that merely "exposure to a target language is not in itself a sufficient condition for language acquisition" to occur (Pica,Young & Doughty 1987:737). In addition, despite Krashen's claims, exposure to comprehensible input alone, although helpful, has not been empirically proven as a causative factor in language acquisition (Long 1985). What has been observed is that when learners are given the opportunity to interact in order to modify the input until s/he understands what is being said, that this interactional process in which learners negotiate meaning, results in the necessary comprehensible input. Long (1985) concludes that "it can be deduced that the interactional adjustments promote acquisition." (Long 1985:85) However, it is important to note that "few studies have actually linked negotiation features found during task interaction with acquisition processes themselves" (Pica et. al. 1990:11). Rather, the link between negotiation in interaction and language acquisition processes has been inferred (Doughty & Pica 1990). Interactional adjustments have been defined as those strategies used to modify language input in some way (linguistically or conversationally) in order to help the participants understand the intended meaning of what is being said (Long 1980; Long & Porter 1985; Doughty & Pica 1986). For classroom research purposes, these strategies have been operationalized as confirmation checks, comprehension checks, clarification requests and repetition (Long 1980, Doughty & Pica 1986). Confirmation checks enable the listener to make sure s/he has understood the speaker correctly. These are usually in the form of repeating (with rising intonation) the speaker's preceding utterance. Clarification requests include verbal strategies the listener uses to indicate non-comprehension of the speaker's utterance and indicate a desire for speaker modification. These are usually direct questions such as "What do you mean . . .?" Comprehension checks are used by the speaker to confirm that the listener has understood the preceding utterance and includes questions such as "Do you know what I mean?". Finally, repetitions involve either the speaker repeating his/her own utterance (or part) or the listener repeating (Pica & Doughty 1985:119-121). Studies have shown that these strategies of clarification & confirmation requests and confirmation checks provide a "mechanism for input redundancy and repetition, which, in turn enhanced learners' comprehension for carrying out the directions of the task" (Pica et. al. 1990:12). Studies investigating these interactional modifications have indicated two significant benefits of small group tasks for language learning. First, small group work (including pairs) is more likely to promote the use of these strategies than whole class instruction. Second, these strategies are particularly prominent in tasks which require participants who individually hold specific pieces of information to pool this in order to complete the task successfully (Long 1983, Doughty & Pica 1984). Based on findings from interactional modification studies. Pica et. al. (1990) have formulated a typology of group interaction tasks. This provides a means through which tasks can be "identified, classified and compared with regard to their roles in meeting instructional and research purposes" (Pica et.al.1990). Two key features of group tasks are evident in this typology: i) There is a product or outcome that learners are expected to accomplish; ii) Tliere is a type of activity involved which requires active participation on the part of those involved. Findings from L2 interaction research have contributed to our understanding of how task variables and language learning are related and have provided input to the formulation of this typology. To begin with, research findings have indicated that meeting the activity and goal criteria in task design is not necessarily sufficient for tasks to stimulate language learning processes (as defined by interactional modification). What appears to be critical is the role relationship between the participants in terms of information-exchange. In this regard, Long refers to one-way and two-way tasks. This distinction refers to the way information is distributed at the outset of a task and the requirement that the structure of the task imposes on participants to exchange that information if they are to complete the task successfully. (Long 1989:13) A one-way task is one in which one participant holds all the information. The information receiver's role is relatively passive linguistically and usually involves a 'hands-on' activity such as drawing a picture or diagram according to the speaker's directions. In contrast, in a two-way task, each participant is given a piece of the total information and must share this in order to complete the task successfully. In reference to this distinction, the task typology (Pica,Kanagy & Falodun 1990) differentiates between independent and mutual participant relationships. An independent relationship refers to a task in which one participant is the sole information holder, whereas in a mutual relationship, each participant holds some of the information and they must communicate these 'pieces' in order to complete the task. In relation to this information-exchange feature, research findings highlight the fact that it is not sufficient for information merely to "facilitate or improve task completion" (Long 1989:13). This information exchange must be required in order to accomplish the task. In this case, participants need to use interactional strategies in order to achieve mutual understanding of the information. Research has also indicated that a closed task (one which requires a single, predetermined outcome) promotes more negotiation of language than an open task (one which allows for a wide range of possible outcomes) (Long 1989:18). In a closed task situation, participants need to reach a single solution or "one of a small finite set of correct solutions determined beforehand by the task designer" (Long 1989:18). Further, Long stresses that it is crucial that learners know whether the task is open or closed. Closed tasks stimulate more negotiation work in that learners know that successful task completion depends on their finding the pre-determined solution rather than being satisfied with any solution they choose. Through the process of negotiating comprehensible input in a closed task, there will be more adjustments and reformulations which in turn lead to "interlanguage destabilization" (Long 1989:18). The notion of open/closed tasks is accounted for in the task typology by the category of communication goal. This is classified into two components: goal orientation and outcome options (Pica.Kanagy & Falodun 1990). Goal orientation refers to whether the end goal necessitates participants reaching a consensus or not. A shared goal necessitates 'convergent interaction' (Duff 1986) since participants are required to reach a consensus. On the other hand, in a task allowing for independent goals, there is 'divergent interaction' because participants are arguing different points of view. Long and Crookes (1990) argue that a task involving a shared goal promotes more negotiation since learners must work at understanding each other's input in order to reach a mutually acceptable decision. In contrast, in an independent goal task, it is not so critical for learners to understand each other as there is no group goal incentive. In this case there will likely be less negotiation for comprehensible input. (Pica, Kanagy & Falodun 1990:7) The second component of the communication goal category is outcome options. This relates to the number of options participants are able to produce; that is, is only one outcome acceptable or are there multiple outcomes possible? Pica, Kanagy & Falodun (1990) identify five different task types resulting from various ways in which the task features discussed above are inter-related in the task design. These types include jigsaw, information-gap, problem-solving, decision-making and opinion-exchange. A jigsaw task is one in wliicli eacli group member hiolds part of the total information and this must be pooled in order to accomplish a single goal. Of the five task types, this one is considered to be the most useful for providing opportunities "for interactants to work towards comprehension, feedback and interlanguage modification processes" (Pica,Kanagy & Falodun 1990:9). The key features promoting these language acquisition benefits are mutual relationship (i.e. two-way information exchange) and single outcome. The information-gap task, on the other hand, is a one-way task in which one participant holds the key information needed for accomplishing the task. The receiver's role is to respond in some way (e.g., drawing a diagram) to this information. In this type of task, there is a single, convergent goal and a single outcome. Information-gap tasks can be beneficial for language acquisition if a version of the task is repeated so that the information holder and receiver roles are switched. This will provide opportunities for both learners to give and receive information (Pica, Kanagy & Falodun 1990:9). A problem-solving task involves a single goal and outcome so that participants need to work collaboratively. However, participants share the given information equally (therefore there is no information gap). Similarly, in decision-making tasks, learners have the same information and there is a shared goal. However, because there is more than one outcome possible, there is less incentive for individuals to make themselves understood or help each other. Finally, opinion-exchange tasks allow for participants to choose whether they arrive at a consensus opinion or not as there are numerous acceptable outcomes possible. It is important to note that, as Pica and colleagues (1990) point out, task types can change by simply modifying one feature. For example, a decision-making task can become a problem-solving one by merely requiring that participants arrive at a single decision. This has implications for the way in which teachers make decisions to modify commercially designed (text) tasks to meet specific instructional goals. Pica and colleagues (1990) argue that a jigsaw task is the most likely type to promote interactional modification as the participants need to comprehend each other's information in order to produce the single expected product or outcome. They state: Under such conditions, learners and their interlocutors must work to understand and be understood by each other and to supply each other with feedback when mutual comprehension becomes difficult or impossible. Opportunities for negotiation and its consequent activation of acquisition processes are more likely to occur under these task conditions than those in which participants have no obligation to exchange information or have a variety of goals and task outcome options available to them. (Pica,Kanagy & Falodun 1990:8) Task-based interaction studies have found that not only the task type but also the way in which students are grouped is critical for promoting the type of language use believed to be important for language acquisition. For example, one study (Long et al.,1976) comparing teacher-led, whole class discussion to unsupervised pair work on the same task, found that the learners in the latter situation talked more and focused on communication for meaning. In contrast, in the whole group format, the teacher limited learner practise of language by the use of display of knowledge questions, and focused on form rather than communication through error correction. Similarly in other studies investigating the effects of task type and participation structure on student language use, findings have shown that student grouping format is critical. In studies comparing the effect of one-way versus two-way type tasks on language use, findings have shown that there is increased student-negotiation work occurring in small groups working on two-way tasks. However, findings have also indicated that there is no significant effect on negotiation work between one-way and two-way tasks in the teacher-fronted format (Pica & Doughty 1985; Doughty & Pica 1986). This suggests that the use of small group participant structure is a critical factor. Based on these research findings, Long (1989) observes that "a task's true potential may not be realized at all in a lockstep format" (p. 19) and concludes that two conditions appear to be necessary to realize the full benefits of task work: the use of two-way, information-exchange tasks and small group format. In conclusion, the use of tasks in small groups has been a central component of L2 instruction since the Communicative Language Teaching approach was introduced in the 1970's. Recently, task-based classroom research has been investigating the benefits of this approach in relation to both second language acquisition research and the way in which second languages are learned in the formal instructional setting. Based on these research findings, there is empirical evidence to show that small group tasks are beneficial for promoting language acquisition through the increased quantity and quality of student language use. In addition, two-way information-exchange tasks have been found to be the most useful task type for promoting negotiation of language opportunities and "consequent activation of acquisition processes" (Pica, Kanagy & Falodun 1990:8). Four critical features are included in this task type: 1) each participant holds a certain portion of the total information which must be shared in order to complete the task; 2) each participant is required to give and receive information; 3) participants have a shared goal; and 4) there is only one acceptable outcome (Pica et al. 1990:7). Both the regular and L2 classroom research reviewed in this section have examined task and participant structure from various perspectives. Regular classroom researchers have looked at the contextual factors impacting on task implementation and completion. These have included the influence of the evaluative framework of classrooms (Doyle 1983), and the socio-linguistic norms established (Erickson 1986). In addition, Doyle has formulated an academic task typology based on the cognitive skills involved in completing tasks. In L2 research, tasks have been approached from a language learning perspective. These studies have shown that small group participation structure is more beneficial for promoting student language use believed to aid language acquisition. L2 researchers have also formulated a task typology but this is based on categories research has indicated influence student language use. Task and participant structure were also central to the present study. Of particular interest were the following questions: 1. How do L2 teachers conceptualize 'task' ? 2. What factors do L2 teachers consider when planning and implementing tasks? 3. What role do L2 teachers adopt while students are engaged in small group/pair tasks? 4. How do individual teacher beliefs, theoretical knowledge and context factors impact on teacher decisions for task and participation structure? CHAPTER IV TEACHER DECISION-MAKING Introduction Researchers conducting teacher thinking studies view decision making as fundamental to teaching and as such, focus on the thought processes of teachers as they plan and teach lessons. In reference to the premise guiding this research tradition, Shavelson (1973) states: Any teaching act is the result of a decision, whether conscious or unconscious, that the teacher makes after the complex cognitive processing of available information. This reasoning leads to the hypothesis that the basic teaching skill is decision making. (in Clark & Peterson 1986:273) In order to provide some perspective on the numerous studies that have been conducted on teacher decision making, it is useful to look at the summary framework formulated by Shavelson & Stern (1981:461). This framework is helpful in two ways. First, it synthesizes the variety of variables that have guided these studies. Second, it suggests how these variables are related and lead to instructional planning and implementation in the classroom. Research on Teacher Decision-Making Teacher Characteristics Antecedent Conditions * — Teacher Evaluation of Decisions Teacher Cognitive Processes Instructional Planning + interaction with Students Consequences for Students Figure 4. Shavelson & Stern 1981:461 The present study is concerned with what factors influence 12 teacher planning and teaching decisions; therefore research findings from studies investigating Antecedent Conditions (information about students, the nature of the instructional task, and the setting) and Teacher Characteristics (beliefs and perceptions) were particularly relevant in developing the conceptual framework. It is important to note that for the present study these two groups of variables were not considered static entities but rather in a dynamic relationship with teachers' evaluation of their planning and implementation of lessons. In other words, ongoing experience in the classroom serves as input to confirm, add to or modify teacher beliefs about language teaching in general and perceptions of the specific situation (including the students, the nature of the instructional task and the setting). This reciprocal relationship among these variables is suggested in the Shavelson & Stern framework by the cyclical connection of the various categories of variables. It is also important to point out that the 'Consequences for Students' category in tlie framework was only considered in this present study in terms of how the teachers involved perceived the learning outcomes of their students. There were no empirical data collected in an effort to correlate teaching techniques with learning outcomes. This focus was not the intent of this study. Before examining the decision-making literature, there are several limitations of this research which need to be considered in terms of the applicability of findings to the present research framework. First, the majority of pre-active/inter-active decision-making studies have been conducted in the regular classroom at the elementary level. The extent to which these findings can be generalized to either secondary or adult levels has not been established. In addition, there has been little interest in examining experienced teacher decision-making in the E S L context apart from two recent studies on planning decisions (Woods 1989; Gumming 1990) with university-level E S L teachers. Secondly, most studies in teacher thinking have focused on either planning decisions or decisions made while implementing plans in the classroom. There have only been a few studies relating planning and classroom teaching. The present study contributes to further understanding of this planning-action link by examining not only what teachers consider when planning lessons but also how and why these plans might be modified during teaching. Given these limitations for the research discussed in this section, there are a number of findings consistent across a variety of studies that have been significant in the conceptualization of the present study. The first part of this chapter reviews regular classroom research on teacher decision-making research: 1) studies concerned with planning (pre-active/post-active) decisions; 2) studies investigating decisions made while teaching (interactive studies); and 3) the "smallest and youngest" group (Clark & Peterson 1986:287) - studies examining the content and organization of teachers' practical knowledge. The second section presents findings from L2 decision-making research. A. Research on Teachers' Pre-active Decision-making This research has been concerned with the way teachers plan for teaching. A consistent finding has been that experienced teachers do not follow the planning process prescribed in most teacher education programmes. This prescribed linear model (Tyler 1950) specifies the following sequenced approach to planning: i) the objectives for the lesson are selected; ii)activities to accomplish those objectives are chosen; iii)procedures for evaluating the effectiveness of the plan are detemnined. Findings have shown that teachers do not necessarily follow this sequence, nor explicitly include all these components (Clark & Peterson 1986). In terms of the attention to objectives in planning, there are conflicting results. Naturalistic studies (those conducted with teachers planning for their own classes) have consistently shown that activities, not objectives, are the guiding focus in teacher planning (Yinger 1977; Morine-Dershlmer 1977). On the other hand, researchers conducting laboratory simulation studies (those in which teachers plan for a hypothetical group of students) claim that teachers do consider objectives (e.g. Zahorik 1975; Marx & Clark 1978) In considering these differing results, it has been suggested that they may have more to with the research methodology employed than actual differences in teacher planning approaches (Clark & Peterson 1986). For example, in the lab simulation planning studies referred to above, teachers were explicitly asked to indicate which goals from a given list they considered while developing the lab study lesson plan. However, in naturalistic studies researchers did not focus attention on any pre-selected planning variables. Rather teachers merely reported on what they had considered while planning a lesson. Given that the researchers specifically focused on objectives in the lab simulation studies, it is not surprising that researchers found objectives as a major planning variable. An additional interpretation not suggested in the literature relates to the significant role that activities play in planning. This suggests that the role of objectives should perhaps be considered from a slightly different perspective. That is, rather than looking at only the existence/non-existence of overall lesson goals, we might want to concentrate on the guiding influence of micro or task-based goats in planning lessons. Given that in both regular and second language teaching literature the definition of 'task' includes goals as a prime component, it seems reasonable to suggest that even if teachers do not designate 'global' lesson goals, they may very well have explicit or implicit task-based goals. This micro-goal level does not appear to have been considered in the planning studies. This may be due to the fact that these planning studies have not linked this particular focus in pre-active and inter-active decision making studies. In such a study, the reason why teachers might modify tasks in the lesson itself may reveal a concern that the underlying but perhaps implicit task goals were not t)eing met, necessitating a change mid-lesson. In addition, while teachers may not write down specific goals for each lesson or each task, by examining teacher beliefs about teaching and learning and their perceptions of their particular teaching context, it may become evident that they choose certain task types and content with very clear objectives in mind. For example, in the second language classroom, teachers may choose small group, problem-solving tasks when they want to promote language fluency development, whereas they may decide on whole class tasks for explanations of a grammar point. A feasible interpretation of such choices concerning tasks and participation formats is that they are guided by a teacher's view of what is most effective for achieving different language learning outcomes and his/her related objectives for a particular class of students. A focus of the present study was to examine the role of teacher objectives (and the underlying beliefs and perceptions) in selecting the types of tasks and associated participation structure for their instructional context. A second interesting finding from naturalistic studies is that experienced teachers rarely write full lesson plans. Rather they usually write down only a list of the activities they want to include in a lesson. The full plan for the lesson is more often a "mental image" (Morine-Dershimer 1977) or "script" (Abelson 1976) plan which is implemented in a lesson. Full written lesson plans seem to be most often developed only when administrators request them or for substitute teachers (McCutcheon 1980). The extent to which teachers plan 'full' lessons either on paper or as mental scripts was also examined in the present study. This may have implications for the flexibility teachers demonstrate towards mid-lesson modifications during teaching; that is, are teachers who globally plan more likely to modify intentions in response to student or contextual situations than those who are detailed planners? This was suggested to be the case in a study by Zahorik (1970) who found that teachers working from well-developed plans (using a linear planning model) "exhibited less honest or authentic use of the pupils' ideas during the lesson" (Clark & Peterson 1986:267). There have been few studies investigating the link between planning and classroom action. Clark & Peterson's (1986) review of research on decision making include references to only four such studies. In addition, the relevance of the findings to the adult context (as in the present study) is questionable since they were conducted at elementary (two studies), pre-school and junior high levels. The most relevant of these studies to the present research is Carnahan's (1980) naturalistic study of nine fifth grade Mathematics teachers. In this study, teachers' written plans were compared to what they did in the classroom. A similar approach was taken in the present study in that both teacher verbal reports about their planning and Implementation as well as lesson plans were analysed in relation to what had occurred in the lessons observed. Findings from Carnahan's study indicated that teacher plans concerning the organisation of students for learning (whole class, small groups etc.) and content to be covered positively correlated with classroom implementation (Carnahan 1980); that is, teachers did what they said they were going to do in these two areas. In response to the findings from these few studies, Clark & Peterson (1986) suggest the following in terms of the link between planning and teaching: Planning shapes the broad outline of what is possible or likely to occur while teaching and is used to manage transitions from one activity to another. But once interactive teaching begins, the teacher's plan moves to the background and interactive decision making becomes more important. (p. 267) Clark & Peterson's view of planning as serving an organizational function seems consistent with the planning research that has found teachers limit plans to listing the activities or tasks to be included in a lesson. The way in which teachers might elaborate this global plan or 'fine tune' it, has been a focus of interactive studies. B. Research on Teachers' interactive Decision-Making Interactive studies have examined teachers' thought processes while teaching. In order to elicit these, teachers are shown video tapes of their lessons and asked to comment on decisions they made during the lesson (stimulated recall technique). Two main research foci guide