LEARNER AUTONOMY PERCEPTION AND PERFORMANCE: A STUDY ON VIETNAMESE STUDENTS IN ONLINE AND OFFLINE LEARNING ENVIRONMENTS

Submitted by Dang Tan Tin B.A., M.A. in TESOL

A thesis submitted in total fulfillment of the requirements for the degree of Doctor of Philosophy

Faculty of Education La Trobe University Bundoora, Victoria 3086 Australia

March, 2012

STATEMENT OF AUTHORSHIP

This thesis integrates the publications authored by me or with my supervisor during my candidature. Most of them, as included in the end of this page, are modified where appropriate to fit into the structure of the thesis.

Other than that, except where reference is made in the text of the thesis, this thesis contains no materials published elsewhere or extracted in whole or in part from a thesis by which I have qualified for or been awarded another degree or diploma.

No other person’s work has been used without due acknowledgement in the main text of this thesis.

This thesis has not been submitted for the award of any degree or diploma in any other tertiary institution.

All research procedures reported in the thesis were approved by the Human Research Ethics Committee, La Trobe University.

Dang Tan Tin Date: 16/03/2012

Publications integrated in the thesis

Dang, T. T. (2010). Learner autonomy in EFL studies in Vietnam: A discussion from socio- cultural perspective. English Language Teaching, 3(2), 3-9.

Dang, T. T., & Robertson, M. (2009). Online communications: Students’ habits and community formation. Paper presented at the AARE 2009 Conference: Inspiring Innovative Research in Education, Canberra.

Dang, T. T., & Robertson, M. (2010a). E-behaviors and E-community Formation: An Investigation on Vietnamese EFL Students. Asian EFL Journal. Professional Teaching Articles, 46, 4-27.

Dang, T. T., & Robertson, M. (2010b). Impacts of Learning Management System on Learner Autonomy in EFL Learning. International Education Studies, 3(3), 3-11.

Dang, T. T., & Robertson, M. (2010c). Pedagogical lessons from students’ participation in Web 2.0. TESOL in Context, 20(2), 5-26.

Dang, T. T., & Robertson, M. (2010d). Responses to learning management system: A case study in higher education in Vietnam. Paper presented at the ACEC2010: Digital Diversity Conference, Melbourne.

i ACKNOWLEDGEMENTS

I highly appreciate the invaluable advice, guidance and suggestions that my principal supervisor Professor Margaret Robertson provided throughout this study. Her dedicated supervision, various detailed comments and encouragement from the research group initiative helped me a lot to get the thesis in shape. Professor Robertson’s supervision was also significantly important for helping me turn part of my thesis into publications in academic journals and conference presentations.

In addition, I would like to acknowledge the important guidance and dedicated supervision of Dr. Alan Williams, my co-supervisor, during the first phase of this study. I also acknowledge the much appreciated assistance and consultations of Professor Ramon Lewis on statistical procedures conducted in this study. Professor Lewis’ professional support is extremely valuable for the reliability and validity of the study.

I would like to express my sincere thank to Dr. Howard Nicolas, Dr. Christine Brew, Associate Professor Wan Ng, Dr. Keith Simkin, and Dr. Ria Hanewald, who patiently listened to my research interest and commented on my research focus. I am further indebted to the cooperation of colleagues and students from five universities in Vietnam for participating in this study.

This thesis would have never been possible without the financial sponsorship of La Trobe University Postgraduate Research Scholarship and Endeavour International Postgraduate Research Scholarship.

Finally, I would like to thank my family members, officemates, friends, colleagues, the family research group, the Vietnamese research group, the Q-research group, and the Qualitative research group, who saved time to comment on my study and encouraged me during this academic journey.

ii TABLE OF CONTENTS

Statement of Authorship i Acknowledgements ii Table of Contents iii List of Tables vii List of Figures ix List of Appendices x Abstract xi

Chapter One INTRODUCTION 1. General context of the study 2 1.1. The centralised mechanism of the national education system and its features 3 1.2. The national education objectives from 1945 to 2015 3 1.3. The national investments in education 5 1.4. The Vietnamese philosophy in education 6 2. Specific context of the study 7 2.1. General issues of EFL education in Vietnam 8 2.2. General description of University of Social Sciences and Humanities 9 2.3. General description of the EFL program 10 2.4. General description of the first year EFL students 11 3. A problem of the context 12 4. Objectives of the study 14 5. Research questions 15 6. Significance of the study 16 7. Limitations of the study 18 8. Definition of terms 18 9. Thesis organization 19 10. Summary 21

Chapter Two LITERATURE REVIEW 1. Introduction 23 2. Learner autonomy in foreign/second language education 24 2.1. Definition of learner autonomy 24 2.2. Roles of learner autonomy 25 3. Conceptualization of learner autonomy 28 3.1. Learner autonomy as an attribute of the learner 28 3.2. Learner autonomy as a reflection of the learning situation 31 3.3. Learner autonomy as an outcome of the learner’s interactions 34 3.4. Learner autonomy as a desire for more access, agency, and power 35 3.5. Establishing the perspective adopted in this study 36 4. Models of learner autonomy 38 4.1. Models of learner autonomy regarding stages of development 38 4.2. Models of learner autonomy regarding areas of control 40 4.3. Discussion on the models 42 4.4. Summary 42 5. Perceptions of learner autonomy 43 6. Performance of learner autonomy 45 6.1. Learner autonomy promoting practices in traditional classrooms 45 6.1.1. Using classroom activities to foster learner autonomy 45

iii 6.1.2. Innovating teaching methods to foster learner autonomy 47 6.1.3. Modifying curriculum to foster learner autonomy 49 6.1.4. Summary 49 6.2. Learner autonomy promoting practices with ICT support 50 6.2.1. Using ICT-supported activities in face-to-face classroom 50 6.2.2. Using ICT-supported activities outside of the class 51 6.2.3. Summary 54 7. Conceptual framework of learner autonomy adopted in the current study 55 7.1. Dimensions of learner autonomy 55 7.2. Mediating factors of learner autonomy 58 7.3. Conceptual overview of learner autonomy 59 8. Conclusion 60 9. Summary 61

Chapter Three METHODOLOGY 1. Research worldview 64 2. Research design 65 3. Study One 67 3.1. Objectives and hypotheses 67 3.2. Participants 68 3.3. Instrument development 69 3.4. Procedures 72 3.5. Results 72 4. Study Two 73 4.1. Objectives and hypotheses 73 4.2. Participants 74 4.3. Course design 75 4.3.1. Course structure 75 4.3.2. Course assessment method 76 4.3.3. LMS design 77 4.3.3.1. Theoretical framework of the LMS design 77 4.3.3.2. Design procedure and structure of the LMS 79 5. Instruments 81 5.1. Measuring students’ perceptions of learner autonomy 81 5.2. Investigating students’ LMS engagement 84 5.3. Investigating mediating factors on students’ learner autonomy 84 6. Procedures 86 6.1. LMS introduction and log generation 86 6.2. Questionnaire administration 88 6.3. Interview protocol 88 7. Issues of validity and reliability 89 7.1. Reliability 89 7.2 Validity 90 8. Ethical considerations 93 9. Summary 96

Chapter Four RESULTS 1. Results of Study One 97 1.1. Perceptions of learner autonomy 100 1.2. Similarities and differences in learner autonomy perceptions 106

iv 1.2.1. Similarities and differences among the four dimensions 106 1.2.2. Effect of gender, living place, and computer proficiency 108 1.3. Summary 111 2. Results of Study Two 111 2.1. General description of the sample 112 2.2. Reliability of the questionnaire instrument 116 2.3. Revisiting the similarities and differences in students’ perceptions 120 2.4. Relationship between learner autonomy perception and performance 125 2.5. Changes in the relationships between perception and performance 129 2.6. Mediating factors of learner autonomy performance 133 2.6.1. Effect of gender, computer proficiency, and teaching 134 2.6.2. Effect of other mediating factors on learner autonomy performance 139 2.6.2.1. Descriptions of students’ engagement in the LMS 140 2.6.2.2. Mediating factors in offline learning environments 145 Effect of preference on learner autonomy performance 145 Effect of motivation on learner autonomy performance 150 Effect of attitude on learner autonomy performance 154 Summary 156 2.6.2.3. Mediating factors in online learning environments 157 Effects of technological competence 158 Effects of attitude on learner autonomy performance 159 Effects of goal orientation on learner autonomy performance 161 3. Summary 163

Chapter Five DISCUSSION 1. Perceptions of learner autonomy 166 1.1. Overview of the four-dimension model of learner autonomy 166 1.2. Items in the four dimensions of learner autonomy 169 1.3. Analysis of the four-dimension model of learner autonomy 175 2. Similarities and differences in learner autonomy perceptions 178 2.1. Similarities and differences among the four dimensions 178 2.2. Effect of gender, living place, and computer proficiency 181 3. Relationship between learner autonomy perception and performance 183 4. Changes in the relationships between perception and performance 185 5. Mediating factors of learner autonomy performance 188 5.1. In the offline learning environment 188 5.1.1. Effect of gender, computer proficiency, and teaching practice 189 5.1.2. Effects of preference 191 5.1.3. Effects of motivation 194 5.1.4. Effects of attitude 196 5.2. In the online learning environment 199 5.2.1. Effects of teaching practice 199 5.2.2. Effects of gender and computer proficiency 200 5.2.3. Effects of attitude 204 5.2.4. Effects of goal orientation 206 6. Conclusions 208 7. Summary 210

v Chapter Six CONCLUSIONS AND IMPLICATIONS 1. Summary 211 2. Theoretical contributions 213 3. Methodological implications 214 4. Pedagogical suggestions 216 5. Limitations 218 6. Directions for further research 219 7. Conclusions 221

REFERENCES 222

APPENDICES 245

vi LIST OF TABLES

Table 2.1: Five-level model of learner autonomy 38 Table 2.2: Adapted model employed in the current project 42 Table 2.3: Index of learner autonomy dimensions in three processes 56

Table 3.1: Sample of the 51-item list categorized in three core processes 71 Table 3.2: Statements related to processes in ICT-supported environment 71 Table 3.3: Course assessment weighting load 76 Table 3.4: Extracted items from the pre-test questionnaire in respective dimensions 82 Table 3.5: Timeframe of the main study procedure 86

Table 4.1: Distribution of gender and living place in the sample 98 Table 4.2: An extract from the descriptive statistics of the 62 questionnaire items 99 Table 4.3: KMO and Bartlett’s Test of the sample 100 Table 4.4: An extract of the total variance explained 101 Table 4.5: Extract of factor analysis of the 25 items on learner autonomy perceptions 103 Table 4.6: Internal consistency of the four factor scales 104 Table 4.7: Positive correlation coefficient among the four dimensions 105 Table 4.8: Descriptive statistics on mean difference among dimensions (perception) 107 Table 4.9: Statistical mean difference between dimensions (perception) 107 Table 4.10: Effect of gender and living place on computer proficiency 109 Table 4.11: Descriptive statistics of gender and living place on computer proficiency 109 Table 4.12: Effect of the three general factors on perceptions of learner autonomy 110 Table 4.13: Distribution of students in groups 113 Table 4.14: Students’ level of self-reported computer proficiency 113 Table 4.15: Test of gender ratio on different groups 114 Table 4.16: Descriptive ratio of females to males in three groups 114 Table 4.17: Test of gender and group on self-reported computer proficiency 114 Table 4.18a: Cronbach’s alpha of Dimension 1 in Study One and Two 117 Table 4.18b: Cronbach’s alpha of Dimension 2 in Study One and Two 118 Table 4.18c: Cronbach’s alpha of Dimension 3 in Study One and Two 118 Table 4.18d: Cronbach’s alpha of Dimension 4 in Study One and Two 119 Table 4.19: Number of items and Cronbach’s alpha of the four dimension scales 120 Table 4.20: Correlations among the four dimensions (perception) 121 Table 4.21: Correlations among the four dimensions (perception) in two studies 121 Table 4.22: Levels of the four dimensions in learner autonomy perception 122 Table 4.23: Level differences in the four dimensions of learner autonomy perception 122 Table 4.24: Effect of gender on the four dimensions of learner autonomy perception 123 Table 4.25: Effect of computer proficiency on learner autonomy perception 123 Table 4.26a: Effect of student groups on learner autonomy perception 126 Table 4.26b: Effect of student groups on pre-test learner autonomy performance 126 Table 4.27: Correlations between perception and pre-test performance 127 Table 4.28: Level comparison between perception and pre-test performance 128 Table 4.29: Descriptive statistics on levels of perception and pre-test performance 129 Table 4.30: Correlations between perception and post-test performance 130 Table 4.31: Level comparison between perception and post-test performance 131 Table 4.32: Mean comparison of learner autonomy between pre-test and post-test 132 Table 4.33: Descriptive statistics on students’ levels of learner autonomy performance 133 Table 4.34: Effect of gender on learner autonomy performance 134

vii Table 4.35: Statistics on the difference between males and females in Dimension 2 135 Table 4.36: Effect of computer proficiency on learner autonomy performance 136 Table 4.37: Statistics on the levels of Dimension 2 by computer proficiency 136 Table 4.38: Mean difference of three groups on performance in the post-test 138 Table 4.39: Statistics on the level of performance in the post-test across three groups 138 Table 4.40: Comparison of students’ level of performance in three groups 138 Table 4.41: Total number of thread starters and replies in the LMS 141 Table 4.42: Detailed number of thread replies to thread starters 142 Table 4.43: Time of postings in the LMS 142 Table 4.44: Students’ number of postings on the LMS from week 3 to week 16 143

Table 5.1: Dimensions identified from the current study with those in the literature 167 Table 5.2: Thirteen items included in both studies 168 Table 5.3: Items in Dimension 1 and their interpreted attributes 170 Table 5.4: Conceptual aspects and attributes contributed by the items in Dimension 2 171 Table 5.5: Conceptual aspects and attributes contributed by the items in Dimension 3 172 Table 5.6: Conceptual aspects contributed by the items in Dimension 4 174 Table 5.7: Dimensions of learner autonomy developed from Vietnamese samples 175 Table 5.8: Leaner autonomy dimensions and their attributes 176 Table 5.9: Allocation of items to the model of learner autonomy from literature 177 Table 5.10: Correlations among the dimensions of learner autonomy performance 179

viii LIST OF FIGURES

Figure 2.1: Four perspectives on the construct of learner autonomy 37 Figure 2.2 : Processes of learner autonomy 57 Figure 2.3: Learner autonomy and its mediating factors 59 Figure 2.4: Theoretical position 59

Figure 3.1: Diagram of the sequential mixed methods design employed in the project 66 Figure 3.2: A snapshot of the LMS interface 80 Figure 3.3: The virtual course layouts of different classes 87

Figure 4.1: Scree plot test with Eigenvalues greater than 1 100

Figure 5.1: Four dimensions of learner autonomy 208 Figure 5.2: Mediating factors of learner autonomy 209

ix LIST OF APPENDICES

Appendix 3A: The Fifty-Nine Items Categorized in Three Core Processes 245 Appendix 3B: Survey Questionnaire of Study One 247 Appendix 3C: Pre-test Questionnaire in Study Two 253 Appendix 3D: Post-test Questionnaire in Study Two 258 Appendix 3E: Numbers Associated with Questionnaire Items in Two Studies 260 Appendix 3F: Semi-Structured Interview Questions for Students 261 Appendix 3G: Ethics Approval and Related Documents 263

Appendix 4A: Descriptive Statistics of the 62 Questionnaire Items 280 Appendix 4B: Factor Analysis Procedures 282 Appendix 4C: Internal Consistency Tests of the Four Dimension Scales 292 Appendix 4D: Effect of Three General Factors on Perceptions of Learner Autonomy 293 Appendix 4E: Dimensions of Learner Autonomy and Other Factors 294 Appendix 4F: Two Excerpts of the Interview Transcripts 295 Appendix 4G: Examples of Themes and Coding 306

Appendix 5A: Allocation of Items to the Model of Learner Autonomy 308

x ABSTRACT

Part of the dilemma facing educational reforms in Vietnam as in other Asian contexts is how to encourage more independence in students’ learning approaches. With the objective of fostering learner autonomy in the higher education sector in Vietnam, this study was designed to understand the perception and performance of learner autonomy from the perspective of undergraduate students majoring in English as a Foreign Language (EFL). Personal and socio- cultural factors are acknowledged as mediating st udents’ learning approaches. Using a mixed methods design, the research consisted of two sequential studies. To understand students’ perceptions of learner autonomy Study One relied on a 62-item questionnaire. Valid responses were received from 562 EFL undergraduate students in four universities. The analyses identified four dimensions of learner autonomy, namely Monitoring learning processes, Goal-setting and evaluating learning, Using ICTs in learning, and Initiating learning opportunities. These dimensions appeared to be interrelated but perceived at different levels. The gender difference was only found in Goal-setting and evaluating learning. Females tended to perceive this dimension at a higher level than males. Study Two used the four-dimension model of learner autonomy observed in Study One to investigate the relationship between perception and performance of learner autonomy. It was conducted with a cohort of 247 EFL first-year students in one university in Vietnam. A revised version of the questionnaire from Study One was used to measure the participants’ perception and performance of learner autonomy at the beginning and end of the course. The online learning management system associated with the course provided additional information on students’ engagement. Finally, semi-structured interviews were conducted with eleven students at the end of the course to gain more insights into the mediating factors on their autonomous learning behaviours. Statistical analyses showed that students’ perception and performance of learner autonomy in each dimension were positively correlated with each other at the beginning of the course. However, this relationship was not confirmed at the end of the course, probably due to examination constraints. Other theme analyses suggested that preference, motivation, and attitude prominently contributed to the shaping of students’ autonomous learning behaviours in the offline learning environment. Students’ performance of learner autonomy in the online space was likely mediated by technological competency and goal orientations. These findings call for a holistic approach in educational reforms to foster learner autonomy, particularly in such an examination-oriented context as Vietnam.

xi

Chapter One

INTRODUCTION

______

Learner autonomy in language learning has been increasingly researched over the last three decades. It has often been described as enabling students to be more active, engaged, motivated, responsible and efficient in their learning process. However, most of these descriptive claims have not been consistently supported by empirical research (Nguyen, 2009). In addition, learner autonomy has been considered a complicated and multifaceted construct (Smith & Ushioda, 2009). It can be manifested in different learning behaviours, depending on the particular socio-cultural contexts. The way a student exercises learner autonomy in one situation is not necessarily the same in another similar situation. Therefore, it is important to understand how students perceive learner autonomy and how they perform as autonomous learners in a local context to design effective methods to foster this capacity.

Investigations into local students’ perceptions and performance of learner autonomy are of even greater importance in situations undergoing change, such as the contemporary higher education sector in Vietnam. Education in Vietnam has been centralised since the mid-1940s. This has restricted the ability of educators to respond appropriately to local variations (Pham, 2010). Learning materials, for example, are rarely negotiated at the university level; and standardized examinations are the norm. Meanwhile, learner autonomy capacity is generally a requirement of the labour market. As a result, the development of learner autonomy has been included in the national education objectives, including some initial steps to decentralise the higher education sector (Pham, 2010). However, local institutions have not actually been given much autonomy (Dao & Hayden, 2010) and reforms continue to be on the discussion agenda (Harman, Hayden, & Pham, 2010). Hence, it is important to make changes to nurture students’ learner autonomy capacity.

The changing situation in the Vietnamese higher education sector is also reflected through teaching and learning practices. Vietnamese students are traditionally known as passive rote ______Chapter 1 Page 1 learners; and teachers are used to dictating to the class (Pham, 2010). Technological infrastructure at universities is limited; and the integration of technology into the curriculum is scarce (Dang, 2010). However, most university students have internet access at home these days; and many demand a more learner-oriented teaching approach. Others still tend to receive their lessons passively due to their familiarity with the knowledge transmission teaching approach at the high school level. Thus, both active and passive learning behaviours currently co-exist in this changing system.

This research study is designed to investigate local students’ perceptions of learner autonomy and their performance as autonomous learners. Both offline and online environments of learning English as a Foreign Language (EFL) are taken into consideration. The study aims at understanding how learner autonomy is valued and actualized in such a changing context. Such insightful understandings can then provide pedagogical suggestions for the local educational system and contribute to the process of promoting learner autonomy as one of the national education objectives. The following section of this chapter presents an overview of the study. It starts with a description of the study context so as to identify the challenges related to the goals of developing learner autonomy in education. It then proposes the objectives, research questions, and significance of the study. Finally, it addresses the limitations of the study and the organization of the thesis.

1. General context of the study

Vietnam is well known throughout the world, particularly in relation to the Vietnam War, which is evidenced by more than thirteen million entries on Google with the key words “Vietnam war.” Its past turbulent history is one of the reasons for different strategic changes in the national education system. This section describes the centralised mechanism that the country has adopted for its educational system since its declaration of independence in 1945. It focuses on the education objectives that Vietnam has targeted in different periods of the country’s development. The section also highlights the capital investment in education to illustrate the country’s efforts to improve its human resources. Finally, it introduces the educational philosophy adopted in Vietnam in relation to the country’s history and the importance of Western materials in the contemporary era of globalization to depict a changing Vietnam.

______Chapter 1 Page 2

1.1. The centralised mechanism of the national education system and its features

Vietnam, a socialist country in South East Asia, has employed a centralised management mechanism in education for almost six decades. The Ministry of Education and Training (MOET) is the authority which manages both the managerial and academic activities of the national education system (Dang, 2010). There are concurrently two types of schools in Vietnam, public and non-public. The former is totally funded by the government while the latter is not. However, both are monitored by MOET in relation to student recruitment processes, student intake quotas, areas of training, curriculum, testing schemes, and even teaching methods. All strategic developments and policies of the schools are also regulated by MOET.

Standardized examinations are an important feature of the local education system that shapes classroom interactions. National examinations are organized every year for high-school graduation and public university entrance. Several other standardized examinations are also run by the Department of Education and Training in each local province or equivalent area every semester. Therefore, classroom teaching and learning activities are often regulated by these examinations. Teachers are normally required to follow certain classroom procedures and students are expected to learn from textbooks and their teachers. As a result, interactions between teachers and students are usually much higher than those among students themselves.

1.2. The national education objectives from 1945 to 2015

Reflecting the historical and developmental landmarks of the country, over the last six decades, the Vietnamese education authority has shifted its major goals several times (World Bank, 2006). After the independence declaration in 1945, fighting against illiteracy was the goal of the national education system. In 1950, the first education reform started, focusing on educating the young generation to become future citizens, loyal to people’s democracy regime, competent to serve people, and resistant to war. During the temporary separation of North and South (1955-1975), the second reform was launched, aiming at fostering young people to become those who had all-round development in all aspects, and were good citizens and officials. In the pre-reform phase of the third reform (1975-1980), the objectives of the education authority were to remove the leftover influences from the colonial education system

______Chapter 1 Page 3 and implement anti-illiteracy activities for people in the age of 12-50. The third reform (1981- 1985) focused on providing the younger generation with knowledge of science, technology, culture, and ideology.

The fourth reform (1986-2005) was a period of significant changes which resulted from the major reform of the nation as a whole. Suffering from the economic crisis in 1986, the country moved from the centralised operating system to the socialist-oriented market mechanism. In general education, it continued to make modifications to curricula objectives, update textbooks and improve the faculty. In vocational education, the main objective of human resource training was shifted from serving state-owned companies to satisfying the market economy where private companies were encouraged. In higher education, the training content was diversified to meet the learning needs of the society and the labour market.

Since 2005, the national education objectives have aimed at enhancing every citizen’s knowledge base, indicating the local authority’s increased attention to the importance of knowledge in this era of globalization. In the period of 2005-2010, building a learning society was the goal of the national education authority. People of different ages were encouraged to develop a learning capacity to sustain life-long learning activities. Learning how to learn effectively was strongly emphasized. Linkages among different existing types of training were therefore developed, making multiple learning pathways available for learners. Similarly, education for all is the objective of the education sector for the period of 2010- 2015. It focuses on improving childcare and early education, internationalizing primary and secondary education, meeting the learning demands of youngsters and adults, creating a knowledge environment, and achieving equity in access to education quality.

Particularly in higher education, the objective indicated in the Higher Education Reform Agenda of 2006-2020 is to train highly qualified human resources that meet the country’s socio-economic development requirements and people's learning needs (MOET, 2005). Of the seven specific aims of the agenda, there are four radical improvements which involve academic staff development; quality assurance and an accreditation system; greater autonomy and accountability for training institutions, and the use of information technology and media in teaching. These have been reflected in different strategies such as sending academic staff abroad for further training, calling for teaching methodology renovation, designing core

______Chapter 1 Page 4 training programs for each area of specialization, and providing every public school with modern computers, fibre-optic internet connection, and other educational applications.

Reflected through the education objectives over the last decades are the government’s efforts in equipping its citizens with suitable knowledge and skills for a certain period of the country’s development. While the aim of producing perfectly knowledgeable citizens in the late 1970s significantly reflected a political ideology, the aim of facilitating sustainable learning in the twenty-first century is demanded by the age of information, globalization, and the knowledge economy. An ability to learn from and deal with new challenges is more important than the acquisition of some existing knowledge. Therefore, the contemporary structure is designed to provide multiple learning pathways for learners. The proposal is that learners need to develop an autonomy capacity to navigate through their learning paths successfully. To achieve these educational objectives, the government has increasingly invested in the education sector. The following section describes these investments in recent years to provide a more solid background for the study.

1.3. The national investments in education

To achieve the educational goals and meet the requirements of national industrialization and modernization by 2015, the education sector has increasingly received investments for renewal during the last decade. The budget share for this sector started at 3.5 percent of the national GDP (Gross Domestic Product) in 1994, increased to 4.6 percent in 2004 (World Bank, 2006) and reached 6.5 percent in 2006 (Vu, 2007). The education sector also received a larger share of the total public expenditure than other sectors; and this share continued to increase, being 14 percent in 1997, increasing to 18.6 percent in 2005 (World Bank, 2006) and stabilizing at 20 percent during 2007-2009 (Mai, 2009). The amount of funding which is directed to education is expected to reach more than 20 percent in the next few years (Mau & Quoc, 2007), placing Vietnam in the group of countries with the highest level of investment in education (Hong, 2010). These budget changes illustrate the government’s strong commitment to improving the education sector.

The vocational and higher education subsectors received an allocation of 26.7 percent of the total budget for education in 1998, and this amount decreased to 22.3 percent in 2002.

______Chapter 1 Page 5

Accordingly, public expenditure on these two subsectors also declined from 12.4 percent to 9.7 percent during this period (World Bank, 2006). The declining trend was the result of the government’s policy to share the costs of tertiary education with learners and their families. Since 2006, it is expected that the budget share for this subsector will increase to be aligned with the Higher Education Reform Agenda 2006-2020 as indicated in the previous section. However, there has been neither published nor official data on the level of cost contributed to education by learners. The issue of cost-sharing has been the subject of many debates; and tuition fees at tertiary education institutions have increased almost every year.

While financial support is necessary for the Vietnamese higher education sector to achieve its objective of the development of learner autonomy, the cultural traits and professional practices of the sector must be considered for they are decisive factors for future success. Therefore, it is important to understand the teaching and learning practices which can significantly contribute to the shaping of learner autonomy in the local context.

1.4. The Vietnamese philosophy in education

Being part of Eastern culture, the Vietnamese educational philosophy is traditionally more associated with absorbing and memorizing than experimenting in the learning process and producing knowledge. Teachers are the class authority; and students are expected to personally and academically obey their teachers (Nguyen, 2002). Negotiations and collaborations between teachers and students or among students themselves are rarely seen in traditional classrooms (Pham, 2010). Deeply rooted in Confucian ideology, students are typically believed to develop best if they behave morally, come to class regularly, listen to the teachers attentively, understand the lectures clearly, and memorize well every single word. Therefore, there are many standardized examinations to test students’ memorized knowledge. The ability to apply knowledge learnt at school to real life situations and produce new knowledge is not the main focus of schooling.

However, with different bodies of knowledge imported from different countries in the era of globalization, Vietnam is currently in a stage of transition to the modern philosophy of Western education. Advantages of modern pedagogies developed from the West are acknowledged by local educators; and different training models from other countries are taken

______Chapter 1 Page 6 into consideration. However, the adoption of Western practices in the local context has resulted in a number of difficulties. For example, when group work activities were employed to enhance students’ active participation in class, teachers could not monitor the situation effectively because of the large class size (Dang, 2010). In addition, when the communicative language teaching method was used in EFL training, students were in danger of failing the standardized examination at the end of the course. Such issues become more serious with increasing pressure from the public.

Another difficulty has emerged in the design of training content. MOET has started to encourage teachers at the tertiary level to take charge of designing course materials. However, their design is restricted within tight guidelines. As a result, it has been suggested that most of the curriculum content is still not suitable for students (Pham & Ngo, 2008). Some teachers have attempted to change some parts or even the whole content of the lesson to foster students’ active learning, without officially indicating any change. In addition, there are no reports of students’ voices being taken into account in the process of designing the training content. That is, there are limited opportunities for teachers to customize the training content. Students are told to develop their learner autonomy, but their participation in the program design is not seriously considered. Consequently, many students fail to meet the program objectives (Hoang, 2008; Nguyen, 2008; Thanh, 2008).

The following section specifically describes the conflicts in one particular context, that is, in the area of EFL training. It considers specific issues relating to EFL education programs in the Vietnamese higher education sector in general and in the University of Social Sciences and Humanities, Ho Chi Minh City, in particular. These situational descriptions serve as a background to the discussion about students’ perceptions and performance of learner autonomy toward the end of the research.

2. Specific context of the study

EFL education in Vietnam has undergone several changes, reflected in the national language policy of different periods (Phan, 2009). However, EFL has recently gained dominance in the education system, and the University of Social Sciences and Humanities, Ho Chi Minh City, is one of the largest universities in the south of Vietnam delivering such programs at both

______Chapter 1 Page 7 undergraduate and postgraduate levels. Students are often prepared with macro skills when entering an undergraduate program and streamed into a specialization after that. As this study focuses on undergraduate students of such EFL programs, a more detailed description of them is presented to profile the study context.

2.1. General issues of EFL education in Vietnam

The position of EFL training in Vietnam has fluctuated during the development of the country due to political changes. English was not the foreign language taught in schools when the country declared its independence in 1945. English only started to be taught in certain schools in 1970 while Chinese, Russian, and French were the main foreign languages being taught (Le, 1998). English then became more popular under the influence of the United States of America until 1975 (Nguyen & David, 1999). However, together with French, English was significantly diminished by Russian in the period of 1975-1986 because of the political situation. Since 1986, the collapse of the Soviet Union, and the establishment of the Open- Door Policy have enabled English to become the most favoured foreign language in Vietnam. English is the language that the Vietnamese use to communicate with international partners for trading, scientific and technical purposes (Phan, 2009). As a result, the majority of students have selected EFL to study in schools and universities since the 1990s. These days, English proficiency is one of the key criteria for job opportunities and promotion in both public and private sectors.

Given the importance of EFL in both academic advancement and the labour market, the delivery of EFL training has undergone several changes, reflecting the influence of the culture, the national language policy, and the import of learning materials in English. Initially, in English language classes, students were generally expected to simply learn by rote, as they did in other subjects, because this learning approach had been used for a long time (Nguyen, 2010). In addition, the lack of qualified EFL teachers at both the school and tertiary education level during the 1990s resulted in a lack of improvement to the quality of EFL training (Vu, 2000). Recently, with the development of technology and the trend towards globalization, EFL stakeholders are trying to update and improve the quality of EFL training. However, this reform process has been slow due to different contextual reasons such as the centralised mechanism of the education system, classroom settings, and the conflict between the new

______Chapter 1 Page 8 teaching methods and traditional beliefs. As a result, the current situation in regard to EFL training is very challenging for teachers, students, and even the educational authority.

The next section describes the specific context of the study, the EFL program in the Faculty of English Language and Literature in the University of Social Sciences and Humanities, Ho Chi Minh City. It is necessary to note that being a public university, the context of EFL training in the University of Social Sciences and Humanities is very similar to that in other public universities across Vietnam.

2.2. General description of the University of Social Sciences and Humanities

As one of about 400 tertiary institutions in Vietnam (Tien, 2009), the University of Social Sciences and Humanities, Ho Chi Minh City National University, is directly under the management of MOET. Therefore, its general strategic policies and development directives are developed and determined by this centralised government authority. This public university is funded by the government and the staff are paid from the national salary budget. It consists of twenty-two faculties and schools, employing more than 450 teaching and research staff, and serving more than 31,000 students. It offers both undergraduate and postgraduate programs in areas of Social Sciences and Humanities such as Vietnamese studies, sociology, linguistics, English, French, and journalism.

Given the centralised budget mechanism, the university infrastructure currently looks fairly traditional, offering limited resources to students. Blackboards, chalk, fixed table rows, and classes of up to sixty students are seen in most of the classrooms. Only a few classes are equipped with moveable single study desks. Teachers frequently have to use the sound system installed in the classroom to be heard by students. In addition, the library is too small in comparison to the number of students, and one can always see many students using the corridors or the corners of any building as private study spaces. Although some national policies on the establishment and usage of Information and Communication Technologies (ICTs) have been issued, the current status of available technology and its application in the classroom is very low. The university website is basically used to give announcements; and there is no online space for student-teacher interactions. There has been no evidence of integrating technology applications such as Blackboard or Moodle in school programs.

______Chapter 1 Page 9

Given the classroom setting and available infrastructure, the teaching and learning practices have not changed much during the last decade. Teacher dominated and rote learning styles are seen in most of classes. Although there have been several attempts to update these pedagogies by shifting to a more student-centred approach and encouraging students’ active participation in class, this transition has been very slow. It is almost impossible for a teacher in a class of over fifty students to pay adequate attention to individual students, especially when they frequently expect to be evaluated. In addition, the fixed table settings neither support students to form work groups nor provide suitable space for teachers to monitor group activities. As a result, the majority of teachers revert to teacher-focused practices, even though many of them would prefer to use other approaches.

In addition, the evaluation of students’ achievement has been principally based on individual assessment. Students are often required to pass a mid-term test and a final exam to complete a course. Classroom participation, co-operative skills or the capability to engage in project work contributes very little or nothing to the course evaluation weighting. Although recognized as a shortcoming, this evaluation method has basically remained unchanged simply because it is the procedure required by the centralised body. Evaluation, in general, needs to be monitored and, if possible, standardized. Teachers are given limited authority to design course assessment methods. Hence, teachers have to draw students’ attention to achievement on individual tests in order for them to pass the course.

2.3. General description of the EFL program

The Faculty of English Language and Literature is one of the largest units in the University of Social Sciences and Humanities, Ho Chi Minh City. It consists of about eighty teaching staff (both permanent and contract) and aims to prepare undergraduate students with academic skills such as a high level of English proficiency, a good understanding of English and American literature, translation competence, and/or pedagogical knowledge. It also offers postgraduate programs such as the Postgraduate Diploma and Master of Teaching English to Speakers of Other Languages (TESOL), Master of Applied Linguistics, and Doctor of Education. Some of these programs are delivered in cooperation with international education providers.

______Chapter 1 Page 10

All the undergraduate programs in the Faculty of English Language and Literature generally consist of two phases. Courses offered in the first phase, called EFL courses, aim to increase students’ English competence. The second phase includes courses for a particular specialization such as literature, translation or pedagogy. Students must select a specialized major once they successfully complete the required courses in the first phase. The language of instruction used in these English-related courses is English; while Vietnamese is used for the others such as Communist Party History and World Civilization History, which are compulsory for all students across the university. This thesis focuses on the students in the EFL courses.

Being able to work with resources in English, EFL teachers and students obtain a rather different set of beliefs and practices in comparison with those in other faculties. They are often more open and adaptive, especially when teachers in these EFL courses are in their twenties or early thirties. Some of them have even received overseas training. Therefore, they tend to be more active in modernising their teaching practices and are willing to accept challenges as they aim to improve the quality of their teaching and course delivery compared to their own experiences in school. Similarly, students in these courses are generally more active and interactive. Forming study groups and looking for collaborative opportunities is often integrated in their learning. More students have begun to practice peer evaluation instead of relying solely on their teachers.

2.4. General description of the first year EFL students

Many of the first-year EFL students in the Faculty of English Language and Literature experience difficulties adjusting to the university learning environment because it is so different from high school. Insufficient oral English competence is a common problem experienced by the majority of students. They also tend to be unfamiliar with learning activities and assignments where there is a low level of teachers’ control. However, most students adjust to this teaching style after one semester. Another problem facing students is finding learning materials. Unfortunately, the resources offered by the university are quite limited. In an individual interview (reported in Dang, 2010), a student said that he did not know if there was a computer lab or a similar place for learning activities. Another student who had been in the EFL program for two years also gave the same response. Students often

______Chapter 1 Page 11 have to buy reference books, newspapers, magazines, and other materials themselves as the library materials are very limited or do not suit their purposes.

As not all students have computer experience prior to entering tertiary education, their online activities are rather varied. Most of those who come from Ho Chi Minh City or another large city have used computers and the Internet at home, probably on a daily basis. However, the majority of students who come from other provinces are not very familiar with computers and internet resources. As a result, some students frequently take part in the virtual environment via email, instant messenger, discussion boards, and social networking sites, to name a few. However, some do not, as going online is not part of their day-to-day routine (Dang & Robertson, 2010a). In addition, most of the students who come from a large city seem to be more confident in their first semester, while the others take longer to adjust to the pace of life in their new surroundings. These differences have resulted in variations in students’ learning perceptions and behaviours.

3. A problem of the context

Given the situation regarding educational funding described in the previous section, centralised mechanism, teaching practices, learning behaviours and social contexts, overall the Vietnamese education system has achieved certain encouraging accomplishments (for more details, see World Bank, 2006). However, graduates from Vietnamese universities have generally been criticised for not having adequately achieved the training objectives. Therefore, this labour force is not ready for the current era of globalisation and economic development especially since Vietnam was admitted to the World Trade Organization in 2006 (T. V. Nguyen, 2008). This failure is considered to be due to the way in which the educational objectives have been designed. Nguyen reported that most of the training objectives in higher education are general and do not match the demands of the labour market because the training objectives have not been developed from contemporary social situations. As a result, UNESCO warned Vietnam that its education system was unlikely to achieve the educational goals, including the improvement of the quality of higher education by 2015 as planned (Luu, 2008).

______Chapter 1 Page 12

These shortcomings can be empirically illustrated by the following statistics. In 1999, the quality of education and training of Vietnam National University (Hanoi) was ranked the seventy-fourth out of seventy-nine of the best multi-disciplinary institutions in Pacific Asia (Asiaweek, 1999). In mid-2008, it was ranked the sixtieth in the top 100 universities in South East Asia, and its counterpart in Ho Chi Minh City was not on the list at all (Cybermetrics Lab, 2008). In addition, since 2000, only around 50 percent of graduates were able to find employment, and only 30 percent of these obtained work in the areas for which they were trained (T. V. Nguyen, 2008). In foreign language education, a survey in 1996 affirmed that only 35-37 percent of graduates from Bachelor programs in foreign languages achieve the level of expertise claimed by their qualifications (Tran, 1996). This number is even smaller in other general estimates which indicate that less than ten out of every fifty graduates from such English programs have the skills needed to work as interpreters, translators, tour guides or teachers of English (Pham, 1999). Recently, in another survey of fifty-nine universities offering programs other than English, only 49.3 percent of students were reported to meet the objectives of their English training programs (Thanh, 2008).

The immediate reason for such a low number of graduates obtaining jobs in their area of training is generally regarded as being due to their low level of competency (Bui, 2006). They are often observed to lack three main interrelated attributes (Pham & Ngo, 2008). First, they do not have appropriate contemporary knowledge because the training content was out of date (also in Hoang, 2008). Second, they do not have the necessary professional techniques in their specialization to satisfy the general expectations of the labour market (also in Hoang, 2008). Third, they do not perform well in a number of generic skills such as being independent, pro- active, critical, and collaborative as these capacities are not sufficiently cultivated and properly nurtured in schools (also in P. Nguyen, 2008). While the first two weaknesses might be more related to curriculum and technology infrastructure, the third is more associated with teaching practices and learning behaviours (Pham & Ngo, 2008).

The students’ weakness in generic skills has therefore resulted in the inclusion of the ability to learn independently in the national education objectives for 2005-2010. The students’ low level of ability in this area, often called learner autonomy capacity, is of concerned to many educational stakeholders (Nguyen, 2006; Thanh, 2009) who are discussing ways to improve this ability in the contemporary socio-cultural environment. Such discussions also include

______Chapter 1 Page 13 critiques and sometimes criticisms of traditional teaching and learning practices. The discussions often conclude with an urge for changes in educational policy or certain behaviours of teachers and students. However, these arguments include little, if any, of the students’ voices.

It can be seen that the importance of learner autonomy capacity has been recognized in the local EFL training context; and pedagogical practices need to foster the development of this capacity in students. Once students become more autonomous in their learning, they can navigate different learning environments, both formal and informal, more successfully. Learner autonomy is even more important these days, as students have more access to informal online learning spaces with little supervision, given the support of the technological infrastructure. Fostering this ability would be more effective if the students’ current perceptions and practices in relation to learner autonomy were taken into account and better understood. This is particularly important in a changing environment, such as in contemporary Vietnam.

4. Objectives of the study

Given very scarce research on learner autonomy in Vietnamese EFL education, this study is designed to provide insights into this construct in the local context. It appears that only two studies have been conducted in this area in Vietnam. However, both (Nguyen, 2009; and Trinh, 2005) take a more confirmatory perspective when investigating this variable. The participants’ perspective in these two studies was not much elicited for inclusion in the process of conceptualizing learner autonomy although learner autonomy has primarily been suggested to be socially situated (Smith & Ushioda, 2009). Students in different socio-cultural contexts may perceive this concept differently and express it in different learning behaviours. To address this gap in the literature, this study employs a more exploratory perspective. The thesis sets out to achieve three main objectives.

First, it aims to understand how learner autonomy is perceived by Vietnamese EFL students. After synthesizing contemporary literature on learner autonomy, it attempts to develop an inventory of learner autonomy attributes taking into account different perspectives. This inventory is then validated, tested, and implemented to produce the dimensions of learner

______Chapter 1 Page 14 autonomy as perceived by local students. This process also helps establish a reliable instrument for learner autonomy to be measured in the local context.

Second, it examines how a cohort of EFL undergraduate students in a local Vietnamese university perceives learner autonomy and exercises it. The investigation concerns both online and offline classroom learning environments. While quantitative data are used to measure their learner autonomy patterns and levels, follow-up qualitative data are drawn on to provide insights into these patterns and behaviours. Integrating these two data sets helps profile the situation of learner autonomy perceptions and performance in the socio-cultural context of Vietnam.

Third, it investigates the relationship between EFL students’ perceptions and their performance of learner autonomy. Provided with different situational dilemmas in contemporary Vietnam, this objective is to identify if there are any differences between what students think and what they do. In other words, it identifies the enablers and the inhibitors of the development of learner autonomy, thereby, opening a discussion on pedagogical implications, organizational policies, and further research.

5. Research questions

This research investigation is structured into two consecutive studies, Study One and Study Two. Each is designed to answer specific questions derived from the study objectives. Study One is designed to develop an instrument to measure learner autonomy in the context of EFL learning in Vietnam and to answer the two following questions:

1. What are Vietnamese EFL undergraduate students’ perceptions of learner autonomy? 2. What are the similarities and differences in their learner autonomy perceptions?

Study Two aims to answer three research questions below:

3. What is the relationship between their understandings of learner autonomy and learning behaviours? 4. Does this relationship change during a semester-long course?

______Chapter 1 Page 15

5. What are the factors that mediate students’ performance of learner autonomy in the local context?

6. Significance of the study

First, this study provides necessary insights into patterns of learner autonomy from local students’ perspectives, directly contributing to the process of nurturing this capability in EFL learning environments. Although many educational advocates and stakeholders in Vietnam have acknowledged the essential role of learner autonomy, there has been very limited empirical research into students’ views on this learning variable. Hence, contemporary learner autonomy promoting practices often reflect either the arbitrary viewpoints of classroom teachers or policy makers from the top-down system. Failing to take into account students’ perceptions in designing classroom activities may result in unexpected outcomes.

Second, understanding how students perceive and exercise learner autonomy can help stakeholders design appropriate pathways for the development of this capacity. Several of the current attempts to promote learner autonomy are often made locally, immediately, or spontaneously and are poorly informed by research evidence. Socio-cultural characteristics and learning style stereotypes are often inappropriately incorporated. No comprehensive development profile of learner autonomy across the different school levels has been reported. So, this study initially provides the significant dimensions of a learner autonomy profile for the context of Vietnam. It can also inform and offer evidence-based advice on autonomy fostering practices, potentially contributing to the process of meeting the national education goals.

Third, understanding students’ behaviours and participation in an online learning environment significantly contributes to establishing a set of sound principles for autonomy cultivating approaches in virtual learning environments. This initiation becomes more important particularly in contexts similar to the Faculty of English Language and Literature, where a learning management system (LMS) or similar applications have not yet been used for classroom practices. Although Moodle, an open-source courseware, is encouraged in all schools (MOET, 2008), this technology is not yet available in the Faculty of English Language and Literature. Therefore, the results of this study can be used as an initial example

______Chapter 1 Page 16 and create an opportunity for critiques. It can also increase teachers’ confidence in working with LMS, challenge their traditional practices and trigger the process of ICT integration in the local context.

Fourth, understanding the development patterns of learner autonomy can significantly contribute to the success of communicative activities that are being encouraged in EFL classes nowadays. As teachers organize learning activities in their daily classes to provide students opportunities to develop communicative competence, it is important to ensure that students are capable of controlling these activities properly to maximize the outcome. Insights into students’ levels of autonomy and patterns of development can help teachers design communicative activities with a suitable range of flexibility. The quality of students’ engagement in such activities can therefore increase. It is predicted that if properly informed by students’ autonomy profile, the measures of facilitation and support during class activities will be more effective and able to ensure students’ successful performance.

Fifth, the instrument developed and validated during the study can be used as a reliable reference for measuring learner autonomy quantitatively in Vietnam. Although this is the first attempt in quantifying the construct of learner autonomy in the local context, the instrument is justified and refined during the study with a reasonably large sample. The rigorous process employed in the study increases the validity and reliability of the instrument. Thus, it can be used to develop an evaluation framework for learner autonomy fostering attempts. It can also be an important reference for future research in both EFL and other areas of study.

Finally, research on learner autonomy has been conducted in a variety of contexts in several countries (as documented in Barfield & Brown, 2007; Gardner, 2007a; Lamb & Reinders, 2008; Miller, 2007) but very little has been conducted in Vietnam. Although it can be argued that the Vietnamese EFL classroom stereotype is rather similar to its Chinese or Japanese counterparts, it has its own socio-cultural and educational context which makes it unique. This becomes even more important for the investigation of a socio-cultural learning variable such as learner autonomy. Therefore, this research potentially plays a role in contributing research evidence to other empirical understandings of the examined construct in the area of EFL in particular and in education in general. Comparing and contrasting the patterns of learner

______Chapter 1 Page 17 autonomy identified in the context of Vietnam with those in other countries can provide more insights into this socially situated construct.

7. Limitations of the study

Learner autonomy in this study is investigated from students’ perspectives only. Perspectives of teachers, school management members, parents, and other stakeholders are not included. It is expected that these groups of people would perceive and perform learner autonomy in different ways because it is a multifaceted construct. Nevertheless, as the ultimate objective of the study is to provide pedagogical implications to promote the students’ capacity for learner autonomy, it is more important to start with the students’ voice in the investigation. Such an approach also allows for the design and development of a learner-oriented pedagogy. Understandings of students’ views on learner autonomy should also be the primary reference for investigations into other perspectives. Effective examination of these perspectives is crucial for the design of professional development agendas and policy reforms.

Furthermore, the current investigation restricts itself to Vietnamese EFL undergraduate students to establish a sound model. International students studying in Vietnam and those taking majors other than EFL were excluded in the study. Students studying in small provinces in Vietnam were also not considered. These sampling criteria potentially facilitated the generation of more focused findings. Hence, the extent to which the study findings are generalisable is limited, but the study instrument can be adapted and used for any research in the area.

8. Definition of terms

For the purpose of this study, several terms need to be defined to provide a specific perspective on the issues discussed. First, learner autonomy is defined as an “ability to take charge of one’s learning” (Holec, 1981, p. 3) and is used in the area of foreign/second language education. This ability enables learners to control their learning activities effectively and efficiently. It has multiple attributes and can be expressed through three processes, namely initiating, monitoring, and evaluating learning. The verbs perform, exercise, manifest, and self-direct are used to refer to the behavioural operation of the learner autonomy capacity.

______Chapter 1 Page 18

Second, EFL and ESL stand for English as a Foreign Language and English as a Second Language, respectively. In Vietnam, English is currently used as a foreign language; however, for the purpose of the discussion in this study, these two terms are used interchangeably to refer to the situation of English learning in Vietnam. Third, there are three types of classroom environments mentioned in this study. Traditional, face-to-face, or offline classroom is used interchangeably to mean the physical learning environment with tables, chairs, and blackboards. ICT class refers to the traditional class with the availability of some technological equipment such as computers and printers. Online class refers to a virtual learning space such as an LMS where students can access it anywhere anytime through the Internet.

9. Thesis organization

This thesis is organized into six chapters. Chapter One is the introduction of the thesis. It first provides an overview of the educational system in Vietnam. It then shifts the focus to EFL education in higher education, with particular reference to the situation of EFL first-year and second-year university students in the Faculty of English Language and Literature, University of Social Sciences and Humanities, Ho Chi Minh City. Following this, the chapter uses these contextual descriptions to link the context with learner autonomy, one of the objectives set in the higher education reform agenda. After briefly describing research on learner autonomy in the local context, it formulates the research questions and proposes two sequential studies for the investigation. Finally, the significance and limitations of the study are considered to help shape the research design and acknowledge associated matters during the investigation.

Chapter Two includes a literature review of learner autonomy in EFL/ESL education. It starts with a description of the definition and roles of learner autonomy. It then continues with a critical analysis on the four perspectives adopted in learner autonomy research. After this, it analyses the similarities and differences of this construct as perceived by different socio- cultural groups of students. Next, it documents and analyses different models of learner autonomy. The discussion includes critiques on learner autonomy promoting practices in traditional classrooms and those with ICT support. Finally, the chapter presents a theoretical framework to convey principal processes of learner autonomy and an analytical framework to understand the manifestation of this construct at the operationalised level.

______Chapter 1 Page 19

Chapter Three describes the methodology of the investigation. It first develops a research perspective consistent with the theoretical framework adopted from Chapter Two. It then argues for the employment of mixed methods to be aligned with the research perspective and generate necessary data for the investigation. The investigation is then designed into two sequential studies to cover the five research questions proposed in Chapter One. A set of procedures for the implementation of these studies is also reported together with the method of data integration. Accordingly, the participants, objectives, data collection procedures, and instrument design of each study are described. The chapter finishes with considerations on issues of reliability, validity and ethical considerations.

Chapter Four reports on the results of the two studies conducted in the current investigation. The first half presents the results of Study One, and the second half presents the results of Study Two. Study One uses the data generated from a questionnaire to answer the first two research questions. Study Two analyses the data generated from the pre-test and post-test questionnaires, LMS records, and semi-structured interviews to answer the other three research questions. A multiple-phased process is conducted to establish the reliability and validity of the instruments. After this, a model of learner autonomy is constructed, and its components are analysed. The relationship between perception and performance of learner autonomy is then investigated, and mediating factors of learner autonomy performance are elicited. The chapter finishes with a general description of learner autonomy in the study context by integrating the results from both studies.

Chapter Five analyses the results integrated from different data sets in relation to the contemporary research in learner autonomy in EFL education. The four-dimension model generated from the current investigation is compared and contrasted with those previously suggested for the study context. It is also paired with other theoretical models in the literature on learner autonomy. The presentation continues with a critical analysis of the relationship between what students think and what they do regarding learner autonomy in the local context with considerations on personal and situational enablers and disablers. After this, the development pattern of learner autonomy during the course is investigated to highlight the impact of mediating factors. The chapter finishes with a critique of the relationship among mediating factors of learner autonomy performance and proposes suggestions for pedagogical practice.

______Chapter 1 Page 20

Chapter Six includes the conclusions and implications of the investigation. It starts with a summary of the whole research, highlighting significant results and critical analyses. Theoretical contributions to the area of learner autonomy are described and followed by methodological implications. The chapter continues with a set of pedagogical practices and implications to promote learner autonomy in EFL education in the local and regional contexts. It then addresses the limitations of the study, considers the possibility of generalizing the study findings, and puts forward directions for further research. The chapter then includes a critical reflection on the whole process of the study formulation and implementation. It finishes with an overview of important conclusions for policy makers and educational stakeholders drawn from the investigation.

10. Summary

This chapter reviews a significant problem in the context of Vietnam, where understanding and facilitating students’ learner autonomy in EFL learning is of great concern. The first five sections attempt to locate the topic by examining four contextual aspects, namely the centralised mechanism of the education system, the objectives proposed by the local authorities, the contemporary situation of EFL training in higher education, and the general learning behaviours of Vietnamese students. These together constitute a critical rationale for investigating the construct of learner autonomy in the local socio-cultural context. It then narrows down the topic to students’ perspectives only to increase the study feasibility. As a result, research questions are proposed to explore EFL students’ perceptions and performance of learner autonomy. These questions are also designed to understand possible associated factors which shape these patterns and behaviours.

The last five sections of the chapter present the significance, limitations, and organization of the research study. Being primarily driven by the context and literature suggestions, this study is argued to serve as an initial investigation on patterns of learner autonomy perceived and exercised by Vietnamese EFL students. Upon completion, it can provide a number of recommendations for stakeholders to foster and evaluate this capacity in the local context. These recommendations should contribute to the process of achieving the planned national education objectives. The limitations of the study are also examined and justified. The organization of the chapters is then presented. To provide a more critical and insightful

______Chapter 1 Page 21 understanding to the construct of learner autonomy, the following chapter synthesizes reports of prior research in the area and relates them to the situation in Vietnam.

______Chapter 1 Page 22

Chapter Two

LITERATURE REVIEW

______

1. Introduction

Learner autonomy has been an important focus of educational practices and research for more than three decades. The word autonomy is derived from auto-nomos, a Greek word (Voltz, 2008), auto meaning “self” and nomos meaning “rule or law.” Auto-nomos refers to a state where one gives oneself his/her own law. Although this concept originates from politics (Boud, 1981), it can be used in any field, including philosophy, medicine, and psychology to indicate a capacity that allows a person or organization to be able to justify reasons for doing things (Dearden, 1972). In education, this capacity is often used for schools, teachers, and learners, referred to as school autonomy, teacher autonomy, and learner autonomy, respectively. While the first two are related to education policies and teachers’ professional development, the third, which is the focus of this research, is about students’ learning attributes.

This chapter provides a synthesized overview of learner autonomy in foreign/second language education in prior research. It starts with a presentation of the definition and roles of learner autonomy. It then continues with a critical analysis of the four perspectives adopted in learner autonomy research. After this, it discusses the perceptions of this construct by different socio- cultural groups of students. Next, it presents and analyses different models of learner autonomy. The discussion continues with critiques of leaner autonomy promoting practices in traditional classrooms and those with ICT support. Finally, the chapter presents a theoretical framework to convey the principal processes of learner autonomy and an analytical framework to understand this construct at the operationalised level.

______Chapter 2 Page 23 2. Learner autonomy in foreign/second language education

Learner autonomy in foreign/second language learning is described as a notoriously ‘complicated’ (Little, 2003) or ‘multifaceted’ construct (Smith & Ushioda, 2009). The notion of learner autonomy has been modified and even transformed during its development. In addition, it has been advocated to be a fundamental goal of education for several decades (Benson, 2001; Waterhouse, 1990). As a result, the definition of this concept has varied widely. Currently, it is considered part of the ‘orthodoxy’ of language education, “an idea that researchers and teachers ignore at their peril” (Benson, 2009, p. 14). This orthodoxy is important for individuals’ personal development and learning processes as it plays a key role in driving humans’ free choice of goals and relations in life (Raz, 1986) regardless of others’ will (Young, 1986). The two following sections attempt to specify the definition of learner autonomy and its roles in the area of second language education.

2.1. Definition of learner autonomy

Learner autonomy in language education was originally defined as an “ability to take charge of one’s own learning” (Holec, 1981, p. 3), and this definition has become the most cited in the literature (Benson, 2009; Dang, 2010). It was driven by the concept of freedom and autonomy in philosophy and initiated by the Council of Europe’s Modern Languages Project in the early 1980s. Later, ‘ability’ and “take charge of” were often replaced by ‘capacity’ and “take responsibility for” respectively. These word replacements seem to be a matter of linguistics only, and the semantic aspects of the construct remain unchanged (Dang, 2010). This ability was further elaborated as not being “inborn but must be acquired,” mostly by formal education practices (Holec, 1981, p. 3). This is a milestone that has triggered different interpretations for this “buzz word” (Little, 1991).

Learner autonomy has therefore been perceived and translated into practice in several ways, depending on particular political, social and contemporary situations (Dang, 2010). First, it is generally considered as an ability of knowing how to learn (Wenden, 1991). Second, it is regarded as an ability to ‘control’ one’s learning activities (Cotterall, 1995). Third, it is seen as an ability for ‘detachment’ (Little, 1991) or ability to learn “without the involvement of a teacher” (Dickinson, 1987, p. 11). Fourth, it is said to be a “capacity to make and carry out choices” (Littlewood, 1996, p. 428) or an ability to perform rational decision-making

______Chapter 2 Page 24 processes over learning activities (Hunt, Gow, & Barnes, 1989). More specifically, it is viewed as an ability to give responses beyond usual instructions (Boud, 1988). These examples suggest that although different aspects of this ability can be the focus in each definition, they always maintain the central core of this construct which is the ability to understand and manage learning processes responsibly and effectively.

These variations in the definition of learner autonomy reflect a normal developmental trend that one should expect. As the notion of every learning construct is embedded in and developed from a situational occasion, its interpretation should be modified by users according to changes in political belief, language learning theory, technology, employment demand, and learning objectives. This becomes even more important for learner autonomy and it has been argued in contemporary literature that this multifaceted capacity needs to be localized and addressed in particular social contexts (Smith & Ushioda, 2009). In other words, people in different socio-cultural situations may have different views on learner autonomy, and this, logically, leads to a different set of practices to foster this capacity. This social turn is, therefore, allocated appropriately in the notion of Vygotsky’s (1986) socio-cultural theory and provides important theoretical foundations for research in this area (Toohey, 2007).

A more comprehensive discussion of different aspects of learner autonomy derived from different definitions of learner autonomy is presented in Section 3 of this chapter. This arrangement is due to the relationship between each definition and its adopted perspective. Before continuing with this discussion, it is important to present the roles of learner autonomy in second and/or foreign language education.

2.2. Roles of learner autonomy

Learner autonomy has been claimed to be an ultimate goal of education in general and second/foreign language education in particular for a long time (Benson, 2001, 2009; Dang, 2010; McClure, 2001; Waterhouse, 1990). It is often identified to signify students’ active participation in learning activities (Benson, 2007), a view which is supported by a number of studies in different contexts (such as Aoki, 2001; Christopher, 2006; Hart, 2002; Miller, Hopkins, & Tsang, 2005; Smith, 2001, 2003c). It has also been reported to result in better productivity, higher level of motivation, higher rate of knowledge retention, and less frustration (Bachman, 1964; Dickinson, 1987; Ellis, 1994; Gardner & MacIntyre, 1991;

______Chapter 2 Page 25 Holec, 1987; Rivers, 2001). This indicates that learner autonomy directly contributes to both processes and outcomes of learning activities. In addition, it indirectly accelerates the level of other learning variables. This section discusses specific roles of learner autonomy in relation to changes in language teaching methods, educational approaches, educational management trends, technology innovation, and socio-economic development.

Learner autonomy is particularly important for the success of the communicative language teaching method (CLT). Learner autonomy supports the process of improving communicative competence. CLT currently receives significant recognition from stakeholders and shares a similar view of learner autonomy regarding the issue of supporting students, not teachers, to stand in the centre of all learning processes (Nunan, 1988, 1991; Tarone & Yule, 1989). Therefore, students who learn under the CLT method are required to actively participate in customized communicative situations to develop their language repertoire and competence. They also have to evaluate their learning regularly to increase language competence (Dam & Legenhausen, 2010). To do this successfully, students need to possess an adequate level of learner autonomy as this can help them become willing to take risks (Pellegrino, 1996) and use the target language to communicate in real situations. In other words, students’ effective engagement in CLT activities is mostly mediated by their learner autonomy capacity (Breen & Mann, 1997; Littlewood, 1996, 1997; Nunan, 1997).

The promotion of learner autonomy corresponds with the shift towards a student-centred approach in modern education (Geng, 2010; Wenden, 2002). Over the past two decades, student centeredness has been popularly claimed and identified to be an effective learning trend in both Western and non-Western contexts. More power is removed from the teachers’ domain of authoritative knowledge (Rivers, 2011). Courses are designed to serve particular groups of learners instead of attempting to make one for all. Therefore, students are given more opportunities, responsibilities, and power to deal with their learning activities. To effectively benefit from this process, students need to be capable of taking control over all the activities provided to them during each learning process. If they are appropriately supported to develop this capacity, defined as learner autonomy, they can take better advantage of the benefits which the modern approach offers. However, if students are not provided with this support, this may, in turn, lead to a failure of the whole approach.

______Chapter 2 Page 26 The orientation towards learner autonomy development is also in line with the privatization, commercialization, and public participation trends of educational services such as those in Vietnam. Different from mass or centralised training systems whose institutions are totally subsidized or funded by a national body of authorities in many countries such as Vietnam, schools are considered as knowledge producers. Therefore, to produce a more competitive product, schools need to offer students more personalized choices and freedom for individual development. Students are also respected for their personal learning styles and preferences, opportunities for content and material negotiations as well as socio-economic and cultural values. Such customized training programs, in return, expect students to exercise their autonomy capacity to effectively engage in these opportunities in order to develop valuable skills for the contemporary labour market.

Furthermore, cultivating and nurturing learner autonomy capacity in foreign language education aligns with the learning attributes required by an ICT-supported educational environment. With the rapid development of ICTs and their increasing uses in foreign language programs these days, students may encounter certain technical problems which need to be overcome in order to take part in the learning environment. It has been shown that, for instance, learner autonomy can help students to face the challenge of technical difficulties (Rubin, 1975). In addition, resources on the Internet are very rich, scattered, and are constantly expanding. Students need to be able to plan in order to engage in effective learning. Research has shown that those who can plan their learning activities better in this interactive environment tend to engage more actively in these learning processes and produce better learning outcomes (Dam, 1990). Better planners also potentially engage in a higher degree of review and reflection on their learning processes.

Learner autonomy capacity is especially important for knowledge construction and sustainable learning in today’s globalized world. As information becomes more abundant and accessible, the ability to synthesize and negotiate with others’ knowledge to produce our knowledge becomes a more crucial process in learning (Dam, 1990). Students need to be trained to be capable of being reflective in the learning process, critical and creative in thinking, and flexible in learning methods (Benson, 2005). Therefore, learning-to-learn skills have been strongly encouraged in different education systems to obtain a better quality labour force (e.g., Benson, 2004; Little, 1996b; Trinh, 2008). Students’ success is judged by their

______Chapter 2 Page 27 capacity to instruct and train themselves, not by their responsiveness to teachers’ instructions (Benson, 2007).

3. Conceptualization of learner autonomy

Given the significantly important role of learner autonomy in foreign language education, it has been perceived from four different perspectives, namely psychological, technical, socio- cultural, and political-critical (Benson, 1997, 2006; Healy, 2007; Oxford, 2003; Sinclair, 2000). The psychological perspective values the personal attributes of the learners; the technical perspective values attributes in the learning environment; the socio-cultural perspective emphasizes the interactions between learners and their environment; and the political-critical perspective focuses on learners’ access, control, power and ideology in their community. This section is devoted to a discussion based on these four interpretations and their impact on pedagogical practices.

3.1. Learner autonomy as an attribute of the learner

According to the psychological perspective, learner autonomy is regarded as a mental and emotional attribute (Oxford, 2003) that enables an individual to exercise his/her learning activities effectively. Autonomous learners are those who believe that they are capable of organizing and performing a course of action required to achieve success (Bandura, 1997). This ability is reinforced by Little (1990) who defines it as “a capacity for detachment, critical reflection, decision-making, and independent action” (p. 7). The aspect of individual psychology becomes clearer when Little (2003) combines the definition of Holec (1981) and his own to suggest that autonomous learners are fully aware of and understand their course objectives, accept the learning goals, and agree to take responsibility.

Although the psychological attributes are indicated in the definition, the learning management activities described in the performance of this capability drives its conceptualization a little away from the origin. If learner autonomy includes a series of learning actions, it should be classified as behaviour rather than capacity or attitude. In addition, if an autonomous learner exercises his/her learner autonomy by performing a number of learning management behaviours, he/she is able to take control rather than take charge of his/her learning processes. So the manifestation of learner autonomy is on the actual performance of a series of actions,

______Chapter 2 Page 28 not just the initial and superficial awareness recognized in the mind. In other words, a combination of proper attitudes, knowledge and skills would potentially enable the performance of learner autonomy (Smith, 2000; Ushioda, et al., 2011).

Taking this argument into account, Horváth (2005) suggests that there are two interrelated elements in this construct, namely cognitive and behavioural. The first is related to the psychological aspect (what learners believe); and the second is related to their behavioural actions (what they do). Specifically, autonomous learners possess a capacity to initiate, monitor, and constantly evaluate their own learning activities (Little, 2003). They perform these processes with clear purposes in mind and employ effective strategies in a persistent manner (Purdie, Hattie, & Douglas, 1996). Importantly, these courses of learning activities are mentally driven; and subsequent behaviours are regulated (Schunk & Zimmerman, 1994) to produce better outcomes.

In an investigation on the behavioural and cognitive aspects of learner autonomy, Horváth (2005) provides more specific examples of each element. The first is associated with management behaviours such as choosing materials, methods, time, and partners with which to learn (as proposed by O'Malley & Chamot, 1990; Oxford & Shearin, 1994). The latter is connected with the conscious efforts of learners such as plan making, monitoring, reflecting, and assessing which trigger and drive the performance of learning behaviours (Dickinson, 1992, 1995; Littlewood, 1996). Horváth also suggests that the most important values of the cognitive component are reflection, self-reflection, evaluation, and self-evaluation. These are constantly used by an investigated sample of postgraduate interpreter trainees. In addition, the relationship among these elements is identified to be cyclic. The reflective activities continually deconstruct what learners have already achieved and reconstruct it. Going through these processes enables students to manage and control their learning to gain better knowledge and skills.

To achieve quality learning, learners need to exercise their behavioural and cognitive processes effectively and efficiently. These processes include resolving conflicts in learning styles and preferences, strategic options, and the material demands that they face (Rivers, 2001). They might also be “associated with a radical restructuring of language pedagogy,” a “rejection of the traditional classroom,” and even an “introduction of wholly new ways of working” (Allright, 1988, p. 35). Contributions of autonomous learners to classroom activities

______Chapter 2 Page 29 may go well beyond what teachers can predict and even eliminate teachers’ lesson plans. Traditional roles of teachers as knowledge transmitters and authoritarians are therefore no longer appropriate and need to be changed to suit autonomous learners. This conceptualization of autonomy challenges ideologies of organizing educational practices and urges significant reforms from teacher-oriented systems.

The significant impact of learner autonomy on learning procedures and the learning environment is also acknowledged by Rivers (2001). He argues that autonomous learners demonstrate their learner autonomy capacity through their “requesting and demanding substantive changes to every aspect of the course, especially to course content and structure” (p. 287). It allows them to self-direct their language learning process effectively and use appropriate learning strategies. Autonomous learners are more active than those who are just mentally responsible for their learning. Their engagement in a learning activity is driven more by their interest rather than the obligations set by their parents (Rivers, 2011).They personally attempt to modify learning-related attributes during learning processes because this can help them achieve their goals more conveniently and appropriately.

However, Rivers (2001) argues that the psychological element is the controller of its behavioural counterpart. His argument is drawn on a strong connection between the concept of learner autonomy in education (e.g., Holec, 1981; Little, 1990, 1991) and metacognition or executive functions as named by Denckla (1996) in developmental neurology and neuropsychology (Dennis, et al., 1996; Taylor, Schatschneider, Petrill, & Owens, 1996). Rivers suggests that executive functions control the autonomous learning process. They consist of metacognitive self-assessment and metacognitive self-management. The former is the ability to assess one’s own cognition; and the latter is the ability to manage one’s further cognitive development (Brown & Palinscar, 1982; Flavell, Miller, & Miller, 1993; Paris, Lipson, & Wixson, 1983). These two processes dictate and regulate the learners’ learning behaviours.

The comparative analysis between the central roles of the human brain and the controlling function of the psychological component in the construct of learner autonomy pushes its conceptualization back to its origin. The underlying aspect of psychology embedded in taking charge of one’s learning is therefore scaffolded and reaffirmed. It particularly goes in line with the use of attitude as an overarching attribute of this construct as argued by Allwright

______Chapter 2 Page 30 (1990). Learner autonomy is defined as an attitude of being willing to take responsibility; demonstrate necessary ability, and perform concrete actions. Although attitude is not a typically representative notion for the psychological attribute, this definition covers most of the important aspects of the automisation process and emphasizes the driving force of learners’ cognition and metacognition.