these studies: i) the points during the lesson when teachers routinely make Interactive decisions and ii) the factors which influence momentary decisions to modify the planned intentions for the lesson. Decisions teachers routinely need to make during a lesson include such examples as selecting students to respond to teacher questions in large group Instruction and responding to student questions or comments. However, there are unpredictable situations which arise, and depending on the teacher's assessment of the situation, require managerial or instructional decisions. These decisions might result in a modification or even abandonment of the intended plan. Peterson & Clark (1978) hypothesized four alternative decision-making "paths" teachers might follow during a lesson. These ranged from no changes at all from the planned route to a completely different plan being implemented mid-lesson (p. 275). Findings from their study indicated that teachers do not often Implement alternative strategies during teaching as long as s/he "judges that the students are understanding the lesson and participating appropriately" (p.274). This, according to these researchers, supports the notion that teachers work from a 'mental lesson script' consisting of "well established routines" which "minimize conscious decision making during teaching" (p. 275). However, \Nhen teachers do decide on alternative courses of action in response to an unexpected situation, research findings suggest that there are three prime factors: student behaviour (including misbehaviour as well as Indications of non-understanding); contextual difficulties (e.g., shortage of time or materials); and the teacher's affective state at the moment (e.g., exasperation over homework not t>eing completed) (Peterson & Clark 1978:276). Responses to these factors may be in terms of managerial decisions (usually in response to student misbehaviour) or instructional decisions. A research focus of the present study was the linkage between what teachers plan for the observed lessons and how these plans might change during the lesson itself. Of particular interest was the teacher's perspective of why these changes were made. It was expected that the findings from studies reported above concerning the influence of contextual, student and affective factors would play a similarly influential role in the interactive decisions teachers make in the L2 classroom. It was not expected that student misbehaviour would be a factor since this study involved adult learners. Teachers' decision-making also involves practical knowledge accumulated through ongoing experience in the classroom. Research on teachers' practical knowledge differs significantly from the pre-active and inter-active decision-making research in both the premises underlying the research and the methodological approach employed. C. Research on Teachers' Practical Knowledge Studies on teachers' practical knowledge (Elbaz 1983, Clandinin 1986) focus on identifying how teachers "attempt to personalize ...and 'make' practical ...theoretical ideas" (Court 1988:74). According to these researchers a teacher's theoretical knowledge (of teaching and learning) serves as input to classroom practice. As such both theory and practice constantly interact; that is, teachers operationalize theory in the classroom by personalizing and making it practical in relation to their own instructional contexts. Teachers' practical knowledge is built from classroom experience and shaped by the "particular mix of knowledge, beliefs and values that each teacher holds" (Court 1988:64). Practical knowledge researchers have criticised the main body of research on teacher thinking for not considering the central role of teachers' practical knowledge in instructional decision making. Clandinin (1986) suggests two main reasons for what these researchers believe is a major shortcoming in the research. First, teachers are considered not as holders of knowledge in their own right but only as "conduits of knowledge (theoretical and cultural) as embodied in curricula, teaching approaches and policies" (p.3). Thus while there may be acknowledgement that practical knowledge influences the implementation of curricula (particularly if curricula are not successfully implemented), teachers are considered only as agents of presenting someone else's intentions (in this case the curriculum designer). Secondly, Clandinin (1986) argues that the conception of what is 'valid' knowledge has been limited by researchers to theoretical knowledge. The validity of experiential knowledge has not been acknowledged in its own right but only in relation to the successful classroom Implementation of curricula. In explaining their perspective, practical knowledge researchers frequently allude to Schwab's (1969) distinction between the functions of practical and theoretical knowledge in curriculum. For Schwab, the theoretical Is associated with the global level of curricula, whereas the practical deals with specific concerns. In relating this to curricula, the designers develop curricula on a global enough level to allow applicability across a wide range of teachers, students and settings. On the other hand, teachers are more concerned with the practical level in that they modify the curricula to meet specific learner and contextual needs. Schbn (1983,1987) also stresses the role that experiential knowledge plays in the instructional process. However, in contrast to the practical knowledge researchers' perspective, he does not perceive theoretical knowledge interacting with experiential knowledge. For Schon professional practice is not improved by increased theoretical input, but by encouraging teachers to reflect more on their classroom practices.. He proposes two stages of reflection as central to professional development; refiection-ln-action and reflection-on-action. The former refers to teachers' formulating responses to unique and puzzling situations encountered in lessons. In this case, Schbn argues, practitioners use past actions and the momentary situational factors In order to hypothesize and generate possible actions which they then Implement. Reflection-on-action, however, occurs when teachers think about the lesson afterwards. At this point, teachers have the opportunity to think about the unique incidents "when action can no longer make a difference to the situation" (Court 1988:144-5). An important element of Schbn's reflection hypothesis is the view that 'problems' are continually reframed using experiential knowledge, rather than solved. This process of reframing (both during and after the lesson) constitutes reflection and new understanding of the situation. According to Schon it is this reflection process, not theoretical knowledge, that informs and improves practice (Grimmett 1988:9). Based on this view, Schon argues for a redirection in teacher education programs, particulary in the supervision of practicums, towards the development of teachers' skills in reflecting on their practices. Critics of Schon's advocacy for the role of reflection not theory in improving teaching practices, have taken issue with the resulting dichotomy between theory and practice his view presents and the limited usefulness of theory he espouses. One critic suggests that by negating the place that theory has in informing practice can lead to "wholly idiosyncratic practitioners whose primary way of operating is to invent unique solutions to problems" (Gillis 1988:50). Similarly, Fenstermacher (1988) argues that both theory and practice inform classroom practice; therefore, the focus should not be limited to only one extreme in devetoping professional competence. While Schon prescribes teacher reflection for informing and improving professional practice, other researchers (Clandinin 1986: Elbaz 1983) view theoretical knowledge as the base which is then modified by experiential knowledge. For these researchers "practice reflects theory in a constant cycle of disjunction and reconciliation of ideas" (Court 1988:70). In this way, there is no separation of theory and practice but, in fact, teaching is theory in action (Connelly & Clandinin 1986). Practical knowledge studies seek to understand the content and organization of teachers practical knowledge through long term, small sample, descriptive research using some classroom observation but mainly open-ended conferencing. The difficulty for these researchers was developing a ' language' or conceptual system for describing teachers' practical knowledge (Court 1988). Elbaz's (1983) approach was to analyze the practical knowledge of the one teacher in her study, Sarah, in three hierarchical levels: rules of practice, participating principles and images. Rules of practice include teacher statements regarding what she usually does in certain situations. These might be compared to Morine-Dershimer's notion of Voutines' teachers implement almost intuitively. An example of this would be waiting for all students to be quiet before beginning a new phase of a lesson. A practical principle is a more abstract generalization which indicates underlying values or beliefs. These principles " may be derived quite formally from theoretical viewpoints, they may grow intuitively out of experience or (most likely) may develop from some conjunction of theory and practice" (Elbaz 1983:134). A teacher statement expressing the view that small group learning promotes a positive affective classroom climate might be an example of a practical principle. Finally the concept of image is the most abstract level of this hierarchy. According to Elbaz, images provide an overall orientation to teaching rather than specific actions and reflect "feelings, values, needs and beliefs ...of how teaching should be" (Elbaz: 1983:134). For example, a teacher might perceive the classroom as analogous to a home environment rather than an Institutional one. Elbaz views the three levels of practical knowledge inter-relating so that, for example, a principle or an image might result In a rule of practice. Rules and principles of practice, however, are usually concerned with Instructional knowledge; whereas images are more encompassing in that they tend to organize all aspects of practical knowledge (Elbaz: 1983:138-9). A major concern raised about this line of research is the interpretation of teacher report data on classroom practices. In the first place, mis-interpretation of the data, particularly In the case of identifying images, could lead to an inaccurate analysis of teacher belief systems about teaching. Second, since beliefs and values "are often implicit and difficult to verbalize" (Court 1989:3), teachers' comments may appear inconsistent with their classroom practices and actions. In response to this apparent inconsistency, Hargreaves (1979) suggests that there are two levels of values - abstract and contextualized (Court 1988). Abstract refers to values stated in isolation from context. On the other hand, contextualized values are embedded in classroom practice and can only be Interpreted in relation to the specific instructional context on which the teacher is commenting. An inconsistency may occur if a teacher is asked about his/her values but this is not grounded in particular classroom actions. In this case comments may reflect an abstract level so that there may appear to be a gap between what teachers claim in isolation from context and what their classroom practices imply. These practical knowledge studies have been small sample-based (Elbaz studied one teacher) and conducted over an extended period of time (Elbaz's study took two years). The primary data collection technique is open-ended interviewing. In reading these studies, one is struck by the subjective stance assumed by the researcher in these conferences. As a result, we learn a great deal about the researcher's beliefs and ideas about teaching when reading the interview data. This 'agent-central' role of the researcher in the study exemplifies the view that the researcher's own values and purposes are central to the research study and accordingly the "knowledge claims made (are)...dependent upon the shared meaning created by the researcher and teacher participants" throughout the study (Clandinin 1986:13). In contrast, in an 'agent-free' role, the researcher adopts a "neutral, objective stance " in relation to the teacher participants and uses methodological techniques to ensure objective distance (e.g., anonymous questionnaires) (Clandinin 1986:12). However, as Clandinin points out, many studies include elements of both researcher role extremes as detennined by data collection techniques. For example a study might include an established observation instrument (e.g., Flanders Interaction Analysis) for observing classes (imposing a more objective perspective than anecdotal notes) and this observation might be followed by an unstructured conference on the lesson (allowing for a more shared interpretation of the lesson). In addition, Clandinin's and Elbaz's research focused on the teachers' interpretation of their classroom practices as opposed to how these practices fit into a theoretical framework. A s such, "the teachers gave their accounts in their own terms, not in terms imposed by the researchers" (Clandinin 1986:p.11). Moreover, data were not collected to 'fit' pre-selected, theory-based variables. The intent in the present study was to examine teacher planning and classroom practises. Data collected through unstructured classroom observation and subsequent in-depth conferences, provided a basis for comparing teacher actions with teacher comments about planning and implementation decisions. However, key concepts in language teaching theory also played an important role in guiding the analysis of the data. Teachers' interpretations of these were examined both through their comments and observation of their classroom practices. Therefore, the Inter-play between teachers' theoretical knowledge and practical knowledge was central to this research. This has not been a focus in second language research to date with the exception of the studies conducted by Cumming (1991) and Woods (1989) discussed below. D. Research on Teacher Decision-making in the L2 Context Cummlng's 1991 study examined how experienced E S L teachers planned curricula for two specialised E S L courses: a preparatory course for Chinese chemistry teaching assistants and an intensive course for visiting Japanese university students. The findings that Cumming reports are significant in that ttiey Indicate the range of concems teachers address when planning. Cumming identified eleven "inquiry processes" that the teachers participating in his study employed in the pre-active decision-making process. These included: L gathering sources; 2. analyzing materials to create pedagogic components; 3. analyzing the social situations students are or will be in; 4. deciding values to place on aspects of course content (emphasis on one area or another) or group processes (e.g., focus on individual vs. whole class); 5. identifying dilemmas in individual student's lives (e.g., problems, goals, experiences); 6. considering indicators of students' learning; 7. considering how the curriculum may be useful to others (e.g., sharing with colleagues); 8. evaluating the effectiveness of the course; 9. considering alternatives (e.g., different presentation styles); 10. assessing value for own personal development; 11. mediating collégial "politics" (with other teachers, assistants,bureaucracy). (Gumming 1991) It is clear from this extensive list of planning behaviours that these teachers attended to more than the pedagogical components (planning tasks and selecting materials) of the course; they were also concerned with meeting the affective and social needs of the students, contextual factors (colleagues and the administration) and personal professional development. Woods' study focused not only on what a group of E S L teachers considered as they planned but how these decisions reflected individually held beliefs about teaching/learning a second language. This decision-making study was conducted with eight university-level E S L teachers in Eastern Canada who were teaching courses for the first time. Teachers kept planning logs In which they "noted important decisions they made or issues they considered" (Woods 1989:4). In addition, each teacher was interviewed on a weekly basis. An attempt was made to link teacher self reports (through the togs and interviews) with classroom practices by video-taping one lesson and eliciting teacher comments on this lesson while viewing it. Woods concluded from this study that the planning decisions reported by these teachers concerning tasks, materials and objectives revealed; 1) internal consistency in individual teachers' planning across lessons and the whole course; 2) coherence between the assumptions teachers held about language teaching/learning and the instructional decisions they made; and 3) a "dramatic" inconsistency between teachers in the instructional decisions made. (Woods 1989:7) Woods' also found that in addition to internal consistency between beliefs and decisions, there was coherence in how decisions were related to each other. Woods states: ... decisions were not randomly related or independent from one another, but rather continually inter-woven; i.e., made in the context of other decisions. (Woods 1989:6) Teacher decisions were inter-related sequentially (one activity follows another) or hierarchically (one activity builds on another). Woods concluded that this approach to decision-making was one way in which teachers coped with "huge number of things that are impinging on the teacher's consciousness and sub-consciousness at any point in time" (Woods 1989:11). Woods' supports his conclusion that teacher decisions are "coherent with deeper underlying assumptions held by each teacher about language learning and teaching" by examples from the study data for two of the teachers participating in the research. For example, one teacher emphasized the importance of learner independence and another prioritized authentic communication. These perspectives were evident in their planning and classroom practices although not components explicitly stated in the prescribed curriculum. While one teacher was primarily concerned with following the institutionally prescribed curriculum, the second teacher adopted a more learner-based approach. This second teacher based most planning decisions on what he perceived the students wanted and needed as the course proceeded "even if it involves reinterpreting or sacrificing aspects of the preset curriculum (Woods 1989:15). This learner-based focus was further supported in the classroom observation. In that lesson. Woods observed that the students seemed to play a major role in determining the extent to which the teacher stayed with the initial lesson plan. On the other hand, the first teacher, whom Woods described as a "curriculum-based" teacher, did not 'allow* students to detour her from the planned route. Consistent with findings from pre-active planning research with regular classroom teachers, the L2 teachers in Woods' study did not develop detailed lesson plans. Even though the teachers in Woods' study were teaching new courses, lesson plans were very global and remained "extremely tentative" and open until the lesson itself. They therefore served as a general guide so that the teachers "had a good idea of what they were doing and why they were doing it" (Woods 1989:13). Woods' study is particularly significant for the present research not only because it was concerned with the types of instructional decisions teachers make in relation to individually held beliefs but because it included some classroom-based data on how these pre-active decisions were implemented in practice in an equivalent setting to this present study. In conclusion, this chapter has reviewed two research perspectives on teacher decision-making: research investigating teachers' pre-active/inter-active decisions and research examining the role of teacher beliefs and practical knowledge in teachers' the decision-making process. These studies have significantly contributed to the conceptual framework for this present study (Table 1-2) and the following questions relating pre-active and context factors to teacher decision-making: 1. What role do teachers' beliefs and contextual factors play in pre-active and inter-active decision-making; 2. To what extent are teachers' decisions internally consistent with stated beliefs about the nature of l_2 teaching/learning and externally consistent with L2 theory. CHAPTER V THE RESEARCH DESIGN This study was concerned with teacher decision-making in the E S L setting and the role that teachers' individual beliefs and perceptions, context factors and L2 theory play in their planning and implementation decisions. This was a qualitative study which focused on the individual teacher's perspective of his/her instructional decisions. Underlying such a research approach is the belief that "the social scientist cannot understand human behaviour without understanding the framework within which the subjects interpret their thoughts, feelings and actions"(Wilson 1977:248). The "framework" in this particular study included both the theoretical and practical knowledge teachers hold about second language instruction and the actual setting (classroom and institution) in which they were working. The first section of this chapter provides an overview of qualitative research methods with particular reference to the educational setting. This general discussion is followed by a description of the research design used in this present study and the specific data collection and analysis techniques employed. A. DATA COLLECTION AND ANALYSIS: A GENERAL PERSPECTIVE A qualitative approach adopted by a number of researchers in the educational field is ethnography. Educational ethnographies are concerned with how events in the smaller communities of schools and/or classrooms are "socially organized and ... conducted for certain purposes" (Mehan 1981:49). Ethnographic research is characterized by several basic features. First, researchers enter the research setting with an 'open mind'; that is, rather than pre-determined hypotheses beginning the data collection process, prepositional statements are generated from data analysis in the initial stages. These serve to "sensitize one to the nature of the setting..."(Bogdan & Taylor 1975:80). Further data are then collected to "demonstrate the plausibility of (these) hypotheses" (p.80). This has been referred to as a "bottom-up" approach (Green & Wallatt 1981) because the formulated typology of behaviour emerges from observation in the field. In contrast, a "top-down" approach involves selecting variables and establishing hypotheses to test before the research begins. In a "top-down" approach, "little or no attempt is made to ... construct new variables or modify existing variables to reflect what is occurring in the specific situation" (Green & Wallatt 1981 :xiv) during the research process itself. The extent to which an ethnographer can enter the field with "no prior conceptual expectations" (Erickson 1986) is an issue addressed in the literature (viz.,Erickson 1986; Sevigny 1981;Wilson 1977). Erickson, for example, argues that the notion of the ethnographer entering the research setting 'tabula rasa', learning the method as s/he goes along and after "tremendous emotional strain", formulating an analysis scheme is an "extremely romantic" one (Erickson 1986:140). He argues that a more deliberative approach to field work enables the researcher to frame research questions and seek relevant data. The researcher's prior knowledge of and experience in similar settings provides an etic perspective which can constructively contribute to the initial, organizing framework for observation. As the study proceeds, and the ethnographer becomes aware of the "internal orderings" (or emic perspective), the initial framework can be altered so that there is a "blend of emic and etic perspectives" (Sevigny 1981:76). In the present study, the initial research framework (Table 1-2, Framework of Relationships) was informed by both regular and L2 classroom research findings, key concepts in current L2 methodological literature and the researcher's own teaching experiences. As the study progressed, the participants' actions and perspectives served to elaborate and refocus this framework. A second characteristic of ethnographic research is that it involves a cyclical process of generating and testing prepositional statements. In the initial stages, research questions are formulated according to the researcher's experiential and theoretical frameworks. As the study proceeds, prepositional statements generated by the data are confirmed, modified and/or elaborated by checking the participants' perspectives until the researcher is sure the account is both complete and valid. In the present study, initial analysis of classroom Observation data provided a focus for the subsequent conference on each observation. In turn, the conference data served to validate, clarify or modify propositional statements arising from observational data and to guide the next phase of observation and conference data collection. A more detailed explanation of the data analysis procedures employed is provided in the next section of this chapter. A third characteristic of ethnographic research is the collection of a variety of data in order to provide as holistic a description of the setting as possible. Wilson (1977) suggests the following as relevant data: 1. form and content of verbal interaction between participants 2. form and content of verbal interaction with the researcher 3. non-verbal behaviour 4. patterns of action and non-action 5. traces, archival records, artifacts, documents (Wilson 1977:255) These data are collected through multiple techniques primarily involving observation and interviewing. However, researcher's field notes and documents collected also provide significant information for the ethnographer as she tries to "experience the world as members of the culture in question do" (Guthrie & Hall 1983:3). Multiple data collection techniques were employed In this present study. Primary data were collected by means of classroom observations, post-observation conferences and final interviews. Additional data were provided by field notes, documents and classroom materials. Burgess argues that this multiple data collection approach (triangulation) enables the researcher to address a frequent criticism of field research ~ the problem of Internal accuracy ("How far does the researcher's presence influence the generation of data?") and external accuracy (Can the data that are obtained in one situation be generalized to another?") (Burgess 1984:145). In reference to establishing external validity, Sevigny maintains that triangulation "offers stronger potential for generalization through built-in mechanisms which rule out rival hypotheses" (Sevigny 1981:73). Through observation, the researcher "attempts to immerse himself in the daily lives and activities of the people being observed" (Guthrie & Hall 1983:9). Ethnographic observation in the educational setting Is focused on the school and/or classroom. In this case, the researcher might adopt one or a combination of four possible roles (Gold 1958). These include: 1) the complete participant, 2) the participant-as-observer, 3) the observer-as-participant, and 4) the complete observer (Burgess 1984:80). The researcher might maintain one of these roles throughout the study, or may adopt varying roles in different phases (Burgess). It is also important to note that the researcher might be "simultaneously perceived in different terms by different members of the institution" (Burgess 1984:88). A more focused research perspective adopted by many ethnographers conducting studies in the educational setting is referred to as microethnography (Erickson 1976)."* Microethnographies differ from the more traditional ethnographies (macroethnographies)in that they are concerned with the social interaction of a smaller group (a class or a group in a class). One concern raised about this narrow focus is the potential exclusion of significant, broader issues pertaining to the wider context (e.g. the school community) which may "account for much of what goes on (or doesn't go on) in schooling" (Lutz 1981:53). Lutz concedes that a great deal can be learned by studying face-to-face interaction in the classroom; however, he argues that these events and behaviours cannot be thoroughly understood unless they are viewed in light of the broader context in which they are embedded (pp. 53-54). In the present study, the institutional contexts were familiar to the researcher, A different perspective from Lutz's is offered by Guthrie and Hall (1983) who consider the controversy concerning micro or macro ethnography promotes a false dichotomy. For Guthrie and Hall, microethnography constitutes one methodological stage in the broader study. This stage with its fine-tuned focus on social interaction allows for a more in-depth examination of classroom life. One particular feature of the microethnographic focus is the use of audio and visual recordings to provide a permanent record of observational data. There are two major advantages in using this technology. First, recording allows for repeated viewing and analyses of the data "from a variety of attentional foci This approach has also been referred to as constitutive ethnography (Mehan 1981) and ethnographic monitoring (Hymes 1981). '5 The researcher has taught in two of the institutions in this study and is regularly involved on a professional level with various faculty from all three. and analytic perspectives" (Erickson 1986:145). Second, it enables the researcher to review data and reflect on her interpretation of them after leaving the field rather that trying to capture everything during the observational period. By not having to note all behaviours during observation, the ethnographer has time to write down more reflective comments about what s/he is observing. In addition, information about what Is happening 'off camera' can be noted so that during analysis this can inform Interpretation of behaviours. In the present study, observational data were video recorded and a running account of the lesson was written. This account included both reflective comments and contextual information. The analysis phase of the observational data, as previously mentioned, is a cyclical procedure In that analysis of data generates hypotheses to be validated through future observations and conversations with the participants. A classification system "either developed for the study or borrowed from previous ones" (Guthrie & Hall 1983:15) is used for data analysis. This system may focus on verbal or non-verbal interactions or a combination of the two. Ethnographers criticize the use of quantification schemes during observation because of the limitations they impose on the observation focus. Mehan (1981) for example, argues that by using quantification schemes during observation, only the frequency of certain, pre-determined t)ehaviours are captured. As a result, the researcher may lose "critical verbal and non-verbal information as well as the sequential flow of events" (in Guthrie & Hall 1983:20). In contrast, by recording classroom observations and using a classification scheme after to organize the data, the researcher is able to make decisions about what Is significant t>ased on the complete corpus of observational data. Therefore, a classification scheme which is informed by the data can be developed during the analysis. The approach employed In the present study was to use a classification system developed for L2 classroom research (COLT, 1984) but to modify it so that it was more reflective of the focus of the study and the data collected. Besides observation, interviewing is a primary data collection technique in ethnographic research. This is important as It provides the researcher with a means of clarifying, elaborating and/or validating hypotheses generated from the observation data. The participants (or key informants) provide the necessary emic perspective of what is observed by the researcher in the setting. Interaction with the key informant can occur informally such as when the researcher "sees something interesting or peculiar ... (and) immediately question(s) the informant about it" (Guthrie & Hall 1983:11) or more formally through scheduled interviews. These interviews can either be structured (with set questions so that data are more easily quantified and compared) or unstructured (which involves a more open-ended format). The less formalized approach, which Burgess (1984) refers to as "conversations with a purpose" (p. 102), enables the researcher to explore topics emerging during the interview itself. Although the unstructured interview is guided by themes and topics that the researcher wishes to pursue, it is largely controlled by the key informant's input. Both post-observation conferences and final interviews provided this type of data in the present study. Field notes also provide important data in an ethnographic study. These notes include not only an accounting of what happened during observations and interactions with the key informants, but a "record of ...(the researcher's) own feelings, preconceptions and future plans" (Bogden & Taylor 1975:66). These reflective comments often direct the researcher to Issues that need to be explored further. Miles and Huberman (1984:65) suggest the following as useful reflective data: 1. what the relationship with the respondent was Uke; 2. second thoughts on the meaning of what a respondent is saying; 3. doubts about the quality of data being recorded; 4. a new hypothesis explaining what was happening; 5. a mental note to pursue an issue further in another part of the data; 6. cross-allusions to something in another part of the data; 7. own feelings about what was said and done; and 8. elaboration or clarification of something said or done. In the present study, field notes focused on questions to pursue in the next contact with the teacher and in the ongoing data analysis. These were usually concerned with inconsistencies noted between teacher comments and classroom behaviours. In addition, notes were written about themes and patterns of behaviour emerging from the analysis of observational and conference data. These notes were useful in focusing subsequent observations and conferences with each teacher as well as contributing to the direction of data collection and analysis foci for all nine teachers. One final point to address before turning to the specific data collection and analysis techniques employed in this study, concerns the process of generating assertions during data analysis. This involves a twofold task: first, all the data collected (through observations, interviewing, field notes and documents) are searched in order to Identify patterns and generate assertions; second, the validity of these is examined by again searching for confirming or disconfirming evidence. If there are more disconfirming instances, then the assertion cannot be considered valid. Even if there is more confirming evidence so the assertion Is validated, the discrepant cases "should be noted for subsequent analysis" (Erickson 1986:146). In searching the data, the researcher should be looking for "key linkages" (p. 148) which indicate patterns of behaviour. According to Erickson (1986), the objective of this pattern analysis "is to discover and test those linkages that make the largest number of connections to items in the corpus" (p. 148). The more Instances of the behaviour pattern found across all the data, the more confident the researcher can be in claiming an assertion to be valid. This section of the chapter has presented a general overview of a qualitative research perspective and ethnographic methodology in particular. It has also alluded to the data collection and analysis techniques used in this study. What follows Is a detailed description of the subjects and sites involved and the specific methods of data collection and analysis employed. B: DATA COLLECTION PROCEDURES 1. Data Sample Nine ESL teachers in three adult education institutions were involved in this study. The teachers were selected through the process of judgment sampling. This involved selection "according to the number of criteria established by the researcher ... that endows them with specialized knowledge" (Burgess 1984:55). For the sample in this study, criteria included a minimum of three years experience teaching adult learners, prior experience teaching in the present Institution and prior experience in teaching courses which focus on developing speaking and listening skills. There was no attempt to 'screen' teachers according to methodological preferences for teaching speaking/listening skills; for example, teachers were not selected on the basis of expressing preference for a communicative language teaching approach. Rather, a criterion for selection was that the course each was teaching was described by the institution as one which promoted speaking and listening skill development. The basis for this sampling focus was the researcher's interest in how individual teachers 'operationalized' courses with similar skill development objectives.!^ This allowed for a comparative analysis of individual approaches to teaching speaking and listening across all nine teachers. Secondly, it was not a focus of the study to evaluate how strictly individual teachers adhered to practices identified with a specific approach (for example, the communicative approach). Rather, the aim of the study was to examine Individual teacher's decisions for classroom practices and how these related to teacher beliefs concerning 12 instruction and perceptions of the instructional context. Selecting three instructors from each institution enabled the researcher to examine how Individual perceptions of institutional features (e.g., type of students, administrative expectations) on planning and teaching might differ for teachers working in the same setting. This provided a basis of comparison in terms of contextual variables within and across the three institutions. The three institutions involved in the study were also selected on the basis of particular criteria. First, each one is a publicly funded, adult education institution In a metropolitan setting. Second, each one has a well-established E S L programme. Third, each E S L programme in the three Institutions offers courses designed to promote speaking and listening skill development. It should be noted at this point that only one of the institutions in the sample offers courses specifically focused on development of speaking and listening skills. The courses In the other two Institutions provide a more integrated approach In that all four language skills (speaking, listening, reading and writing) are included Woods' L 2 study (1989) was also CMicerned with how teachers individually operationaUze course objectives which have been institutionally prescribed; however, the teachers in Woods' study appear to have been given pre-planned curricula by the institudon. In this present study, teachers were given a set of objectives but developed their own curricula for the courses they were teaching. in the course description. Although this presented a potential limitation in terms of comparative analysis, in the classes observed in these two institutions, reading and writing activities played a very minor role. They were mainly used as stimuli for speaking/listening activities rather than being a focus of skill development in themselves. Initial access to these three sites involved contacting the directors of the selected E S L programmes by letter and a subsequent phone call. This initial contact provided the directors with a brief overview of the purpose of the research and requested permission to conduct the study in that institution. Following this, two institutions invited the researcher to attend a faculty meeting and explain the research project to the instructors. At this meeting, instructors who were interested in participating in the research informed the researcher. Teachers were then selected from this group based on the criteria for sample selection mentioned earlier. In the third institution, the director of the E S L programme decided to explain the research project to the faculty herself. Unfortunately, out of this initial request to faculty, no teachers expressed readiness to participate in the project at that time. This necessitated the researcher personally contacting instructors at this institution after talking to the director and receiving her permission to do this. Through these personal contacts, three teachers who satisfied the selection criteria and who were interested in participating in the project were chosen. In addition, approval was requested and received from the research "Ethics" Committee at the University of British Columbia to carry out this research. Having provided this general description of the instructors and institutions, the criteria used in selecting these and how the researcher initially approached the sites, a more specific description of each site and teacher participant follows. ^ ' This university ethics committee is formally known as the Behavioural Sciences Screening Committee for Research and Other Studies Involving Human Subjects. See Appendix 2 for insitutional course descriptions and course objectives. 2. Description of Sites and Participating Teacliers Site A: English Bay Community College English Bay Community College was established through local plebiscite in 1965. This institution has three separate campuses; the site involved in this study is the major one for E S L programmes. At this site there are 2,500 (including part-time and full-time) E S L students enrolled in three English language training programmes. These include courses in general E S L instruction (both on and off campus), English for Academic Purposes and E S L & Vocational training. The three teachers involved in the study from this site are instructors in the general language instruction programme which offer courses at beginner, intermediate and advanced levels. The Academic Preparation Programme was not selected for two reasons. First, this study was concerned with students at an intermediate level of language proficiency and only students assessed at an advanced level are eligible for the Academic Programme. Second, these courses focus on reading and writing skills not speaking and listening. Teacher 1: CLAIRE At the time of the study, Claire had been teaching part-time at English Bay College since 1986. She has since moved to a different institution where she has a full-time position. Claire has been teaching E S L for seven years and has had a variety of teaching experiences. These have included teaching immersion E S L courses in the summer (in the United States), tutoring individual students at all levels, teaching academic reading & writing skills to advanced students and teaching integrated skill courses to Intermediate students. Claire's academic and teacher training background include university E S L teacher training (In the United States) and a Masters degree in Education (TESL) from a university in Western Canada. Actual names of the institutions and the teachers participating in this study have been changed. Teacher 2: MARK Mark has been teaching E S L for ten years. This experience has included teaching all levels of students (beginners, intermediate and advanced) at several different Institutions. He has been teaching at this particular institution since 1987. Before teaching ESL , Mark was involved in monitoring in French immersion classes while attending university In Eastern Canada. This stimulated an Interest in second language instruction and upon returning to the West, he began a university teacher training programme in T E S L (Teaching English as a Second Language). He Is currently completing an instructors' diploma programme sponsored by the Provincial Government and designed for Individuals wishing to teach at the college level. Teacher 3: STAN Stan has been teaching E S L for eighteen years and has been teaching mainly In the general English programme at English Bay Community College since 1975. He has also taught in various other institutions in the city as well as abroad in Japan. In addition to teaching, this instructor has been involved In writing a Ministry curriculum guide for E S L adult learners. This teacher's teacher training programme includes a Secondary School certificate (History) and undergraduate university T E S L training courses. He has also completed an Instructor's Diploma for college instructors. Site B: Pacific Language Studies Centre This centre was established in 1969. The Centre offers five different types of E S L programmes including Communications and Academic, as well as special programmes designed for specific groups of students who usually attend In the summer term. Courses at all levels are offered in both the Communications and Academic divisions, although there is a threshold entry criteria for the academic programme. Students' eligibility for this is assessed by the Centre. Instructors participating In the study from this site were teaching in the Communications Programme. The Academic Programme was not chosen for this study as the course content Is specifically focused on speaking and listening skills for the academic setting (such as listening to academic lectures and participating In seminar discussions). By selecting courses In the Communications Programme with Its more general skill development focus, a more comparative basis was possible across the three Institutions. All the students In this programme attend classes focusing on general communication skills in the morning but can choose from a number of elective courses in the afternoon. Teacher 4: HELEN Helen has been teaching E S L for seven years and has been at the Centre since 1986. Before teaching, she had experience working with E S L students both through monitoring and 'subbing' while completing her teacher training. She has taught at various locations in this city as well as for a year In Japan. At the Centre she has taught in both the Communications and Academic Divisions. Helen's E S L training background includes a teacher preparation course in T E S L at the university level and an intensive teacher training course at a local institution. In addition, she is currently working towards a Masters of Education degree in T E S L at a university in Western Canada. Teacher 5: EDIE Edie has been teaching E S L since 1981. Her E S L background includes teaching academic preparation and conversation classes to both secondary and adult students In Canada and abroad (Japan and Taiwan). At the time of the study, she had fc>een working at the Centre for eighteen months. She has since left and Is now teaching ESL students at the secondary level. This teacher has a secondary school teaching certificate and has completed her Masters of Education in T E S L from a university in Western Canada. She has also worked with T E S L student teachers in local institutions supervising their E S L practicums. Teacher 6: FRAN Fran has been teaching E S L since 1983 and at the Centre since 1987. She has taught a variety of E S L courses at both adult and secondary levels in several locations (including abroad in Japan and Indonesia). Her experience at the Centre has included teaching in both the Communications and Academic divisions. Prior to teaching E S L she was involved in E S L on a volunteer basis. Fran's E S L training includes a T E S L teacher preparation course at the university level and an intensive training programme at a local institute. She is currently completing her Masters of Education in T E S L at a university in Western Canada. Site C: South Centre Community College South Centre Community College was established in 1970 and currently there are approximately 7,500 students enrolled in courses offered on two campuses. This college offers Diploma and Certificate programmes in academic studies as well as non-credit programmes in the community programmes area. The E S L division provides non-credit courses for both community and international students. These courses are designed for intermediate to advanced level students with the most advanced courses focusing on college/university preparatory skill development. Courses are offered at four levels (lower intermediate, upper intermediate, advanced, college preparatory) and each course focuses on particular language skills (speaking/listening, reading, writing). While at the time of this study, speaking and listening skills were both addressed in a single course, these skill areas have since been separated administratively into two courses. The majority of students (both international and community) enrolled in the E S L programme intend to take university transfer credit courses at the college upon completion of their E S L coursework. Teacher 7: ANNIE Annie l ias been teaching E S L for twelve years. She has also had a variety of other teaching experiences: French as a Second Language at the elementary level; English for Special Purposes with adult business and vocational students, and English for Academic Purposes at the adult level. Prior to teaching ESL , she was a monitor in an E S L programme while completing her undergraduate university degree. Annie's educational background Includes a B.ED, with a T E S L major and a recently completed Masters of Education in T E S L from a university in Western Canada. Teacher 8: SANDY Sandy has a varied background of teaching experiences. She has taught English and History at the secondary level, social studies in Adult Basic Education, worked in the k-12 system with a Moms and Tots programme, and taught in French immersion summer programmes. Sandy has also been very involved with immigrant services groups In her local community. She has been teaching E S L to adult learners for the last fourteen years. This instructor's educational background includes a secondary teaching certificate (English & History) and a Masters in Adult Education. She is currently completing a certificate programme In T E S L at a university in Western Canada. Teacher 9: PAM Pam has been teaching E S L since 1978 and has been at South Centre Community College since 1983. Prior to teaching E S L , she taught Language Arts, Math, and Drama at both the elementary and high school levels and Language Arts in Adult Basic Education programmes. Pam's teacher training background includes a Secondary School certificate and undergraduate university T E S L coursework. She is currently completing a Masters of Education in T E S L at a university in western Canada. Summary The three public institutions involved in this study provide the major E S L programmes for adult E S L learners at a post-secondary level in this metropolitan area. Each of these three institutions offers courses in speaking, listening, reading and writing skill development at various levels. Secondly, as the preceding teacher descriptions indicate, the nine instructors participating in the study have a wide variety of experiences with second language students. This has included not only teaching E S L (experience ranges from 7 to 18 years) but also backgrounds in monitoring and volunteer work (in both French immersion and E S L programmes). In addition, many of the teachers have taught E S L abroad as well as other subjects in the school system. The majority of the participating teachers have T E S L training background at the university level; many have also taken additional intensive courses at institutions offering more classroom-based training programmes in TESL . In addition, six of the nine teachers have completed or are in the process of completing a Masters of Education in T E S L . One of the teachers has a Masters in Adult Education. Given the length and variety of E S L teaching experiences and the academic qualifications they hold, this is a highly qualified group of teachers and well above the base criteria set for the sample selection in this study. 3. Program Selection The E S L courses taught by the instructors participating in this study were all at the intermediate level and emphasized the development of students' speaking and listening skills. While at one of the institutions (South Centre College), the speaking/listening course is taught separately from reading and writing, at the other two sites (Pacific Language Centre and English Bay College), a more integrated skill approach is evident in the institutional course descriptions. However, as previously mentioned, in the classes observed at these two sites, activities emphasized speaking and listening with the other two Skills serving mainly to provide a stimulus for oral activities. Activities involving reading and writing were used in a similar way in the classes observed at South Centre Community College where the courses were described as focusing on only speaking and listening. Therefore, although the institutional organization of these courses was not the same for all three sites, in actual practice, there were more similarities than differences. An important sample selection decision concerned the choice of focusing on courses in the more general E S L programmes at Pacific Language Centre and English Bay College as opposed to the Academic Programmes. Initially it had been decided to focus on courses in the academic preparation programmes offered by the three institutions. However, this became problematic In terms of how these institutions organize their academic programmes. While Pacific Language Centre and English Bay College offer academic courses in separate divisions from the more general language stream, all courses at South Centre College are considered part of an academic preparation programme. Therefore at South Centre students enrolled In these courses are Intermediate to advanced E S L learners. On the other hand, only advanced level students can enrol In this programme at Pacific Language Centre and English Bay College. It had already been decided that teachers working with advanced level students would not t>e implementing the types of speaking/listening tasks that this study was particularly interested in examining. This is because students at this advanced level are already quite fluent orally and therefore, speaking/listening activities tend to be more concerned with developing specific skills required for the academic setting (such as seminar participation skills) rather than oral skills per se. The Intermediate level was selected because students at this stage are still working on fluency development and so classes at this level would provide richer data on the planning and Implementation of communicatively focused tasks. As a result, intermediate level courses In the general programmes at Pacific Language Centre and English Bay College, and intermediate level courses at South Centre College were selected for the study. A second factor Influencing the selection of intermediate classes concerned the extent to which instructors teaching at this level have freedom in determining objectives and planning curricula. At the intermediate level, these three institutions tend to allow their instructors greater freedom in specific planning choices than at the more specialised, college preparatory level. In the advanced academic courses there are specific, academically focused goals for teachers to meet and often required texts to use. On the other hand, the intermediate level courses allow teachers more decision-making latitude in the planning of courses. While there are institutionally established course objectives for the courses taught, these are very general and serve mainly as guidelines and bases of articulation between the various levels. Therefore, teachers appear to have much more individual choice in setting specific course objectives, planning curricula, choosing materials and deciding on classroom activities. In other words, there seems to be less institutional 'interference' in planning and implementation decisions. This was an important consideration in this study which was concerned with teacher decision making. It was also felt to be a significant factor in providing a comparative basis for data analysis across the nine teachers in three different sites while still providing a basis for assessing contextual factors. 