In short, the psychological perspective views learner autonomy as a capacity which is regulated by the learners. Therefore, autonomous learners are often described as those who attempt to manage their behaviours and modify the environment to achieve learning goals. It does not appropriately acknowledge the influences of learning environment attributes such as technology, resource availability, and social ties back on the learners. However, these environmental contributions are not ignored by researchers in the field. In contrast, they are significantly taken into account and provide foundations for another strand of autonomy, namely the technical perspective. The following section is devoted to a discussion of learner autonomy from this point of view.

3.2. Learner autonomy as a reflection of the learning situation

The technical perspective views learner autonomy as a contribution of the learning environment on learners. Instead of focusing on the learners’ personal attributes, it strongly emphasizes the prominent effects of situational aspects. Learner autonomy is defined as a “situation in which the learner is totally responsible for all of the decisions concerned with his learning and the implementation of those decisions” (Dickinson, 1987, p. 11). The immediate learning environment is seen as the principal factor that drives learners’ behaviours. It can enable learners to select what, when and how to learn to achieve their targets (Oxford, 2003; Pennycook, 1997). In other words, learner autonomy is activated and developed by an other- created situation; and any students in an autonomous situation can self-direct their own learning activities automatically. This strand of learner autonomy has led to the establishment of a number of self-access learning centres during the 1990s (Benson, 2006) and it is the focus of the discussion here.

Within the supportive context of self-access learning centres, it is argued that learners are given adequate agency to carry out their learning activities effectively. The most autonomous learners do not need any involvement from teachers, consultancy experts, learning facilitators,

______Chapter 2 Page 31 or prepared materials (Dickinson, 1987). A resource-rich environment can also increase their motivation and empower them to adopt suitable learning strategies. They can therefore make all instructional decisions and perform all learning activities on their own. These seem to suggest that autonomous learners tend to work alone in bookish situations such as self-access learning centres rather than in collaboration and negotiation with other peers in more relaxing environments such as playgrounds or coffee shops.

Prior research has shown that there are different ways of using self-access learning centres to foster learner autonomy. For example, authentic materials in a multimedia library are reported to regulate students’ independent learning activities autonomously (Riley & Zoppis, 1985). Integrating such resources into a training program as an additional activity is also indicated to facilitate students’ learning participation (Miller, et al., 2005). Students have opportunities to personalize their learning materials to suit their levels of proficiency and topic preferences. As a result, they become more active, motivated, and confident in learning processes (Brandon, 2003), and their learning outcome is then improved.

However, these investigations also indicate that it is not purely the resources themselves that foster learner autonomy; rather, it is the appropriate design and employment of activities organized in a well-equipped environment. These activities include sufficient advisory support from expert facilitators; productive collaboration between teaching staff and the management team (as in Brandon, 2003), customized material design (Riley & Zoppis, 1985), and suitable learning tasks in a consistent course structure (Miller, et al., 2005). They provide supportive consultation and guidance to help students navigate their learning processes rather than leaving them alone to deal with the resources in self-access learning centres. Successful navigation enables students to manage their learning process effectively and develop their autonomy capacity.

These mixed results and interpretations of the contribution of self-access learning centres to learner autonomy development imply a lack of distinction between learning with support and without support. Although they are not the same, over the past three decades, research on learner autonomy has often equated self-access learning with self-instructed learning (Benson, 2001). Despite both being used to refer to learning without teachers’ instructions, the former is provided with adequate support in the learning situation such as technical help and language expert advice; the latter is not. As mentioned previously, there has been no clear

______Chapter 2 Page 32 evidence that suggests a relationship between self-instruction in self-access learning centres and learner autonomy development (Benson, 2001). Most of the indicators of learner autonomy development are fostered and maintained by timely support from the surroundings.

In short, the focus on the situational attributes of the technical perspective has been supported by research despite the fact that many of these do not distinguish the difference between ways of learning. It can be seen that this strand of learner autonomy shares the same view as the psychological perspective on learners’ ability to control their learning process. However, they differ in the argument on the driving factors. Personal psychology seems to form a stronger driver than an external situation in regulating one’s learning behaviours because it better helps to explain human’s control of action. Different from the technical perspective, the psychological perspective does not assume that learners can simply start to self-direct their learning when being located in a situation where they have no other choice than learning. Learner autonomy is not an object or a tool which can be easily granted to students simply by changing the environment. In certain situations, giving students full responsibility can even lead to the avoidance of learner autonomy development as students may not take that responsibility for their learning.

Therefore, the other-created situation alone seems to be insufficient for fostering learner autonomy. In other words, the technical perspective does not take into account students’ personal variations. Similarly, the psychological perspective does not acknowledge the influence of students’ living environment although human psychology should be mediated by the surroundings. This suggests that either the psychological or technical perspective alone may not result in a complete understanding of the construct of learner autonomy. A combination of these two perspectives appears to provide a more complete picture as they seem to be complementary to one another. The two perspectives acknowledge the two significant sources of attributes on human learning behaviours, namely individual psychology and the learning situation. Taking this into account, the socio-cultural perspective focuses on the interactions between an individual and his/her immediate learning context on exercising or enhancing learner autonomy. The next section is devoted to this discussion.

______Chapter 2 Page 33 3.3. Learner autonomy as an outcome of the learner’s interactions with the environment

The socio-cultural perspective, sometimes called the sociocognitive or social-interactionist perspective, perceives learner autonomy as a socially-situated construct (Smith & Ushioda, 2009). This means it is formed and developed by the interactions between learners and their learning situations. Any attribute provided by an external environment has to be filtered and internally processed before creating an impact on a learner. This internalization process is mediated by the psychological factors of each individual and produces behaviours that enable a learner to control his/her learning activities. Therefore, learner autonomy should be an attribute of neither the learner nor the learning situation alone, but an interaction of both.

The theoretical argument for this strand of learner autonomy is driven by socio-cultural theory (Lantolf, 2000; Lantolf & Thorne, 2006; Vygotsky, 1986) and community of practice (Lave & Wenger, 1991; Wenger, 1999; Wenger, McDermott, & Snyder, 2002). They claim that every learner has specific ties with his/her living environment; and every learning activity organized in that environment is associated with specific social, cultural and economic factors of that environment. Therefore, the mediating learning variables are crucial for language development and learner autonomy. The process of using individual ability to participate, interact and negotiate with situational influences can nurture learner autonomy capacity.

The space for students to navigate their learning activities and acquire learner autonomy is called the Zone of Proximal Development (Lantolf, 2000; Lantolf & Thorne, 2006; Vygotsky, 1986). The Zone of Proximal Development sets a boundary between the learner’s performance with support and without support. Successfully moving through this space enables learners to achieve learning outcomes. In other words, participating in a community of practice allows less capable learners to apprentice the knowledge and skills that are constructed by the senior (Wenger, 1999). Learners’ desire for greater competence is then increased. The senior can be resources, peers, mentors or experts available in the situation to facilitate the learners’ participation process and motivation, enabling them to better demonstrate their identity of the “imagined communities” (Kanno & Norton, 2003; Norton, 2000, 2001). The learners’ participation in the community then changes from peripheral beginners to expert controllers of different learning activities as they practice more interactions with the situation.

______Chapter 2 Page 34 To gain a better understanding of learners’ interactions, Schwienhorst (2003) describes these processes in three interrelated dimensions. The individual cognitive dimension characterizes students’ ability to regulate their learning and achieve better language awareness. The social interactive dimension indicates the degree to which students use the target language in communicative learning activities. The participatory dimension refers to students’ level of participation in all community activities. The first two dimensions can together scaffold the success of students’ learning engagement in the third dimension. These three dimensions of learners’ autonomy performance present three roles of learners, namely intentional learner, communicator, and researcher, respectively (as suggested in Little, 1996a). These descriptions provide greater insights into the formation of learner autonomy from a socio- cultural perspective.

Being able to draw both personal and situational factors into the argument on learner autonomy formation, the socio-cultural perspective has recently created a social turn in this area of research (Riley, 2009; Smith & Ushioda, 2009). This trend has gained dominant support primarily because it allows learner autonomy investigations to take into account local and cultural differences (Pemberton, Toogood, & Barfield, 2009) whose attributes are significantly important for shaping learners’ perceptions and performance of learner autonomy as well as contributing to the process of sustainable learning. However, the socio- cultural perspective is not sufficient to address the reasons for an individual to develop a learner autonomy capacity at a more philosophical level; and this is why the political-critical perspective becomes important. The next section documents a discussion on this strand in detail.

3.4. Learner autonomy as a desire for more access, agency, and power in the community

A political-critical perspective argues that learner autonomy is shaped by constraints of access, control, power and ideology (Pennycook, 1997). As the concept of learner autonomy is driven from freedom and power, the development of this construct arises from a desire to become “the author of one’s own world” (Pennycook, 1997, p. 45). Therefore, students constantly seek possible alternatives to counter current situations; demonstrate their superiority, and conquer the situation in which they are involved. They take advantage of any insurgent language aspect that they can identify during the learning process to gain a more powerful life (as in the cases of refusniks reported in Pennycook, 1997). In addition, they try

______Chapter 2 Page 35 to transform themselves, that is, erasing former identities, changing beliefs and modifying cultural routines if they believe that those values are no longer appropriate for their own development in the target community (Ogbu, 1991; Oxford, 2003).

Autonomous learners are described as those who have power to control their learning situations; have choices in learning activities, and are free from oppressive forces (Oxford, 2003). However, research has shown that the desire to gain access to power structures is not the same for all individuals or groups. Some may think that gaining power in a socio-political condition is out of their reach and do not want to strive for this (Ogbu, 1991). This suggests that the formation of learner autonomy is constrained by specific locations, spaces, situations, communities and by individual characteristics such as age, gender, belief, religion, and culture (Pennycook, 1997). It is situation-biased and non-universal.

Although the political-critical perspective does not explicitly present a set of learner autonomy dimensions or describe what learner autonomy looks like, it provides a more philosophical foundation for the investigated construct. It takes into account values arising from psychological, familial, situational, cultural, social, and ideological factors, indicating that the concept of learner autonomy should be understood at local, regional, and global levels. Striving for learner autonomy can be similar for most students who receive equal rights in schooling in similar contexts. However, their degree of perceiving and performing this capability is supposed to be different due to their local situations.

3.5. Establishing the perspective adopted in this study

In brief, learner autonomy is a capacity, enabling learners to control their learning activities more effectively and engage in them more actively in both school and life-long learning (Dam, 1990; Little, 1991). It is illustrated by their ability to initiate, monitor, and evaluate learning processes (Little, 2003). Learner autonomy is shaped by students’ readiness to learn; their access to peer support through the learning community, and opportunities for topic selection in resource-rich language laboratories (Healy, 2007). The formation of learner autonomy is also about how these attributes interact with one another and the level of control over these processes that learners wish to take. Learners are seen as both independent individuals who want to modify their living situation for their own sake and members of generalized socio-cultural groups who are attached to specific values and relationships. As a

______Chapter 2 Page 36 result, learner autonomy is identified as a socially and culturally situated variable, affirming the multifaceted nature of this construct.

The concept of learner autonomy has principally been examined from four different perspectives (Figure 2.1), constituting a better understanding of this ‘susceptible’ construct (Smith, 2003b). They may look divergent and even opposite at first, but actually they are complementary. Impacts from the learning environment (technical perspective) and personal characteristics (psychological perspective) are the two important interrelated aspects of learner autonomy formation and development. Negotiations and interactions between an individual with his/her situation (socio-cultural perspective) is in a shared space. These interactive behaviours in a community are driven by a desire for more effective learning outcomes, more agency and better life quality (political-critical perspective). It has been suggested that research should cover as many perspectives as possible as they are not antithetical to one another (Oxford, 2003).

Psychological perspective Technical perspective

Learner Autonomy

Socio-cultural perspective Political-critical perspective

Figure 2.1: Four perspectives on the construct of learner autonomy

Apart from the perspective distinction analysis, it is also necessary to acknowledge that there are other ways of examining the construct of learner autonomy. For example, Holiday (2003) focuses on the importance of cultural differences and proposes three approaches, namely native-speakerist, cultural-relativist, and social autonomy. These are driven by different assumptions on cultural superiority which it is argued to create different impacts on the conceptualization of learner autonomy. Other examples are personal autonomy (Young, 1986), weak and strong autonomy (Smith, 2003a), narrow and broad autonomy (Kumaravadivelu, 2003), divergence and convergence (Ribe, 2003), and interactions in dynamic inter-relational space (La Ganza, 2001, 2002). The main reason for these several analyses is that they employ different viewpoints and assumptions on teaching practices, learners, and situational aspects. However, they are similar in describing the performance of learner autonomy capacity.

______Chapter 2 Page 37 Driven by these four perspectives on learner autonomy and taking into account the contemporary social turn, this research study acknowledges the significant contributions of socio-cultural factors in shaping students’ perceptions and their performance of learner autonomy. Given the changing circumstances in Vietnam (Dang & Robertson, 2010c), it is almost impossible at this stage to understand how Vietnamese EFL students value learner autonomy and their desire to fit into the local current socioeconomic environment or control their own learning activities in their own world (the argument is in Benson, 2009). Therefore, the following section attempts to argue for an appropriate model of learner autonomy to be employed in the current research.

4. Models of learner autonomy

Similar to the diverse interpretation of the concept of learner autonomy, there are six different models of learner autonomy that can be identified in the literature. They can be classified into two types, which focus on the stages of development and areas of control. The first type attempts to create a general index for each level of learner autonomy development. The second type describes the performance of learner autonomy in different aspects of the students’ learning environment.

4.1. Models of learner autonomy regarding stages of development

The first person attempting to index learner autonomy attributes into phases of development was Nunan (1997). Based on learners’ actions, he proposed a model of five degrees, namely

Table 2.1: Five-level model of learner autonomy (Nunan, 1997, p. 195)

Level Learner Action Content Process Learners identify strategy Learners are made aware of the implications of pedagogical tasks and 1 Awareness pedagogical goals and content of the identify their own preferred learning materials they are using. styles/strategies. Learners are involved in selecting Learners make choices among a 2 Involvement their own goals from a range of range of options alternatives on offer. Learners are involved in modifying 3 Intervention and adapting the goals and contents Learners modify/adapt tasks. of the learning program. Learners create their own goals and 4 Creation Learners create their own tasks. objectives. Learners go beyond the classroom Learners become teachers and 5 Transcendence and make links between the content researchers. of classroom learning and the world.

______Chapter 2 Page 38 awareness, involvement, intervention, creation and transcendence. Conceptually, this model is similar to the stages of a learning process, which is how learners are guided to perform a sequential series of behavioural and cognitive actions in learning a language item. First, learners need to be aware of learning goals and prepare learning strategies. They then adopt their own goals and select suitable tasks to perform. After this, they modify, create their new learning goals, and design their new tasks. More details are presented in Table 2.1 (page 38).

The model provides a useful index in a gradual developmental order of learner autonomy. Learning behaviours and processes in each stage are clearly specified. However, this order of development is not necessarily true for learners in different contexts. For example, Chinese learners favour a flexible space for creating new content and attempt to design new tasks (Level 4 and 5), but they are confused at selecting from a range of alternatives (Level 2) (Sinclair, 2009). This means a learner does not necessarily have to achieve Level 2 before Level 4 of learner autonomy. This demonstrates the non-universal nature of learner autonomy and the disadvantage of using such a language learning theory-driven model for this socially- mediated construct (compare with Nunan & Lamb, 2001).

The second model of learner autonomy is derived from the definition of learner autonomy as a self-regulation process. It includes two levels of self-regulation, namely reactive and proactive autonomy (Littlewood, 1999). A reactive autonomy regulates the learning activity after clear directive instructions have been provided. It “enables learners to organize their resources autonomously in order to reach their goal” (p. 75). A proactive autonomy regulates both the activity and the instructions. It allows learners to “affirm their individuality and [set] up directions which they themselves have partially created” (p. 75). A reactive autonomy is either preliminary to or independent of the proactive level. Given relatively specific requirements of each training course, educational practices are designed to empower students’ reactive autonomy rather than the other. The distinction between reactive and proactive levels has provided important insights, suggesting autonomy nurturing in education without significant reforms (Benson, 2006) and specifying a benchmark for measurement. However, this distinction is relatively broad.

The third model was proposed by Scharle and Szabo (2000), consisting of three stages, namely raising awareness, changing attitudes and transferring roles. Raising awareness is the initial cognitive process that enables learners to be conscious about learning objectives; define

______Chapter 2 Page 39 the contents, and determine the progression. Changing attitudes refers to the stage of transition where learners attempt to replace the previous learning behaviours with some new way of learning. They rationally select appropriate methods and techniques for their learning path (also in Little, 1991). Transferring roles is the highest level of learner autonomy when learners can fully control their learning process with little structured instructions and support. Learners are involved in a process of checking the ongoing learning activities and evaluating the knowledge gained. It seems that this model is a simplified version of Nunan’s (1997) with quite similar indexes.

In brief, these three models have successfully allocated many attributes proposed in the concept of learner autonomy in developmental stages. They can be used to indicate possible progression in capability from low to high (Benson, 2006) in education contexts and broader social contexts. However, their theoretical foundation based on language learning development has challenged assumptions of the non-linear development of learner autonomy. This means different attributes of learner autonomy can be developed independently from one another. Therefore, researchers have suggested another set of models, based on the areas of learning control.

4.2. Models of learner autonomy regarding areas of control

The literature has identified three models of learner autonomy which tend to focus on students’ areas of control. The first is a three-stage model, namely autonomy as a communicator; autonomy as a learner, and autonomy as a person (Littlewood, 1996). The three roles in the model are associated with the three situational dimensions of learners, namely contexts of language acquisition, learning approach, and personal development, respectively (Benson, 2006). Autonomy as a communicator in the context of language acquisition involves the ability to communicate successfully in particular situations by creatively employing the language and properly adopting strategies. Autonomy as a learner in the context of foreign language learning involves the ability to use suitable strategies to participate in learning activities inside and outside the class. Autonomy as a person in a broader context involves the ability to communicate personal thoughts and personalize the learning environment. Although there seems to be a distinction between these three contexts of autonomies, their attributes are interrelated. The first two are assumed to contribute to the characteristics of the more advanced person-related (Littlewood, 1996).

______Chapter 2 Page 40 The second model of this type was developed by Macaro (1997, 2008) and also consists of three aspects, namely autonomy of language competence, autonomy of language learning competence, and autonomy of choice and action. The first refers to communicative ability after mastering second language rules at a certain level. The second refers to the ability to reproduce the language skills which have been gained and transfer them to other similar situations. The third refers to the ability to develop learning options such as short-term and long-term objectives and identify preferred personal learning strategies as well as an ability to perform higher-order thinking skills in the target language such as constructing argumentative essays and providing justifications for a particular matter. It can be seen that each aspect is associated with a specific group of skills and behaviours which are not totally comparable to each other. For example, acquiring a language rule is not necessarily a prerequisite for the development of learning strategies. Therefore, the latter aspect is not necessarily more advanced than its former counterpart and overlapping between them is possible.

Instead of focusing more on learners’ behaviours and critical thinking skills, Benson (2001) proposes another three-aspect model, specifying three broader areas of control, namely learning management, cognitive processes and learning content. These are closely interdependent and associated with learning behaviours, the psychology of learning, and the learning situation, respectively. The development of control in one area can support performance of the others. For example, effective control over the content of learning can result from positive learning attitudes and appropriate learning strategies. A high level of cognitive awareness of learning may accelerate the level of control over the learning management process. Pairing the model with the four perspectives indicates a relatively match between the three areas of control with the perspectives’ foci.

In brief, the three models regarding areas of control have several features in common. The attributes of the first two models are very similar. For example, the attributes of autonomy as a learner in the first model are not very different from the autonomy of language learning competence in the second model. Both concern the learner’s ability to use language for communication in different contexts. Similarly, several attributes suggested in autonomy of choice and action in model two are included in the controlling areas of learning management and learning content in model three. The main reason for these similarities is simply due to the interrelated nature of categories within each model. Therefore, there are more overlaps when pairing one model with another.

______Chapter 2 Page 41 4.3. Discussion on the models

The summary of the six aforementioned models indicates two important indexes for the construct of learner autonomy. The first is related to the areas of performance, and the second concerns the levels of performance. Each index generally consists of three categories, except Nunan’s model (1997) which has five categories. This section attempts to draw a connection between these two indexes in relation to the general perspective of learner autonomy employed in this study. The areas of performance, namely cognitive processes, demonstrated behaviours, and situation management, serve as an overall index. Each category in this index consists of three typical progressive levels, characterized by groups of skills indicated in the other index. Table 2.2 represents this model.

Table 2.2: Adapted model employed in the current study

Cognitively identifying learning styles Psychological Cognitive processes Cognitively modifying tasks perspective Cognitively creating new task Socio-cultural Learner Performing selected learning styles perspective performed Demonstrated behaviours Autonomy Modifying tasks Political-critical in (classroom-like situation) Creating new tasks perspective Indentifying resources from contexts Technical Situation management Modifying the resources perspective (beyond classroom) Creating new resources

This model can be read from left to right but the vertical order is not important. For example, learner autonomy from the psychological perspective is performed primarily in cognitive processes. It can be described by a number of attributes such as the ability to identify preferred learning styles, modify learning tasks, and design new learning activities. The reference entries for the level index are presented in a relatively progressive order. However, it does not mean that the latter skills are higher and should be developed after the former. The three levels of progression is simply a way of representing the development of this construct. In addition, the three areas of the performance index overlap with one another. Success in one area can support the improvement of the others, although it is not always necessarily the case.

4.4. Summary

This section presents a comparative and contrastive analysis of the six models of learner autonomy under two types of classification. Each model has its own legitimacy and includes

______Chapter 2 Page 42 important indicators for the construct. However, there are several overlaps among the categories of each respective model. Grouping them under level and controlling area types results in two prominent indexes which are useful for the development of a more comprehensive model to be employed in the current study. It consists of three broad areas of control driven by the four perspectives on learner autonomy. Each area of control is characterized by a relatively progressive index of attributes. However, the level index of the model is kept open because it has not been integrated with empirical research. Therefore, the next section draws on prior research to identify the attributes valued and performed by learners in different contexts for model index enrichment.

5. Perceptions of learner autonomy

Although the literature on learner autonomy significantly acknowledges different variations in its definition and the impact of socio-cultural factors on learner autonomy, there has been little research on students’ perceptions of learner autonomy. In a series of focused group conversations in the United Kingdom, Lamb (2009) indicated that secondary school students generally want to control their learning activities but they had different perceptions of their ability and responsibility. This is very much in line with a report on Chinese EFL teachers in mainland China (Sinclair, 2009). These student teachers valued opportunities for critical reflection on their teaching practices but did not always appreciate the choices offered during the course. In other words, they wanted to be given freedom to carry out their learning processes but did not prefer to totally control the learning situation.

Similarly, in a context of self-access learning centres in Hong Kong, Gardner (2001, 2009) presents a complex perception pattern of both language teachers and students. The teachers’ perceptions of learner autonomy are varied; while their students’ patterns are quite similar and remain unchanged towards the end of the ten-week course. The investigations indicated that the learner autonomy dimensions recognized by students include customizing and personalizing learning processes (Gardner, 2007b). Other dimensions suggested by Japanese EFL students are self-initiative, making plans, self-control, flexibility, taking actions and concentration (Yang, 2007). These are among the most comprehensive indexes of learner autonomy dimensions documented in the contemporary literature.

______Chapter 2 Page 43 Apart from these studies, learner autonomy is often investigated through students’ responses to an interpretation of a literature-driven definition and an introduction of some teaching effect during a course. For example, in the context of EFL higher education in Vietnam, Trinh (2005) defined learner autonomy as a self-regulating ability, including three dimensions, namely planning, monitoring and regulating processes. He then modified the curriculum and used a task-based learning approach, providing the participants better learning choices to enhance their attitude towards and level of learner autonomy. As a result, the data analysis primarily focused on the effects of the curriculum and pedagogical innovation. Statistical tests were used to confirm the internal consistency of the three learner autonomy dimensions rather than generating a localized index of dimensions for the study participants and using it for other analyses.

Similarly, Nguyen (2009) defined learner autonomy as learner self-initiation and learner self- regulation and used strategy-based instructions to promote learner autonomy. These two dimensions relate to research outcomes in contexts other than Vietnam. Hence, the questionnaire instrument did not allow opportunities for any emerging dimensions of learner autonomy to be included. In other words, the study did not aim to empirically justify the pre- defined number of learner autonomy dimensions, although it is important to do so for such a sophisticated and socially-biased construct. Using an exploratory approach to understand learner autonomy from Vietnamese students’ perspective becomes even more important because students in Vietnam and Western contexts should be different. While Western students are often said to be active and independent, Vietnamese are stereotypically described as being passive and obedient learners (Le, 2008; Pham & Ngo, 2008; Wong, 2004). They also prefer to work in a big group of twenty or thirty rather than in smaller groups (Kramsch & Sullivan, 1996). Thus, their responses to learner autonomy may indicate dimensions other than those that have been suggested in the literature, especially when the contemporary Vietnam education is rapidly changing (Dang & Robertson, 2010c).

In short, learners’ perceptions of learner autonomy have not been significantly investigated in different contexts despite the contextually-driven nature of this construct. Nevertheless, several dimensions generated in different learning environments have been documented such as planning, initiating, regulating, monitoring, controlling, and personalizing learning activities. It seems that these studies do not aim to differentiate between cognitive and behavioural processes as indicated in the four perspectives. They assume that people do what

______Chapter 2 Page 44 they think best, despite a few opposite suggestions (such as Woodsong & Koo, 1999). Actually, the dimensions of these two processes can be understood at both cognitive and behavioural levels. Taking this into account, the following section aims to document more behavioural dimensions exercised by learners that have been identified in prior research. All of the dimensions, perceived and exercised by the learners, are then analysed and categorized to develop a framework for the current investigation.

6. Performance of learner autonomy

This section documents the learner autonomy dimensions performed by learners in different educational contexts to produce a more comprehensive index of the dimensions. As the development of learner autonomy needs to be fostered systematically and deliberately (Holec, 1981), educationists and researchers have used various practices, depending on particular situations, to enable learners to take more control of their learning. They include those with and without the involvement of ICT support.

6.1. Learner autonomy promoting practices in traditional classrooms

Identified in the literature are various practices which have been used to cultivate and nurture learner autonomy in language education classrooms. What researchers have often done in prior studies is to examine the contexts; introduce some effects into their classrooms, and analyse the outcomes. These effects are associated with some modifications of classroom activity, curriculum, and/or teaching method. This section looks critically into these attempts to document all the dimensions of learner autonomy that have been identified and their relationships with other associated variables in particular contexts.

6.1.1. Using classroom activities to foster learner autonomy

As learning activities are central to the language learning process, they have been designed to suit particular learning objectives (Conole et al., 2008; Lockyer, et al., 2008). Driven by different perspectives of learner autonomy, research has used both collaborative tasks such as group work, project work, and debate and individual activities such as reflective journals and extensive reading to promote learner autonomy. The former focuses more on opportunities for interaction and negotiation. The latter focuses more on opportunities for individual work.

______Chapter 2 Page 45 Although these appear to be opposed to each other, it is suggested that they both contribute positively to the enhancement of learner autonomy dimensions.

Group-based activities built on a cooperative learning approach have been identified to enable students to maintain an appropriate working agenda, as well as effectively implementing it, and constantly evaluating learning outcomes (Smith, 2001, 2003a). In his longitudinal study with different groups of undergraduate students majoring in languages other than English in Japan, Smith began by eliciting students’ interest in daily activities outside of classroom. Those who shared similar interests in activity types were allocated to work together. Each group was required to work out an action learning plan suitable for their favourite activities. For example, those who liked learning the language through topic discussion had to work out a list of topics that they would like to discuss and the way that the group discussion should be organized. The process was the same with those interested in debating and watching movies. They were then required to implement their plans until the end of the course.

Despite certain difficulties in adapting to the cooperative learning environment, these students became better controllers of their learning activities towards the end of the course. When given more power to manage their learning, they were willing to take the opportunity to do so. Most of them demonstrated their preference for working in small groups and expressed their desire to continue in the same fashion in the following semester. They also explicitly expressed positive satisfaction for their achievements during the course. However, others indicated that they needed more instruction, advice, and strategy training in certain learning activities. This result appears to negate the stereotypical assumption that culturally, Japanese learners tend to passively wait for knowledge to be transferred from teachers (Maiko, 2003; Rundle, 2007). It also called for a shift in the teachers’ roles, from a more teacher-fronted position to a more student-centred position.

Group-based activities with a wide range of choices are also identified as increasing students’ learning engagement and achievement (Aoki, 2001). When being asked to participate in designing content for a methodology course, a student teacher in this study was strongly resistant to the initiative. She did not like sharing ideas with other peers during group discussion sessions. However, she became involved in the activity and developed a close relationship with other group members towards the end of the course. Working with the group also enabled her to identify learning goals and increase motivation. Similarly, other students

______Chapter 2 Page 46 reported that their attitudes towards the course became more positive as they could freely participate in and manage their learning process. It can be seen that collaborative activity can empower learning even with those who are only familiar with individual activity. Interactive learning can be the preference of any students if the appropriate process of instruction and adaptation is undertaken.

However, this does not mean that individual learning activities are not important in developing the students’ ability to control their learning. Hart (2002) indicates that a suitable combination of both collaborative and individual activities for learning can improve reactive learner autonomy. In this study, over ninety college students formed their own groups, chose a topic and outlined a number of sub-topics in which they were interested. After that, each member selected a sub-topic and looked for materials from different resources to investigate it. They then synthesized all the details collected, gave a presentation on the topic during the course and made a poster by the end of the course. Apart from these group work activities, each student was asked to give an oral presentation and write a critical personal reflection on his/her learning process. Although there was no statistical evidence for the students’ overall progress, their reflective diaries clearly suggested that these activities positively improved their learning attitudes, strategy adoption, and language authenticity.

Similarly in another study, over twenty language teachers from different institutions in Hong Kong suggest that project work, group work, reflective journals, and extensive reading are among the most appropriate learning activities for learner autonomy enhancement (Voller, 2005). They can raise students’ learning awareness, provide more learning choices, offer better opportunities for negotiation, increase the use of metalanguage, and trigger more reflections on learning. These suggested activities are based on particular constraints in various local working environments such as syllabus structure, exam pressure, resource limitation, school system, and social situation. They again reaffirm that raising awareness, interacting with others, reflecting on one’s learning, and employing suitable strategies are essential dimensions of learner autonomy.

6.1.2. Innovating teaching methods to foster learner autonomy

In addition to designing class activities, innovating teaching methods is a popular practice adopted by educationalists to facilitate learner autonomy. Braine (2003) used a learner-

______Chapter 2 Page 47 centred approach with Chinese undergraduate students and suggested that it can enhance their learner autonomy at both cognitive and behavioural levels. Despite being exposed to a predominantly teacher-centred approach since early schooling, the participants in this research gradually became active when peer feedback during a writing course was offered. They perceived the method as an opportunity to express their opinion, negotiate it with other peers, and evaluate their writing. As a result, their satisfaction with the course increased. However, it is necessary to note that their negotiation comments are culturally bound. They tend to avoid criticism and disagreement with their classmates’ writings, which is different from Western students (Carson & Nelson, 1996) who are more direct in peer reviewing regardless of potential conflicts.

In relation to the teacher’s role in introducing innovative teaching methods, evidence suggests that a more facilitative and consultant role is better for learner autonomy enhancement (Smith, 2001). In an empirical experiment in China (Naizhao & Yanling, 2004), English language students working with teachers as facilitators, counsellors, and resource persons consistently achieved higher scores in the final exam than those working with teachers as knowledge transmitters regardless of proficiency levels and out-of-class learning time. In addition, students in the former groups became more self-confident, collaborative, and aware of learning goals and strategies. Although improvements in learner autonomy dimensions were based on qualitative rather than statistical comparisons, a positive relationship between the facilitative teacher, learner autonomy dimensions, and learning outcome was acknowledged.

Another innovative attempt relating to teaching practice is the use of informed goal-setting as in Koda-Dallow and Hobbs (2005). The qualitative interview data from the study indicate that undergraduate students who were required to explicitly set out weekly learning goals during a five-week Japanese course could personalize their learning process more easily than others. They even adopted goals that were challenging and beyond the course requirement but interesting to them. This illustrates a positive relationship between cognitive awareness and learning behaviour. Students become more committed to achieving their goals if they are explicitly set out (compare with Locke & Latham, 1990). However, the quantitative data on students’ perceptions of different learner autonomy dimensions such as selecting materials, correcting mistakes, looking for resources, and identifying suitable strategies did not show any significant improvement. This suggests that the treatment and time length of the study were inadequate for learner autonomy development.

______Chapter 2 Page 48 6.1.3. Modifying curriculum to foster learner autonomy

Another way of fostering learner autonomy suggested in the literature is related to curriculum modification such as changing materials or adding an extra component to the existing program. For example, Fonseka (2003) replaced several sections of the course textbook with English songs and suggested that it enhanced children’s motivation and communicative competence. Given the poor resource learning environment in public schools in Sri Lanka, authentic English songs would have drawn students’ learning interest. Singing together in a cheerful spirit and carnivalesque atmosphere can help students remember the conversation lyrics easily. These extracts were naturally replicated in daily interactive communications.

In another example, students in a collaborative project in Hong Kong were asked to undertake several activities apart from their normal curriculum, resulting in improvements in different dimensions of learner autonomy (Miller, et al., 2005). These activities include collecting authentic materials for a self-access centre, conducting extra tasks in relation to the class topics, and conducting project work. For instance, students were encouraged to organize a restaurant and act as working staff in that simulated environment after studying the topic ‘Food’ or develop a number of songs for a school show. Students reported that they liked such open activities and became more actively engaged in learning. They also became more motivated and confident after collaboratively managing these processes on their own.

6.1.4. Summary

The dimensions of learner autonomy that have been reviewed in the previous sections are extracted from investigations conducted in traditional learning environments, which have identified various ways of promoting learner autonomy in different learning contexts. These can be related to organizing learning activities, changing teaching methods, and modifying curriculum. However, they all aim to create a learning situation where learners have more choices and freedom to develop their capabilities through interacting with peers and/or tasks. It also means that teachers are required to step back and provide facilitative support to help students complete a given task successfully. The section which follows continues to look at such attempts which have been conducted with the support of ICTs. Any dimensions of learner autonomy manifested in the technological environment are accumulatively documented to enrich the dimension index.

______Chapter 2 Page 49 6.2. Learner autonomy promoting practices with ICT support

With the increasing trend of integrating ICTs into teaching and learning EFL (Hubbard, 2005; Jung, 2005; Kanniah & Krish, 2010; Levy & Stockwell, 2006), learner autonomy has been investigated through e-learning and blended learning environments. Recent research (such as Healy, 1999; Littlemore, 2001; Murray, 1999; Schwienhorst, 2003, 2008a, 2008b) has suggested that computer-assisted language learning (CALL) can provide facilitative opportunities for learners to develop learner autonomy. This section draws on different studies in this area to understand the particular dimensions of learner autonomy that can be fostered through an ICT-supported learning space. Included is the use of ICT-supported activities inside and outside the classroom although the former is less popular than the latter. The distinction between these two categories is only for ease of structuring the section, as most ICT-supported attempts have been conducted beyond the boundary of the physical classroom.

6.2.1. Using ICT-supported activities in face-to-face classroom

In a technology-rich learning classroom, Christopher (2006) shows that interacting with technology enables students to exercise their controlling ability; tailor the learning process, and increase their engagement. This project uses El Investigador en Español, a software application for learning Spanish, to assist undergraduate students navigate through the learning activities. The software offers students a number of topics from which they are free to choose one in which they are interested. They then use the classroom computers to research the topic; develop and revise it; and send to their friends for peer-assessment. This gives them the opportunity to work with a topic related to their interest and directly shape the content of the course in which they are involved. As a result, they receive more specific advice from peers and teachers in relation to their learning process. In addition, this activity can reinforce an authentic link between the problems that they try to solve in a learning task and their concerns in daily life.

ICT-supported activities are also found to provide students with opportunities for reflection, negotiation, and language rule implementation. Blin (2004) reported on a case study of about forty first-year students studying French as a foreign language for twelve weeks. They were asked to collaboratively build a website for learning French. Therefore, they had to perform a series of activities such as retrieving information, generating thematic materials, writing up

______Chapter 2 Page 50 meeting minutes, and negotiating on labour distribution. Engaging in these activities enabled students to collectively reflect on their language learning and try different linguistic structures that they come across. They could also develop skills that help them learn from others in the community.

However, virtual collaboration on particular activities is not always useful for learner autonomy development. As students are required to complete tasks, they may interact more with the computers than with their peers (Christopher, 2006). This reduces their time for negotiation and discussion activities which are more important for managing their learning processes (Blin, 2004). In addition, students sometimes take advantage of such a supervision- free virtual space to produce a final product without proper attention to the process of doing it (Blin, 2004). Such a problem becomes even more serious when this virtual instruction method is not favoured by students who prefer more face-to-face grammatical lessons or exercises (Christopher, 2006). They can regard being involved in such activities as a waste of time for their language learning.

In addition to students’ perceptions on teaching and learning practices, students’ efforts and the nature of the course plays a central role in shaping opportunities for learner autonomy development. It has been observed that tutor talk time dominates in online synchronous tutorials, and most of the interactions are between tutors and an individual student (Heins et al., 2007). It is harder to get students to work in groups in such a less structured class. In addition, students’ responses are found to be more structured. The result is that opportunities are diminished for informal and spontaneous chit chats which are considered important for autonomy development.

6.2.2. Using ICT-supported activities outside of the class

In contrast to the limited number of studies using ICT support in face-to-face classrooms, there have been a lot of investigations conducted online. Empirical research in tandem learning has indicated that both synchronous and asynchronous communication channels can help students effectively reflect on their learning process, as well as engage in learning through interacting with peers, and experiment with the target language in authentic situations (Schwienhorst, 2003, 2008b). As the tandem exchanges occur between students, they gradually become successful in controlling the activity. The text-based environment is

______Chapter 2 Page 51 motivating for students to choose preferred topics and lead the conversations to meet their learning needs. In addition, these archived conversations give students an opportunity to re- read any message, evaluate it, and keep track of their learning (Little, 2001). Their metalinguistic awareness is therefore increased, especially with the support of appropriate feedback from their partners.

Students’ learning reflections, interactions and motivation are also promoted through the LMS, blog, discussion forum, and wiki environment. Gitsaki (2005) integrated an online Backboard component in an intercultural undergraduate course for ESL Japanese students. Different levels of collaboration on authentic resources motivated students to challenge their understanding of cultural issues from multiple perspectives. Similar suggestions are also made by Pinkman (2005) and other researchers (Kessler & Bikowski, 2010), who investigated learner autonomy development through students’ participation in blog and wiki spaces, respectively. Opportunities to exchange opinions with and receive feedback from others can encourage students to experiment with the target language more frequently and successfully complete the activities. Students then become better at managing their learning activities (Sidhu & Embi, 2009). However, it has been observed that students’ participation is varied. Some contribute a lot more than expected; others simply want to meet the course requirements (also in Blin, 2004).

In addition, online collaboration in a relaxing atmosphere has been identified as a way to promote students’ monitoring and personalizing learning processes. Attempts such as the e- China projects (Sinclair, 2009), iEARN and the Global Classroom Project (Wells, 2007) have used the learner-oriented approach to facilitate cross-cultural communication and project- related skills. Choices and opportunities offered during these courses enable learners to reflect critically on their teaching experience and exercise their metacognitive knowledge. However, these opportunities are not always appreciated. Some learners in the e-China project were happy with their choices; but some others indicated that there were too many choices. One learner even studied every topic in the presented order so as not to miss any. This indicates that the students were not familiar with having wide choice and freedom to choose which they found misleading in such a material-rich environment (Figura & Jarvis, 2007).

Further investigations on such out-of-class learning environments where the target language is a second or foreign language have indicated that students always tend to choose receptive

______Chapter 2 Page 52 skill-based activities rather than productive skill-based activities with which to work (such as Freeman, 1999; Hyland, 2004; Mary, Gillian, & Victoria, 2002; Pickard, 1996; Pill, 2001). This is because these kinds of passive activities are easier for students to engage in individually. They tend not to choose productive activities mainly because they are afraid of having to interact with others. Other factors contributing to this tendency, as suggested by the qualitative data, may also be social and political factors (Hyland, 2004). Nevertheless, these active out-of-class students are considered successful language learners who make use of different dimensions of learner autonomy.

Another study on a group of Junior High School students in Indonesia (Lamb, 2004) also presents rather similar results on these kinds of out-of-class learning activities. Passive learning activities such as watching television or videos, listening to songs, reading books or magazines are commonly reported by these highly motivated participants. However, the most frequently undertaken activity identified in the data is attending private courses. This is interpreted as working in a collaborative community although no insightful examination of this type of out-of-curriculum course is made. Students are also reported to use the Internet sometimes for learning purposes, but qualitative data fail to specify the kind of activities adopted. In contrast, teacher learners are suggested to actively engage in an online social network specifically designed for reflective learning and professional development (Ushioda, et al., 2011). They appear to use the virtual community to reflect on their teaching practices and share their experience with others. These examples suggest further research on the relationship among learners’ age, interactive patterns, and affordances of the communicative platform.

Nevertheless, it seems to be clear that these active out-of-class learners demonstrate certain levels of ability to control their learning processes. Performance related to planning courses of actions, monitoring learning activities and evaluating learning progress is recorded. When learner autonomy is not fostered in the curriculum and through teaching practices in such a stereotyped passive learning system, personal psychological attributes and social factors are assumed to play a greater role in shaping autonomous learning behaviours. This also reflects social pressure which is somehow contradictory to traditional cultural values, a phenomenon commonly seen in developing countries.

______Chapter 2 Page 53 Other investigations on reasons for choosing certain types of activities for learning in a self- access centre with a group of international students taking General English or English for Academic Purposes courses in New Zealand indicate that their decisions are driven by their perceptions on the importance of the activities and their weaknesses (Wallis, 2005). Even though they adopt individual or collaborative, and receptive or productive activities, they all take some degree of control over their learning processes except for evaluation. In addition, out-of-class learning is often considered to be of either equal or less importance than formal learning although qualitative data suggests that some students do not really value formal class lessons. Again, personal beliefs on learning play a key role in exercising and implementing out-of-class activity options.

6.2.3. Summary

In summary, learner autonomy has been promoted differently, even in the ICT-supported learning environment. These variations are generally due to dissimilar interpretations of the construct, the availability of technological infrastructure, and other learner-related variables (Littlemore, 2001; Richards, 2005). In addition, fostering learner autonomy potentially entails modifications to curriculum design, material development, pedagogical paradigm, management ideology and strategic development (Benson, 2006). Therefore, any local practice is constrained by institutional and ideological factors. The linkage between learner autonomy and ICTs is largely due to the freedom and flexibility that the ICT environment creates for students to learn at their own pace and convenience both inside or outside class (Boulton et al., 2008; Duda, 2005).

Particularly, several technologies documented in the learner autonomy promoting practices in this section include standalone language learning applications, email, chat, discussion forums, wiki, and blogs. These interactive and authentic environments support different dimensions of learner autonomy, depending on particular contexts. Students have demonstrated that they become better at planning, tailoring, personalizing, and monitoring their learning process when interacting with peers and materials in virtual space. Their motivation and participation in the learning activities is therefore increased. However, the level of their engagement and collaboration is not the same across all participants. Some are not ready to perceive the degree of power transferred to them from the instruction method; others do not value such an informal learning space.

______Chapter 2 Page 54 From an overview of learner autonomy dimensions derived from theoretical perspectives and identified in empirical research as described previously, the following section attempts to systematically compile all of these dimensions to produce a conceptual framework of this construct. As learner autonomy has been exercised in both the traditional classroom and the ICT-supported space, it is necessary for this framework to be shaped in such a way that dimensions in both types of learning environments are included. This general framework which is applicable for both online and offline contexts will then be used as a theoretical background for the current study.

7. Conceptual framework of learner autonomy adopted in the current study

Most of the empirical investigations documented in the previous sections either take a single perspective or focus on promoting certain dimensions when examining learner autonomy. This is because this construct can be recognized in different forms (Benson, 2001). Given the exploratory purpose and the rhetoric context of this research, it aims to develop a comprehensive framework, consisting of two layers. The first one, looking into the inner side of the construct, is called dimensions of learner autonomy. The second, looking at the outer side of learner autonomy, is called mediated factors.

7.1. Dimensions of learner autonomy

Presented in Section 5 and 6 of this chapter are a number of studies investigating learner autonomy in different research situations. Although they may take different perspectives and interpretations of the construct, most of them use the phrase controlling and/or managing learning processes to refer to the performance of learner autonomy (e.g., Christopher, 2006; Lamb, 2009). In addition, several dimensions of learner autonomy are repeatedly identified in different investigations. Some of these are only linguistically different. For example, personalizing, customizing, and tailoring are respectively used in Sinclair (2009), Gardner (2007b) and Christopher (2006). Further examination also shows that some dimensions can be included in some others. For example, selecting appropriate materials (in Blin, 2004 and Miller, et al., 2005) and employing suitable learning strategies (in Hart, 2002 and Voller, 2005) can be argued to be under personalizing learning.

______Chapter 2 Page 55 It can be seen that the term dimension is not consistently used in the literature. It is used to refer to either a specific or general aspect of learner autonomy. To overcome this terminology problem, the term attribute is used to refer to all of these general and specific dimensions. These attributes are then categorised to conceptually formulate the construct of learner autonomy. A thematic analysis indicates that these attributes can be grouped into three types of processes, namely initiating, monitoring and evaluating learning. This categorization matches the proposal by Little (2003). Therefore, the attribute index of learner autonomy is arranged into three categories as presented in Table 2.3.

Table 2.3: Index of learner autonomy attributes in three processes

Processes Attributes Source examples being aware of learning goals and strategies identifying goals and setting goals Aoki (2001) Initiating making plans or work agendas Nguyen (2009) looking for resources Yang (2007) maintaining agendas and keeping track of learning identifying and employing suitable strategies Christopher (2006) selecting appropriate materials Conole, et al. (2008) tailoring, customizing and personalizing learning Gardner (2007b; 2009) Monitoring being flexible and regulating learning Lamb (2009) taking actions or implementing agendas Nguyen (2009) concentrating on learning Trinh (2005) collaborating and interacting with others Yang (2007) expressing opinions and negotiating with peers reflecting critically Sinclair (2009) Evaluating evaluating learning outcomes Little (2003) correcting mistakes

Although these three processes seem to be distinct from one another, they actually overlap when being performed. To be able to monitor a learning process successfully, a student may need to evaluate his/her current practices and initiate new learning opportunities. Similarly, an attempt to initiate a learning process may originate from an evaluative purpose. For example, an EFL student approaching two native English speakers can be interpreted as his/her initiating a learning opportunity or evaluating his/her English proficiency, or even both. In addition, the development of one process may be dependent or independent from the development of the others. A student good at monitoring learning processes may also be good at evaluating learning processes, but this is not always necessarily the case. These three processes can also be considered complementary to one another in cyclic order during a course of learning activities.

______Chapter 2 Page 56 The interrelated pattern of the three learning processes is therefore reflected in the attribute index, although each process has its own distinct attributes. The initiating learning process involves attributes related to understanding personal learning preferences, setting goals, preparing study plans, and creating learning opportunities. The monitoring learning process includes attributes related to learning engagement and maintenance such as selecting appropriate strategies, modifying learning paths, and negotiating with others. This is probably the longest process, where most of the learning takes place. The evaluating learning process anticipates attributes about assessing learning outcomes such as proofreading an assignment and evaluating a piece of writing. It can be seen that attributes across the three processes may be operated sequentially or simultaneously during a learning act. For example, students may plan to read one book every two weeks. They then choose some titles that they like and maintain the reading activity. However, after three weeks, they find that reading these books is not very useful and decide to read other titles or change to watching movies.

Figure 2.2 visually summarizes the three learning processes of learner autonomy. The construct of learner autonomy is considered to consist of cognitive and behavioural components as argued in Section 3, 4, 5, and 6 of this chapter. The exercise of behavioural components can be in the context of the classroom and beyond as indicated in the adapted model discussed in Section 4. These two components are responsible for controlling the learning activity. Particularly, they control the three processes, namely initiating, monitoring, and evaluating learning. These three processes may take place in a cyclic order to generate effective learning outcomes. These processes are, of course, mediated by other factors which are discussed in the next section.

Initiating

Cognition Behavior : Controlling direction Evaluating Monitoring

LEARNER AUTONOMY

Figure 2.2: Processes of learner autonomy (adapted from Dang & Robertson, 2010b)

______Chapter 2 Page 57 7.2. Mediating factors of learner autonomy

While the previous section investigates the inner structure of learner autonomy, this section synthesizes the external factors that may have direct influence on the performance of this construct. As suggested in Section 6 of this chapter, personal preference, attitude and motivation appear to be the three major mediating factors. Students with different learning preferences may be interested in different levels of activity flexibility and interaction (e.g., Braine, 2003; Figura & Jarvis, 2007; Nguyen, 2009; Smith, 2001; 2003a). Some may prefer to work on their own, while others like to cooperate with peers in groups. In addition, some are more active in the online learning space, while others tend to engage more in offline learning activities.

Similarly, students with different learning attitudes and beliefs may also perceive similar learning opportunities differently (e.g., Trinh, 2005; Wallis, 2005). For example, while some students with rote learning habits probably favour grammatical lessons and drills, others may prefer learning by completing projects. Part of the students’ autonomous learning capacity is associated with attitude (Dam & Legenhausen, 2010), and it can contribute to students’ EFL proficiency. Benson (2010) argues that positive attitudes can lead to high expectation, and this enables students to implement their learning plans and monitor their learning progress. Although learner autonomy and attitude is not always observable and measurable, a positive relationship between the two appears to be obvious.

Furthermore, the relationship between learner autonomy and motivation has been argued to be strong (Murray, Gao, & Lamb, 2011). Motivation is suggested to be influential on students’ performance of learner autonomy (e.g., Aoki, 2001; Miller, et al., 2005; Schwienhorst, 2003, 2008b). Those who are more motivated may be more eager in looking for learning opportunities and maintaining more effective learning engagement (Reinders & Balcikanli, 2011). Motivated students are also suggested to take part in autonomous learning behaviours more frequently (Tok, 2011). Other research shows that the relationship between learner autonomy and motivation is very complex (Lamb, 2010). Hozayen (2011) argues that motivation can empower learner autonomy, while others (Lamb, 2001; Lamb & Fisher, 1999) suggest that an increase in learner autonomy can enhance students’ motivation. It seems that this relationship is interactive rather than unidirectional. Figure 2.3 (page 59) visualizes the relationship between the three mediating factors and learner autonomy.

______Chapter 2 Page 58

Preference Motivation

LEARNER AUTONOMY

: Interactive influence Attitude

Figure 2.3: Learner autonomy and its mediating factors

7.3. Conceptual overview of learner autonomy

Learner autonomy has been argued to be shaped by attributes from personal psychology, learning environments, interactions between personal psychology and environment, and the desire for access and power. Thus, the current study employs Vygotsky’s notions of socio- cultural theory (Lantolf, 2000; Lantolf & Thorne, 2006) and community of practice (Lave & Wenger, 1991; Wenger, 1999; Wenger, et al., 2002) to understand human behaviours, that is, students’ learner autonomy in particular. This approach positions learners into a local environment whose identities, resources, and practices construct its members’ capacity of learner autonomy (Toohey, 2007). Psychological attributes of learner autonomy are primarily shaped from one’s access to and interaction with the ideologies and desires of the local community. In addition, technical attributes are basically mediated by the local resources. Hence, the performance of learner autonomy is internally generated after one’s multi- directional negotiations with his/her community enablements and constraints (Dang, 2010). These are illustrated in Figure 2.4.

Resources Practices LA: Learner Autonomy LA

Persons

Figure 2.4: Theoretical position (adapted from Toohey, 2007, p. 233)

______Chapter 2 Page 59 Interactions and negotiations of a person with his/her community can be one-on-one, immersion or delegation (Wenger, 1999), depending on his/her situation in each activity. The level of an individual’s participation in an activity also mirrors both his/her personal characteristics, constraints, and enablements formed by that contemporary environment. In other words, moving one’s position from a peripheral participant to an insider, from a passive to an active member, and from a spontaneous to a controlling learner in a community requires a lot of facilitative practices from the context. Therefore, learner autonomy is a socially-bound capacity, and its development needs to be examined in relation to other associated factors in the context (Dang, 2010). This theoretical position also properly conforms to the mixed perspective adopted in this research. It helps decrease the probability of individualistic perspective on learner autonomy (Pemberton, et al., 2009). On the one hand, it values the significant impact of personal attributes and independent internalization. On the other hand, it acknowledges external contributions from the context. Therefore, the highest level of learner autonomy should be still described via dialogic negotiations and interactions within the immediate situation. It should not be considered as an ability to work alone solely.

8. Conclusion

Drawing on both theoretical and empirical studies to discuss the construct of learner autonomy, this chapter has aimed to shape a proper definition and theoretical position to be employed in the current investigation. Learner autonomy is argued as a capacity which enables learners to take charge of their learning. Consisting of cognitive and behavioural components, this capacity controls learners’ course of learning through three core cyclic processes, namely initiating, monitoring and evaluating learning. The performance of each process is manifested by different attributes. The attributes in each process includes three levels, namely identifying, modifying, and creating learning-related variables. These attributes can be fostered in both the traditional classroom and the ICT-supported environment. In addition, students’ performance of these attributes is characterized by three mediating factors, namely personal preference, attitude, and motivation.

In addition to these attributes and mediating factors, historically, learner autonomy in language learning has been interpreted from four perspectives. Each perspective focuses on particular factors that potentially shape the performance of this construct. The psychological perspective emphasizes personal attributes of the learners; the technical perspective involves

______Chapter 2 Page 60 attributes from the learning environment; the socio-cultural perspective focuses on the interactions between learners and their environment; and the political-critical perspective considers learners’ access, control, power and ideology in their community.

The current study employs a theoretical position derived from socio-cultural theory and the communities of practice literature. It takes into account contributions from personal and contextual aspects, as well as the interactions between the two.

Provided with the conceptual framework and the theoretical position of learner autonomy, the study seeks to understand how learner autonomy is perceived and exercised in each particular learning context. The purpose for this is to design effective learner autonomy promoting practices. Research in the area of EFL learning in higher education in Vietnam is scarce. Only limited understanding on students’ perspectives has been documented. Therefore, the current research sought to identify local Vietnamese students’ perceptions of learner autonomy and the factors that mediate their performance of this capacity.

Such an understanding potentially enriches the merit of Vietnamese students’ perspectives in the area of learner autonomy in the contemporary literature. On one hand, it potentially provides more insights into the conceptualization of this socially-shaped construct by enabling research in the field to compare and contrast the Vietnamese patterns of learner autonomy with those in other contexts. On the other hand, it suggests pedagogical implications for the learner autonomy fostering practices to be taken in the local context.

9. Summary

Learner autonomy has been identified as an important goal of education in general and EFL education in particular. Empirical research has suggested that the capacity of learner autonomy has different dimensions and can be perceived differently in different socio-cultural learning contexts. In other words, the interactions between learners and their learning environment are essential for the formation and development of learner autonomy. Hence, the current study aims at understanding learner autonomy from the perspective of EFL students in higher education in Vietnam. The investigation also examined students’ performance of learner autonomy in the local context and the factors that mediate students’ autonomous learning behaviours. Findings from the study can potentially contribute to the shaping of

______Chapter 2 Page 61 learner autonomy promoting practice in Vietnam and similar contexts. The study design and methodology are further presented in the following chapter.

______Chapter 2 Page 62

Chapter Three

METHODOLOGY

______

Chapter One focused on the need to investigate EFL students’ perceptions of learner autonomy and their performance as autonomous learners in the context of Vietnamese higher education. The literature review and analyses in Chapter Two drew on different theoretical arguments on the formation and development of learner autonomy. Empirical research documented the attributes of this construct in both traditional and ICT-supported learning spaces. From these analyses, a theoretical position based on Vygotsky’s (1986) socio-cultural theory and Wenger’s (1999) community of practice construct was established for the current investigation. This chapter describes the methodology employed for this project to achieve the objectives proposed in Chapter One and indicates how evidence will be gathered to answer the following research questions:

1. What are Vietnamese EFL undergraduate students’ perceptions of learner autonomy? 2. What are the similarities and differences in their learner autonomy perceptions? 3. What is the relationship between their understandings of learner autonomy and learning behaviours? 4. Does this relationship change during a semester-long course? 5. What are the factors that mediate students’ performance of learner autonomy in the local context?

This chapter starts with a discussion of research perspectives and mixed methods in relation to these research questions. It then reports on the design and methods for two sequential studies developed for this research investigation. In each study, the chapter describes the participants, objectives, data collection procedures, and instrument design. The chapter then finishes with discussions of issues of reliability, validity and ethical considerations.

______Chapter 3 Page 63 1. Research worldview

The philosophical worldview employed in this research project acknowledges the assumptions on ontology, epistemology, and axiology of different paradigms. As learner autonomy has been perceived through four different perspectives, the employment of a single paradigm is insufficient for the investigation of this construct. For example, the technical perspective only acknowledges the external impact on the construction and development of learner autonomy, suggesting that there is only one reality as argued by post-positivism (Denzin & Lincoln, 2005; Tashakkori & Teddlie, 2010). This means students will become autonomous when staying in an autonomous learning situation. However, research has shown that this is not necessarily the case (Benson, 2006). Moreover, students’ performance of learner autonomy may be mediated by personal characteristics and other momentary factors. A certain learner autonomy promoting practice may work successfully with some students but may have negative effects with some others. This indicates that multiple realities exist, supporting the assumptions of constructivism (Guba & Lincoln, 2005; Paul, 2005; Slife & Williams, 1995).

In addition, the theoretical position employed in this project allows multiple views on knowledge construction and human-related phenomenon interpretation. The socio-cultural and community of practice theories can help interpret the development of learner autonomy, based on interaction and participation in a community. Therefore, members of the same communities may obtain some shared patterns of behaviours and beliefs. However, such general patterns may not necessarily be accurate for every single member in that community or those in other communities. In other words, knowledge is constructed through negotiation and agreement within a group of people. This means it is necessary to actively collaborate with students in a particular situation to understand their perceptions and behaviours of learner autonomy. In addition, it is important to adopt an outsider’s perspective and critically reflect on the phenomenon. These different ways of approaching a problem can provide more complete insights for any investigation. Such a position shift is reflected in the description of the research design below.

______Chapter 3 Page 64 2. Research design

It is the complexity of learner autonomy that calls for a multiple approach research design rather than a single set of data to increase the level of confidence in answering the research questions. Therefore, this project employed a mixed methods approach, characterized by a process of generating and interpreting both quantitative and qualitative data (Tashakkori & Teddlie, 2010). A combination of different data sets, collected at different stages within the research project, can capture more details of the situation being investigated (Creswell & Clark, 2011). Especially for a multi-dimensional construct such as learner autonomy in language education, a mixed methods approach can result in a more informative picture than either alone (Hubbard, 2005). The quantitative data allows the project to identify local students’ patterns of learner autonomy perceptions and performances; and the qualitative data can elicit specific aspects of the local learning situation that contribute to the shaping of these patterns. This combination of data sets can also be complementary to one another, supporting the project to extract more insightful understanding of the research problem (Johnson, Onwuengbuzie, & Turner, 2007; Maxwell & Mittapalli, 2010; Tashakkori & Teddlie, 2010).

The specific research questions in this study involved seeking quantitative data that could be further informed from follow-up qualitative evidence. Such a mixed methods design followed a “sequential explanatory” model (Creswell & Clark, 2011, p. 57). Hence, the current research project consisted of two studies with different participant groups from the same population. It initially collected quantitative data, interpreted it, and generated themes to inform the qualitative data collection procedure that followed. The analyses of the qualitative data were then used to explain and elaborate assumptions concluded from the quantitative data set.

In the current project, Study One used the quantitative data generated from a survey questionnaire to identify the dimensions of learner autonomy perceived by local students. The instrument was then revised to reflect the dimensions of learner autonomy perceived by local students only. After that, Study Two was conducted with a focused cohort of students during a 16-week course. It consisted of two phases. The first phase collected two sets of quantitative data, using the revised questionnaire and log records generated from the students’ virtual learning space. The questionnaire was designed to measure the students’ perceptions and performances of the identified learner autonomy dimensions; and the log records were used to profile their online learning engagement. The second phase generated the qualitative data

______Chapter 3 Page 65 from the semi-structured interviews with students. This qualitative data set was designed to understand factors that mediated the students’ performance of the identified learner autonomy dimensions in the local context.

The weighting focus of the data interpretation is equally distributed among the data sets because each data set aims to answer its respective research questions. However, they were collected in a sequential order because these research questions are logically interrelated. Satisfactorily answering the first is a prerequisite for considering the next. For example, in order to understand how students exercise learner autonomy, it is necessary to understand

Phase Procedure Product

Study One Quantitative Cross university survey Mainly numeric data data collection (n=562)

Data screening Descriptive statistics Quantitative Frequencies Factor loadings data analysis Exploratory factor analysis Items identified Validity Four dimensions identified Survey revised

Study Two One cohort of students in a Mainly numeric data Quantitative university data collection (n=247)

Data screening Descriptive statistics Quantitative Frequencies Level of each dimension data analysis Reliability Relationship between Correlations perceptions & performances Students’ LMS engagement Generating themes from the Semi-structured interview Qualitative instrument & protocol four dimensions questions development

Selecting randomly on a Interview transcripts Qualitative data voluntary basis (n=11) collection Face to face interviews

Thematic coding Patterns of mediating Qualitative data Within-case and cross case factors analysis theme development Variations across cases Cross theme analysis

Interpretation and analysis Discussion Integration of of quantitative and Implications results generated from qualitative results Limitations all the data sets Further research

Figure 3.1: Diagram of the sequential mixed methods design employed in the project

______Chapter 3 Page 66 what dimensions of learner autonomy they value, given the multidimensional and socially driven nature of this construct. Similarly, in order to understand the factors fostering and inhibiting local students’ development of learner autonomy, it is important to understand how students exercise this capacity in particular learning environments. Finally, the discussion draws on the integration of both quantitative and qualitative data sets (Creswell & Clark, 2011; Morse, 2010) to provide an insightful profile of learner autonomy in the investigated context. This sequential procedure also allows the study to increase the reliability and validity of the instruments. Figure 3.1 (page 66) visually presents procedures of the mixed methods sequential explanatory design employed in the current study. The procedures of both Study One and Study Two are presented in the following sections.

3. Study One

This section describes Study One in detail, featuring its objectives, hypotheses, participant recruitment, instrument development, and data collection procedures. To maintain the structure of the report, the main results of the study are included at the end of this section. It briefly mentions the main perceptions of learner autonomy identified from the participants’ responses. The analysis process to generate these perceptions is presented in Chapter Four; and a follow-up discussion is given in Chapter Five.

3.1. Objectives and hypotheses

Study One aimed to investigate Vietnamese EFL undergraduate students’ perceptions of learner autonomy in both online and offline learning environments. These perceptions were categorized into groups and called dimensions of learner autonomy valued by local students. A general understanding of these dimensions was used as a core baseline for further investigations of this construct in the local context. Any identified relationship between these dimensions also provided important directions for the construction of this concept. In addition, the resulting instrument became a useful tool for measuring this variable at both cognitive and behavioural levels in follow-up studies or similar contexts.

Study One also aimed to examine the relationship between learner autonomy perceptions and general variables such as gender, socio-economic factor, and level of computer proficiency. It was hypothesized that most of the dimensions perceived by local students were significantly

______Chapter 3 Page 67 correlated with the socio-economic factor because learner autonomy is a situation-biased construct. However, only the dimension associated with ICT usage was influenced by students’ general level of computer proficiency. The study did not expect any correlation between learner autonomy and gender because no salient difference between males and females has been documented in the field.

These research objectives called for a research approach that could generate systematic and replicable data. This means the patterns of learner autonomy perceptions generated from the study needed to reflect the perspective of a reasonably large sample of the population. These results then needed to be applicable for the sample of a follow-up study, Study Two. Such requirements can be successfully met with quantitative research. Therefore, a questionnaire instrument was employed. This quantitative method could also produce rigorous and reliable research outcomes for the investigation (Dornyei, 2007). In addition, it enabled the researcher to investigate the existence and strength of the relationship between variables through correlation analysis (Mackey & Gass, 2005). In Study One, it allowed examinations of the relationships between different perceptions of learner autonomy and the effects of general variables on these perceptions.

3.2. Participants

The target population of this study was Vietnamese undergraduate students who were currently majoring in an English-related major such as EFL, Teaching English to Speakers of Other Languages, British and American literature, British and American culture, or English translation and interpretation in public universities in Vietnam. As the questionnaire consisted of 62 items and one of the study aims was to run exploratory factor analysis to identify themes, sample size was important. The study was aimed at surveying from 400 to 600 participants (the requirement for a particular number of cases in exploratory factor analysis is discussed in Cohen, Manion, & Morrison, 2007). The study also wished to include different groups of the population from varying socio-cultural situations to examine its impact. Therefore, it included participants from four typical regions of Vietnam, namely the North, Central, South, and Mekong Delta. Although students across the country are all under the same centralised education system, it has been suggested that the living tempo and local norms are different from one region to another. Therefore, a selection of students from these

______Chapter 3 Page 68 four regions can provide a reasonably typical representation of socio-cultural differences in Vietnam.

After a phone invitation was communicated with seven major English language education universities in these four regions, four of them (one from each region) agreed to voluntarily participate in the study. A lecturer coordinator in each university was arranged to support the process. Communication between the researcher and each of the four coordinators was established to make sure that the coordinators understood every question in the questionnaire and the administration process. They then invited all English-related major students in their universities, hereafter called EFL students, and distributed the questionnaire to them. All participants were informed that their participation in the study was voluntary and it would not result in any consequences in relation to their study. The questionnaires were then completed in class by the participants. The coordinators were available in class to provide all necessary explanations to the participants. After that, the questionnaires were collected and returned to the researcher.

Once cases associated with invalid responses were removed, the data screening process showed that there were 562 valid responses from the participants (71 males and 491 females). The distribution of the sample was 26.7 percent, 41.1 percent, 14.8 percent, and 17.4 percent in the North, Central, South and Mekong Delta regions, respectively. The participants ranged from nineteen to twenty-five years of age and followed a similar curriculum framework prescribed by the Vietnam national education authorities. All participants were in the first, second, or third year of their candidature and were primarily studying language skills courses. The study excluded the fourth year students, whose courses are significantly different from one another due to their specific majors.

3.3. Instrument development

The questionnaire instrument for Study One was developed to collect data on the local students’ perceptions of learner autonomy and some general variables defined in the study design. Therefore, it consisted of two parts. The first part was to collect information on gender, their living place, and self-reported level of computer proficiency. The second part included 62 statement items, starting with “Students who succeed best with learning English” and followed by verb phrases. Each verb phrase referred to a behaviour related to learner

______Chapter 3 Page 69 autonomy, developed from the overview of the dimensions on this construct in the contemporary literature, for example, “use time effectively” and “want to communicate with foreigners in English.” The participants were asked to indicate their degree of agreement with these statements, using a five-point Likert scale, ranging from never or almost never true (one point), occasionally true (two points), sometimes true (three points), to usually true (four points) and always or almost always true (five points). In addition, three open entry items were added to the end of the questionnaire to ask the participants to include other behaviours related to learner autonomy that they thought appropriate. The development process for these 62 statements was as follows.

The questionnaire items were initially adapted from the Learner Autonomy Inventory, the most comprehensive instrument available in the literature. It was developed by Yang (2007) to identify the dimensions of learner autonomy in the traditional learning context in Japan. It consists of 56 items, collaboratively generated by a cohort of fifteen postgraduate students and twice tested with 197 and 396 EFL intermediate students. After examining these 56 items, the current project identified five pairs of items whose meanings were almost similar. Therefore, each pair was combined, resulting in 51 items. These items were then selected on the basis of their relevance to the three core processes of learner autonomy as proposed in Chapter Two. For example, “try to find as many ways as they can to improve their English, study things which were not from their class, and check to make sure that they understood the lesson” refer to the initiating, monitoring, and evaluating learning processes respectively.

All of the 51 items were then modified and reworded to be suitable for the EFL learning context of Vietnam. Two EFL lecturers in Vietnam and an Asian PhD student working on a similar topic were asked to categorize the 51 statements into the three processes if possible. Each statement could be allocated in more than one process or even no process, depending on its notional relationship with the process interpreted by the evaluators. The results indicated that their ratings associated every item in the list with at least one of the processes, confirming the validity of the instrument. In addition, several items were rated as associated with two processes, reflecting the susceptible nature of the construct. For example, “want to communicate with foreigners in English” can be associated with both initiating and evaluating processes; or “check to make sure that they understood the lesson” can refer to both monitoring and evaluating processes. Table 3.1 (page 71) is an extraction of the groups of

______Chapter 3 Page 70 statements in the three processes, based on the three evaluators’ ratings (see Appendix 3A for a complete list of statements).

Table 3.1: Sample of the 51-item list categorized in three core processes of learner autonomy

(1) Initiating (2) Monitoring Statement (3) Evaluating (1) (2) (3) x look for opportunities to use English as much as possible x try to find as many ways as they can to improve their English x find information about English by themselves x make their schedule so they will have enough time to study English x dream of being good English speakers x want to be good English learners x x practice English with people outside class x x want to communicate with foreigners in English x x learn things that the teachers do not give as a task x carry out the learning plans once they have been made x x write down their feelings towards English studies in a language learning diary x x notice their mistakes and use that information to help them do better x x check to make sure that they understood the lesson x check their English proficiency by taking TOEIC, TOEFL or IELTS voluntarily x think about their progress in learning English

Based on these 51 statements, 11 items indicating the behaviours of learner autonomy in an ICT-supported learning environment were generated. These 11 items were the replications of those in the 51-item inventory. For example, “learn things that the teachers do not give as a task” and “check their English proficiency by taking TOEIC, TOEFL or IELTS voluntarily” in the traditional learning environment were replicated as “use the Internet to learn things which are not taught in class” and “check their English level by comparing it with the skills of others on the Internet” respectively. Appropriate attention was paid to the process of replicating items regarding their meanings and distribution across the three core processes.

Table 3.2: Statements related to processes in an ICT-supported environment

Process Statement Initiating look for different resources on the Internet use the Internet to learn things which are not taught in class will use English to communicate if seeing a stranger on the Internet Monitoring try to do some online activities even with limited time share their feelings towards English studies with friends online pay more attention when they see an English website want to improve English by taking part in online communities such as forums, blogs, chat rooms go online as a way of learning English like to study with computers Evaluating check their English level by comparing it with the skills of others on the Internet will do a search on the Internet if they have a question about English

______Chapter 3 Page 71 Table 3.2 (page 71) arranges these 11 items into three processes. Altogether, the final questionnaire was made up of 62 items which were then randomly ordered. Thirteen of these (about 20 percent of the questionnaire) were randomly selected to be in negative form; and three open questions were added to the end to collect additional autonomous learning behaviours suggested by participants (see Appendix 3B for the whole questionnaire).

3.4. Procedures

After two parts of the questionnaire were completed, it was translated into Vietnamese to facilitate the responding processes of the participants as they were all native speakers of Vietnamese. Back-translation was then employed. The Vietnamese version was sent to two university lecturers in Vietnam and a doctoral student in Australia. They were all Vietnamese and their work was associated with lecturing EFL at university level, either currently or formerly. They were asked to translate the Vietnamese version of the survey back into English. All the differences between the original English version and the three translated English versions were carefully examined in order to produce a final version in Vietnamese. As a result of the back-translation procedure, several changes were made to both versions in terms of word choice.

The final Vietnamese version was then piloted with two twelfth-graders, two first-year EFL students, three second-year EFL students, and a university graduate (three males and five females) in Vietnam. These Vietnamese subjects were not among those selected for the main sample of the study. The pilot trials showed that these subjects did not have any problem understanding and responding to the questionnaire. Therefore, no further amendment was made, and it was accepted as the final Vietnamese version. The questionnaire was then distributed to the participants, and data were entered into SPSS 18. Any case associated with missing values was removed; and negatively worded items were re-coded positively before analysis.

3.5. Results

After a procedure of data screening and processing, statistical tests were conducted to answer the first two research questions of the project. A preliminary factor analysis identified 41 items whose factor loadings in each respective factor were worth considering. Further

______Chapter 3 Page 72 analyses identified four dimensions of learner autonomy perceived by the sample. They are (1) Monitoring learning processes, (2) Goal-setting and evaluating learning, (3) Using ICTs in learning, and (4) Initiating learning opportunities. Then, the correlation between these dimensions was addressed for similarities and differences in students’ learner autonomy perceptions, presenting more insightful understandings of the construct. After that, the influence of gender, living place, and self-rated level of computer proficiency on these dimensions was analysed. Finally, the questionnaire was then revised to reflect the four identified dimensions of learner autonomy. It was comprised of 25 items which were later used in Study Two. All of the details related to the analysis procedure of this data set are presented in Chapter Four.

4. Study Two

After the data collected in Study One were processed, Study Two was conducted with a cohort of undergraduate students at University of Social Sciences and Humanities, Ho Chi Minh City. The participants worked with both traditional face-to-face learning environment and ICT-supported space, that being, a learning management system (LMS), during an EFL course. Three data sets were collected to answer the three latter research questions of the project. The first was quantitative data generated from a revised version of the questionnaire; the second was log records generated from the course LMS; and the third was qualitative data generated from semi-structured interviews at the end of the course. This section provides an overview of Study Two procedures, including objectives, participants, instruments, course design, and data collection processes.

4.1. Objectives and hypotheses

Study Two aimed to investigate the relationship between Vietnamese EFL undergraduate students’ perceptions of learner autonomy and their learning behaviours during a semester. The study also attempted to understand the factors that foster and inhibit the performance of this capacity in both online and offline learning spaces in the local context. It was hypothesized that the perception of learner autonomy and learning behaviours were not always positively correlated in every dimension due to the effect of certain mediating factors. For instance, all students might think that Using ICTs in learning was good, but not all of them could or wanted to exercise this dimension because of variations in their level of ICT

______Chapter 3 Page 73 competence; the infrastructure availability, and the paper-based mode of the exam. It was also hypothesized that such socio-cultural and momentary constraints in the local situation might have had a significant impact on students’ learning behaviours at the beginning and the end of the course. In other words, students’ development patterns of learner autonomy are supposed to reflect situational factors.

As the study focused on the relationship between different constructs and insightful understandings of a phenomenon, it suggested a suitable combination of both quantitative and qualitative research enquiry. While a quantitative design could generate numeric data and establish correlations; data collected from the qualitative perspective could enable fuller insights and better interpretations of a complex phenomenon in a local context (Creswell, 2009; Dornyei, 2007; Punch, 2009). How local students perceived learner autonomy was potentially reflected in their learning behaviours; therefore, a quantitative questionnaire instrument resulting from Study One was used to explain this association. In addition, Study Two used a set of qualitative data generated from semi-structured interviews to gain a better understanding of the factors that mediated students’ learning processes in their historical, cultural, social, and experiential learning environments (Guba & Lincoln, 2005). Such a data collection method also brought results which were specific to the study context and allowed for emerging themes. Another method of generating data for the study was the LMS logs. This data source provided general patterns of students’ learning engagement in the virtual class and partly supported the interpretations of the other two data sets.

4.2. Participants

Participating in Study Two was a cohort of 247 undergraduate students at University of Social Sciences and Humanities, Ho Chi Minh City. All had passed the national standardized entrance exam to enrol in the program as was the case for any other student in the public university system. As reported in Dang and Robertson (2010a), most of these students were eighteen to nineteen years old and in the second semester of their Bachelor program. They were taking an EFL Listening – Speaking course, one of the English proficiency courses that they were required to successfully complete in the first four semesters before pursuing a particular area of specialization such as English teaching, translation and interpretation or literature and culture. The cohort comprised five classes, taught by three teachers. The first two teachers worked with two classes each; and the other teacher worked with one class. The

______Chapter 3 Page 74 teachers were in their mid-twenties and all were female. Neither the teachers nor the students had used any online learning space as part of their school work prior to this investigation.

As the course started, the first two teachers made LMS participation compulsory in their classes. Students’ participation in the virtual class was counted as part of the course evaluation. Meanwhile, the third teacher made the LMS component optional for her students. They could take part in the virtual class, but it was not included in the course evaluation. During the course, all students were invited to answer two questionnaires, and 147 valid responses were collected for analysis. Towards the end of the sixteen-week course, students were invited to participate in an individual semi-structured interview. Eighteen responded to the email invitation, but none were from the fifth class. The first fourteen were selected; but ultimately, only eleven interviews (two from the first class and three from each of the other three classes) were successfully completed and included in the analysis.

4.3. Course design

The EFL upper-intermediate Listening – Speaking course in which this study was done was designed to improve students’ English oral skills in particular and their overall English skills in general. The philosophical assumption of the curriculum design was that all macro skills are closely related to one another. The development of one skill significantly depends on the support of the others. The macro skills are taught separately in each course and simultaneously across courses to conform to the framework provided by the national authorities and to demonstrate the teaching of integrated skills. The structure and assessment method of the Listening – Speaking course is presented in the next section.

4.3.1. Course structure

The upper-intermediate Listening – Speaking course class met once a week over a period of 16 weeks. The duration of each meeting was four hours. Mosaic 1 and 2: Listening-Speaking (Hanreddy & Whalley, 2002) were used as textbooks. The teachers sometimes provided supplementary materials for the class. As there were 45-55 students in each class, most of the teaching practices were teacher-dominated, and teachers had to use a sound system in the classroom to be heard by students. Although collaborative activities such as group work and pair work were not the primary mode of class interactions, they were sometimes used by the

______Chapter 3 Page 75 teachers. Some entertainment-oriented activities, such as listening to English songs and watching movies, were also conducted apart from the regular lessons. Native speakers of English were invited to have informal talks with students once or twice during the semester, depending on each teacher’s network of contacts.

4.3.2. Course assessment method

The course assessment consisted of three parts, namely a mid-term examination (30 percent), a final examination (40 percent), and class participation (30 percent). The two examinations included an oral and written component. The oral component aimed to evaluate the students’ speaking skills and was marked by two examiners. The written component aimed to evaluate the students’ listening skills and was marked by one examiner. The average of these two components constituted the final score of each examination. Class participation included engagement in face-to-face class meetings and LMS space. The face-to-face class participation included engagement in class activities and group presentation. The presentation required students to form groups, with each group being responsible for developing an assigned topic into a presentation. The presentation was done during class time and evaluated by the class teacher.

Before the course started, discussions took place among the faculty deans, the three participating teachers, and the webmaster. Subsequent to the meeting, the teachers were encouraged to integrate the LMS activities into their course structure. They could choose to include these virtual activities in the course assessment. When the course began, the first two teachers who taught the first four classes used LMS participation as a compulsory component for their course. This component was weighted 15 percent in the course evaluation (see Table 3.3 for the evaluation weightings). However, the teacher who taught the fifth class made the

Table 3.3: Course assessment weighting load

Assessment component Class 1 & 2 Class 3 & 4 Class 5 Mid-term exam 30% 30% 30% Final exam 40% 40% 40% Face to face class participation 15% 15% 30% Virtual class participation 15% 15% 0%

LMS component optional and virtual class participation was not evaluated in any way. It was observed that the first two teachers carefully designed the LMS virtual space for their classes

______Chapter 3 Page 76 and made the requirements clear to students, while the third teacher just left her online class structure as a default. The login records showed that she did not go to her virtual class after the course started. These virtual LMS classes are presented in detail in the following section.

4.3.3. LMS design

Although the LMS environment was formally provided to the participants for the first time, many were expected to have been familiar with this virtual space. It was observed that most of them had maintained some kind of online profile and engaged in different social networks and/or special interest groups (Dang & Robertson, 2010a). Therefore, they often associated the online space with purposes other than learning. In order to properly use this space in the local learning context, the study developed a strategy which was informed by both empirical research and contextual factors. Hence, this section starts with a description of the theoretical foundation for constructing the LMS and how it fits into the situation being investigated. It then finishes with a summary of the design procedure and site structure.

4.3.3.1. Theoretical framework of the LMS design

Study Two attempted to construct a theoretical framework for the online learning space that could match with the theoretical position of learner autonomy employed in the current study. As a result, the seven good pedagogical principles for undergraduate education suggested by Chickering and Gamson (1987) were used. These principles are empirically considered to be one of the most appropriate guidelines for understanding the quality of online learning environments (Bailey & Card, 2009; Brew, 2008; Morris & Finnegan, 2008-2009; Palloff & Pratt, 2005; Young, Cantrell, & Shaw, 2001). These principles, as expanded in Dang and Robertson (2010c), are:

1. Interactions between students and teachers, in and out of classes, need facilitating. 2. Team learning results in better outcomes than solo learning. 3. Structured exercises, challenging discussions, team projects, and peer critiques can enhance learning engagement. 4. Learning development needs proper and timely feedback. 5. Time management is a critical skill for both students and professionals. 6. Higher expectations need to be negotiated.

______Chapter 3 Page 77 7. Diversifying course delivery is necessary, and different talents are expected. (Chickering & Gamson, 1987)

With the achievements of internet technology and Web 2.0 which was employed in the current study, it would appear to indicate that technically, these seven principles could be satisfied in the LMS platform. Classroom time and space are no longer restricted in the virtual world when internet access can be comfortably achieved via computers and various handheld devices. Web 2.0 also supports the process of building knowledge from large groups of people (Surowiecki, 2004). The interactive and collaborative dimensions of both synchronous and asynchronous tasks can be quickly enabled, thanks to the availability of various platform infrastructures.

Pedagogically, these principles are aligned with those embedded in student-centred approaches to teaching, communicative language teaching, learner-autonomy promoting methods, and the knowledge construction theory of constructivism; all of which play a role in the area of second and foreign language education. They underpin reflective and collaborative interactions among teachers and students both inside and outside the class to develop learner autonomy capacity and communicative competence. The virtual space gives students an opportunity to stand in the centre of their learning processes, manage their communicative activities in real situations, and control their learning processes. Such an opportunity is sympathetic, too, to the communicative language teaching approach (Canale, 1983; Canale & Swain, 1980) which was part of the educational philosophy employed by the teachers in the study context. In addition, structured tasks and appropriate feedback have been empirically researched and identified to be important for language learning and acquisition (e.g., Bitchener, Young, & Cameron, 2005; Guardado & Shi, 2007; Skehan, 2003).

Contextually, the principles of Chickering and Gamson (1987) are also aligned with the contemporary situation of teaching and learning foreign languages in Vietnam. Unlike traditional class spaces, where teachers are often prevented from giving students opportunities to actively participate in their learning activities due to big size classes (Pham, 1999), the virtual learning environment is not restricted. This means that students have time and space for collaborative activities, feedback, and free discussion. This is particularly important because, as indicated earlier, it has been argued that Vietnamese students value these opportunities in the learning processes (Wong, 2004). It is also important to note that these

______Chapter 3 Page 78 issues have been acknowledged in the local context for a long time, but they have not been effectively resolved.

Therefore, it is appropriate to apply these principles effectively in each local learning context when investigating students’ behaviours. Morris and Finnegan (2008-2009) argue that teachers need to be actively involved in the program at the beginning to trigger students’ engagement and identify suitable individual assistance. Students should also be encouraged to bring their prior online experience to their academic life by working collaboratively with others to organize, share, modify and publish the content on which they are working (Lomicka & Lord, 2009a). As working cooperatively is always associated with negotiations for knowledge construction, the factors contributing to students’ online negotiation processes become crucial for achievement and learner autonomy development.

4.3.3.2. Design procedure and structure of the LMS

Given the theoretical framework for the online learning space and the contextual characteristics, Study Two used an LMS built around a Moodle site to create a collaborative and friendly learning environment in which students could learn (Dang & Robertson, 2010c). It served as a virtual extension of the face to face classroom to give students more opportunities to produce better quality communication for learning purposes (Beauvois, 1992). It was also used as a connection, linking the in-class and out-of-class learning activities together, assisting students to control their learning activities (Dang & Robertson, 2010b). The site was hosted in the local city to facilitate loading speed and was trialled for six months prior to the study for technical quarantine.

The LMS design generally aligned with the seven principles of best practice (as reported in Dang & Robertson, 2010b, 2010c). Structured activities, opportunities for feedback, and space for reflection, negotiation, and collaboration were provided. It used various forms of the online discussion tool as the primary mode for students’ interactions because it has been argued that this tool supports the constructivist approach to learning (Malikowski, Thompson, & Theis, 2006) and is preferred by students (Deepwell & Malikb, 2008). Other tools such as synchronous chat and collaborative Wiki writing were also introduced. In addition, the teachers were made aware of different methods to foster students’ engagement in the learning

______Chapter 3 Page 79 process such as encouraging students to critically consider others’ perspectives and the use of their own experiences to facilitate interpretation (Merriam, Caffarella, & Baumgartner, 2007).

The LMS was kept fairly clean with clear signposts to ease navigation (Dang & Robertson, 2010a). Players for the British Broadcasting Corporation (BBC) and Voice of America (VOA) radio channels were integrated and placed on the home page. A randomly selected clip box from YouTube was also placed on the right of the home page. These add-ons were to provide updated resources for listening practice. Furthermore, the default blog module on the Moodle was replaced by the Open University blog for Moodle, providing students with more flexible configurations, such as comment adding and visibility setting for each blog entry. A utility for posting voice messages with the support of NanoGong technology was also integrated into the website. This tool allowed participating students to record their oral message, save it in an audio file, replay it and embed it in their posting easily with a few clicks. This was done to support speaking practice.

The site was password-protected and structured into three main sections: a personal section, a course section, and a public section (Dang & Robertson, 2010a, 2010b). The first enabled

Figure 3.2: A snapshot of the LMS interface

______Chapter 3 Page 80 each student to update several personal details such as nickname, favourites, instant messenger username, and blog URL. The second section was made accessible to students of each respective class only. It featured a number of activities designed and facilitated by the class teachers; and students’ learning performance in this place contributed to their course assessment. The public section consisted of a notice board, a technical support forum, a general discussion forum, a chat room, and a global blog (see Figure 3.2 for a snapshot of the LMS interface, page 80). In addition, every site member could always track the most recent login record of the others with a simple click. It was expected that enabling this feature for students would be a culturally appropriate way to encourage their participation.

5. Instruments

In the context of the previously mentioned participants and course design, Study Two included three instruments to collect data to answer the three latter research questions of the current project. The first was a questionnaire, designed to investigate the participants’ perceptions of learner autonomy and their autonomous learning behaviours at the beginning and the end of the course. The second was a set of log files, automatically generated from the LMS, providing students’ engagement records in the LMS. The third was a set of guided questions used in the semi-structured interviews to extract students’ autonomous learning behaviours in detail and factors that mediated their performance process. Described in the following section is the development and rationale of the three instruments.

5.1. Measuring students’ perceptions of learner autonomy and autonomous learning behaviours

As the first two research questions of this Study Two investigated the relationship between learner autonomy perceptions and learning behaviours over a semester, it was necessary to understand students’ perceptions of learner autonomy and their autonomous learning behaviours at the beginning and at the end of the course. The way these were measured are described in the sections that follow.

Students’ perceptions of learner autonomy and their autonomous learning behaviours were measured by two questionnaires. The first was administered at the beginning of the course to measure students’ perceptions of learner autonomy and their autonomous learning behaviours.

______Chapter 3 Page 81 The second questionnaire was administered at the end of the course to measure students’ autonomous learning behaviours. As a result, one set of measurements on students’ perceptions of learner autonomy and two sets of measurements on students’ autonomous learning behaviours were generated. The former dataset was paired with each of the two latter datasets to establish the associational relationship between perceptions of learner autonomy and autonomous learning behaviours in the study context. Examining this relationship at the beginning of the course and at the end of the course would provide insights into the changes in this relationship over a semester. These changes could also be used to explain the development patterns of autonomous learning behaviours which were generated by pairing the two latter datasets.

The first questionnaire, called the pre-test, consisted of three parts, collecting data on students’ general details, perceptions of learner autonomy, and autonomous learning behaviours. The general section included information on gender, age, and self-reported level of computer proficiency. The second part of the questionnaire was used to measure the participants’ perceptions of learner autonomy. It included 41 items, derived from the preliminary analysis of Study One. However, only 25 valid items related to the four identified dimensions of learner autonomy, were included in the analysis. The other 16 items from Study One were retained, but treated as dummy items. Table 3.4 presents some of the 25 items from the questionnaire according to their respective dimensions. All of these items started with “Students who succeed best with learning English” and were answered using a five-point Likert scale as in Study One.

Table 3.4: Extracted items from the pre-test questionnaire in respective dimensions

Dimension Some extracted items monitoring learning processes use time effectively (9 items) make good use of materials & resources when studying E at home know the method which suits them best and use it goal-setting and evaluating want to be good English learners learning (7 items) think about their progress in learning English are aware of their studies using ICTs in learning like to study with computers (4 items) go online as a way of learning English will do a search on the Internet if they have a question about E initiating learning opportunities try to find as many ways as they can to improve their English (5 items) want to find a job where only English is used in the future look for opportunities to use English as much as possible

______Chapter 3 Page 82 The third part of the pre-test questionnaire was to measure the participants’ level of autonomous learning behaviours or their performance of learner autonomy. It consisted of the 41 items (16 of them were treated as dummy items) addressing the four dimensions as in the second part. All of these items started with “I” instead of “Students who succeed best with learning English” to elicit information on what the participants actually do, not what they think as in the second part. Accordingly, the five points on the Likert scale were modified to range from never or almost never true for me (one point), occasionally true for me (two points), and sometimes true for me (three points), to usually true for me (four points) and always or almost always true for me (five points). The pronouns in each item, if any, were also modified to match with the subject “I”. Three open-ended entries were also added to the end of the second and third part to collect additional information.

Distracting devices were then inserted between the last two parts of the pre-test questionnaire to minimize confounded responses. Because the items in these two parts are almost identical, the participants’ responses to one part could provide a reference to the other. After reading about successful learners’ behaviours, some participants may want to report themselves as a successful learner; while others may prefer an opposite image, depending on their personalities. This was very likely to happen in Vietnamese culture. Therefore, distracting devices were used to enable participants to clear the contents of and their responses to Part two from their short-term memory before working with Part three. This means their responses in Part three would be minimally mediated by those in Part two. The distracting devices included ten items organized into two sections, requiring participants to perform some mathematical calculations and answer some general questions such as the estimated population of Vietnam and the most crowded city in the world. The complete pre-test and post-test questionnaires are presented in Appendices 3C and 3D, respectively. As the questionnaire in Study One has more items than those in Study Two, the number associated with each item is different, and this is presented in Appendix 3E.

The second questionnaire, called the post-test, was a copy of the third part of the pre-test questionnaire. It consisted of the same 41 items (16 of them are dummy items), measuring participants’ level of autonomous learning behaviours regarding the four dimensions of learner autonomy by the end of the course. Pairing the level of learner autonomy perception with the level of learner autonomy performance in the pre-test could reveal the relationship between perception and performance of learner autonomy at the beginning of the course.

______Chapter 3 Page 83 Similarly, pairing the level of learner autonomy perception with the level of learner autonomy performance in the post-test could reveal the relationship between perception and performance of learner autonomy at the end of the course. Any differences in these results could inform the change of the relationship between students’ understanding of learner autonomy and their autonomous learning behaviours during a 16-week course in the local context. Their development pattern of learner autonomy was therefore also presented through their performance of learner autonomy reported in the pre-test and post-test.

5.2. Investigating students’ LMS engagement

In order to answer the third research question in Study Two, it was necessary to gather data on the possible factors that mediated students’ performance of learner autonomy. Hence, the study needed to be informed of students’ autonomous learning behaviours in both online and offline learning environments in advance. Such prior understandings of students’ behaviours would contribute to the design of the instrument which could later be used to elicit mediating factors. The following paragraph describes the instrument employed to understand students’ participation in their virtual classes.

Students’ LMS engagement was documented by the log records automatically generated by the LMS. Such records could provide technically accurate data on students’ LMS engagement. They could also be used to describe the level of students’ participation over a certain period of time. The records employed in the investigation included login error statistics, global forum details, and course activity logs. The login error files provided the exact date, time and IP address associated with each error. Other statistics reported the total number of postings in each section of the LMS with specific times. These log records contributed to the description of students’ technical problems and degree of engagement when working with the online learning environment. The students’ virtual learning profile was also supported by relevant communication between participants and the webmaster via email and instant messenger.

5.3. Investigating mediating factors on students’ learner autonomy

The section which follows continues the description of the instrument used to collect data for the third research question of the study. In order to gain insights into students’ performance of

______Chapter 3 Page 84 learner autonomy and possible mediating factors, the data collection process of the study was designed to be able to accommodate the widest possible range of responses from the participants. An individual semi-structured interview (Bogdan & Biklen, 2007) was the tool selected to give students opportunities to report their learning behaviours and judgments in both online and offline contexts according to a set of guided questions. Interview questions were kept fairly open to give students enough space for expression while taking into consideration the local socio-cultural insights such as relationship maintenance, a positive atmosphere, and suitable motivating stimuli during the interviews (Dang & Robertson, 2010a). Closed-ended questions were often used first to direct the conversation and culturally build up the participants’ confidence and motivation around the topic. In addition, virtual observation conducted during the course added to the information collected. This supporting data source was then used to provide general background information and inform the interview management process, thus increasing the reliability of the data collected.

The interview questions for students to reflect on, in the semi-structured interview, focused on the four identified dimensions of learner autonomy in Study One with particular consideration to those reported by Martinez (2008) and Smith and Erdogan (2008). For example, to address the Monitoring learning processes dimension, the study used questions such as “Do you often stick to a learning plan? Do you often make modifications to your original plans such as changing the deadline of a task or replacing an activity with another? How often do you do that? Why do (not) you do that? What does that mean to your learning? Can you give some examples?” Similarly, to address the dimension of Using ICTs in learning, employed questions were “Do you often use the LMS? Is there anything that you are particularly interested in? Do you have a plan for studying with the LMS? Do you use it to keep tracks of what you have done or remind you of your work? Does the communication from that environment make you spend more time on this course than the others? What are the specific examples that you can give?” (All of these questions are included in Appendix 3F.)

Given the instruments just described, the study then developed a set of procedures to use during the investigation for the study purposes. The following section presents these procedures in detail and how they were conducted to align with the research questions.

______Chapter 3 Page 85 6. Procedures

The procedures of the three instruments employed in Study Two are separately described to maintain the clarity of the presentation. The first is about how the LMS was introduced and log records were generated; the second is about how the questionnaires were administered; and the third is about how the semi-structured interviews were conducted. Before the course started, the virtual class layout was designed to provide a learning environment for students later to engage in during the course. In the first week of the course, the LMS was introduced; and the pre-test questionnaire was administered. After the LMS demonstration in week two, the LMS activities started. When the course finished in week 16, the LMS activities also finished; and the post-test questionnaire was administered. In the next three weeks, the individual semi-structured interviews were conducted. All of these events are briefly summarized in Table 3.5 and described in detail in the following sections.

Table 3.5: Timeframe of the main study procedure

Time line Activity Prior to the course Teachers working on the virtual class design Week 1 Course started Course description, LMS introduction, students’ LMS account generation Pre-test questionnaire Week 2 Demonstration of LMS in class Week 3 LMS activities started Week 16 Course finished, LMS activities finished Post-test questionnaire Week 17-19 Semi-structured interviews

6.1. LMS introduction and log generation

To collect data to understand how students responded to the online learning space, the LMS was introduced to the cohort. Prior to the course commencement, three teachers of the upper intermediate Listening – Speaking cohort worked with the webmaster to decide on the course layout for their classes. The LMS features were technically and pedagogically introduced to the teachers; and a space for follow-up discussions was created in the LMS. The teachers had the opportunity to try all the LMS functions, practice designing their own virtual course space, and choose what they thought appropriate for their course. After three weeks of critical analyses on the different technical and pedagogical aspects of the LMS, trialling practices, and the course objectives, the third teacher divided her virtual class outline into two sections, namely an administration section and a discussion forum.

______Chapter 3 Page 86 In contrast, the first two teachers worked on their virtual class design and finally agreed on a six-category layout. These categories were Course administration, Sharing learning experience, Improving listening skills, Improving speaking skills, Group presentations and Short test and quizzes. Each category consisted of subcategories, depending on each teacher. As one teacher was in charge of two classes, the LMS designs of the classes taught by the same teacher looked identical, but they were slightly different across the teachers. As presented in Figure 3.3, Classes 1 and 2 had Audio files sharing in Improving listening skills and Presentation submissions in Group presentation. Meanwhile, the teacher of Classes 3 and 4 attempted to make Quiz 1 in Short tests and quizzes, but there was nothing in that category in Classes 1 and 2. Although Quiz 1 was not finally made available to students, it was worth noting.

Class 1 & 2 (Teacher 1) Class 3 & 4 (Teacher 2)

Figure 3.3: The virtual course layouts of different classes

When the course commenced, the course description, including the LMS and assessment, was introduced to all students. Their email addresses were collected in the first class meeting for LMS account generation. At the end of the first week, they received an email, indicating their own account information and general details about how to access, navigate the site and seek support. In the second week, a demonstration tutorial about how the website functioned was conducted in each class by the webmaster. This tutorial session was then followed by a

______Chapter 3 Page 87 question and answer section. Technical support was provided to both students and teachers during the course via email and instant messenger to minimize technical barriers. The LMS activities began in the third week and finished by course end. After this, log records were generated and collected for analysis.

6.2. Questionnaire administration

As the pre-test and post-test questionnaires were developed to measure students’ perceptions of learner autonomy and their autonomous learning behaviours over the course, the questionnaires were distributed to the participants during their first and last class meeting respectively. The questionnaires were in Vietnamese and students were required to complete them page by page without referring back to the previous one. They were also told that their responses to the questionnaires would not have any effect on their study. They were also assisted when completing the questionnaires with necessary explanations and support as required. International students whose native language was not Vietnamese were not invited to participate in the study. It took the participants around 35-40 and 15-20 minutes to complete the two questionnaires, respectively. Only responses from those who completed both questionnaires were included in the analysis.

6.3. Interview protocol

The interview questions which were designed to understand the mediating factors of students’ performance of learner autonomy were translated into Vietnamese. When the course was about to finish, a student from the cohort was invited to pilot the interview process. This interview was not included in the analysis. After interview skills were reinforced through the pilot, data collection commenced. At the beginning of every interview session, it was again made clear to the participants that interview participation was voluntary and it would not be used for any purpose other than the current study. At the interview, participants were given time to express themselves. They were also encouraged to talk about any particular events that were of interest to them, and to provide examples to illustrate their retrospective descriptions. In addition, they were advised that they could use any English words or phrases convenient for them to express their opinions during the interviews, although the language used in the interviews was their mother tongue, Vietnamese (Dang & Robertson, 2010a).

______Chapter 3 Page 88 After pseudonyms were applied, the interview data were transcribed and translated into acceptable English for analysis. An overall investigation of the transcripts was conducted to understand the data as a whole. A code list was developed and revised during the coding process. The personal background and experience of the interviewer was used to support the interpretation (Corbin & Strauss, 2008). General themes emerged from the data along with the participants’ exercise of the four learner autonomy dimensions. Any details that did not conform to any of the identified patterns were also taken into account. In addition, the factors that were indicated to mediate the participants’ performance of learner autonomy, such as assessments, learning habits, and LMS access were categorized and discussed. All the interview results were then integrated with other data sets for discussion.

The instrument development processes and procedures just described were undertaken to generate the most appropriate data for the current project. Consideration was given to the issues of validity and reliability throughout the study design to maximize the confidence level of the results. The following section addresses these concerns.

7. Issues of validity and reliability

In addition to the appropriateness of the mixed-methods design employed, different procedures of data collection and analyses were used to establish reliability and validity for the current research project. While the research design was primarily driven by the project objectives, the procedures of reliability and validity employed were to ensure that every aspect of the design was properly implemented. Although the issues of reliability and validity were interwoven, the following section attempts to describe them separately to maintain the clarity of the discussion.

7.1. Reliability

The research project used exploratory factor analysis and Cronbach’s alpha as methods to increase the reliability of the questionnaire. The former was to identify clusters of variables statistically, called factors. The variables that were more correlated with those in one group and less correlated with those in the other groups should be grouped together to constitute a construct (Cohen, et al., 2007; Field, 2005). The latter was to ensure that the internal consistency of a scale was maintained. A high level of Cronbach’s alpha, the most frequently

______Chapter 3 Page 89 measure of internal consistency in questionnaire research (Brown, 2001), would indicate a reliable instrument of the measured construct. In the current project, factor analysis was used to extract prominent patterns of learner autonomy perceptions in Study One, the latter was to check reliability of these patterns in Study Two.

In Study One, the items in the questionnaire were both adapted from prior research and newly developed. It included 62 items, reflecting different dimensions of learner autonomy. Therefore, exploratory factor analysis was an appropriate reliability test to arrange these items into groups, reflecting only the dimensions perceived by the study sample. Small dimensions or dimensions with less than three items were not taken into consideration. In addition, because the factor loading of an item provides the relative contribution of that item to its respective factor, items with weak loadings in their respective dimensions were also removed from the analysis. Once the prominent dimensions were formed, the Cronbach’s alpha of each dimension was calculated. The alpha if item deleted test was conducted to increase the internal consistency level of each scale and the relatively equal contribution of each item to a scale.

In Study Two, the internal consistency of each dimension scale was again checked when the questionnaire was used for the new sample. The alpha if item deleted test was applied to remove any item which was unlikely to measure the same construct with other items in a scale (Dornyei, 2007). Different from many other studies in which the instrument reliability was only tested once, this study used the same questionnaire items to collect three different data sets. This allowed for items on a scale to be considered three times for retention or removal of an item in a scale by examining its contribution to the scale across three datasets. Therefore, the reliability of the instrument was increased, contributing to a higher level of confidence in the study results. All of these processes are specifically described in Chapter Four where the study results are reported.

7.2 Validity

The current research project employed different theoretical and statistical evidence drawn from various sources to establish its validity (Dornyei, 2007). These sources were used to facilitate the data interpretation within the study context (Bachman, 2004) and draw meaningful conclusions (Creswell, 2009). They enabled the study results to be built up from

______Chapter 3 Page 90 previous research and further the discussion of learner autonomy in both a local and global context. This section particularly addresses the issues of internal, external and content validity of the current study. The first concerns the explanation of learner autonomy being demonstrated by the data collected in the study. The second concerns the extent of the generalisation of the study outcomes to another sample or population. The third concerns the coverage of the measurement instruments regarding the concept of learner autonomy (Cohen, et al., 2007). The issue of triangulation is also considered to provide additional scaffolds for the study validity.

The primary issue to consider in the study context is the validity of using an instrument developed for use in another context. As the questionnaire was first developed by Yang (2007) in the context of Japan, it probably reflects certain aspects of Japanese culture. Similarly, most other research discussed in this study was conducted in places other than Vietnam. Therefore, when the questionnaire was adapted for the current study context, it was necessary to develop a variety of measures that would ensure its context validity. In other words, the questionnaire needed to be sequentially tested more than once to better extract the study participants’ perspective. The interview process designed for Study Two also took into account significant consideration of cultural and situational issues.

As a result, the study developed a multi-phase design to elicit the best possible data for the problems being investigated. To understand local students’ perceptions of learner autonomy, the project started with an exploratory perspective, asking students about a range of learner autonomy behaviours. Each student had opportunities to describe an image of an autonomous learner. Therefore, the pattern of learner autonomy identified in Study One was defined by the local students. This pattern of learner autonomy was then used to measure students’ performance of learner autonomy and develop the interview questions for Study Two. Those attributes which were not associated with this pattern were removed from the measurement. The study only measured the learning behaviours which were consistently perceived as being associated with learner autonomy by the sample. In other words, the data collected could be confidently used to explain the phenomenon of learner autonomy in the local context.

The whole research project involved an appropriate sample size for generalisability to other samples of the population (Cohen, et al., 2007). In Study One, the sampling process enabled participants from different socio-economic backgrounds to include their perspectives. The

______Chapter 3 Page 91 patterns of learner autonomy generated from Study One were then verified three times in Study Two with a different sample from the population. In addition, expert validation was applied at each stage of the project. In Study One, language experts and piloting were employed before the questionnaire was administered. During the process of identifying patterns of learner autonomy, content expert validation was used to examine each attribute of these patterns. Therefore, the final results generated from the sequential design were considered semantically consistent with the measured construct and representative of the population’s views.

In addition, the current project employed several language devices in the questionnaires to minimize problems with response sets or robotic responses. For example, the questionnaire items were arranged randomly instead of being clustered in groups. Both negative and positive forms were used across the questionnaires to help the study to easily identify invalid responses. Dummy items were inserted randomly in each part of the questionnaires to prevent the participants from thinking that all of the questions were the same and hence tending to give identical responses to these questions. Distracting devices were also inserted between similar parts to enable the participants to refresh their minds and stay away from response sets which they might have automatically created based on working on the previous part. These measures were designed to increase the number of carefully considered and honest responses, contributing to better quality responses of students to the following section.

Apart from the aforementioned validity measures, the current project employed three different types of triangulation to obtain a fuller explanation of the richness and complexity of learner autonomy in the study context (Cohen, et al., 2007). These included theoretical triangulation, time triangulation, and methodological triangulation (Brown, 2001; Denzin, 1970). The first was implemented through the integration of different theoretical perspectives on learner autonomy into the study instead of relying on a singular viewpoint. Given the multifaceted characteristics of learner autonomy, such an approach is vital for investigations in the field. The second type of triangulation was expressed through the use of the same measurement at different stages. The study was designed to observe students’ autonomous learning behaviours at the beginning and the end of the course. Their participation in the LMS was also recorded and captured during the course. This method allowed the study to take into account the effects of situational factors and better understand students’ learning process. The third type of triangulation was manifested through the use of different data sources of students’ learning

______Chapter 3 Page 92 engagement which were generated from different methods to understand the construct being researched. All the information gathered from the survey questionnaires, LMS records, semi- structured interviews, and informal communication with students and teachers were combined and synthesized to produce a better profile of learner autonomy in the study context.

In short, the current research project employed a number of reliability and validity measures to better address the research questions proposed and to enhance the confidence in the data analyses. These measures were then integrated into the analysis of the study results and are presented in Chapter Four where all the study data are reported. In addition, the process of implementing these measures took into account relevant issues relating to ethics. Such considerations which are described in the next section could reserve the participants’ right to take part in learning activities during the course of the study, minimizing the number of distorting responses to the study instruments.

8. Ethical considerations

This research project was approved by the La Trobe University Ethics Committee, Australia; and ethical principles of research were properly followed. It is necessary to note that ethics has not been part of the research culture in Vietnam; and a proper Vietnamese term for ethics has not been developed. Ethics can be literally equivalent to morals in Vietnamese; but moral in Vietnamese generally refers to a set of criteria that people should follow as a default. It is not a procedure for approval; and there is no Ethics Committee in education research in Vietnam. Most of the research conducted in this setting is based on mutual trust and respect between the researchers and participants. Teaching is considered the noblest occupation in Vietnamese society. It includes teaching both knowledge and moral lessons to students; therefore, it is unlikely that a teacher would do anything harmful to the students. It is part of the cultural values rather than legislative rules.

In the context of Vietnamese education, if a teacher agrees for someone to conduct a piece of research with his/her students, these students will generally participate in it without hesitation. They will think that it is safe and necessary to do so because their teacher has already approved it. Such a practice partly explains the teaching philosophy and learning behaviours in the contemporary Vietnamese situation. Quite often, teachers unintentionally impose a certain learning method on students and students accept it as the best one for them. They are

______Chapter 3 Page 93 not always guided to make their own judgments in relation to a problem. Therefore, they do not receive many opportunities to activate and exercise their learner autonomy capacity. Of course, deciding to take part in a study is different from agreeing to employ a learning strategy, but the cultural relationship between students and teachers does mediate students’ learning behaviours.

The ethical practice of educational research in the local context also affected the students’ reaction to the consent form of the current research project. As most of the students had never heard about consent form in relation to participating in educational research, they were surprised when receiving the consent form. In Vietnamese, consent refers to obligations other than rights; it is often supposed to require someone to follow or obey some procedure. Signing a consent form is attached to doing some duty. However, the study consent form was to protect students’ rights rather than force them to do something. Therefore, a detailed explanation was always given to the participants and related parties when it came to the issues of ethics.

In Study One, all communications relating to the data collection process were given in Vietnamese to ensure the comprehension of those involved. All of the participants understood that their participation in the study was voluntary; and that it would not have any effect on their teaching or learning. They could withdraw their participation from the study, and they knew how to do this. They also understood that their personal information would not be revealed in any way, and their responses to the questionnaire were only used for the study. In addition, they were told that the questionnaire was to understand what they thought about learning English. As Vietnamese students are very concerned about being right or wrong, it was explained to them that there was no right or wrong answer for any question. As three of the four coordinators, who supported administering the questionnaire, had conducted research in Vietnam for their postgraduate studies in Australia, they understood these procedures well. The other coordinator was given a more detailed explanation. Most of these communications were done through different channels such as telephone, instant messenger, and email.

In Study Two, after some communication via telephone and communication, the Dean and Vice Dean of the Faculty of English Language and Literature, University of Social Sciences and Humanities, Ho Chi Minh City, agreed to the study being conducted in the faculty. It is necessary to note that the Vice Dean, an alumnus of an Australian university, understood the

______Chapter 3 Page 94 differences between Australia and Vietnam regarding ethical issues in education research. Therefore, he facilitated and processed this procedure properly and quickly. After that, the three teachers of the Upper Listening – Speaking course were invited to a meeting; and the Vice Dean explained the study to them. As it involved the LMS component and probably some modification in the assessment criteria, the three teachers were asked to think carefully about participating in the study. They were also told that their participation was totally voluntarily; and it would not have any consequence on the evaluation by the faculty on their teaching performance. After these three teachers reached an understanding of the study and agreed to participate in it, the course schedule was given to the study researcher. The researcher was then invited to the classes to administer the questionnaires.

Similar to the procedure implemented in Study One, the student participants in Study Two had the study explained to them and were invited to participate in it during their first class meeting. The pre-test questionnaire was then distributed to them and responses were collected. The post-test questionnaire was also administered at the end of the course in a similar procedure but with less explanation because the students already knew about the study. The consent form was obtained from all of the teachers and students involved. Two weeks before the course finished, students were invited to take part in a semi-structured interview through email and LMS. At the beginning of each interview session, the student participants were told that the interview would be treated confidentially and that their responses would not be revealed to any third party, including their teachers. They could also withdraw their participation from the study by email or telephone.

In brief, given the differences of ethical practices in education research between Australia and Vietnam, the study gave more consideration to this issue than it would normally have received in Vietnam. Ethical principles of research were properly complied with. All the Vietnamese participants and involved parties were informed about the ethics procedure of the current research project. They were told in every situation that their participation was confidential and that pseudonyms would always be used instead of their real names. These processes were approved by La Trobe University Human Ethics Committee before the commencement of the data collection for the investigation (see Appendix 3G for more details).

The mixed methods design set up in the current research investigation allowed for the collection of a variety of data to be able to respond to the five research questions of the

______Chapter 3 Page 95 current project. The data collection procedures, reliability and validity measure, and ethical considerations were all designed to capture a more insightful and richer description of the investigated construct in the context of Vietnam higher education. The sequence of the five research questions and procedure structure presented in this chapter are carried out to the result and analysis section, Chapter Four, to increase the clarity of the report.

9. Summary

This chapter has described the research design and methods used in Study One and Study Two, which were conducted as part of the current research project. It presented the objectives, participant recruitment, instrument development and data collection procedure of each study. In Study One, there were 562 EFL undergraduate students from four universities across Vietnam. With the 62-item questionnaire developed from the contemporary literature, four dimensions of learner autonomy that were nominated by local students were identified. The questionnaire was then revised to reflect the four dimensions and used to measure 247 EFL students’ perceptions and performances of learner autonomy during a 16-week course of upper intermediate Listening – Speaking in Study Two. At the end of the course, the semi- structured interviews were organized with eleven volunteer students to provide more insights into their learning behaviours and the factors that mediated their performance. Log records from the LMS were also used to profile the study context. All of the results generated from these designs are presented in Chapter Four.

______Chapter 3 Page 96

Chapter Four

RESULTS

______

Chapter Three described the research design and methods used to design a research study to answer the five research questions proposed in the current project. Given the theoretical position presented in Chapter Two and the project objectives, a mixed methods design, consisting of two sequential studies to collect multiple data sets, was employed. This chapter continues to report on the results of these two studies. The first half of the chapter presents the results of Study One to answer the first two research questions. These results are primarily based on the quantitative data collected from 562 EFL students. The analysis process addresses the reliability and validity of the instrument during an exploratory factor analysis procedure to identify four dimensions of learner autonomy perceptions. The second half of the chapter reports on the results of Study Two to answer the last three research questions. These results are based on the three data sets, generated from the pre-test and post-test questionnaires, LMS logs, and semi-structured interviews. The relationships between learner autonomy perceptions and learning behaviours at the beginning and at the end of the course are investigated. Finally, the factors that mediated students’ autonomous learning behaviours are drawn from a thematic synthesis of semi-structured interview transcripts across cases.

1. Results of Study One

This section presents the results of Study One based on an analysis of the questionnaire data. These results were used to answer the first two research questions of the project, that is, to identify the Vietnamese EFL undergraduate students’ perceptions of learner autonomy and the relationships between these perception patterns. The section also examines the roles of three independent variables, namely gender, living place, and self-reported level of computer proficiency in mediating these perceptions. As discussed in Chapter Three, the questionnaire consisted of two parts. The first collected information about the three independent variables; and the second collected perceptions on 62 autonomous learning behaviours using in a five-

______Chapter 4 Page 97 point Likert scale format with additional information being provided through three open ended questions at the end of the questionnaire.

The data from the first section of the questionnaire returned valid responses from 562 participants from four universities scattered geographically across Vietnam. The descriptive statistics showed that there were more females (87.4 percent) than males (12.6 percent) in the sample, indicating the female dominated nature of the EFL area in the local context. Regionally, there were more participants from the Central (231) and North Regions (150) than from the Mekong Delta (98) and the South (83). Table 4.1 presents the distribution of these two general variables in detail. Although the imbalanced distribution of gender and living place in the sample was not ideal for the study, it reflected the contemporary realities and difficulties in data collection.

Table 4.1: Distribution of gender and living place in the sample

Gender Living Place Total Female Male North (University A) 138 12 150 (26.7%) Central (University B) 206 25 231 (41.1%) South (University C) 67 16 83 (14.8%) Mekong (University D) 80 18 98 (17.4%) 491 71 Total 562 (100%) (87.4%) (12.6%)

The participants’ self-rating level of computer proficiency was generally average or good (50.4 percent and 35.9 percent, respectively); with only a small number of participants who thought that their level of computer skills was bad, very bad, and excellent (accounting for 8.5 percent, 1.2 percent, and 3.9 percent, respectively). These numbers generally matched with the local situation where most of undergraduate students were confident with their computer skills.

The data from the second section of the questionnaire showed that the participants’ ratings for each of the 62 autonomous learning behaviours in the questionnaire used the full range of possible responses, from “never or almost never true” to “always or almost always true,” represented by values 1 and 5, respectively. The behaviour mean ranged from 2.56 to 4.64 (2.56 ≤ M ≤ 4.64); and the standard deviation ranged from .88 to 1.39 (.88 ≤ SD ≤ 1.39). As partly extracted in Table 4.2 (page 99) and fully presented in Appendix 4A, 59 of the 61 behaviours investigated were skewed within the range between -2 and +2. Only three

______Chapter 4 Page 98 remaining behaviours had a skew level beyond this range. These were items 45, 55, and 58. Despite some items being skewed, the data indicated that a normal distribution for the total sample of items was reasonably achieved. Therefore, all of these items were included in the exploratory factor analysis. It is also necessary to note that items 45, 55, and 58 were all negatively skewed, meaning that these behaviours tended to be more highly rated by the participants.

Table 4.2: An extract from the descriptive statistics of the 62 questionnaire items

No Question items in brief Min* Max* Mean SD Skew 1 study English voluntarily 1 5 4.18 .929 -1.204 2 give themselves a reward or treat when they do well in E 1 5 2.73 1.088 .320

45 want to be good English learners 1 5 4.64 .912 -2.857 47 do things actively 1 5 4.25 1.076 -1.348

54 try to complete the things they have decided to do 1 5 3.64 1.056 -.506 55 think English is important for their future 1 5 4.62 .956 -2.640 56 have plans about how to learn English 1 5 4.41 1.076 -1.852 57 check their E profi by taking TOEFL or IELTS voluntarily 1 5 3.23 1.394 -.323 58 are aware of their studies 1 5 4.52 1.005 -2.208

61 find information about English by themselves 1 5 4.36 1.063 -1.706 62 know the method which suits them best and use it 1 5 3.99 1.067 -.938 * Min and Max values rated by the participants (1 = never or almost never true; 5 = always or almost always true) Shading: items with a skew value out of the range between 2 and -2

The three open entries at the end of the questionnaire, which were items 63, 64, and 65 (see Appendix 3B), did not attract many responses; and most of them were simply repetitions of some items in the questionnaire. These responses did not clearly indicate any newly emergent autonomous learning behaviour. Therefore, the data generated from these open entries were not included in the analysis. As a result, the data processing procedures described the information drawn from the questions about the 62 autonomous learning behaviours, gender, living place, and computer proficiency.

Given the reasonably normal distribution of the data on the 62 behaviour items, a factor analysis was conducted to identify the principal dimensions of learner autonomy perceptions. The measures of reliability and validity were taken into account to maximize the confidence level of the results generated. Subsequently, correlations between these dimensions were tested to understand how strong these dimensions are correlated. Finally, correlations between each perception dimension with the three general factors were examined to understand the

______Chapter 4 Page 99 roles of these independent variables on learner autonomy perceptions, using multivariate ANOVA. All of these processes are described in the sections which follow.

1.1. Perceptions of learner autonomy

The analysis procedure described in this section was designed to answer the first research question, identifying the participants’ perception dimensions of learner autonomy. Therefore, an exploratory factor analysis was conducted with the data collected from the 62 questionnaire items to extract possible clusters of these items. As presented in Table 4.3, the strong partial correlations (Kaiser-Meyer-Olkin measure = .921) and statistically significant correlations (Bartlett’s test p<.01) among the 62 items suggested the possibility that the data could be factored (Coakes, Steed, & Ong, 2009). A scree plot test with Eigenvalues greater than one was used to get an estimate of the number of factors for the sample. According to Cohen et al. (2007), the number of factors was to be selected depends upon the number of eigenvalues on the curve to the left of the scree plot which could either include or exclude the break point. Thus, Figure 4.1 suggests the possibility of either three or four factors for the data sample.

Table 4.3: KMO and Bartlett’s Test of the sample

KMO and Bartlett's Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy .921 Bartlett's Test of Sphericity Approx. Chi-Square 12718.769 df 1891 Sig. .000

Figure 4.1: Scree plot test with Eigenvalues greater than 1

______Chapter 4 Page 100 Principal Component Analysis, the most popular extraction method that allows extracted factors to account for the maximum possible amount of variance (Gorsuch, 1983), was used to extract these four factors for examination. Four factors were extracted, accounting for 20.394, 8.281, 5.148, and 3.084 percent of the total variance respectively, a total of 36.907 percent of the total variance explained (see Table 4.4 for an extract). Factor four accounted for 3.084 percent of the total variance explained. Hence, the four factor solution provided another 3.084 percent to the total variance explained by the three factor solution. Although such an amount of variance was small for the whole sample, it was not small for the total amount of 36.907 percent variance explained by four factors. The contribution of the fourth factor was relatively equivalent to the amount of variance explained by the third factor (3.084 percent vs. 5.148 percent, respectively). Therefore, to maintain the exploratory nature of the current factor analysis procedure, all the four factors were taken into account. However, this decision would be considered again at a later stage for inclusion in further analyses.

Table 4.4: An extract of the total variance explained

Total Variance Explained Initial Eigenvalues Extraction Sums of Squared Loadings Component Total % of Variance Cumulative % Total % of Variance Cumulative % 1 12.645 20.394 20.394 12.645 20.394 20.394

2 5.134 8.281 28.675 5.134 8.281 28.675 3 3.192 5.148 33.823 3.192 5.148 33.823 4 1.912 3.084 36.907 1.912 3.084 36.907

dimension 5 1.876 3.025 39.932

62 .242 .390 100.000 Extraction Method: Principal Component Analysis.

In order to adopt an appropriate rotation method for the current factor analysis, it was necessary to understand the relationship among these factors. If these factors were essentially unrelated to one another, the data would be orthogonal and Varimax rotation should be applied. If these factors were essentially related, the data would be oblique and Direct Oblimin rotation should be applied. In the current study, as the dimensions of learner autonomy were assumed to be correlated with one another, the Direct Oblimin rotation method was selected.

As a result, principal component analysis with Direct Oblimin rotation and Kaiser Normalization was conducted to extract the four factors. To enhance the conceptual interpretation of the factors, both reliability and validity processes were taken into account.

______Chapter 4 Page 101 First, a primary analysis attempted to remove items with factor loadings smaller than .40. After that, the factor analysis was conducted again with the remaining items, and factor loadings of each item were examined until all items with a factor loading smaller than .40 were removed. Second, a cutoff point of factor loadings was assigned for each factor to produce factors with distinct concepts (Hardy & Reynolds, 2009). Third, expert validation was employed to remove items whose meanings were not essentially related to the majority of items in each respective factor. Finally, the internal consistency of items in each factor was examined to maximize the consistency level of each factor (Cohen, et al., 2007).

When principal component analysis with Direct Oblimin rotation was employed in the factor analysis, the four factors extracted from the 62 items accounted for 36.907 percent of the total variance explained (see Table 1, Appendix 4B). This preliminary extraction identified 12 items with loadings smaller than .40 in their respective factors. They were items 34, 40, 14, 16, 2, 13, 11, 3, 29, 1, 8 and 18 (see Table 2, Appendix 4B). The same procedure of factor analysis was conducted again with the remaining 50 items; and the four extracted factors accounted for 41.063 percent of the total variance explained (see Table 3, Appendix 4B). The factor loadings of each item were examined again and item 53 was removed because all of its loadings were smaller than .40 (see Table 4, Appendix 4B). The factor analysis was rerun with the remaining 49 items; and the extracted four factors accounted for 41.487 percent of the total variance explained (see Table 5, Appendix 4B). No more items were removed at this stage because they all obtained at least a factor loading of greater than .40 (see Table 6, Appendix 4B).

An examination of the factor loadings of the 49 items presented in Table 6 of Appendix 4B showed that there were 21, 12, 11, and 5 items in factor one, two, three, and four respectively. As the study was interested in discrete factors, a cutoff point was applied for the factor loadings of each factor. Only items with factor loadings as follows were selected for inclusion in their respective factors: ≥.585 (factor 1), ≥.560 (factor 2), ≥.498 (factor 3), and ≥.446 (factor 4). As a result of this, 17 items were removed and 32 items were retained (see Table 7, Appendix 4B). These 32 items continued to be further analysed. The same factor analysis procedure was again conducted and the four extracted factors accounted for 47.543 percent of the total variance explained (see Table 8, Appendix 4B). Factors one and two consisted of 10 items each; factor three consisted of 7 items; and factor four consisted of five items (see Table 9, Appendix 4B).

______Chapter 4 Page 102 To ensure that the meaning of every item in a factor referred to some similar construct, an expert validation process was employed. Two professors of education examined the meaning of each item in its respective factor. One item in factor one, three items in factor two, and three items in factor three were eliminated because their meanings were essentially different from other items in their respective factors. A total of 7 items (items 30, 56, 47, 61, 28, 27, and 54) were removed (as indicated in Table 10, Appendix 4B). Hence, 25 items remained for inclusion in the four factors.

These 25 remaining items continued to be factored for the last time. Four factors were extracted and accounted for 26.794, 9.588, 8.405, and 5.420 percent of the total variance respectively, a total of 50.207 percent of the total variance explained (see Table 11, Appendix 4B). The factor loadings are partly presented in Table 4.5 and fully presented in Table 12 of Appendix 4B. The four factors or dimensions of learner autonomy perception were named,

Table 4.5: An extract of factor analysis of the 25 items on learner autonomy perceptions

Structure Matrix Component Cumulative % Students who succeed best with learning English 1 2 3 4 of Variance Factor 1: Monitoring learning processes 26.794 37: use time effectively .797 32: make schedule so they’ll have engh time to study E .726 49: study things which were not from their class .725 62: know the method which suits them best and use it .712 48: try to study English regularly even with limited time .684

Factor 4: Initiating learning opportunities 50.207 4: want to communicate with foreigners in E -.714 9: try to find as many ways as they can to improve their E -.678 12: want to study in an E-speaking envi if having a chance -.643 7: look for opportunities to use E as much as possible -.597 41: want to find a job where only E is used in the future -.524 Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalization. Factor loadings sorted by size. Other weaker loadings of each item were removed for clarity. and expert validation was again employed. The naming process suggested that factor names should be shaped from the accumulative meanings of the items in each respective factor and able to demonstrate the distinction of one factor to another. As a result, the items of the first factor referred to behaviours of carrying out learning plans and making learning activities effective and efficient. Therefore, it was called Monitoring learning processes. The second factor consisted of items associated with students’ awareness of their EFL learning and

______Chapter 4 Page 103 judgment of their learning behaviours. Hence, it was named Goal-setting and evaluating learning. The third factor consisted of items which were associated with the use of technology for learning purposes. Therefore, it was named Using ICTs in learning. Finally, the items in the fourth factor shared a common behaviour of looking for learning opportunities in different ways. It was named Initiating learning opportunities.

As a result of the naming process, the four factors were named Monitoring learning processes; Goal-setting and evaluating learning; Using ICTs in learning; and Initiating learning opportunities respectively in order. The factor analysis procedure was conducted in multiple phases to be able to extract the four discrete dimensions of learner autonomy. They all account for more than five percent of the total variance explained, indicating their significant dominance in the perceptions of learner autonomy in the local context. More discussion on the concept of learner autonomy resulting from this procedure is presented in Chapter Five.

To examine the internal consistency reliability of the instrument, the internal consistency of each dimension was tested. As summarized in Table 4.6, the Cronbach’s alpha values for the four dimensions were .874, .788, .659, and .683 respectively. Each item was then examined

Table 4.6: Internal consistency of the four factor scales

No. of Cronbach’s Dimensions of learner autonomy Items Alpha D1: Monitoring learning processes 9 .874 D2: Goal-setting and evaluating learning 7 .788 D3: Using ICTs in learning 4 .659 D4: Initiating learning opportunities 5 .683 using the item deleted method, and this indicated that these scales produced the highest possible alphas for their respective factors (see Appendix 4C). This means they would not be any higher if any item in their respective dimensions was removed. The internal consistency of the first two dimensions was highly reliable and reliable respectively, allowing them to be retained for further analyses (Cohen, et al., 2007). Based on the argument of Bynner and Stribley (1979), the internal consistency of the fourth dimension was also acceptable because it was greater than .670. The internal consistency of the third dimension was a little lower than the expected level of .670. However, provided with only four items in a newly developed dimension, an internal consistency level of .659 was acceptable. Hence, all of these four

______Chapter 4 Page 104 dimensions were worth considering and accepted as the four important dimensions of learner autonomy perceptions identified in the study.

As presented in Table 4.6 (page 104), the number of items in the four dimensions was 9, 7, 4, and 5 respectively. This nature of exploratory factor analysis suggested that the first two dimensions accounted for a greater portion of the total variance explained. They were the more dominant dimensions of learner autonomy as perceived by Vietnamese EFL students. Having more items also contributed to a higher level of internal consistency for the first two dimensions. In contrast, the internal consistency of the third dimension was the lowest because only 4 items comprised this factor. This result suggested that there might be various ways of using ICTs for learning purposes for which the study instrument could not adequately account. Several instrument items replicated for this dimension might also be perceived as other dimensions. For example, try to do some online activities even with limited time (item 16) was loaded almost equally on Dimension 2 and 3 (see Table 2, Appendix 4B) and finally removed. Similarly, this happened with items 3 and 11.

Such an overlap in the way learner autonomy was perceived prompted further investigations on the relationship among the four identified dimensions of learner autonomy perceptions. This is also in line with the assumption of the dimension relationship adopted for the factor rotation solution, the Direct Oblimin. Therefore, a Pearson-product moment correlation test was conducted between the four dimensions, and as expected, a statistically significant correlation was found between every pair of the four factors (.178 ≤ r ≤ .582, p ≤ .01, 2-tailed, see Table 4.7). Five of the six correlations among these dimensions were at a moderate level,

Table 4.7: Positive correlation coefficient among the four dimensions

Correlations Dimension of D2. Goal-setting & D3. Using ICTs D4. Initiating learning learner autonomy perception evaluating learning in learning opportunities D1. Monitoring Pearson Correlation .380** .327** .537** learning processes Sig. (2-tailed) .000 .000 .000 N 562 562 562 D2. Goal-setting & Pearson Correlation .178** .300** evaluating learning Sig. (2-tailed) .000 .000 N 562 562 D3. Using ICTs in Pearson Correlation .382** learning Sig. (2-tailed) .000 N 562 **. Correlation is significant at the 0.01 level (2-tailed).

______Chapter 4 Page 105 indicating a common ground shared by these dimensions. The correlation between Dimension 2 and 3 was weak, suggesting that using ICTs in learning was limitedly associated with goal setting and study evaluating activities. The whole factor analysis procedure explained the tremendous reduction of items from 62 to 25 items when the study was interested in extracting discrete factors.

In short, the analysis procedure described in this section identified four interrelated dimensions of learner autonomy perceptions in the context of EFL learning in Vietnamese higher education. They were arrived at using a multiphase factor analysis with appropriate integrations of reliability and validity processes. The four extracted dimensions were named respectively Monitoring learning processes (9 items), Goal-setting and evaluating learning (7 items), Using ICTs in learning (4 items), and Initiating learning opportunities (5 items). They accounted for 26.794, 9.588, 8.405, and 5.420 percent of the total variance respectively, a total of 50.207 percent of the total variance explained. The internal consistency of the four dimensions was .874, .788, .659, and .683 respectively, confirming the reliability of the instrument. The next section attempts to analyse the levels of these four dimensions to understand the similarities and differences in students’ perceptions of learner autonomy.

1.2. Similarities and differences in learner autonomy perceptions

To provide more insights into the four dimensions of learner autonomy identified from the factor analysis solution, this section focuses on the second research question of the current project. The section is structured into two subsections. The first considers similarities and differences among these four dimensions. The second considers the effect of gender, living place, and computer proficiency on each of these four dimensions.

1.2.1. Similarities and differences among the four dimensions of learner autonomy perception

The data analysis procedure presented in this subsection was utilized to answer the second research question, which is identifying the similarities and differences in students’ perceptions of learner autonomy. As learner autonomy was generally perceived in the four significant dimensions identified in the previous factor analysis procedure, an understanding of students’ ratings on these four dimensions could reveal the similarities and differences in their levels of learner autonomy perceptions. The analysis in this section also investigated the effects of

______Chapter 4 Page 106 gender, living place, and self-reported level of computer proficiency on each of the four dimensions of learner autonomy. Such an examination could provide a more insightful understanding of the relationships among these four dimensions. A descriptive analysis of the four dimensions of learner autonomy followed by further inferential analysis is given below.

As a result of the multiphase factor analysis procedure on perceptions of learner autonomy, the four dimensions were identified and computed. The scores of items within each factor were totalled and averaged for each individual subject. Presented in Table 4.8 is the descriptive statistics of these four dimensions. Each of dimension received ratings from the lowest or almost the lowest to the highest level of the Likert scale (1.00 ≤ Min≤ 1.20, Max = 5.00). The mean of the items in each of these four dimensions ranged from moderately high to

Table 4.8: Descriptive statistics on mean difference among the dimensions (perception)

Dimensions (N=562) Min Max Mean S D Variance D1: Monitoring learning 1.11 5.00 3.786 .74149 .550 D2: Goal-setting and evaluating learning 1.00 5.00 4.381 .67554 .456 D3: Using ICTs in learning 1.00 5.00 3.670 .75572 .571 D4: Initiating learning opportunities 1.20 5.00 4.183 .63252 .400

Table 4.9: Statistical mean difference between the dimensions (perception)

Paired Samples Test Paired Differences 95% Confidence Interval Sig. Std. Error t df Mean SD of the Difference (2-tailed) Mean Lower Upper Pair 1 Dimension 1 – 2 -.595 .79101 .03337 -.66030 -.52922 -17.825 561 .000 Pair 2 Dimension 1 – 3 .116 .86847 .03663 .04439 .18831 3.176 561 .002 Pair 3 Dimension 1 – 4 -.397 .66793 .02817 -.45234 -.34165 -14.090 561 .000 Pair 4 Dimension 2 – 3 .711 .91986 .03880 .63489 .78732 18.327 561 .000 Pair 5 Dimension 2 – 4 .198 .77441 .03267 .13360 .26193 6.054 561 .000 Pair 6 Dimension 3 – 4 -.513 .77832 .03283 -.57783 -.44886 -15.636 561 .000 high in comparison with an average of 3.00 (3.67 ≤ M ≤ 4.38). This preliminary observation suggested that although the four dimensions of learner autonomy were all rather highly valued by local students, their perceptions probably varied with different dimensions. To test this hypothesis, a paired sample T-test was conducted with every pair of the four dimension means, and it showed statistically significant differences between every pair of means (- 17.825≤ t ≤ 18.327, p <.01, Table 4.9). This result indicated that all the four dimensions of learner autonomy perceptions were different from one another. There was no similarity identified in students’ levels of the four dimensions. Goal-setting and evaluating learning was

______Chapter 4 Page 107 most valued (Mean= 4.381) followed by Initiating learning opportunities (Mean=4.183). Monitoring learning (Mean=3.786) and using ICTs in learning (Mean=3.670) were less appreciated.

1.2.2. Effect of gender, living place, and computer proficiency on learner autonomy perception

To provide further understanding on students’ perceptions of learner autonomy, the analysis described in this subsection focused on the effect of gender, living place, and computer proficiency on each of the four dimensions. It aimed to test the three hypotheses proposed in the study: an insignificant effect of gender on perceptions of learner autonomy, a significant effect of computer proficiency on Using ICTs in learning, and a significant effect of living place on every perceived dimension of learner autonomy. As students in different living places and of different genders might have different levels of computer proficiency, the analysis procedure described below examined the relationship among these three independent variables to inform the analysis procedure adopted to examine their effects on the four dimensions of learner autonomy.

To understand the relationship among gender, living place, and computer proficiency, the effect of gender and living place on self-reported computer proficiency was investigated using a Univariate ANOVA analysis. As presented in Table 4.10 (page 109), neither gender nor living place had statistically significant effect on students’ self-reported level of computer proficiency (F=1.900, p=.169 and F=1.894, p=.129 respectively). However, an interaction of these two variables created a statistically significant effect on students’ self-reported level of computer proficiency (F=5.598, p<.01). In other words, males and females in different living places had statistically significantly different levels of computer proficiency. For example, female students in Mekong and the North reported their computer proficiency at the highest and lowest levels among the four regions respectively (MFemaleMekong=3.56 vs. MFemaleNorth= 2.99, Table 4.11, page 109). Meanwhile, male students in the North rated their computer proficiency at the highest level in comparison to the males in other regions (MMaleNorth = 3.67, Table 4.10, page 109). Although the effect pattern of gender and living place on computer proficiency was complicated, it suggested that students’ self-reported level of computer proficiency was mediated by an interaction of gender and living place in the study sample.

______Chapter 4 Page 108 Table 4.10: Effect of gender and living place on computer proficiency

Tests of Between-Subjects Effects Dependent Variable: Self-reported computer proficiency Type III Sum Mean Source df F Sig. of Squares Square Corrected Model 27.994a 7 3.999 7.976 .000 Intercept 2659.323 1 2659.323 5304.011 .000 Gender .953 1 .953 1.900 .169 Living place 2.849 3 .950 1.894 .129 Gender * Living place 8.420 3 2.807 5.598 .001 Error 277.764 554 .501 Total 6528.000 562 Corrected Total 305.758 561 a. R Squared = .092 (Adjusted R Squared = .080)

Table 4.11: Descriptive statistics of gender and living place on computer proficiency

Descriptive Statistics Dependent Variable: Self-reported computer proficiency Gender Living place Mean SD N Female Mekong 3.56 .653 80 South 3.51 .746 67 Central 3.34 .672 206 North 2.99 .710 138 Total 3.30 .720 491 Male Mekong 3.56 .784 18 South 3.06 .854 16 Central 3.64 .860 25 North 3.67 .778 12 Total 3.49 .843 71 Total Mekong 3.56 .675 98 South 3.42 .783 83 Central 3.38 .699 231 North 3.05 .736 150 Total 3.33 .738 562

As self-reported computer proficiency was mediated by the interaction of gender and living place in the sample, the study conducted a MANOVA test (Multivariate Analysis of Variance) to understand the effect of all three variables on each of the four dimensions of learner autonomy identified in the factor analysis solution. As presented in Table 4.12 (page 110, and see Appendix 4D for more details), the interactions of gender, living place, and self- reported level of computer proficiency had no statistically significant effect on any dimension of learner autonomy (.052≤ p ≤.712). This result suggested that Vietnamese EFL students tended to perceive each dimension of learner autonomy as not related to gender, socio- economic environment, and computer using capability. Therefore, only one hypothesis on the non-significant effect of gender on learner autonomy perceptions was supported; the other two hypotheses on the significant effect of living place and self-reported computer proficiency on different dimensions of learner autonomy were rejected. This meant male and

______Chapter 4 Page 109 female students did not perceive each of the four dimensions at significantly different levels. Students in different cities with different levels of computer proficiency tended to perceive an equally important role of every dimension of learner autonomy.