4. Description of Specific Data Collection Procedures Classroom observations of each lesson, post-observation conferences and final interviews provided the primary data for this research. In addition, field notes and documents contributed to the analysis of the primary data. Observations: Each of the nine teachers was observed teaching four consecutive two hour classes. As previously mentioned, these classes were video-taped and notes were written during each observation. The notes included a timed, running account of what was happening in the lessons and information regarding where students were sitting for various activities and the movements of the teacher, particularly during small group work. In addition, reflective comments and questions were written concerning specific aspects of the lesson. These highlighted points to ask the teacher about in the subsequent post-observation conference. During these observations the researcher assumed the role of complete observer although at times she was invited by teachers to comment on a point being discussed in the lesson. Apart from these brief, invited involvements, the researcher remained quietly in a back corner of the classroom for the entire lesson. The students in these classes did not seem to be concerned at all by the presence of the researcher or the video recorder. The study was explained to the students before the observations began and they all had signed forms giving the researcher their permission to come into the four classes (See Appendix 1 for student and teacher permission forms). Students were also invited to view the video tapes once the observations were completed. In those classes where students had expressed interest in this, the researcher returned after the data had been collected and showed them the video tapes. Several of the teachers were interested in having copies of the videos of their classes. These were provided for them if requested. Post-observation conferences: A conference was scheduled for as soon as possible following each class observation. This was frequently the most difficult aspect of the data collection process as teachers had busy teaching schedules and when not actually teaching, needed that time to prepare lessons and mark. In addition it was often difficult to coordinate a time slot which fit the teacher's schedule and the availability of a video playback machine and a free classroom for viewing and conferencing. However, in each case, it was possible to arrange the post-observation conference within a day or two of the class observation to be discussed. The post-observation conferences lasted for approximately two hours. During these conferences, which were audio-taped, the teachers watched the video of the lesson and commented on what was happening. These comments usually related to the following: changes the teacher had made to the lesson 'plan' mid-stream and why; observations of individual student actions which were often accompanied by some background Information about these students; observations about their own teaching behaviours and Instructional decisions; and statements about instructional changes they felt they needed to make based on what they were seeing on the video. Task Grid: Teachers were also asked to analyse the task structure of each lesson in the post-observation conference. A task grid was developed for this purpose. This grid focused on two dimensions which emerged from the second language research and methodological literature review as central to an analysis of task implementation ~ participant structure and language focus (accuracy/fluency). These grids (see Figure 4) required teachers to indicate two features of their tasks. First, they identified whether the task focused mainly on accuracy or fluency language learning objectives. Secondly, they marked whether they considered the task to be teacher-centered or student-centered. Task Grid =3 i_ W c o Q. Q. Accuracy Teacher Centered Fluency Student-centered task language focus Figure 5 The vertical axis of the task grid identifies the participation structure used for task implementation. The horizontal axis identifies whether accuracy of language use or fluency of language use was the focus of the task. These were considered significant task descriptors for several reasons. First, both regular and L2 classroom-based studies have found that participation structure and tasks are inter-related features of classrooms (viz., Erickson 1976; Harker & Green 1981; Pica & Doughty 1985b;Pica et al. 1990). Second, research findings have shown that teacher-centered, whole group Instruction tends to dominate both LI and 12 instruction (viz.. Philips 1972; Edwards & Furlong 1978; Long et.al.1976;). Third, L2 researchers examining classroom organization in relation to student language use have found that the type of student language use is influenced by the participation structure.(e.g. P ica & Doughty 1985b) Studies have shown that while the use of unsupervised student-centered, small group organization Is likely to promote a communicative language use focus (defined in this study as fluency-focused), teacher-centered,whole class organization tends to result In an emphasis on accuracy of language use. However, what L2 research has not investigated is whether teachers associate teacher-centered instruction with accuracy-focused language learning objectives and student-centered tasks with fluency-focused language learning objectives. That is, do teachers determine participation structures for tasks on the basis of the language learning focus they wish to work on? Secondly, L2 research has not investigated the impact on language use when teachers are supervising small group tasks. L2 methodologists argue that small group tasks should focus on communication not accuracy and that teachers should facilitate this communicative Interaction, not correct language errors (e.g., Rivers 1987). However, classroom research has not investigated If this is actually the role teachers adopt during small group work. In other words, if teachers interact with students (as opposed to maintaining a silent monitoring role) during small group tasks, do they focus on students' accuracy of language use or engage the student in a communicatively-focused exchange? By asking teachers to complete task grids, data were gathered on what they perceive their language interaction focus to be in teacher-centered and student-centered tasks. In addition, by encouraging teachers to 'talk-aloud' while completing these task grids, valuable data were collected concerning teacher perceptions of accuracy/fluency and the role each plays in their instructional approach. Teachers, not the researcher, determined what constituted task boundaries. It was therefore their decision as to how many task grids to complete for each lesson. Comments teachers made v&\\\e determining task boundaries and the participation structure/language focus of each task provided rich data concerning how they conceptualized tasks. This process also stimulated teacher comments on the discrepancies they observed between their task planning and what actually happened during task implementation and more general statements about their approach to second language instruction, their students, planning and classroom techniques. The researcher believes that without the task grid component being included in the conferences, the collection of such rich data would not have been possible. In addition, this teacher-directed process enabled the researcher to remain relatively passive as opposed to directly prompting the teacher to comment about their teaching and planning practices. Teachers spontaneously verbalized about these issues while completing the task grids. Interviews : After all the observations and post-observation conferences had been completed for the nine teachers in the study, each teacher was interviewed. The average length of time for each interview was approximately 90 minutes, although several lasted close to three hours. The interviews were structured in that there were set questions asked of each instructor. This allowed for a comparative analysis of the data across all nine teachers. These questions were developed over the observation and conference data collection period so that these data as well as input from L2 theory informed the interview schedule. However, although there were set questions guiding the interview, these stimulated many secondary questions and lengthy discussions of particular points. These enabled the researcher to probe for more in-depth infonnatlon and clarification of certain points noted in the observational and conference data. (See Appendix 2 for the interview schedule). The interview questions focused on six main topics. The first topic was concerned with the teachers' educational and teaching experience backgrounds. Teachers were asked about their teacher training, previous teaching experiences, present teaching position and any involvement they might have had with second language learners on a non-teaching basis. The second topic explored teachers' views and beliefs about teaching E S L . These questions were concerned with what factors they believed to be important for second language learning, what they felt was the most effective approach for teaching speaking and listening skills in the classroom context, and if/how their views had changed as a result of experience in teaching. In addition, teachers were asked to 'define' certain terms (e.g., accuracy, fluency, task) which are constantly referred to in the 12 literature. This section of the interview also addressed teachers' views of their Individual instructional settings. Questions were asked about their perceptions of the level of Institutional Involvement and the Impact this involvement had on their planning and teaching. In addition, teachers were asked what they believed to be the impact of student characteristics and classroom factors on their planning and teaching practices. The third topic was concerned with teacher planning. This topic included questions concerning the extent to which they planned units and lessons, what they Included in these plans and what factors they considered while planning. One section of this topic focused on planning small group activities in particular. The fourth topic addressed teachers' classroom practices. Specific questions included what teachers believed their role to be in both whole class and small group activities; the language learning focus they associated with particular task types; the type of feedback they felt was important to provide to students and how this might be influenced by the task and the participation structure; and what factors they considered in deciding on participation structures for various tasks. The final, fifth topic addressed teachers' views of participation structures specifically. These questions were concerned with what percentage of class time teachers believed they spent in whole class, and small group or pair organizational formats. In addition, teachers were asked what factors they considered when selecting the participation structures for various activities. Interviewing all nine teachers after the completion of the collection and initial analysis of the observation and conference data, allowed the researcher to probe issues emerging from both types of data not only In relation to Individual teachers but across the whole sample. This was particularly important in the latter case since as the study progressed, patterns of behaviour and themes common to all the participants became more evident and these therefore could be explored in the final interviews. In addition, propositions generated by the data from the last group of teachers could be checked with the first group in the final interviews. The interview stage also provided the opportunity to collect detailed information about the teachers' educational and teaching experience and more significantly, teacher comments about their individual teaching approach and planning/instructional decisions in a context that was not specifically linked to a particular lesson. Although this type of data might have been collected in an initial interview, by waiting until the final stage of the study, the researcher had a clearer understanding of the individual teacher's views and practices and a more purposeful exploration of teachers' theoretical and practical knowledge was possible in the interviews. It was also felt that a full interview conducted at the beginning of the study might affect the observational and conference data collection in that teachers would be more alerted to the particular constructs the study was interested in examining. Documents: Both institutional and teacher documents were collected. Institutional documents included descriptions of the specific courses involved in the study and of the institution itself (See Appendix 3). More detailed information such as student enrolment figures was obtained through phone contact with the Registrar's Office at the individual sites. Teacher documents included copies of materials (including text and teacher-made materials) used in the lessons observed, as well as any unit and lesson plans written for these lessons. Field Notes: Throughout the study, field notes were written. These focused on items that the researcher wanted to pursue in conferences with individual teachers, in the ongoing data analysis and in the final interview. Other notes dealt with researcher reflections during the different phases of the study and served as input for focusing data collection and analysis. C. DATA ANALYSIS PROCEDURES The analysis of the data began after the first observation had been completed and continued throughout the data collection period. In this way, data were concurrently collected and analysed. This ongoing process enabled the researcher to identify themes and patterns of behaviour emerging from both classroom practices and teacher comments. This concurrent collecting and analysis process also served as a check that sufficient and appropriately focused information was being gathered before completion of the field work (Bogdan & Taylor 1975). Initial analysis of the recorded observation data involved searching through the data to obtain an overall understanding of the lesson framework. Particular attention was paid to the tasks and participation structures teachers implemented and the teacher roles adopted. During these first searches through the data, notes were made about specific points to pursue with the teacher in the conference about the lesson and in the subsequent, more in-depth analysis of the data. After the initial process of becoming familiar with the data, a more thorough analysis was undertaken. In order to provide a descriptive framework for analysing the lessons, a classification system was used. This was a modified version of the C O L T (Communicative Orientation to Language Teaching) system developed by a group of researchers for a study investigating French instruction at the university level (Allen et.al.1984). This particular system was selected as an initial organisation framework because it provided a broader set of categories than other L2 observation systems which were more focused on classroom discourse. The C O L T system includes categories for participation structures, materials used and the content focus in lessons. Since all of these aspects of lessons were important to the present study, this system provided a useful starting point for classifying the observation data. As the analysis proceeded, modifications were made and categories added in order to reflect more appropriately, the nature of the lessons observed and the focus of the research. In addition, the initial single classification scheme See Appendix 4 for die COLT Observation System (Allen, Frohlich & Spada 1984). was re-organized into two separate cliarts: 1) a detailed analysis of teacher-student interaction for each lesson task and 2) one for a summary of tasks for each lesson activity. These charts were used in the initial in-depth analysis stage to classify the observational data. In the second stage of the analysis, observational data were compared with information gathered through post-observation conferences and interviews. In order to reduce threats to internal reliability of the coding of the data, several techniques were employed. First, the classification schemes used to analyse the lessons contained a number of low-inference descriptors (time and participation structure, language skills, input materials). The coding of the data into the other categories which were more open to researcher interpretation (language focus, accuracy focus and task control), was validated in the conferences following each observation. In addition, since each lesson was video-taped, there was a visual and verbatim account of what had occurred. By showing these tapes to the participant teacher, the researcher was able to not only corroborate what had been observed and recorded but to validate researcher interpretations of what had occurred with the participants (McMillan & Schumacher 1989:191). This validation was provided both through teacher comments on the lessons but also through the completion of task analysis charts by each teacher. The task analysis charts directly focused on the teacher participant's analysis of the task structure of each lesson and the participant structure and language focus (accuracy/fluency) of each task. 1. First Stage of Analysis: Lesson Tasks Interaction Segments: The first analysis chart - Task Analysis:Interaction Segments -provided a description of each lesson task. The categories included in this analysis chart were derived from the C O L T system (1984) and included the length of time for each interaction segment (column 1), the type of participation structure (column 2) and details of teacher-student interaction (columns 3-7). For the coding of the data into the first two categories (time and participation structure), whole group instruction was considered as a single interaction segment. During student-centered tasks, each time the teacher moved to a different group or pair was coded as a new interaction segment. Within each interaction segment, the activity type was coded as either monitoring only {when there was no interaction between the teacher and students), or in the case of interaction occurring, whether the teacher or a student initiated the interaction. The names of the students involved in the interaction segments were also noted in the explanation column. Task Analysis: Interaction Segments Teacher: Activity: -Lesson: Task: — 1 Time Participation T <--> S : Language ' Accuracy Structure ' Focus ; Focus Focus 6 7 Feedback ; Explanation W S G P M T - S S - T A F G P V. I key Participation Structure W = whole group S G = small group P = pairs Language Focus A = accuracy P = fluency Activity Focus M = monitors T-S = teacher-student S-T = student-teacher Accuracy Focus G = grammar P = pronunciation V = vocabulary I = information Figure 6 In the Language Focus column (4) the interaction was noted as either accuracy-focused or fluency-focused. An accuracy focus was defined as one which predominantly dealt with grammar, pronunciation or vocabulary. Fluency-focused interaction involved a more communicative exchange (e.g. "Who do you think is the best person for the job?") in which the teacher and student were concerned with exchanging ideas or opinions. In addition to these language use foci, another category was added to reflect interaction concerned with task management; that is, interactions which focused on the implementation of the task. These included teacher comments such as reminding students to complete a worksheet; teacher checking to see if individual students or groups were finished; students asking for re-explanation of the task directions. The fifth column detailed accuracy-focused interactions in terms of specific focus on grammar, pronunciation and/or vocabulary. An additional descriptor, accuracy of information, was added to this category in order to code teacher-student exchanges which focused on whether students had heard the correct content information (for example, key details in a taped conversation). In these cases, the information was known by the teacher so it was not a communicative, information-seeking exchange. In addition, the concern was not specifically directed at linguistic accuracy but accuracy of information. The sixth column identified the type of teacher feedback provided in each teacher-student exchange such as rephrasing or repetition of student answer, correction or praise. The final column in this chart. Explanation, provided space for detailed information on each exchange. These notes were descriptive and included information about the actual comments made or questions asked. In addition, particular aspects of the interaction segment, considered useful for analysis, were noted. These included such observations as teacher or student actions (e.g. teacher distributes a worksheet), contextual details (e.g. level of noise during a segment), actions indicating teacher/student affective state (e.g. tone of voice indicating mood) and pedagogical aspects (e.g. teacher use of multiple questions). Task Analysis- Activity Level: Once the interaction segments within each task had been analysed using the above system, the second analysis chart was used to provide a summary analysis of all the tasks implemented for each activity in the lesson. All tasks which were focused on the same topic (e.g., review of adjective clauses or description of an object) were coded as one activity. For example, one activity might be concerned with a problem-solving situation. The teacher could break this down into several stages (or tasks) each involving a different participation structure, language skill focus and specific objective(s). These stages could include first having students listen to and take notes on a taped conversation providing the information needed to complete the problem-solving situation (the final task). Second, the students, in pairs, compare notes and share the information they heard. Next, in a whole group discussion, the teacher makes sure the students have all the necessary information from the tape. Finally, using the information, students in small groups discuss the problem and decide on possible solutions. In each of the above stages, the content focus remains the same (therefore one activity) but the participation structures, specific objectives and student actions change several times. The above example would be coded as one activity involving four different tasks. Task Analysis: Activity Level Teacher: Lesson: Activity: task# time participation language^ ^ task content structures skills input materials control language focus w: t-c/s-c s/l/r/w/ a/v/w L1 L 2 A H S N A I F gp/pr/i key: Participation Structure Task Content w = whole group materials language focus gp = snnall group p = pairs i = individual A = authentic A = accuracy F = fluency 1 = information s-c = student-centered t-c = teacher-centered input A = audio Language Skills s = speaking V = visual W = written 1 = listening r = reading w = writing Figure 7 The first column of the Activity Analysis chart noted the length of time and the number of the task. The second column described the type of participation structure (whole class, small groups, pairs) and whether the task was student-centered or teacher-centered. The second broad category analysed the task content. First it was noted what language skills (listening.speaking,reading,writing) were involved and which one(s) were primary. Second, the materials used were described. This description included the following: the medium (audio, visual, written text); the source of the materials (developed for LI instruction, developed for 12 instruction or authentic such as a newspaper article); the extent of teacher or text control for task implementation (high, semi, none); and the language focus (accuracy, fluency,accuracy of information). In the final column, the specific objectives of the task were noted; for example, students listen to the tape and take notes using a worksheet that provides information categories. This Activity Analysis chart thus provided a summary of the activities included in each lesson as well as the participation structures and content of tasks implemented within each activity. 