Table 4.12: Effect of the three general factors on perceptions of learner autonomy

Tests of Between-Subjects Effects Type III Sum Mean Source Dependent Variable df F Sig. of Squares Square Corrected D1: Monitoring learning processes 28.142a 33 .853 1.606 .019 Model D2: Goal-setting & evaluating learning 73.752b 33 2.235 6.474 .000 D3: Using ICTs in learning 38.126c 33 1.155 2.161 .000 D4: Initiating learning opportunities 19.836d 33 .601 1.551 .028 Intercept D1: Monitoring learning processes 1011.420 1 1011.420 1905.216 .000 D2: Goal-setting & evaluating learning 1333.373 1 1333.373 3862.643 .000 D3: Using ICTs in learning 991.165 1 991.165 1854.015 .000 D4: Initiating learning opportunities 1270.172 1 1270.172 3277.750 .000 Gender D1: Monitoring learning processes .544 1 .544 1.026 .312 D2: Goal-setting & evaluating learning .556 1 .556 1.611 .205 D3: Using ICTs in learning 1.011 1 1.011 1.891 .170 D4: Initiating learning opportunities .003 1 .003 .006 .936 Living place D1: Monitoring learning processes 2.863 3 .954 1.797 .147 D2: Goal-setting & evaluating learning 26.791 3 8.930 25.870 .000 D3: Using ICTs in learning 1.032 3 .344 .643 .587 D4: Initiating learning opportunities 1.185 3 .395 1.019 .384

Gender * D1: Monitoring learning processes 2.889 4 .722 1.360 .247 Comp. D2: Goal-setting & evaluating learning 3.519 4 .880 2.548 .039 proficiency D3: Using ICTs in learning 3.638 4 .909 1.701 .148 D4: Initiating learning opportunities 3.828 4 .957 2.469 .044 a. R Squared = .091 (Adjusted R Squared = .034) d. R Squared = .088 (Adjusted R Squared = .031) b. R Squared = .288 (Adjusted R Squared = .244) Shading: significant level of smaller than .05 c. R Squared = .119 (Adjusted R Squared = .064)

Although an interaction of the three general factors did not yield a significant difference on the perceptions of learner autonomy, some significant effects were found with other interaction terms. For example, living place alone had a statistically significant difference in students’ levels of valuing Goal-setting and evaluating learning (p<.01, see Table 4.12 for an extract and Appendix 4D for more details). Similarly, an interaction of gender and computer proficiency indicated a statistically significant effect on the perceiving level of Goal-setting and evaluating learning and Initiating learning opportunities (p<.05, see Table 4.12 for an extract and Appendix 4D for more details). In other words, the effect of the three general factors of the study on learner autonomy perceptions appeared to be complex and indicated a need to continue such an investigation in further research with a sample that can eliminate the statistically significant relationship among the three factors.

______Chapter 4 Page 110 In short, all the four dimensions of learner autonomy were perceived at different levels although certain aspects of their notions were shared with one another. No similarity was found in the level of students’ perceptions across the four dimensions. Goal-setting and evaluating learning and Initiating learning opportunities were appreciated at a higher level than Monitoring learning processes and Using ICTs in learning. It could be seen that despite the availability of ICT infrastructure at home, ICT-supported learning was valued at the lowest level compared to those of the traditional learning environment because such a practice was not properly integrated into the school curriculum. The analysis also showed that students’ level of learner autonomy perceptions was not mediated by gender, socio-economic environment, and self-reported computer capability.

1.3. Summary

The data analysis procedure described in Study One answered the first two research questions of the current project. The first was about students’ perceptions of learner autonomy, and the second was about the relationship among these perception patterns. Four discrete dimensions of learner autonomy were perceived by Vietnamese EFL undergraduate students, namely Monitoring learning processes, Goal-setting and evaluating learning, Using ICTs in learning, and Initiating learning opportunities respectively. The first two dimensions were more dominant in the construct of learner autonomy perceptions. In addition, although there were notional overlaps among these dimensions, each of them was perceived at a significantly different level. The second and fourth dimensions were more highly valued than the other two. These levels of appreciation were suggested not to be associated with differences in gender, living environment, and self-reported computer proficiency; however, these relationships needed to be further researched. Therefore, the instrument was included in Study Two for further investigation.

2. Results of Study Two

This section describes the analysis procedures of the data collected in Study Two to answer the three latter research questions of the current project. Specifically, the three questions were regarding: (i) the relationship between students’ perceptions of learner autonomy and their autonomous learning behaviours; (ii) the change of this relationship over a 16-week course; and (iii) the mediating factors of learner autonomy performance. The analyses drew on the

______Chapter 4 Page 111 three data sets, namely quantitative, qualitative, and log records. The quantitative data were generated from a pre-test and post-test questionnaire, developed based on the results of Study One. Data were used to report on the relationship between perception and performance of learner autonomy. Three scales were identified, namely learner autonomy perceptions, learner autonomy performance on the pre-test, and learner autonomy performance on the post-test. The qualitative data, generated from eleven individual semi-structured interviews with students, and the log records, generated from the LMS with which the students were working during the course, were used to understand the factors that mediated students’ performance of learner autonomy in the study context.

To maintain the clarity of the data presentation, the following section was structured into six sub-sections. The first is a general description of the data collected in the study sample. The distribution of gender and self-reported computer proficiency in the sample were reported. The study did not collect data on students’ living place because they all lived in the same city where the university was located. The second addresses the reliability of the three measurement scales: perception of learner autonomy, pre-test performance of learner autonomy, and post-test performance of learner autonomy. This provided foundational credentials for the study instruments and allowed the results elicited from this procedure to be employed in further analyses. The third is a revisit to the similarities and differences in students’ level of learner autonomy perceptions as stated in the second research question of the study. It also addressed the complicating effect of gender and self-reported computer proficiency on learner autonomy perception as proposed at the end of Study One. The last three sub-sections consider the three research questions respectively.

2.1. General description of the sample

Study Two was conducted with a cohort of 247 students enrolled in the Listening – Speaking course in the Faculty of English Language and Literature, University of Social Sciences and Humanities, Ho Chi Minh City. However, many of them did not turn up to the first and last class meeting when the pre-test and post-test questionnaires were administered. Several cases with missing data were also found and had to be removed from the data set. As a result of the screening process, 147 participants (118 females and 29 males, 59.5 percent of students) were identified as having completed the required questions in both questionnaires on learner autonomy perception and performance. These participants, whose responses were included in

______Chapter 4 Page 112 the quantitative analysis, were distributed fairly equally among the five classes, ranging from 10.9 percent to 26.5 percent of the sample (see Table 4.13).

Table 4.13: Distribution of students in groups

No of No of Class Group Percent Percent Participants Participants Class 1 A 16 10.9 55 37.4 Class 2 (Teacher 1) 39 26.5 Class 3 B 38 25.9 66 44.9 Class 4 (Teacher 2) 28 19.0 C Class 5 26 17.7 26 17.7 (Teacher 3) Total 3 147 100.0 147 100.0

As the first four classes were taught by two teachers each, it was observed that students sometimes moved between the classes of the same teachers. Therefore, the classes taught by the same teacher were considered as one group, resulting in three groups. They were thus called Group A, B, and C, taught by Teacher 1, 2, and 3, respectively (see Table 4.13). These codes, A, B, and C, were used in the study analysis. Most of the students were 18 and 19 years old (80.3 percent and 13.6 percent respectively); only 9 of them (6.1 percent of the sample) were aged from 20 to 24 years old. A majority of students thought that their level of computer proficiency was average and good (63.3 percent and 21.1 percent, respectively); only a few students rated their computer skills as very bad, bad, and excellent (4.1 percent, 8.2 percent, and 3.4 percent, respectively). This pattern is relatively similar to what was found in Study One (see Table 4.14).

Table 4.14: Students’ level of self-reported computer proficiency

Self-reported Computer Proficiency Study One Study Two Frequency Percent Frequency Percent Very Bad 7 1.2 6 4.1 Bad 48 8.5 12 8.2 Average 283 50.4 93 63.3 Good 202 35.9 31 21.1 Excellent 22 3.9 5 3.4 Total 562 100.0 147 100.0

To provide guidelines for further analyses conducted in the latter part of the study, the relationship among gender, student group, and self-reported computer proficiency was examined. The analyses started with the distribution of males and females across the three

______Chapter 4 Page 113 groups. Therefore, a Chi-square test was conducted with gender and student group. As presented in Table 4.15, there was no statistically significant difference between the number of males or females in the three groups (X2 = .255, df = 2, p = .880). In other words, the ratio of females to males was similar across the three groups. The number of females ranged from 3.6 to 4.2 per male in each group (Table 4.16). The examination continued to look at the effect of gender and group in students’ self-reported level of computer proficiency. A Univariate ANOVA was conducted, and it showed that gender, group, or an interaction of the two had no statistically significant effect on students’ self-reported level of computer proficiency (1.022 ≤ F ≤ 2.281 and .106 ≤ p ≤ .314, Table 4.17). This suggested that students’ level of self-reported computer proficiency was similar between males and females across the three groups.

Table 4.15: Test of gender ratio on different groups

Chi-Square Tests Asymp. Sig. Value df (2-sided) Pearson Chi-Square .255a 2 .880 Likelihood Ratio .254 2 .881 Linear-by-Linear Association .136 1 .712 N of Valid Cases 147 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 5.13.

Table 4.16: Descriptive ratio of females to males in three groups

Crosstabulation of Group and Gender Gender Ratio Total Female Male (Female : Male) Group A 55 43 12 3.6 : 1 Group B 66 54 12 4.5 : 1 Group C 26 21 5 4.2 : 1 Total 147 118 29 4.1 : 1

Table 4.17: Test of gender and group on self-reported computer proficiency

Tests of Between-Subjects Effects Dependent variable: Self-reported level of computer proficiency Type III Sum Source df Mean Square F Sig. of Squares Corrected Model 2.989a 5 .598 1.028 .404 Intercept 792.611 1 792.611 1362.164 .000 Gender .595 1 .595 1.022 .314 Group 1.668 2 .834 1.433 .242 Gender * Group 2.654 2 1.327 2.281 .106 Error 82.045 141 .582 Total 1512.000 147 Corrected Total 85.034 146 a. R Squared = .035 (Adjusted R Squared = .001) ______Chapter 4 Page 114 Although only 147 students provided valid responses to the two questionnaires, all 247 students of the cohort participated in the LMS as a learning activity. Together with their three teachers, they made over 2000 postings on the LMS during the course. Both public and class sections attracted postings from students. Most of these postings were made in the middle of the course, from week 5 to 11. It was also observed that several students encountered technical problems as they began to work with the LMS. In addition, only about a half of students from Group C logged onto the LMS, and they only made a few postings because unlike Groups A and B, participating in the LMS was not made compulsory for students in Group C. All of the LMS records were documented to profile students’ learning behaviours in an ICT-supported environment.

In addition to the questionnaire data and log records, students’ autonomous learning experiences were documented through interviews with eleven students at the end of the course. Students were invited to talk about their personal learning behaviours related to the four dimensions of learner autonomy. They generally indicated a complicated pattern of autonomous learning behaviours. Making learning plans for a semester, setting up certain targets, and arranging strict schedules were usually implemented. However, modifying these plans and even ignoring some tasks were what they did every day. Similarly, they would like to participate in the discussion threads on the LMS at the beginning, but they ended up browsing some other websites. The reasons reported for such alterations in their autonomous learning behaviours included social constraints, personalities, school workload, and technical problems.

In brief, participating in Study Two were 247 EFL students of an upper intermediate Listening – Speaking course. They were divided into three groups, taught by three teachers. The quantitative data collected from two questionnaires consisted of valid responses from 147 students. The data showed that the ratio of male to female students was similarly maintained across three groups. The self-reported computer proficiency of both males and females in different groups was also reported to be at a similar level. The LMS attracted postings from both teachers and students, particularly in the middle of the course. The data drawn on the individual interviews with eleven students indicated that they all performed a certain level of learner autonomy throughout their course. Different factors that mediated their learning behaviours were also revealed.

______Chapter 4 Page 115 Before considering each of the three research questions separately, the following section describes a reliability process employed for the questionnaire instrument. Although the questionnaires used in Study Two resulted from Study One, they were checked again with the new sample to increase the reliability of the data.

2.2. Reliability of the questionnaire instrument

This section describes an analysis procedure used to test the reliability of the questionnaire instrument which was employed in Study Two to measure students’ levels of learner autonomy perception and performance. As presented in Section 5.1 of Chapter Three, two questionnaires were employed in the study. The first was administered at the beginning of the course to measure students’ perception and performance of learner autonomy (see Appendix 3C). The second questionnaire was administered at the end of the course to measure students’ performance of learner autonomy again (see Appendix 3D). Students’ performance of learner autonomy at the beginning and at the end of the course was called pre-test performance and post-test performance, respectively. In summary, there were three measurement scales employed, one for learner autonomy perception and two for learner autonomy performance.

Each of these three measurement scales consisted of 25 items, 16 dummy items, and three open entries. The 25 items indicated the autonomous learning behaviours associated with the four dimensions of learner autonomy identified from the factor analysis procedure presented in Study One (see Table 4.5, page 103). The three open entries at the end of each measurement scale were to provide space for students to add more autonomous learning behaviours. However, these three open entries received only a few responses from students, and most of them were similar to what was already indicated in the 25 items of their respective measurement scales. Hence, only the data collected from the 25 items across the three measurement scales were included in the study analysis, and those collected from the three open entries were excluded.

As each of the three 25-item measurement scales consisted of four dimensions, the analysis that follows examined the internal consistency of these dimensions across the three measurement scales to ensure the reliability of the data collected. The four dimensions identified from the factor analysis in Study One were Monitoring learning processes, Goal- setting and evaluating learning, Using ICTs in learning, and Initiating learning opportunities

______Chapter 4 Page 116 respectively. The number of items in these four dimensions was nine, seven, four, and five, respectively. An internal consistency reliability test was then conducted with the items in each of these four dimensions across the three measurement scales.

In order to examine if any of the items in each dimension should be removed to increase the reliability level of that dimension, an internal consistency test with Cronbach’s alpha, if an item was deleted, was employed. As shown in Table 4.18a, the alpha value of the first dimension reached the highest level in each respective measurement scale. It could not become greater if any of the nine items was removed. It was concluded that the instrument of the first dimension conducted in the study was most reliable if all of the nine items were included. With the alpha level ranging from .779 to .803 across the three measurement scales, the reliability of the first dimension was accepted to be included in further analyses.

Table 4.18a: Cronbach’s alpha of Dimension 1 in Study One and Two

Cronbach's Alpha if Item Deleted Dimension 1: Study One Study Two Monitoring learning processes (9 items) Pre-test Post-test Perception Perception Performance Performance No Cronbach's alpha of the scale .874 .779 .782 .803 23 use time effectively .852 .741 .763 .778 18 make schedule so they’ll have engh time to study E .857 .760 .776 .792 32 study things which were not from their class .860 .745 .753 .774 7 know the method which suits them best and use it .860 .758 .763 .798 31 try to study E regularly even with limited time .860 .761 .752 .787 3 make good use of materials & res when study E .866 .778 .769 .769 24 notice mistakes & use that info to hlp them do better .861 .761 .769 .780 22 reflect on what they learn and look for sth important .861 .755 .755 .791 27 check to make sure that they understood the lesson .866 .762 .753 .788

The same reliability procedure was conducted with the other three dimensions, and ultimately all items were retained as indicated in the analyses of Study One. Dimension 2 had seven items, and its alpha level was .689, .758, and .736 in the perception, pre-test performance and post-test performance measurement scales, respectively (Table 4.18b, page 118). If item 29 was removed, the alpha level of the dimension increased to .721 and .762 in the perception and pre-test performance measurement scales, respectively. However, the alpha level of the dimension in the post-test performance measurement scale decreased to .681. Similarly, if items 10 or 41 were removed, the alpha level of the dimension might increase in one or two measurement scales but decrease in others. These variations were identified to be small, and the removal of any item in Dimension 2 did not result in a better level of internal consistency

______Chapter 4 Page 117 across the three measurement scales. Therefore, the seven items in Dimension 2 with an alpha level ranging from .689 to .758 were the best possible for scale reliability and were thus retained for further analyses.

Table 4.18b: Cronbach’s alpha of Dimension 2 in Study One and Two

Cronbach's Alpha if Item Deleted Dimension 2: Study One Study Two Goal-setting and evaluating learning (7 items) Pre-test Post-test Perception Perception Performance Performance No Cronbach's alpha of the scale .788 .689 .758 .736 36 think English is important for their future .755 .669 .751 .714 39 are aware of their studies .749 .629 .687 .663 29 want to be good English learners .757 .721 .762 .681 10 try to improve their weaknesses .756 .696 .739 .733 41 practice English with people outside class .768 .643 .778 .778 12 think about their progress in learning English .766 .601 .691 .684 9 know their good points and weaknesses .778 .615 .679 .675 Shading: Values higher than the current Cronbach’s alpha of the dimension

Dimension 3 had four items, and its alpha level was .622, .725, and .626 in the perception, pre-test performance and post-test performance measurement scales, respectively (Table 4.18c). If item 30 was removed, the alpha level of the dimension increased to .786 and .668 in the pre-test and post-test performance measurement scales, respectively. However, the alpha level of the dimension in the perception measurement scale decreased to .598, a very low level. To maintain a reasonably balanced level of internal consistency for the dimension across three measurement scales, item 30 needed to be included. Apart from this variation, there was no other change that could result in an increase in the internal consistency level of the dimension. Given the small number of items for such a diversified dimension which easily overlapped with other dimensions as indicated in Study One, the dimension was maintained, and a more insightful understanding of this dimension was supported by the qualitative data collected through interviews.

Table 4.18c: Cronbach’s alpha of Dimension 3 in Study One and Two

Cronbach's Alpha if Item Deleted Dimension 3: Study One Study Two Using ICTs in learning (4 items) Pre-test Post-test Perception Perception Performance Performance No Cronbach's alpha of the scale .659 .622 .725 .626 17 like to study with computers .568 .497 .631 .468 26 go online as a way of learning English .564 .605 .589 .442 19 will do a search on internet if have a question abt E .587 .485 .571 .579 30 pay more attention when they see an E website .643 .598 .786 .668 Shading: Values higher than the current Cronbach’s alpha of the dimension

______Chapter 4 Page 118 Dimension 4 had five items, and its alpha level was .668, .686, and .763 in the perception, pre-test performance and post-test performance measurement scales, respectively (Table 4.18d). Consideration was paid to items 40, 4, and 8 because if one of them was removed, the respective alpha level of the dimension in the perception, pre-test performance and post-test performance measurement scales increased. However, the alpha level of the other two measurement scales decreased in each situation. In addition, these increases were small, while their respective decreases were big. For example, if item 8 was removed, the alpha level of the dimension in the post-test performance scale increased from .763 to only .768, but in the perception and pre-test performance measurement scales, the alpha level of the dimension decreased from .688 to .648 and from .686 to .632, respectively. In summary, the internal consistency level of this dimension was best across the three measurement scales when all five items were retained. With an alpha level of .668 and above, the reliability of the scale was ensured.

Table 4.18d: Cronbach’s alpha of Dimension 4 in Study One and Two

Cronbach's Alpha if Item Deleted Dimension 4: Study One Study Two Initiating learning opportunities (5 items) Pre-test Post-test Perception Perception Performance Performance No Cronbach's alpha of the scale .683 .668 .686 .763 40 want to communicate with foreigners in E .613 .688 .650 .730 5 try to find as many ways as they can to improve E .602 .605 .595 .685 8 want to study in an E-speaking envi .624 .648 .632 .768 4 look for oppor to use E as much as possible .620 .598 .699 .717 33 want to find a job where only E is used in future .698 .531 .602 .693 Shading: Values higher than the current Cronbach’s alpha of the dimension

In short, the procedure described to establish the reliability of the questionnaire instrument suggested that all the four dimensions adopted from the factor analysis procedure in Study One were worth using for further investigations. After an examination of the internal consistency level of the four dimensions across the three measurement scales, the 25 items loaded in the four dimensions were all retained to achieve the highest level of reliability for the instrument. With most of the Cronbach’s alpha values being around .670 and greater, the internal consistency of the four dimensions was established across the three measurement scales (Table 4.19, page 120). Although Dimension 3 achieved an alpha level of .622 and .626 in two measurements, it was also included in further analyses and further explained by the qualitative data.

______Chapter 4 Page 119 Table 4.19: Number of items and Cronbach’s alpha of the four dimension scales

Cronbach’s Alpha No of Learner autonomy dimensions Pre-test Post-test items Perception Performance Performance D1. Monitoring learning processes 9 .779 .782 .803 D2. Goal-setting & evaluating learning 7 .689 .758 .736 D3. Using ICTs in learning 4 .622 .725 .626 D4. Initiating learning opportunities 5 .668 .686 .763 Shading: Values smaller than .670

2.3. Revisiting the similarities and differences in students’ perceptions of learner autonomy

The second research question investigated the similarities and differences in the students’ level of learner autonomy perception. As indicated in the factor analysis procedure in Study One, local students identified four significant dimensions of learner autonomy. These four dimensions of learner autonomy were indicated to be interrelated (see Table 4.7, page 105), and it appeared that some common ground was shared by these dimensions. However, each dimension was perceived at a different level by Vietnamese EFL students (see Table 4.9, page 107). In addition, gender and computer proficiency were reported to have a complicating effect on these four dimensions. Therefore, the analysis procedure described in this section attempted to revisit this relationship among the four dimensions of learner autonomy perception in the context of data collected for Study Two. The effect of gender and self- reported computer proficiency across these four dimensions were also examined. This revisit was to reconfirm the responses to the second research question presented in Study One, contributing to the reliability of the research project.

To confirm that there were correlations among the four dimensions of learner autonomy perception as identified in Study One, a Pearson-product moment correlation test was conducted with the data collected in Study Two. As presented in Table 4.20 (page 121), all of these dimensions were statistically significantly correlated with one another (.182 ≤ r ≤ .522, p≤ .05, 2-tailed) except the one between Dimension 2 and 3 (p=.122). By comparison, the correlation between these two dimensions was at a weak level of .178 in Study One although it was statistically significant. This difference may be derived from the sample size difference between Study One and Study Two (n=562 vs. n=147). It could also be seen that all of the significant correlation coefficients yielded in Study Two were at a relatively similar level to their respective coefficients which were noted in Study One (Table 4.21, page 121). These results suggested that the four dimensions of learner autonomy were essentially perceived as

______Chapter 4 Page 120 being related to one another, sharing some common learning aspects. For example, students tended to think that when they set up learning goals, they should look for learning opportunities and monitor their learning activities to achieve those goals. However, it seems that they were less likely to think that ICTs should be used for setting learning goals and evaluating EFL learning or vice versa.

Table 4.20: Correlations among the four dimensions (perception)

Correlations Dimension of D2. Goal-setting & D3. Using ICTs in D4. Initiating learning learner autonomy perception evaluating learning learning opportunities D1. Monitoring Pearson Correlation .436** .284** .405** learning processes Sig. (2-tailed) .000 .000 .000 N 147 147 147 D2. Goal-setting & Pearson Correlation .128 .182* evaluating learning Sig. (2-tailed) .122 .027 N 147 147 D3. Using ICTs in Pearson Correlation .522** learning Sig. (2-tailed) .000 N 147 **. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).

Table 4.21: Correlations among the four dimensions (perception) in two studies

Correlations of Dimensions in Learner Autonomy Perceptions Study One Study Two Dimensions of learner autonomy perception N Correlation Sig. N Correlation Sig. Pair 1. Dimension 1 & Dimension 2 562 .380 .000 147 .436 .000 Pair 2. Dimension 1 & Dimension 3 562 .327 .000 147 .284 .000 Pair 3. Dimension 1 & Dimension 4 562 .537 .000 147 .405 .000 Pair 4. Dimension 2 & Dimension 3 562 .178 .000 147 .128 .122 Pair 5. Dimension 2 & Dimension 4 562 .300 .000 147 .182 .027 Pair 6. Dimension 3 & Dimension 4 562 .382 .000 147 .522 .000 Shading: significant value greater than .05 Dimension 1: Monitoring learning processes Dimension 3: Using ICTs in learning Dimension 2: Goal-setting & evaluating learning Dimension 4: Initiating learning opportunities

Given the significant correlation between almost every pair of the four dimensions of learner autonomy perception, their level of difference was tested. As the descriptive statistics indicated in Table 4.22 (page 122), the four dimensions of learner autonomy were perceived at different levels, ranging from a moderately high to a high level (3.665 ≤ M ≤ 4.710,

Maverage= 3). This prompted a mean comparison test across the four dimensions of learner autonomy. A paired-samples t-test was conducted with every pair of the four dimensions. Similar to the conclusions drawn in Study One, a statistically significant difference was found between every two dimensions of learner autonomy perception (-14.743≤ t ≤18.080, p<.01, 2-

______Chapter 4 Page 121 tailed, Table 4.23). This meant each of the four learner autonomy dimension was perceived at a different level by students in the study context. Goal-setting and evaluating learning was most valued, followed by Initiating learning opportunities and Monitoring learning processes. Using ICTs in learning was appreciated at the lowest level among the four. This result was in line with what was found in Study One.

Table 4.22: Levels of the four dimensions in learner autonomy perception

Descriptive Statistics Dimensions of learner autonomy N Min Max Mean SD D1. Monitoring learning processes 147 2.00 5.00 4.094 .53630 D2. Goal-setting & evaluating learning 147 3.00 5.00 4.710 .38996 D3. Using ICTs in learning 147 2.00 5.00 3.665 .63468 D4. Initiating learning opportunities 147 2.40 5.00 4.321 .54575

Table 4.23: Level differences in the four dimensions of learner autonomy perception

Paired Samples Test Paired Differences Dimensions of learner 95% Confidence Interval autonomy perceptions Mean SD Std. Error of the Difference Sig. Mean Lower Upper t df (2-tailed) Pair 1. Dimension 1 & 2 -.617 .50713 .04183 -.69934 -.53401 -14.743 146 .000 Pair 2. Dimension 1 & 3 .429 .70503 .05815 .31384 .54369 7.373 146 .000 Pair 3. Dimension 1 & 4 -.227 .59021 .04868 -.32357 -.13115 -4.671 146 .000 Pair 4. Dimension 2 & 3 1.045 .70105 .05782 .93116 1.15971 18.080 146 .000 Pair 5. Dimension 2 & 4 .389 .61012 .05032 .28986 .48876 7.736 146 .000 Pair 6. Dimension 3 & 4 -.656 .58209 .04801 -.75101 -.56124 -13.666 146 .000

Provided that all of the four learner autonomy dimensions were perceived at different levels, the analysis aimed to test the effect of gender and computer proficiency on these levels. As gender and self-reported computer proficiency were not correlated in this sample, they were separately tested with the four dimensions of learner autonomy perception. To understand the effect of gender on the level of learner autonomy perception, an independent samples t-test was conducted. The calculation showed that gender had a statistically significant effect only on Dimension 2 of learner autonomy perception (t=2.089, p=.044, Table 4.24, page 123). Accordingly, female students were found to have a statistically significantly higher level of

Goal-setting and evaluating learning than the male counterparts (Mfemale=4.7542 and

Mmale=4.5320, Table 1, Appendix 4E). However, in the other three dimensions, there seemed to be no gender difference.

______Chapter 4 Page 122 Table 4.24: Effect of gender on the four dimensions of learner autonomy perception

Independent Samples Test Levene's Test for Equality of t-test for Equality of Means Variances 95% Confidence Sig. Mean Std. Error Interval of the F Sig. t df (2-tailed) Differ. Differ. Difference Lower Upper D1 Equal variances assumed .447 .505 .319 145 .750 .03562 .11150 -.18475 .25599 Equal variances not assumed .291 38.688 .773 .03562 .12255 -.21232 .28355 D2 Equal variances assumed 9.990 .002 2.814 145 .006 .22222 .07897 .06613 .37831 Equal variances not assumed 2.089 33.072 .044 .22222 .10636 .00584 .43860 D3 Equal variances assumed .004 .949 .092 145 .926 .01220 .13199 -.24868 .27308 Equal variances not assumed .091 42.102 .928 .01220 .13394 -.25807 .28248 D4 Equal variances assumed 2.991 .086 1.803 145 .073 .20240 .11225 -.01946 .42425 Equal variances not assumed 1.514 36.058 .139 .20240 .13364 -.06862 .47341 D: Dimension Shading: significant value smaller than .05

In Study One, because there was a correlation between gender and computer proficiency, the effect of these two factors on the four dimensions of learner autonomy perception was investigated together. The analysis suggested that the effect of an interaction between gender and computer proficiency on Dimension 2 and 4 was statistically significant. In Study Two, because gender and computer proficiency were found to be uncorrelated, their effect on the four dimensions was investigated separately. Gender was identified to have a significant effect on Dimension 2. The results of the two studies were generally aligned with each other, and this called for an investigation on the effect of computer proficiency on the four dimensions of learner autonomy perception in Study Two. Therefore, an ANOVA test was conducted, and as presented in Table 4.25, self-reported level of computer proficiency had no

Table 4.25: Effect of computer proficiency on learner autonomy perception

ANOVA Sum of Mean df F Sig. Squares Square D1. Monitoring learning processes Between Groups .312 4 .078 .266 .900 Within Groups 41.681 142 .294 Total 41.993 146 D2. Goal-setting & evaluating learning Between Groups .649 4 .162 1.069 .374 Within Groups 21.553 142 .152 Total 22.202 146 D3. Using ICTs in learning Between Groups 2.224 4 .556 1.396 .239 Within Groups 56.588 142 .399 Total 58.812 146 D4. Initiating learning opportunities Between Groups .802 4 .200 .667 .616 Within Groups 42.683 142 .301 Total 43.485 146

______Chapter 4 Page 123 statistically significant effect on any of the four dimensions (p>.05). Students with different levels of computer skills did not tend to perceive each dimension of learner autonomy at different level.

The absence of statistically significant effects for computer proficiency on the four dimensions of learner autonomy perception raised two possible interpretations. First, students of the contemporary world might think that they know enough about computers to perform their autonomous learning behaviours in EFL learning. They did not believe that computer skills could either improve or diminish their autonomous learning behaviours such as looking for learning opportunities or setting learning goals. Second, they probably did not think that they should use computers or ICT-supported environments to perform their learner autonomy capacity. ICT devices were thought to be essentially associated with purposes other than learning. However, the second suggestion did not seem to match with the study context where ICTs were becoming part of students’ daily routine, and a lot of them were using technologies for their learning activities. Therefore, the first suggestion seemed to be logical, reflecting the common place of ICT devices and younger learners’ technology capabilities.

It seemed important to note that self-reported computer proficiency did not yield a significant effect on the learner autonomy dimension related to the use of ICTs in learning. As learning EFL with ICTs normally required a certain level of technology competence, students were expected to think that their computer skills would significantly affect their learning in ICT- supported learning environments. Prior research during early 2000s also acknowledged the roles of computer competence in students’ attitude, engagement, and achievement in such a learning space (Al-Jarf, 2007; Chen, 2003; Lin, 2004). However, students’ responses in this study indicated a new trend in which they did not see that general computer proficiency affected their autonomous learning behaviours even though these behaviours needed to be supported by some computer skills.

In short, the analyses presented in this section revisited some of the conclusions and further developed some of the suggestions made in Study One regarding the similarities and differences among the four dimensions of learner autonomy perception. Accordingly, these four dimensions were similarly confirmed to be perceived at different levels in both studies. The most recognized dimension was about setting goals and evaluating learning processes. Initiating and monitoring learning processes were rated in the second and third position,

______Chapter 4 Page 124 respectively. The dimension related to the use of ICTs for learning purposes was perceived at the lowest level. These four dimensions were also confirmed to share some notional grounds except for the independent relationship between the dimension associated with the use of ICTs for learning purposes and the one on goal setting and learning evaluation. This was in line with the conclusion that students’ computer proficiency did not affect their levels of appreciating any learner autonomy dimension. Only the dimension of setting goals and evaluating learning was perceived at different levels by males and females.

At this stage, a thorough understanding of the relationships among the four dimensions of learner autonomy perception was achieved. They were generally correlated with one another but each of them was perceived at a different level. The relationships between each dimension with gender and self-reported computer proficiency were also clarified. Males and females were only different in their perceptions of goal setting and learning evaluation, while computer proficiency did not trigger any significant effect on leaner autonomy perception. The next three sections are devoted to gradually addressing the three research questions of the project. This means they seek an understanding of the relationships between perception of and performance of learner autonomy dimensions over the course by analysing the data collected from the three measurement scales, namely perception, pre-test performance and post-test performance. They also aim to look for insights into the factors that mediated students’ autonomous learning behaviours through the interview data.

2.4. Relationship between learner autonomy perception and performance

This section considers the third research question proposed in the current project. The analyses sought to understand the relationship between the students’ understanding of learner autonomy and their autonomous learning behaviours. Using the same four dimensions of learner autonomy identified in Study One, the measurement of students’ understanding of learner autonomy, or perceptions of learner autonomy, and autonomous learning behaviours, or performance of learner autonomy were all based on these four dimensions. The perception measurement and pre-test performance were done at the beginning of the course, while the post-test performance was done at the end of the course. Pairing data collected from the perception scale with those from the two performance scales could reveal any changes in the relationship between learner autonomy perception and performance over the course.

______Chapter 4 Page 125 To inform the analysis procedure of investigating the relationships between perception and performance of learner autonomy employed in the following section, it was important to take into account the effect of student groups. This was because students from different groups might have different levels of learner autonomy perception and performance. Therefore, an ANOVA test was conducted to compare students’ levels of learner autonomy perception and performance among the three groups. As presented in Tables 4.26a and 4.26b, no statistically significant difference in the four dimensions of learner autonomy perception and performance across the three groups was identified (p>.05). This meant that students in different groups could be viewed as having similar levels on the four dimensions of learner autonomy perception and performance. Students appeared to be normally distributed across the three groups in terms of learner autonomy perception and performance.

Table 4.26a: Effect of student groups on learner autonomy perception

ANOVA Sum of Mean df F Sig. Squares Square D1. Monitoring learning processes Between Groups 1.001 2 .501 1.759 .176 Within Groups 40.991 144 .285 Total 41.993 146 D2. Goal-setting & evaluating learning Between Groups .303 2 .152 .997 .372 Within Groups 21.899 144 .152 Total 22.202 146 D3. Using ICTs in learning Between Groups 1.849 2 .925 2.337 .100 Within Groups 56.963 144 .396 Total 58.812 146 D4. Initiating learning opportunities Between Groups .610 2 .305 1.024 .362 Within Groups 42.875 144 .298 Total 43.485 146

Table 4.26b: Effect of student groups on pre-test learner autonomy performance

ANOVA Sum of Mean df F Sig. Squares Square D1. Monitoring learning processes Between Groups 1.107 2 .554 1.606 .204 Within Groups 49.657 144 .345 Total 50.764 146 D2. Goal-setting & evaluating learning Between Groups 1.473 2 .737 1.833 .164 Within Groups 57.876 144 .402 Total 59.350 146 D3. Using ICTs in learning Between Groups .092 2 .046 .067 .935 Within Groups 98.830 144 .686 Total 98.922 146 D4. Initiating learning opportunities Between Groups .849 2 .424 1.547 .216 Within Groups 39.498 144 .274 Total 40.347 146

______Chapter 4 Page 126 In addition to the students’ similar levels of learner autonomy perception and performance across the three groups, the analyses from the general description of the sample also showed that there was no statistically significant difference in the student ratio of males to females among the three groups and that their self-reported levels of computer proficiency were not different across the groups. Students were normally distributed across the three groups regarding their gender, computer proficiency, perception, and performance of learner autonomy. The effect of the student groups was independent from the relationship between learner autonomy perception and pre-test performance. Therefore, the analysis of this relationship did not take into account the group effect, and the data from the three groups were put together for analysis instead of being investigated separately. A paired samples t-test was conducted to understand the correlations and level differences between perception and pre-test performance of learner autonomy regarding the four dimensions.

As presented in Table 4.27, students’ perceptions of learner autonomy were statistically significantly correlated with their performance of learner autonomy in every respective dimension (.288 ≤ r ≤ .531, p<.01). This meant that students’ understanding of learner

Table 4.27: Correlations between perception and pre-test performance

Paired Samples Correlations N Correlation Sig. D1. Monitoring learning processes Perception & 147 .301 .000 Pre-test Performance D2. Goal-setting & evaluating learning Perception & 147 .288 .000 Pre-test Performance D3. Using ICTs in learning Perception & 147 .531 .000 Pre-test Performance D4. Initiating learning opportunities Perception & 147 .475 .000 Pre-test Performance autonomy was consistently related to their autonomous learning behaviours regarding their EFL study. In addition, the correlations between perception and performance of learner autonomy were all positive, suggesting that students tended to do what they thought important for their learner autonomy. The dimensions which were perceived as being more important were performed more often. However, the correlation coefficients between perception and performance were rather small, ranging from .288 to .531. As a square of a correlation coefficient between two variables indicated the percentage of variance shared by those two variables, the coefficient range between .288 and .531 suggested that only 8.29 percent to 28.20 percent of variance was commonly shared by learner autonomy perception and

______Chapter 4 Page 127 performance in each dimension, respectively. Such correlation levels were not very useful for group prediction (Cohen, et al., 2007).

Provided with the consistency between what students thought and what they did for their learner autonomy, the analysis focused on how they reported their levels of learning behaviours towards their expected profile of successful autonomous learners. This called for a test to compare students’ level of learner autonomy perception with their level of learner autonomy performance. As presented in Table 4.28, a statistically significant difference between perception and performance of learner autonomy was found in the first three dimensions (2.047 ≤ t ≤ 11.858, p<.05, 2-tailed) but not in the fourth one (t=.991, p=.323, 2- tailed). These suggested different patterns of the relationship between learner autonomy perception and performance.

Table 4.28: Level comparison between perception and pre-test performance

Paired Samples Test Paired Differences 95% Confidence Sig. Std. Interval of the t df (2-tailed) Mean SD Error Difference Mean Lower Upper D1. Monitoring learning processes .652 .66695 .05501 .54359 .76102 11.858 146 .000 D2. Goal-setting & evaluating learning .363 .64438 .05315 .25745 .46753 6.820 146 .000 D3. Using ICTs in learning .123 .72526 .05982 .00423 .24067 2.047 146 .042 D4. Initiating learning opportunities .045 .54937 .04531 -.04465 .13445 .991 146 .323

To better understand the patterns of relationship between perception and performance of learner autonomy, a closer examination of the means was conducted. In the pre-test, students tended to self-rate their autonomous learning levels in the first three dimensions as being significantly lower than those that they expected from autonomous learners (3.441 vs. 4.094, 4.348 vs. 4.710, and 3.543 vs. 3.665, respectively, Table 4.29, page 129). These three dimensions included Monitoring learning processes, Goal-setting and evaluating learning, and Using ICTs in learning. In contrast to the first three dimensions, students self-rated their performance level of Initiating learning opportunities as high as the level of autonomous learners. This suggested that at the beginning of the course, students believed that they did not reach the highest level of setting goals, monitoring, evaluating learning, and using ICTs for learning. However, they achieved the highest level of looking for learning opportunities.

______Chapter 4 Page 128 Table 4.29: Descriptive statistics on levels of perception and pre-test performance

Paired Samples Statistics Std. Error Mean N SD Mean D1. Monitoring learning processes Perception 4.094 147 .53630 .04423 Pre-test Performance 3.441 147 .58966 .04863 D2. Goal-setting & evaluating learning Perception 4.710 147 .38996 .03216 Pre-test Performance 4.348 147 .63758 .05259 D3. Using ICTs in learning Perception 3.665 147 .63468 .05235 Pre-test Performance 3.543 147 .82313 .06789 D4. Initiating learning opportunities Perception 4.321 147 .54575 .04501 Pre-test Performance 4.276 147 .52569 .04336

In short, the analyses in this section indicated that students’ self-rated autonomous learning behaviours were consistently informed by their understanding of this construct at the beginning of the course. If a learning behaviour was considered good and useful, it was likely to be carried out by students such as Monitoring learning processes and Using ICTs in learning. However, more broadly, the strength of the relationship between these two constructs was relatively weak, suggesting that it could be mediated by other factors. For example, they thought that their level of behaviours such as goal setting, monitoring and evaluating learning processes, and using ICTs for learning purposes was significantly lower than that of an autonomous learner in their mind. These suggestions called for further investigations of the relationship between students’ understanding of learner autonomy and their autonomous learning behaviours at the end of the course, as well as their learner autonomy achievement during the course.

2.5. Changes of the relationships between learner autonomy perception and performance over a semester

The fourth research question was related to the changes of the relationship between learner autonomy perception and performance in an EFL learning context over a 16-week course. To investigate such changes, the analyses described in this section compared this relationship at the beginning of the course with that at the end of the course. The relationship at the beginning of the course was already investigated in Section 2.4 of this chapter. It suggested that students consistently tended to do what they valued in learner autonomy. They also achieved a lower level of learner autonomy than what they expected from an autonomous learner in three of the four dimensions. Therefore, the analyses continued to look at this

______Chapter 4 Page 129 relationship at the end of the course. After this, these two results were integrated to indicate the changes of this relationship over the course.

Similar to the analysis procedure described in Section 2.4 to understand the relationship between learner autonomy perception and performance at the beginning of the course, tests of correlation and mean comparison were conducted. Accordingly, a paired samples t-test was conducted to compare students’ level of learner autonomy perception with their self-rated level of learner autonomy reported in the post-test. As presented in Table 4.30, students’ perceptions of learner autonomy were statistically significantly correlated with their performance of learner autonomy in Dimension 2, 3, and 4 (p<.01) but not in Dimension 1 (p=.121). This suggested that students tended to do what they thought important for learner autonomy in three dimensions only. Their understanding of the first dimension of learner autonomy did not inform their autonomous learning behaviours or vice versa. This result was a little different from what was reported at the beginning of the course, where it was suggested that factors other than students’ perception of learner autonomy played a more important role in informing students’ autonomous learning behaviours in Dimension 1.

Table 4.30: Correlations between perception and post-test performance

Paired Samples Correlations N Correlation Sig. D1. Monitoring learning processes Perception & 147 .129 .121 Post-test Performance D2. Goal-setting & evaluating learning Perception & 147 .345 .000 Post -test Performance D3. Using ICTs in learning Perception & 147 .478 .000 Post -test Performance D4. Initiating learning opportunities Perception & 147 .253 .002 Post -test Performance

As the relationship between learner autonomy perception and performance at the beginning of the course was relatively different from that at the end of the course, more investigations of students’ level of learner autonomy at the end of the course were called for. To do this, students’ level of learner autonomy perception was paired with their level of learner autonomy performance at the end of the course. As presented in Table 4.31 (page 131), a statistically significant difference was found between students’ level of learner autonomy perception and their level of learner autonomy performance at the end of the course in three dimensions (p<.01). No statistically significant difference resulted between these two measures of the third dimension (p=.467). This meant that students thought that their levels of

______Chapter 4 Page 130 Monitoring learning processes, Goal-setting and evaluating learning, and Initiating learning opportunities at the end of the course were still lower than those of a truly autonomous learner (Table 2, Appendix 4E). However, they reported that they achieved a similar level of Using ICTs in learning like other autonomous learners.

Table 4.31: Level comparison between perception and post-test performance

Paired Samples Test Paired Differences 95% Confidence Sig. Std. Error Interval of the t df Mean SD (2-tailed) Mean Difference Lower Upper D1. Monitoring learning processes .894 .76554 .06314 .76939 1.01897 14.162 146 .000 D2. Goal-setting & evaluating learning .364 .55660 .04591 .27370 .45516 7.938 146 .000 D3. Using ICTs in learning .043 .70733 .05834 -.07278 .15782 .729 146 .467 D4. Initiating learning opportunities .154 .69679 .05747 .04016 .26732 2.675 146 .008

In brief, students’ self-rated performance of Dimension 2, 3, and 4 at the end of the course were found to be linked with their understanding of learner autonomy (see Table 4.30, page 130). This meant that the st udents’ understanding of learner autonomy of these three dimensions was informed by their learning experience during the course. However, the strength of their relationship was low, potentially allowing for other factors to trigger effects on their learning behaviours. This partly explained the insignificant correlation between the students’ understanding of learner autonomy and their autonomous learning performance in Dimension 1 at the end of the course. In addition, students reported that they performed at a truly autonomous-like level of Dimension 3 at the end of the course, although their learning behaviours of Dimension 1, 2 and 4 were significantly lower than what they expected from a truly autonomous learner.

Pairing the relationship between learner autonomy perception and performance at the beginning of the course with that at the end of the course resulted in a few suggestions. First, students’ learning behaviours were all informed by their understanding of learner autonomy at the beginning of the course. However, the level of such associations was weak. This suggested an explanation for why students’ learning behaviours of one dimension were driven by factors other than their understanding of learner autonomy at the end of the course. Second, students generally rated their level of learner autonomy lower than what they expected from an ideal autonomous learner at the beginning of the course. This pattern was

______Chapter 4 Page 131 reported to be relatively unchanged at the end of the course. However, there were two exceptions. Students’ self-rated level of Dimension 4 was similar to that of a truly autonomous learner at the beginning of the course (Table 4.28, page 128), but was lower than that of an autonomous learner at the end of the course (Table 2, Appendix 4E). This suggested a decrease in students’ level of autonomous learning behaviours over the course. In contrast, students’ level of Dimension 3 was reported as being lower than that of an autonomous learner at the beginning of the course (Table 4.29, page 129), but it was rated as being equivalent to that of an autonomous learner at the end of the course (Table 4.31, page 131). This suggested an increase in students’ level of autonomous learning behaviours over the course.

At this stage, the analyses showed that some components in the relationship between learner autonomy perception and performance were maintained over the course, while some others were changed. These differences called for an investigation of the development pattern of students’ autonomous learning behaviours over the course. To understand the differences in students’ learner autonomy performance between the beginning and the end of the course, a pair samples t-test was conducted with the data generated from the pre-test and post-test measurements on learner autonomy performance. As presented in Table 4.32, a statistically significant difference in the level of Dimension 1 and 4 was found between the pre-test and post-test (t=4.693, p<.01 and t=2.453, p=.015 respectively). No statistically significant difference resulted in the level of Dimension 2 and 3. Students’ level of Dimension 1 and 4 at the end of the course was significantly lower than that at the beginning of the course (Mpre=

3.441 and 4.276, Mpost= 3.200 and 4.167, respectively, Table 4.33, page 133), suggesting a level decrease throughout the course. However, the level of the two other dimensions was similar between the beginning and the end of the course.

Table 4.32: Mean comparison of learner autonomy between pre-test and post-test

Paired Samples Test Paired Differences 95% Confidence Sig. Std. Error Interval of the t df Mean SD (2-tailed) Mean Difference Lower Upper D1. Monitoring learning processes .242 .62486 .05154 .14002 .34373 4.693 146 .000 D2. Goal-setting & evaluating learning .002 .55731 .04597 -.08890 .09279 .042 146 .966 D3. Using ICTs in learning -.080 .69980 .05772 -.19400 .03414 -1.385 146 .168 D4. Initiating learning opportunities .109 .53800 .04437 .02115 .19654 2.453 146 .015

______Chapter 4 Page 132 Table 4.33: Descriptive statistics on students’ levels of learner autonomy performance

Paired Samples Statistics Std. Error Mean N SD Mean D1. Monitoring learning processes Pre-test Performance 3.441 147 .58966 .04863 Post-test Performance 3.200 147 .61959 .05110 D2. Goal-setting & evaluating learning Pre-test Performance 4.348 147 .63758 .05259 Post -test Performance 4.346 147 .55402 .04569 D3. Using ICTs in learning Pre-test Performance 3.543 147 .82313 .06789 Post -test Performance 3.622 147 .73871 .06093 D4. Initiating learning opportunities Pre-test Performance 4.276 147 .52569 .04336 Post -test Performance 4.167 147 .59267 .04888

In summary, the analysis procedure described identified the different patterns of relationships between students’ understanding of learner autonomy and their learning behaviours. At the beginning of the course, students’ self-rated learning behaviours were all consistently related to their understanding of learner autonomy. However, this was not consistent at the end of the course. Therefore, further investigations of the development patterns of learner autonomy throughout the course were conducted. The analyses showed that the levels of Dimension 2 and 3 of learner autonomy performance were similar at the beginning and at the end of the course, respectively. Meanwhile, the level of two others, Dimension 1 and 4, was higher at the beginning than the end of the course. All of these findings suggested that there were factors other than the students’ understanding of learner autonomy triggering effects on their autonomous learning behaviours. This meant mediating factors of students’ learner autonomy performance in the study context needed to be investigated.

2.6. Mediating factors of learner autonomy performance

The fifth research question of the current project considered the mediating factors of learner autonomy performance. The data analysis conducted in this section aimed at gaining an understanding of important factors that mediated students’ autonomous learning behaviours in the local context. The analyses drew on both quantitative and qualitative data collected through questionnaires, LMS log records, and eleven semi-structured interviews with students. Themes considered in the following section are gender, computer proficiency, and teaching on the four dimensions of learner autonomy identified in Study One through statistical analysis. Also considered are the effects of personal, institutional, and social factors on the students’ performance of learner autonomy through students’ descriptions of their learning processes in both online and offline environments.

______Chapter 4 Page 133 2.6.1. Effect of gender, computer proficiency, and teaching on learner autonomy performance

As presented in Section 2.4 on sample description, males and females were normally distributed among the three groups of students. Students’ levels of computer proficiency were also similar between males and females across the three groups. Therefore, the mediating effect of these three variables on the four dimensions of learner autonomy performance identified in the current project needed to be examined separately. The role of gender was first considered, followed by computer proficiency and teaching effect.

The effect of gender on learner autonomy performance at the beginning and at the end of the course was investigated, using a repeated measures test. The test compared the female students’ level of learner autonomy with that of their male counterparts measured in the pre- test and post-test questionnaires. As presented in Table 4.34, the level of learner autonomy performance was statistically significantly different between males and females in Dimension 2 only (F=3.909, p=.05). In the three other dimensions, both males and females achieved similar levels of learner autonomy in both pre-test and post-test. In brief, the mediating effect of gender was only significant in students’ ability to set learning goals and evaluate their learning processes. Female students tended to achieve a higher level of goal-setting and learning evaluating than their male counterparts (MFemale=4.412 vs. MMale=4.089, and

MFemale=4.367 vs. MMale=4.261, in the pre-test and post-test, respectively; Table 4.35, page 135). Table 4.34: Effect of gender on learner autonomy performance

Tests of Between-Subjects Effects Measure: MEASURE_1 Transformed Variable: Average Type III Mean Source df F Sig. Sum of Squares Square Dimension 1: Monitoring learning processes Intercept 2053.053 1 2053.053 3801.490 .000 Gender 2.269E-5 1 2.269E-5 .000 .995 Error 78.309 145 .540 Dimension 2: Goal-setting & evaluating learning Intercept 3414.730 1 3414.730 6239.843 .000 Gender 2.139 1 2.139 3.909 .050 Error 79.351 145 .547 Dimension 3: Using ICTs in learning Intercept 2429.030 1 2429.030 2473.118 .000 Gender .428 1 .428 .436 .510 Error 142.415 145 .982 Dimension 4: Initiating learning opportunities Intercept 3247.446 1 3247.446 6782.144 .000 Gender 1.071 1 1.071 2.237 .137 Error 69.429 145 .479

______Chapter 4 Page 134 Table 4.35: Statistics on the difference between males and females in Dimension 2

Descriptive Statistics Dimension 2: Gender Mean SD N Goal-setting & evaluating learning Female 4.412 .58446 118 Pre-test Performance Male 4.089 .77818 29 Total 4.348 .63758 147 Female 4.367 .51338 118 Post-test Performance Male 4.261 .69993 29 Total 4.346 .55402 147

This effect of gender on students’ performance of learner autonomy was consistent with the effect of gender on learner autonomy perception analysed in section 2.3 of this chapter (page 120). Females and males were found to have different levels of both perception and performance in Goal-setting and evaluating learning. Females tended to value goal setting and learning evaluation at a significantly higher level than their male counterparts. Similarly, females achieved a higher level of goal setting and learning evaluation than males. Such consistency was aligned with the conclusion that students tended to do what they thought best for their learning. This is reflected in the positive correlation between learner autonomy perception and performance as indicated previously in section 2.5 of this chapter (page 129).

Driven by the effect of gender on learner autonomy performance, the effect of self-reported computer proficiency was investigated. Similar to the analysis method applied to gender, a repeated measures test was conducted to identify the differences in students’ level of learner autonomy regarding their level of self-reported computer proficiency. As presented in Table 4.36 (page 136), a statistically significant difference on the levels of Dimension 2 and 3 was found among students with different levels of self-reported computer proficiency (F=3.955, p<.01 and F=5.383, p<.01, respectively). This indicated that self-reported computer proficiency played a role in students’ level of setting learning goals, evaluating learning processes, and using ICTs for learning. Students at different levels of self-reported computer proficiency tended to perform at a similar level in Monitoring learning processes and Initiating learning opportunities but at different levels in Goal-setting and evaluating learning and Using ICTs in learning.

______Chapter 4 Page 135 Table 4.36: Effect of computer proficiency on learner autonomy performance

Tests of Between-Subjects Effects Measure: MEASURE_1 Transformed Variable: Average Type III Mean Source df F Sig. Sum of Squares Square Dimension 1: Monitoring learning processes Intercept 1127.105 1 1127.105 2080.702 .000 Gender 1.389 4 .347 .641 .634 Error 76.921 142 .542 Dimension 2: Goal-setting & evaluating learning Intercept 1769.094 1 1769.094 3426.158 .000 Gender 8.168 4 2.042 3.955 .004 Error 73.322 142 .516 Dimension 3: Using ICTs in learning Intercept 1232.272 1 1232.272 1410.746 .000 Gender 18.808 4 4.702 5.383 .000 Error 124.036 142 .873 Dimension 4: Initiating learning opportunities Intercept 1690.999 1 1690.999 3628.290 .000 Gender 4.320 4 1.080 2.317 .060 Error 66.180 142 .466

Further examination of the mean difference in students’ levels of Goal-setting and evaluating learning and Using ICTs in learning did not indicate a unidirectional pattern. The homogeneous subsets test did not provide a clear suggestion for pattern forming either. For example, regarding Goal-setting and evaluating learning in the post-test (see Table 4.37), students who rated their level of computer proficiency as extreme, that is, very bad and excellent performed better than the others (MVeryBad= 3.259 and MExcellent=3.467 vs. 3.120

≤MOthers≤ 3.258). However, this pattern was not consistent with what was indicated in the pre- test, given that the groups of students with extreme levels of self-reported computer proficiency were smaller than other groups in the study sample.

Table 4.37: Statistics on the levels of Dimension 2 by computer proficiency

Descriptive Statistics Computer Mean SD N Proficiency Very bad 3.500 .48048 6 Bad 3.157 .64172 12 Average 3.443 .58002 93 Pre-test Performance Good 3.538 .61336 31 Excellent 3.422 .62559 5 Dimension 2: Goal- Total 3.441 .58966 147 setting & evaluating Very bad 3.259 .69448 6 learning Bad 3.120 .59451 12 Average 3.172 .62822 93 Post-test Performance Good 3.258 .60205 31 Excellent 3.467 .69567 5 Total 3.200 .61959 147

______Chapter 4 Page 136 Similarly, the effect pattern of students’ computer proficiency on their level of Using ICTs in learning was not linear. At first, it was indicated that students with a higher level of self- reported computer proficiency tended to use more ICTs in their learning (MVeryBad=2.542,

MBad=3.146, MAverage=3.497, and MVeryBad= 2.958, MBad=3.521, MAverage=3.573 respectively in the pre-test and post-test, Table 3, Appendix 4E). However, this pattern was not consistent with students who rated their computer proficiency as being excellent because it was shown that these students used ICTs for their learning less than those who rated their computer proficiency as being good (MGood= 3.960 vs. MExcellent= 3.950 and MGood= 3.911 vs. MExcellent= 3.800 respectively in the pre-test and post-test, Table 3, Appendix 4E).

Related to the effect of computer proficiency on the perception of learner autonomy indicated in Section 2.3 of this chapter (page 120), it appeared that computer proficiency triggered mediating effects on two of the four dimensions of learner autonomy performance, but it did not have any effect on any dimension of learner autonomy perception. This suggested that although students with different levels of self-reported computer proficiency perceived learner autonomy at a similar level, they performed some dimensions of learner autonomy at different levels. Students with a higher level of computer proficiency tended to use more ICTs in their learning although it was not confirmed for those who thought that their computer skills were excellent. A better understanding of th ese effect patterns of computer proficiency on learner autonomy is gained from the analysis of qualitative data and LMS log records discussed in Section 2.6.2.1 of this chapter (page 140).

As students were in three teacher groups, each worked with one teacher, students in different groups could possibly have received different ways of teaching associated with each teacher’s personality and expertise. Hence, the teaching effect on students’ self-rated learner autonomy performance was investigated. Given that students’ level of learner autonomy perception and performance at the beginning of the course was similar across the three groups, students’ ratings on their self-rated levels of learner autonomy at the end of the course were used for this purpose. An ANOVA test was conducted to compare the levels of learner autonomy performed by different groups.

As presented in Table 4.38 (page 138), the teaching effect appeared to result in a statistically significant difference on students’ level of Initiating learning opportunities only (F=4.365, p=.014). After the 16-week course, students’ ability to look for learning opportunities was

______Chapter 4 Page 137 Table 4.38: Mean difference of three groups on performance in the post-test

ANOVA Sum of Mean df F Sig. Squares Square D1. Monitoring learning processes Between Groups .693 2 .346 .901 .408 Within Groups 55.355 144 .384 Total 56.048 146 D2. Goal-setting & evaluating learning Between Groups 1.061 2 .530 1.745 .178 Within Groups 43.753 144 .304 Total 44.813 146 D3. Using ICTs in learning Between Groups .359 2 .180 .326 .722 Within Groups 79.312 144 .551 Total 79.671 146 D4. Initiating learning opportunities Between Groups 2.932 2 1.466 4.365 .014 Within Groups 48.352 144 .336 Total 51.283 146

Table 4.39: Statistics on the level of performance in the post-test across three groups

Descriptive Statistics 95% Confidence Interval for Mean N Mean SD Std. Error Min Max Lower Upper Bound Bound D4. Initiating learning Group A 55 4.007 .65993 .08898 3.8289 4.1857 1.60 5.00 opportunities Group B 66 4.209 .56069 .06902 4.0713 4.3469 2.80 5.00 Group C 26 4.400 .41952 .08228 4.2306 4.5694 3.40 5.00 Total 147 4.167 .59267 .04888 4.0707 4.2640 1.60 5.00

Table 4.40: Comparison of students’ level of performance in three groups

Multiple Comparisons Tukey HSD 95% Confidence Mean Group Group Std. Interval Dependent Variable Difference Sig. (I) (J) Error Lower Upper (I-J) Bound Bound D4. Initiating learning Group A Group B -.20182 .10579 .140 -.4524 .0487 opportunities Group C -.39273* .13791 .014 -.7193 -.0661 Group B Group A .20182 .10579 .140 -.0487 .4524 Group C -.19091 .13417 .332 -.5087 .1268 Group C Group A .39273* .13791 .014 .0661 .7193 Group B .19091 .13417 .332 -.1268 .5087 *. The mean difference is significant at the 0.05 level. performed at different levels by different groups. Meanwhile, their levels of other learner autonomy dimensions were similar across the three groups. Further examination of the descriptive statistics and multiple comparisons on the mean of Initiating learning

______Chapter 4 Page 138 opportunities in three groups suggested that only the difference between Groups A and C was statistically significant (Table 4.40, page 138). Students in Group C reported achieving a higher level of looking for learning opportunities than those in Group A (MGroupC=4.4000 vs.

MGroupA=4.0073, Table 4.39, page 138).

To summarise, this section investigated the roles of gender, computer proficiency and teaching practice on learner autonomy performance. All of these three factors had certain effects on some dimensions of learner autonomy. Female students were identified as performing higher in activities associated with goal setting and learning evaluation than their male counterparts. These goal setting and evaluating activities were also reported to be mediated by students’ self-reported levels of computer proficiency, but there was no linear pattern between the two. In addition, students with a higher level of self-reported computer proficiency tended to use ICTs for learning purposes more often. However, this was not confirmed for those who thought their computer proficiency was excellent. Finally, the teachers’ teaching effect was found to be significant on students’ ability to look for learning opportunities. It seems to suggest that none of these three factors could trigger significant effects across the four dimensions of learner autonomy, confirming the complicated nature of this construct.

Given the limited impact of gender, computer proficiency and teaching effect on students’ levels of learner autonomy performance, it was necessary to identify other mediating factors of learner autonomy performance to gain more understanding of this multifaceted construct. Therefore, the data generated from LMS log records and interviews with students were analysed. Factors related to students’ EFL learning preference, motivation and attitude at both personal and contextual levels were extracted. These analyses are presented in the following section.

2.6.2. Effect of other mediating factors on learner autonomy performance

Apart from gender, computer proficiency, and teaching effect, another purpose of the study was to identify other mediating factors on students’ performance of learner autonomy. To do this, LMS log records of both teachers and students were collected. These records provided quantitative data on students’ login and posting details. In addition, semi-structured interviews were conducted at the end of the course. Four groups of questions, addressing the four

______Chapter 4 Page 139 dimensions of learner autonomy identified in the factor analysis of Study One, were employed. Eleven students were asked to reflect on their autonomous learning activities and provide justifications for those learning activities. In a pressure-free conversational atmosphere, students described their normal learning processes associated with setting goals, initiating learning, monitoring learning, evaluating learning, and using ICTs in learning (see Appendix 4F for two excerpts of the interview transcripts).

Preliminary analyses indicated that each mediating factor originated from both personal and situational attributes. In addition, students often participated in both online and offline learning environments during a learning process. They tended to switch between these two spaces for different purposes to study the same content. Therefore, the mediating factors of learner autonomy in both learning environments were not always separately considered in the interviews. To provide background information for these analyses, the following section provides some general descriptions of students’ participation in the LMS during the course.

2.6.2.1. Descriptions of students’ engagement in the LMS

The analyses presented in this section drew on the data collected from the LMS log records of all 247 students participating in the study. As presented in the research design and procedures of the LMS employment in Section 4 of Chapter Three, five classes, taught by three teachers, participated in Study Two. As the first two teachers taught two classes each and students often moved between classes conducted by the same teacher, the analyses put these five classes in three groups. These groups were called A, B, and C, and were taught by Teachers 1, 2, and 3, respectively. The LMS participation was made compulsory to Groups A and B, but not Group C. An examination of the postings in the LMS showed that the content created on the LMS was primarily initiated by students. Only a few discussion threads were started by the group teachers. The students and the teachers altogether made a total of 2296 postings during the course, equivalent to less than one posting per student per week. The teachers did not participate much in the discussion, but it was evident that they were following the discussions closely.

The synchronous chat room in the public section was not used much by students (Dang & Robertson, 2010c). The chat log showed only a few chat lines because students did not log in to the chat session at the same time to be able to communicate with each other although it was

______Chapter 4 Page 140 observed that many of them were in the LMS at the same time. Further investigations of students’ use of technology indicated that their chat community was through other channels such as Facebook, instant messenger, or mobile phone. In contrast to the quietness of the chat room, the global blog, which was also made public, attracted 259 postings from students. This was an unexpected outcome, given that this section was only for general communication, and not part of the course requirements. There were several new entries and comments every day, and many students even found this public section more interesting than their course section. It appeared to meet their need to communicate or perhaps to socialize with their peers.

Students’ participation in the LMS reflected the outcomes of the teachers’ attempts to nurture the virtual learning environment (Dang & Robertson, 2010d). As Teacher 3 only introduced the LMS to her students but did not try to integrate it into the course nor include it in the course assessment as the other two had done, only fourteen out of 36 students in Group C attempted to log in to the LMS, and no posting was found in Teacher 3’s virtual group. One of students in Group C emailed the webmaster, requesting to move her LMS account to another group because she would like to have opportunities to engage in the online learning space. In contrast, all students in Groups A and B logged into the LMS and made a total of 1715 and 322 postings respectively (Table 4.41). Further examination of the number of postings made by Groups A and B showed that both thread starters and replies were found more in Group A than in Group B (741 vs. 46 starters and 974 vs. 276 replies, respectively, see Table 4.41) although the number of students in these two groups was relatively similar. This difference was perhaps due to the teachers’ requirements. Teacher 1 required a specific number of postings made by each student, while Teacher 2 did not.

Table 4.41: Total number of thread starters and replies in the LMS

No. of No. of Section Total starters replies Public 134 125 259 Group A (Classes 1&2) 741 974 1715 Group B (Classes 3&4) 46 276 322 Group C (Class 5) 0 0 0 Total 921 1375 2296

Analyses of the number of thread starters and replies suggested that students interacted with one another in the LMS. As presented in Table 4.42 (page 142), 54.4 percent of thread starters did not receive any reply. This meant that 45.6 percent of the discussion threads received at

______Chapter 4 Page 141 least one reply. This number was smaller in the class section than the public section (45.2 percent and 39.1 percent vs. 50.0 percent, respectively). In other words, students seemed to be more interactive in the public domain than the class section, indicating a moving trend towards the larger community where a more active audience was found. However, the maximum number of replies in a discussion thread was much higher in the class section than the public section (17 and 13 vs. 6, respectively). These suggested that although students generally tended to engage more in the public space, their discussions were shorter than those in the class section.

Table 4.42: Detailed number of thread replies to thread starters

Thread starters that get Maximum Total number no reply at least one reply number of Section of thread replies in a starters No. Percent No. Percent thread Public 134 67 50.0 67 50.0 6 Group A 741 406 54.8 335 45.2 17 Group B 46 28 60.9 18 39.1 13 Total 921 501 54.4 420 45.6 17

Analyses on students’ engagement time in the LMS indicated that students placed a lot fewer postings during working time than after working time. In Vietnam, working hours often started at 7:00 am and finished at 5.00 pm, including a lunch break. The LMS log records showed that 66.1 percent of the postings in the LMS were made after working hours (Table 4.43). This number was relatively similar between the public section and class section (78.8 percent and 64.5 percent, respectively). These numbers confirmed that students often

Table 4.43: Time of postings in the LMS

Postings in public Postings in classes Total Time No. Percent No. Percent No. Percent 7:00 am – 5:00 pm 55 21.2 724 35.5 779 33.9 5:01pm – 6.59 am 204 78.8 1313 64.5 1517 66.1 participated in the virtual learning environment at home due to the limitation of ICTs infrastructure at their school. Another examination of students’ online learning time showed that fewer postings were placed at the beginning of the course. After that, more postings were added, particularly from week 5 to week 11 (Table 4.44, page 143) when students had become more familiar with one another and they could identify their own favourite content and peers to interact with. The number of postings were significantly reduced towards the end of the

______Chapter 4 Page 142 course, probably because most of the students had completed the course requirements and needed to save time to study for the examinations.

Table 4.44: Students’ number of postings on the LMS from week 3 to week 16

Public Group A Group B Group C Time Total Section Section Section Section Week 3 45 4 11 0 60 Week 4 24 87 26 0 137 Week 5 56 97 98 0 251 Week 6 43 113 57 0 213 Week 7 22 208 22 0 252 Week 8 11 253 23 0 287 Week 9 7 208 22 0 237 Week 10 8 305 20 0 333 Week 11 3 337 8 0 348 Week 12 3 49 16 0 68 Week 13 2 11 8 0 21 Week 14 0 12 0 0 12 Week 15 0 11 1 0 12 Week 16 1 8 0 0 9 Total 225 1703 312 0 2240

As expected, reading, reflecting and posting were the activities that students reported having done most in the LMS. Most of them said that they were very careful when placing a posting in the LMS. For a few others, simply meeting the course requirements was their only purpose. There were those who liked reading the messages and resources only. There were also those who preferred reading to sharing materials and those who were interested in doing both. Those who had postings reported that they often engaged in a cyclic process of placing a posting, following it up, reading the comments if there were any, responding to comments, and following up again. Going through this process also triggered a lot of other activities, such as reading the comment-makers’ profiles and postings in the LMS (Student 5), searching Google for further details of the posting topics (Students 1 and 8), and going to the original sources of the postings for other materials (Student 2).

Interestingly, students in all the five classes did not post any voice message on the website. They reported several reasons for this, presenting different levels of participation awareness. Some students said that they did not know about the availability of the voice message tool. Some indicated that they knew about the tool but did not have the necessary equipment such as microphone and speakers to be able to record their oral messages. Some others recorded their oral messages but did not post the files because they were not confident enough and afraid of potential criticism from their classmates. For example, Student 8 said:

______Chapter 4 Page 143 [I did not post any voice message because] it is fine to type my messages. In addition, other people may not understand what I say [in my voice message]. I also don’t like to listen to my voice. Online text messages are just the same for everyone, no personality can be revealed. (Student 8)

These comments were also indicated by Student 7, who said:

The main reason [that I did not post any voice message] is that I am not confident enough with my pronunciation. […If] I become the first person posting voice messages, others will think that I am too arrogant because my voice is not sweet enough, and my pronunciation is not totally accurate. (Student 7)

(also in Dang & Robertson, 2010a, p. 14-15)

In short, the preliminary analyses described in this section provided general descriptions of the students’ engagement in the LMS. The LMS log records showed that a few students in Group C attempted to log in to the LMS, but no posting was found in their class section. Therefore, the content generated in the LMS was basically from Groups A and B. Most of the postings were placed between week 4 and 11 of the course, when students become more comfortable with the facility and their peers. The LMS records also showed that students generally preferred to respond to the public discussions rather than the class discussions, but longer discussions were found in the class section. In addition, the interview data suggested that students behaved differently in their online learning space due to different circumstances.

Therefore, a triangulation among the data collected from the interviews with eleven students, the log records generated from the LMS, and the communication between the webmaster and students was conducted to contribute to the process of identifying the mediating factors on learner autonomy. To maintain the clarity of the data presentation, the mediating factors on students’ learner autonomy in the offline and online learning spaces are separately described. The data collected for the three dimensions associated with the learning behaviours in an offline or traditional learning environment are analysed first. After this, the data collected for the dimension associated with the learning behaviours in an ICT-supported environment are presented.

______Chapter 4 Page 144 2.6.2.2. Mediating factors of learner autonomy performance in offline learning environments

To extract groups of factors contributing to students’ performance of learner autonomy in the offline learning environment, a theme analysis was conducted with the interview transcripts collected from students’ reflective descriptions of their learning behaviours (refer to Appendix 4G for examples of themes and coding). The analysis showed that personal preference, motivation, and attitude were the most important factors that influenced students’ autonomous learning behaviours in the offline learning space. These three mediating factors were reflected at both personal and situational levels. For example, students might be motivated by a friend’s impressive achievement and make a learning plan, but they would probably ignore it soon after that because they preferred a more flexible learning arrangement to working with a set plan. It could be seen that the motivation triggered by situational attributes contributed to the shaping of students’ autonomous learning behaviour. However, this behaviour was not successfully maintained because it did not match with the students’ preference.

Given the important roles of preference, motivation, and attitude in mediating students’ learner autonomy performance which emerged from the theme analysis, each of these three factors are separately analysed in the following section. The analysis of each factor focused on both personal and situational attributes in particular learning contexts. Specific examples were also presented to illustrate how an autonomous learning behaviour was driven by each mediating factor.

Effect of preference on learner autonomy performance

A between subjects theme analysis indicated that personal preference was one of the most influential factors in mediating students’ autonomous learning behaviours. Of the eleven students participating in the interviews, ten said that they performed planning activities as part of their monitoring learning processes. Their plans could be made for a year, a semester, a month, a week, or even a day. They fully understood that a learning plan could help them monitor their learning activities successfully. However, they were not always interested in doing this. For example, Student 9 was a very task-oriented female. She made a semester plan for her EFL study at the beginning of the course and weekly plans based on the school schedules and her living place, given that she shared a room with some others in the school dormitory. She said:

______Chapter 4 Page 145 I made a master learning plan when the semester started… Every two months, I made weekly plans in detail. I scheduled to learn listening and reading in the afternoon and evening because my room was usually quiet at that time… If I could not complete a task in a plan, I would try to finish it on the weekend. (Student 9)

Whilst Student 9 appeared to be very well-organized, Student 10 seemed more relaxed and only made learning plans sometimes, depending on particular situations. He said:

I make no plan for my learning because it is me. I learn what I like at a particular time, [not guided or regulated by a plan]. However, if the examinations come, I’ll make a detailed plan… [because] I’ll be more motivated to study… I am not sure if I’ll make learning plans in the future because I am an unplanned person. (Student 10)

Student 10 knew that making learning plans was useful for time management and effective learning. However, this learning behaviour was not aligned with his personality, and he therefore performed it only in special situations where his personally driven methods were not effective enough. Other situational factors, such as time constraints and entertainment activities, also influenced students’ preferences for making plans. For example, Student 7 indicated that she had to give up planning activities because she commuted to school every day, and the local bus service was not reliable. It always took her more time than expected. Even if she had a learning plan, it could not help her monitor her study effectively. Meanwhile, Student 3 said that she was not a well disciplined person and her learning plans were often delayed due to other social activities.

In contrast to the ten students who had at least some level of planning for their learning activities, Student 6 did not do any planning at all. This female described herself as a romantic person and claimed that her learning activities could be only driven by her psychological mood. She said:

I work totally upon my mood. If I just buy an interesting book, I’ll try to finish reading it before preparing for an examination scheduled in the following day… I usually study when I am free and feel like studying. So only a learning mood, [not a learning plan], works for me. (Student 6)

It could be seen that planning was not always one of the local students’ preferences for learning activities. Those who were more task-oriented and practical, such as Student 9, tended to perform more planning activities for their learning. Meanwhile, those who were

______Chapter 4 Page 146 more relaxed, such as Student 6, did not value planning much. These two extreme approaches of students tended to demonstrate who they were in their learning activities. In contrast, those who did not have a strong personality, such as Student 10, tended to use different learning preferences in different situations. These examples suggested that both personal and situational attributes played a role in their preference for the adoption of an autonomous learning behaviour.

Personal preference was also reflected in the students’ adoption of other autonomous learning behaviours such as setting goals, looking for learning opportunities, and evaluating learning. Ten of the eleven students in the interviews said that they set goals for their EFL learning. Their goals were usually general, such as reading a few books and improving their spoken skills. However, three students presented very specific goals. Student 3, an ambitious girl with a tight learning plan, targeted a grade point average of 8.0 for the semester to win a scholarship. Student 1 aimed to increase his grade point average by 0.2, and Student 11 wanted to achieve a score of more than 750 in a TOEIC test. It seemed that strongly task- oriented and competitive students tended to set specific goals for their learning. Meanwhile, those who thought that their learning performance was lower than the others in the class did not make their goals specific. However, based on the data collected in this study, there was insufficient evidence to make any firm conclusions on goal setting behaviours.

The only student who did not set any learning goals described herself as a shy girl. She came from a province and stayed in the school dormitory. Therefore, her social activities in the city where she studied were limited. She indicated that it was not necessary to set goals because learning EFL was her interest. She said:

English is my favourite and passion. Actually I do not think that it is important to set learning goals. For example, if I am interested in any books, I’ll look for them and read them. It is not necessary to have a goal of reading a certain number of books in a fixed period of time. (Student 5)

This student seemed to suggest that goal setting was only necessary for those who did not want to study hard. They needed such learning goals to remind them of their EFL learning activities. In her case, because she was fully aware of her EFL study and always tried her best, she did not need to set any learning goals. She also implied that it was better to be guided by a personal desire than a goal in EFL learning.

______Chapter 4 Page 147 In addition to individual differences in adopting goal setting activities, students’ preferences were evidenced in their choice of learning initiatives and evaluation behaviours. Nine of the eleven students interviewed indicated that they either went to an English speaking club to looking for speaking opportunities or formed a group of friends to practice English speaking skills. They also often asked their friends to evaluate their learning performance. Although these nine students were all interested in working with others, their engagement in these initiating and evaluating learning opportunities was different. For example, Student 2, who appeared to be extroverted in her behaviours and served on the organizing committee of an English speaking club, was very interested in the speaking activities of the club. She said:

I think that my English is not good yet. The English speaking club is a very good environment for me because I can speak English and improve my communication skills there… [As a committee member], I also have opportunities to practice public speaking skills. (Student 2)

This preference to speak with others was also reported by Student 4, who said:

I am very passionate about English so I want to have more environments to interact with. If I just study alone, I do not have opportunities to interact and exchange ideas with others. In addition, I want to demonstrate myself. I need a playground to perform my capabilities. (Student 4)

Clearly, Students 2 and 4 were very fond of speaking English with other people. Such opportunities for interactions were also enjoyed and used by other students in evaluating their learning. Student 1, for example, liked to check her English proficiency with foreigners in informal contexts, although she was very considerate in approaching someone. She said:

I always want to evaluate my English speaking skills with foreigners around here. However, they may be busy when I approach them or they may be hostile to Vietnamese... [So,] I often go to a coffee shop and start talking to those who do not look sad and stressed. (Student 1)

Her perceptive outlook was also reflected in other learning behaviours. As she was interested in writing short essays, she wanted her essays to be evaluated. However, she was afraid that her friends were not interested in looking at her writings or they might copy her ideas. She therefore ended up evaluating it by herself.

______Chapter 4 Page 148 Learning alone, particularly, appeared to be the learning preference of Students 5 and 11. Although they were somewhat confident about their English competence, they were not interested in interacting with others. Student 5 initiated her learning processes by speaking and reading to herself and recording it. She then listened again to the recordings and evaluated her own performance. Student 11 often looked for new learning opportunities in books and practice tests. He evaluated his learning by checking his answers to a test with the answer key and continued to learn from that. Both Students 5 and 11 knew about the availability of opportunities for interactive learning around them, but they were not attracted to participate. They seemed to be introvert in their learning behaviours.

Apart from those who were very interested in interacting with others and those who preferred to work alone, there were students who wanted to do both, depending on the context. For example, Student 10 said:

I rarely talk to my friends in English… [but] I sometimes go to an English speaking club… I once passed by the Youth House and saw a club there. So I joined it… However, [I don’t join it regularly because] it is a little far from my home and I have a part-time job… [Instead,] I bought TOEFL, TOEIC books, and Special English magazine to learn on my own. (Student 10)

Such contextually driven learning behaviours were also reported by Student 8. She indicated that she had heard about an English speaking club in her school and would like to go there for more learning opportunities. However, she did not do this because none of her friends were interested.

It was suggested that these students were flexible in their preference of a learning activity. As a result, they could adopt various ways to perform autonomous learning behaviours. In short, students’ preferences played an important role in their adoption of each autonomous learning behaviour. Although they all acknowledged the behaviours associated with goal setting, initiating, monitoring, and evaluating learning, they performed these behaviours in different and sometimes opposite ways. Strongly task-oriented and confident students tended to have detailed plans and specific goals for their EFL learning. Meanwhile, those who appeared less focused on the end goals and more motivated by personal desire to learn tended to spend less time planning activities and make their goals more general. Similarly, students who appeared to be more social in outlook tended to perform their initiating and evaluating learning activities through interactions with peers or friends. In contrast, students who appeared to be

______Chapter 4 Page 149 timid tended to perform these activities alone. Regarding the students who were not extreme in personality, their preferences for autonomous learning behaviours tended to be influenced by situational attributes such as company availability, physical distance, and examinations. For example, they might be interested in going to a speaking session and planned to do so, but if a close friend did not come, they stayed home instead.

Effect of motivation on learner autonomy performance

Analyses on the interview data collected from the eleven students suggested three sources of motivation that contributed to the shaping of their learner autonomy performance. The first was personal interests. Some students were interested in some areas of knowledge or skills, and they would perform any autonomous learning activities to satisfy their interest. The second came from the learning successes of their friends or peers. Students with this type of motivation like to be as successful as some other people that they know. The third was associated with a bright picture of high achievement profiles. These students wanted to attain high scores and become an outstanding member in their group. It could be seen that these three primary sources were both internally and externally motivated. While the one originating from personal interests seemed to be more internal; the one associated with friends’ successes seemed to be more external; and the one of being outstanding in a group implied both internal and external attributes.

Of the eleven students interviewed, only Student 6 reported that her learning behaviours were not much motivated by learning achievements. She was primarily motivated by her interest in the literature of different nations in the world. Therefore, she always looked for literature works and read them no matter the language in which they were written. She sometimes read both Vietnamese and English versions of a work to comprehend it better. She also added that:

I rarely pay attention to my classmates’ learning achievements. I don’t care much about that… I sometimes check with my teachers if I do not feel happy with my assignment grades… but I never check with my friends. (Student 6)

What Student 6 said indicated a non-interactive pattern of learning behaviours. She seemed to be alone in pursuing her own interests because it might be difficult to find someone who could share the same interests. In addition, she seemed to enjoy reading alone and did not want to be bothered. Her learning behaviours were mainly directed by the motivation for a

______Chapter 4 Page 150 better understanding of literature works. The motivation associated with gaining better learning achievements than friends and becoming outstanding in the class did not have a strong impact on her learning behaviours. The following examples further illustrate the effects of the three sources of motivation on students’ variations in adopting learning behaviours.

As the study participants majored in EFL, it was understandable that they all had a certain degree of interest in mastering English language skills and developing their understanding of English speaking countries. They often looked for friends with the same interests and formed study groups to enhance their English language competence. For example, Student 2 reported that:

A friend of mine from the University of Finance and Banking was very interested in improving her English skills... Therefore, she and I invited two others, and we meet every Friday and Saturday evening. Each of us presents a topic to practice English speaking skills in general and public speaking skills in particular. (Student 2)

Similar behaviour of creating collaborative learning opportunities was also adopted by Students 3 and 9 in more casual situations. These two often paired with another friend and talked about their concerns in daily life. If it was difficult to find a suitable partner, they tended to study alone or interact with the internet. For example, Student 1 affirmed that she would prepare her lessons very carefully in advance if the topic was one of her interests such as Hemingway or English and American literature. She also looked for more information on the internet. However, if she was not interested in the topic, she would only have a glance at it. This pattern of learning behaviour resembled many others such as Students 3, 4, and 7. Given a cultural tradition of community orientation in Vietnam, the cohort of students in the study generally knew about one another. In addition, students’ learning results were always publicly released. Therefore, students generally knew of their friends’ learning performance, and this had an impact on their autonomous learning behaviours. Apart from Student 6 who was internally motivated only, nine of the other ten reported that they often used their friends’ learning achievements to inform their own learning behaviours. Different aspects of their friends’ learning performance were taken into consideration. For example, in class, when a friend said something, Students 1 and 3 often paid attention to his/her pronunciation and intonation; meanwhile Students 2 and 4 focused on word usages, tenses, and grammatical structures.

______Chapter 4 Page 151 In addition to these specific attributes of language competence, students gave considerable attention to their friends’ overall performance such as fluency, strengths, and weaknesses. For example, Student 8 said that she often compared her friends’ performance with her own to identify what they did well and what they did badly. She would then use this for her motivation. However, students did not always fall into these two distinct groups. Student 4, for example, evaluated her language competence by considering her friends’ competence at both general and specific levels. These suggested that differences in students’ learning behaviours were motivated by their friends’ achievements and how these achievements were assessed.

Further analyses on the differences in students’ attention to language aspects suggested different patterns of learning behaviours. Those who were more motivated by their friends’ specific level of language competence tended to perform very specific learning behaviours such as pronouncing a word many times or figuring out a structure to express an idea (Student 3) and asking a friend to check their pronunciation or recording their speech (Student 10). These behaviours seemed to be short-term and could only meet students’ immediate demands for language competence. Meanwhile, those who were more motivated by their friends’ general competence performed various patterns of learning behaviours. The behaviours could be used for either short-term or long-term learning goals. This was understandable because an overall evaluation of language competence should be collectively constituted by the judgments on different specific language skills.

It is also necessary to note that students’ motivation triggered by their peers’ learning performance might be associated with their own language competence self-evaluation. This meant that if a student thought that he/she had problems with pronunciation or opinion expression, he/she would tend to pay more attention to his/her peers’ performance of these skills. For example, Student 9 said that she listened to her friends carefully because she thought that she was not as fluent as her friends. However, as Student 8 implied, if she found that her fluency was better than her friends, it was unlikely that she would perform an autonomous learning behaviour to improve her fluency. Pairing these two comments suggested that students’ levels of language competence could probably direct their attention to their peers’ language performance, but their peers’ performance could motivate different learning behaviours, depending on the students’ judgment of their peers’ performance.

______Chapter 4 Page 152 In addition to personal interests and friends’ achievements, three students reported that their autonomous learning behaviours were motivated by the possibility of securing a bright future by holding many successful positions. This source of motivation was interpreted as being triggered by both internal and external attributes. Students might develop intrinsic motivation for being successful during their growth in a certain socio-cultural situation. This kind of motivation could also be developed during their EFL learning as they sought to compete with all of their peers. Student 5, for example, looked for more learning challenges by taking extra EFL classes in another language school. She liked to catch up with her friends and attain good scores. She wanted to successfully enhance her language competence to secure a good job. A similar example was reported by Student 6, whose desire for learning success was expressed in a continuous learning plan.

As I aim to be a high achiever, I often prepare the lessons in advance. I examine the topic, look up new words, and read through the lessons… [After each lesson,] I look for additional information about that topic from other resources. (Student 6)

It seemed that the motivation triggered by the desire to have a highly successful career tended to drive students away from their current group. They tended to adopt autonomous learning behaviours which were not associated with those against whom they were competing. Such a tendency was also reported by Student 2, who said:

If I meet a person and realize that he/she is better than I, I would ask myself why he/she could be successful like that and when I would be able to be that good… [However,] I would not ask him/her about his/her learning methods... because I don’t want to be just like him/her. (Student 2)

It was obvious that Student 2 might potentially be motivated by her peers’ impressive achievements. However, she probably did not interact with them because she wanted to be different. Apart from Student 6, who described herself as a romantic female, Students 2 and 5 seemed to be stricter in terms of resolving unaccomplished learning plans. Student 2 said that if she failed to carry out a learning plan, she would give it another day or so to get it done. Such a measure was also adopted by Student 5. Meanwhile, other students seemed to be more relaxed in similar situations. Students 7 and 11, for example, said that they would ignore any plan that they could not finish and went on with others. They only insisted on doing it if it was very important. These comments suggested that although the students’ learning behaviours of

______Chapter 4 Page 153 dealing with unaccomplished plans might be driven by the motivation associated with a bright future, there was also an impact from other attributes such as justification and attitudes.

In summary, students’ autonomous learning behaviours were reported to be motivated by personal interests, friends’ achievements, and the desire to be outstanding. Students’ learning behaviours became more diverse if they obtained different sources of motivation. Those who were motivated primarily by personal interests tended to study alone and did not care much about their peers’ progress. In addition, those who were motivated by their friends’ specific language competence, such as pronunciation and intonation, tended to adopt specific learning behaviours to achieve short-term goals such as repeating a phrase and checking with a friend. Meanwhile, those who were motivated by their peers’ general performance of language competence tended to adopt more various learning behaviours and set long-term plans. In addition, those who aimed to be outstanding in a group tended to detach from their groups and looked for learning opportunities in other places. They also tried harder to complete their learning plans, and such a reaction to unaccomplished plans might be partly mediated by students’ attitude. Therefore, it was necessary to investigate the mediating effects of attitude on students’ adoption of learning behaviours.

Effect of attitude on learner autonomy performance

Analyses of the interview transcripts with students indicated that attitude was an important factor in mediating students’ performance of learner autonomy. Nine of the eleven students reported that their school textbooks were boring, and that was why they did not care much about reading them. For a typical example, both Students 2 and 10 agreed that the textbooks were not interesting enough for them to read and they did not find them useful for their learning. Therefore, they did not enjoy re ading them. Such comments were also firmly acknowledged by Student 11, who said:

I think there is nothing to read in [those textbooks]. I just need to improve my four language skills [speaking, listening, reading, and writing]. Attending classes and reading the textbooks do not help me learn better. Really, nothing at all… So it is better to make my own learning plan and follow it. (Student 11)

It appeared that because these students had a negative attitude towards the school textbooks, they tended to look for learning opportunities in other places. Student 10 looked for

______Chapter 4 Page 154 interesting materials in local bookshops, while Students 1 and 4 were more interested in materials on the internet. In contrast, because Students 5 and 6 did not hold a negative attitude towards the textbooks, their learning behaviours were developed around the textbook activities. Therefore, the learning behaviours of Students 5 and 6 tended to be less interactive but more tasked oriented. Meanwhile, students such as 1 and 4 preferred more interactive learning behaviours and even aimed to develop soft skills such as teamwork and public speaking. These differences also suggested that students’ attitudes towards the school textbooks could mediate their levels of working with reference materials.

The impact of attitude on autonomous learning behaviours was also evident in the students’ activities of checking their work. Student 9 was the only one who reported that she did not check her work before submission because she did not want to read her work again and felt disappointed with it. Her comments suggested that she might never think that she did her work well, and self-evaluating her work could not make her work better. However, the other ten students disagreed with this idea. They thought that reviewing their own work before submission was necessary and they always did this. Further analyses indicated that the students’ levels of review were dissimilar. Seven students said that they often looked over their work for spelling and grammatical mistakes because they thought such mistakes had to be avoided. Meanwhile, three other students said that they reviewed their work very carefully because they loved reading their own work (Student 4) or wanted to express their respect to their teachers (Student 1). Student 11 even sent his work to his peers for evaluation because he thought that an independent reader could more easily identify his mistakes.

The mediating effect of attitude was also reflected in several other learning behaviours of students. Student 10, for example, indicated that one of the reasons that he did not regularly participate in the speaking activities in English speaking clubs was his uncertainty about its effectiveness. Similarly, both Students 1 and 2 said that they would give up writing small essays and planning activities in the future if their language competence was not improved. This suggested that an attitude towards a learning behaviour was constituted by reflective judgments. Students often became interested in a learning activity, tried it for a while and evaluated the outcome. If they were satisfied with that outcome, it was likely that they would develop a positive attitude towards that learning behaviour and continue to perform it. Otherwise, they would substitute it with another one of their interests.

______Chapter 4 Page 155 Importantly, the requirements from class teachers had a strong impact on students’ learning behaviours. Nine of the eleven students participating in the interviews agreed that the textbooks were boring, and generally their learning activities were developed independently from the textbooks. However, all of them said that they would read the textbooks carefully if it was required by their teachers. Such an obedient behaviour suggested the important role of the course requirements and the teachers themselves in students’ learning behaviours. Students seemed to think that the requirements from the teachers were those of the course and needed to be fulfilled. They might also tend to strongly believe that their teachers’ advice was important, and they should follow this. In either of these two cases, the teachers’ advice played a much more important role in mediating students’ learning behaviours.

In short, the analyses identified that variations in students’ attitudes contributed to the differences of students’ autonomous learning behaviours in the study context. Those who had positive attitudes towards a textbook tended to develop learning behaviours based on it. Meanwhile, those who thought that textbooks were boring and useless to them tended to perform the learning behaviours associated with their peers or other learning resources. Students’ attitude formation was also identified to be negotiated during their learning processes. If a learning behaviour was effective after being tried for a while, students’ attitudes towards it would be reinforced and it would be potentially maintained. Otherwise, another learning behaviour would be taken into consideration and tested.

Summary

The analyses described in this section identified three mediating factors for students’ autonomous learning behaviours in the offline learning environment. These three factors were preferences, motivation and attitudes. A theme analysis of between-subjects and within- subjects drawn on the semi-structured interview data with eleven students suggested that these three factors were often interrelated with one another. In addition, each one was constituted from different attributes shaped by personal and situational contexts. For example, some students appeared to be more socially withdrawn, some were more socially active, while others were in between. Similarly, some might have positive attitudes towards a source of learning materials but others might not. As a result, each of these three factors triggered different patterns of autonomous learning behaviours and highlighted the complexity of the construct.

______Chapter 4 Page 156 Regarding personal preferences, task-oriented and confident students tended to made detailed plans and specific goals, while students with a more relaxed approach were likely to spend less time on planning activities and only set general goals. Regarding the effect of motivation, the analyses suggested that students primarily motivated by personal interests tended to study alone, and those aiming to become outstanding tended to detach from their groups in learning activities. Meanwhile, those motivated by their peers’ learning performance tended to set short-term and/or long-term learning goals, depending on the language competency aspects that they heeded or wished to improve. Regarding the effect of attitude, students in the study reported that their attitude towards a particular autonomous learning behaviour was negotiated during their learning processes. Those who thought that textbooks were useful tended to interact more with textbooks in their learning behaviours, while others tended to interact more with their peers or other learning resources.

Given the effects of preferences, motivation, and attitudes on students’ performance of learner autonomy in the offline learning environment, the following section was designed to investigate the effects of other mediating factors in the online learning space. As the virtual space anticipated the use of technology, it was expected that the factors that mediated students’ autonomous learning behaviours in the online environment should be different from those in the offline learning environment. Understanding the effects of mediating factors in the online space would provide more insights into students’ usage of ICTs in learning, one of the four major dimensions of learner autonomy identified in the factor analysis in Study One.

2.6.2.3. Mediating factors of learner autonomy performance in online learning environments

This section is designed to identify the factors that mediated students’ performance of learner autonomy in the online learning space. The analyses primarily drew on the data collected from a section in the eleven semi-structured interviews in which students reported on their learning behaviours associated with ICT-supported learning environments. These interviews provided descriptive and explanatory evidence for students’ autonomous learning behaviours. In addition, students’ login details and their communication with the webmaster were examined. The analysis showed that students’ autonomous learning behaviours in the online environment were primarily mediated by their attitude, technological competency, and goal orientations. To maintain the clarity of the presentation, how these three mediating factors

______Chapter 4 Page 157 were reflected in the students’ performance of learner autonomy is specifically described in the following sections.

Effects of technological competence on learner autonomy performance

Log file analyses indicated that many students had problems using their accounts to log in to the LMS. There were 221 login attempts using odd username patterns by 80 student turns. These usernames included full email address, part of email address, full name with blanks, and username or nickname from somewhere else. In several cases, up to seven or eight consecutive attempts of an odd username were identified. Some students used their own email accounts, instead of their LMS accounts, to log in the LMS and had to seek technical support after several failures. One student even forwarded her account details to another person and asked him to communicate with the webmaster, although she did not have any problems. She was not aware of revealing her personal details and violating the course requirements when doing so.

Examining the communication between the webmaster and students via email and instant messenger also brought the issue of technological competence into account. Several students insisted that their email addresses did not include .vn at the end, although this was clearly indicated in their mailboxes. Many others who failed to seek technical support via IM due to the unavailability of the service at certain times did not use the alternative communication mode. They did not leave any details that enabled the webmaster to track their account details later to provide support either. Therefore, many technical conversations were unintentionally divided into different pieces, carried out time by time. These instances showed that some students encountered even basic technical problems, while the majority of them did not. This result was similar to what was reported in the questionnaire of the study in which 8.2 percent and 4.1 percent of students self-rated their levels of computer proficiency as being bad and very bad respectively.

Students’ technological competence was reported to be closely related to their level of computer access, and this affected their online learning behaviours. Of the eleven students participating in the interviews, nine indicated that they had laptops or desktops and used them on a daily basis. Therefore, they seemed to be more confident with their computer skills, and their learning behaviours were more attached to the ICT-supported environment. For example,

______Chapter 4 Page 158 Student 10 said that he set the LMS site as his start-up homepage and came to the site every day. In a similar sense, Student 3 said:

I often go to my virtual class to see if there was any update […]. I found out that there were many postings that received many replies. Therefore, I came to them, read them, and posted replies, asking for further details. (Student 3)

(also in Dang & Robertson, 2010b, p. 5)

In contrast, Students 5 and 11 reported that they did not use a computer nor come to the LMS every day because they did not have a personal computer with them. Student 5 also said that unlike many of her friends, she did not know how to use blogging and instant messenger. She went online every three or four days for news and learning activities. Participation in the online learning space was much less with Student 11, who had problems using browsers and rarely used computers for learning purposes. He thought that offline learning was enough for him and the online space should be for news updates or other purposes. It seemed that the learning participation of this student in the online environment was also affected by his attitudes towards this learning space.

In brief, the data analyses suggested that students’ levels of computer proficiency affected their learning behaviours in the online environment. Those who were more capable of using computers generally tended to engage more in the online learning space. Meanwhile, those who were less proficient in computer skills tended not to take part in this learning environment much or even to avoid using this environment for learning purposes. In addition, they indicated that the online learning environment was not very important for their EFL learning. This suggested that students’ attitudes towards the online learning environment contributed to the shaping of their online learning behaviours. Such an effect of attitudes is more closely examined in the following section.

Effects of attitude on learner autonomy performance

Students’ responses in the study presented mixed attitudes towards the LMS, ranging from neutral to positive and very positive. Three of the eleven interviewees indicated that they initially participated in the LMS because it was part of the course requirements. However, after taking part in the online activities, they became interested in the diversity of friendly

______Chapter 4 Page 159 topics and relevant content available online. As a result, their attitudes towards this learning environment became more positive. For example, Student 10 said that because regular engagement in virtual class activities was required, he had to come to the LMS every time he connected to the Internet. However, this did not stop him from enjoying reading the LMS, placing postings, sharing video clips, and commenting on his friends’ postings. These activities indicated a move from an outsider to an insider of the community, gradually establishing relationships and controlling the situation (as presented in Bhappu, Ebner, Kaufman, & Welsh, 2009).

Five other students said that they were interested in participating in these online interactions right from the beginning, and their attitudes were even more enhanced as the course went on, as a result of the useful discussions and reflections they experienced with the LMS. For example, Student 6 said:

I often come to the LMS to see if there is anything new. However, I gradually visit the site more frequently because there are so many postings there… The LMS made the course more interesting… I could know a lot about [my friends] when reading their postings on the LMS... (Student 6)

However, there were two comments suggesting a casual use of the LMS. For example, Student 11 indicated that because he placed limited value on the online materials, he only logged on to the LMS a few times during the course. Similarly, Student 7 reported that she did not pay much attention to the LMS and “always came to it at last” every time she went online. She sometimes turned off the computer without coming into the LMS even though her initial plan was to check what was going on in her virtual class. She considered it as a leisure activity rather than a learning activity.

There were also two opposing perspectives regarding the kinds of comments that teachers should make on students’ postings. Student 8 did not expect to have many specific comments from the teachers because “that would make students have a feeling of being checked for mistakes all the time”. In contrast, Students 2 and 10 said they would like to have as many as possible specific comments from their teachers, for example about spelling, grammar, word use, content or personal thoughts. They believed that the teachers were more experienced than them and could clearly point out the mistakes for them to learn from. Although they may have felt a little embarrassed once their mistakes could be seen by all of their classmates, they

______Chapter 4 Page 160 thought it would be good for their progress. Moreover, the LMS was the only opportunity to obtain teachers’ specific corrections because there was no time for that during in-class hours. These comments were similar to those which were reflected in the teachers’ attitudes towards the online space.

It seemed that students’ learning attitudes towards the online learning space mediated their learning behaviours in the LMS. A more positive attitude generally led to a higher level of engagement in this learning environment, while a negative attitude might prevent students from working with this environment. In addition, different attitudes towards learning activities potentially led to variations in students’ behaviours. Those who did not want to be monitored in the online environment tended not to expect teachers’ specific responses, while those who wanted to be corrected tended to expect more specific comments from teachers. This also suggested an effect of students’ goal orientation in the online learning space. Therefore, more investigations were conducted in the following section to gain a better understanding of the effect of this variable.

Effects of goal orientation on learner autonomy performance

Analyses on students’ learning behaviours in relation to their goals suggested three patterns of learning participation in the online learning space, namely task-oriented, content-oriented, and community-oriented participation. Task-oriented participants were those who logged in to the LMS only to meet the course requirements of the online component (also suggested in Kessler & Bikowski, 2010). After placing a posting, they probably stayed in the LMS to meet the onsite time requirement, but they did not actively participate in the LMS activities. They did not cooperate with their peers as documented in previous studies (e.g., Lomicka & Lord, 2009b). Very often, their postings were not well-prepared, their interaction with the LMS was one-way directional, and their tie with the online space was very loose (as suggested by Kent & Facer, 2004). For example, Student 4 said that she seldom read her postings in the LMS again in any situation. Similarly, Student 1 revealed that:

Because my teacher only required the number of postings… I just posted anything. I did not have to proofread it carefully because even if I did, no one would read my postings. (Student 1)

______Chapter 4 Page 161 The second type of participation was content or information-oriented. These students tended to target the postings related to their interests. They preferred to work with high quality content and often ignored short messages for relationship building or maintenance. They also became frustrated easily and left the LMS if they could not see anything appealing to them. They often paid attention to the thread titles and the length of the messages before considering the details (Students 2 and 8). They did not care much about the authors of the postings (Student 3). Their postings were often initiated by what they read on the LMS. They particularly prepared their postings carefully before hitting the post button and followed them up (Student 4). Sometimes, they reported that they first typed their message but did not post it because they did not believe it to be sufficiently interesting.

The third pattern of engagement involved community-oriented participants, who tended to come to the LMS to interact with the peers they knew. They often targeted messages written by familiar people (Student 4) or those attracting many people (Student 10). Also, they often expressed their personal feelings and used dialoguing connectors such as adding “hello” or “thank you” in responding to the authors. They appeared not to pay much attention to a discussion topic if they could not see any of their close friends participating in it. In other words, their online participation appeared to be significantly mediated by their relationship with their online peers and the number of people online. For example, Student 11 said:

I do not go online very often… However, if there is a close friend of mine posting something in the LMS, I’ll have to come to see what he/she writes about… For the postings of the others, I may look at them later. (Student 11)

Student 5 added:

I look at the number of people participating in each discussion thread [to decide if I need to read it]… If the thread has already attracted two or three comments, I will come to see. If the thread has not received any comment, I will not come to it... Therefore, never am I the first person responding to a posting. (Student 5)

(also in Dang & Robertson, 2010c, p. 17)

These analyses on the three patterns of participation suggested that students’ learning behaviours in the online learning environment were mediated by their goal orientations. Task- oriented students tended to take part in the LMS to meet the course requirements only. They did not seem to use the LMS for their learning purposes. In contrast, content-oriented students

______Chapter 4 Page 162 tended to look for learning opportunities in the LMS and engage in the resources which were collaboratively constructed by their peers and teachers. Meanwhile, community-oriented students were more interested in making friends and maintaining social contacts with their peers. They tended to consider the LMS as an environment to colour their academic life. Of these three goal orientations, it seemed that content orientation would potentially lead to the most effective learning. However, it would become even more effective if students held all of the three orientations at the same time for their online learning.

In brief, the analysis process described in Section 2.6 was to answer the fifth research question regarding the mediating factors of students’ autonomous learning behaviours. Students’ learning behaviours in both the offline learning environment and online learning space were investigated. As a result, apart from gender and teaching, three other mediating factors in each learning environment were identified. Preference, motivation and attitude were the three prominent factors in the offline learning space, and the other three in the online space were computer proficiency, attitude, and goal orientation. This outcome seemed to suggest that attitude was a commonplace factor in mediating students’ autonomous learning behaviours because it was evidenced in both learning spaces. However, it did not imply that attitude could have a greater effect on students’ learning behaviours than the others. The analyses also showed that different attributes at a situational, institutional, and personal level contributed to the shaping of students’ mediating factors. In addition, it would be superficial to conclude that these factors were totally independent in triggering students’ learning behaviours. Therefore, understanding the relationship among these factors would need to be investigated in further research.

3. Summary

This chapter uses the data collected from students’ responses to the questionnaires, the LMS, and the interview to answer the five research questions proposed in the current investigation. The analyses of students’ perception in Study One suggest a model of learner autonomy which consists of four dimensions, namely Monitoring learning processes, Goal-setting and evaluating learning, Using ICTs in learning, and Initiating learning opportunities. These dimensions are likely to be dependent on one another but perceived and performed at different levels. Students seem to study autonomously what they think important for their learning.

______Chapter 4 Page 163 However, this association is suggested to be weak and different between the beginning and the end of the course.

Study Two further examined the relationship between students’ perception and performance of learner autonomy and the mediating factors in the context of EFL learning in Vietnam higher education. The analyses showed that students’ self-rated performance of learner autonomy was positively related to their perceptions of learner autonomy. However, this relationship was fairly weak, indicating that it was mediated by many other factors, for example the involvement of examinations in the end of the course. Further investigations suggested that apart from gender and teaching, students’ performance of learner autonomy in the offline learning environment was generally mediated by their learning preference, motivation and attitude. Similarly, their autonomous learning behaviours in the ICT supported space were generally mediated by their computer proficiency, attitude, and goal orientation. These findings have highlighted the manifestation of learner autonomy and local factors in shaping this construct. They are further analysed in relation to those reported in prior research and presented in the following chapter.

______Chapter 4 Page 164

Chapter Five

DISCUSSION

______

Based on the data generated from the research methodology described in Chapter Three, Chapter Four reported results and different analysis procedures to consider the five research questions proposed in Chapter One. The exploratory factor analysis suggested that Vietnamese EFL undergraduate students tended to perceive learner autonomy in four dimensions, namely Monitoring learning processes, Goal-setting and evaluating learning, Using ICTs in learning, and Initiating learning opportunities. Other statistical analysis processes were then conducted to understand the relationship between students’ perceptions and self-reported performance of learner autonomy. Finally, the mediating factors of students’ performance of learner autonomy were both quantitatively and qualitatively investigated. Apart from gender and teaching, results indicated that students’ performance of learner autonomy in the offline learning environment tended to be mediated by their learning preference, motivation and attitude. The results also suggested that students’ performance of learner autonomy in the online learning space tended to be mediated by their computer proficiency, attitude, and goal orientation.

This chapter attempts to integrate the results of Study One and Study Two reported in Chapter Four and analyses them in relation to the previous research findings in the area of learner autonomy in second language education. The discussion is built around the five themes proposed by the five research questions of the current investigation. The first section relates the four dimensions of learner autonomy identified in the Vietnamese context with those identified in other contexts. The second section discusses the similarities and differences among these four dimensions of learner autonomy to gain a better understanding of the construct. The third section considers the relationship between perception and performance of learner autonomy to predict students’ autonomous learning behaviours in the local context. The fourth section analyses changes in students’ performance of autonomous learning behaviours during a semester course. The fifth section examines the mediating effects of the

______Chapter 5 Page 165 seven factors on students’ performance of learner autonomy identified in the current investigation.

1. Perceptions of learner autonomy

In an analysis of the four-dimension model of learner autonomy perceptions, reference was made to other models developed in previous research. In this context, it should be stated that the current investigation was conducted in a Vietnamese context and was based on an adaptation of the questionnaire instrument for data collection from a Japanese based context (Yang, 2007). As learner autonomy is a complex construct, it could be expected to be perceived differently in different contexts. Hence, a comparison of the dimensions of learner autonomy extracted from the two study contexts is considered. Hence, the four-dimension model of the current investigation is paired with other theoretically-proposed and empirically- identified models. Similarities and differences among these models will contribute to the shaping of a learner autonomy model for the Vietnamese context. All of these analyses are specifically described in the following section.

1.1. Overview of the four-dimension model of learner autonomy

The exploratory factor analysis presented in Chapter Four primarily drew on the quantitative data collected from 562 participants, and it suggested four dimensions of learner autonomy from the perspective of Vietnamese EFL undergraduate students. These four dimensions were Monitoring learning processes, Goal-setting and evaluating learning, Using ICTs in learning, and Initiating learning opportunities (refer to the results on page 103-105). Whist the third dimension refers to the learning behaviours in the online space, the other three dimensions refer to the learning behaviours in the offline environment. It can be seen that the three dimensions associated with the offline learning space are similar to the three core processes indicated in the conceptual framework of learner autonomy developed in the literature review (Table 5.1, page 167). This suggests that the pattern of learner autonomy perceived by Vietnamese EFL students is relatively similar to what is proposed and initially developed in Western contexts where students are often considered as being more active and independent (Hassan, et al., 2010).

______Chapter 5 Page 166 Table 5.1: Dimensions identified from the current study with those in the literature

Dimensions Core processes generated from the current study synthesised from the literature (see results on page 104) (see Table 2.3, page 56) Monitoring learning processes Monitoring Goal-setting and evaluating learning Evaluating Initiating learning opportunities Initiating

The data of the current investigation suggest that despite being known as passive and obedient (Pham & Ngo, 2008), Vietnamese EFL students acknowledge different learning alternatives. Their silence in class which was often observed in previous research (e.g., Le, 2008; Wong, 2004) may not necessarily indicate a preference for rote learning or that they do not want to control their learning processes. This inference supports the argument that students may not have been given enough opportunities to be active or control their learning through class activities. The data suggest that they were trained to be passive and silent and behaved in this expected manner although such behaviours were not always their preference. This argument aligns with the socio-cultural perspective of learner autonomy which was the focus of the current investigation. Students in the local context appreciated autonomous learning behaviours but they did not look to be autonomous in class because they did not appear to be supported to act independently (Le, 2000). Teaching practices and other cultural norms may have constrained students’ performance of learner autonomy in the local context.

The learner autonomy pattern of Vietnamese students observed in the current investigation shows some similarities and differences with the pattern observed in Yang’s Japanese based study (2007). Although the questionnaire instrument of the current investigation was primarily developed from Yang’s learner autonomy inventory, only the first four dimensions, namely self-initiative, making plans, self-control, and flexibility, suggested in her study were generally shown to occur in the current investigation. Accordingly, thirteen of twenty-six items in these four dimensions received a high level of consistent ratings by the Vietnamese sample and thus were included in the three dimensions of the current study model (see Appendix 5A). However, only the structure of the dimension associated with learning initiative activities is basically similar between the two studies. This dimension is called Initiating learning opportunities in the current study and Self-initiative in Yang’s (see Table 5.2, D4 and Y1 respectively, page 168). The factor loadings of the other items were different, constituting differences in the pattern of learner autonomy between the two. For example, item 38, notice mistakes and use that information to do better, was grouped in the monitoring

______Chapter 5 Page 167 dimension in the current investigation, but grouped in the planning dimension in Yang’s study. As a result, the items which were grouped in the dimension associated with monitoring activities of the current study are from three dimensions of Yang’s study.

Table 5.2: Thirteen items included in both studies

Current Study Yang’s Study Items D1 D2 D4 Y1 Y2 Y3 Y4 37: use time effectively x x 32: make schedule so they’ll have engh time to study E x x 62: know the method which suits them best and use it x x 6: make good use of materials & res when study E x x 38: notice mistakes & use that info to help them do better x x 36: reflect on what they learn and look for sth important x x 43: check to make sure that they understood the lesson x x 19: try to improve their weaknesses x x 25: think about their progress in learning English x x 9: try to find as many ways as they can to improve their E x x 12: want to study in an E-speaking envi if having a chance x x 7: look for opportunities to use E as much as possible x x 41: want to find a job where only E is used in the future x x The items in the factor associated with the use of ICTs (Dimension 3) in the current study were removed for clarity because these items were newly developed used in the current study only. The wording of the items was a bit different between the two studies. The wording presented in this table was from the current study. D1: Monitoring learning processes Y1: Self-initiative D2: Goal-setting and evaluating learning Y2: Making plans D4: Initiating learning opportunities Y3: Self-control Y4: Flexibility

Further investigations on the dimension associated with learning initiative which is shared by both studies suggest a sizable difference. While self-initiating was the most prominent dimension identified in Yang’s study, it was the least prominent in the current investigation. This suggested that the autonomous learning behaviours linked with learning initiative took a smaller portion in the model of learner autonomy perceived by the Vietnamese than the Japanese counterparts. In other words, in the context of Vietnam, students were less likely to generate learning opportunities when learner autonomy was concerned. Meanwhile, in the context of EFL learning in Japan, this dimension was the most dominant. Such a difference could be due to certain situational factors, but at this stage, there is inadequate evidence to make a proper comparison.

The pattern difference of learner autonomy between the two contexts was also reflected in the absence of two Japanese dimensions in the Vietnamese sample. All items from these two dimensions, taking actions and concentration, were eliminated during the analysis of the data from the Vietnamese sample of the current study. Two possible interpretations are proposed.

______Chapter 5 Page 168 First, perhaps because in Yang’s study, these two dimensions only accounted for 4.41 percent and 4.40 percent of the total variance respectively, they were too small to be reflected in the current study sample. This assumption is statistically logical because the analysis procedure employed in the current study tended to extract discrete and strong dimensions only. Second, students in these two contexts had different perspectives on these two dimensions. The Vietnamese students tended to be more relaxed in class and did not value the class lectures as indicated in the interview data. Therefore, they tended not to appreciate the attribute of concentration in their EFL learning. Meanwhile, Japanese students might tend to be more focused on class lectures and attach greater importance on class attendance. This may explain why they valued the behaviours associated with concentration.

The separation of the learning activities associated with the use of ICTs from the others was also an unexpected outcome. The literature of learner autonomy seemed to suggest that the three core learning processes of learner autonomy should be applicable to both traditional and ICT supported learning environments. Nevertheless, students in the current study essentially perceived the learning processes in these two environments differently. The learning activities associated with the use of ICTs did not tend to be integrated with other learning processes in the traditional learning space. Prior research often investigated learner autonomy in each learning space separately (Lamb, 2008), and critical comparisons of autonomous learning behaviours between the two learning spaces were scarce. With the data generated from this investigation, it was suggested that different autonomous learning behaviours were perceived in different learning spaces.

1.2. Items in the four dimensions of learner autonomy

Of the four dimensions of the model identified in the current study, the one associated with monitoring learning behaviours was the most dominant. This reflects the nature of the exploratory factor analysis and the complexity of the dimension. The statistical formulae of the exploratory factor analysis are designed to extract and arrange factors in relation to the percentage of variance that can be explained. The factor, accounting for the greater percentage of variance explained, tends to consist of more items. In the current study which employed a four-factor solution, the most dominant factor, Monitoring learning processes, accounted for 26.794 percent of the variance and included nine items (see Table 4.5, page 103). These items appeared to relate to different activities such as time management, planning, collaborating,

______Chapter 5 Page 169 and reflecting (as summarized in Table 2.3, page 56). This result highlights the complexity of the dimension because monitoring learning often requires the involvement of different learning attributes.

Investigations of the nine items included in the monitoring dimension of the current study suggested a number of attributes in students’ learning behaviours. As presented in Table 5.3, learning behaviours associated with time management were reflected in items 37, 32, and 48; behaviours associated with planning or maintaining learning agendas were reflected in items 32 and 48; behaviours associated with reflecting, reviewing, and evaluating were included in items 49, 38, 36, and 43; and behaviours associated with selecting appropriate learning methods were indicated in items 62 and 6. Pairing these behavioural attributes with those indicated in prior research (such as Hart, 2002; Miller, Hopkins, & Tsang, 2005; Sinclair, 2009; Smith, 2001; 2003a) showed a lot of similarities. Notwithstanding, the attribute index of the dimension generated in the current study was more comprehensive. It covered a wider range of behavioural attributes which prior studies could not do because they focused on much smaller samples.

Table 5.3: Items in Dimension 1 and their interpreted attributes

Dimension 1: Monitoring learning processes Attributes 37: use time effectively Time management 32: make their schedule so they will have engh time to study E Planning, time management 49: study things which were not from their class Reviewing, making decisions 62: know the method which suits them best and use it Evaluating, selecting methods 48: try to study English regularly even with limited time Planning, time management 6: make good use of materials & res when studying E at home Selecting methods 38: notice their mistakes & use that info to help them do better Evaluating, engaging purposefully 36: reflect on what they learn and look for something important Reflecting, analysing 43: check to make sure that they understood the lesson Evaluating, engaging

Further examination of each of the items in Dimension 1 suggested that most of the items referred to more than one behavioural attribute. For example, item 38, notice their mistakes and use that information to help them do better, included the activity of recognising mistakes and using these mistakes to improve learning. Arguably, if students only noticed their mistakes and did not use this information in their next learning activities, they did not seem to monitor their learning process effectively. Similarly, item 32, reflect on what they learn and look for something important, integrated two activities of reflecting on learning and identifying important information into a learning process. These examples reconfirm the

______Chapter 5 Page 170 multi-phases of a monitoring process which often include a series of learning activities, serving different but interrelated purposes.

Whilst items in Dimension 1 focus more on the behavioural aspect of learning processes, items in Dimension 2 cover both behavioural and cognitive aspects. As presented in Table 5.4, items 55, 58, 45, 25, and 15 refer to the cognitive aspect of the dimension. They indicate what students think, know, or want for their learning. Meanwhile, items 19 and 60 refer to the behavioural aspect of the construct. They indicate what students actually do in their learning process. The inclusion of two conceptual aspects in Dimension 2 suggests that goal-setting and learning evaluation emphasize the important role of the cognitive aspect in the implementation of this dimension. In other words, it is likely that students need to explicitly understand the importance of setting goals and evaluating learning and really want to do this before they can actually carry out those activities.

Table 5.4: Conceptual aspects and attributes contributed by the items in Dimension 2

Aspects Attributes Dimension 2: Goal-setting and evaluating learning Cognitive Behavioural Goal-set Eval. 55: think English is important for their future x x 58: are aware of their studies x x 45: want to be good English learners x x 19: try to improve their weaknesses x x 60: practice English with people outside class x x 25: think about their progress in learning English x x 15: know their good points and weaknesses x x

It is important to note that goal setting and learning evaluation were perceived as one dimension in this investigation. As described in Table 5.4, three items tended to refer to goal setting activities and four items tended to refer to learning evaluation activities. The inclusion of two types of learning activities in one dimension seems to indicate that students do not separate these two types of activities and suggests a very close association between the two. If a learning goal is set, students want to know about their learning progress. This result is different from what has been suggested in prior research. For example, Rivers (2001) found that students set their goals first and attempt to modify the learning activities to achieve their goals. Meanwhile, Aoki (2001) showed that students participate in learning activities prior to setting their goals. It seems that goal setting is often indicated to be associated with learning engagement rather than learning evaluation.

______Chapter 5 Page 171 Previous research also suggests that the relationship between goal setting and learning evaluation is at the two opposite endpoints of a learning process. Both Nunan (1997) and Littlewood (1999) advocated in their models of learner autonomy that an autonomous learning process should start with goal setting activities and finish with evaluation activities. That is, autonomous students are expected to perform a series of learning activities between goal setting and evaluation activities, and it is unlikely that they would associate goal-setting with learning evaluation so directly and closely. Nonetheless, students in the current study were different, perhaps because they studied in a very exam-oriented system where both social and institutional adjustments on students’ learning outcomes were totally based on examination scores (Le & Barnard, 2009a; Tran & Baldauf, 2007).

The third dimension of learner autonomy identified in the current investigation was associated with the use of ICTs in EFL learning and included four items. As suggested in Table 5.5, one of these items in the instrument referred to the cognitive aspect of the dimension construct, and three others referred to the behavioural aspect. This suggests that similar to goal setting and learning evaluation, the students’ usage of ICTs for learning purposes is mediated at the cognitive level (also suggested by Mynard, 2009 in a blogging environment). For example, the analysis of the interview data shows that positive attitudes towards or preferences for this learning environment may trigger more autonomous learning behaviours. This argument is aligned with the stage models proposed by Nunan (1997) and Scharle and Szabo (2000) for the offline learning environment in which autonomous learning behaviours are suggested to be driven by a certain level of awareness or attitude. However, the application of such models in the online learning environment has received limited discussion in previous research.

Table 5.5: Conceptual aspects and attributes contributed by the items in Dimension 3

Aspects Dimension 3: Using ICTs in learning Attributes Cognitive Behavioural 31: like to study with computers x Selecting methods 42: go online as a way of learning English x Selecting methods 33: will do a search on internet if have a question x Initiating 46: pay more attention when they see an E website x Initiating

Further examination of the four items (see Table 5.5) also suggests that they describe only two attributes, namely initiating learning and selecting methods. This seems contradictory to the various platforms and opportunities that the online environment can possibly offer for learning. In addition, the behaviours indicated in the four items of this dimension seem to

______Chapter 5 Page 172 suggest a hierarchical relationship rather than a parallel order. The first two refer to students’ selection of computer and online engagement for learning purposes in general, while the second two refer to two specific learning behaviours, namely searching the Internet and targeting English websites. These two behaviours serve as an illustration for the overarching concept of online engagement and interest in using computers. This probably contributes to the low internal consistency level of the dimension and suggests that the instrument was inadequate to obtain an overall picture of learner autonomy from the ICT-supported learning space.

Given the limitation of the quantitative instrument used for understanding the dimension of learner autonomy associated with the online learning space, insights were sought from LMS log records and interview data. Apart from searching for information on the Internet and targeting English websites, the interview data suggest several other learning behaviours were occurring in the online space. These include reviewing, reflecting, interacting, responding, maintaining connections, and building networks. The analysis even identified some negative behaviours in this learning space that suggest learning avoidance or superficial participation. The conclusion is that the learning behaviours in the virtual space in the study context are various and should be expected to be different from those in the offline learning environment. Therefore, future research could make better use of this insight by a subsequent update of this instrument. This would be able to capture a more sufficient picture of autonomous learning behaviours in the ICT-supported learning environment, increasing the reliability of the data collected.

Dimension 4 of the learner autonomy model analysis was associated with the students’ ability to look for learning opportunities and consisted of five items. Similar to Dimensions 2 and 3, the anticipation of both cognitive and behavioural aspects in the dimension construct was clearly evidenced, even though this was the smallest dimension of the study model (s ee Table 5.6, page 174). Students’ learning initiation was also suggested to be underpinned by their cognitive drive. For example, the interview data showed that students who described themselves as being timid tended not to initiate any oral communication with their peers or foreigners. By comparison, those who reported themselves as being interactive and competent tended to initiate conversations with others with an appropriate manner.

______Chapter 5 Page 173 Table 5.6: Conceptual aspects contributed by the items in Dimension 4

Aspects Dimension 4: Initiating learning opportunities Cognitive Behavioural 4: want to communicate with foreigners in English x 9: try to find as many ways as they can to improve their E x 12: want to study in an E-speaking envi if having a chance x 7: look for opportunities to use English as much as possible x 41: want to find a job where only E is used in the future x

Investigations of the five items in the instrument of Dimension 4 show that all of them refer to the learning initiative attribute only. This contributes to the distinctiveness of the dimension. The instrument items describe the different ways that were employed by students to create learning opportunities. The targeted learning environments ranged from a less structured one such as casual communication with foreigners and friends (items 4 and 7, respectively) to more established and long-term situations such as in an English-speaking school or workplace (items 12 and 41, respectively). By comparison, the items of the first three dimensions always referred to at least two attributes of learning behaviours. On one hand, Dimension 4 could be considered the most unified of the four, and aligned with the analysis procedure which aimed at extracting discrete dimensions. On the other hand, it did not reflect the complexity of the concept. It can be seen that the instrument adapted from Yang’s (2007) study is useful to describe the learning initiative dimension, but it also has certain limitations.

Further analyses associated with the interview data collected in the current study suggest another important side of the behaviours linked with learning initiative activities. Some students reported that they interacted with new learning materials. For example, Student 10 in the interview said that he looked for new learning opportunities in bookshops, and Students 1 and 4 went for internet resources. These bookish behaviours of learning initiation were briefly addressed in prior research (e.g., Dickinson, 1987; Riley & Zoppis, 1985) but they were not thoroughly discussed because the activities associated with learning initiation were argued to take place primarily in a rich-resource lab or with a learning peer (Brandon, 2003; Miller, et al., 2005). Perhaps, the absence of these less social interactive behaviours of learning initiative in the quantitative instrument prevented the dimension instrument from achieving a higher level of internal consistency.

In brief, the analysis shows that both cognitive and behavioural aspects of the model identified in the current investigation are evident in three of the four dimensions. In the

______Chapter 5 Page 174 dimension associated with activities of learning monitoring, the cognitive aspect was not clearly represented (see Table 5.3, page 170). Perhaps this is because this dimension included a few different attributes which students often performed in their learning routine. Therefore, the anticipation of the cognitive drive in the dimension was automatically understood. Other analyses of both the quantitative and qualitative data collected in the study indicated the complexity of each dimension. The following section further explores this conclusion.

1.3. Analysis of the four-dimension model of learner autonomy

Different from most of previous studies, the current investigation was designed in two phases. This provided opportunities for the learner autonomy model to be constructed and validated in the first and second phase with different samples of the same population, respectively. The internal consistency levels of the four dimensions in both phases, named Study One and Study Two, (see Table 4.19, page 120) establish the reliability of the instrument and confirm the identified structure of the four-dimension model.

The four-dimension model of the current investigation seems to account for the added dimensions when compared with previous studies conducted in Vietnam. Trinh (2005) suggested a model of three dimensions, namely planning, monitoring and regulating. Nguyen (2009) only included two dimensions, namely self-initiation and self-regulation. Regardless of any difference between these two models, pairing them with the one identified in the current investigation showed similarities and differences. As presented in Table 5.7, two dimensions of the model in the current study, namely monitoring and initiating learning, reflect the structure of Nguyen’s and Trinh’s models. The other two dimensions, associated with using ICTs and evaluating, are newly identified. The attributes of items in the first dimension of the current study (Table 5.3, page 170) describe the three dimensions of planning, monitoring, and regulating in Trinh’s model. This could also be the reason that these three dimensions were considered as one in Nguyen’s model.

Table 5.7: Dimensions of learner autonomy developed from Vietnamese samples

Current Study Nguyen (2009) Trinh (2005) Planning Monitoring Self-regulation Monitoring Regulating Initiating Self-initiation Goal-setting & evaluating Using ICTs

______Chapter 5 Page 175 The three models indicate some chronological development. Trinh’ model (2005) primarily focused on the dimension of monitoring or regulating learning processes. Four years later, self-initiation was included in Nguyen’s model (2009). The current study now identifies two other dimensions and adds them to the model. Although the number of dimensions has increased within each study, the weighting of dimensions has generally remained stable. The most dominant dimension across the three models is monitoring. It is also important to note that the model observed in the current study is primarily driven by the data collected, particularly in the local context. Hence, it is more contextually based and should reflect the nature of learner autonomy as experienced by Vietnamese students. The two previous models are essentially driven by the literature which was largely developed from other contexts.

In addition, the model identified in the current study is conceptually reinforced by the attributes described in its four dimensions. As presented in Table 5.8, several dimension attributes of the current study reflect findings from previous studies. For example, in the

Table 5.8: Leaner autonomy dimensions and their attributes

Current Study Literature Review (as in Table 2.3, page 56) Dimensions Attributes Dimensions Attributes planning maintaining agendas & keeping track selecting methods identifying and employing strategies making decisions selecting appropriate materials Monitoring analysing tailoring, customizing & personalizing (Aoki, 2001; Monitoring engaging purposefully being flexible and regulating learning Gardner, 2009; managing time taking actions or implementing agendas Lamb, 2009) reviewing concentrating on learning evaluating collaborating and interacting reflecting expressing opinions & negotiating being aware of learning goals & strategies looking for learning Initiating identifying goals and setting goals Initiating opportunities in different (Yang, 2007; making plans or work agendas environments Voller, 2005) looking for resources Evaluating reflecting critically Goal-setting evaluating (Braine, 2003; evaluating learning outcomes & evaluating setting goals Little, 2001) correcting mistakes initiating learning Using ICTs selecting methods dimension described monitoring learning processes, the attributes of planning in the current study are likely to be similar to maintaining learning agendas and keeping track in prior research (e.g., Smith, 2003a; Trinh, 2005). Similarly, selecting methods and making decisions in the current study reflect the attribute of identifying and employing suitable strategies in the literature (e.g., Benson, 2006; Rivers, 2001). In addition, a few attributes are classified into

______Chapter 5 Page 176 different dimensions between the current study and the others. For example, in the current study, setting goals is loaded in Goal-setting and evaluating, while being aware of learning goals and strategies is argued to be in Initiating in the literature review. Such differences are understandable because they reflect the conceptual variations of the model in different contexts.

As shown in Table 5.8 (page 176), the dimension Using ICTs is not integrated into other dimensions although its two attributes, namely initiating learning and selecting methods, are included in dimensions Monitoring and Initiating, respectively. That is, the attributes of the dimension related to online behaviours are similar to those of the dimensions related to offline behaviours, but the dimension related to online behaviours is independently perceived from the others. Sharing ideas, reviewing postings, and reflecting on learning in the virtual learning space seem to require a different set of skills and interest from offline behaviours. Perhaps this is one of the reasons that research into students’ autonomous engagement in the online learning environment (such as Jeon-Ellis, Debski, & Wigglesworth, 2005; Wells, 2007) pays limited attention to students’ learning behaviours in the offline context.

Further examination of the learning behaviours in both the online and offline space indicated in the current model confirms the theoretical model of learner autonomy developed in the literature review. As presented in Figure 2.2 (page 57), an analytical synthesis of six models

Table 5.9: Allocation of items to the model of learner autonomy from literature

Benson (2001), Littlewood (1996, 1999), Macaro (1997, 2008), Nunan (1997), & Scharle and Items in the four-dimension model Szabo (2000). See Figure 2.2 (page 57) observed in the current study Cognitive Demonstrated Situation processes behaviours management 37: use time effectively x 32: make schedule so they will have engh time to study E x 62: know the method which suits them best and use it x x 48: try to study English regularly even with limited time x 36: reflect on what they learn and look for sth important x x 43: check to make sure that they understood the lesson x 55: think English is important for their future x 58: are aware of their studies x 19: try to improve their weaknesses x x 7: look for opportunities to use E as much as possible x x in the literature suggested three principal areas of control in the concept of learner autonomy. They include cognitive processes, demonstrated behaviours (classroom-like situation), and

______Chapter 5 Page 177 situation management (beyond classroom). These three areas are consistently reflected by the 25 items of the four-dimension model extracted by the study data. For example, as partly shown in Table 5.9 (page 177), items 55 and 58 illustrate the learning behaviours taking place in the cognitive process; items 36 and 43 indicate the autonomous learning behaviours in a classroom-like situation; while items 37 and 32 are more associated with behaviours of situation management (see Appendix 5A for full details). In addition, there are items expressing more than one area of control at the same time such as items 62 and 7. This is aligned with the argument that these three areas should be overlapping and interrelated.

To sum up, the four-dimension model analysed in the current investigation shows a complex pattern of learner autonomy. Each dimension seems to be discrete, but it reflects different attributes. An attribute of this dimension can be linked with another attribute, and both of them become included in another dimension. Although the study model seems to be consistent with the one regarding areas of control developed in the literature review, further examination of the model structure indicates overlaps and intertwinements among these areas of control. An autonomous learning behaviour may be associated with more than one area of control, suggesting an interrelated relationship among different areas of control and among different dimensions of the model. In brief, a students’ level of one dimension may be dependent on their level of another dimension. Such differences and similarities are further analysed in the following section.

2. Similarities and differences in learner autonomy perceptions

Drawing on the results in Study One and Study Two of the current investigation, the analysis presented in this section considers the independence and interdependence of one dimension to the other. In addition, similarities and differences in the levels of appreciation that local students hold for each dimension are also investigated. The section finishes with an analysis on the effect of gender, living place, and computer proficiency on each of these four dimensions.

2.1. Similarities and differences among the four dimensions

Analyses on the perceptions of the four dimensions included in the model generally indicate that these dimensions are correlated with one another in both Study One and Study Two (see

______Chapter 5 Page 178 Table 4.21, page 121). Further analyses on students’ self-rating performance of these four dimensions also show similar outcomes. As presented in Table 5.10, students’ performance of one dimension is statistically significantly correlated with their performance of other dimensions (.198 ≤ r ≤ .611, p<.05). That is, students’ performance of one dimension can be used to predict their performance of another dimension. These results suggest that the four dimensions of learner autonomy observed in the current investigation are interrelated. This can be seen as a contextual response to the contemporary argument in the literature which proposes both dependent and independent relationships among the dimensions of learner autonomy (Littlewood, 1999; Sinclair, 2009).

Table 5.10: Correlations among the dimensions of learner autonomy performance

Correlations of Dimensions in Learner Autonomy Performance Pre-test Post-test Dimension of learner autonomy performance N Correlation Sig. N Correlation Sig. Pair 1. Dimension 1 & Dimension 2 147 .218 .008 147 .423 .000 Pair 2. Dimension 1 & Dimension 3 147 .309 .000 147 .352 .000 Pair 3. Dimension 1 & Dimension 4 147 .548 .000 147 .611 .000 Pair 4. Dimension 2 & Dimension 3 147 .198 .016 147 .352 .000 Pair 5. Dimension 2 & Dimension 4 147 .461 .000 147 .570 .000 Pair 6. Dimension 3 & Dimension 4 147 .444 .000 147 .393 .000 Dimension 1: Monitoring learning processes Dimension 3: Using ICTs in learning Dimension 2: Goal-setting & evaluating learning Dimension 4: Initiating learning opportunities

The interrelatedness of the four dimensions of the study model can be explained by the conceptual structure of the construct. Several attributes are shared among the dimensions. Therefore, the perception and performance of one attribute in a dimension can entail the perception and performance of that attribute in another dimension. For example, a behaviour of talking to a peer for evaluative purposes is accounted for evaluating learning, but this behaviour can also be accounted for monitoring learning if it is followed by a behaviour of using the mistakes in that conversation for better learning. In addition, because both cognitive and behavioural aspects, that is, what students think and do respectively, are included in each dimension of the model, it seems to be logical to have a similar pattern of relationship among the four dimensions in both perception and performance scales. Students’ appreciation of learner autonomy seems to be in line with their behaviours of autonomous learning. The interrelatedness of the four dimensions of the learner autonomy model emphasizes the involvement of different attributes in a learning process. As Healy (2007) stated, although a positive learning attitude and an appropriate learning environment can facilitate some learning activity, the actual operation of an autonomous learning behaviour depends a lot on the

______Chapter 5 Page 179 interactions of these factors. In addition, the sustainable maintenance of such a behaviour may be influenced by the level of control that a learner would like to take in a particular social- cultural situation (Benson, 2009). For example, Student 10 in the interview reported that he was interested in joining an English Speaking Club to improve his speaking skills. However, he could not make it regularly because he did not set a plan for it nor set a particular goal for himself.

Despite being all correlated with one another, the four dimensions of the model are consistently perceived at different levels of importance in both Study One and Study Two. As presented in Table 4.8 (page 107) and Table 4.22 (page 122), Goal-setting and evaluating learning is most valued, followed by Initiating learning opportunities, and Monitoring learning. Using ICTs in learning is judged least true of the four dimensions by students. This consistently perceived pattern of the four dimensions provides more insights into the structure of learner autonomy from students’ perspectives. Although the four dimensions are acknowledged to be important, the frequency levels for each dimension to be performed in learning contexts are expected to be different. This is probably why opportunities for choices and critical reflections made Chinese learners confused (Sinclair, 2009), while such practices were favoured in other contexts (Conole et al., 2008; Lockyer et al., 2008; Smith, 2003a).

Variations in students’ levels of appreciation for each dimension also suggest the impact of contextual factors. Students tend to believe that autonomous learners almost always set their learning goals and constantly evaluate their learning. The former attribute may be derived from the Vietnamese cultural tradition of aiming high in education (Cao, 2001). The latter may be influenced by the current exam-oriented system in which assessment is primarily used for ranking and achievement. This is contrary to countries like Finland where assessment is primarily used for school and student development (Finnish National Board of Education, 2010). Similarly, despite being also considered important, using ICTs for learning purposes is least valued because most contemporary teaching and learning practices involve a minimum use of digital technologies. It has been observed that students do not need to actually interact with any ICT-supported platform for their learning (Dang, 2010). Computers and projectors are occasionally used in classroom but only for lecture presentations.

In addition, students’ different levels of appreciation of the two dimensions associated with learning initiative and monitoring reflect the impact of pedagogical practices in the local

______Chapter 5 Page 180 context. Being aware of the passive atmosphere created in the class, students generally appreciate the capacity of looking for new learning opportunities. Behaviours associated with this capacity are even suggested to be performed more regularly than those associated with monitoring learning. It seems that they tend to focus on triggering the learning activities and examining the outcome rather than engaging in the learning process. This suggests that Vietnamese students, like their Asian counterparts, are not culturally interested in critiquing, questioning and negotiating during their learning process (Mulligan & Kirkpatrick, 2000). They tend to maintain harmony with their peers rather than being critical (Hofstede & Hofstede, 2005). This may also suggest that the current teaching approach, which is often used to prepare students for examinations, does not enable students to realize the importance of the learning process. Perhaps this is the main reason for a lack of soft skills such as problem-solving, negotiation, and collaboration that have been discussed seriously in the contemporary context (Hoang, 2008; Thanh, 2008).

In brief, an overall examination of the levels of learner autonomy perceived by EFL students in the current investigation reflects the interactions of students’ psychological attributes and socio-cultural factors. They tend to value the evaluation activities more because the teaching practices do not seem to focus on the learning process and good scores are always expected. This seems to be opposite to students in other parts of the world, where deep-learning has been implemented and students tend to focus on the learning processes and experiences (Entwistle, 2000; Kevin, 2003; Wyatt, 2011). Similarly, appreciation of using technology for learning purposes is the lowest because technology has not been properly integrated into the curriculum yet. In contrast, students’ recognition of learning initiative seems to be opposite to the cultural tradition of learning. They greatly appreciate the ability to look for new learning opportunities. These differences in students’ perceptions in such a changing situation with multi-directional variations of the local context continue to be examined in relation to gender, living place and computer proficiency in the following section.

2.2. Effect of gender, living place, and computer proficiency

Of the three independent variables investigated in the current investigation, gender is identified to have the most salient effect on learner autonomy perception. Females (N=118) particularly perceive the dimension associated with goal-setting and evaluation behaviours at a higher level than their male counterparts (see Table 4.24, page 123 and Table 1, Appendix

______Chapter 5 Page 181 4E). Regardless of age differences, this is consistent with what was reported by Kissau (2006) in the Canadian context. Kissau asked around 500 students learning French as a second language at a secondary school and reported that the females set their learning goals more regularly than the males. Given that Kissau’s study was driven by the literature on motivation and attitudes (Dornyei, 2001, 2003; Gardner, 2001), a close connection can be seen between the concept of learner autonomy and learning motivation as indicated in the current investigation.