2. Second Stage of Analysis Comparison of Tasks across Four Lessons: The next stage of the analysis involved examining the task structure across the four lessons observed. This was analysed for both consistencies/inconsistencies and suggested patterns of teaching practices. One particular area of research interest was the extent to which a teacher's analysis of the language learning focus of a task was coherent with what the implementation data analysis indicated. For example, a teacher might state that the language learning focus for a task had been primarily concerned with learners' communication of ideas and not their accuracy of language use. Accordingly, the teacher might code the task as fluency focused on the task grid. If the Interaction segment analysis also revealed that the majority of teacher-student Interaction had focused on information-seeking questions and attention to content of utterances (as opposed to linguistic accuracy), this would be considered a consistency between teacher analysis and implementation data analysis. However, If the interaction segment data Indicated that the teacher-student interaction had primarily focused on correct grammatical or pronunciation usage, this would be considered an inconsistency between teacher analysis and Implementation data analysis. Instances of both consistencies and inconsistencies would then be explored further with the teacher participant particularly in relation to teacher stated beliefs about second language instruction and individual teaching practices. The comparative analysis between Task Implementation and Teacher Task Analysis is illustrated in Figure 8. Comparative Task Data Analysis Imolementation Data Teacher Task Analysis participation structure consistencies participation structure language learning focus and language focus interaction language focus (T-S ) inconsistencies task objectives Figure 8 Comparison of Observational and Conference/Interview Data: The consistencies and inconsistencies between task implementation analysis and teacher task analysis of individual lessons were examined in relation to teacher beliefs statements and their perceptions of the contextual factors which impacted on decision-making. Figure 9 illustrates the components compared in this analysis phase. Teacher Perceptions & Beliefs and Task Implementation Task Implementation Teacher Beliefs and Perceptions -task content consistencies -about teaching & -task implementation and learning E S L -teacher task inconsistencies -of the instructional context analysis -of the instructional task Figure 9 Summary of Data Collection and Analysis Procedures: To provide an overall summary of the procedures used in this study, Table 5-1 lists the observations, conferences and interviews that were completed for the nine teachers, and Table 5-2 indicates the relationship between the data collection techniques and data analysis. Table 5-1 Data Collection OBSERVATIONS CONFERENŒS INTERVIEWS TEACHF,RS 1-9 36 36 9 TOTAL T I M E 72 hours 72 hours 18 hours DATA -video tape -anecdotal notes - documents -audio tape -task grids -audio tape -charts (in interview schedule) Table 5-2 Relationship between Data Collection and Analysis F o c u s T Y P E O F D A T A 1. Lesson/Task Analysis Observation Data 2. Comparative Task Analysis: Observation -t- Conference Data Implementation/Teacher 3. Comparative Analysis: Observation Data Implementation /Beliefs & Perceptions Conference -i- Interview Data In conclusion, this chapter has discussed the methodological approach used in the present study and the specific data collection and data analysis procedures employed. The thesis now turns to a discussion of the research findings. Chapter VI, Individual Teacher Findings, presents the findings for each of the nine teachers and the context factors pertaining to each of the three institutions. Chapter VII, Comparative Findings, takes a cross-sectional perspective and discusses the similarities and differences in decision-making which emerged from the data across all nine teachers and the three sites. CHAPTER VI INDIVIDUAL TEACHER FINDINGS Introduction Chapters VI and VII of this thesis reports the findings of the study from two different perspectives. Chapter VI reports on how individual teachers approached the instructional task and the impact of contextual factors on decisions. Chapter VII takes a more global perspective and addresses the similarities and differences in decision-making that emerged for all nine teachers and the consistencies/inconsistencies in relation to L2 theory. An examination of the findings from both these perspectives allows for a more in-depth reporting of the rich data gathered in this study. The first part of the chapter provides background information on 1) individual teacher experience and T E S L training/ academic background (teacher characteristics); and 2) context factors specific to each instructional setting. Individual teacher findings are reported in the second part of the chapter.2i These findings are organized by sites and in the following order: English Bay Community College, Pacific Studies Language Centre and South Centre Community College. In each of these site sections, individual teacher findings are examined in three areas: 1) Approach to Instruction; 2) Perceptions of the Instructional Context; and 3) Instructional Task: Perceptions and Implementation. A. PRE-ACTIVE FACTORS: TEACHER CHARACTERISTICS 1. TESL Experience and Academic Background Practical knowledge researchers (viz.,Elbaz 1983; Clandinin 1986) have argued that both theoretical knowledge and practical knowledge inform" the decision-making process. In order to consider these factors in the present study, data on ESL teaching experience, T E S L training and academic background were gathered. These data are presented in Table 6-1, Teacher Characteristics. 2! Detailed lesson charts and teacher task grids for each instructor are included in Appendix 5. Table 6-1 PRE-ACTIVE FACTORS: Teacher Characteristics experience experience educational in teaching at present background ESL institution in TESL English Bay Claire 7 years 2 years -undergraduate TESL courses -M.A./TESL (completed) Mark 10 years 6 years -undergraduate TESL courses Stan 18 years 16 years -undergraduate TESL courses Pacific Centre Helen 7 years 5 years -undergraduate TESL courses -M.Ed../TESL (currently) Edie 10 years 1.5 years -undergraduate TESL courses -M.Ed../TESL (completed) Fran 8 years 4 years -undergraduate TESL courses -M.Ed../TESL (currently) South Centre Annie 12 years 3 years -undergraduate TESL courses -M.Ed../TESL (completed) Sandy 14 years 7 years -undergraduate TESL courses -[M.A.. history] -TESL certificate (currently) Pam 13.5 years 8 years -undergraduate TESL courses -M.Ed../TESL (currently) TESL Experience: As the preceding table shows, this group of teachers had extensive experience in the E S L classroom. What is not possible to summarize in a table such as this is the richness and variety of background experiences these teachers brought to their present teaching situations. First, the majority had been involved in some type of volunteer work in the second language classroom before beginning their teaching careers. For example, while working on his undergraduate degree in Eastern Canada, Mark volunteered his time in a French Immersion Programme at his university. Also, while taking T E S L training, Helen and Annie spent time helping adult learners in the E S L classroom. Several of these teachers have also been involved in helping New Canadians adjust to life in a new culture. Sandy was particularly active In this area while on the executive of a local community immigrant services society for several years. This Involved her in linking New Canadian families to vital resources in the community and setting up/delivering needed language programmes (e.g., a 'Moms & Tots' programme) in the community. Helen spent volunteer time with a local immigrant services society and helped refugees settle into the local community. These teachers have also had a wide range of teaching experiences. Sandy, Edie and Pam have taught both E S L and a variety of other subjects at the secondary level; Sandy has also worked at the elementary level with E S L learners. Annie had recently taught an undergraduate T E S L course in a university teacher training programme. A number of these teachers had also been sponsors for teachers in training. Most have taught E S L in a variety of settings ranging from local community classes held in church basements to private schools In Japan, Indonesia and Taiwan. The majority also speak another language fluently; for Claire, English is her second language and at the time of the study she was learning a third language, Farsl. It is more common than not for experienced E S L teachers to have such a wide-ranging background, particularly those holding teaching positions at established further education colleges. The problem in quantifying these data as has been done in the Table 6-1, is that it presents a very simplified picture of what Is actually a multi-faceted background of practical knowledge that these teachers brought to the study context. TESL Training and Academic Background: The second type of data presented in this table relates to teacher training and graduate work in TESL. These data show that all these teachers completed university level T E S L training and in addition, several had also participated in short, intensive training programmes (at adult Institutions). According to the teachers, the emphasis in these intensive programmes had been 'hands-on' experience in the classroom with a major portion of the time spent working with a sponsor teacher. In addition, at the time of this study, the majority of these teachers had also completed, or were in the process of completing their Masters of Education in TESL. This indicates what seems to be a current phenomenon in E S L -- a professionalization of the discipline with established teachers pursuing Masters of Education degrees. These teachers were asked how they felt their initial T E S L training and particularly, their more recent graduate work, had influenced their views on E S L teaching/learning. In reference to graduate work, most felt this had been a positive experience in that it provided them with a theoretical framework for what they were already doing in the classroom. For example, Edie noted, -"All of a sudden, I could hang my experience on this tree of theory." Others felt that graduate work had "recharged" them professionally. Annie, for instance, explained that she now thought more in terms of theory when she was making decisions for her teaching. She also commented that both her teaching experience and recent graduate work had given her a more critical perspective on classroom practices; "I find I'm more interested in why things work, how they work and in what context they work." A common theme in all their responses related to the central role that the practical component plays in improving classroom skills. They stressed that the most important element in their continued growth as professionals was observing colleagues teaching. In addition, a number commented on the need for a more 'hands-on' approach in teacher training as opposed to an emphasis on theory at this stage; --"There are so many things to figure out when you're first teaching. You're not thinking about theory. You're just thinking about getting through the day." These comments provided some insight about how theory and practice contribute to professional development from experienced teachers' perspectives. Teachers felt that the theoretical input had contributed current research knowledge for informing their planning/teaching decisions; at the same time, they also valued opportunities to observe colleagues teaching and share ideas and materials. 2. Context Factors Before examining how specific setting factors impacted on teacher decision-making, it is useful to provide pertinent background information on the three sites involved in this study. The following table (6-2), Contextual Characteristics of the Three Sites, summarizes tine features of tlie programs, courses, students and administrative input for eacl i of tfie tliree sites. Table 6-2 Contextual Characteristics of the Three Sites SITES ENGLISH BAY PACIFIC CENTRE SOUTH CENTRE PROGRAM general + academic general + academic academic this study general general general CLASSES time day/evening day/evening day/evening this study day/evening day day levels beginners to advanced beginners to advanced intermediate to college prep. this study intermediate + advanced intermediate + advanced intermediate + advanced skill focus integrated + specialised integrated speaking/listening reading/writing this study integrated + discussion (Helen's class) integrated speaking/listening ADMINISTRATIVE input student evaluation list of objectives + course description institutional exams list of objectives + course description teacher evaluation course description (including objectives) teacher evaluation STUDENTS international + community international international + community Contextual Characteristics programs: While English Bay and Pacific Centre offer both General and Academic programs, South Centre (with its relatively small E S L department), offers only an Academic Preparation Program. The teachers participating in this study from the first two sites were teaching courses in the General Program. As discussed in the Methodology Chapter (site descriptions), the General program with its classroom focus on speaking/listening at the intermediate level made this a better program choice than the Academic programs at these two sites. The courses taught by the teachers at South Centre, were intermediate to advanced speaking/listening classes. classes: a) time: All three sites offer day and evening classes; the teachers in this study (with the exception of Claire), taught daytime classes. b) levels : Classes at English Bay and Pacific Centre ranged in levels from beginners to advanced; those at South Centre ranged from intermediate to college preparatory. The courses in this study were intermediate to lower advanced levels. c) skill focus: Both English Bay and Pacific Centre use an integrated skill approach in the General Program; that is, these courses include all four skills of speaking, listening, reading and writing. On the other hand, the courses at English Bay are skill-based; each course focuses on a specific skill. Although on paper, this indicates a significant difference for course content between English Bay and the other two colleges, in reality, this was not the case. Speaking/listening activities were the major components of the classes observed at both English Bay and Pacific Centre; thus there was greater consistency in the actual implementation of these courses than is apparent in the institutional descriptions. institutional guidelines: All three institutions outlined objectives (structural, functional and skill-based) for these courses in course descriptions. While these descriptions were also those given to the students at English Bay and Pacific Centre, the course descriptions at South Centre were administrative documents and only used by the instructors. The students were given instructor-developed course descriptions. These college (administrative) descriptions had been developed by a departmental curriculum committee in a format which conformed with college policy. In addition to course objectives, college policy requires other details of the course to be included such as class size, pre-requisites and evaluation procedures. These course descriptions must also be approved by a college-wide committee before they can be implemented at the departmental level. student evaluation: English Bay tests all students at the end of term with an institutional set of exams; teachers also write reports on their students, but the exam results determine whether a student moves up to the next level or not.^^ At Pacific Centre and South Centre, individual teachers are solely responsible for the evaluation process. At South Centre there are general guidelines for this in the departmental course descriptions, but teachers can decide to evaluate students through a series of assignments or in-class tests (mid-term and end-of-term), or a combination of both. At Pacific Centre teachers write reports on their students at the end of term; at South Centre teachers do not write reports for the administration but there is an informal procedure whereby individual teachers can write anecdotal notes on particular students to help with placement at registration time. students: All three sites have both international students and community students enrolled in their courses. However, most of the classes in this study were predominantly composed of international students. The one exception was Claire, in whose evening class there were mainly community students. However, it should be noted that the course content/objectives for this evening class was the same as for the day classes. This first section has provided background information on teacher characteristics and context characteristics of the study settings. The next section reports the findings for the groups of teachers in each of the three sites. For each teacher there is a brief discussion of i) approach to the instructional task; ii) perceptions of the instructional context; and iii) perceptions and Since this study, the E S L Department at English Bay College has decided to place more emphasis on teacher reports/evaluation for determining whether students should be promoted to the next level or not. implementation of the instructional task. In addition, at the end of each individual teacher section, there is a table summary of the activities, segments and participation structures in the four lessons observed. More detailed summaries of each lesson and the task grid analyses teachers completed are provided in Appendix 5. The findings related to teacher beliefs and perceptions and how these impacted on decision-making will be discussed in greater detail and in relation to all nine teachers in Chapter VII. B. SITES AND INDIVIDUAL TEACHERS 1. English Bay Community College English Bay Community College offers a varied selection of E S L courses. The largest programme is one which provides general English instruction in both daytime and evening classes. The three teachers participating in this study from English Bay — Claire, Mark and Stan ~ were teaching in this programme. While Mark and Stan were teaching full-time in the day, Claire was a part-time evening instructor. Although the institutional objectives remained the same for both day and evening courses, the type of student who enrolled tended to be somewhat different. Stan and Mark's students were primarily international students. Many of these students intended to complete the General English Programme, enrol in the Academic Preparation Programme at the college and eventually attend a Canadian university. The other students in these two classes were French Canadians, most of whom were here for a short time and intended to return to Quebec to attend university. The students in Claire's class were mainly community students. Many of them were already Canadian citizens and the rest either had landed immigrant status or refugee status and were intending to become Canadian citizens. A number of these students were working full-time and had families to take care of so they had little time to spend on homework assignments. However, unlike the international students in the daytime classes, these evening students used English on a daily basis in the community. They were therefore quite fluent. Some of these students wanted to continue on with their E S L studies in the Academic Programme; other students just wanted to improve their general English skills. Despite these differences in the type of students, the same curriculum guideline was used for both day and evening classes. In addition, all the students in these classes were at the intermediate- lower advanced level. The institution tests the students in this programme with a set of exams. These consist of an oral interview, a composition and multiple choice questions. These are marked by instructors in the department. There is a 'criteria sheet' provided for marking the interview and the composition although Mark noted that "it comes down to teachers interpreting the criteria sheet through their own biases". He felt there was some discrepancy therefore in how these exams were marked; some instructors focused primarily on student grammatical errors but others "go for content." Mark also noted that while this battery of tests used to be the only assessment considered for student promotion to the next level, the department had recently "opened up" the grading system so that a percentage of a student's mark was now based on classroom work and individual instructor assessment. CLAIRE : English Bay Community College Approach to Instruction Claire's instructional goals were primarily grammar-based. Accordingly her curriculum and the majority of her lesson tasks were focused on accomplishing structural objectives. Claire articulated a number of reasons for her product-oriented planning. First, she felt that her students were "already quite fluent" and at this stage of their language learning (advanced level) needed to work on developing linguistic competence. Second, Claire, who had learned English as a second language herself, viewed her knowledge and mastery of English structure as a particular strength she brought to the teaching situation. Third, Claire was particularly concerned that her students were adequately prepared for the institutional end-of-term exams. She felt these exams primarily tested student mastery of the structural objectives listed for the course. However, Claire's implementation approach was interactive based on her view that students should be "actively engaged in meaningful language use." She therefore provided many opportunities in her lessons for students to co-operatively work in small groups. The tasks that Claire implemented were not the typical 'communication gap' activities found in L2 texts since Claire did not find these particularly relevant for her students -- "we don't have time to think about how to order something from a menu." For Claire, a discussion of a grammar point was "just as valuable as any other type of communication gap activity." Indeed, there were no traditional communication gap activities used in the lessons observed; however, there were many small group tasks in which students collaboratively worked on grammatically-focused worksheets and exercises. Claire's interactive approach to accomplishing structural objectives is inconsistent with previous research findings. These studies have indicated that grammar and accuracy are predominantly addressed in teacher-fronted instruction (e.g.. Long 1977; Kramsch 1985, 1987). In addition, Claire's analysis of the student-centered tasks in her lessons indicated an inconsistency with the theoretical view that student interactive tasks should focus on meaning not form (e.g., Johnson & Morrow 1981; Littlewood 1981; Savignon 1983). Claire analysed most of these tasks as encompassing both accuracy and fluency components; the task product required accurate use of language whereas the process of student collaboration in accomplishing that product involved fluency or "getting the message across." Further, in assessing the success of her student-centered tasks, Claire clearly judged this in terms of the task product not the task process; that is, whether the students were able to successfully complete the grammar exercise. This perspective is not consistent with the emphasis task-based researchers place on the process of student interaction and in particular, the language of negotiation required to complete the task (Long 1977; Pica e t al. 1990). Another significant finding for this teacher was the impact of institutional factors on her decision-making. Claire was particularly concerned about covering all the institutional objectives in time for the end-of-term exams (set by the institution). She frequently expressed frustration over the limited time she felt she had in which to do this. There seemed to be a constant struggle for Claire between addressing externally set criteria for the course and at the same time meeting her students' affective and language learning needs. While her lessons were planned in reference to these external objectives (five or six were covered in a two week plan), the way in which a lesson actually unfolded was often determined by the students. For example, in several lessons, Claire made spontaneous decisions to review a grammar item when students were having some difficulty with exercises focusing on these. This tension between meeting curriculum-centered goals on the one hand and learner-based needs on the other, suggests that interpreting a teacher's approach as either one or the other (Johnson 1982; Richards 1984) is too simplistic. Perceptions of the Instructional Context institution: Although there were institutional objectives for the course, Claire felt she had a fair amount of freedom in her decision-making. However, she perceived the institution's major objective was the development of grammatical competence and planned her course accordingly. At the same time, she felt there was an "unwritten" expectation that classroom instruction would be primarily interactive. Since Claire personally viewed grammatical accuracy as a critical language learning goal for her students, her views were consistent with what she perceived as an institutional priority. In addition, Claire's concern for student interaction and collaboration in the classroom, was also consistent with what she perceived as the approach endorsed by the institution. students: Claire was very conscious of the fact that her evening students were "tired and had a lot on their minds." She therefore tried to incorporate a lot of variety in her lesson activities. In addition, since most of her students were working and/or had young families, they did not have time to work on home assignments so "all the work had to be done in class time." The students in Claire's class had a variety of language learning goals and needs. Some wanted to continue on to the college preparatory programme; others just wanted to improve their spoken English. Claire commented that this division made planning difficult in terms of meeting individual needs. instructional Tasl<: Perceptions and Implementation Participation Structure: Teacher Perception of % of Participation Structure Used whole class small groups pairs individual 20% 70% 5% 5% Analysis of the implementation data shows that across the four lessons observed, 37% of class time was spent in whole class format as opposed to 63% in student-centered (42% in small groups; 0% pair; 2 1 % individual). Therefore, Claire's perception of the general distribution of teacher-centered to student-centered organization was consistent with the data although slightly more time was spent in whole class than she thought. types of activities: Claire viewed whole class format as most useful for introducing/taking up tasks and for reviewing particular grammar points. In the lessons observed, Claire also used this format for working on taped listening exercises. Small group work was the primary format for collaborative work on structurally-focused tasks, comparing answers on completed exercises and pre-writing/writing revision discussions. Individual format was used for tasks focused on listening, reading and writing exercises although these usually served as a pre-activity or follow-up to small group tasks. teacher role: Claire did not specifically comment on her role during whole class work; however, in the lessons observed, her usual approach was to employ a question/answer technique. These sessions were quite lively since Claire did not solicit particular students to respond. Frequently, there was student-interaction with Claire sitting back until the students involved had reached a consensus or an impasse. In small group work, Claire saw her role as a resource and a monitor. She felt that the monitoring role was particularly important as she needed to know how students were coping with tasks. If she observed that a number of students were having similar difficulties, she would then go over these in the whole class follow-up. Claire also noted that monitoring student work enabled her to "make all those last minute decisions." Indeed, Claire's post-observation comments on her lessons indicated that she frequently modified her plans while teaching in response to the students. Reflecting on these interactive changes, Claire noted that there was a lot of "improvisation" in her lessons. However, it was clear that this was not due to lack of planning but Claire's concern for meeting student needs and responding to student affective states. In one lesson for example, Claire abandoned a vocabulary exercise when she realized students were too tired to cope. She switched to a song activity which she had brought to class as â contingency activity. The implementation data also confirmed Claire's perception of herself as a facilitator and resource during small group tasks. Even if'students attempted to draw Claire into the group discussion, she would usually feed the question back to another student in the group so that a student-directed focus was maintained. One's sense of Claire's view of her role was that she was there to help but that small group tasks were 'student-owned' and she should not become actively involved. Claire/ Lessons 1-4: Activities, Segments & Participation Structures Activities Participation Stractures & Segments whole class student-centered T-C/S-C G/P/I LESSON #1: Activity 1: Adjective Clauses time: 83 min. 42 min. 41 min. # of segments: 3 2 %o fP .S . 