Different from gender, living place and self-reported computer proficiency are indicated to only have a very marginal impact on students’ perception of learner autonomy. For example, findings from Study One seem to suggest that students living in different places across Vietnam perceive goal-setting and evaluation activities at different levels. Similarly, students with different levels of self-reported computer proficiency tend to perceive the behaviours of learning initiative at different levels. Nevertheless, such effects seem to be very small, probably because the differences in living places of students in the study sample are not distinct enough to trigger significant effects on students’ perceptions. It may also be that students do not actually experience different technological platforms for their learning initiatives to be able to encounter technical difficulties. These assumptions need to be further investigated for a better understanding.

The suggestion that perceptions of learner autonomy are not greatly mediated by gender, living place and self-reported computer proficiency reinforces the importance of socio- cultural aspects in this construct. Learner autonomy can be non-universal as argued by Pennycook (1997), but differences should only be expected from distinctly socio-cultural contexts because learner autonomy is shaped at the cognitive level. A certain level of variation in living situations does not create significant differences in people’s awareness and understandings of the world on a large scale if the same social structure is applied in those situations. In other words, students in different parts of Vietnam may have different local habits and customs, but their recognition of learning values is not saliently different because they experience the same education system. Only changes in the management policy and social expectation at the national level can produce changes in students’ perception.

In brief, the four dimensions of learner autonomy perceived in the current investigation are consistently related to one another although they are not seen as equally true. That is, the

______Chapter 5 Page 182 acknowledgement of one dimension is dependent on that of the others, but they are not considered of equal importance. This reaffirms the complexity of this multifaceted construct. The behaviours associated with achieving learning scores, like learning evaluation, are most valued, whilst the behaviours associated with learning experience and exploration, such as monitoring and engagement, are less valued. Such evidence seems to indicate a pattern of surface learning which primarily aims at achieving purely score-oriented success. Consequently, quality learning is not properly encouraged (T. V. Nguyen, 2008). This calls for a change in pedagogical practices to guide students into deep learning (Kevin, 2003) for the sustainable development of a quality education system.

The current investigation also suggests that of the three independent variables investigated, only gender has an effect on the perceived levels of one dimension of learner autonomy. The other two, namely living place and self-reported computer proficiency do not trigger any statistically significant difference. This seems to indicate that students of different ethnic backgrounds in the same country still have a similar level of recognition of learner autonomy dimensions. Similarly, a high level of computer proficiency does not entail a higher level of appreciation of using ICTs for learning purposes. Such evidence suggests that the perceptions of learner autonomy are shaped by complex interactions among different factors in a particular context rather than a single factor. In turn, this suggests the value of further investigations into the relationship between students’ perception and performance of learner autonomy in the local context.

3. Relationship between learner autonomy perception and performance

Perception and performance of learner autonomy have rarely been integrated into an investigation. It is understandable that human perceptions and behaviours in ordinary situations should be aligned. Notwithstanding, more insightful knowledge on this relationship can shed better light on research into the roles of other socio-cultural factors that mediate human thoughts and behaviours. More effective methods of learner autonomy promotion can therefore be developed. Prior research studies by Gardner (2009) and Lamb (2009) only focus on the perception of learner autonomy and its changes after some kind of intervention. Alternatively, those such as Hart (2002), Smith (2003a), and Trinh (2005) only focus on the performance of learner autonomy during the implementation of some intervention. To fill this

______Chapter 5 Page 183 gap, the current investigation was designed to understand the relationship between perception and performance of learner autonomy.

Based on the four dimensions of the model identified in the current investigation, the analysis shows that students’ perceptions of learner autonomy are positively correlated with their autonomous learning behaviours (see Table 4.27, page 127). Students tend to do what they think is important for their EFL learning. This seems to show that students’ autonomous learning behaviours are directed by their understandings of learner autonomy. They are able to cognitively react to their learning activities although their behaviours may be different from institutional and socio-cultural expectations. For example, although the use of ICTs for learning purposes has not been provided in the learning context, students appreciate the importance of this learning technology and perform their learning activities in this environment. This partly demonstrates students’ desires and actions for better control of their learning world and opportunities (Pennycook, 1997). However, this also means that students’ uses of ICTs for learning purposes in the local context are contemporarily developed in a more informal fashion.

Analyses of the current investigation show that the relationship between perception and performance of learner autonomy is not strong. The impact that learner autonomy perception has on learner autonomy performance seems to be small. In other words, students’ performance of learner autonomy is suggested to be largely mediated by other factors. This evidence supports the argument that learner autonomy is non-universal (Pennycook, 1997) and contextually constructed (Benson, 2009). Learners cannot always do what they want to do for their learning even though they understand that it is necessary and useful to do so. It has been observed from the interview data that students sometimes fail to perform their learning activities because of travelling problems, part-time job commitments, or friend-related issues. Typical comments reported include having to stay at a friend’s house because of missing a bus home (Student 3), going to work (Student 10), going out with friends (Student 8), or staying on the phone with friends (Student 2).

Such reasons probably contribute to the differences between students’ level of perception and self-rated performance of learner autonomy. At the beginning of the semester, students suggested in the questionnaire that their performance of learner autonomy was generally lower than what was expected from an autonomous learner. This is, of course, to be expected

______Chapter 5 Page 184 from these first year students. The result also enforces the validity of the study as it matches with the interview data randomly collected from the same sample. Students generally report that they sometimes perform different autonomous learning behaviours, and this seems relatively contradictory to the Vietnamese stereotype of passive learners described earlier in the literature. The conclusion is that Vietnamese EFL students’ learning behaviours are changing. Traditionally, they have not been autonomous and independent in their learning, but they are also no longer passive and rote learners.

This conclusion is also supported by the students’ descriptions of their learning initiatives. Their capacity to create learning opportunities is self-rated as equivalent to that of an autonomous learner. This suggests that local students are particularly keen on looking for opportunities for their EFL learning. Pairing this result with their perceptions of learning initiative shows a unified pattern. Students tend to value the behaviours of learning initiatives at the highest level, and actually behave close to that perception. It seems that these students, who choose to study English language and literature for a degree, tend to be more autonomous in this dimension of learner autonomy. Such a conclusion calls for further investigations into the relationship between perception and performance of learner autonomy at the end of the course. They can potentially reflect the effects of the teaching approach and situational constraints that can be used to provide suggestions for pedagogical practices in the local context.

4. Changes in the relationships between perception and performance of learner autonomy over a semester course

At the beginning of the course, students’ perception and performance of the learner autonomy dimension are consistently correlated with each other across the four dimensions (see Table 4.27, page 127). Nonetheless, such a correlation is only found in three of the four dimensions at the end of the course (see Table 4.30, page 130). Students’ perception and performance of the dimension associated with monitoring processes are not correlated with each other at the end of the course. That is, students’ perception of learner autonomy cannot always predict their performance of learner autonomy. Given that the perception of learner autonomy in the current study is only investigated at the beginning of the course, there are two possibilities. First, students’ perception of the dimension associated with monitoring behaviours changes over the course; therefore, their perception at the beginning of the course does not match with

______Chapter 5 Page 185 their behaviours of this dimension at the end of the course. Second, students’ monitoring behaviours at the end of the course are mediated by factors other than their perception of this dimension.

Given that the three teachers in the study did not use any intervention or new teaching approach, it is not very persuasive to suggest that students’ perception of monitoring processes is changed over the course. It is more logical to suggest that students’ performance of monitoring processes is mediated by other factors. This is supported by the very low correlation between the perception and performance of this dimension at the beginning of the course (see Table 4.27, page 127). This argument is also aligned with the learning situation in which students tend to be more relaxed and focus on enriching their learning experience at the beginning of the course. However, it does not seem to happen towards the end of the course because they have to study for the examinations. It is also necessary to note that the dimension associated with monitoring processes is valued at the second lowest level of the four (see Table 4.22, page 122); therefore, it cannot contribute much to shaping students’ learning behaviours, especially in situational circumstances. It seems that students may like to engage in different learning processes at the beginning of the course. However, when the examinations come, they tend to focus more on the learning activities that can help them achieve a better score.

Notwithstanding, such a relationship pattern does not occur with the dimension associated with the use of ICTs for learning purposes. Although this dimension is valued at the lowest level of the four, the correlation between perception and performance of this dimension is maintained at the beginning and the end of the course. Perhaps this is because this correlation is strong (see Table 4.28, page 127), that is, students’ learning behaviours associated with the use of ICTs are strongly linked to their perception of these behaviours. In addition, they have opportunities to work with the LMS during the course, and this may reinforce their recognition of the ICT-supported learning environment. This suggests that teaching practices can either foster or hinder the alignment between perception and performance of learner autonomy.

Such a conclusion is also reflected in the two dimensions associated with goal-setting, evaluating, and initiating learning processes. As students greatly appreciate these two dimensions, and the learning situation seems to encourage them to perform these learning

______Chapter 5 Page 186 behaviours, their performance and perception of these two dimensions are both reinforced and consistently linked with each other. For example, studying in such an examination-oriented system as Vietnam since primary school level, students understand that passing examinations is the only way to demonstrate their capabilities. A university degree is often considered a survival passport to life. Therefore, they always need to make themselves familiar with tests and regularly evaluate their learning progress. They even engage in these activities more frequently when the examinations are coming. The interview data also suggest that students tend to look for more learning materials and opportunities in the hope of scoring a better result in the examinations.

The impact of teaching practices on students’ autonomous learning behaviours is particularly reflected in the students’ self-reported level of using ICTs in learning. At the beginning of the course, students report that their level of using ICTs for learning purposes was significantly lower than what they expected from an autonomous learner (see Tables 4.28 and 4.29, page 128 and 129, respectively). During the course, the LMS is introduced, and students formally have an opportunity to work with an ICT platform. Therefore, at the end of the course, they rate their level of using ICTs for learning purposes as high as what they expected from an autonomous learner (see Tables 4.31, page 131 and Table 2, Appendix 4E). Such an increased engagement in an ICT-supported environment is also reported in Christopher (2006) where students are given an opportunity to work with a software application that enables them to navigate through their Spanish learning activities.

Further analyses on the interview data collected from the current study particularly describe the complexity of the students’ engagement process with the LMS during the course. It seems that students do not just take part in the LMS activities simply because this learning environment is made available to them. Some report that they become involved in a cyclic process of posting messages on their virtual class, reading follow-up comments from peers, and reflecting on their learning. These results are very encouraging and reconfirm the facilitative role of technology in fostering reflective and quality learning as well as engagement, as previously discussed in the literature (Blin, 2004; Kessler & Bikowski, 2010; Schwienhorst, 2008a). Nonetheless, some other students indicate only a partial participation role in the LMS. They go to the LMS occasionally and do not actually interact with their peers (also in Christopher, 2006). The LMS log records even show that the learning activities of students from one class seem to be totally detached from the LMS platform of the course.

______Chapter 5 Page 187 These findings generally suggest that the relationship between perception and performance of learner autonomy cannot always be maintained because of other mediating factors in each learning context. For example, some students in the interviews report that they dislike planning and do not make learning plans, but they certainly do this when the examinations approach. Similarly, Aoki (2001) reports on a Japanese study where a pre-service teacher in a methodology course indicated that she prefers to work alone. However, after working with other peers in a group-work activity, she changed her mind. In another situation of EFL learning in China, students’ autonomous learning is mediated by their cultural traits. They tend to avoid critical comments on their peers’ work, and this tendency affects their learning engagement (Braine, 2003). Investigations into students’ participation in the online space further suggest that motivation and goals may hinder students’ learning behaviours and affect their learning quality (Gitsaki, 2005; Sidhu & Embi, 2009). These indications call for further understandings on the factors that potentially mediate students’ performance of learner autonomy. This is particularly important for research due to the constant change in our contemporary world (Deuze, 2011) in which positive learning behaviours can be easily hindered or fostered (Chen, Quadir, & Teng, 2011; Lew & Schmidt, 2011).

5. Mediating factors of learner autonomy performance

Drawing on different data sources collected within the current investigation, the analyses suggest different mediating factors on EFL students’ performance of learner autonomy. These factors are not always similar between the online and offline learning environments. Gender, teaching effect, preference, motivation, and attitude seem to trigger the most dominant effects on students’ autonomous learning behaviours in the offline learning environment. Meanwhile, technological competence, attitude, and goal orientation tend to be the most dominant mediating factors in the online learning space. Such a difference in the mediating factors of the two learning spaces should be expected, given that learner autonomy is a contextually- biased construct and each learning space is often associated with different characteristics. Further illustrations are presented in the following sections.

5.1. In the offline learning environment

Analyses on students’ performance of learner autonomy in the traditional or offline learning environment indicate different mediating factors. Similar to what has been argued from the

______Chapter 5 Page 188 research literature in Chapter Two, data from the current investigation suggest that learning behaviours of an individual student can be automatically autonomous in one situation but totally inhibited in another situation. This seems to be contradictory to the argument that autonomous learning behaviours can be transferred across similar learning situations (Macaro, 2008). Therefore, it is important to question the adequate level of similarity among the learning situations for autonomous learning behaviours to be transferred. Nevertheless, such different manifestations of learner autonomy in each offline learning context are reported to be likely associated with several personal and cultural attributes of the learners. This contributes to the complicated relationship among the mediating factors of learner autonomy performance as described in the sections which follow.

5.1.1. Effect of gender, computer proficiency, and teaching practice

Gender, computer proficiency, and teaching practice have been argued as having a role in mediating students’ autonomous learning behaviours. Notwithstanding, this role is likely to be small. Female students tend to perform goal-setting and evaluating behaviours during their learning processes at a higher level than their male counterparts (see Table 4.35, page 135). Such a similar pattern is also suggested for students’ perception of this dimension in both Study One and Study Two. This consistency highlights the impact of gender in goal-setting and evaluating learning that has been argued for by Pennycook (1997) but seldom investigated in the area of learner autonomy research. This finding is further illustrated by the stronger performance of female Taiwanese ESL students in reading and interpreting graphic displays than their male counterparts (Cheng, 2011).

Furthermore, computer proficiency triggers different levels of students’ behaviours related to goal setting and learning evaluation (see Table 4.36, page 136). In addition, the relationship between these two variables is suggested to be nonlinear, and this contributes to the complexity of the issue. Generally, computer skills should be irrelevant to students’ offline learning activities because no ICT-related device is involved in this learning environment. Hence, computer competence should be expected to have an indirect effect on the students’ level of goal-setting and evaluating. For example, differences in students’ computer proficiency may reflect variations in students’ socio-economic background and familial expectation. The interaction of these factors may then influence students’ learning behaviours. Such an assumption is more aligned with the contemporary situation of the study context,

______Chapter 5 Page 189 where students from wealthy families tend to have more opportunities to access computers (Dang & Robertson, 2009). Furthermore, when parents buy computers for their children, they often expect their children to learn better.

Different from gender and computer proficiency, teaching practice is not among the foci of the current investigation, but the analysis in Study Two suggests that teaching practice seems to cause differences in the students’ levels of initiating learning (see Table 4.38, page 138). Students in Group C seem to have performed the behaviours associated with learning initiatives at a higher level than those in Group A at the end of the course (see Table 4.39, page 138), although all students have a similar level of this dimension at the beginning of the course. It seems that this difference is due to the learning opportunities that students are given during the course. As students in Group C did not use the LMS as a learning space, they seem to have tried harder than those in the other groups to use the learning opportunities in the offline environment. Nonetheless, the data collected in the current study are insufficient to draw such a conclusion.

Research studies that aim at enhancing the students’ ability to initiate learning activities often create a supportive learning space for conducting learning activities. For example, group and project activities are organized to encourage interactions among students (Lockyer, et al., 2008; Voller, 2005). Thus, discussions, idea exchanges, and reflections are initiated. Similarly, extra curriculum activities such as material collection and music performance are suggested to create more learning opportunities for students (Miller, et al., 2005). However, a general recommendation from these initiatives is that the learning playground needs to be adjusted in accordance with students’ learning preferences. A successful attempt in a particular context is not necessarily applicable to another context. As the current study does not aim to make observations on teaching practice and classroom interaction, an appropriate initiative for better learning opportunities needs to be informed by better documentation on contemporary classroom practice and students’ learning behaviours in the local context.

In brief, despite triggering a small effect, the role of gender and computer proficiency on the performance of learner autonomy needs to be acknowledged. Both of these factors contribute to the students’ performance of goal-setting and learning evaluation activities only. The other three dimensions of the learner autonomy model observed in the current investigation are not mediated by these two factors. In addition, teaching practice seems to contribute to variations

______Chapter 5 Page 190 in the students’ learning initiative activities although this is not included in the study focus. Nevertheless, this assumption emphasizes the importance of pedagogical practice in promoting learner autonomy in a local context. It also calls for further observations on the contemporary classroom interactions and contextual attributes for the formation of an appropriate learner autonomy enhancing approach. Therefore, the effect of the three other mediating factors is further analysed in the following section.

5.1.2. Effects of preference

The interview data collected in the current investigation suggest personal preference is an important factor that can contribute to the shaping of students’ learner autonomy performance. Variations in students’ personal preference can direct any autonomous learning behaviour of the four dimensions into different orientations. Students who are confident about their EFL competence prefer to overcome challenging learning tasks. Therefore, they tend to make their learning goals and plans specific (Students 9, 11, and 1, for example, see page 146-148). It should be expected that specific goals and plans entail better task-oriented learning activities, and these tasks can produce better learning outcomes (Mulligan & Garofalo, 2011). It seems that these students are likely to focus more on the final achievements because they want to conquer difficult assignments during their learning journey. Alternatively, they do it for the sake of course credits (McCarthy, 2010). They want to demonstrate their learning success or just simply complete the course successfully.

In contrast, students who are passionate about EFL learning appear to focus less on planning activities and form general goals. It seems that they enjoy the learning process rather than the final outcome. They perceive EFL learning as part of their lives. Student 6, for example, reported that she learned whatever she liked, and a plan was unnecessary for her learning activities. Although most EFL teaching practices such as communicative language teaching, cooperative teaching, or student-centeredness aim at facilitating interactive and collaborative learning opportunities for students, it is difficult to measure if such process-oriented learning behaviours can produce better quality learning than the outcomes-oriented approach. The former shifts the control to students, and they can directly deal with their learning problems to improve their skills. Notwithstanding, this method may also drive students away from the learning focus because they are given too many alternatives. Similarly, the latter can help students concentrate on their task objective and score well in the examinations, but they may

______Chapter 5 Page 191 be tempted to ignore the benefits from the negotiation and reflection processes for their knowledge construction.

It can be seen that any attempt at promoting learner autonomy may lead to a divergence in the learning pathway. Depending on personal preferences, empowered students may aim at achieving a set of goals expected by the current system or engaging in the learning activities and potentially producing a new set of their own objectives. For example, Students 11 and 1 in the current study aimed to score high in the examinations, while Student 6 aimed to learn what she liked. This is a choice that any educational philosophy needs to make when learner autonomy promoting practices are concerned (Benson, 2009). In the contemporary centralised system of higher education in Vietnam, it seems that a combination of these two approaches is more appropriate because it can help teachers handle large-sized classes with better quality learning outcomes. In addition, it supports the management of quality control and equality across the country because public universities need to work according to a common agenda.

Another aspect of preference that influences students’ performance of learner autonomy is their interaction outlook. Some prefer to carry out their learning activities with someone, while others prefer to learn individually. As indicated in Section 6 of Chapter Two (page 45), autonomous learning behaviours can be promoted in different ways. If students are given adequate assistance within a learning space, they can successfully navigate through the learning processes to develop their EFL competence. They then become responsible for their learning and capable of managing these processes. It does not matter whether they do it individually or interactively. Although it has been argued that learning with a partner should produce better results than learning alone (Lantolf & Thorne, 2006; Wenger, 1999), it may not necessarily apply to learner autonomy promoting practices simply because some students do not like to work in pairs or groups.

Other findings on students’ preference of interactive learning patterns suggest the importance of personality and trust. For example, Students 2 and 4, who favour interactive learning, appear to be sociable. In contrast, Students 5 and 11, who favour individual learning, appear to be reluctant to work with others. Prior research in the context of Japan (Nelson & Carson, 1998) and China (Yang, Badger, & Yu, 2006) emphasizes that students have sceptical perceptions of working with peers. They value their friends and peers’ contribution to their learning, but they do not always think that it is useful. Therefore, it is likely that they tend to

______Chapter 5 Page 192 detach from their peers when given a chance. As they do not trust their friends’ collaboration for their EFL development, they often do the learning task alone, as reported by Student 1 in the interview. This also implies that the method of member allocation in group work is critical. An inappropriate group formation technique can negatively affect the quality of learning engagement and hinder the promotion of learner autonomy.

Apart from individuals with extreme personalities, the mediation of learner autonomy performance is described as being conditional. For example, students would have joined English speaking clubs regularly if they had lived near the clubs (Student 10) or if they had had a friend with whom to go (Student 8). Hence, they sometimes chose to stay at home and work with other materials. These students tend to be flexible in their learning behaviours which are primarily shaped by the contemporary situation. It seems that these students are more capable of managing different learning contexts. Nonetheless, this does not necessarily guarantee optimum success, especially for students with a low level of learner autonomy because the situational factors play a decisive role in their learning behaviours. This also means these students may respond positively to different teaching approaches. They can shift their focus between the engagement process and the final outcome of a learning activity easily as required. They can also adopt an interactive or individual learning behaviour without hesitation for the completion of a learning task.

In brief, personal preference appears to directly trigger different autonomous learning behaviours. Such behaviours may be flexible enough to be adaptive with learning opportunities provided in each learning context. However, these behaviours may be mismatched with certain learning conditions, and this may discourage students from willingly taking a controlling role in their learning process. Therefore, learner autonomy promoting teaching practice needs to take into account students’ preference of learning goals and interaction mode.

5.1.3. Effects of motivation

Motivation is another important mediating factor on the performance of learner autonomy identified in the current study. Findings suggest that there are three sources of motivation which contribute to the shaping of students’ motivation pattern. Specifically, these three sources originate from personal interests, peers’ achievement, and desire for being

______Chapter 5 Page 193 outstanding, respectively. The first appears to be more internal, the second appears to be more external, while the third seems to be associated with both. Gardner (1985) and Ely (1986) term these three types of motivation as intrinsic motivation, extrinsic motivation, and motivation intensity, respectively. These terms are later used in the literature (Carreira, 2011; Dornyei, Csizér, & Németh, 2006; Liu, 2009). Although the driving forces of motivation intensity seem to be a combination of what constitute both intrinsic and extrinsic motivation, reports from students in the current study indicate that these three patterns of motivation tend to trigger different patterns of autonomous learning behaviours.

Students with intrinsic motivation in an area of knowledge, for instance, seem to prefer to work alone with the materials of their interest. They tended to detach from the class activities and did not care much about their classmates’ learning performance, as illustrated by the case of Student 6. She planned, created and engaged in the learning activities that could help her satisfy her thirst for knowledge, regardless of the constraints from the learning environment. Nevertheless, it seems that such students will be able to engage very actively in class activities if they find a match between their learning motivation and the course objectives. To do this, apart from the suggestion by Aoki (2001) to engage students in a learning activity to gradually increase their motivation, the current study suggests developing mutually shared learning content and objectives during the course. Given that the curriculum in the study context is centralised, students’ voices are often ignored in the course material design. Therefore, students’ intrinsic motivation is often not properly negotiated, and this may drive students’ learning behaviours away from the course objectives.

Different from internally motivated students, those who appear to be more motivated by external factors, such as peers’ success, opt for more diverse learning behaviours. As they would like to be better than their classmates, they often form pairs or groups with close friends and initiate learning opportunities. If finding a partner from their class is difficult, they tend to use other channels or even just study alone. This again implies the importance of grouping techniques and activity organization to facilitate students’ learning engagement. It also reflects a shortcoming of the current teaching practices in creating study groups. Notwithstanding, it is important to note that students’ purpose in cooperating with a peer in a learning activity can be very short term and does not necessarily lead to sustainable learning development. For example, Student 10 (see Appendix 4D) expresses a desire to have a partner simply for pronunciation checking. Therefore, learner autonomy practice needs to

______Chapter 5 Page 194 communicate the process of achieving some learning goals rather than just leaving students with a pre-described goal without much guidance.

Although it is generally agreed that motivation can shape students’ learning achievement (Gardner, Moorcroft, & Metford, 1989; Vandergrift, 2005), the current study suggests that this relationship is interactive and reflected via learning behaviours. If students are externally motivated by a classmate’s great pronunciation, as indicated in the interview data with Student 10 for example, they can pair with someone for pronunciation drills. Nonetheless, if they think that their skills are better, they are unlikely to look for a partner and initiate a drill activity. Instead, they may focus on something else and no longer be motivated by their peers’ achievement. This suggests that a certain level of variation in language competence among members in a study group or class can enhance each individual’s learning motivation. Therefore, pedagogical practice needs to facilitate opportunities for students to recognise the competence gap between group members. This can encourage students to continue to evaluate their peers’ performance and their own learning activities for the development of other processes such as initiating and monitoring.

Students motivated by both internal and external factors, termed motivation intensity (Ely, 1986), appear to perform other patterns of autonomous learning behaviours. Similar to those who are primarily internally motivated, students with both internal and external motivation tend to detach from their classmates in their learning behaviours. They seem to like learning alone with their highly personalized plan and in their own way or collaborating with people other than their classmates. However, different from their intrinsically motivated counterparts, these students are always concerned about their peers’ performance and progress. They use their peers’ achievement for the development of their own learning goals. They also seem more committed to their learning plans regardless of social distracters or obstacles.

It can be seen that students’ perceived level of EFL competence in relation to that of their peers is important in predicting their autonomous learning behaviours, and this is applicable for different types of motivation. In such a collectivist society as Vietnam, learners often value the relative performance of an individual in comparison with others within the group rather than in reference to an absolute set of indicators. Therefore, students tend to regulate their learning behaviours according to their friends’ performance. If they think that they are making better progress than their peers, they probably become more motivated in their

______Chapter 5 Page 195 learning (Masgoret & Gardner, 2003). Nevertheless, if they think their peers are more successful in learning achievements, their motivation needs to be fostered by teaching practice. Prior research shows that explicit instructions (Wacho, 2006) and motivational strategies (Guilloteaux & Dornyei, 2008; Papi & Abdollahzadeh, 2011) can increase students’ learning empowerment and motivation. However, this has not been validated in the context of Vietnam.

In brief, the analyses in the current investigation suggest variations in autonomous learning behaviours are shaped by different patterns of motivation. Although the data are insufficient for a proper comparison of these learning behaviours regarding learning quality, the study provides different suggestions for learner autonomy pedagogy. These analyses also show a close relationship between motivation and preference in mediating autonomous learning behaviours. The intrinsic motivation originating from a desire for knowledge is closely related to a preference in an area of knowledge. Similarly, the external factors constituting extrinsic motivation may affect students’ preference for a learning mode. Students can set their learning goals based on their preference and motivation, such as friends’ competence, their knowledge interest, or both. These differences in goal-setting, in turn, trigger a non- interactive or interactive learning pattern, reflecting both their learning preference and motivation orientation.

5.1.4. Effects of attitude

Analyses on students’ reported autonomous learning behaviours suggest the important role played by attitude, as proposed in the literature (Allwright, 1990; Scharle & Szabo, 2000). Students holding a positive attitude towards course textbooks tend to develop their learning behaviours based on the textbook activities. Meanwhile, other students’ learning behaviours are initiated and shaped by people and resources elsewhere. Perhaps this is one of the reasons for the development of various learning attempts that may not necessarily match with the course objectives. As observed in Study Two, students were not specifically guided to reference books. They seemed to simply pick up anything that they thought appropriate, as in the case of Students 6 and 10. On one hand, the absence of a direct instruction from the teacher seemed to give students freedom to choose the learning materials that they preferred, and this is good for their learning achievement. On the other hand, students may be confused

______Chapter 5 Page 196 due to a lack of support, especially when they are not competent to judge the quality of a learning material or activity.

One typical example regarding students’ problems with learning attitude is reflected in their participation in English speaking clubs. As such an activity is organized for a public audience, it is a useful arena in which EFL students can improve their English competence. Nonetheless, it is not always suitable for a student’s EFL development due to a possible mismatch between his/her proficiency level and the target participants of the club. This is probably the reason that Student 10 in Study Two became confused about the benefit of his engagement in this learning opportunity (see page 155). His attitude towards this learning space seems to be positive at the beginning but gradually becomes unclear. It seems that students’ attitude towards a learning activity can be reshaped during their learning process, which changes their autonomous learning behaviours. This conclusion is an addition to the argument that different students can have different attitudes towards the same learning activity (Hart, 2002). In other words, teaching practice can play an important role in changing students’ attitudes towards a learning space by helping them recognize the advantages of taking part in it.

Further analyses also suggest that even when students hold the same attitude towards a learning activity, their autonomous learning behaviours can be performed at different levels. For example, some students only do a spelling and grammatical check when reviewing their assignments, while others tend to focus more on the content although they all have a positive attitude towards this learning activity (see page 155). This suggests either a difference in the level of attitude or the anticipation of other factors. However, an interactive combination between these two alternatives should be more logical, as attitude can be governed by both individual and situational factors (Dornyei, 2003). More noticeably, it can be changed according to experience (Elyildirim & Ashton, 2006). These suggestions again emphasize the complex pattern of mediating factors on the performance of learner autonomy.

Of the factors influencing students’ attitude, the teachers’ instruction appears to be the most significant. It is likely that students in the study context strongly believe their teachers’ advice. Although nine of the eleven interviewees reported that they did not develop their learning activities around the textbook, if their teachers asked them to do an activity in the book, they would definitely do so (see page 156). They seem to automatically modify their

______Chapter 5 Page 197 attitude as instructed by the teachers. Their obedience seems to reflect the cultural and educational philosophy which is deeply rooted in the Vietnamese culture. Of course, there should be the anticipation of a power relationship and course requirement, but students did not seem to perceive any zone for learner autonomy development when teachers are involved. On one hand, this makes the role of teachers even more critical in nurturing learner autonomy because they need to help students understand the possibility of negotiating and debating with them. On the other hand, it is easier for teachers to facilitate the class activities in the way they want. Students may have a negative attitude towards an initiative at first, but they have a chance to modify it because they actually take part in that initiative.

In brief, attitude is suggested to mediate the students’ grounds for learning activity development and their levels of learner autonomy. If they have a positive attitude towards a learning channel or resource, they tend to develop more learning activities around that. This was illustrated by Student 11’s preference in books with practice tests, while Students 1 and 6 were more interested in literature books. Nevertheless, their attitude can be modified, depending on how they perceive the effectiveness of these activities. If they find these activities useful, their attitude is reinforced and they may become more interested in utilizing that learning channel. Otherwise, they may become confused, and their autonomous learning behaviours are more mediated by other factors. Given the socio-cultural distance between students and teachers, it seems that students do not tend to negotiate with teachers. Therefore, it is necessary to facilitate more space culturally and academically for students to be more confident in taking the controlling power from teachers. Notwithstanding, this process needs to be aligned with students’ learning goals because research shows that students may want to be tightly instructed by their teachers to develop their language skills, rather than collaborating with peers in groups for a unit of knowledge (Cheng et al., 2010).

In summary, this section considered the mediating factors of learner autonomy performance in EFL learning in the offline environment. It appears that different patterns of autonomous learning behaviours are triggered by variations in gender, computer proficiency, preference, motivation, and attitude. Although the analyses attempted to separate these factors when examining the performance of learner autonomy to maintain the clarity of the presentation, it is suggested that these factors interact with one another before having an effect on a learner. Such evidence is also aligned with the argument that there is a positive relationship between motivation and attitude (Dornyei, 2003) or an association among preference, belief, and

______Chapter 5 Page 198 attitude (Bullock, 2011). These suggestions call for a large-scale project on learner autonomy promoting pedagogies, particularly in a context where both deep and surface learning orientations co-exist.

The analyses also show that there are different mediating factors, and each factor has different effects on the performance of learner autonomy in the offline learning environment. Thus, it is necessary to consider the mediating factors on this capacity in the online learning space because students currently interact with both online and offline domains almost on a daily basis. Such an investigation becomes more important when the transferability of learner autonomy capacity between different learning environments is questionable. Identifying the mediating factors in both learning spaces can shed light on further research into their interactive effects on students’ performance of learner autonomy. This can also provide pedagogical implications for local practices.

5.2. In the online learning environment

Virtual space is expected to provide a ubiquitous learning environment where students’ learning behaviours can be dynamically illustrated (Jeong & Hong, 2011). Using the LMS as a virtual platform for EFL learning, the current investigation considers variations in students’ performance of learner autonomy. Further analyses show that these variations are triggered by differences in teachers’ class practices and students’ computer proficiency, attitude, and goal orientation. The effects of these mediating factors are specifically presented in the following section.

5.2.1. Effects of teaching practice

Reports from students in the current investigation suggest that their level of using ICTs for learning purposes is not mediated by teaching practice. Although these three groups of students worked with three different teachers, they all employ a similar level of ICT usage for learning purposes (see Table 4.38, page 138). Given that the classroom practice of these three teachers was not observed during the study, no assumptions can be made. At the same time, the students’ participation in the LMS is different across the three groups (see Table 4.41, page 141). The number of postings made by each group varied according to the course

______Chapter 5 Page 199 requirements set by each teacher. The students’ level of engagement in the LMS activities appeared to be linked to the course assessment criteria set by individual teachers. Students’ similarities in using ICTs and their differences in LMS participation seem to reinforce the proposition that students generally have regular access to virtual space. If a course requires them to take part in a certain online platform such as the LMS, they tend to act as required. Otherwise, they tend to look for opportunities in other virtual learning platforms. As a result, their total engagement time in the virtual world seems to be similar, as reported by Students 1, 3, 4, 6, 7, and 11. This suggests that integrating students’ activities in the virtual world into their school activities to enhance learning quality is an appropriate trend in contemporary education in Vietnam as indicated in the higher education reform agenda (Pham, 2010). Such an attempt can also nurture opportunities for learner autonomy development and sustainable learning as students become better at managing their learning activities beyond the classroom.

These findings also indicate the importance of teaching practices and assessment design in shaping the students’ learning platform. Of course, encouraging students to engage in any online learning space is useful for their EFL development. Nevertheless, they should be able to benefit more from such a space if better activities are appropriately designed and tailored for their own level of proficiency and needs. The potential value of such a recommended learning space seems to be promising but also implies a lot of challenges for higher education in Vietnam. Given the current heavy teaching load and the teacher-led practice that is popularly employed, it seems impossible for teachers to handle the additional work arising from the virtual learning space. This means if the sector wants to improve learning quality, the problem of workload, assessment procedures, and professional development needs to be resolved (Harman & Nguyen, 2010).

5.2.2. Effects of gender and computer proficiency

While gender is identified as mediating some of the students’ autonomous learning behaviours in the offline learning context, it does not appear to trigger any difference in students’ performance of learner autonomy in the ICT supported learning space. Both males and females are reported to perceive and perform at a similar level of using ICTs for their learning (see Table 4.12, page 110 and Table 4.34, page 134, respectively). This finding on students’ usage of ICTs for EFL learning seems contradictory to the dominance of males in computer sciences and the intuitive understanding that working with computers is more

______Chapter 5 Page 200 favoured by males. Notwithstanding, it shows a difference between studying computer technology and using computers for learning purposes. Male students may prefer to learn about computer sciences, but it does not necessarily mean that they engage in more online activities for their EFL learning than their female counterparts. Nonetheless, generalizations based on this assumption should be limited to EFL students only, and more gender-balanced samples need to be investigated in further research.

As prior research in learner autonomy has rarely considered the issue of gender differences in the online learning context, the similarity in the students’ level of autonomous learning behaviours in the ICT-supported learning environment is a useful contribution to gender research. In the general context of higher education in Israel, female students are reported to place more postings in the LMS than males (Caspi, Chajut, & Saporta, 2008). However, this result is different from the studies conducted earlier in the United States (Davidson-Shivers, Morris, & Sriwongkol, 2003) and South Africa (Masters & Oberprieler, 2004) in which both males and females are found to have a similar level of engagement in the virtual space of the course. It seems that online communication by gender is influenced by learning interest, cultural background, and the development of technology in a certain context. This implies a cross-cultural study with participants from many disciplines.

Different from gender, self-reported computer proficiency is identified as triggering differences in students’ performance of learner autonomy in the ICT-supported learning space (see Table 4.36, page 136). This is understandable because learners generally need to have achieved a certain level of computer literacy to be able to initiate and engage in this learning environment. It is also aligned with the positive relationship between computer skills and online engagement (Chen, Lambert, & Guidry, 2010), and between computer skills and academic success (Rakap, 2010). Further examination of the mean of students’ learner autonomy performance indicates that such a correlation generally distinguishes those with very poor computer skills from those with better computer skills (see Table 4.37, page 136 and Table 3, Appendix 4E). Students with a very low level of computer proficiency tend to engage less in the online learning space than those with higher levels. In other words, a moderate level of computer skills is enough to take part in the online learning space. Given that students’ computer proficiency can be enhanced during their participation in the online learning space (Robinson & Hullinger, 2008), pedagogical practice needs to nurture students’ achievement both academically and technically.

______Chapter 5 Page 201 Other analyses of students’ LMS failed login records and understanding of their own email addresses also show the impact of limited computer proficiency on students’ learning. While a majority of students understood that different systems are unlikely to use the same login account, many others did not. Therefore, they kept failing when attempting to log on to the LMS and had to seek technical support. This is a cause of delay or even creates a technical barrier for their participation in the LMS. Perhaps this can be explained in that students often use only one favourite channel for online communication, so they automatically type in the same account whenever being asked for one. Similarly, they are not fully aware of the full form of their email addresses as they often access their mailbox from an instant messenger program such as Yahoo! Messenger. Meanwhile, they self-rate their level of computer skills as a little over average (see Table 4.14, page 113). These data appear to suggest a discrepancy between students’ actual technology skills and their self-rated level of technology competence. They tend to think that they know how to use it because they use it often. Nevertheless, they might not understand the concepts behind it (Dang & Robertson, 2010a).

The main reason for the discrepancy between students’ perceived and actual competence in ICT may come from the local context where students generally achieve internet skills through informal learning (Dang & Robertson, 2010b). Internet technology is popularly available at home and in internet cafes, but it is not employed much in school activities for learning purposes. Hence, it is suggested that basic technical understanding is not necessarily acquired even when frequent self-access to the Internet and cyber communication is achieved. It also seems to suggest that these students may not need a certain period of training to be familiar with the technology adopted because they have been using it, as presented in Xie (2002), for instance. Notwithstanding, some critical sessions on technology usage with explicit explanations of the concepts behind this would be more suitable for them.

In addition, it seems necessary to note that regardless of technical competence, students tend to like being connected, and being instantly and conveniently in touch with others when working online. Although mutual visibility is not obtained in the virtual communication modalities examined in this investigation, interlocutors often expect prompt responses once they start an online conversation. If their expectation is not met, the communication process breaks down, and an alternative means is unlikely to be sought for compensation (Dang and Robertson, 2010a). This breakdown in communication transactions should be expected to happen regularly as many people are reported to prefer being connected only. They do not

______Chapter 5 Page 202 want to respond promptly even when they can (Baron et al., 2005; Jacobs, 2008). Thus, the students participating in the current research often targeted other interlocutors or references instead. In addition, technical difficulties may result in hesitation in using certain types of technologies even though the desire to be connected is strong.

The mediating impact of computer proficiency and access on the performance of learner autonomy continued to be reflected in the interview data regardless of the study procedure on technical support. Students 11 and 5, both living away from home, reported that they had technical difficulties with and limited access to a computer. Nonetheless, they reacted differently to the LMS activities. Student 5 only encountered technical problems at the beginning of the course. She then quickly learned how to use the LMS and participated in the virtual class regularly even though she had to go to a computer service. In contrast, Student 11 indicated that his computer skills had not improved and he had the same problems till the end of the course although he went home almost every weekend and had access to a home computer. These two specific examples provide more insights into the argument that students should be able to improve their ICT skills during the course (Lin, 2004; Robinson & Hullinger, 2008) and that ICT skills should not be an obstacle for this generation of students (Yang, 2001).

In brief, while gender differences do not have any effect on students’ performance of learner autonomy in the online learning environment, computer proficiency appears to mediate their initiation and engagement in this learning space. Difficulties with login and email communication were indicated to be some of the most prominent hurdles because some students did not understand some basic technical concepts regardless of their regular use of computers. More detailed examples also show that engaging in the virtual section of the course cannot always minimize students’ technical problems. If they are really interested in the online activities, they tend to overcome these hurdles. For example, Student 5 reported that using the LMS became easy for her after a few weeks although she did not know much about websites and the Internet at the beginning of the course (refer to page 159). However, if the online activities are not interesting enough to motivate students to learn more about computers, their ICT competence is unlikely to improve and they tend not to take part in this learning space much. These findings imply a pedagogical integration of technical and academic issues in fostering students’ learner autonomy in the online learning space.

______Chapter 5 Page 203 5.2.3. Effects of attitude

Students’ differences in trying to overcome technical difficulties when participating in the LMS seem to be associated with differences in their attitude towards this learning environment. This assumption is supported by the interview data which indicated a positive relationship between students’ attitude towards the LMS and their participation patterns. If students recognize the benefits of the LMS to their learning, they are more active in that environment as indicated by Belz (2003). In addition, the more they interact with their peers in the LMS, the greater sense of belonging they have with it (Kessler & Bikowski, 2010). Of course, the quality of engagement is also essential in mediating these relationships. Importantly, these responses reflect the diversification of Vietnamese students’ online participation styles which were different from their traditional stereotype such as their culturally working alone preference (Dixon, Dixon, & Siragusa, 2007; Tharp, 1989) or passive learning only.

Students’ significant use of Global blog and very little use of Chat room indicate their attitude towards online communication. It seems that a larger audience with opportunities for more community ties can have a positive influence on students’ attitude. Although these two sections were not included in the course evaluation, the students’ participation in these two spaces was completely opposite. Blog entries about topics other than those prescribed in the course seemed to trigger a lot of voluntary communicative exchanges, raising the issue of community building and learning opportunities. The growth of this spontaneous community was probably fostered by friendly topics and the supervision-free atmosphere. Similarly, as the content in the synchronous Chat room was only spread to those concurrently in the room, and it was not archived, only a very small audience could access that content. Therefore, this tool was almost completely ignored by the students. It seems that a positive attitude is initially important for participation in this learning space, and topics of interest can maintain the engagement process.

Students’ attitude towards the online space was also reflected in their careful consideration of the message content that they posted on the LMS (see page 144). As they were fully aware that their personality and capability demonstrated in their online messages was probably being evaluated by some audience, they tried to minimize their mistakes, demonstrate their competence and maintain a cheerful community atmosphere to promote their identity (refer to

______Chapter 5 Page 204 page 144. They even want to be recognized by the community, which is similar to what Mortensen (2008) acknowledged in her blogging experience. This indicates that the local students consider the virtual learning environment as formal as their traditional offline class (Dang & Robertson, 2010a). In addition, they use the anonymity feature of online space to conform to the culture of cyber world rather than taking advantage of this for rebellious or destructive purposes. In other words, students tended to build up a positive and unique identity or netizenship (as in Burgess, 2006; Robertson, Webb, & Fluck, 2007) that potentially distinguished them from the others, without damaging the community.

The reflection of attitude on students’ autonomous learning behaviours becomes complicated when comparing their engagement in the LMS class section with the LMS public section. In principle, virtual class participation is basically to fulfil the course requirements, while the LMS public section is more about sharing feelings and demonstrating personalities. Therefore, it is expected that students would enjoy and engage in the relaxing public space in the public section rather than the formal assessed class space. Nevertheless, it did not happen in such a manner. Some students became interested in their online class during the course and engage more intensively in that small community. They found it easier and had more confidence to demonstrate their capabilities (Kern, 1995; Lam, 2004) academically and socially. This participation trend in the virtual environment indicates that larger and informal communities seem to positively affect students’ attitude, but a level of sustainable engagement is associated with the participants’ sense of belonging which can gradually be built up.

In brief, the analyses showed that students in the study context have very different attitudes towards the online learning space. Some perceive it as a serious learning environment or a compulsory set of activities to which they have to comply. Some consider it a relaxing and enjoyable space for out-of-class learning activities. Some others even see it as an extra or unnecessary learning space in which they do not wish to take part. These attitude patterns can result in different levels of students’ engagement in the LMS. Notwithstanding, these patterns can be changed during their participation in an online community from which they can benefit academically, socially, or even both. When a mixed attitude is developed, students may attempt to include informal content in their online learning space. Such participation behaviour potentially facilitates the learning process in the virtual class and challenges the serious atmosphere in the traditional offline class where only formal content is allowed.

______Chapter 5 Page 205 5.2.4. Effects of goal orientation

While students’ attitudes seem to be important for their initiating learning behaviours in the ICT-supported learning space, the quality of their interactions with this environment is likely characterized by the participating roles that they adopt (Dang & Robertson, 2010c). It appears that task-oriented participants took part in the LMS at a peripheral level, given that the task provided in this study only counted the number of postings, not the quality of the postings. Meanwhile, content- and community-oriented participants seemed to engage extensively in and even tried to control the virtual environment. While Farmer (2006) indicates that the LMS could force the participants to work on “shared communication spaces, rather than on the individuals” (p. 95), this study suggests that it depends on the participants’ learning goals. Content-oriented participants are interested in shared spaces, while community-oriented participants focus more on interacting with fewer individuals with whom they have good relationships.

Students’ adopted participation roles can also be among the indicators for their development patterns of attitude towards and engagement in the LMS (Dang & Robertson, 2010c, 2010d). For example, those whose adopted goals are either content-oriented or community-oriented tend to move from peripheral to integral levels of participation because more interesting content was produced and better community ties were developed as the course went on. Once their participation level increased, they found themselves more associated with the environment and developed more positive attitudes towards that space. It also means that they had more opportunities to interact with appropriate interlocutors (Kurata, 2007) for their learning purposes. This observation raises pedagogic considerations on the activity design, not just on the learning styles as proposed in previous studies (Dixon et al., 2007; Stepp-Greany, 2002), to effectively engage task-oriented participants in the virtual learning space.

Differences in students’ goal orientation are also reflected in their different expectations to the degree of teachers’ comments and participation in the LMS. An analysis of the interview data shows that that those who are more task-oriented (such as Student 4, refer to page 161) do not seem to expect a lot of comments from the teachers. Meanwhile, those who are more content- oriented and community-oriented (such as Students 2 and 10, refer to page 162) tend to prefer to receive more comments from teachers. These preferences are probably because task- oriented participants do not often prepare their work carefully, while community-oriented

______Chapter 5 Page 206 participants are interested in the quality of their postings and opportunities for community interactions. Nonetheless, if the requirement of the online task is more about the quality of the postings, not just the quantity as in this study, task-oriented participants may probably have to expect more comments from teachers to improve their score. These assumptions therefore provide suggestions for further research in task design and students’ goals.

The relationship among the three patterns of participation driven by different goal orientations is complicated. The interview data show that students often report characteristics consistent with more than one pattern of participation. For example, Student 11 indicated that he was a task-oriented and a community-oriented participant, whilst Student 4 was content-oriented and community-oriented. The relationship among these adopted/shifted roles also depends considerably on the quality of each goal orientation. For example, as indicated by Student 6, a good posting was able to draw attention from different people, and enabled comments to be added. Similarly, when a student came across a thread with many comments, he/she expected that the posting would be interesting (e.g., Student 10). However, these assumptions are not always accurate as a lot of students responded to a thread simply because of their relationship with the author. These preliminary findings propose suggestions for further investigations on students’ role changes in online learning environments.

In summary, the mediating impact of several factors on learner autonomy performance is considered in relation to the socio-cultural characteristics of the study context. Gender, computer proficiency, preference, attitude, motivation, goal orientation, and teaching practice are identified to have different effects on different dimensions of the four-dimension model of learner autonomy observed in the current investigation. Although these factors are analysed separately, it is suggested that they all interact with one another before generating an effect on a pattern of autonomous learning behaviours. It is also suggested that several personal, situational, and cultural attributes may contribute to the shaping of these mediating factors. This indicates an interactive effect of personal psychology and learning environments on the performance of learner autonomy, confirming the proposition of the theoretical position employed in the current investigation.

Further examination indicates that these mediating factors can be positively or negatively developed during students’ performance of learner autonomy. When a learning practice is adopted in an environment with certain resources, if students can technically navigate the

______Chapter 5 Page 207 space and successfully build up their identity to collaborate with other peers, they tend to engage in and control their learning processes better. Such positive outcomes are likely to, in turn, enhance their navigation skills, attitude, preference, motivation and so on. In contrast, if students employ an autonomous learning behaviour in a learning space and fail to receive benefits socially or academically, their learning motivation and attitude towards that space are hindered. As a result, they have to initiate their learning processes in another environment or adopt a different strategy. This acknowledges the multidirectional interactions among identities, resources, and practices in constructing, transforming, and developing autonomous learning behaviours.

6. Conclusion

This chapter analyses the findings of the current investigation at the local and global level. These results are compared and contrasted with previous research to suggest the study contributions. The learner autonomy model observed in the study adds two new dimensions to the two models suggested earlier for the context of EFL learning in Vietnam, constituting a model with four dimensions (see Figure 5.1). The attributes of these dimensions conform to a number of learner autonomy attributes described in different research contexts. Both cognitive and behavioural aspects are identified in each dimension of the model. Although these dimensions appear to overlap with one another, each of them is perceived at different levels. The dimension associated with goal-setting and evaluating activities is more valued than the one associated with monitoring learning activities. It seems that students are more interested in doing practice tests for better course credits than enjoying the learning process exploration, given that education in contemporary Vietnam is examination-oriented.

D1: Monitoring learning processes D1 D3 D2: Goal-setting and evaluating learning D3: Using ICTs in learning D4: Initiating learning opportunities

D4 D3

Learner Autonomy

Figure 5.1: Four dimensions of learner autonomy

______Chapter 5 Page 208 Further investigations into students’ perceptions of learning initiatives and using ICTs for learning purposes suggested another aspect of Vietnamese EFL students. Although being generally exposed to teacher-centred approaches since primary school, these students show a relatively high level of interest in creating learning opportunities. They do not expect to be spoon-fed as traditionally known. They also perceive the importance of employing technology to support their learning although this is not required in their course. This indicates a lag between school and social life regarding technology integration. This finding urges curriculum innovations to use available technology resources for learner autonomy development.

An examination of the perception and performance of learner autonomy generally shows a positive relationship between the two. Students tend to do what they think is best for their learning. However, this relationship appears to be influenced by situational constraints. Students’ performance of learner autonomy may be different at the beginning and the end of the course. This leads to an examination of the mediating factors of learner autonomy performance. As a result, variations in gender, computer proficiency, attitude, preference, motivation, and goal orientation are suggested to trigger different behaviours and levels of learner autonomy dimensions. The analyses also indicate that these mediating factors seem to interact with one another in fostering or inhibiting the performance of an autonomous learning behaviour. In addition, the process of implementing a learning behaviour to engage in a learning activity and receiving some outcome can, in turn, reshape the state of the different mediating factors. Figure 5.2 visually presents the relationship between mediating factors and learner autonomy.

Goal orientation

Gender LEARNER Attitude

AUTONOMY Preference Motivation

Computer proficiency

: Interactive influence

Figure 5.2: Mediating factors of learner autonomy

______Chapter 5 Page 209 7. Summary

The discussion presented in this chapter further analyses the research findings reported in Chapter Four. The four-dimension model of learner autonomy suggested by responses from the study participants were paired with other models indicated in prior research. Based on these four dimensions, students’ performance of learner autonomy in both online and offline learning environments was investigated. As what the students thought about learner autonomy did not always seem to be aligned with their practice of learner autonomy, the study proposed and examined the roles of mediating factors. Further analyses suggested that these factors were associated with personal, situational, and social attributes. These findings on the construct and manifestation of learner autonomy imply several pedagogical practices to foster this capacity in the study context. A summary of the study with key contributions and implications is considered in the final chapter in relation to the objectives of facilitating long- life learning and sustainable education set out by the higher education reform agenda in Vietnam.

______Chapter 5 Page 210

Chapter Six

CONCLUSIONS AND IMPLICATIONS

______

1. Summary

The study reported in this thesis investigated learner autonomy as an important capacity for the operation of learning processes. The research was motivated by a concern that there was insufficient practice of a student-centred approach in the context of Vietnamese education and the resulting impact on students’ learning behaviours. In addition, there was a lack of integration of new technologies into tertiary curricula, which lagged behind the rapid rise of networked computer use in contemporary Vietnam. The investigation examined in particular EFL undergraduate students’ perception and performance of learner autonomy. The factors that mediated students’ performance of learner autonomy were also taken into account. Participants of this mixed-methods designed study included 562 EFL students from four universities across Vietnam in the first phase (Study One) and a cohort of 247 first-year students in the second phase (Study Two). Research instruments were developed from contemporary research in learner autonomy with consideration of the Vietnamese context. Data were generated from questionnaire surveys and individual semi-structured interviews.

A procedure was developed and conducted to analyse the study data. Regarding the survey data collected from students’ perception of learner autonomy, a multi-phased exploratory factor analysis was employed together with an expert validation process. This method allowed the study to integrate reliability and validity tests into the same procedure. After establishing the reliability and validity of the instrument, the analysis identified four patterns of learner autonomy in the sample responses. These appeared best described as Monitoring learning processes, Goal-setting and evaluating learning, Using ICTs in learning, and Initiating learning opportunities. These dimensions accounted for about 50 percent of the total variance explained in the factor analysis (see page 103). Thus, these four factors were taken to be a good representation in the Vietnamese context, given that learner autonomy is a

______Chapter 6 Page 211 multidimensional construct. However, it needs to be stated that there might be other dimensions which were not included in this analysis, as several aspects of this construct might not be either observable or measurable (Benson, 2010).

Further analyses showed that students’ perception of one learner autonomy dimension was dependent on their perception of other dimensions. In other words, these four dimensions were interrelated. Furthermore, learning activities associated with goal-setting and evaluation appeared to be the most valued, while those associated with the use of ICTs for learning purposes seemed to be the least appreciated. These findings are likely to reflect the contemporary situation of learning in Vietnam, where students tend to be driven by an ultimate score rather than the learning process experience. It also acknowledges the importance of socio-cultural influence on shaping learner autonomy. Other analyses indicated that female students valued goal-setting and evaluation activities more than their male counterparts. Socio-economic differences in students’ places of residence did not appear to produce variations in students’ levels of learner autonomy perception.

The current investigation suggested an interrelated relationship between perception and performance of learner autonomy. If students perceived an autonomous learning behaviour as important, they tended to implement it in their learning process. This seems understandable as human’s actions are supposed to be guided by their thinking. However, this relationship is not very strong when learner autonomy is concerned. Statistical analyses of the current research showed that the probability of using students’ perception to predict their level of learner autonomy performance was small. Important factors from the immediate context such as examinations and other life commitments could override students’ autonomous learning behaviours. These findings probably contribute to the argument that learner autonomy is non- universal (Pennycook, 1997), contextually shaped (Benson, 2009) and may have a mask (Benson, 2010). Students cannot sometimes do what they want because of constraints and expectations from their community. They may also perform an activity simply to satisfy some stakeholder.

Investigations into the mediating factors of students’ performance of learner autonomy indicated both personal and situational attributes. Gender appeared to exert some limited influence on a dimension of learner autonomy. In addition, preference, motivation, and attitude were reported to prominently contribute to the shaping of students’ autonomous

______Chapter 6 Page 212 learning behaviours in the offline leaning context. For example, students who were intrinsically motivated by a particular area of knowledge tended to detach from the class activities. On the contrary, those who preferred to be interactive seemed to look for collaboration with peers, while others appeared to be interested in working alone. Nevertheless, these factors perhaps interacted with one another in a certain context to have a certain effect on students’ autonomous learning behaviours.

In the online learning space, students’ autonomous learning behaviours seemed to be mediated by their technological competence, learning attitude, and goal orientation. Students with poor computer skills tended to detach from the ICT-supported learning environment, while those who were computer proficient tended to engage in the virtual learning space. Similarly, students with positive attitudes towards this learning space tended to look for learning opportunities in this environment while others did not. Furthermore, those who considered the course LMS a serious learning space expected detailed comments from others in the course and wanted their postings to be specifically evaluated. Meanwhile, those who considered this environment to be a social venue did not want their online exchanges monitored and their postings explicitly assessed by teachers. In a similar sense, students whose goals were oriented in task completion, knowledge development, and community building tended to engage superficially, interactively, and socially in the online learning space respectively.

In brief, the current investigation considered EFL undergraduate students’ learner autonomy in the contemporary context of Vietnam higher education. Four dimensions of learner autonomy were identified, and relationships among them were investigated. Different mediating factors of learner autonomy performance at both personal and situational levels were also proposed. The study findings reflect different socio-cultural characteristics of the context and provide several contributions to the area of learner autonomy. These contributions are specifically presented below.

2. Theoretical contributions

The four-dimension model of learner autonomy identified and validated in the current investigation can be considered a reference for research on learner autonomy in EFL education. Analyses on the differences and similarities between this model and others in prior

______Chapter 6 Page 213 research suggest that learner autonomy is a developmental process (Benson, 2010) and can be changed according to socio-cultural and situational conditions (Smith & Ushioda, 2009). However, this capacity seems to be strongly recognized by students even when it is not deliberately promoted by the study context. Therefore, although the current research was conducted in Vietnam, the model should be useful for studies conducted in other contexts, particularly those with similar socio-cultural characteristics.

Furthermore, as both cognitive and behavioural aspects appear to be integrated in each dimension, the study suggests that both aspects need to be included in any research on learner autonomy for a better understanding. It is likely that the exercise of learner autonomy requires learners to perform learning activities and to think of them critically. Learners constantly need to reflect on what they are doing and look for better alternatives. To effectively maintain learners’ control of their action, this process needs to be continuous. These two aspects should also be supplementary to one another as the outcome of one aspect is arguably served as a basis for the development of the other. When learners think that a learning behaviour is good, they potentially do it. Then, when they actually do it, they can know if that behaviour is good and whether they should keep doing it.

The current investigation into students’ perceptions of learner autonomy also reflects an integration of multiple perspectives on the concept of this construct. The importance of the learning situation indicated by technical perspective such as resource availability and virtual learning space is reported to trigger variations in students’ learning behaviours. In addition, the impact of personal attributes from the psychological perspective such as intrinsic motivation and learning preference is identified to drive students to adopt interactive or individual learning patterns. Negotiations between students and their learning environment for the performance of a learning activity are also evidenced in the study data. Students tend to consider different factors in a certain learning context to come up with an appropriate learning activity. It seems that learner autonomy is perceived by an intertwinement among different perspectives rather than a single perspective.

3. Methodological implications

As suggested in the literature review (see page 28), there have been multiple perspectives on learner autonomy. This suggests that using multiple data sets and employing different data

______Chapter 6 Page 214 collection methods in a single study is necessary. Students’ autonomous learning behaviours may be motivated and shaped by immediate enablements and constraints in that specific learning situation. Such behaviours may not necessarily be performed in the same tradition when that learning situation is replicated. A student who may look autonomous in a classroom context is not necessarily considered an autonomous learner in other contexts. Hence, students’ capacity for learner autonomy needs to be observed and documented in different learning environments. Related observations can use different methods as certain aspects of learner autonomy capacity are not measurable or observable (Benson, 2010). A mixed- methods design with appropriate integration of data analyses employed in the current investigation should be a useful reference for other studies.

In this regard, research in the area of learner autonomy is very limited in Vietnam. The questionnaire instrument developed and validated in the current study is a useful reference for further research in the local context. Items in the questionnaire may also be modified and used in other similar learning contexts such as China, Indonesia, and Malaysia, where the education system is centralised. Similarly, this may apply to multiple disciplines. These items can be revised to reflect learning behaviours in other fields such as mathematics and social sciences. Then, students’ autonomous learning behaviours in these fields can be investigated with the same approach.

In constructing an effective virtual learning environment, the current study confirms the usefulness of a Web 2.0 platform, built around the seven principles of Chickering and Gamson (1987). These principles are reported to be applicable to the study context although they were initiated a while ago. Structured opportunities for interactive communication, collaborative activities, negotiation, and feedback continue to be acknowledged for better learning and different talents. The connection between school and home activities created by the online platform can enrich both the learning and social space of students (compared with Kent & Facer, 2004). This environment appears to be promising for the promotion of learner autonomy. These potentially contribute to the conceptualization of an effective online community for academic purposes and challenge the complex roles of teachers as facilitators in technology enhanced environments.

______Chapter 6 Page 215 4. Pedagogical suggestions

The findings reported in the current research highlight EFL students’ acknowledgement of learner autonomy and the socio-cultural influence of the context. This is aligned with prior research in Vietnam, China, and Japan (e.g., Nguyen, 2009; Sinclair, 2009; Yang, 2007). Given the dominance of textbook-based, teacher-fronted practice (Thomas & Inkson, 2003; Wang, 2008), and the importance of test scores (Le, 2000; Warden & Lin, 2000), students seem to become demotivated (Tran & Baldauf, 2007) and passive. To some extent, the impact of Confucius ideology in these contexts also drives students to obedience and teacher dependence (Pham, 2010; Sao & Wu, 2007). Teachers are reported not to be ready for the promotion of learner autonomy (Nakata, 2011). Nevertheless, similar to Wang (2011), the current study shows that students want to control their learning processes. These socio- cultural and ideological conditions suggest a holistic pedagogical approach for learner autonomy to be promoted properly.

A strategic policy needs to be developed for the facilitation of learner autonomy at the classroom level. Curriculum negotiation should be allowed at the university level, and assessment measures should take into account the importance of the learning experience. The policy should be open enough for teachers to conduct learning activities for quality learning even when such a method conflicts with the centralised policy (Rivers, 2011). Teachers will no longer worry about how to complete an assigned textbook in a given time frame (Le & Barnard, 2009b). They also have opportunities to nurture students’ engagement in learning activities and encourage learning enquiries. Once achieving a high score is not the ultimate aim, students will probably agree to participate more actively in learning for knowledge development. Learner autonomy is therefore fostered, knowledge production becomes of greater concerned, and diverse success can be achieved. Creative and life-long learners can also be expected (Hozayen, 2011).

As students’ performance of learner autonomy can be mediated by personal, situational, and social factors, teachers need to allow opportunities for students to tailor learning activities to suit their own preferences. Of course, this learning approach needs to be negotiated with students. They can be encouraged to use their prior learning experience and personal strategies flexibly during learning processes. They also have opportunities to reflect on their performance and modify their behaviours, if necessary, to obtain better outcomes (Dam &

______Chapter 6 Page 216 Legenhausen, 2010). Hence, when carrying out a learning activity, they are guided to target linguistic objectives (Lamb, 2010) and enhance their learner autonomy capacity. To design these activities properly, it is necessary to offer teachers professional development opportunities, as such an approach is relatively new in the context of Vietnam and most pre- service teacher training programs do not significantly address this issue (Reinders & Balcikanli, 2011).

An attempt to employ ICTs in an educational context such as in the current study might also lead to a different pedagogy of which both teachers and students are unfamiliar (Dang & Robertson, 2010a). Students are given more power and control over their learning processes, and teachers are expected to be ubiquitous and prompt facilitators. Hence, students need to be informed in order to express themselves more confidently in this environment; and teachers need to be prepared to facilitate a friendly learning environment, offer necessary incentives, and regularly monitor activities to increase learning engagement. In addition, some students may concurrently maintain more than one social network, and others may not be interested in the course platform very much. It is not always easy to predict how students use this channel for their learning purposes (Ushioda, et al., 2011). Thus, it is important to use appropriate strategies to nurture the formation of an online community for the course.

In the online context, it can be expected that course community will develop. This needs to be maintained effectively and take into account students’ preferred participation styles (Dang & Robertson, 2010c, 2010d). Provided with the three engagement patterns reported in the current study, namely task-oriented, content-oriented and community-oriented, different forms of facilitation need to be created for effective interactions. For example, online tasks need to be designed in such a way that their completion requires a certain level of student interaction. General guidelines for a posting to be counted for evaluation need to be provided and possibly negotiated with students at the beginning of the course. In addition, the relationship among online community members has significant impact on the quality of the online learning process. Therefore, taking advantage of students’ offline social ties (Haythornthwaite, 2005) to develop the online community and using their online relationships to promote new offline connections helps to enrich the learning community. The level of teachers’ comments in the online environment is particularly critical as it can either enhance or inhibit further interactions. Thus, there is a need for teachers to pay attention to students’ individual differences in making online comments.

______Chapter 6 Page 217 5. Limitations

Although the current study was carefully designed, and associated issues were properly considered, it could not avoid some limitations. First, the survey items and interview questions employed in the study were basically developed from studies in contexts other than Vietnam. Some were initially developed in Asian contexts, while others were conducted in Western education situations. This is primarily due to the fact that the reported studies in the local context are not adequate for the current study’s instruments. Being aware that the instrument origins can be critical for investigations of such a socially mediated construct as learner autonomy, the study performed a legitimate procedure to maximize the level of validity and reliability. The process of the instrument development places significant considerations on the socio-cultural characteristics of Vietnam.

Second, it is the scope of the investigation that allows it to limitedly consider several other factors that potentially mediate the examined variables. These include individual learning path, home computer access, parental support, participation in activities other than those in the course, teaching styles and the degree of formal teaching about learner autonomy that the participants received prior to the study. This does not mean that the study makes light of these associated factors. It only aims to focus more on the general patterns of learner autonomy and attempts to illustrate them by examples extracted from students’ course-related activities. Any possible impact of these associated variables on perceptual and behavioural patterns of learner autonomy is acknowledged and suggested for further research.

Third, the study restricts itself to young adolescents learning EFL in Vietnamese public universities only. Their English proficiency ranges from upper intermediate to pre-advanced levels. It excludes students of other levels of English proficiency, in other types of universities, and in other age ranges. In addition, the participating students are located in four cities in Vietnam where life styles and cultural customs are supposed to be varied. To obtain relatively comparable groups of students, the study sample only draws on those from large cities. As a result, the actual quantified values of students’ ratings are only valid for comparative purposes within the study. The extent to which the study findings are generalised should also be confined to this type of participant only. Other follow-up studies focusing on provincial students will be necessary for a more comprehensive scenario on how learner autonomy is perceived and exercised in Vietnam.

______Chapter 6 Page 218 Fourth, the investigation of patterns of learner autonomy is conducted from the students’ perspective only, not from the perspectives of lecturers, school managers, and other stakeholders. Teaching practices in classrooms and other interrelated constraints are acknowledged, used as part of the situational descriptions and proposed for further research. Hence, it can be argued that the research produces a superficial examination of such a complicated construct as learner autonomy. However, the study itself does not aim to explicitly investigate issues from the school management board and policy making authorities. Instead of looking at perspectives from both students and teachers, the study selects the former because it lays more importance on how students perceive a set of learning activities than how teachers conduct them. In addition, examining students’ learning experience and roles can provide descriptive behaviours of teachers because they interact with each other during the course of each activity. Feedback from the participating teachers also indicates that they did not have any major problem during the course delivery.

Fifth, the investigation of the development of students’ learner autonomy was conducted within one semester only. This time length can be criticised as it may take a longer period of time for learner autonomy to change significantly in a certain context. Most prior research also shares this problem. A few other longitudinal studies provide descriptive changes other than statistical changes on students’ learner autonomy. However, these analyses are not supported by any level indicator index. In other words, there has been no empirical evidence that addresses the amount of estimated time necessary for a meaningful change of learner autonomy capacity.

6. Directions for further research

The model of learner autonomy reported in the current study includes four dimensions, and each dimension is appreciated at a different level. As learner autonomy is socially-situated, it is supposed that the model may be changed when situational conditions are different. There should be other dimensions too. Thus, more research needs to be conducted to understand how the contemporary model may change in the context of Vietnam when socio-cultural and ideological conditions change. Such investigations are important for Vietnamese education, particularly when the nation is implementing different strategic plans and reforms in both economy and education. This research direction is also necessary for any learning context in the world. Differences in the model of learner autonomy across different countries can

______Chapter 6 Page 219 demonstrate how the learning context mediates students’ autonomous learning behaviours. This can significantly contribute to the agenda of promoting learner autonomy.

The complex relationship among dimensions of learner autonomy also needs further research. As one dimension likely interacts with the others, the performance of one dimension can be inhibited or nurtured by others. Therefore, understanding the operation of the dimensions can help educational stakeholders design appropriate pedagogical suggestions or guidelines for a specific context. Similarly, the pattern of mediating factors on learner autonomy is complicated. Factors such as motivation, attitude, and preference seem to be interrelated with each other. In addition, each of them probably has a different impact on each dimension of learner autonomy, and a combination of these factors may even trigger a more complex pattern of impact. These complexities should be potential proposals for future studies.

Further investigations also need to identify the influence of other mediating factors that have not been demonstrated in the current sample. For example, students’ learning belief and their parents’ expectation may be important drivers for the shaping of students’ learning behaviours. Prior research has shown that students simply revert to aiming at course credits even when learner autonomy enhancing practice such as project-based activities are employed (McCarthy, 2010). Meanwhile, other studies suggest that students’ belief and attitudes can be positively developed when they actually have a chance to work in the new learning environment (Al-Jarf, 2007; Chen, 2003; Felix, 2001; Lin, 2004). It seems that learning behaviours and beliefs interact with one another during the implementation of a pedagogical practice. A positive belief or attitude is unlikely to be as important as how a learner autonomy enhancing practice is applied.

Further considerations should also be placed on the relationship between online and offline learning behaviours. Those who are more silent in offline class may be more active in online space and vice versa (c.f., Burgess, 2006). However, in many cases reported in the current study, students’ engagement appears to be similar in both online and offline learning environments. Hence, it is necessary to understand how students’ online learning behaviours are shaped. The development of a new online community, such as a course LMS, also remains unexplored. In addition, further research needs to examine how offline relationships contribute to the establishment of online connections and if these two environments are

______Chapter 6 Page 220 perceived differently. Empirical evidence needs to be collected to understand how that perception affects students’ autonomous learning behaviours.

7. Conclusion

The study reported in this thesis examined the perception and performance of learner autonomy in EFL learning, an important capability that Vietnam wants to promote in its higher education. Learner autonomy appears to consist of four interrelated dimensions and be mediated by socio-cultural conditions. This suggests that students need integrated skills and cognitive thinking for the performance of autonomous learning behaviours in either the online or offline learning space. Both students and teachers play an important role in fostering learning activities in an environment (Thomas, 2005). Negotiations between novice (students) and expert others (teachers) on a certain type of content (course requirements) are vital for the establishment and maintenance of a learning community. Dialogues among novices are also significantly essential and should be the central place for learning to happen. Therefore, a combination of Vygotsky’s (1986) novice-expert emphasis and Wenger’s (1999) apprenticeship among peers would make a complete framework for learning community analysis. These findings potentially shed light on a research agenda for the shaping of pedagogical practice to nurture learner autonomy in the context of Vietnam and beyond.

______Chapter 6 Page 221 REFERENCES

Al-Jarf, R. (2007). Teaching vocabulary to EFL college students online. CALL-EJ, 8(2), Retrieved 15 Feb 2011, from http://callej.org/journal/8-2/al-jarf.html.

Allright, R. L. (1988). Autonomy and individualization in whole-class instruction. In A. Brookes & P. Grundy (Eds.), Individualization and autonomy in language learning (pp. 35-44). London: Modern English Publications and the British Council.

Allwright, D. (1990). Autonomy in Language Pedagogy CRILE Working Paper 6. Lancaster: University of Lancaster, Centre for Research in Education.

Aoki, N. (2001). The institutional and psychological context of learner autonomy. AILA Review, 15, 82-89.

Asiaweek. (1999). Asia's Best Universities 1999, Asiaweek, p. 3. Retrieved 15 June 2009, from http://cgi.cnn.com/ASIANOW/asiaweek/universities/schools/index.html

Bachman, J. (1964). Motivation in a task situation as a function of ability and control over task. Journal of Abnormal and Social Psychology, 69, 272–281.

Bachman, L. F. (2004). Statistical analyses for language assessment. Cambridge: Cambridge University Press.

Bailey, C. J., & Card, K. A. (2009). Effective pedagogical practices for online teaching: Perception of experienced instructors. The Internet and Higher Education, 12(3-4), 152-155.

Bandura, A. (1997). Self-efficacy: The exercise of control. New York: W.H. Freeman.

Barfield, A., & Brown, S. H. (Eds.). (2007). Reconstructing autonomy in language education: Inquiry and innovation. Basingstoke: Palgrave Macmillan.

Baron, N. S., Squires, L., Tench, S., & Thompson, M. (2005). Tethered or mobile? Use of away messages in instant messaging by American college students. Retrieved from http://www1.american.edu/tesol/Grimstad-Baron.pdf

Beauvois, M. H. (1992). Computer assisted classroom discussion in the classroom: Conversation in slow motion. Foreign Language Annals, 25(5), 525-534.

Belz, J. A. (2003). Linguistic perspectives on the development of intercultural competence in telecollaboration. Language Learning & Technology, 7(2), 68-99. Retrieved 25 June 2009, from http://llt.msu.edu/vol7num2/belz/default.html

Benson, P. (1997). The philosophy and politics of learner autonomy. In P. Benson & P. Voller (Eds.), Autonomy and independence in language learning (pp. 18-34). London: Longman.

Benson, P. (2001). Teaching and researching autonomy in language learning. London: Longman.

______References Page 222

Benson, P. (2004). Autonomy and information technology in the educational discourse of the information age. In C. Davison (Ed.), Information technology and innovation in language education (pp. 173-192). Hong Kong: Hong Kong University Press.

Benson, P. (2005). Autonomy and information technology in the educational discourse of the information age. In C. Davison (Ed.), Information technology and innovation in language education (pp. 173-191). Hong Kong: Hong Kong University Press.

Benson, P. (2006). Autonomy in language teaching and learning. State-of-the-art Article. Language Teaching, 40(1), 21-40.

Benson, P. (2007). Autonomy and its role in learning. In J. Cummins & C. Davison (Eds.), International handbook of English language teaching (pp. 733-745). New York, NY: Springer.

Benson, P. (2009). Making sense of autonomy in language learning. In S. Toogood, R. Pemberton & A. Barfield (Eds.), Maintaining control: Autonomy and language learning (pp. 13- 26). Hong Kong: Hong Kong University Press.

Benson, P. (2010). Measuring autonomy: Should we put our ability to the test? In A. Paran & L. Sercu (Eds.), Testing the Untestable in Language Education (pp. 77-97). Bristol: Multilingual Matters.

Bhappu, A. D., Ebner, N., Kaufman, S., & Welsh, N. (2009). Online communication technology and relational development. In C. Honeyman, J. Coben & G. D. Palo (Eds.), Rethinking Negotiation Teaching: Innovations For Context And Culture (pp. 239-249). Saint Paul: DRI Press.

Bitchener, J., Young, S., & Cameron, D. (2005). The effect of different types of corrective feedback on ESL student writing. Journal of Second Language Writing, 14(3), 191- 205.

Blin, F. (2004). CALL and the development of learner autonomy: Towards an activity- theoretical perspective. ReCALL, 16(02), 377-395.

Bogdan, R., & Biklen, S. K. (2007). Qualitative research for education: An introduction to theories and methods (5th ed.). Boston, Mass.: Pearson A & B.

Boud, D. (1981). Toward student responsibility for learning. In D. Boud (Ed.), Developing student autonomy in learning (pp. 21-37). New York: Nichols Publishing Company.

Boud, D. (Ed.). (1988). Developing Student Autonomy in Learning (2nd ed.). New York: Kogan Page.

Boulton, A., Chateau, A., Pereiro, M., & Azzam-Hannachi, R. (2008). Learning to learn languages with ICT - But how? CALL-EJ Online, 9(2). Retrieved 15 July 2010, from http://callej.org/journal/9-2/boulton.html

Braine, G. (2003). From a teacher-centered to a student-centered approach: A study of peer feedback in Hong Kong writing classes. Journal of Asian Pacific Communication, 13(2), 269-288.

______References Page 223

Brandon, K. (2003). Best practice in guided individual learning in Australian ELICOS colleges. Paper presented at the Inaugural Independent Learning Association Conference, University of Melbourne, Australia. Retrieved from 15 July 2009, http://independentlearning.org/ILA/ila03/ila03_papers.htm

Breen, M. P., & Mann, S. (1997). Shooting arrows at the sun: Perspectives on a pedagogy for autonomy. In P. Benson & P. Voller (Eds.), Autonomy and Independence in Language Learning (pp. 132-149). Harlow: Longman.

Brew, L. S. (2008). The role of student feedback in evaluating and revising a blended learning course. The Internet and Higher Education, 11(2), 98-105.

Brown, A., & Palinscar, L. (1982). Inducing strategic learning from texts by means of informed, self-control training. Topics in Learning and Learning Disabilities, 2, 1-18.

Brown, J. D. (2001). Using surveys in language programs. Cambridge: Cambridge University Press.

Bui, H. T. M. (2006). Teaching speaking skills at a Vietnamese university and recommendations for using CMC. The Asian EFL Journal. The Innovative Practices & Professional Teaching, 14.

Bullock, D. (2011). Learner self-assessment: an investigation into teachers’ beliefs. ELT Journal, 65(2), 114-125.

Burgess, J. (2006). Blogging to learn, Learning to blog. In A. Bruns & J. Jacobs (Eds.), Uses of blogs (pp. 105-114). New York: Peter Lang.

Bynner, J., & Stribley, K. M. (Eds.). (1979). Social research: Principles and procedures. London: Longman and the Open University Press.

Canale, M. (1983). From communicative competence to communicative language pedagogy. New York: Longman.

Canale, M., & Swain, M. (1980). Theoretical bases of communicative approaches to second language teaching and testing. Applied Linguistics, 1, 1-47.

Cao, H. X. (2001). Tieng Viet, Van Viet, Nguoi Viet [Vietnamese language, Vietnamese literature, and Vietnamese people]. Ho Chi Minh City: Youth Publishing House.

Carreira, J. M. (2011). Relationship between motivation for learning EFL and intrinsic motivation for learning in general among Japanese elementary school students. System, 39(1), 90-102.

Carson, J. G., & Nelson, G. L. (1996). Chinese students’ perceptions of peer response group interaction. Journal of Second Language Writing, 5(1), 1-9.

Caspi, A., Chajut, E., & Saporta, K. (2008). Participation in class and in online discussions: Gender differences. Computers & Education, 50(3), 718-724.

______References Page 224

Chen, N.-S., Quadir, B., & Teng, D. C. (2011). Integrating book, digital content and robot for enhancing elementary school students’ learning of English. Australasian Journal of Educational Technology, 27(3), 546-561.

Chen, P.-C. (2003). EFL student learning style preferences and attitudes toward technology- integrated instruction. Unpublished EdD Dissertation. University of South Dakota, South Dakota.

Chen, P.-S. D., Lambert, A. D., & Guidry, K. R. (2010). Engaging online learners: The impact of Web-based learning technology on college student engagement. Computers & Education, 54(4), 1222-1232.

Cheng, M. C., Jui-Chuan, C., Yi-Chen, C., & Ying-shu, L. (2010). “Do they want the same thing?” Learner perspectives on two content-based course designs in the context of English as a foreign language. Asian EFL Journal, 12(4), 67-84.

Cheng, T.-Y. (2011). The ability of Taiwanese college freshmen to read and interpret graphics in English. The Asian EFL Journal Quarterly, 13(1), 321-347.

Chickering, A. W., & Gamson, Z. F. (1987). Seven principles for good practice in undergraduate education. Washington DC: Washington Center News.

Christopher, L. L. (2006). Fostering learner autonomy in a technology-enhanced, inquiry- based foreign language classroom. Foreign Language Annals, 39(1), 71.

Coakes, S. J., Steed, L. G., & Ong, C. (2009). SPSS: Analysis without anguish: Version 16 for Windows. Milton, Qld.: John Wiley & Sons Australia.

Cohen, L., Manion, L., & Morrison, K. (2007). Research methods in education (6th ed.). London; New York: Routledge.

Conole, G., Brasher, A., Cross, S., Weller, M., Nixon, S., Clark, P., & Petit, J. (2008). A new methodology for learning design Proceedings of World Conference on Educational Multimedia, Hypermedia and Telecommunications (EDMEDIA), 30 June - 4 July 2008. Vienna.

Corbin, J. M., & Strauss, A. L. (2008). Basics of qualitative research: Techniques and procedures for developing grounded theory (3rd ed.). Thousand Oaks, CA.: Sage.

Cotterall, S. (1995). Readiness for autonomy: Investigating learner beliefs. System, 23(2), 195-205.

Creswell, J. W. (2009). Research design: Qualitative, quantitative, and mixed methods approaches (3rd ed.). Thousand Oaks, CA: Sage.

Creswell, J. W., & Clark, V. L. P. (2011). Designing and conducting mixed methods research. Thousand Oaks, CA: Sage.

Cybermetrics Lab. (2008). Regional and Global Ranking of South-East Asia Region. Madrid, Spain: CCHS-CSIC.

______References Page 225

Dam, L. (1990). Learner autonomy in practice: An experiment in learning and teaching. In I. Gathercole (Ed.), Autonomy in language learning: Papers from a conference held in Janaury 1990 (pp. 16-37). London: Centre for Information on Language Teaching & Research.

Dam, L., & Legenhausen, L. (2010). Learners reflecting on learning: Evaluation versus testing in autonomous language learning. In A. Paran & L. Sercu (Eds.), Testing the untestable in language education (pp. 120-139). Clevedon: Multilingual Matters.

Dang, T. T. (2010). Learner autonomy in EFL studies in Vietnam: A discussion from socio- cultural perspective. English Language Teaching, 3(2), 3-9.

Dang, T. T., & Robertson, M. (2009). Online communications: Students’ habits and community formation. Paper presented at the AARE 2009 Conference: Inspiring Innovative Research in Education, Canberra.

Dang, T. T., & Robertson, M. (2010a). E-behaviors and e-community formation: An investigation on Vietnamese EFL students. Asian EFL Journal. Professional Teaching Articles, 46, 4-27.

Dang, T. T., & Robertson, M. (2010b). Impacts of learning management system on learner autonomy in EFL learning. International Education Studies, 3(3), 3-11.

Dang, T. T., & Robertson, M. (2010c). Pedagogical lessons from students’ participation in Web 2.0. TESOL in Context, 20(2), 5-26.

Dang, T. T., & Robertson, M. (2010d). Responses to learning management system: A case study in higher education in Vietnam. Paper presented at the ACEC2010: Digital Diversity Conference, Melbourne.

Dao, K. V., & Hayden, M. (2010). Reforming the governance of higher education in Vietnam. In G. Harman, M. Hayden & N. T. Pham (Eds.), Reforming higher education in Vietnam: Challenges and priorities (1st ed., pp. 129-142): Springer.

Davidson-Shivers, G., Morris, S. B., & Sriwongkol, T. (2003). Gender differences: Are they diminished in online discussions? International Journal on e-learning, 2(1), 29-36.

Dearden, R. F. (1972). Autonomy and education. In R. F. Dearden, P. H. Hirst & R. S. Peters (Eds.), Education and the development of reason (pp. 448-465). London: Routledge & Kegan Paul.

Deepwell, F., & Malikb, S. (2008). On campus, but out of class: an investigation into students’ experiences of learning technologies in their self-directed study. ALT-J, Research in Learning Technology, 16(1), 5-14.

Denckla, M. (1996). Research on executive function in a neurodevelopmental context: Application of clinical measures. Developmental Neuropsychology, 12, 5-15.

Dennis, M., Barnes, M., Donnelly, R., Wilkinson, M., & Humphreys, R. (1996). Appraising and managing knowledge: Metacognitive skills after childhood head injury. Developmental Neuropsychology, 12, 77-103.

______References Page 226

Denzin, N. K. (1970). The research act in sociology: A theoretical introduction to sociological methods. London: Butterworth.

Denzin, N. K., & Lincoln, Y. S. (Eds.). (2005). The SAGE handbook of qualitative research (3rd ed.). Thousand Oaks, CA: Sage.

Deuze, M. (2011). Media life. Media, Culture & Society, 33(1), 137-148.

Dickinson, L. (1987). Self-instruction in language learning. Cambridge [Cambridgeshire]; New York: Cambridge University Press.

Dickinson, L. (1992). Learner autonomy 2: Learner training for language learning. Dublin: Authentik.

Dickinson, L. (1995). Autonomy and motivation a literature review. System, 23(2), 165-174.

Dixon, R., Dixon, K., & Siragusa, L. (2007). Individuals' perceptions of online environments: What adult learners are telling us. In R. J. Atkinson, C. McBeath, S. K. A. Soong & C. Cheers (Eds.), ICT: Providing choices for learners and learning. Proceedings ascilite Singapore 2007 (pp. 207-218). Centre for Educational Development, Nanyang Technological University, Singapore, 2-5 December.

Dornyei, Z. (2001). New themes and approaches in second language motivation research. Annual Review of Applied Linguistics, 21, 43-59.

Dornyei, Z. (2003). Attitudes, orientation, and motivations in language learning: Advances in theory, research, and applications. Language Learning, 53(S1), 3-32.

Dornyei, Z. (2007). Research methods in applied linguistics. Oxford: Oxford University Press.

Dornyei, Z., Csizér, K., & Németh, N. (2006). Motivation, language attitudes and globalisation: A Hungarian perspective. Clevedon, England; Buffalo: Multilingual Matters.

Duda, R. (2005). Assumptions and hidden agendas in ICT materials: How does autonomization come in? Mélanges CRAPEL, 28, 67-75.

Ellis, R. (1994). The study of second language acquisition. Oxford: Oxford University Press.

Ely, C. M. (1986). Language learning motivation: A descriptive and causal analysis. The Modern Language Journal, 70(1), 28-35.

Elyildirim, S., & Ashton, S. (2006). Creating positive attitudes towards English as a foreign language. English Teaching Forum, 44(4), 2-21.

Entwistle, N. (2000). Promoting deep learning through teaching and assessment: Conceptual frameworks and educational contexts. Paper presented at the TLRP Conference, Leicester. Retrieved 5 July 2010, from http://www.tlrp.org/acadpub/Entwistle2000.pdf

______References Page 227

Farmer, J. (2006). Blogging to Basics: How Blogs are Bringing Online Education Back from the Brink. In A. Bruns & J. Jacobs (Eds.), Uses of blogs (pp. 91-104). New York: Peter Lang.

Felix, U. (2001). A multivariate analysis of students' experience of web based learning. Australian journal of educational technology, 17(1), 21-36.

Field, A. (Ed.). (2005). Discovering statistics using SPSS. London: Sage.

Figura, K., & Jarvis, H. (2007). Computer-based materials: A study of learner autonomy and strategies. System, 35(4), 448-468.

Finnish National Board of Education. (2010). Education. Retrieved 15 March 2011, from http://www.oph.fi/english/education

Flavell, J. H., Miller, P. H., & Miller, S. A. (1993). Cognitive development (3rd ed.). Englewood Cliffs, N.J.: Prentice-Hall.

Fonseka, E. A. G. (2003). Autonomy in a resource-poor setting: Enhancing the carnivalesque. In D. Palfreyman & R. Smith (Eds.), Learner autonomy across cultures: language education perspectives (pp. 147-163). New York: Palgrave Macmillan.

Freeman, M. (1999). The language learning activities of students of EFL and French at two universities. Language Learning Journal, 19(1), 80 - 88.

Gardner, D. (2001). Making self-access centres more effective. In D. Kember, S. Candlin & L. Yan (Eds.), Further Case Studies of Improving Teaching and Learning from the Action Learning Project (pp. 161-174). Hong Kong: HK Polytechnic University.

Gardner, D. (2007a). Integrating self-access learning into an EAP course. In D. Gardner (Ed.), Integration and support (pp. 8-32). Dublin, Ireland: Authentik.

Gardner, D. (2007b). Integration and support. Dublin, Ireland: Authentik.

Gardner, D. (2009). Understanding autonomous learning: Students’ perceptions. Paper presented at the Independent Learning Association 2007 Japan Conference: Exploring theory, enhancing practice: Autonomy across the disciplines, Kanda University of International Studies, Chiba, Japan. Retrieved 15 September 2010, from http://independentlearning.org/ILA/ila07/files/ILA2007_011.pdf

Gardner, R. (1985). Social psychology and second language learning: The role of attitudes and motivation. London: Edward Arnold.

Gardner, R. (2001). Integrative motivation and second language acquisition. In Z. Dornyei & R. Schmidt (Eds.), Motivation and second language learning (pp. 1-19). Honolulu: University of Hawaii.

Gardner, R., & MacIntyre, P. (1991). Motivational variables in second language acquisition. Studies in Second Language Acquisition, 13, 57-72.

______References Page 228

Gardner, R., Moorcroft, R., & Metford, J. (1989). Second language learning in an immersion program: Factors influencing acquisition and retention. Journal of Language and Social Psychology, 8(5), 287-305.

Geng, J. (2010). Autonomy for English teaching and learning in China. Journal of Language Teaching and Research, 1(6), 942-944.

Gitsaki, C. (2005). Course design to promote student autonomy and lifelong learning skills: A Japanese example? Paper presented at the 2nd Independent Learning Association Oceania Conference, Manukau Institute of Technology, Auckland, New Zealand. Retrieved 11 February 2010, from http://independentlearning.org/ILA/ila05/ila05_papers.htm

Gorsuch, R. L. (1983). Factor analysis (2nd ed.). Hillsdale, N.J: L. Erlbaum Associates.

Guardado, M., & Shi, L. (2007). ESL students' experiences of online peer feedback. Computers and Composition, 24(4), 443-461.

Guba, G., & Lincoln, Y. S. (2005). Paradigmatic controversies, contradictions, and emerging confluences. In N. K. Denzin & Y. S. Lincoln (Eds.), The SAGE handbook of qualitative research (3rd ed., pp. 191-215). Thousand Oaks, CA: Sage.

Guilloteaux, M. J., & Dornyei, Z. (2008). Motivating language learners: A classroom-oriented investigation of the effects of motivational strategies on student motivation. TESOL Quarterly, 42(1), 55-77.

Hanreddy, J., & Whalley, E. (2002). Mosaic: Listening/Speaking (4th ed.). New York: McGraw-Hill. Books 1 & 2.

Hardy, M., & Reynolds, J. (2009). Incorperating categorical information into regression models: The utility of dummy variables. In M. Hardy & A. Bryman (Eds.), The handbook of data analysis (pp. 209-236). London: Sage.

Harman, G., Hayden, M., & Pham, N. T. (2010). Higher education in Vietnam: reform, challenges and priorities. In G. Harman, M. Hayden & N. T. Pham (Eds.), Reforming higher education in Vietnam: Challenges and priorities (1st ed., pp. 1-13). London: Springer.

Harman, K., & Nguyen, T. N. B. (2010). Reforming teaching and learning in Vietnam’s higher education system. In G. Harman, M. Hayden & N. T. Pham (Eds.), Reforming higher education in Vietnam: Challenges and priorities (1st ed., pp. 65-86). London: Springer.

Hart, N. (2002). Intra-group autonomy and authentic materials: A different approach to ELT in Japanese colleges and universities. System, 30(1), 33-46.

Hassan, A., Jamaludin, N. S., Sulaiman, T., & Baki, R. (2010). Western and Eastern educational philosophies. Paper presented at the 40th Annual Conference of the Philosophy of Education Society of Australasia Inc., Murdoch University.

Haythornthwaite, C. (2005). Social networks and Internet connectivity effects. Information, Communication, & Society, 8(2), 125-147. ______References Page 229

Healy, D. (1999). Theory and research: Autonomy and language learning. In J. Egbert & E. Hanson-Smith (Eds.), CALL environments: research, practice, and critical issues (pp. 391-402). Alexandria, Va.: Teachers of English to Speakers of Other Languages.

Healy, D. (2007). Theory and research: autonomy and language learning. In J. Egbert & E. Hanson-Smith (Eds.), CALL environments: research, practice, and critical issues (2nd ed., pp. 377-388). Alexandria, Va.: Teachers of English to Speakers of Other Languages.

Heins, B., Duensing, A., Stickler, U., & Batstone, C. (2007). Spoken interaction in online and face-to-face language tutorials. CALL, 20(3), 279 – 295.

Hoang, T. (2008). Educational crisis: Reasons and solutions in globalization challenges Paper presented at the Comparative Education Conference 2008, Ho Chi Minh City.

Hofstede, G., & Hofstede, J. (2005). Cultures and organizations software of the mind. New York: McGraw-Hill.

Holec, H. (1981). Autonomy in foreign language learning. Oxford: Pergamon.

Holec, H. (1987). The learner as manager: Managing learning or managing to learn. In A. Wenden & J. Rubin (Eds.), Learner strategies in language learning (pp. 145–157). Englewood Cliffs, New Jersey: Prentice-Hall Inc.

Holiday, A. (2003). Social autonomy: Addressing the dangers of culturism in TESOL. In D. Palfreyman & R. Smith (Eds.), Learner autonomy across cultures: language education perspectives (pp. 110-126). New York: Palgrave Macmillan.

Hong, H. (2010). Viet Nam thuoc nhom nuoc co ty le chi cho giao duc cao nhat the gioi [Vietnam belongs to the group of countries with the highest level of investment in education] Retrieved 10 July 2010, from http://dantri.com.vn/c25/s25-392976/viet- nam-thuoc-nhom-nuoc-co-ty-le-chi-cho-giao-duc-cao-nhat-the-gioi.htm

Horváth, I. (2005). The cognitive components of autonomous learning in postgraduate interpreter training. Paper presented at the 2nd Independent Learning Association Oceania Conference, Manukau Institute of Technology, Auckland, New Zealand. Retrieved 22 May 2009, from http://independentlearning.org/ILA/ila05/ila05_papers.htm

Hozayen, G. (2011). Egyptian students’ readiness for autonomous language learning. In D. Gardner (Ed.), Fostering Autonomy in Language Learning (pp. 115-125). Gaziantep: Zirve University.

Hubbard, P. (2005). A review of subject characteristics in CALL research. Computer Assisted Language Learning, 18(5), 351 - 368.

Hunt, J., Gow, L., & Barnes, P. (1989). Learner self-evaluation and assessment - A tool for autonomy in the language learning classroom. In V. Bickley (Ed.), Language Teaching and Learning Styles Within and Across Cultures (pp. 207-217). Hong Kong: Institute of Language in Education, Education Department.

______References Page 230

Hyland, F. (2004). Learning autonomously: Contextualising out-of-class English language learning. Language Awareness, 13(3), 180-202.

Jacobs, G. E. (2008). People, purposes, and practice: Insights from cross-disciplinary research into instant messaging. In J. Coiro, M. Knobel, C. Lankshear & D. J. Leu (Eds.), Handbook of research on new literacies (pp. 467-490). New York: Lawrence Erlbaum Associates/Taylor & Francis Group.

Jeon-Ellis, G., Debski, R., & Wigglesworth, G. (2005). Oral interaction around computers in the project-oriented call classroom. Language, Learning & Technology, 9(3), 121-145.

Jeong, H., & Hong, B. (2011). A learning system using user preference in ubiquitous computing environment. In J. J. Park, L. T. Yang & C. Lee (Eds.), Future Information Technology (Vol. 184, pp. 240-247). London: Springer Berlin Heidelberg.

Johnson, R. B., Onwuengbuzie, A. J., & Turner, L. A. (2007). Toward a definition of mixed methods research. Journal of Mixed Methods Research, 1(2), 112-133.

Jung, U. (2005). CALL: Past, present and future - A bibliometric approach. ReCALL, 17(1), 4-17.

Kanniah, A., & Krish, P. (2010). Collaborative learning skills used in Weblog. CALL-EJ Online, 11(2). Retrieved 12 June 2009, from http://callej.org/journal/11- 2/kanniah_krish.html

Kanno, Y., & Norton, B. (2003). Imagined communities and educational possibilities: Introduction. Journal of Language, Identity & Education, 2(4), 241 - 249.

Kent, N., & Facer, K. (2004). Different worlds? A comparison of young people’s home and school ICT use. Journal of Computer Assisted Learning, 20, 440-455.

Kern, R. G. (1995). Restructuring classroom interaction with networked computers: Effects on quantity and characteristics of language production. Modern Language Journal, 79(4), 457-476.

Kessler, G., & Bikowski, D. (2010). Developing collaborative autonomous learning abilities in computer mediated language learning: attention to meaning among students in wiki space. Computer Assisted Language Learning, 23(1), 41 - 58.

Kevin, W. (2003). Deep learning and education for sustainability. International Journal of Sustainability in Higher Education, 4(1), 44.

Kissau, S. (2006). Gender differences in motivation to learn French. The Canadian Modern Language Review, 62(3), 401-422.

Koda-Dallow, T., & Hobbs, M. (2005). Personal goal-setting and autonomy in language learning. Paper presented at the 2nd Independent Learning Association Oceania Conference, Manukau Institute of Technology, Auckland, New Zealand. Retrieved 22 June 2009, from http://independentlearning.org/ILA/ila05/ila05_papers.htm

Kramsch, C., & Sullivan, P. (1996). Appropriate pedagogy. ELT Journal, 50(3), 199-212.

______References Page 231

Kumaravadivelu, B. (2003). Beyond methods: Macrostrategies for language teaching. New Haven: Yale University Press.

Kurata, N. (2007). Language choice and second language learning opportunities in learners' social networks. Australian Review of Applied Linguistics, 30(1), 05.01-05.18.

La Ganza, W. (2001). Out of sight - not out of mind: Learner autonomy and interrelating in online teaching. Information technonogy, Education and Society, 2(2), 27-46.

La Ganza, W. (2002). Maintaining learner autonomy online: The teacher’s role. In L. Mak, S. Chang, P. Foo, J. Hunter, M. Keung, J. Lee, M. Wa & N. Noakes (Eds.), Meeting the challenges of research and practice. Hong Kong: Hong Kong University of Science and Technology.

Lam, W. S. E. (2004). Second language socialization in a bilingual chat room. Language Learning and Technology, 8(3), 44-65.

Lamb, M. (2004). ‘It depends on the students themselves’: Independent language learning at an Indonesian state school. Language Culture and Curriculum, 17(3), 229-245.

Lamb, T. (2001). Metacognition and motivation - Learning to learn. In G. N. Chambers (Ed.), Reflections on Motivation (pp. 85-93). London: CILT.

Lamb, T. (2008). Leaner autonomy and teacher autonomy: Synthesising an agenda. In T. Lamb & H. Reinders (Eds.), Learner and teacher autonomy: concepts, realities, and responses (pp. vi, 286 p.). Amsterdam; Philadelphia: John Benjamins.

Lamb, T. (2009). Controlling learning: Learners’ voices and relationships between motivation and learner autonomy. In S. Toogood, R. Pemberton & A. Barfield (Eds.), Maintaining control: Autonomy and language learning (pp. 67-86). Hong Kong: Hong Kong University Press.

Lamb, T. (2010). Assessment of autonomy or assessment for autonomy? Evaluating learner autonomy for formative purposes. In A. Paran & L. Sercu (Eds.), Testing the untestable in language education (pp. 98-119). Clevedon: Multilingual Matters.

Lamb, T., & Fisher, J. (1999). Making connections: Football, the Internet and reluctant language learners. Language Learning Journal, 20, 32-36.

Lamb, T., & Reinders, H. (2008). Learner and teacher autonomy: concepts, realities, and responses. Amsterdam; Philadelphia: John Benjamins Pub. Co.

Lantolf, J. P. (2000). Sociocultural theory and second language learning. Oxford: Oxford University Press.

Lantolf, J. P., & Thorne, S. L. (2006). Sociocultural theory and the genesis of second language development. Oxford: Oxford University Press.

Lave, J., & Wenger, E. (1991). Situated learning: legitimate peripheral participation. Cambridge [England]; New York: Cambridge University Press.

______References Page 232

Le, B. K. (1998). Vai net chinh ve dinh huong phat trien ngoai ngu trong nhung nam dau cua the ky 21 [Some main points about the development directions of foreign language at the beginning of the 21st century] Nhung van de chien luot trong phat trien giao duc thoi ky hien dai hoa va cong nghiep hoa [Stretegic matters of education development in the time of modernization and industrialization] (pp. 117-127): Vu Giao Duc Thuong Xuyen, Bo Giao Duc Dao Tao.

Le, C. V. (2000). Language and Vietnamese pedagogical contexts. In J. Shaw, D. Lubelska & M. Noullet (Eds.), Partnership and interaction: Proceedings of the Fourth International Conference on Language and Development (pp. 73–79). Bangkok: Asian Institute of Technology.

Le, C. V., & Barnard, R. (2009a). Curricular innovation behind closed classroom doors: A Vietnamese case study. Prospect: An Australian Journal of TESOL, 24(2), 20-33.

Le, C. V., & Barnard, R. (2009b). Teaching grammar: A survey of teachers’ attitudes in Vietnam. The Journal of Asia, 6(3), 245-273.

Le, H. P. (2008). Su bao cap da troi buoc giao vien [The system of subsidies has tied up its teachers], Tuoi Tre. Retrieved 12 February 2009, from http://www.tuoitre.com.vn/Tianyon/Index.aspx?ArticleID=287486&ChannelID=13

Levy, M., & Stockwell, G. (2006). CALL dimensions: Options and issues in computer- assisted language learning. Mahwah, NJ: Lawrence Erlbaum.

Lew, D. N. M., & Schmidt, H. G. (2011). Writing to learn: can reflection journals be used to promote self-reflection and learning? Higher Education Research & Development, 30(4), 519-532.

Lin, Y.-C. A. (2004). An assessment of the international students' attitudes toward technology-based learning: English as a second language (ESL) implication. Unpublished EdD thesis. Mississippi State University, Mississippi.

Little, D. (1990). Autonomy in language learning: Some theoretical and practical considerations. In I. Gathercole (Ed.), Autonomy in language learning: papers from a conference held in January 1990 (pp. 106 p.). London: Centre for Information on Language Teaching & Research.

Little, D. (1991). Learner autonomy 1: Definitions, issues, and problems. Dublin: Authentik.

Little, D. (1996a). Freedom to learn and compulsion to interact: promoting learner autonomy through the use of information systems and information technologies. In R. Pemperton, E. S. L. Li, W. W. F. Or & H. D. Pierson (Eds.), Taking control: Autonomy in language learning. Hong Kong: Hong Kong University Press.

Little, D. (1996b). The politics of learner autonomy. Learning Learning, 2, 7-10.

Little, D. (2001). Learner autonomy and the challenge of tandem language learning via the internet. In A. Chambers & G. Davies (Eds.), ICT and language learning: a European perspective (pp. 29-38). Lisse, Netherlands ; Exton, (PA): Swets & Zeitlinger Publishers.

______References Page 233

Little, D. (2003). Learner autonomy and second/foreign language learning. Guide to Good Practice. Retrieved 12 April 2009, from http://www.llas.ac.uk/resources/gpg/1409

Littlemore, J. (2001). Learner autonomy, self-instruction and new technologies in language learning: Current theory and practice in higher education in Europe. In A. Chambers & G. Davies (Eds.), ICT and language learning: a European perspective (pp. 39-52). Lisse, Netherlands ; Exton, PA: Swets & Zeitlinger.

Littlewood, W. (1996). “Autonomy”: An anatomy and a framework. System, 24(4), 427-435.

Littlewood, W. (1997). Self-access: why do want it and what can it do? In P. Benson & P. Voller (Eds.), Autonomy and Independence in Language Learning (pp. 79-92). Harlow: Longman.

Littlewood, W. (1999). Defining and developing autonomy in East Asian contexts. Applied Linguistics, 20(1), 71-94.

Liu, M. (2009). Reticence and Anxiety in Oral English Lessons: Peter Lang Pub Inc.

Locke, E. A., & Latham, G. P. (1990). A theory of goal setting & task performance. Englewood Cliffs, N.J.: Prentice-Hall.

Lockyer, L., Bennett, S., Agostinho, S., & Harper, B. (Eds.). (2008). Handbook of research on learning design and learning objects: Issues, applications and technologies: Ideas Group Publications.

Lomicka, L., & Lord, G. (2009a). Introduction to social networking, collaboration, and web 2.0 tools. In L. Lomicka & G. Lord (Eds.), The next generation: social networking and online collaboration in foreign language learning (pp. 1-11). San Marcos, TX: CALICO.

Lomicka, L., & Lord, G. (2009b). The next generation: social networking and online collaboration in foreign language learning . San Marcos, TX: CALICO.

Luu, A. T. (2008). Viet Nam dung thu 79 ve chi so phat trien giao duc [Vietnam's EDI ranks 79th], Tuoi Tre. Retrieved 11 June 2009, from http://www.tuoitre.com.vn/Tianyon/Index.aspx?ArticleID=286274&ChannelID=13

Macaro, E. (1997). Target language, collaborative learning and autonomy. Clevedon: Multilingual Matters.

Macaro, E. (2008). The shifting dimensions of language learner autonomy. In T. Lamb & H. Reinders (Eds.), Learner and teacher autonomy: concepts, realities, and responses (pp. 47-62). Amsterdam; Philadelphia: John Benjamins.

Mackey, A., & Gass, S. M. (2005). Second language research: Methodology and design. Mahwah, N.J.: Lawrence Erlbaum.

Mai, L. (2009). Van la cau chuyen tien giao duc [The story of education budget again] Retrieved 10 February 2010, from http://tuoitre.vn/Giao-duc/341528/Van-la-cau- chuyen-tien-cho-giao-duc.html

______References Page 234

Maiko, H. (2003). Using computer-mediated communication in second language classrooms. Osaka Keidai ronshu, 54(3), 115-125.

Malikowski, S. R., Thompson, M. E., & Theis, J. G. (2006). External factors associated with adopting a CMS in resident college courses. The Internet and Higher Education, 9(3), 163-174.

Martinez, H. (2008). The subjective theories of student teachers: Implication for teacher education and research on learner autonomy. In T. Lamb & H. Reinders (Eds.), Learner and teacher autonomy: concepts, realities, and responses (pp. 103-124). Amsterdam; Philadelphia: John Benjamins.

Masgoret, A. M., & Gardner, R. C. (2003). Attitudes, motivation, and second language learning: A meta-analysis of studies conducted by Gardner and associates. Language Learning, 53(S1), 167-210.

Masters, K., & Oberprieler, G. (2004). Encouraging equitable online participation through curriculum articulation. Computers & Education, 42(4), 319-332.

Mau, C., & Quoc, B. (2007). Hoi nghi trien khai ke hoach kinh ke xa hoi va ngan sach nha nuoc nam 2008: Phan dau hoan thanh co ban cac chi tieu chu yeu giai doan 2006-2010 [Meeting on the implementation of socio-economic plans and national budget in 2008], Tap chi Cong nghe Thong tin va Truyen thong. Retrieved 5 June 2009, from http://www.tapchibcvt.gov.vn/News/PrintView.aspx?ID=17805

Maxwell, J. A., & Mittapalli, K. (2010). Realism as a stance for mixed methods research. In A. Tashakkori & C. Teddlie (Eds.), SAGE handbook of mixed methods in social and behavioral research (2nd ed., pp. 145-167). Thousand Oaks, CA: Sage.

McCarthy, T. (2010). Integrating Project-based learning into a traditional skills-based curriculum to foster learner autonomy: An action research. The Journal of Kanda University of International Studies, 22, 221-244.

McClure, J. (2001). Developing language skills and learner autonomy in international postgraduates. ELT Journal, 55(2), 142-148.

Merriam, S. B., Caffarella, R. S., & Baumgartner, L. (2007). Learning in adulthood: a comprehensive guide (3rd ed.). San Francisco: Jossey-Bass.

Miller, L. (2007). Autonomy in the classroom. Dublin, Ireland: Authentik.

Miller, L., Hopkins, M., & Tsang, E. (2005). Self-access language learning in Hong Kong secondary schools. Paper presented at the 2nd Independent Learning Association Oceania Conference, Manukau Institute of Technology, Auckland, New Zealand. Retrieved 1 June 2009, from http://independentlearning.org/ILA/ila05/ila05_papers.htm

Ministry of Education and Training (MOET). (2005). Vietnam higher education renovation agenda - Period 2006-2020. Resolution No.14/2005/NQ-CP on the Fundamental and Comprehensive Reform of Higher Education in Vietnam 2006-2020.

______References Page 235

MOET. (2008). Huong dan thuc hien nhiem vu CNTT nam hoc 2008-2009 [Official Guides to Task Implementation of the School Year 2008-2009]. Ha Noi: MOET.

Morris, L. V., & Finnegan, C. L. (2008-2009). Best practices in predicting and encouraging student persistence and achievement online. Journal of College Student Retention, 10(1), 55-64.

Morse, J. M. (2010). Procedures and practice of mixed method design: maintaining control, rigor, and complexity. In A. Tashakkori & C. Teddlie (Eds.), SAGE handbook of mixed methods in social and behavioral research (2nd ed., pp. 339-352). Thousand Oaks, CA: Sage.

Mortensen, T. E. (2008). Of a divided mind: Weblog literacy. In J. Coiro, M. Knobel, C. Lankshear & D. J. Leu (Eds.), Handbook of research on new literacies (pp. 449-466). New York: Lawrence Erlbaum Associates/Taylor & Francis Group.

Mulligan, C., & Garofalo, R. (2011). A collaborative writing approach: Methodology and student assessment. The Language Teacher, 35(3), 5-10.

Mulligan, D., & Kirkpatrick, A. (2000). How much do they understand? Lectures, students and comprehension. Higher Education Research & Development, 19(3), 311 - 335.

Murray, G. (1999). Autonomy and language learning in a simulated environment. System, 27(3), 295-308.

Murray, G., Gao, A., & Lamb, T. (Eds.). (2011). Identity, Motivation, and Autonomy in Language Learning Bristol: Multilingual Matters.

Mynard, J. (2009). How blogging can promote learner autonomy. Paper presented at the Independent Learning Association 2007 Japan Conference: Exploring theory, enhancing practice: Autonomy across the disciplines, Kanda University of International Studies, Chiba, Japan. Retrieved 12 June 2010, from http://independentlearning.org/ILA/ila07/files/ILA2007_030.pdf

Naizhao, G., & Yanling, Z. (2004). An empirical investigation of learner autonomy in some EFL classes in China. Paper presented at the AARE 2004 Conference, Melbourne. Retrieved 12 December 2008, from http://www.aare.edu.au/04pap/nai04930.pdf

Nakata, Y. (2011). Teachers' readiness for promoting learner autonomy: A study of Japanese EFL high school teachers. Teaching and Teacher Education, 27(5), 900-910.

Nelson, G., & Carson, J. G. (1998). ESL students’ perceptions of effectiveness in peer response groups. Journal of Second Language Writing, 7(2), 113-131.

Nguyen, B., & David, C. (1999). The design of English language textbook in Vietnamese secondary schools. Paper presented at the The Fourth International Conference on Language and Development, Hanoi, Vietnam. Retrieved 12 June 2009, from http://www.languages.ait.ac.th/hanoi_proceedings/crabbe.htm

Nguyen, L. V. (2010). Computer mediated collaborative learning within a communicative language teaching approach: A sociocultural perspective. The Asian EFL Journal Quarterly, 12(1), 202-233. ______References Page 236

Nguyen, N. T. (2006). Bo truong Giao duc dang dan Quoc hoi [Education Minister's Report in the Congress Meeting]. Ha Noi.

Nguyen, P. (2008). Chat luong giao duc dai hoc: Doi moi tu goc den ngon [Quality in higher education: An entire renovation], Tuoi Tre. Retrieved 17 July 2009, from http://www.tuoitre.com.vn/Tianyon/Index.aspx?ArticleID=237211&ChannelID=13

Nguyen, T. C. L. (2009). Learner autonomy and EFL learning at the tertiary level in Vietnam. Doctoral Thesis. Victoria University of Wellington. Retrieved from 12 March 2011, http://researcharchive.vuw.ac.nz/handle/10063/1203

Nguyen, T. H. (2002). Vietnam: Cultural Background for ESL/EFL Teachers. The Review of Vietnamese Studies, 2(1).

Nguyen, T. V. (2008). Chat luong giao duc: Tu thay den tro [Educational quality: From teachers to students], Nguoi Lao Dong. Retrieved from 12 October 2010, http://www.nld.com.vn/tintuc/giao-duc/213086.asp

Norton, B. (2000). Identity and language learning: gender, ethnicity and educational change. Harlow, England; New York: Longman.

Norton, B. (2001). Non-participation, imagined communities, and the language classroom. In M. Breen (Ed.), Learner contributions to language learning: New directions in research (pp. 159–171). Harlow, England: Pearson Education.

Nunan, D. (1988). The learner-centred curriculum. Cambridge; New York: Cambridge University Press.

Nunan, D. (1991). Language teaching methodology: a textbook for teachers. New York: Prentice-Hall.

Nunan, D. (1997). Designing and adapting materials to encourage learner autonomy. In P. Benson & P. Voller (Eds.), Autonomy and Independence in Language Learning (pp. 192 - 203). Harlow: Longman.

Nunan, D., & Lamb, C. (2001). Managing learning process. In A. Hewings & D. Hall (Eds.), Innovation in English language teaching: a reader (pp. 27-45). London; New York: Routledge in association with Macquarie University and the Open University.

O'Malley, J. M., & Chamot, A. U. (1990). Learning strategies in second language acquisition. Cambridge [England]; New York: Cambridge University Press.

Ogbu, J. (1991). Immigrant and involuntary monorities in comparative perspective. In M. A. Gibson & J. U. Ogbu (Eds.), Minority status and schooling: a comparative study of immigrant and involuntary minorities (pp. 3-33). New York: Garland.

Oxford, R. (2003). Toward a more systematic model of L2 learner autonomy. In D. Palfreyman & R. Smith (Eds.), Learner autonomy across cultures: language education perspectives (pp. 75-91). New York: Palgrave Macmillan.

Oxford, R., & Shearin, J. (1994). Language learning motivation: Expanding the theoretical framework. The Modern Language Journal, 78(1), 12-28. ______References Page 237

Palloff, R. M., & Pratt, K. (2005). Collaborating online: learning together in community (1st ed.). San Francisco: Jossey-Bass.

Papi, M., & Abdollahzadeh, E. (2011). Teacher motivational practice, student motivation, and possible L2 selves: An examination in the Iranian EFL context. Language Learning, 61(2), 1-24.

Paris, S., Lipson, M., & Wixson, K. (1983). Becoming a strategic reader. Contemporary Educational Psychology, 8, 293-316.

Paul, J. L. (2005). Introduction to the philosophies of research and criticism in education and the social sciences. Upper Saddle River, NJ: Pearson Education.

Pellegrino, V. (1996). At the risk of speaking... Risk management in second language use. ACTR Letter, 22, 6-13.

Pemberton, R., Toogood, S., & Barfield, A. (Eds.). (2009). Maintaining Control: Autonomy and Language Learning. Hong Kong: Hong Kong University Press.

Pennycook, A. (1997). Cultural alternatives and autonomy. In P. Benson & P. Voller (Eds.), Autonomy and independence in language learning (pp. 18-34). London: Longman.

Pham, H. H. (1999). The key socio-cultural factors that work against success in tertiary English language training programs in Vietnam. Paper presented at the The fourth International Conference on language and Development, Ha Noi. Retrieved from 12 November 2010, http://www.languages.ait.ac.th/hanoi_proceedings/hiep.html

Pham, H. T., & Ngo, A. D. (2008). Phat trien nguon nhan luc cong nghiep o Viet Nam trong giai doan moi cua cong nghiep hoa [Human development for Vietnam industry in the period of industrialization]. Retrieved from 12 December 2008, http://www.vdf.org.vn/Doc/2008/VDFConf_WIPHoangVie.pdf

Pham, N. T. (2010). The higher education reform agenda: A vision for 2020. In G. Harman, M. Hayden & N. T. Pham (Eds.), Reforming higher education in Vietnam: Challenges and priorities (1st ed., pp. 51-64). London: Springer.

Phan, H. T. T. (2009). Impacts of Vietnam’s social context on learners' attitudes towards foreign languages and English language learning: Implications for teaching and learning. The Asian EFL Journal Quarterly, 11(4), 169-188.

Pickard, N. (1996). Out-of-class language learning strategies. ELT Journal, 50(2), 150-159.

Pill, T. J. H. (2001). Adult learners' perceptions of out-of-class access to English. Master of Arts in Applied Linguistics, Hong Kong University, Hong Kong. Retrieved 18 June 2009, from http://sunzi.lib.hku.hk/hkuto/record/B3194517X

Pinkman, K. (2005). Using blogs in the foreign language classroom: Encouraging learner independence. The JALT CALL Journal, 1(1), 12-24.

Punch, K. F. (2009). Introduction to research methods in education. Los Angeles; London: SAGE.

______References Page 238

Purdie, N., Hattie, J., & Douglas, G. (1996). Student conceptions of learning and their use of self-regulated learning strategies: A cross-cultural comparison. Journal of Educational Psychology, 88(1), 87-100.

Rakap, S. (2010). Impacts of learning styles and computer skills on adult students' learning online. The Turkish Online Journal of Educational Technology, 9(2), 108-115.

Raz, J. (1986). The morality of freedom. Oxford: Oxford University Press.

Reinders, H., & Balcikanli, C. (2011). Learning to foster autonomy: The role of teacher education materials. Studies in Self-Access Learning Journal, 2(1), 15-12.

Ribe, R. (2003). Tramas in the foreign language classroom: autopoietic networks for learner growth. In D. Little, J. Ridley & E.Ushioda (Eds.), Towards Autonomy in Foreign language Learning (pp. 11-28). Dublin: Authentik.

Richards, C. (2005). The design of effective ICT-supported learning activities: Exemplary models, changing requirements, and new possibilities. Language learning and technology, 9(1), 60-79.

Riley, P. (2009). Discursive dissonance in approaches to autonomy. In R. Pemberton, S. Toogood & A. Barfield (Eds.), Maintaining Control: Autonomy and Language Learning (pp. 45-63). Hong Kong: Hong Kong University Press.

Riley, P., & Zoppis, C. (1985). The sound and video library. In P. Riley (Ed.), Discourse and Learning (pp. 114-131). London: Longman.

Rivers, D. J. (2011). Strategies and struggles in the ELT classroom: Language policy, learner autonomy, and innovative practice. Language Awareness, 20(1), 31 - 43.

Rivers, W. P. (2001). Autonomy at all costs: An ethnography of metacognitive self- assessment and self-management among experienced language learners. The Modern Language Journal, 85(2), 279-290.

Robertson, M., Webb, I., & Fluck, A. (2007). Seven steps to ICT integration. Camberwell, Vic.: ACER Press.

Robinson, C. C., & Hullinger, H. (2008). New benchmarks in higher education: Student engagement in online learning. Journal of Education for Business, 84(2), 101-109.

Rubin, J. (1975). What the “good language learner” can teach us. TESOL Quarterly, 9, 41-51.

Rundle, C. F. (2007). Teacher approaches to Japanese culture: Implications for student autonomy. Paper presented at the Proceedings of the Independent Learning Association 2007 Japan Conference: Exploring theory, enhancing practice: Autonomy across the disciplines, Kanda University of International Studies, Chiba, Japan. Retrieved 22 June 2009, from http://independentlearning.org/ILA/ila07/files/ILA2007_034.pdf

Sao, H., & Wu, Z. (2007). Nurturing language learner autonomy through caring pedagogic practice. In A. Barfield & S. Brown (Eds.), Restructuring autonomy in language education: Inquiry and innovation. Basinstoke: Palgrave. ______References Page 239

Scharle, A., & Szabo, A. (2000). Learner Autonomy: A Guide to Developing Learner Responsibility. Cambridge: Cambridge University Press.

Schunk, D., & Zimmerman, B. J. (1994). Self-regulation of learning and performance: Issues and educational applications. Hillsdale, NJ: Erlbaum.

Schwienhorst, K. (2003). Neither here nor there? Learner autonomy and intercultural factors in CALL environments. In D. Palfreyman & R. Smith (Eds.), Learner autonomy across cultures: language education perspectives (pp. 164-179). New York: Palgrave Macmillan.

Schwienhorst, K. (2008a). CALL and autonomy: Settings and context variables in technology-enhanced language environments. Independence, 43, 13-15.

Schwienhorst, K. (2008b). Learner autonomy and CALL environments. New York; London: Routledge.

Sidhu, R. K., & Embi, M. A. B. (2009). Learner autonomy among Malaysian adult learners through asynchronous online discussions. ASEAN Journal of Teaching & Learning in Higher Education, 1(2), 30-34.

Sinclair, B. (2000). Learner autonomy: The next phase. In I. Mcgrath, B. Sinclair & T. Lamb (Eds.), Learner Autonomy, Teacher Autonomy: Future Directions (pp. 15-23). London: Longman.

Sinclair, B. (2009). The teacher as learner: Developing autonomy in an interactive learning environment. In R. Pemberton, S. Toogood & A. Barfield (Eds.), Maintaining Control: Autonomy and Language Learning (pp. 175-198). Hong Kong: Hong Kong University Press.

Skehan, P. (2003). Task-based instruction. Language Teaching, 36(1), 1-14.

Slife, B. D., & Williams, R. N. (1995). What's behind the research? Discovering hidden assumptions in the behavioral sciences. Thousand Oaks, CA: Sage.

Smith, R. (2000). Starting with ourselves: Teacher-learner autonomy in language learning. In B. Sinclair, I. McGrath & T. Lamb (Eds.), Learner autonomy, teacher autonomy: New directions (pp. 89-99). Harlow: Longman.

Smith, R. (2001). Group work for autonomy in Asia: insights from teacher-research. AILA Review, 15, 70-81.

Smith, R. (2003a). Pedagogy for autonomy as (becoming-) appropriate methodology. In D. Palfreyman & R. Smith (Eds.), Learner autonomy across cultures: language education perspectives (pp. 129-146). New York: Palgrave Macmillan.

Smith, R. (2003b). Postcript: Implications for language education. In D. Palfreyman & R. Smith (Eds.), Learner autonomy across cultures: language education perspectives (pp. 254-260). New York: Palgrave Macmillan.

______References Page 240

Smith, R. (2003c). Teacher education for teacher-learner autonomy. In J. Gollin, G. Ferguson & H. Trappes-Lomax (E ds.), Symposium for Language Teacher Educators: Papers from Three IALS Symposia. Edinburgh: IALS, University of Edinburgh.

Smith, R., & Erdogan, S. (2008). Teacher-learner autonomy: Programme goals and student- teacher construct. In T. Lamb & H. Reinders (Eds.), Learner and teacher autonomy: concepts, realities, and responses (pp. 83-102). Amsterdam; Philadelphia: John Benjamins Pub. Co.

Smith, R., & Ushioda, E. (2009). 'Autonomy': under whose control? In R. Pemberton, S. Toogood & A. Barfield (Eds.), Maintaining Control: Autonomy and Language Learning (pp. 241-253). Hong Kong: Hong Kong University Press.

Stepp-Greany, J. (2002). Student perceptions on language learning in a technological environment: Implications for the new millennium. Language, Learning & Technology, 6(1), 165-180.

Surowiecki, J. (2004). The wisdom of crowds: why the many are smarter than the few and how collective wisdom shapes business, economies, societies and nations. London: Little, Brown.

Tarone, E., & Yule, G. (1989). Focus on the language learner. Oxford: Oxford University Press.

Tashakkori, A., & Teddlie, C. (Eds.). (2010). SAGE handbook of mixed methods in social and behavioral research (2nd ed.). Thousand Oaks, CA: Sage.

Taylor, H., Schatschneider, C., Petrill, S., Barry, C., & Owens, C. (1996). Executive dysfunction in children with early brain disease: Outcomes post Haemophilus Influenzae meningitis. Developmental Neuropsychology, 12, 35-51.

Thanh, H. (2008). Giang day tieng Anh trong cac truong DH: kem vi thieu chuan [Teaching English in Universities: Ineffective Due to Lack of Standard], Tuoi Tre. Retrieved 23 September 2009, from http://www.tuoitre.com.vn/Tianyon/Index.aspx?ArticleID=291254&ChannelID=13

Thanh, T. (2009). Thay doi cho dai hoc the ky 21 [Renovations for University in the 21st Century], Tuoi Tre. Retrieved 12 June 2011, from http://tuoitre.vn/Giao- duc/342861/Thay-doi-cho-dai-hoc-the-ky-21.html

Tharp, R. G. (1989). Psychocultural variables and constants: Effects on teaching and learning in schools. American Psychologist, 44(2), 349-359.

Thomas, A. (2005). Children online: learning in a virtual community of practice. E–Learning, 2(1), 27-38.

Thomas, D. C., & Inkson, K. (2003). Cultural intelligence: People skills for global business. San Francisco, CA: Berrett-Koehler.

Tien, D. (2009). 30 nam chat luong giao duc dai hoc bi “bo ngo” ["Ignorance" of higher education quality over 30 years]. VnExpress. Retrieved from 12 March 2010, http://vnexpress.net/gl/xa-hoi/giao-duc/2009/10/3ba151a0/ ______References Page 241

Tok, H. (2011). Autonomous language learning: Turkish tertiary students’ behaviours. In D. Gardner (Ed.), Fostering Autonomy in Language Learning (pp. 136-146). Gaziantep: Zirve University.

Toohey, K. (2007). Conclusion: Autonomy/Agency through socio-cultural lenses. In A. Barfield & S. H. Brown (Eds.), Reconstructing autonomy in language education: inquiry and innovation (pp. 231-242). Basingstoke: Palgrave Macmillan.

Tran, T. D. (1996). So phan cac sinh ngu thoi hien dai [The fate of foreign languages in modern time], Giao Duc Thoi Dai, p. 8.

Tran, T. T. T., & Baldauf, R. B. (2007). Demotivation: Understanding resistance to English language learning: The case of Vietnamese students. The Journal of Asia TEFL, 4(1), 79-105.

Trinh, Q. L. (2005). Stimulating learner autonomy in English language education: a curriculum innovation study in a Vietnamese context. PhD Thesis. University of Amsterdam. Retrieved 1 May 2009, from http://dare.uva.nl/document/102346

Trinh, Q. L. (2008). Phat trien nang luc tu hoc trong hoan canh Viet Nam [Developing learner autonomy in the context of Vietnam]. Tap Chi Khoa Hoc, 10. Retrieved 1 May 2009, from http://www.ctu.edu.vn/departments/dra/journal/vol10/21.pdf

Ushioda, E., Smith, R., Mann, S., & Brown, P. (2011). Promoting teacher-learner autonomy through and beyond initial language teacher education. Language Teaching, 44(01), 118-121.

Vandergrift, L. (2005). Relationships among motivation orientations, metacognitive awareness and proficiency in L2 listening. Applied Linguistics, 26(1), 70-89.

Voller, P. (2005). Teachers, facilitation and autonomy. Paper presented at the 2nd Independent Learning Association Oceania Conference, Manukau Institute of Technology, Auckland, New Zealand. Retrieved from 11 May 2010, http://independentlearning.org/ILA/ila05/ila05_papers.htm

Voltz, D. (2008). Autonomy. New World Encyclopedia. Retrieved 15 June 2009, from http://www.newworldencyclopedia.org/entry/Autonomy?oldid=795378

Vu, V. Q. (2007). Thu phan tich so lieu cong bo cua Bo Giao duc ve Dau tu va Co cau Tai chinh [An attempt to analyse the statistics about investement and budget structures released by the Ministry of Education and Training] Retrieved 20 December 2008, from http://www.viet-studies.info/VQViet_PhanTichSoLieuCongBo.htm

Vu, V. T. (2000). The higher education. In MOET (Ed.), Vietnam education and training directory (pp. 34-40). Vietnam: Education Publishers.

Vygotsky, L. S. (1986). Thought and language (A. Kozulin, Trans. ed.). MA: The MIT Press.

Wacho, P. (2006). Methods and materials for motivation and learner autonomy. Reflections on English Language Teaching, 5(1), 93-122.

______References Page 242

Wallis, R. (2005). Independent learning: What do our students do and why do they do it? Paper presented at the 2nd Independent Learning Association Oceania Conference, Manukau Institute of Technology, Auckland, New Zealand. Retrieved 17 May 2009, from http://independentlearning.org/ILA/ila05/ila05_papers.htm

Wang, H. (2008). Language policy implementation: A look at teachers’ perceptions. The Asian EFL Journal Quarterly. Professional Teaching Articles, 30, 1-38.

Wang, H. (2011). Promoting university English majors' learner autonomy in the Chinese context. Journal of Language Teaching and Research, 2(2), 408-412.

Warden, C. A., & Lin, H. J. (2000). Existence of integrative motivation in an Asian EFL setting. Foreign Language Annals, 33(5), 535-545.

Waterhouse, P. (1990). Supported self-study across the curriculum. In I. Gathercole (Ed.), Autonomy in language learning: papers from a conference held in Janaury 1990 (pp. 4-6). London: Centre for Information on Language Teaching & Research.

Wells, M. (2007). Collaborative online projects in a global community. In T. Townsend & R. J. Bates (Eds.), Handbook of teacher education: globalization, standards and professionalism in times of change (pp. 657-674). Dordrecht: Springer.

Wenden, A. L. (1991). Learner Strategies for Learner Autonomy. Hemel Hempstead: Prentice Hall.

Wenden, A. L. (2002). Learner development in language learning. Applied Linguistics, 23(1), 32-55.

Wenger, E. (1999). Communities of practice: learning, meaning, and identity. Cambridge, MA: Cambridge University Press.

Wenger, E., McDermott, R. A., & Snyder, W. (2002). Cultivating communities of practice: a guide to managing knowledge. Boston: Harvard Business School Press.

Wong, J. K.-K. (2004). Are the learning styles of Asian international students culturally or contextually based? International Education Journal, 4(4), 154 - 166.

Woodsong, C., & Koo, H. P. (1999). Two good reasons: women's and men's perspectives on dual contraceptive use. Social Science & Medicine, 49(5), 567-580.

World Bank. (2006). Education in Vietnam: Development history, challenges and solutions. Washington, DC.

Wyatt, M. (2011). Becoming a do-it-yourself designer of English language teaching materials. Forum: Qualitative Social Research, 12(1).

Xie, T. (2002). Using Internet Relay Chat in teaching Chinese. CALICO Journal, 19(3), 513- 524.

Yang, M., Badger, R., & Yu, Z. (2006). A comparative study of peer and teacher feedback in a Chinese EFL writing class. Journal of Second Language Writing, 15(3), 179-200.

______References Page 243

Yang, S. C. (2001). Language learning on the World Wide Web: An Investigation of EFL Learners Attitudes and Perceptions. Journal of Educational Computing Research, 24(2), 155 - 181.

Yang, T. (2007). Construction of an inventory of learner autonomy. On CUE, 15(1), 2-9.

Young, R. (1986). Personal autonomy: Beyond negative and positive liberty. London: Croom Helm.

Young, S., Cantrell, P. P., & Shaw, D. G. (2001). Online instruction: New roles for teachers and students. Academic Exchange Quarterly, 5(4), 11-16.

______References Page 244

Appendix 3A

THE FIFTY-NINE ITEMS CATEGORIZED IN THREE CORE PROCESSES

(1) Initiating (2) Monitoring Statement (3) Evaluating (1) (2) (3) x look for opportunities to use English as much as possible x try to find as many ways as they can to improve their English x find information about English by themselves x will ask their teacher if they have a question about English x will talk with their friends/relatives if they have a question about English x want to study in an English-speaking environment if they have a chance x want to find a job where only English is used in the future x pay attention when someone is speaking in English

x have plans about how to learn English x make their schedule so they will have enough time to study English x consider a lot about study/work before doing it x try to have a good environment to study English before studying E at home

x study English voluntarily x are aware of their studies x learn English because they enjoy studying it x have a strong will and are not easily influenced by others x are determined to achieve the target which they set for their English study

x dream of being good English speakers x want to be good English learners x think English is important for their future x x practice English with people outside class x x will talk to a foreigner in English if they come across him/her in the street x x want to communicate with foreigners in English

x x learn things that the teachers do not give as a task x carry out the learning plans once they have been made x try to study English regularly even with limited time x try to complete the things they have decided to do x will act immediately if they decide to do something x use time effectively

x study things which were not from their class x pay more attention to the lessons that they are not good at x change their study contents and target according to their needs x will work out a question about English by themselves if they have

x try to develop their good points x try to improve their weaknesses x know the method which suits them best and use it x make good use of materials and resources when studying English at home

______Appendix 3A Page 245

(1) Initiating (2) Monitoring Statement (3) Evaluating (1) (2) (3) x do things actively x concentrate on their studies even if they happen to hear others talking x feel like studying once the class starts x study English no matter what mood they are in

x x try to use other methods if one method of English study does not suit them x x write down their feelings towards English studies in a language learning diary x x notice their mistakes and use that information to help them do better x x check to make sure that they understood the lesson x check their English proficiency by taking TOEIC, TOEFL or IELTS voluntarily

x think about their progress in learning English x know their good points and weaknesses x reflect on what they learn and look for something important x think about their methods of learning E and whether those methods are good x give themselves a reward or treat when they do well in English 24 22 13 Total

______Appendix 3A Page 246 Appendix 3B SURVEY QUESTIONNAIRE OF STUDY ONE

English Version ======SURVEY QUESTIONNAIRE FOR STUDENTS

I. Please make a cross (x) in the appropriate boxes to indicate your choice

1. Your gender: Male Female 2. Your major in university: English ………………………………….. 3. Your general computer proficiency (MS Office, web, chat, email, blog…): very good good ok bad very bad

4. Your university is in …………….……………………………………………….. city.

II. The statements below are designed to explore the characteristics of EFL students. To what extent do you think that these statements are true? Please indicate the degree of your agreement by circling the corresponding number for each item, using the scale below.

1. never or almost never true 4. usually true 2. occasionally true 5. always or almost always true 3. sometimes true

STATEMENTS DEGREE OF No Students who succeed best with learning English: AGREEMENT 1 study English voluntarily. 1 2 3 4 5 2 give themselves a reward or treat when they do well in English. 1 2 3 4 5 3 use the Internet to learn things which are not taught in class. 1 2 3 4 5 4 want to communicate with foreigners in English. 1 2 3 4 5 5 carry out the learning plans once they have been made. 1 2 3 4 5 make good use of materials and resources when studying English at 6 1 2 3 4 5 home. 7 look for opportunities to use English as much as possible. 1 2 3 4 5 do not want to improve English by taking part in online communities 8 1 2 3 4 5 such as forums, blogs, chatrooms... 9 try to find as many ways as they can to improve their English. 1 2 3 4 5 10 will ask their teacher if they have a question about English. 1 2 3 4 5 11 do not look for different resources on the Internet. 1 2 3 4 5 12 want to study in an English-speaking environment if they have a chance. 1 2 3 4 5 think about their methods of learning English and whether those methods 13 1 2 3 4 5 are good. try to have a good environment to study English, e.g. cleaning up the 14 1 2 3 4 5 table before studying English at home. 15 do not know their good points and weaknesses. 1 2 3 4 5 ______Appendix 3B Page 247 16 try to complete some online activities even with limited time. 1 2 3 4 5 17 try to develop their good points. 1 2 3 4 5 18 will work out a question about English by themselves if they have. 1 2 3 4 5 19 do not try to improve their weaknesses. 1 2 3 4 5 20 do not learn things that the teachers do not give as a task. 1 2 3 4 5 write down their feelings towards English studies in a language learning 21 1 2 3 4 5 diary. 22 learn English because they enjoy studying it. 1 2 3 4 5 23 share their feelings towards English studies with friends online. 1 2 3 4 5 24 have a strong will and are not easily influenced by others. 1 2 3 4 5 25 do not think about their progress in learning English. 1 2 3 4 5 26 concentrate on their studies even if they happen to hear others talking. 1 2 3 4 5 27 consider a lot about study/work before doing it. 1 2 3 4 5 28 pay attention when someone is speaking in English. 1 2 3 4 5 29 change their study contents and target according to their needs. 1 2 3 4 5 are determined to achieve the target which they set for their English 30 1 2 3 4 5 study. 31 like to study with computers. 1 2 3 4 5 32 make their schedule so they will have enough time to study English. 1 2 3 4 5 33 will do a search on the Internet if they have a question about English. 1 2 3 4 5 34 will act immediately if they decide to do something. 1 2 3 4 5 check their English level by comparing it with the skills of others on the 35 1 2 3 4 5 Internet. 36 reflect on what they learn and look for something important. 1 2 3 4 5 37 use time effectively. 1 2 3 4 5 38 notice their mistakes and use that information to help them do better. 1 2 3 4 5 try to use other methods if one method of English study does not suit 39 1 2 3 4 5 them. 40 feel like studying once the class starts. 1 2 3 4 5 41 want to find a job where only English is used in the future. 1 2 3 4 5 42 go online as a way of learning English. 1 2 3 4 5 43 check to make sure that they understood the lesson. 1 2 3 4 5 44 pay more attention to the lessons that they are not good at. 1 2 3 4 5 45 do not want to be good English learners. 1 2 3 4 5 46 pay more attention when they see an English website. 1 2 3 4 5 47 do not do things actively. 1 2 3 4 5 48 try to study English regularly even with limited time. 1 2 3 4 5

______Appendix 3B Page 248 49 study things which were not from their class. 1 2 3 4 5 50 will talk with their friends/relatives if they have a question about English. 1 2 3 4 5 51 study English no matter what mood they are in. 1 2 3 4 5 will talk to a foreigner in English if they come across him/her in the 52 1 2 3 4 5 street. 53 will use English to communicate if seeing a stranger on the Internet. 1 2 3 4 5 54 try to complete the things they have decided to do. 1 2 3 4 5 55 think English is not important for their future. 1 2 3 4 5 56 have no plans about how to learn English. 1 2 3 4 5 check their English proficiency by taking TOEIC, TOEFL or IELTS 57 1 2 3 4 5 voluntarily. 58 are not aware of their studies. 1 2 3 4 5 59 dream of being good English speakers. 1 2 3 4 5 60 do not practice English with people outside class. 1 2 3 4 5 61 do not find information about English by themselves. 1 2 3 4 5 62 know the method which suits them best and use it. 1 2 3 4 5

Other characteristics:

63. ……………………………….……………………………….……………………………….……………………………….……………………………….