51% 49% Activity 2: Listening time: 11 min. 11 min. # of segments: 2 %o fP .S . 100% Lesson Total time: 94 min. 53 min. 41 min. # of segments: 7 5 2 %o fP .S . 57% 43% LESSON #2: Activity 1: listening time: 35 min. 20 min. 15 min. # of segments: 3 2 %o fP .S . 57% 43% Activity 2: defmitions time:24 min. 4 min. 20 min. # of segments: 3 2 %o fP .S . 17% 83% Lesson Total time: 59 min. 24 min. 35 min. # of segments: 10 6 4 %o fP .S . 41% 59% Activities Participation Stinctures & Segments whole class student-centered T-C/S-C G/P/I LESSON #3 Activitv l:readinp time: 39 min. # of segments: %o fP .S . 12 min. 2 31% 27 min. 2 69% Activitv 2:song time: 35 min. # of segments: %o fP .S . 5 min. 2 14% 30 min. 1 86% Activitv 3: listenine rime: 21 min. # of segments: %o fP .S . 6 min. 2 29% 15 min. 1 71% Lesson Total total rime:95 min. # of segments: 10 %o fP .S . 23 min. 6 24% 72 nun. 4 76% LESSON #4 Activitv l:adjective clauses time: 55 min. 7 min. # of segments: 2 %o fP .S . 13% 48 min. 1 87% Activity 2:test time: 20 min. # of segments: %o fP .S . 20 min. 1 100% Activitv 3: reading time: 23 min. # of segments: %o fP .S . 13 nun. 2 57% 10 min. 1 43% Lesson Total Total time:98 min. # of segments: 7 %o fP .S . 30 min. 4 31% 78 min. 3 79% STAN: English Bay Community College Stan has been teaching at English Bay Community College since 1975. He became very interested in the functional approach to teaching E S L in the late 1970's and several years ago was part of a team of curriculum planners who developed an adult functionally-based curriculum for the Ministry of Education. The class observed in this study was an intermediate level group with the majority of students from Japan and Hong Kong. There was also a small group of students from Quebec. Approach to Instruction Stan approaches E S L teaching/learning through a functional/structural approach. He feels that in order for his students to acquire socio-linguistic competency, they need to be actively engaged in language situations which have a purpose beyond merely manipulating the language. For Stan, the most effective approach for developing speaking/listening skills is "interaction in the classroom and interaction with native speakers." One way in which he provides meaningful interaction opportunities with native speakers is through extensive use of contact assignments outside of class. In class, Stan believes that classroom activities should be predominantly student-centered and interactive -"there's no reason why any task has to be teacher-centered." Stan organized his course and lesson activities in a functional framework. Although lesson tasks included some attention to grammatical items, these were always embedded in a functional framework rather than practised as isolated elements. This was consistent with Stan's view -"I don't believe in teaching grammar without a function to use it in." The role Stan adopted during student-centered tasks was of particular interest in the lesson observations. This was significant in terms of how much he stepped "out of the limelight" (Rivers 1987). More than any of the teachers participating in this study, Stan illustrated the non-participant role that communicative methodologists (e.g. Littlewood 1981; Johnson 1981) argue teachers should adopt during student-centered tasks. Apart from responding to student questions and infrequent exchanges with individual students about topics unrelated to the task, Stan maintained a monitoring role throughout these tasks. Stan frequently commented that he became quite "bored" during these activities and seemed to spend his teaching life just "waiting for people to finish things." 23 Perceptions of the Instructional Context Institution: According to Stan, the instructors working at English Bay College, were given "a great deal of latitude" in their planning and teaching decisions. He did not feel that the institutional course objectives or the end-of-term exams restricted his professional autonomy. Stan did note, however, that the students were very conscious of the exams and worried about them, particularly in the last part of the course. Stan felt that the institutional objectives were addressed "in dramatically different ways" by the various instructors in the programme. In contrast to Claire, he did not think that there was an institutional norm in terms of teaching approach. He commented that instructors in the department "would never agree anyway" on a particular instructional approach. students: Stan commented that students came to the programme with very different objectives. Some were there because they wanted to continue onto the academic programme; others were there "because it's the fashionable thing to do." He found that many of them needed to be motivated and encouraged to think for themselves. In general, he thought the students were "a nice bunch of students ... who were fairly responsive." Stan made a point of getting to know his students. He would often quietly chat to individual students during small group tasks about personal interests and lives. Stan felt that one of the advantages of small group work was the opportunity it gave him to make personal contact with his students. •^^ Interestingly, many of the teachers stated that they actively participated in student discussions because of the teacher boredom factor during student-centered tasks. Instructional Task : Perceptions and Implementation Participation Structure Teacher Perception of % of Participation Structure Used whole class small groups pairs individual 10% 30% 55% 5% Although Stan felt he rarely used whole class format, the implementation data shows that this organization comprised 67% of the time over the four lessons observed. The fact that in these lessons students were giving oral presentations, accounts for some of the discrepancy between Stan's perception of his use of whole class format and what happened in these classes. Even more surprising, given Stan's perceived ratio of the various formats, was the fact that group work only occupied 2 % of class time. In terms of student-centered participation structure, Stan used both pair tasks (24%) and individual tasks (7%) more often than small group work. He noted that he rarely had students working on individual tasks and when he did, it was more of "a concession" to his students since "many of them want to use the pen and paper they've been clinging to all the time." It should be noted that when commenting about participation structures, Stan was not referring specifically to these four classes but to his teaching approach in general. types of activities: Stan considered whole class format useful for introducing/ explaining tasks and "debriefing" tasks. In the classes observed, Stan also used whole class for role playing contact assignments, listening/pronunciation practise and student oral presentations. Stan noted that small group and pair tasks in his lessons usually Involve one of the following; problem-solving, "consensus-building", discussions and practising structures/functions. In the lessons observed, most of the student-centered tasks involved pair interaction. These tasks included pronunciation work (in dialogue form), describing, asking for personal information and decision-making. teacher role: One problem that Stan found with adopting a more student-centered approach is that the teacher "doesn't have a lot to do" and consequently spends a great deal of time "waiting for other people to do things." However, he felt that this student-interaction approach was "the best way to teach " a second language. He noted that one of the problems with an student-centered approach was that it "relies on student initiative and motivation" and students "often get sidetracked". He therefore saw one of his main roles during student interaction work was to keep students on task and moving towards completion. He commented that he usually stops student-centered tasks once three or four students have finished. Although this tends to "rush" students through tasks, he felt that most instructors give students too much time and as a result, "half the students are twiddling their thumbs and are bored." Stan also commented that he liked the opportunity this format gave him to interact with individual students. This was evident in the observed lessons; he would frequently chat to students about items not related to the task. He would also monitor and help students when asked. Unlike the majority of the teachers in this study, Stan did not become involved in student discussions but for the majority of the time maintained a monitor/resource role. Stan/Lessons 1-4 : Activities, Segments & Activities Participation Structures & Segments Whole Class Student-centered T-C/S-C G/P/I LESSON #1 Activitv 1 assignment time: 3 min. Activitv 2 problem-solving time: 92 minutes # of segments: %o fP .S . 28 min. 7 35% 64 min. 6 65% Activity 3 lesson closure time: 2 minutes 2 Lesson Total time: 97 minutes # of segments: 15 %o fP .S . 33 min. 9 34% 64 min. 6 66% LESSON #2 Activitv 1 oral presentations time: 26 minutes 26 min. Activitv 2 listening/pronunciation time: 33 minutes 25 min. # of segments 5 %o fP .S . 88% 8 min. 2 12% Activitv 3 contact assignment rime: 5 minutes 5 min. Activity 4 describing time: 28 minutes # of segments %ofP .S . 20 min. 2 71% 8 min. 1 29% LESSON TOTAL time: 100 minutes #of segments: 15 %o fP .S . 84 min. 12 84% 16 min. 3 16% Participation Structures Activities Participation Structures & Segments whole class student-centered T-C/S-C G/P/I LESSON #3 Activity 1 oral presentations time: 27 minutes 27 min. Activity 2 T.Monologue 5 min. Activity 3 proverbs (homework) time:25 minutes 25 min. # of segments 4 Lesson Closure 1 min. Lesson Total time: 58 minutes 58 min. # of segments 8 %o fP .S . 100% LESSON #4 Activity 1 presentations time: 23 minutes 23 min. Activity 2 describing time: 55 minutes 25 nun. 30 min. # of segments 3 1 %o fP .S . 45% 55% Activitv 3 adjective word order time: 19 minutes 11 min. 8 min. # of segments 2 2 %o fP .S . 58% 42% Lesson Closure 1 nun. Lesson Total time: 98 minutes 60 min. 38 min. # of segments 8 3 %ofP .S . 61% 39% MARK: English Bay Community College Approach to Instruction Mark described himself as an "oral, grammar-oriented" teacher. He believed that one of the "pitfalls" in language teaching theory was the view that 12 learners -can get by, both as a speaker and as a listener, with a fraction of the accuracy that would be used by a native speaker and communication will happen. Mark commented that he did not understand how anyone could learn a second language without grammar, particularly adults learners since they need to be able to "seize the language intellectually instead of scrambling around in the dark." Mark frequently referred to his own experience in learning French and how the classes that had been the most productive and enjoyable for him were those which had focused on "exploring" the formal system of the language. He had also found that through learning the formal system of French, he had discovered the French culture. In the same way, his goal for his teaching was to involve students in exploring the language and culture. He felt it was important that his students realized that learning a second language is not simply "mastering different words" but understanding "different thought patterns." Mark's product-based view of L2 teaching/learning was illustrated in the observation data. He not only based all classroom activities on grammatical objectives, but emphasized corrective feedback to students. He explained that he did not have much use for "this communicative stuff" because he felt what the students really needed was to develop accurate language use. At the same time, Mark's classes were not devoid of communication. Work in class was orally-based and included a great deal of student input (both solicited and spontaneous). Mark usually opened questions up to the class and this often resulted in quite lively interactions in the lesson between Mark and his students. One of the interesting findings from the implementation data was the amount of 'tangential' teaching Mark included in his lessons. Although his formal 'plan' usually consisted of checking the homework and working on the particular Structural focus of the lesson, in actual fact, his lessons included a number of 'mini-lessons'. None of these were planned but were spontaneously prompted by errors that students had made in their work. As a consequence of these 'side' lessons, the timeframe for taking up a homework exercise expanded considerably. Indeed, Mark never seemed to be concerned about the time factor in his lessons. If a student had a question or if there was a grammar point that a student did not understand, Mark did not hesitate to stop what they were doing at the moment and launch into an explanation of the item in question. Frequently this involved quite a full review of the grammar point including examples and/or timelines on the blackboard. Mark seemed to particularly relish these linguistic forays and one's sense was that the homework exercise was seen by him as a springboard for these more enjoyable lesson tangents. Mark obviously felt very comfortable 'thinking on his feet' and preferred letting the lesson evolve from within rather than imposing a preplanned structure beyond a very basic planned framework. Mark's lessons were also extensively teacher controlled. For the major portion of each lesson observed, Mark used whole class format (81% across the four lessons). Student-centered tasks that did take place (19%), only occurred at the very end of the lesson. These was also quite controlled in that students were usually given a set dialogue by Mark to practise. Since Mark viewed structural accuracy as critical for his students' language development, this implementation approach was consistent with his second language instruction beliefs. It is also consistent with L2 classroom research which shows that accuracy tends to be the predominant focus in teacher-fronted instruction (Long 1977; Kramsch 1985, 1987). Perceptions of the instructional Context Institution: Mark felt that the institution allowed instructors a fair amount of professional freedom in planning --""there's nobody coming around and saying you can't do that" - and teaching ~ "once I close the door, it's my world." Mark viewed the institutional list of course objectives as a useful guide for planning but, in contrast to Claire, did not feel that instructors were necessarily expected to include these in their course plans. He noted that the guide offered "all sorts of suggestions, but "you are allowed to create your own course objectives if you're not happy". Also in contrast to Claire, Mark did not focus his planning/teaching on preparing his students for the end-of-term exams set by the institution. He also noted that while this battery of tests used to be the only assessment considered for student exit from a level, the department had recently "opened up" the grading system so that a percentage of a student's mark was now based on classroom work and individual instructor assessment. students: In reference to student expectations, Mark commented that since many of them come with an "Oriental idea" of what happens in the language classroom, "their expectations are usually wrong." Mark added that his Asian students needed a lot of "weaning from these classroom expectations" so that they develop a more independent language learning approach. Mark feels that the problem with adopting a more communicative approach (as opposed to grammatical) to language teaching is the independence, self-reliance and confidence required on the part of the students. Mark noted, "that's not the type of student I am dealing with." Mark commented that he tended to have two basic types of students. One group is comprised of the "totally off the wall students" who, no matter what the instructor does, are not going to learn. At the other extreme is the group of students who is going to learn no matter what the teacher does because of independent learning skills. A basic requirement for learning, Mark believes, is student motivation; othenwise s/he is not going to learn. Mark believes that 90% of language acquisition comes from the student and only 10% from the teacher. The teacher's role is to create or enhance student motivation; "if the teacher doesn't do that", whatever else happens in the classroom "will be to no avail". Mark also commented that one of the problems with the programme was the lack of contact his Asian students had with Canadians. Although the students can live with a host family, most prefer to live on their own; "it's their first crack at freedom, their own place without parents around." However, this usually results in these students being quite isolated and having minimal contact with native speakers. On the other hand, those students who live with Canadian families or who have made Canadian friends, "obviously make more headway" in their language development. Instructional Task : Perceptions and Implementation Participation Structure Teacher Perception of % of Participation Structure Used whole class small groups pairs individual 50% 2 5 % ( 2 5 % ) Mark felt that he divided class time equally between whole class format and student-centered format. However, in the lessons observed, Mark's instructional format was predominantly teacher-centered (81% across the four lessons). The student-centered work he included in lessons tended to be at the end of the class and always involved small group format. In these activities, the focus was on practising a particular structure in a controlled dialogue exercise. Mark believed that to master a structure, it was necessary for students to say it over and over again. He referred to this as the "ugly" but necessary part of language learning. types of activities: Mark's lessons followed a routine of first taking up grammar exercises completed for homework followed by a teacher-fronted explanation/ review of the lesson structure. Mark referred to this part of his lessons as the "controlled" phase. Towards the end of the lesson he would switch to student-centered table work, which involved students' practising the structure embedded in a dialogue. For these "table dialogue" activities, Mark would usually expand a two-sided dialogue from the text Side bv Side into a three or four-sided dialogue so that each student at the table would have at least one line to practise. When asked about pair or individual activities, Mark commented that he used pair work in the computer lab. Visits to the computer lab constituted approximately one hour out of sixteen over the term. Individual work was usually reserved for the language lab every two weeks. Mark also taught a number of 'mini-lessons' while taking up homework. These varied in length depending on the complexity of the grammatical item, but would usually be prompted by errors students had made in their homework or student questions about an item. Mark commented that he liked these 'tangents' and felt they served a useful purpose. He particularly likes students having questions because it indicates that the student "is trying to deal with the material" and is "exploring" the language. teacher role: Mark maintained a teacher-directed role throughout the lessons observed. However, he did so with a sense of humour and an obvious enjoyment for what he was doing. Therefore, while the atmosphere was a serious, work-oriented one, it was also frequently entertaining. Mark's concern for grammatical accuracy was evident not only in his thoroughness in taking up homework but his approach to spontaneous errors students made while going through exercises in class. He would often rap on the desk while a student was speaking to indicate that s/he had made a mistake. The students were obviously used to this system as they would stop and try to correct their mistake right away. Mark also noted that if a student made a mistake during a whole class discussion, he did not usually stop to correct "unless it's something we've done recently." He added that if he stopped students every time a mistake was made, "there would be no communication happening at all." When students were in small groups, Mark circulated from table to table. Mark sat at each table, monitored the task, corrected student grammar errors and/or provide mini-lessons to the group on a particular point. Mark/Lessons 1-4: Activities.Segments & Participation Structures Activities & Segments Participation Structure whole class student-centered T-C/S-C G/P/I LESSON 1 Activity 1: homework time: 24 min. # of segments %o fP .S . 24 min. 2 100% Activity 2: homework time: 49 min. # of segments %o fP .S . 34 min. 2 69% 15 min. 2 31 % Activitv 3: film discussion time: 5 min. # of segments 5 min. 1 Activitv 4 2nd condition rime: 39 min. # of segments %o fP .S . 24 min. 2 62% 15 min. 1 38% Lesson Total time: 117 min. # of segments %o fP .S . 87 min. 9 74% 30 min. 4 26% LESSON 2 Activitv 1: skiing time: 15 min. # of segments 15 min. 1 Activitv 2: homework time: 24 min. # of segments 24 min. 1 Activity 3:conditional time: 42 min. # of segments %ofP .S . 27 min. 3 64% 15 min. 1 36% Lesson 2 Total time: 118 min. # of segments % o f P.S. 94 min. 5 80% 24 min. 2 20% Activities Participation Structure & Segments whole class student-centered T-C/S-C G/P/I Activity l:skiing time: 8 min. 8 min. # of segments 1 Activitv 2: homework time: 30 min. 30 min. # of segments 1 Activitv 3: homework time: 26 min. 17 min. # of segments 1 % of P.S. 65% 9 min. 1 35% Activitv 4: conditional time: 52 min. 36 min. # of segments 3 % of P.S. 61% 16 min. 2 39% Lesson Total time: 116 min. 91 min. # of segments 6 % of P.S. 78% 25 min. 3 22% LESSON 4 Activitv 1: ski t i ip time: 22 min. 22 min. # of segments 1 Activitv 2: homework rime: 26 min. 26 min. # of segments 1 Activitv 3: homework time: 35 min. 24 min. # of segments 1 % of P.S. 69% 11 nun. 1 31% Activitv 4: conditional time: 36 min. 36 min. # of segments 1 Lesson Total time: 119 min. 108 min. # of segments 5 % of P.S. 91% 11 min. 1 9% English Bay Community College : Concluding Comments Claire, Mark and Stan shared similar views about the institutional list of objectives; that is, they all considered these objectives as suggestions rather than requirements for teaching. However, there were differences in terms of how they used this list for planning decisions. While Claire adopted this list as her curriculum plan, Stan and Mark used course plans they had developed over a number of terms they had taught this course and more in line with their own views concerning what this course should entail. Mark and Stan also differed from Claire in their relative lack of concern for the institutional end-of-term exams. They were both quite dismissive in their attitude towards these and seemed to view this departmental evaluation process of students as a minor concern in relation to their planning and teaching decisions. On the other hand, Claire was particularly concerned that her students were prepared for these exams. She frequently expressed frustration over having to "teach to the test" and the limited time in which she had to do this. These three instructors were quite different in their approach to the instructional task. While both Claire and Mark were structurally-based in their beliefs about 1_2 learning and in their course planning approach, their classroom implementation techniques were considerably different. Mark adopted a teacher-fronted, mastery-learning approach which focused on practising discrete structural items through grammar exercises and four-line dialogues. He also emphasized learner understanding of the linguistic rules governing these structural items. Claire, on the other hand, used a more student-centered approach with a great deal of class time being spent in collaborative small group work. Although this was a communicative framework, her primary concern was the successful completion of structurally-focused task products. Unlike Mark and Claire, Stan approached his teaching through a functional framework. He believed that students needed to be provided with opportunities to use the language meaningfully through interaction activities in class and contact assignments outside of class. Although Stan included some work on structures, this component was always embedded in a functional framework so that only those grammar items related to a given function were included in a lesson. 2. PACIFIC STUDIES LANGUAGE CENTRE This institution offers both a general English language programme and an Academic Preparation programme. In addition, there are short, intensive language courses in the summer. The three teachers participating in this study, Helen, Edie and Fran, were all teaching in the general programme at the time. However, while Edie's and Fran's courses were part of the morning core programme and therefore required for all students, Helen's course was one of a number of elective courses offered in the afternoon. The particular course that Helen was teaching was a discussion course focusing on social issues. Teachers at the Centre were provided with course descriptions and a list of functional/structural objectives for the various levels and courses in this programme (See Appendix 3). For Helen's course, there was only a course description and a list of suggested topics; however, there was a more comprehensive set of objectives provided for the courses in the core programme. Another difference between Helen's course and the core course related to the institutional expectation that the morning courses would include more of a grammatical focus. Helen commented that the afternoon classes were supposed to be more "relaxing" and intended as a "break" from the more rigorous focus in the morning. A contextual difference between this site and the other two was the administrative requirement for instructors to submit a curriculum plan for their courses at the beginning of the term. The Centre provided planning charts (See Appendix 3) which instructors used to note their weekly themes, structural/ functional content and materials. Helen's curriculum plan was more general; the curriculum planning chart for her 'specialized' class included only three categories: content, skills & student activity and materials. The three instructors participating in the study did not feel this was restrictive in any way; in fact, they all commented that they found this useful. These instructors also noted that they were not necessarily bound by these initial plans and often made changes as the course proceeded and they responded to their own learners' needs. The students in the three classes observed were predominantly from Asian countries. In Helen's class there was more of a mixture of nationalities in that there were also students from Quebec and Mexico. Helen: Pacific Studies Language Centre Approach to Instruction This course was unique in this study given the minimal external constraints imposed by the institution on the curriculum. Apart from the expectation that the subject content would focus on social issues and the language component would emphasize discussion, Helen had a fair amount of autonomy in how she designed and implemented the course. What seemed to be a potentially limiting factor was the type of student who enrolled in the course. Helen implied that it was critical for the success of the course that students were self-directed, outgoing individuals who were genuinely interested in learning more about these issues and discussing points of view with classmates. Although Helen described the class observed for this study as a "tough" class to teach, she also enjoyed their lively spirit and comradeship. The fact that Helen was not obliged to cover a certain number of language objectives in the course, also meant that she could respond more easily to student needs and interests. She commented that she had incorporated much more of a vocabulary development focus after several of the students had requested this. She could also lengthen or shorten the time spent on any one topic in response to student interest. In contrast, the instructors teaching the core courses in the mornings, were expected to teach a certain amount of grammar and to incorporate the four language skills in their lesson activities. Perceptions of Instructional Context institution: This particular course is advertised by the institution as a conversation course focusing on social issues. As such, there is a specific content focus and an expectation that classroom activities will be "communicative". However, beyond those general expectations, Helen felt she had a great deal of freedom in terms of how she planned and taught the course. There are suggested topics but no explicit language learning objectives. Although Helen was required to submit a projected 12 week curriculum plan for the term, she admitted that she did not necessarily keep to these projected plans. However, she viewed this planning a useful exercise as it "forces me to think about what I'm going to do at the beginning of the term." Helen was expected to write final reports for her students but these are anecdotal and more concerned with class participation and general progress over the term than evaluation of specific language learning objectives. Helen noted that she had never failed anyone in this course although she has had to talk to individual students about lack of participation in class. When asked if she ever gave tests to these students, Helen said she did "periodically" but found that students in this class did not usually study. She added that she tended to give collaborative vocabulary quizzes at the end of each topic. These usually consisted of matching words and definitions. Fewer institutional constraints enabled Helen to be more guided by student expectations. For example, the students not only selected the topics but their interest in a topic would be the deciding factor in terms of how long they would spend on it. Helen was also able to respond to specific requests as the course proceeded. When some of the students asked for more emphasis on vocabulary for example, Helen was able to incorporate more of an emphasis on this. An institutional constraint that Helen noted was the expectation for instructors to organize and participate in social activities outside of class time. Helen commented that these activities could mean "an extra eight hours" added on to her time commitment to the class. She felt that the institution did not fully realize "how many extra hours" were actually spent in fulfilling this expectation and although she usually enjoyed these dinners or outings, they were also "very tiring." students: Helen commented that she particularly enjoyed teaching the class observed for this study because the students were friends outside of the class and this positively impacted on classroom atmosphere. They were really good friends. They socialized outside of class and so they enjoyed coming just to see their friends. No matter what I did, they had fun. Everyone was really easy-going and supportive. I had such fun teaching them because it was a social gathering in a way. Helen explained that the student nationality mix she had in a social issues class significantly impacts on how successful the course is. In order to be successful, it requires motivated students who are genuinely interested in the issues and willing to actively participate in class discussions. In reflecting on the class observed, Helen noted that the group of Mexican students were very outspoken and at times difficult to keep on task. Although she enjoyed their lively spirit, it was necessary to keep the pace moving with a variety of activities and at times impose more teacher control than she wanted. Helen commented that if she let an student-centered task continue for too long, these students would start speaking Spanish and "get completely off track". Instructional Task: Perceptions and Implementation participation structure Teacher Perception of % of Participation Structures Used whole class small groups pairs individual 15% 70% 10% 5% In the lessons observed, Helen used whole class format 28% of class time over the four lessons. This slightly higher percentage than the 15% Helen specified, was largely due to the 'town meeting' activity (coded as whole group) in lesson 4. In terms of "teacher talk", Helen spent relatively little time at the front of the classroom. Most of her interaction with students occurred during group and pair work. Student-centered tasks comprised 72% of classtime across the four lessons. These were predominantly small group tasks (37%) and discussion focused. Helen finds whole class format useful for certain activities such as introducing a new topic (providing background information), or eliciting student knowledge of a new topic. She also commented that towards the end of term, she often uses whole class format for discussions since the students are "fed up" with talking to each other in small groups and are happier with whole class interaction. Helen commented that she used small group tasks more frequently than pair tasks. She mainly used the latter in preparation for small group work or for accomplishing a more specific task objective such as vocabulary exercise. She also found pair work useful in classes where students "didn't talk much" as in this format each student "had to talk". types of activities: Helen commented that one of her goals for this class was to try to get away from basic questions of 'what happened' and to encourage students "to look at both sides of an issue." By including such activities as debates and 'town meetings' in which students adopted particular roles, Helen involved students in considering and arguing from different perspectives on social issues. Within the social issues content framework, Helen also focused on vocabulary building. In order to provide background information on a topic, Helen began each unit with a mini-lecture or a newspaper article. Once students had this basic content information, classroom tasks (pairs and small groups) primarily focused on exchanging opinions and cross-cultural exchanges of information. Helen believed that information exchange tasks were particularly important for this type of class since "students should have a reason for communicating". teacher role: Helen noted that her classroom approach was less language code focused in this discussion class than when she taught in the morning core classes. In the core classes, she was more likely to address grammatical problems through individual feedback and task objectives. On the other hand, in this discussion class, she rarely worked on structure and accuracy of language use. Content was the primary activity focus and so Helen tended to respond to "what students are saying not how". This was evident in the lessons observed. Helen became actively involved in student discussions as an equal participant, sharing experiences and ideas with the students. The only language code focus Helen specifically dealt with in her lesson exchanges with students was vocabulary. In the observed lessons, Helen's approach in whole class format was quite relaxed and informal. When introducing a new topic or following up on a class activity, she usually encouraged all students to participate by asking information-seeking questions and not soliciting particular students to respond. As a result these were really brainstorming sessions with Helen and the students exchanging information. On the other hand, when giving task instructions, Helen was more businesslike and efficient in explaining what they were to do. If students were not clear about the task, Helen would explain the instructions on an individual basis. In small group and pair tasks, Helen saw her role as a facilitator and resource person. She felt it was important to circulate and make sure the students understood "what the task was all about". She added that she helped with vocabulary, answered questions and frequently participated in discussions. Helen commented that she shouldn't participate so much in student discussion tasks, but added 1 slip into it more by accident. 1 try not to but 1 tend to give my opinion, especially if they ask me. Helen noted that she "should just stand back and listen" rather than get so involved. However, despite feeling a little guilty about this participation, she also believed students enjoyed hearing her anecdotes (often related to her visits to their countries) and cultural point of view on certain issues,-"lt makes the conversation more real". Helen/ Lessons 1-4 : Activities, Segments & Participation Structures Activities & Segments Participation Structure whole class student- centered T-C/S-C G/P/I LESSON #1: introduction to environment topic time: 101 min. # of segments: 11 % of P.S. 25 min. 6 25% 76 min. 5 75% LESSON #2: Activitv l:planning social activity time: 6 min. # of segments: % of P.S. 6 min. 1 100% Activitv 2:.stations time: 101 min. # of segments: % of P.S. 4 min. 2 4% 97 min. 3 96% Lesson Total time: 107 min. # of segments: 6 % of P.S. 10 min. 3 9% 97 min. 3 91% Activities Participation Structure & Segments whole class Student-centered T-C/S-C G/P/I LESSON #3 Activitv l:recvclins rime: 55 min. # of segments: % of P.S. 16 min. 4 29% 39 min. 3 71% Activitv 2 Fihn:"Cry for Freedom" time: 47 min. 4 min. # of segments: 2 % of P.S. 9% 43 min. 2 91% Lesson Total time: 102 min. # of segments: % of P.S. 20 min. 6 20% 82 min. 5 80% LESSON 4 Activitv 1: social activity plans time: # of segments: % of P.S. 10 min. 1 100% Activitv 2: review vocabulary/content time: 35 min. # of segments: % of P.S. 15 min. 4 43% 20 min. 3 57% Activitv 3: 'town meeting' time: 63 min. # of segments: % of P.S. 41 min. 4 65% 22 min. 1 35% Lesson 4 Total time: 108 min. # of segments: % of P.S. 66 min. 9 58% 42 min. 4 42% EDIE: Pacific Studies Language Centre Approacli to Instruction Three inter-related features of Edie's approach are particulary significant: the interactive role Edie adopted during whole class instruction; her communicative approach to both grammar and content activities; and the predominance of teacher-fronted tasks . Evident in Edie's whole class teaching was the 'negotiated' (Long 1989) nature of the lesson flow. Edie encouraged active student participation through a series of questions, both teacher and student initiated and teacher responses focused on clarification and confirmation. Secondly, Edie's belief in the importance of encouraging student language 'output' was clearly reflected in her approach to grammar-based tasks. Students were encouraged to compare and discuss grammar work they had completed for homework as opposed to being merely given the correct answers. In addition, Edie's 'discovery' approach to grammar teaching was consistent with the view that "students can talk just as meaningfully about the language system as they can a content topic". Edie felt very strongly that drilling a grammar point was "counter-productive" because "just memorizing something is not really helping students learn the language." Although Edie believed it was important for students to be given information about the patterns in the language, she felt that students internalize these patterns at different stages. She therefore did not worry if individual students were not using a structure correctly since this would happen when students reached an appropriate stage in their language development. The important thing for her was that students were given the opportunity to use the language for communication. Finally, despite Edie's perception that whole class instruction was a minor focus in her classroom organization patterns, the implementation data indicates that this was not the case. However, the data also suggest that this teacher-fronted approach was not inconsistent with her communicative view of 12 instruction. First, Edie's interactive approach in whole class format encouraged student active participation in the flow of the lesson. Second, Edie frequently stated that students need to communicate with someone who knows whether they are accurate and comprehensible (a native speaker) if they are to successfully develop their language skills. As a native speaker Edie therefore provided not only the accurate language model but was able to provide feedback to students on their comprehensibility. Perceptions of the Instructional Context institution: Edie felt that the administration was very supportive and allowed teachers a great deal of freedom in their planning and teaching decisions. Edie did not find that the inventory of institutional course objectives imposed on her planning/teaching decisions. Rather they provided a framework and a guide for assessing whether students were ready for the next level at the end of the term. While Edie appreciated this freedom, she also felt it had some drawbacks in that she was unclear at times as to what the administration's expectations were. students: Edie enjoyed the small classes at the centre because it meant that students could be organized into pairs or small groups easily. On the other hand, a negative feature was the lack of cultural mix. This she felt was problematic particularly from the students' point of view since they often felt "it was like being in Japan." Having a majority of one language group also impacted negatively on the type of interaction opportunities students had in a class. In comments referring to individuals in the class, Edie clearly separated students into two groups; those who needed to work on accuracy and those who needed to work on fluency. In reference to the former, Edie felt these students " were pretty fluent and good communicators but what came out of their mouths was pretty jumbled and garbled". On the other hand, she felt the Japanese students were at the level where the main focus should be fluency and "so that's what I focused on." instructional Task: Perceptions and Implementation participation structure: Teacher Perception of Participation Structures Used whole class small group pairs individual 20% 50% 20% 10% In reflecting upon her use of whole group format with this class, Edie stated, "I did very little actually standing up and talking. I tried my best to get them to learn what they could from working together in groups or pairs." However, although Edie felt she used whole group format only 20% of the time, the implementation findings show whole class format as the predominant participant structure (75%) used in these lessons. Also in contrast to what Edie stated about the ratio of student-centered tasks, small group work was only used for 10% of the class time across these four lessons. Pair tasks and individual tasks occupied the remaining 15% (9% for pairs and 6% for individual). Edie's approach within whole class format, was question-answer and therefore involved a high level of student participation. In analyzing these teacher-fronted tasks in the post-observation conferences, Edie would also frequently point out that although the organization was teacher-centered, the students' input was not entirely controlled so that "the students have the opportunity to ask a question and move (the lesson) in a different direction". Edie stated that she prefers to use small groups for discussions but pairs format is "better for grammar work." This view was reflected in the grammar lesson activities in which Edie used either whole group format for explanation and taking up homework/class work or pair format for working on the grammar point. Small group work was not used when grammar was the focus. activity types: Edie believed that "to really learn how to use the language, you need to have pushed output" ; that is, students need to be given opportunities to organize information and express ideas in longer segments than one or two words. In order to provide these opportunities, Edie incorporated two particular components into her lessons: discussion activities and oral presentations. In preparation for each 'discussion lesson', Edie's students read a newspaper article (provided by Edie) and completed vocabulary and comprehension questions for homework. During the lesson itself, students were given a number of questions on the topic of the article to guide the discussion. Edie was careful to choose articles on current topics and ones she felt her students were interested in- "I feel it is important to use content that involves the students so that they are interested in the discussion." Several of Edie's students also gave oral presentations during the videotaping of these classes. Edie commented that she felt this type of activity provided an important skill development focus for her students in that it gave students the "experience of presenting something and having people understand you." Although individual students were given 'evaluative' feedback from Edie and their peers on the presentation, this was only intended to provide "helpful suggestions" and not part of the student's final grade. Edie felt her role in providing feedback was "only one of 12" and that peer comments were just as important- "It's important that students get feedback from their peers about whether or not they were understood". One lesson a week was designated for grammar instruction. In the other three lessons, Edie predominantly focused on content and fluency development activities. Apart from working on the vocabulary in the discussion articles, there was very little attention to language code in these classes . teacher role: Edie's Interactions with students during small group discussions was primarily content-based and information-seeking rather than corrective -I don't like to correct students. I'll correct them when they're writing because it brings their attention to (the error) but when they're speaking I'm more interested in content. In general Edie 'interfered' very little with students in discussion groups and when she did become involved it was either as a native speaker resource or as one of the participants. In whole class format, Edie usually focused on content (as opposed to language code) and her feedback to students' responses most often involved clarification and confirmation requests. This teacher-student interaction approach often resulted in a number of exchanges as Edie and the students clarified and exchanged information. Attention to student errors in grammar was extremely rare and only became a focus when discussing a particular grammar point during those lessons designated as 'grammar days' by Edie. In this case Edie addressed the errors in the work students had completed for homework not the spontaneous errors made while students were speaking. This predominant 'conversational' approach to teacher-student interaction was consistent for all the whole-class tasks including those focusing on vocabulary and grammar and reflected Edie's concern for meaningful interaction. "When you're talking meaningfully, it doesn't matter what the topic is. You can be talking meaningfully about Aids or about how the English language is put together." The fact that a great deal of this "meaningful interaction" took place in whole class format might be explained by Edie's view that "students not only have to have opportunities to use the language but it's important that they produce the language for somebody who knows whether it's comprehensible or not". Further, Edie commented that students receive feedback on whether they are comprehensible "through the use of confirmation questions" not through accuracy-focused feedback. In group work tasks, Edie saw her role as "going around and monitoring in order to see how well the task was going and " what kinds of things they were coming up with." She also liked to "throw out an idea" if students were having difficulty or "if there was something important they were missing." At times, Edie commented "I would just play devil's advocate just to get things going." Edie often participated in the group discussions but this interaction with students was primarily content-focused. She rarely responded to student language code errors but would provide students with vocabulary if they asked. Edie would circulate from group to group during discussions. Her approach was to listen for several minutes and then either become involved in the discussion or move on to the next group. On the other hand, her role during pair work (grammar) was much more systematic. She would interact with each pair of students pointing out errors in their work and helping them with the exercise. Finally, Edie saw her role during student oral presentations as one of the members of the class. She did not interrupt the speaker at all while s/he was presenting and gave the presenter the sole responsibility for directing the subsequent question-answer session. It was particularly interesting that even when the question-answer portion was 'dying' Edie did not become involved but would let the presenter and his/her classmates deal with this. In both student led presentations/discussions observed, there were a number of occasions during which the presenter had to struggle with getting the question-answer/discussion period going again. This they were successful in doing. Edie/Lessons 1-4: Activities, Segments & Participation Structures Activities Participation Structures & Segments whole class student-centered T-C/S-C G/P/I LESSON #1: Activitv l:wann-iip time: # of segments: % P.S. 15 min. 1 100% Activitv 2:.sones time: 62 min. # of segments: % P.S. 35 min. 4 56% 27 min. 3 44% Activitv 3:pre.sentiition time: 25 min. # of segments: 1 % P.S. 100% Activitv 4:homewoik time: 3 min. # of segments: 1 % P.S. 100% Lesson Total time: 105 min. #of segments: 10 % P.S. 78 min. 7 74% 27 min. 3 26% LESSON #2 Activitv hwarm-up time: 20 min. # of segments: 20 min. 1 Activitv 2:homework time: 40 min. 40 min. # of segments: 1 Activitv 3: grammar time: 50 min. # of segments: 3 % of P.S. 12 min. 2 24% 38 min. 1 76% Lesson Total time: 10 min. # of segments: 5 % of P.S. 72 min. 4 65% 38 min. 1 35% Activities Participation Structures & Segments whole class student-centered T-C/S-C G/P/I LESSON #3: Activitv l:wann-up time: 15 min. # of segments: 15 min. 1 Activity 2: Aids article time: 85 min. # of segments: % P . S . 43 min. 5 51% 42 min. 1 49% Activity 3:monologue time: 5 min. 5 nun. # of segments: 1 Lesson Total time: 105 min. # of segments % P.S. 63 nun. 7 60% 42 min. 1 40% LESSON #4: Activitv 1: warm-up time: 10 min. # of segments: 10 min. 1 Activitv 2: grammar time: 62 min. # of segments: 62 min. 3 Activitv 3:presentation time: 25 min. 25 min. # of segments: 1 Activitv 4:'housekeepin2' rime: 3 min. 3 min. # of segments: 1 Lesson Total time: 100 min. # of segments: % of lesson 100 min. 6 100% FRAN: Pacific Studies Language Centre Approacli to Instruction Fran believed that learning grammar and paying attention to accuracy of language use were critical for successful language learning. This view was reflected in Fran's structural approach to planning and her accuracy-focused approach in lessons. Three features were particularly significant in Fran's instructional approach. First, Fran divided her lessons into two distinct segments. While one segment was explicitly concerned with a particular structural item, the second segment, was comprised of a number of inter-related tasks built around a decision-making content core. This was a consistent lesson organization pattern across the four lessons observed. A second significant feature was Fran's use of text-based communicative activities for explicit practise and reinforcement of structure and vocabulary. In the four lessons observed, three activities from the communicatively-focused 12 text, React/Interact were used as a content base for a wide range of 'scaffolding' language-code tasks. These tasks not only prepared students for the problem-solving component of the activity (usually the final task) by having them work with the information content and the vocabulary of the 'problem', but also served as discrete structural tasks in themselves. Fran's comment "I like to come at things from 100 different ways" was reflected in her "task building" approach for these activities. This use of a "scaffolding" technique has also been found in classroom research (Wong-Fillmore 1985) and L2 teacher planning research (Woods 1989). The third significant characteristic of Fran's approach was the instructional role she adopted during small group/pair tasks in her lessons. Fran viewed student-centered tasks not only as an opportunity for individualizing instruction but providing the structural feedback she believed was critical for her students' language development. In the lessons observed, this view was reflected not only in the structural objectives of the majority of tasks implemented, but the accuracy-focused feedback she consistently provided in interactions with individual students during small group and pair work. Percept ions of the Instructional Context inst i tut ion: Fran felt that the institution imposed few restrictions on her planning and teaching. She appreciated this professional freedom in her planning and teaching decisions because "I like to do what I want." Fran saw the institutional objectives as a useful resource but did not feel they dictated what teachers did or did not do in their courses. students:Fran consistently expressed the view that students needed to be self-disciplined and motivated in order to learn a second language successfully-I don't think learning a second language is a big mystery. You just need to be motivated and disciplined and to me that's all there is to it. For Fran, these student factors are more critical for language learning than teaching approach or materials. In commenting on her students, Fran noted that many seemed to lack the necessary self-discipline and did not take the opportunities to learn; "they just want to learn by being in class every day for 3 hours. They think they are going to learn magically without putting any effort into". This she finds quite disheartening; "I wonder why I'm teaching as the students aren't motivated and not disciplined. I can't relate to them." Fran commented that many of the classes tended to be composed of students from the same cultural background which she felt was a limiting factor. Fran also noted that students come with varying expectations and needs. This necessitates some "juggling" in teaching approach. Fran stated that she tries to modify her teaching strategies in response to these student differences. For example, she commented that she rarely corrects her Asian students because it is too inhibiting for them. However, she tends to correct her Québécois students all the time,- "I stop the Quebec students every 3 or 4 seconds to correct". For the Quebec students she feels that the majority of class time should be spent on teaching grammar. On the other hand, Asian students "already know grammar but they need to speak more." Therefore she feels she needs to provide more group work for her Asian students to get them talking. In contrast, the Quebec Students don't need so much group work because they talk all the time and "it's a waste of time for them to do a lot of talking." in addition, Fran tries to adapt her feedback strategies to the type of student. For example, with students who "clam up if they are given corrective feedback... I don't bother them." In the end, Fran believes that it's "really intuition knowing when correction is impeding or enhancing the flow. Some students I correct all the time; others I don't correct at all." Instructional Task: Perceptions and Implementation participation structure Teacher Perception of Participation Structures Used whole class small group pairs individual 30-40% 30% 30% 0-5% Fran commented that she "feels guilty" using whole class organization and tries to spend "as little time as possible" in whole class activities. She feels students need to be given opportunities for language 'output' in the classroom as students "get enough input on the street". Fran's perceptions of the percentage for whole class to student-centered activities in her classes is consistent with the observational data on the lessons observed. In these four classes, Fran used whole class format 44% across the four lessons. For student-centered activities, Fran used group work 33% of the time, pair tasks 2 1 % of the time and individual tasks only 1% of the time in these lessons. Fran stated that she preferred small group activities since- "it's the only time you get to do corrections... this takes too long in whole group format". She felt that small group work also provided students with the opportunity for language "output" which Fran viewed as critical for language learning. Fran views pair work as a useful format for interaction; however, she noted it can also be problematic because often one partner "does not talk much". Fran also commented she used pairs for grammar-focused tasks although she said often the students don't trust each other's knowledge of the grammar. Fran felt she used Individual format infrequently (5%) because it lacks interaction opportunities. In addition, she felt that if something could be done individually, it was a better activity for homework. In class, Fran would use individual format for grammar exercises, and cloze listening/reading exercises types of activities: Fran stated that she uses whole group organization for introducing an activity, presenting a structure, giving dictations, student oral presentations and taking up homework or classroom tasks. This list is also consistent with what was observed in the four lessons. Small group/pair work consisted of a wide range of activities which focused primarily on language code. For example, students completed exercises (from both grammar texts and communicative texts) and practised specific structures using dialogues, situations and visuals. teacher role: Findings from the observation data show that Fran predominantly used a question-answer format in whole class instruction. In the lessons observed, the majority of Fran's questions were pseudo-questions (that is, there was a particular answer she was looking for) and were focused on accuracy of information or language code (grammar, vocabulary). In student-centered work, Fran constantly circulated while students work in groups or pairs on a given task. Although at times Fran monitored only, it was a more consistent pattern for her to interact with individual students. These interactions were usually language code focused. They sometimes involved student-initiated questions about vocabulary meaning, but most often were teacher-initiated exchanges about accuracy of information, grammar or vocabulary. Pronunciation correction was an infrequent focus. These interactions also included management-focused exchanges such as asking students if they were finished, re-explaining task directions, and 'warning' students that their time was almost finished. In addition, there were affectively-focused exchanges which involved encouraging the quieter students to participate and praising the work students were doing. Fran also interacted with the students about the subject content in the task. These exchanges included both accuracy of information questions/ comments (checking that the students had the correct information) and information-seeking questions/comments (asking student opinions about the content). The focus of the content-based exchanges depended a great deal on the task the students were doing. If, for example, it was a task in which students were sharing information they had from a taped monologue, Fran's focus was centered on whether students had the correct information (accuracy of information and vocabulary). On the other hand, if students were involved in a decision-making or problem-solving task, Fran's interactions included questions about the students' opinions (information-seeking). Fran/ Lessons 1-4: Activities, Segments & Participation Structures Activities Participation Structures & Segments whole class student-centered T-C/S-C G/P/I LESSON #1 Activitv IrCloze time: 64min. # of segments: % P.S. 29 min. 2 45% 35 min. 2 55% Activitv 2: "Lady Gerta" time: 35 min. # of segments: % P.S. 20 min. 3 57% 15 min. 2 43% lesson closure time: 1 min. Lesson Total time: 100 min. # of segments: 9 % P.S. 50 min. 6 50% 50 min. 4 50% LESSON #2 Activitv 1: "Kate Lang" time: 66 min # of segments: % P.S. 23 min. 2 35% 43 min. 3 65% Activitv 2: Modals time: 36 min. 18 min. # of segments: 1 % P.S. 50% 18 min. 1 50% Lesson Total time: 102 min. # of segments % of P.S. 41 min. 3 40% 61 min. 4 60% Activities & Segments Participation Structure whole class student-centered T-C/S-C G/P/I LESSON #3 Activitv l:Modals time: 53 min. # of segments: % P.S. 38 min. 5 72% 15 min. 2 28% Activity 2: "Sophie's Choice" time: 47 min. # of segments: % P.S. 25 min. 2 53% 22 min. 2 47% Lesson Total time: 100 min. #of segments: 10 % P.S. 63 min. 6 63% 37 min. 4 37% LESSON #4 Activity 1: "Sophie's Choice" time: 103 min. # of segments: % P.S. 25 min. 2 24% 78 min. 3 76% time: 10 min. of segments: % P.S. 5 min. 1 50% 5 min. 1 50% Lesson Total time: 113 min. # of segments: 7 % P.S. 30 min. 3 27% 83 min. 4 73% Pacific Studies Language Centre: Concluding Comments All three Instructors from this site felt that the institution gave them a fair amount of freedom in their planning and teaching decisions. None of them found the requirement to submit a curriculum plan for courses a negative factor. In fact, all expressed that this was a useful exercise in that it forced them to think about the course in global terms before they started more detailed planning. Institutional expectations for the morning courses were seen by these instructors as more "serious" and more oriented to specific language learning objectives than the afternoon elective courses. Fran and Edie both referred to the language learning objectives set by the institution and the fact that they explicitly used these when planning for their courses; on the other hand, apart form a list of suggested topics, there were no specific objectives for Helen's afternoon discussion course on social issues. Although the institutional list of objectives and requirement for submission of curriculum plans clearly demonstrates that instructors did not have complete freedom in their decision making, instructors seemed to have a fair amount of leeway in how they actually implemented their curriculum in the classroom. This is suggested by the marked difference in how Fran and Edie approached their instructional tasks. For example, Edie designated specific classes for particular language learning goals; on the other hand, Fran adopted an integrated approach in each lesson. One result of this 'segregation' in Edie's classes was that accuracy of language use was only addressed on "grammar days". For the other classes, Edie maintained a content-based, fluency-focused approach. However, in Fran's lessons, each activity consisted of variety of tasks which encompassed grammar objectives, a content focus as well as both accuracy and fluency development. In addition, Edie and Fran approached the issue of accuracy and fluency quite differently in their teaching. Edie not only designated specific days for grammar, but only in these lessons did she address students' accuracy of language use. In other lessons she virtually ignored grammatical errors and responded only to the content of what students were saying. On the other hand, Fran, emphasized accuracy of language use in each lesson in both her task objectives and feedback to students. These contrasting approaches to the teaching of grammar and the role of accuracy and fluency in the classroom, obviously relate to fundamental differences in teacher beliefs about second language teaching and learning. However, these differences also suggest the level of freedom this institution gives to instructors to plan and implement lessons according to individual beliefs about language instruction. 3. South Centre Community College South Centre Community College has the smallest E S L programme of the three institutions in this study. In contrast to English Bay Community College and Pacific Language Studies Centre, the E S L department at this college offers only an Academic English programme. The courses in this programme are skill-based, (speaking,listening, reading,writing) and range from Intermediate to College Preparatory levels. Both international students and community students are enrolled in these courses. One of the attractions of this programme for international students is that they can begin their university coursework while completing the College Preparatory level of the E S L programme. Guiding instructors in their planning of courses, are departmental course descriptions. These have been developed in the department and approved by a college-wide committee. These course descriptions provide a general guide of course content and objectives and serve to ensure articulation between the four levels. Instructors are expected to consult these descriptions when planning but are not asked to submit curriculum plans for their courses. These descriptions are for instructor use in planning; the students are given course outlines which are developed by individual instructors for the specific courses. Each time an instructor teaches a particular course, a new outline is written to reflect any changes the instructor has made in the course design. A particular contextual feature of this site is the informal 'articulation' process that occurs in the department at the beginning of term when instructors meet and discuss the courses they are scheduled to teach with those who taught the same courses the previous term. In addition, most instructors file their lesson plans and teaching materials in course binders which are often 'passed on' to the next person teaching the course. There are no end-of-term departmental exams; however, individual instructors are expected to evaluate students in conformance with the general guidelines stated in the course descriptions and according to the college-wide grading system established for non-credit courses. ANNIE: South Bay Community College Approach to Instruction Several features in Annie's approach to instruction are particulary significant. First, Annie emphasized the importance of providing as much student-centered time in speaking classes as possible. In the post-observation conferences, Annie frequently commented on the fact that she tried to keep teacher-fronted instruction to a minimum. She believed that students' time was "wasted" unless they were given the maximum amount of class time to be involved in student-interaction. It is very clear from the observation data that Annie operationalized this view in her classroom lessons. In the four lessons observed, teacher-fronted tasks comprised 3 3 % of lesson time. In addition, Annie tended to use this format as a transition segment between tasks as opposed to a teaching format. These segments consistently involved a brief introduction to the content in the new activity and/or task, a reference to the language structure (which students had usually worked on for homework) and an explanation with examples of the interaction gambits (usually for the leaders) to be used in the group task. In addition, Annie might briefly follow up on the task the students had just completed but this was not a regular pattern. In the conferences, Annie explained that her decision to follow-up on tasks depended on whether she felt this was necessary. If as she was monitoring the interactive task she saw that they were completing it successfully, she felt it was redundant to then have whole class closure on it. A second marked feature of Annie's instructional approach relates to her organization and implementation of the course. Annie included two main components in her curriculum plan: academic discourse functions (and language structures to express these) and interaction gambits. However, the observational data indicate that she added another organizational focus at the lesson planning level " topics. Within this topic organization, Annie worked on the curriculum level components. Since Annie felt very strongly that the students at an advanced level should be "moving away from concrete activities" and developing critical thinking skills, she tried to select materials that had "no right answers ..." and activities which focused on classifying, comparing and evaluating various points of view on a topic. Annie's approach to teaching language structure was also through content. She would give students articles to read with key language structures underlined and ask students to "study and memorize" these for homework. However, beyond emphasizing the importance of language patterns such as comparison and classification for academic work, there was no explicit lesson follow-up on this language-focused homework. Annie believed that by continually exposing students to these key language patterns through reading content and handouts, students would internalize the patterns and use them in class Interaction activities. Another criterion for selecting materials and activities was Annie's concern for building a positive classroom climate ~ "I never do anything like talking about stress, worry or anxiety because I want it to be a positive experience." Concern for students' anxiety level was also reflected in her reticence to give contact assignments to students since she felt that "going up to someone you don't know" creates anxiety. Perceptions of the Instructional Context institution: Annie felt that the Institutional involvement in setting course objectives and curriculum planning was minimal. Annie felt very positive about this minimal direction coming from the administration; "I don't want them involved beyond setting out basic outlines." In terms of teaching method, Annie commented that although there was not overt direction given, there was an unexpressed department norm for how classes were taught "so that we all teach in the same way". The impact of colleagues' views on Annie's approach was evident in her perception of an instructor-level norm for a fluency focus in speaking classes. She explained that before coming to the college, she used to teach speaking with a structural focus. However, when she began teaching at the college the instructors "who first initiated me into the way things were done felt strongly that even if you teach grammar in a speaking class, the students tend not to pick it up ...so I've changed my approach to a fluency one." students: Annie commented on the fact that there had been a shift to more international students and fewer community students in classes at the college. She felt that this was a limiting factor in that the groups were more homogeneous. Annie also expressed frustration with the lack of motivation for learning that she felt the international students had. She felt this was partly due to their age and not having clear career goals. These factors made it more difficult for them to apply themselves. However, this apparent lack of motivation had changed her attitude towards her students -"now I assume with international students that they're not going to apply themselves until I have evidence to the contrary." Previously she had assumed that because these were adult students, they were responsible and wanted to learn. She also assumed that adult students would approach coursework as she had; "As a student I always did my homework and got good grades. I assumed everybody was like that." Instructional Task: Perceptions and Implementation participation structure: Teacher Perceptions of Participation Structures Used whole class small groups pairs individual 20% 30% 30% 20% Annie believed that there should be very little teacher-fronted instruction in a speaking class since she didn't like "stealing the students' time" and wanted to "give them as much opportunity to interact" in class as possible. The implementation data is consistent with Annie's comments. Across the four lessons observed, whole class format comprised a minor proportion (33%) of total class time . The majority of this teacher-fronted time was used to introduce student-centered tasks. This usually consisted of a brief introduction to the topic, reference to the language pattern students might focus on during the task, and a reminder of the discussion leader's responsibilities (language gambits) for introducing and summarizing the discussion. Annie spent very little time, if any, on task follow-up and when she did, this was often in the form of transition comments moving students from one task to the next. When Annie did spend time on taking up tasks, this usually consisted of asking one or two of the group leaders for their groups' discussion conclusions. There was seldom any emphasis given to the reasons why groups had reached these conclusions. This reporting rather then explanatory focus was interesting given the importance Annie placed on developing students' critical thinking skills. In order to reduce teacher-talk time between student-centered tasks, Annie employed an interesting technique. She would hand out the task sheets for the next interaction activity while students were still involved in the current interaction task. This not only served as a signal to students that the present task was about to end, but meant that students were ready to attend to the next task after Annie's brief transition comments. Small group tasks were used extensively by Annie in her classroom lessons. Although Annie felt that only 30% of class time was spent in small group work, the implementation data indicates that this was in fact closer to 50%. In contrast to Annie's perception that she used pair work as often as group work, (30%), across these four lessons, pair tasks occupied only 18% of the total lesson time. In addition, Individual tasks, which Annie felt she used 10% of the time, comprised only 2 % of these observed lessons. types of activities: In the observed lessons, Annie used whole class format for introducing/taking up tasks and giving homework. Her pattern for Introducing a new topic activity and the first student-centered task, was to explain the content focus, draw student attention to language structures the task involved and provide gambits the group leader could use in the task itself. These gambits were usually ones for introducing the task to the group, and summarizing group member opinions. In addition, Annie frequently listed gambits the leaders could have used during the discussion in the task follow-up. These she usually wrote on the board so that students could copy them into their notebooks. When asked about the abbreviated (or lack of) follow-up for most tasks, Annie explained that if she saw that the students had dealt with the task successfully in their groups, it was "redundant" and a "waste of students' time" to go over the task with the whole group. When Annie did follow up on tasks, it usually involved asking for a brief report of the conclusions reached by one or two groups, and providing students with discussion leader language gambits. In student-centered format, tasks focused on a range of information-sharing activities. The majority were guided by worksheets Annie had taken from communicatively-based L2 texts. In addition, several of the discussions were based on articles that individual students selected as part of their course assignment (leading a discussion). teacher role: Annie perceived her role during student-centered tasks as "purely reactive". This involved "going around and making sure people were on task, ...monitoring and listening" and acting as a resource if students had questions. She believed that it was important for her to remain as a resource only since this "compensates for the teacher dominance in the teacher-fronted portions of the lesson" and "gives (students) more of a feeling of control". However, although Annie saw herself "more as a resource than a participant", observation data show that in fact, she spent a large portion of task time actively involved in group discussions. Her most usual routine was to listen for a few minutes, then ask a clarification question and then provide a related experience of her own. At times Annie would talk to the whole group; at other times she would become involved in talking with only one student while the rest of the group continued on with their own discussion. Annie's role during student-centered tasks also included clarifying task instructions, checking that there was a group leader, reminding leaders of their notetaking and summarizing responsibilities, and generally monitoring how students were doing. Annie/Lessons 1-4: Activities, Segments & Participation Structures Activities & Segments Participation Structures whole class student-centered T-C/S-C G/P/I LESSON #1 Activitv 1: evaluation time: 10 min. 10 min. # of segments 1 % of P.S. 100% Activitv 2: Jobs time: 55 min. # of segments % of P.S. 15 min. 2 27% 40 min. 2 73% Activitv 3:Sopliie's Choice" time: 31 min. 7 min. # of segments 4 % of P.S. 23% 24 min. 3 77% Activitv 4: "cocktail mix" + homewoik time: 17 min. 1 min. # of segments 1 % of P.S. 6% 16 min. 1 94% Lesson Total time: 103 min. # of segments % of P.S. 23 min. 8 22% 80 min. 6 78% LESSON #2 Activitv 1: IJFO's time: 52 min. # of segments % of P.S. 13 min. 2 25% 39 min. 2 75% Activitv 2: Dreams time: 51 min. # of segments % of P.S. 16 min. 4 31% 35 min. 2 69% lesson closm'e time: % of P.S. 2 min. 100% LESSON TOTAL time: 105 min. # of segments % of lesson 31 min. 7 30% 74 min. 4 70% Activities whole class student-centered LESSON #3 Activitv l:Dreams time: 27 min. # of segments % of P.S. 7 min. 2 26% 20 min. 1 74% Activitv 2: Superstitions time: 29 min. 11 min. # of segments 2 % of P.S. 38% 18 min. 1 62% Activity 3:stiip story time: 17 min. 3 min. # of segments 1 % of P.S. 18% 14 min. 2 82% Activitv 4: discussion time: 28 min. 7 nun. # of segments 2 % of P.S. 30% 21 min. 1 70% lesson closure 1 min. Lesson Total time: 103 min. # of segments % of P.S. 30 min. 8 29% 73 min. 5 71% LESSON #4 Activitv l:'hou.sekeeping" time: 10 nun. 10 min. Activitv 2:superstitions/science time: 55 min. 22 min. # of segments 2 % of P.S. 40% 33 nun. 2 60% Activitv 3:classification time: 12 min. 10 min. # of segments 2 % of P.S. 83% 2 min. 1 17% Activitv 3:'problems' Time: 34 min. 12 min. # of segments 3 % of P.S. 35% 22 min. 2 65% Lesson Total rime: 111 min. # of segments % of P.S. 54 min. 8 t9% 57 min. 5 51% SANDY: South Bay Community College Approach to Instruction A concern for meeting students affective needs was a consistent theme in Sandy's comments on her decision-malcing. She felt that collaborative learning activities were particularly important for fostering a positive classroom climate in which students could gain self-confidence in their language learning abilities. Through group interaction tasks she tried to encourage student respect for classmates as well as help them realize that they could learn from each other. I wanted them to know that... collectively they have everything I could teach them as far as being able to work around (problems with vocabulary). Sandy was particularly concerned that students have positive language learning experiences in the classroom. For example, in explaining why she tended to spend a fair amount of time on introducing new activities, Sandy commented that it was important for students to begin a task knowing exactly what to do because ... "they learn more from success". Sandy was very conscious of building confidence and avoiding anything that might undermine this process. When asked about her approach to giving corrective feedback, Sandy stated --I think the first thing you have to measure is the level of self-confidence. I don't want to do any-thing that is going to stifle their developing fluency. A second consistent theme in Sandy's comments about her instructional approach, was her concern for purposeful classroom activities. She felt it was essential that what she did in the classroom should be 'grounded' in the academic and/or 'real life' objectives she had set for the course. It was also important for her that her students felt that lessons involved "serious" language learning activities rather than "fun and games". For this reason many of the interaction tasks, which prim