64. ……………………………….……………………………….……………………………….……………………………….……………………………….

65. ……………………………….……………………………….……………………………….……………………………….………………………………. ======

Note: Thirteen items in shading are in negative form. Responses to these items are positively recoded before the analysis process.

______Appendix 3B Page 249 Vietnamese Version ======

CÂU HỎI KHẢO SÁT DÀNH CHO SINH VIÊN

I. Hãy đánh dấu chéo (x) vào khung ứng với lựa chọn của bạn hoặc điền vào khoảng trống

1. Giới tính của bạn: Nam Nữ 2. Chuyên ngành của bạn là: Tiếng Anh ………………………………….. 3. Tự đánh giá khả năng máy tính cơ bản của bạn (MS Office, web, chat, email, blog…):

giỏi khá trung bình yếu kém

4. Trường của bạn ở thành phố …………….……………………………………………….. .

II. Các ý kiến được ghi nhận bên dưới nhằm tìm hiểu về các đặc điểm và thói quen học tập của các bạn sinh viên chuyên ngành tiếng Anh. Bạn hãy vui lòng cho biết ý kiến của mình về những đặc điểm này bằng cách sử dụng các mức độ bên dưới và khoanh tròn con số tương ứng để chỉ mức độ đồng tình của bạn với từng câu. Lưu ý rằng đây là những nhận xét của riêng cá nhân bạn, thực tế không có đáp án nào là duy nhất đúng cho mỗi câu cả. Đồng thời, ý kiến của bạn cũng không hề ảnh hưởng gì đến việc học hiện tại của bạn.

Các mức độ dùng để đánh giá: 1. không bao giờ đúng hoặc hiếm khi đúng 2. thỉnh thoảng đúng 3. đôi khi đúng 4. thường đúng 5. luôn luôn hoặc hầu như luôn luôn đúng

S Ý KIẾN M Ứ C Đ Ộ TÁN TT Sinh viên học tiếng Anh THÀNH CÔNG NHẤT là những người: THÀNH 1 tự nguyện học tiếng Anh. 1 2 3 4 5 2 tự thưởng cho mình mỗi khi đạt được thành tích gì đó về tiếng Anh. 1 2 3 4 5 3 dùng internet để học những điều không được dạy ở lớp. 1 2 3 4 5 4 muốn giao tiếp với người nước ngoài bằng tiếng Anh. 1 2 3 4 5 5 thực hiện các kế hoạch học tập nếu đã đề ra. 1 2 3 4 5 6 khai thác các tài liệu tốt khi học tiếng Anh ở nhà. 1 2 3 4 5 7 luôn tìm kiếm cơ hội sử dụng tiếng Anh thật nhiều. 1 2 3 4 5 không muốn cải thiện tiếng Anh bằng cách tham gia vào các cộng đồng 8 1 2 3 4 5 trên mạng như diễn đàn, viết blog, chat… 9 cố gắng tìm mọi cách để cải thiện tiếng Anh. 1 2 3 4 5 10 sẽ hỏi giáo viên nếu có thắc mắc về tiếng Anh. 1 2 3 4 5 11 không tìm tư liệu trên internet. 1 2 3 4 5 ______Appendix 3B Page 250 12 muốn học tập ở môi trường nói tiếng Anh nếu có cơ hội. 1 2 3 4 5 suy nghĩ về phương pháp học tiếng Anh của họ và đánh giá xem những 13 1 2 3 4 5 phương pháp ấy có tốt hay không. cố gắng tạo một môi trường tốt để học tiếng Anh, ví dụ như lau bàn sạch 14 1 2 3 4 5 sẽ trước khi học tiếng Anh ở nhà. 15 không biết được điểm mạnh và điểm yếu của bản thân. 1 2 3 4 5 16 vẫn cố thực hiện vài hoạt động trên mạng cho dù chỉ có ít thời gian. 1 2 3 4 5 17 nỗ lực phát triển những điểm tốt của mình. 1 2 3 4 5 18 sẽ tự mình giải quyết khi có thắc mắc về tiếng Anh. 1 2 3 4 5 19 không nỗ lực cải thiện những điểm yếu của mình. 1 2 3 4 5 20 không học những phần mà giáo viên không giao làm. 1 2 3 4 5 21 ghi lại cảm xúc của mình về việc học tiếng Anh ra nhật ký học tập. 1 2 3 4 5 22 thích học tiếng Anh, cho nên họ học nó. 1 2 3 4 5 23 chia sẽ cảm xúc học tiếng Anh của mình với bạn bè trên mạng. 1 2 3 4 5 24 có ý chí mạnh mẽ và không dễ bị người khác gây ảnh hưởng. 1 2 3 4 5 25 không suy nghĩ gì về tiến bộ của mình trong quá trình học tiếng Anh. 1 2 3 4 5 26 vẫn chăm chú vào học tập cho dù tình cờ có nghe ai đó nói chuyện. 1 2 3 4 5 27 suy nghĩ rất nhiều về công việc/học tập trước khi thực hiện chúng. 1 2 3 4 5 28 chú ý lắng nghe khi có ai đó đang nói tiếng Anh. 1 2 3 4 5 29 thay đổi nội dung và mục tiêu học tập tùy theo nhu cầu của mình. 1 2 3 4 5 30 kiên định nhằm đạt được các mục đích học tiếng Anh đã đề ra. 1 2 3 4 5 31 thích học tập với máy tính. 1 2 3 4 5 32 lập thời gian biểu để có đủ thời gian học tiếng Anh. 1 2 3 4 5 33 sẽ lên mạng tìm thông tin khi gặp khó khăn về tiếng Anh. 1 2 3 4 5 34 sẽ hành động ngay nếu đã quyết định làm điều gì. 1 2 3 4 5 kiểm tra trình độ tiếng Anh của mình bằng cách so sánh kỹ năng của 35 mình với các bạn khác trên internet. 36 tự xem lại những điều đã học và tìm kiếm những chi tiết quan trọng. 1 2 3 4 5 37 sử dụng thời gian hiệu quả. 1 2 3 4 5 38 để ý những lỗi của mình và dùng nó để giúp họ học tốt hơn. 1 2 3 4 5 cố gắng dùng các phương pháp học tiếng Anh khác nếu thấy một phương 39 1 2 3 4 5 pháp nào đó không phù hợp với mình. 40 cảm thấy như đang học tập ngay từ đầu buổi học. 41 muốn sau này tìm được việc làm ở nơi sử dụng toàn tiếng Anh. 42 lên mạng là một cách để học tiếng Anh. 1 2 3 4 5 43 kiểm tra để chắc rằng mình đã hiểu bài. 1 2 3 4 5 44 chú ý tới bài học nhiều hơn nếu họ không giỏi phần đó. 1 2 3 4 5

______Appendix 3B Page 251 45 không thích trở thành người học giỏi tiếng Anh. 1 2 3 4 5 46 chú ý hơn khi thấy một trang web bằng tiếng Anh. 1 2 3 4 5 47 học tập không năng động. 1 2 3 4 5 48 cố gắng học tiếng Anh đều đặn cho dù có hạn hẹp về thời gian. 1 2 3 4 5 49 học cả những phần không có trên lớp. 1 2 3 4 5 50 sẽ nói với bạn bè nếu có thắc mắc về tiếng Anh 1 2 3 4 5 51 học tiếng Anh trong bất kỳ tâm trạng nào. 1 2 3 4 5 sẽ nói chuyện với người nước ngoài bằng tiếng Anh nếu tình cờ gặp họ 52 1 2 3 4 5 ngoài đường. 53 sẽ dùng tiếng Anh để giao tiếp nếu gặp một người lạ nào đó trên mạng. 1 2 3 4 5 54 cố gắng hoàn tất những điều mình đã quyết định sẽ làm. 1 2 3 4 5 55 cho rằng tiếng Anh không quan trọng đối với tương lai của mình. 1 2 3 4 5 56 không có kế hoạch cho việc học tiếng Anh của mình. 1 2 3 4 5 kiểm tra trình độ tiếng Anh của mình bằng cách tự nguyện đi thi TOEIC, 57 1 2 3 4 5 TOEFL hay IELTS. 58 không ý thức về việc học của mình. 1 2 3 4 5 59 mơ ước trở thành người nói tiếng Anh thành thạo. 1 2 3 4 5 60 không thực hành tiếng Anh với người khác bên ngoài lớp học. 1 2 3 4 5 61 không tìm kiếm các thông tin về tiếng Anh. 1 2 3 4 5 62 biết được phương pháp học nào phù hợp với mình nhất và dùng nó. 1 2 3 4 5

Những đặc điểm khác của một sinh viên chuyên Anh giỏi:

63. ……………………………….……………………………….……………………………….……………………………….……………………………….

64. ……………………………….……………………………….……………………………….……………………………….……………………………….

65. ……………………………….……………………………….……………………………….……………………………….………………………………. ======

______Appendix 3B Page 252 Appendix 3C PRE-TEST QUESTIONNAIRE IN STUDY TWO

======SURVEY QUESTIONNAIRE FOR STUDENTS

I. Please make a cross (x) in the appropriate boxes to indicate your choice or fill in the blank

1. Gender: Male Female 2. Year of birth: 19………….….. 3. Student No.: …………………………………..…….. 4. Class: LS-3B……………………….….. 5. General computer proficiency (MS Office, web, chat, email, blog…): very good good ok bad very bad

II. The statements below are about habits of students majoring in English. Please indicate the degree of your agreement by circling the corresponding number for each item, using the scale below.

1. never or almost never true 4. usually true 2. occasionally true 5. always or almost always true 3. sometimes true

STATEMENTS DEGREE OF No Students who succeed best with learning English: AGREEMENT 1 use the Internet to learn things which are not taught in class. 1 2 3 4 5 2 do not find information about English by themselves. 1 2 3 4 5 3 make good use of materials and resources when studying English at home. 1 2 3 4 5 4 look for opportunities to use English as much as possible. 1 2 3 4 5 5 try to find as many ways as they can to improve their English. 1 2 3 4 5 6 will ask their teacher if they have a question about English. 1 2 3 4 5 7 know the method which suits them best and use it. 1 2 3 4 5 8 want to study in an English-speaking environment if they have a chance. 1 2 3 4 5 9 do not know their good points and weaknesses. 1 2 3 4 5 10 do not try to improve their weaknesses. 1 2 3 4 5 11 do not do things actively. 1 2 3 4 5 12 do not think about their progress in learning English. 1 2 3 4 5 13 learn English because they enjoy studying it. 1 2 3 4 5 14 have a strong will and are not easily influenced by others. 1 2 3 4 5 15 study English voluntarily. 1 2 3 4 5

______Appendix 3C Page 253 16 are determined to achieve the target which they set for their English study. 1 2 3 4 5 17 like to study with computers. 1 2 3 4 5 18 make their schedule so they will have enough time to study English. 1 2 3 4 5 19 will do a search on the Internet if they have a question about English. 1 2 3 4 5 20 will act immediately if they decide to do something. 1 2 3 4 5 21 try to develop their good points. 1 2 3 4 5 22 reflect on what they learn and look for something important. 1 2 3 4 5 23 use time effectively. 1 2 3 4 5 24 notice their mistakes and use that information to help them do better. 1 2 3 4 5 25 try to use other methods if one method of English study does not suit 1 2 3 4 5 them. 26 go online as a way of learning English. 1 2 3 4 5 27 check to make sure that they understood the lesson. 1 2 3 4 5 28 think about their methods of learning English and if those methods are 1 2 3 4 5 good. 29 do not want to be good English learners. 1 2 3 4 5 30 pay more attention when they see an English website. 1 2 3 4 5 31 try to study English regularly even with limited time. 1 2 3 4 5 32 study things which were not from their class. 1 2 3 4 5 33 want to find a job where only English is used in the future. 1 2 3 4 5 34 consider a lot about study/work before doing it. 1 2 3 4 5 35 try to complete the things they have decided to do. 1 2 3 4 5 36 think English is not important for their future. 1 2 3 4 5 37 have no plans about how to learn English. 1 2 3 4 5 38 change their study contents & target according to their needs. 1 2 3 4 5 39 are not aware of their studies. 1 2 3 4 5 40 want to communicate with foreigners in English. 1 2 3 4 5 41 do not practice English with people outside class. 1 2 3 4 5

Other learning habits of students good at English:

42. ……………………………….……………………………….……………………………….……………………………….……………………………….

43. ……………………………….……………………………….……………………………….……………………………….……………………………….

44. ……………………………….……………………………….……………………………….……………………………….……………………………….

______Appendix 3C Page 254

III. Do the following calculations

1. Eight plus five:

2. Six multiply by four:

3. Three times eleven:

4. One hundred and ten divide by five:

5. Six times twelve plus eight times twenty:

IV. Attempt to answer the following questions, using your own estimations

1. Current number of students in USSH: …………………………

2. Current population of HCMC: …………………………

3. Current population of Vietnam: …………………………

4. Current population of the world: …………………………

5. The most crowded city in the world: …………………………

______Appendix 3C Page 255 V. Here are now some statements about your own learning approach. If a statement is never true for you, choose 1. If it is always true for you, choose 5. Circle your answers in the corresponding numbers.

1. never or almost never true for me 4. usually true for me 2. occasionally true for me 5. always or almost always true for me 3. sometimes true for me

DEGREE OF No STATEMENTS AGREEMENT 1 I use the Internet to learn things which are not taught in class. 1 2 3 4 5 2 I do not find information about English by myself. 1 2 3 4 5 3 I make good use of materials and resources when studying English at 1 2 3 4 5 home. 4 I look for opportunities to use English as much as possible. 1 2 3 4 5 5 I try to find as many ways as I can to improve my English. 1 2 3 4 5 6 I will ask my teacher if I have a question about English. 1 2 3 4 5 7 I know the method which suits me best and use it. 1 2 3 4 5 8 I want to study in an English-speaking environment if I have a chance. 1 2 3 4 5 9 I do not know my good points and weaknesses. 1 2 3 4 5 10 I do not try to improve my weaknesses. 1 2 3 4 5 11 I do not do things actively. 1 2 3 4 5 12 I do not think about my progress in learning English. 1 2 3 4 5 13 I learn English because I enjoy studying it. 1 2 3 4 5 14 I have a strong will and am not easily influenced by others. 1 2 3 4 5 15 I study English voluntarily. 1 2 3 4 5 16 I am determined to achieve the target which I set for my English study. 1 2 3 4 5 17 I like to study with computers. 1 2 3 4 5 18 I make my schedule so I will have enough time to study English. 1 2 3 4 5 19 I will do a search on the Internet if I have a question about English. 1 2 3 4 5 20 I will act immediately if I decide to do something. 1 2 3 4 5 21 I try to develop my good points. 1 2 3 4 5 22 I reflect on what I learn and look for something important. 1 2 3 4 5 23 I use time effectively. 1 2 3 4 5 24 I notice my mistakes and use that information to help me do better. 1 2 3 4 5 25 I try to use other methods if one method of English study does not suit 1 2 3 4 5 me. 26 I go online as a way of learning English. 1 2 3 4 5

______Appendix 3C Page 256 27 I check to make sure that I understood the lesson. 1 2 3 4 5 28 I think about my methods of learning English and if those methods are 1 2 3 4 5 good. 29 I do not want to be good English learners. 1 2 3 4 5 30 I pay more attention when I see an English website. 1 2 3 4 5 31 I try to study English regularly even with limited time. 1 2 3 4 5 32 I study things which were not from their class. 1 2 3 4 5 33 I want to find a job where only English is used in the future. 1 2 3 4 5 34 I consider a lot about study/work before doing it. 1 2 3 4 5 35 I try to complete the things I have decided to do. 1 2 3 4 5 36 I think English is not important for my future. 1 2 3 4 5 37 I have no plans about how to learn English. 1 2 3 4 5 38 I change my study contents & target according to my needs. 1 2 3 4 5 39 I am not aware of my studies. 1 2 3 4 5 40 I want to communicate with foreigners in English. 1 2 3 4 5 41 I do not practice English with people outside class. 1 2 3 4 5

Other habits of my English learning:

42. ……………………………….……………………………….……………………………….……………………………….……………………………….

43. ……………………………….……………………………….……………………………….……………………………….……………………………….

44. ……………………………….……………………………….……………………………….……………………………….……………………………….

======Note: The 16 items in shading are used as dumping items. Responses to these items are not included in the analysis.

______Appendix 3C Page 257 Appendix 3D POST-TEST QUESTIONNAIRE IN STUDY TWO ======SURVEY QUESTIONNAIRE FOR STUDENTS

I. Please fill in the blank

1. Student No.: …………………………………..…….. 2. Class: LS-3B……………………….…..

II. Here are some statements about your own learning approach. If a statement is never true for you, choose 1. If it is always true for you, choose 5. Circle your answers in the corresponding numbers.

1. never or almost never true for me 4. usually true for me 2. occasionally true for me 5. always or almost always true for me 3. sometimes true for me

DEGREE OF No STATEMENTS AGREEMENT 1 I use the Internet to learn things which are not taught in class. 1 2 3 4 5 2 I do not find information about English by myself. 1 2 3 4 5 3 I make good use of materials and resources when studying English at 1 2 3 4 5 home. 4 I look for opportunities to use English as much as possible. 1 2 3 4 5 5 I try to find as many ways as I can to improve my English. 1 2 3 4 5 6 I will ask my teacher if I have a question about English. 1 2 3 4 5 7 I know the method which suits me best and use it. 1 2 3 4 5 8 I want to study in an English-speaking environment if I have a chance. 1 2 3 4 5 9 I do not know my good points and weaknesses. 1 2 3 4 5 10 I do not try to improve my weaknesses. 1 2 3 4 5 11 I do not do things actively. 1 2 3 4 5 12 I do not think about my progress in learning English. 1 2 3 4 5 13 I learn English because I enjoy studying it. 1 2 3 4 5 14 I have a strong will and am not easily influenced by others. 1 2 3 4 5 15 I study English voluntarily. 1 2 3 4 5 16 I am determined to achieve the target which I set for my English study. 1 2 3 4 5 17 I like to study with computers. 1 2 3 4 5 18 I make my schedule so I will have enough time to study English. 1 2 3 4 5 19 I will do a search on the Internet if I have a question about English. 1 2 3 4 5 20 I will act immediately if I decide to do something. 1 2 3 4 5

______Appendix 3D Page 258 21 I try to develop my good points. 1 2 3 4 5 22 I reflect on what I learn and look for something important. 1 2 3 4 5 23 I use time effectively. 1 2 3 4 5 24 I notice my mistakes and use that information to help me do better. 1 2 3 4 5 25 I try to use other methods if one method of English study does not suit 1 2 3 4 5 me. 26 I go online as a way of learning English. 1 2 3 4 5 27 I check to make sure that I understood the lesson. 1 2 3 4 5 28 I think about my methods of learning English and if those methods are 1 2 3 4 5 good. 29 I do not want to be good English learners. 1 2 3 4 5 30 I pay more attention when I see an English website. 1 2 3 4 5 31 I try to study English regularly even with limited time. 1 2 3 4 5 32 I study things which were not from their class. 1 2 3 4 5 33 I want to find a job where only English is used in the future. 1 2 3 4 5 34 I consider a lot about study/work before doing it. 1 2 3 4 5 35 I try to complete the things I have decided to do. 1 2 3 4 5 36 I think English is not important for my future. 1 2 3 4 5 37 I have no plans about how to learn English. 1 2 3 4 5 38 I change my study contents & target according to my needs. 1 2 3 4 5 39 I am not aware of my studies. 1 2 3 4 5 40 I want to communicate with foreigners in English. 1 2 3 4 5 41 I do not practice English with people outside class. 1 2 3 4 5

Other habits of my English learning:

42. ……………………………….……………………………….……………………………….……………………………….……………………………….

43. ……………………………….……………………………….……………………………….……………………………….……………………………….

44. ……………………………….……………………………….……………………………….……………………………….………………………………. ======Note: The 16 items in shading are used as dumping items. Responses to these items are not included in the analysis.

______Appendix 3D Page 259 Appendix 3E

THE NUMBERS ASSOCIATED WITH QUESTIONNAIRE ITEMS IN STUDY ONE AND STUDY TWO

Study Study STATEMENTS One Two 3 1 use the Internet to learn things which are not taught in class. 61 2 do not find information about English by themselves. 6 3 make good use of materials and resources when studying English at home. 7 4 look for opportunities to use English as much as possible. 9 5 try to find as many ways as they can to improve their English. 10 6 will ask their teacher if they have a question about English. 62 7 know the method which suits them best and use it. 12 8 want to study in an English-speaking environment if they have a chance. 15 9 do not know their good points and weaknesses. 19 10 do not try to improve their weaknesses. 47 11 do not do things actively. 25 12 do not think about their progress in learning English 22 13 learn English because they enjoy studying it. 24 14 have a strong will and are not easily influenced by others. 1 15 study English voluntarily. 30 16 are determined to achieve the target which they set for their English study. 31 17 like to study with computers. 32 18 make their schedule so they will have enough time to study English. 33 19 will do a search on the Internet if they have a question about English. 34 20 will act immediately if they decide to do something. 17 21 try to develop their good points. 36 22 reflect on what they learn and look for something important. 37 23 use time effectively. 38 24 notice their mistakes and use that information to help them do better. 39 25 try to use other methods if one method of English study does not suit them. 42 26 go online as a way of learning English. 43 27 check to make sure that they understood the lesson. 13 28 think about their methods of learning English & whether those methods are good. 45 29 do not want to be good English learners. 46 30 pay more attention when they see an English website. 48 31 try to study English regularly even with limited time. 49 32 study things which were not from their class. 41 33 want to find a job where only English is used in the future. 27 34 consider a lot about study/work before doing it. 54 35 try to complete the things they have decided to do. 55 36 think English is not important for their future. 56 37 have no plans about how to learn English. 29 38 change their study contents & target according to their needs. 58 39 are not aware of their studies. 4 40 want to communicate with foreigners in English. 60 41 do not practice English with people outside class.

Note: The 16 items which are shaded are used as dummy items in Study Two. Responses to these items are not included in the analysis.

______Appendix 3E Page 260 Appendix 3F

SEMI-STRUCTURED INTERVIEW QUESTIONS FOR STUDENTS

I. Ability to initiate EFL learning

1. Do you have any foreign friends? How do you know them and how do you often talk to them? Why do you do so? What do you think about that? 2. Do you often look for opportunities to communicate in English such as making friends with English-speaking people, speaking English to friends, going to the English speaking clubs, reading newspapers in English? Why do/don’t you do that? What encourages or prevent you from continuing to do that? 3. Do you often read the textbooks carefully? Why (not)? Do you read other materials? Why (not)? How can you get them? Why do you choose them? How do the extra materials help your learning in school?

II. Ability to monitor EFL learning process

1. Do you make a plan for your EFL learning such as scheduling what to be done within a week, setting deadline for a learning activity, putting important events in personal calendar? Why do/don’t you do that? 2. Do you often stick to a learning plan? Do you often make modifications to your original plans such as changing the deadline of a task or replacing an activity with another? How often do you do that? Why do (not) you do that? What does that mean to your learning? Can you give some examples? 3. When you take part in a learning activity, what do you actually do to manage that activity? Can you give some examples to illustrate?

III. Ability to set learning goals and evaluate learning

1. Do you often include goals in your learning plans? Why and why not? Do you have any general or specific goal for your learning activities? 2. Do you often check your learning progress such as asking friends to evaluate your pronunciation, comparing your work with others’, talking to your teachers about the evaluation of your tasks, taking practice tests that you see? How often do you do that? Why do (not) you do that? Do you learn anything from that? Can you give some examples? 3. Do you often check your tasks/assignments carefully such as reading your assignments carefully before submitting, reading aloud a dialog carefully before schools, recording your pronunciation and evaluate that? Do you find it helpful? What motivates you (not) to do so?

______Appendix 3F Page 261 IV. Ability to use ICTs for EFL learning purposes

1. Do you often use the LMS? Why so? Is there anything that you are particularly interested in? Do you have a plan for studying with the LMS? Do you use it to keep tracks of what you have done or remind you of your work? Does the communication from that environment make you spend more time on this course than the others? What are the specific examples that you can give? 2. Does any online activity of the course help you with an opportunity to start learning some English like getting a new website address, downloading some materials, triggering some communication, providing some clue for further work? How did you find them? Give some examples to illustrate. 3. Does the course website help you evaluate your online activities such as giving you more opportunities to review your tasks, allowing your friends/teacher evaluate your work? What are the situations that the online component can help you evaluate your work? Is it easier for your work to be evaluated online than in class?

______Appendix 3F Page 262 Appendix 3G ETHICS APPROVAL AND RELATED DOCUMENTS

1. Approval of La Trobe University Human Ethics Committee

______Appendix 3G Page 263

______Appendix 3G Page 264 2. Letter of Acceptance from Participating University

______Appendix 3G Page 265 3. Information Sheet

STUDENT PARTICIPANT INFORMATION SHEET

Project Title: Learner Autonomy: The Roles of Activity Type and Delivery Mode in Conducting EFL/ESL Activities Researcher: Tan Tin Dang Supervisors: Prof. Margaret Robertson & Dr. Alan Williams Course: Doctor of Philosophy HREC Number: 08-148

The project “Learner Autonomy: The Roles of Activity Type and Delivery Mode in Conducting EFL/ESL Activities” is conducted by Mr. Tan Tin Dang during his PhD candidature, under the supervision of Professor Margaret Robertson and Dr. Alan Williams, Faculty of Education, La Trobe University, Australia. This project is funded by La Trobe University Postgraduate Research Scholarship (LTUPRS) and Endeavour International Postgraduate Research Scholarship (EIPRS) during the period of 2008 – 2011. It aims at investigating the need for learner autonomy in ESL/EFL learning in Vietnam by examining the impact of activity types and delivery modes.

The project has three principal objectives. First, it examines students’ perception of learner autonomy (that is, what kinds of control do they want to take over their learning and to what degree do they want to do it?) and their actual level of learner autonomy (that is, how much can they currently control their learning processes?) in their EFL/ESL learning activities. Second, it investigates the impact of different activity types on promoting learner autonomy (that is, how much can an activity like role play or group discussion help students control their learning process?). Third, it looks for possible effects of computer-mediated instruction on learner autonomy (that is, do students control their learning in virtual environment differently from in traditional classroom?).

This project will provide an opportunity for students to learn ESL/EFL with different kinds of activities in both classroom and computer-mediated instruction. It also helps them reflect on their learning process and understand more about their learning behaviors. In addition, it provides hints and clues for students to independently take advantage of the technology infrastructure in their university. The ability of controlling their learning is therefore reinforced. This set of skills is potentially not only useful for students during their schooling but also important for their future professional development and social life.

Participation in the project will involve responding to a questionnaire twice, at the beginning and at the end of a 16-18 week course. The questionnaire items ask the participants to give ratings on a number of statements about the way that students control their learning. The questionnaire is written in Vietnamese and will be completed during class time. This will probably take 20 minutes and the researcher will be in the class to answer all the questions or concerns that the participants have. In addition, some of the participants will randomly be

______Appendix 3G Page 266 selected for individual interviews with the researcher. This will probably be 20-30 minutes long, out of class time and in the school area. The interviewees will be asked about their reflections on learning, preferences on different activities, and habitual learning behaviors.

The questionnaire data will be quantitatively analyzed. The interview data will be transcribed and examples be translated into English by the researcher. Only the researcher and possibly his supervisors will know participants’ identity. Their personal information will be kept confidential, and the data will be kept in a locked cabinet in the university and password protected computer file. The results of the research will be published in a thesis and may also be published in a journal article or presented at a conference. However, at no time will participants’ identity be revealed. The results of the study and personal data collected during the course of research will be available to the participants on request.

Participation in this project is entirely voluntary and every participant may withdraw from active participation in the project at any time during the data collection period. In the event that a line of questioning does develop in such a way that participants feel hesitant or uncomfortable, they will be reminded of their right to decline to answer any particular question(s). They also have the right to request that data arising from their participation are not used in the research project, provided that this right is exercised within four weeks of the completion of the data collection period. If they wish to withdraw from the project, they are asked to complete the “Withdrawal of Consent Form” or to notify the researcher by e-mail or telephone that they wish to withdraw their consent for their data to be used in this research project. There are no disadvantages or adverse consequences for not participating or for withdrawing prematurely from the research.

Any questions regarding this project may be directed to the researcher, Mr. Tan Tin Dang, email address: [email protected] or telephone: (+61) 423 591 206, or Prof. Margaret Robertson, Associate Dean Research, Faculty of Education, email address: [email protected] or telephone: (+61 3) 9479 5616.

If the participants have any complaints or queries that the researcher has not been able to answer to their satisfaction, they may contact the Secretary, Human Ethics Committee, La Trobe University, Victoria, 3552, e-mail: [email protected].

In order to participate in this study, participants are requested to complete a signed consent form and return to the researcher.

Thank you for understanding and cooperation.

______Appendix 3G Page 267

LECTURER PARTICIPANT INFORMATION SHEET

Project Title: Learner Autonomy: The Roles of Activity Type and Delivery Mode in Conducting EFL/ESL Activities Researcher: Tan Tin Dang Supervisors: Prof. Margaret Robertson & Dr. Alan Williams Course: Doctor of Philosophy HREC Number: 08-148

The project “Learner Autonomy: The Roles of Activity Type and Delivery Mode in Conducting EFL/ESL Activities” is conducted by Mr. Tan Tin Dang during his PhD candidature, under the supervision of Professor Margaret Robertson and Dr. Alan Williams, Faculty of Education, La Trobe University, Australia. This project is funded by La Trobe University Postgraduate Research Scholarship (LTUPRS) and Endeavour International Postgraduate Research Scholarship (EIPRS) during the period of 2008 – 2011. It aims at investigating the need for learner autonomy in ESL/EFL learning in Vietnam by examining the impact of activity types and delivery modes.

The project has three principal objectives. First, it examines students’ perception of learner autonomy (that is, what kinds of control do they want to take over their learning and to what degree do they want to do it?) and their actual level of learner autonomy (that is, how much can they currently control their learning processes?) in their EFL/ESL learning activities. Second, it investigates the impact of different activity types on promoting learner autonomy (that is, how much can an activity like role play or group discussion help students control their learning process?). Third, it looks for possible effects of computer-mediated instruction on learner autonomy (that is, do students control their learning in virtual environment differently from in traditional classroom?).

This project will provide an opportunity for lecturers to try different kinds of activities in both classroom and computer-mediated instruction during their ESL/EFL courses. It also helps lecturers reflect on their teaching process and understand more about their students’ learning behaviors. In addition, it will provide experimental experience to develop learner autonomy in their students. Moreover, lecturers will have a chance to integrate the current technology infrastructure in the university into their ESL/EFL teaching. These are potentially not only useful for their current teaching practices but also important for their professional development and future experiments.

Participation in the project will involve organizing and conducting different kinds of activities in ESL/EFL courses, ranging from high to low levels of control, during a 16-18 week semester. The high-controlled activity such as repetition and drill requires students to perform the language items or dialogues exactly as indicated in the activities; meanwhile, the medium- and low-controlled activity such as group discussion or topic development provides students an opportunity to look for different references, use their personal ideas, negotiate with their peers, and present it in the way they like.

______Appendix 3G Page 268 There will be four groups in this project. The first one is a control group in which the lecturer just conducts his/her class as usual. There is no treatment from the project to this group. The lecturer of the second group provides his/her students with high- and medium-controlled activities during the course. The lecturer of the third group conducts medium- and low- controlled activities during the course. The lecturer of the fourth group uses all high-, medium- and low-controlled activities in both the classroom and online activities during the course. The online component employs a Moodle package which is installed and developed by the researcher. The online activities will be designed and added in discussion with the involved lecturer along the course. All technical supports will be provided by the researcher.

Participating lecturers will be asked to teach in one of the three modes. Discussion with the researcher will be done throughout the courses to achieve assistance and support when necessary to maintain the appropriateness of the treatment. No other kinds of data will be collected. The description of each kind of treatment will be employed for analyses on students’ learner autonomy within the project; but the identity of the lecturers will be completely removed from statistics. Their personal information will be kept confidential and the data will be kept in a password protected computer file. The results of the research will be published in a thesis and may also be published in a journal article or presented at a conference. However, at no time will participants’ identity be revealed. The results of the study and personal data collected during the course of research will be available to any participant on request.

Participation in this project is entirely voluntary and every participant may withdraw from active participation in the project at any time during the data collection period. They also have the right to request that data arising from their participation are not used in the research project, provided that this right is exercised within four weeks of the completion of the data collection period. If they wish to withdraw from the project, they are asked to complete the “Withdrawal of Consent Form” or to notify the researcher by e-mail or telephone that they wish to withdraw their consent for their data to be used in this research project. There are no disadvantages or adverse consequences for not participating or for withdrawing prematurely from the research.

Any questions regarding this project may be directed to the researcher, Mr. Tan Tin Dang, email address: [email protected] or telephone: (+61) 423 591 206, or Prof. Margaret Robertson, Associate Dean Research, Faculty of Education, email address: [email protected] or telephone: (+61 3) 9479 5616.

If the participants have any complaints or queries that the researcher has not been able to answer to their satisfaction, they may contact the Secretary, Human Ethics Committee, La Trobe University, Victoria, 3552, e-mail: [email protected].

In order to participate in this study, participants are requested to complete a signed consent form and return to the researcher.

Thank you for understanding and cooperation.

______Appendix 3G Page 269

UNIVERSITY PARTICIPANT INFORMATION SHEET

Project Title: Learner Autonomy: The Roles of Activity Type and Delivery Mode in Conducting EFL/ESL Activities Researcher: Tan Tin Dang Supervisors: Prof. Margaret Robertson & Dr. Alan Williams Course: Doctor of Philosophy HREC Number: 08-148

The project “Learner Autonomy: The Roles of Activity Type and Delivery Mode in Conducting EFL/ESL Activities” is conducted by Mr. Tan Tin Dang during his PhD candidature, under the supervision of Professor Margaret Robertson and Dr. Alan Williams, Faculty of Education, La Trobe University, Australia. This project is funded by La Trobe University Postgraduate Research Scholarship (LTUPRS) and Endeavour International Postgraduate Research Scholarship (EIPRS) during the period of 2008 – 2011. It aims at investigating the need for learner autonomy in ESL/EFL learning in Vietnam by examining the impact of activity types and delivery modes.

The project has three principal objectives. First, it examines students’ perception of learner autonomy (that is, what kinds of control do they want to take over their learning and to what degree do they want to do it?) and their actual level of learner autonomy (that is, how much can they currently control their learning processes?) in their EFL/ESL learning activities. Second, it investigates the impact of different activity types on promoting learner autonomy (that is, how much can an activity like role play or group discussion help students control their learning process?). Third, it looks for possible effects of computer-mediated instruction on learner autonomy (that is, do students control their learning in virtual environment differently from in traditional classroom?).

This project will provide lecturers an opportunity to try different kinds of activities in both classroom and computer-mediated instruction during their ESL/EFL courses. It also helps lecturers reflect on their teaching process and understand more about their students’ learning behaviors. In addition, it will provide experimental experience to develop learner autonomy in their students. Moreover, lecturers have a chance to integrate the current technology infrastructure in the university into their ESL/EFL teaching. These are potentially not only useful for their current teaching practices but also important for their professional development and future experiments.

Participation of the lecturers in the project will involve organizing and conducting different kinds of activities in ESL/EFL courses, ranging from high to low levels of control, during a 16-18 week semester. The high-controlled activity such as repetition and drill requires students to perform the language items or dialogues exactly as indicated in the activities; meanwhile, the medium- and low-controlled activity such as group discussion or topic development provides students an opportunity to look for different references, use their personal ideas, negotiate with their peers, and present it in the way they like.

______Appendix 3G Page 270 There will be four groups in this project. The first one is a control group in which the lecturer just conducts his/her class as usual. There is no treatment from the project to this group. The lecturer of the second group provides his/her students with high- and medium-controlled activities during the course. The lecturer of the third group conducts medium- and low- controlled activities during the course. The lecturer of the fourth group uses all high-, medium- and low-controlled activities in both the classroom and online activities during the course. The online component employs a Moodle package which is installed and developed by the researcher. The online activities will be designed and added in discussion with the involved lecturer along the course. All technical supports will be provided by the researcher.

Participating lecturers will be asked to teach in one of the three modes. Discussion with the researcher will be done throughout the courses to achieve assistance and support when necessary to maintain the appropriateness of the treatment. No other kinds of data will be collected. The description of each kind of treatment will be employed for analyses on students’ learner autonomy within the project; but the identity of the lecturers will be completely removed from statistics. Their personal information will be kept confidential and the data will be kept in a password protected computer file. The results of the research will be published in a thesis and may also be published in a journal article or presented at a conference. However, at no time will participants’ identity be revealed. The results of the study and personal data collected during the course of research will be available to any lecturer participant on request.

In addition, this project will provide students an opportunity to learn ESL/EFL with different kinds of activities in both classroom and computer-mediated instruction. It also helps them reflect on their learning process and understand more about their learning behaviors. In addition, it provides hints and clues for students to independently take advantage of the technology infrastructure in their university. The ability of controlling their learning is therefore reinforced. This set of skills is potentially not only useful for students during their schooling but also important for their future professional development and social life.

Participating students will involve responding to a questionnaire twice, at the beginning and at the end of a 16-18 week course. The questionnaire items ask the participants to give ratings on a number of statements about the way that students control their learning. The questionnaire is written in Vietnamese and will be completed during class time. This will probably take 20 minutes and the researcher will be in the class to answer all the questions or concerns that the participants have. In addition, some of the participants will randomly be selected for individual interviews with the researcher. This will probably be 20-30 minutes long, out of class time and in the school area. The interviewees will be asked about their reflections on learning, preferences on different activities, and habitual learning behaviors.

The questionnaire data will be quantitatively analyzed. The interview data will be transcribed and examples be translated into English by the researcher. Only the researcher and possibly his supervisors will know participants’ identity. Their personal information will be kept confidential, and the data will be kept in a locked cabinet in the university and password protected computer file. The results of the research will be published in a thesis and may also be published in a journal article or presented at a conference. However, at no time will participants’ identity be revealed. The results of the study and personal data collected during the course of research will be available to the student participants on request.

______Appendix 3G Page 271 Participation in this project is entirely voluntary and every participant may withdraw from active participation in the project at any time during the data collection period. In the event that a line of questioning or treatment details does develop in such a way that participants feel hesitant or uncomfortable to comply, they will be reminded of their right to decline to perform any particular requirement(s) from the treatment. They also have the right to request that data arising from their participation are not used in the research project, provided that this right is exercised within four weeks of the completion of the data collection period. If they wish to withdraw from the project, they are asked to complete the “Withdrawal of Consent Form” or to notify the researcher by e-mail or telephone that they wish to withdraw their consent for their data to be used in this research project. There are no disadvantages or adverse consequences for not participating or for withdrawing prematurely from the research. An overall report on the result of the study will be available to the school upon request.

Any questions regarding this project may be directed to the researcher, Mr. Tan Tin Dang, email address: [email protected] or telephone: (+61) 423 591 206, or Prof. Margaret Robertson, Associate Dean Research, Faculty of Education, email address: [email protected] or telephone: (+61 3) 9479 5616.

If the participants have any complaints or queries that the researcher has not been able to answer to their satisfaction, they may contact the Secretary, Human Ethics Committee, La Trobe University, Victoria, 3552, e-mail: [email protected].

In order to participate in this study, the school is requested to write to the researcher a letter, indicating that the school would like to take part in the project and allow him to come to the school for data collection.

Thank you for understanding and cooperation.

______Appendix 3G Page 272 4. Invitation letters

STUDENT PARTICIPANT INVITATION LETTER

Project Title: Learner Autonomy: The Roles of Activity Type and Delivery Mode in Conducting EFL/ESL Activities Researcher: Tan Tin Dang Supervisors: Prof. Margaret Robertson & Dr. Alan Williams Course: Doctor of Philosophy HREC Number: 08-148

Dear Students,

My name is Dang Tan Tin, an EFL/ESL lecturer in Vietnam for several years. I am currently conducting a project entitled “Learner Autonomy: The Roles of Activity Type and Delivery Mode in Conducting EFL/ESL Activities” during my PhD candidature, under the supervision of Professor Margaret Robertson and Dr. Alan Williams, Faculty of Education, La Trobe University, Australia. This project is funded by La Trobe University Postgraduate Research Scholarship (LTUPRS) and Endeavour International Postgraduate Research Scholarship (EIPRS) during the period of 2008 – 2011. It aims at investigating the need for learner autonomy in ESL/EFL learning in Vietnam by investigating the impact of activity types and delivery modes.

Part of my project is the collection of data from EFL/ESL second year students. It includes a questionnaire, asking about students’ perception of learner autonomy (that is, what kinds of control do they want to take over their learning and to what degree do they want to do it?) and their actual level of learner autonomy (that is, how much can they currently control their learning processes?) in their EFL/ESL learning activities. The other kind of data involves an individual interview with me, talking about the impact of different activity types and delivery modes on learner autonomy development. Should you be interested in the project, please have a look at the information sheet attached for more details.

If you are an ESL/EFL student in the second year and would like to participate in the project, please keep me informed and you will be invited. Participation in this project is entirely voluntary and every participant may withdraw from active participation in the project at any time during the data collection period. There are no disadvantages or adverse consequences for not participating or for withdrawing prematurely from the research.

Thank you very much for your understanding and cooperation.

Yours sincerely, Dang Tan Tin PhD Student Faculty of Education, La Trobe University Email: [email protected] Phone: (+61) 423 591 206

______Appendix 3G Page 273

LECTURER PARTICIPANT INVITATION LETTER

Project Title: Learner Autonomy: The Roles of Activity Type and Delivery Mode in Conducting EFL/ESL Activities Researcher: Tan Tin Dang Supervisors: Prof. Margaret Robertson & Dr. Alan Williams Course: Doctor of Philosophy HREC Number: 08-148

Dear Lecturers,

My name is Dang Tan Tin, an EFL/ESL lecturer in Vietnam for several years. I am currently conducting a project entitled “Learner Autonomy: The Roles of Activity Type and Delivery Mode in Conducting EFL/ESL Activities” during my PhD candidature, under the supervision of Professor Margaret Robertson and Dr. Alan Williams, Faculty of Education, La Trobe University, Australia. This project is funded by La Trobe University Postgraduate Research Scholarship (LTUPRS) and Endeavour International Postgraduate Research Scholarship (EIPRS) during the period of 2008 – 2011. It aims at investigating the need for learner autonomy in ESL/EFL learning in Vietnam by investigating the impact of activity types and delivery modes.

Part of my project is the implementation of the treatments for data collection in EFL/ESL courses. These treatments involve organizing and conducting different kinds of activities, ranging from high to low levels of control, during a 16-18 week semester. The high-controlled activity such as repetition and drill requires students to perform the language items or dialogues exactly as indicated in the activities; meanwhile, the medium- and low-controlled activity such as group discussion or topic development provides students an opportunity to look for different references, use their personal ideas, negotiate with their peers, and present it in the way they like. An online component, employing a Moodle package, is also included during the treatments. The online activities will be designed and added in discussion with me. I will provide all technical supports during the treatments. Should you be interested in the project, please have a look at the information sheet attached for more details.

If you are an ESL/EFL lecturer, currently working with second year students and would like to participate in the project, please keep me informed and you will be invited. Participation in this project is entirely voluntary and every participant may withdraw from active participation in the project at any time during the data collection period. There are no disadvantages or adverse consequences for not participating or for withdrawing prematurely from the research.

Thank you very much for your understanding and cooperation.

Yours sincerely, Dang Tan Tin PhD Student, Faculty of Education, La Trobe University Email: [email protected] Phone: (+61) 423 591 206

______Appendix 3G Page 274

UNIVERSITY PARTICIPANT INVITATION LETTER

Project Title: Learner Autonomy: The Roles of Activity Type and Delivery Mode in Conducting EFL/ESL Activities Researcher: Tan Tin Dang Supervisors: Prof. Margaret Robertson & Dr. Alan Williams Course: Doctor of Philosophy HREC Number: 08-148

Dear Faculty of English, Ho Chi Minh City University of Social Sciences and Humanities,

My name is Dang Tan Tin, an EFL/ESL lecturer in Vietnam for several years. I am currently conducting a project entitled “Learner Autonomy: The Roles of Activity Type and Delivery Mode in Conducting EFL/ESL Activities” during my PhD candidature, under the supervision of Professor Margaret Robertson and Dr. Alan Williams, Faculty of Education, La Trobe University, Australia. This project is funded by La Trobe University Postgraduate Research Scholarship (LTUPRS) and Endeavour International Postgraduate Research Scholarship (EIPRS) during the period of 2008 – 2011. It aims at investigating the need for learner autonomy in ESL/EFL learning in Vietnam by investigating the impact of activity types and delivery modes.

Part of my project is the implementation of the treatments for data collection in EFL/ESL courses. These treatments, provided by the participating lecturers, will involve organizing and conducting different kinds of activities, ranging from high to low levels of control, during a 16-18 week semester. An online component, employing a Moodle package, will also be included during the treatments. The students in this project will be asked to respond to a questionnaire twice, at the beginning and at the end of a 16-18 week course. The questionnaire items ask them to give ratings on a number of statements about the way that students control their learning. In addition, some of them will randomly be selected for individual interviews with the researcher. They will be asked about their reflections on learning, preferences on different activities, and habitual learning behaviors. All details are provided in the Information Sheet attached.

Therefore, I am writing this letter to ask for your permission to come to your faculty and invite ESL/EFL lecturers and second year students to participate in the project. If you agree, please write me a letter, indicating that I am allowed to work with your teaching staff and students to collect data for my project. Your support is highly appreciated. Thank you very much for your understanding and cooperation.

Yours sincerely, Principal supervisor of the project Dang Tan Tin Prof. Margaret Robertson PhD Student Associate Dean, Research Faculty of Education, La Trobe University Faculty of Education, La Trobe Uni. Email: [email protected] Email: [email protected] Phone: (+61) 423 591 206 Phone: (+61 3) 9479 5616

______Appendix 3G Page 275 5. Consent Form

STUDENT PARTICIPANT CONSENT FORM

Project Title: Learner Autonomy: The Roles of Activity Type and Delivery Mode in Conducting EFL/ESL Activities Researcher: Tan Tin Dang Supervisors: Prof. Margaret Robertson & Dr. Alan Williams Course: Doctor of Philosophy HREC Number: 08-148

I, ……...... …………………………., have read and understood the information sheet provided by the researcher concerning this activity, and any questions I have asked have been answered to my satisfaction. a. I agree to answer the questionnaires. b. I agree to interviews being audio-recorded. c. I agree that the research data collected for the study may be published or provided to other researchers on the condition that my name is not used, and that I cannot be identified in any other way. d. I agree to participate in the project, realizing that I may physically withdraw from the study at any time during the data collection period and may request that no data arising from my participation are used, up to four weeks following the completion of my participation in the research. e. A copy of the information sheet for this project and this form has been provided to me to keep.

Name of Participant ...... Email ...... Signature ...... Date......

Name of researcher ...... Signature ...... Date ......

Names of supervisors...... Signatures ...... Date ......

______Appendix 3G Page 276

LECTURER PARTICIPANT CONSENT FORM

Project Title: Learner Autonomy: The Roles of Activity Type and Delivery Mode in Conducting EFL/ESL Activities Researcher: Tan Tin Dang Supervisors: Prof. Margaret Robertson & Dr. Alan Williams Course: Doctor of Philosophy HREC Number: 08-148

I, ……...... …………………………., have read and understood the information sheet provided by the researcher concerning this activity, and any questions I have asked have been answered to my satisfaction. a. I agree to work with the researcher to implement the treatment provided by the study. b. I agree that the research data collected for the study may be published or provided to other researchers on the condition that my name is not used, and that I cannot be identified in any other way. c. I agree to participate in the project, realizing that I may physically withdraw from the study at any time during the data collection period and may request that no data arising from my participation are used, up to four weeks following the completion of my participation in the research. d. A copy of the information sheet for this project and this form has been provided to me to keep.

Name of Participant ...... Email ...... Signature ...... Date......

Name of researcher ...... Signature ...... Date ......

Names of supervisors...... Signatures ...... Date ......

______Appendix 3G Page 277 6. Withdrawal Form

STUDENT WITHDRAWAL OF CONSENT FORM

Project Title: Learner Autonomy: The Roles of Activity Type and Delivery Mode in Conducting EFL/ESL Activities Researcher: Tan Tin Dang Supervisors: Prof. Margaret Robertson & Dr. Alan Williams Course: Doctor of Philosophy HREC Number: 08-148

I, ……...... …………………………., wish to WITHDRAW my consent to the use of data arising from my participation in the project. Data arising from my participation must NOT be used in this research project as described in the Student Participation Information Sheet and Consent Form. I understand that data arising from my participation will be destroyed provided this request is received within four weeks of the completion of my participation in this project. I also understand that this notification will be retained together with my consent form as evidence of the withdrawal of my consent to use the data I have provided specifically for this research project.

Name of Participant ...... Email ...... Signature ...... Date......

Name of researcher ...... Signature ...... Date ......

Names of supervisors...... Signatures ...... Date ......

______Appendix 3G Page 278

LECTURER WITHDRAWAL OF CONSENT FORM

Project Title: Learner Autonomy: The Roles of Activity Type and Delivery Mode in Conducting EFL/ESL Activities Researcher: Tan Tin Dang Supervisors: Prof. Margaret Robertson & Dr. Alan Williams Course: Doctor of Philosophy HREC Number: 08-148

I, ……...... …………………………., wish to WITHDRAW my consent to the use of data arising from my participation in the project. Data arising from my participation must NOT be used in this research project as described in the Student Participation Information Sheet and Consent Form. I understand that data arising from my participation will be destroyed provided this request is received within four weeks of the completion of my participation in this project. I also understand that this notification will be retained together with my consent form as evidence of the withdrawal of my consent to use the data I have provided specifically for this research project.

Name of Participant ...... Email ...... Signature ...... Date......

Name of researcher ...... Signature ...... Date ......

Names of supervisors...... Signatures ...... Date ......

______Appendix 3G Page 279 Appendix 4A

DESCRIPTIVE STATISTICS OF THE 62 QUESTIONNAIRE ITEMS

(N=562) No Question items in brief Min* Max* Mean SD Skew 1 study English voluntarily 1 5 4.18 .929 -1.204 2 give themselves a reward or treat when they do well in E 1 5 2.73 1.088 .320 3 Use the Internet to learn things which are not taught in class 1 5 3.77 1.062 -.634 4 want to communicate with foreigners in E 1 5 4.37 .894 -1.528 5 carry out the learning plans once they have been made 1 5 3.90 .982 -.589 6 make good use of materials & resources when studying E at home 1 5 3.85 1.070 -.581 7 look for opportunities to use E as much as possible 1 5 4.16 1.050 -1.179 8 want to improve E by joining online communities such as forum, blogs 1 5 4.12 1.156 -1.216 9 try to find as many ways as they can to improve their E 1 5 4.26 .905 -1.241 10 will ask their teacher if they have a question about E 1 5 3.83 1.022 -.579 11 look for different resources on the Internet 1 5 4.48 .884 -1.943 12 want to study in an English-speaking envi if they have a chance 1 5 4.35 .920 -1.683 13 think of their methods of learning E & if those methods are good 1 5 3.77 .964 -.391 14 try to have a good environ. to study E, e.g. cleaning up the table 1 5 2.59 1.161 .336 15 know their good points and weaknesses 1 5 4.11 1.139 -1.135 16 try to complete some online activities even with limited time 1 5 2.97 1.139 .016 17 try to develop their good points 1 5 3.97 .960 -.857 18 will work out a question about E by themselves if they have 1 5 2.80 1.098 .230 19 try to improve their weaknesses 1 5 4.39 1.021 -1.644 20 learn things that the teachers do not give as a task 1 5 4.18 1.037 -1.223 21 write down feelings towards E studies in a language learning diary 1 5 2.57 1.282 .352 22 learn English because they enjoy studying it 1 5 3.69 1.069 -.546 23 share their feelings towards English studies with friends online 1 5 2.77 1.197 .099 24 have a strong will and are not easily influenced by others 1 5 3.33 1.083 -.180 25 think about their progress in learning English 1 5 4.15 1.056 -1.131 26 concentrate on their studies even if they happen to hear others talking 1 5 3.11 1.085 -.081 27 consider a lot about study/work before doing it 1 5 3.57 1.071 -.400 28 pay attention when someone is speaking in English 1 5 4.00 .926 -.767 29 change their study content & target according to their needs 1 5 3.42 1.090 -.344 30 are determined to achieve the target which they set for their E study 1 5 3.88 1.031 -.645 31 like to study with computers 1 5 3.68 1.073 -.385 32 make their schedule so they will have enough time to study E 1 5 3.64 1.091 -.569 33 will do a search on the Internet if they have a question about E 1 5 3.79 1.050 -.596 34 will act immediately if they decide to do something 1 5 3.51 1.128 -.299 35 check E level by comparing it with the skills of others on the Internet 1 5 3.15 1.156 -.225 36 reflect on what they learn and look for something important 1 5 3.70 1.011 -.573 37 use time effectively 1 5 3.74 1.178 -.555 38 notice their mistakes & use that information to help them do better 1 5 3.90 .955 -.678 39 try to use other methods if one method of E study does not suit them 1 5 3.97 .891 -.596 40 feel like studying once the class starts 1 5 2.99 1.080 .006 41 want to find a job where only E is used in the future 1 5 3.78 .985 -.332 42 go online as a way of learning English 1 5 3.83 1.079 -.683 43 check to make sure that they understood the lesson 1 5 3.70 .945 -.418 44 pay more attention to the lessons that they are not good at 1 5 3.96 .986 -.847 45 want to be good English learners 1 5 4.64 .912 -2.857 46 pay more attention when they see an English website 1 5 3.37 1.097 -.309 47 do things actively 1 5 4.25 1.076 -1.348 48 try to study English regularly even with limited time 1 5 3.80 .980 -.587 49 study things which were not from their class 1 5 3.75 1.135 -.698 50 will talk with their friends/relatives if they have a question about E 1 5 3.87 .971 -.797 51 study English no matter what mood they are in 1 5 2.56 1.237 .447 52 will talk to a foreigner in E if they come across him/her in the street 1 5 3.77 1.122 -.651 53 will use E to communicate if seeing a stranger on the Internet 1 5 2.78 1.223 .177 54 try to complete the things they have decided to do 1 5 3.64 1.056 -.506

______Appendix 4A Page 280 55 think English is important for their future 1 5 4.62 .956 -2.640 56 have plans about how to learn English 1 5 4.41 1.076 -1.852 57 check their E proficiency by taking TOEFL or IELTS voluntarily 1 5 3.23 1.394 -.323 58 are aware of their studies 1 5 4.52 1.005 -2.208 59 dream of being good English speakers 1 5 4.42 .982 -1.925 60 practice English with people outside class 1 5 4.24 1.022 -1.297 61 find information about English by themselves 1 5 4.36 1.063 -1.706 62 know the method which suits them best and use it 1 5 3.99 1.067 -.938 * Min and Max values rated by the participants (1 = never or almost never true; 5 = always or almost always true) Shading: items with a skew value out of the range between 2 and -2

______Appendix 4A Page 281 Appendix 4B FACTOR ANALYSIS PROCEDURES

Table 1: An extract of the total variance explained when 62 items were included

Total Variance Explained Rotation Sums Initial Eigenvalues Extraction Sums of Squared Loadings of Squared Component a Loadings Total % of Variance Cumulative % Total % of Variance Cumulative % Total 1 12.645 20.394 20.394 12.645 20.394 20.394 11.087

2 5.134 8.281 28.675 5.134 8.281 28.675 5.498 3 3.192 5.148 33.823 3.192 5.148 33.823 7.743 4 1.912 3.084 36.907 1.912 3.084 36.907 2.308

dimension 5 1.876 3.025 39.932

62 .242 .390 100.000 Extraction Method: Principal Components Analysis. a. When components are correlated, sums of squared loadings cannot be added to obtain a total variance.

Table 2: Loadings of the 62 items on the four factors

Structure Matrix Component Students who succeed best with learning English 1 2 3 4 37: use time effectively .758 49: study things which were not from their class .703 32: make their schedule so they will have enough time to study E .697 .337 30: are determined to achieve the target which they set for their E study .685 .361 48: try to study English regularly even with limited time .662 .368 62: know the method which suits them best and use it .661 6: make good use of materials & resources when studying E at home .655 .352 38: notice their mistakes & use that information to help them do better .639 .435 36: reflect on what they learn and look for something important .630 .433 43: check to make sure that they understood the lesson .617 .327 10: will ask their teacher if they have a question about E .597 .407 7: look for opportunities to use E as much as possible .591 .336 .422 39: try to use other methods if one method of E study does not suit them .578 .433 57: check their E proficiency by taking TOEIC, TOEFL or IELTS voluntarily .569 5: carry out the learning plans once they have been made .535 21: write down their feelings towards E studies in a language learning diary .533 23: share their feelings towards English studies with friends online .518 24: have a strong will and are not easily influenced by others .513 44: pay more attention to the lessons that they are not good at .506 .450 35: check their E level by comparing it with the skills of others on the Internet .490 .307 26: concentrate on their studies even if they happen to hear others talking .467 52: will talk to a foreigner in E if they come across him/her in the street .460 .383 20: learn things that the teachers do not give as a task .448 -.433 50: will talk with their friends/relatives if they have a question about E .445 .412 34: will act immediately if they decide to do something .387 .376 -.339 40: feel like studying once the class starts .308 .308 56: have plans about how to learn English -.682 55: think English is important for their future -.666 .314 58: are aware of their studies -.642 45: want to be good English learners -.628 .351 19: try to improve their weaknesses .326 -.583

______Appendix 4B Page 282 47: do things actively .435 -.580 61: find information about English by themselves -.572 60: practice English with people outside class -.539 25: think about their progress in learning English -.518 15: know their good points and weaknesses .324 -.514 51: study English no matter what mood they are in .490 53: will use E to communicate if seeing a stranger on the Internet .413 14: try to have a good environ. to study E, e.g. cleaning up the table before studying E .392 16: try to complete some online activities even with limited time .383 .353 2: give themselves a reward or treat when they do well in E 12: want to study in an English-speaking environment if they have a chance .685 9: try to find as many ways as they can to improve their E .391 .561 28: pay attention when someone is speaking in English .560 31: like to study with computers .559 46: pay more attention when they see an English website .314 .542 33: will do a search on the Internet if they have a question about E .320 .540 17: try to develop their good points .412 .536 42: go online as a way of learning English .521 22: learn English because they enjoy studying it .494 59: dream of being good English speakers .492 4: want to communicate with foreigners in E .486 .375 41: want to find a job where only E is used in the future .408 13: think about their methods of learning E & whether those methods are good .378 .393 11: look for different resources on the Internet -.381 .393 3: use the Internet to learn things which are not taught in class .351 .357 .338 29: change their study content & target according to their needs .347 -.303 1: study English voluntarily .343 54: try to complete the things they have decided to do .435 .344 -.478 27: consider a lot about study/work before doing it .316 .361 -.424 8: want to improve E by taking part in online communities such as forums, blogs... .309 18: will work out a question about E by themselves if they have Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalization. Factor loadings sorted by size Loadings smaller than .30 were removed for clarity Shading: Items with factor loadings smaller than .40 and should be removed

Table 3: An extract of the total variance explained when 50 items were included

Total Variance Explained Rotation Sums Initial Eigenvalues Extraction Sums of Squared Loadings of Squared Component a Loadings Total % of Variance Cumulative % Total % of Variance Cumulative % Total 1 11.692 23.383 23.383 11.692 23.383 23.383 9.159

2 4.211 8.423 31.806 4.211 8.423 31.806 6.197 3 3.000 6.000 37.806 3.000 6.000 37.806 4.982 4 1.629 3.258 41.063 1.629 3.258 41.063 5.358

dimension 5 1.597 3.193 44.257

50 .259 .519 100.000 Extraction Method: Principal Components Analysis. a. When components are correlated, sums of squared loadings cannot be added to obtain a total variance.

______Appendix 4B Page 283 Table 4: Loadings of the 50 items on the four factors

Structure Matrix Component Students who succeed best with learning English 1 2 3 4 37: use time effectively .747 49: study things which were not from their class .698 32: make their schedule so they will have enough time to study E .685 30: are determined to achieve the target which they set for their E study .638 6: make good use of materials & resources when studying E at home .638 -.406 48: try to study English regularly even with limited time .633 62: know the method which suits them best and use it .630 43: check to make sure that they understood the lesson .606 36: reflect on what they learn and look for something important .587 21: write down their feelings towards E studies in a language learning diary .582 38: notice their mistakes & use that information to help them do better .580 -.472 23: share their feelings towards English studies with friends online .575 57: check their E proficiency by taking TOEIC, TOEFL or IELTS voluntarily .553 10: will ask their teacher if they have a question about E .552 -.495 35: check their E level by comparing it with the skills of others on the Internet .548 39: try to use other methods if one method of E study does not suit them .537 5: carry out the learning plans once they have been made .508 24: have a strong will and are not easily influenced by others .500 26: concentrate on their studies even if they happen to hear others talking .486 44: pay more attention to the lessons that they are not good at .450 52: will talk to a foreigner in E if they come across him/her in the street .448 56: have plans about how to learn English -.748 58: are aware of their studies -.694 55: think English is important for their future -.688 47: do things actively -.673 45: want to be good English learners -.661 19: try to improve their weaknesses -.643 61: find information about English by themselves -.629 60: practice English with people outside class -.594 15: know their good points and weaknesses -.557 25: think about their progress in learning English -.555 20: learn things that the teachers do not give as a task .407 -.520 51: study English no matter what mood they are in .437 53: will use E to communicate if seeing a stranger on the Internet 31: like to study with computers .637 28: pay attention when someone is speaking in English .558 -.407 54: try to complete the things they have decided to do .539 42: go online as a way of learning English .531 33: will do a search on the Internet if they have a question about E .529 27: consider a lot about study/work before doing it .506 46: pay more attention when they see an English website .497 59: dream of being good English speakers .487 50: will talk with their friends/relatives if they have a question about E .418 .431 4: want to communicate with foreigners in E -.704 9: try to find as many ways as they can to improve their E -.664 7: look for opportunities to use E as much as possible .541 -.628 12: want to study in an English-speaking environment if they have a chance .529 -.625 41: want to find a job where only E is used in the future -.455 17: try to develop their good points .426 -.449 22: learn English because they enjoy studying it .403 -.434 Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalization. Factor loadings sorted by size Loadings smaller than .40 were removed for clarity Shading: Items with factor loadings smaller than .40 and should be removed

______Appendix 4B Page 284 Table 5: An extract of the total variance explained when 49 items were included

Total Variance Explained Rotation Sums Initial Eigenvalues Extraction Sums of Squared Loadings of Squared Component a Loadings Total % of Variance Cumulative % Total % of Variance Cumulative % Total 1 11.674 23.824 23.824 11.674 23.824 23.824 9.082

2 4.029 8.223 32.048 4.029 8.223 32.048 6.483 3 2.996 6.115 38.163 2.996 6.115 38.163 5.090 4 1.629 3.324 41.487 1.629 3.324 41.487 4.832

dimension 5 1.541 3.145 44.632

49 .264 .539 100.000 Extraction Method: Principal Components Analysis. a. When components are correlated, sums of squared loadings cannot be added to obtain a total variance.

Table 6: Loadings of the 49 items on the four factors

Structure Matrix Component Students who succeed best with learning English 1 2 3 4 37: use time effectively .746 49: study things which were not from their class .696 32: make their schedule so they will have enough time to study E .686 6: make good use of materials & resources when studying E at home .639 -.403 30: are determined to achieve the target which they set for their E study .637 48: try to study English regularly even with limited time .635 62: know the method which suits them best and use it .630 43: check to make sure that they understood the lesson .608 36: reflect on what they learn and look for something important .590 38: notice their mistakes & use that information to help them do better .585 -.440 21: write down their feelings towards E studies in a language learning diary .583 23: share their feelings towards English studies with friends online .574 10: will ask their teacher if they have a question about E .557 -.478 57: check their E proficiency by taking TOEIC, TOEFL or IELTS voluntarily .544 .400 35: check their E level by comparing it with the skills of others on the Internet .543 39: try to use other methods if one method of E study does not suit them .543 5: carry out the learning plans once they have been made .511 24: have a strong will and are not easily influenced by others .505 26: concentrate on their studies even if they happen to hear others talking .492 44: pay more attention to the lessons that they are not good at .449 .406 52: will talk to a foreigner in E if they come across him/her in the street .439 56: have plans about how to learn English .753 58: are aware of their studies .698 47: do things actively .692 55: think English is important for their future .687 45: want to be good English learners .661 19: try to improve their weaknesses .653 61: find information about English by themselves .639 60: practice English with people outside class .603 15: know their good points and weaknesses .571 25: think about their progress in learning English .560 20: learn things that the teachers do not give as a task .536 51: study English no matter what mood they are in -.407 31: like to study with computers .628 28: pay attention when someone is speaking in English .570

______Appendix 4B Page 285 54: try to complete the things they have decided to do .535 42: go online as a way of learning English .523 33: will do a search on the Internet if they have a question about E .521 27: consider a lot about study/work before doing it .516 46: pay more attention when they see an English website .498 59: dream of being good English speakers .493 17: try to develop their good points .446 -.413 50: will talk with their friends/relatives if they have a question about E .419 .434 22: learn English because they enjoy studying it .423 -.413 4: want to communicate with foreigners in E -.711 9: try to find as many ways as they can to improve their E -.647 7: look for opportunities to use E as much as possible .543 -.628 12: want to study in an English-speaking environment if they have a chance .550 -.598 41: want to find a job where only E is used in the future -.446 Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalization. Factor loadings sorted by size

Table 7: Factor loadings when cutoff points were applied for each factor

Structure Matrix Component Students who succeed best with learning English 1 2 3 4 37: use time effectively .746 49: study things which were not from their class .696 32: make their schedule so they will have enough time to study E .686 6: make good use of materials & resources when studying E at home .639 -.403 30: are determined to achieve the target which they set for their E study .637 48: try to study English regularly even with limited time .635 62: know the method which suits them best and use it .630 43: check to make sure that they understood the lesson .608 36: reflect on what they learn and look for something important .590 38: notice their mistakes & use that information to help them do better .585 -.440 21: write down their feelings towards E studies in a language learning diary .583 23: share their feelings towards English studies with friends online .574 10: will ask their teacher if they have a question about E .557 -.478 57: check their E proficiency by taking TOEIC, TOEFL or IELTS voluntarily .544 .400 35: check their E level by comparing it with the skills of others on the Internet .543 39: try to use other methods if one method of E study does not suit them .543 5: carry out the learning plans once they have been made .511 24: have a strong will and are not easily influenced by others .505 26: concentrate on their studies even if they happen to hear others talking .492 44: pay more attention to the lessons that they are not good at .449 .406 52: will talk to a foreigner in E if they come across him/her in the street .439 56: have plans about how to learn English .753 58: are aware of their studies .698 47: do things actively .692 55: think English is important for their future .687 45: want to be good English learners .661 19: try to improve their weaknesses .653 61: find information about English by themselves .639 60: practice English with people outside class .603 15: know their good points and weaknesses .571 25: think about their progress in learning English .560 20: learn things that the teachers do not give as a task .536 51: study English no matter what mood they are in -.407 31: like to study with computers .628 28: pay attention when someone is speaking in English .570

______Appendix 4B Page 286 54: try to complete the things they have decided to do .535 42: go online as a way of learning English .523 33: will do a search on the Internet if they have a question about E .521 27: consider a lot about study/work before doing it .516 46: pay more attention when they see an English website .498 59: dream of being good English speakers .493 17: try to develop their good points .446 -.413 50: will talk with their friends/relatives if they have a question about E .419 .434 22: learn English because they enjoy studying it .423 -.413 4: want to communicate with foreigners in E -.711 9: try to find as many ways as they can to improve their E -.647 7: look for opportunities to use E as much as possible .543 -.628 12: want to study in an English-speaking environment if they have a chance .550 -.598 41: want to find a job where only E is used in the future -.446 Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalization. Factor loadings sorted by size Shading: Items to be removed after cutoff points were applied

Table 8: An extract of the total variance explained when 32 items were included

Total Variance Explained Rotation Sums Initial Eigenvalues Extraction Sums of Squared Loadings of Squared Component a Loadings Total % of Variance Cumulative % Total % of Variance Cumulative % Total 1 8.293 25.916 25.916 8.293 25.916 25.916 6.594

2 3.118 9.744 35.660 3.118 9.744 35.660 5.558 3 2.320 7.249 42.909 2.320 7.249 42.909 3.262 4 1.483 4.634 47.543 1.483 4.634 47.543 3.231

dimension 5 1.180 3.688 51.231

32 .290 .905 100.000 Extraction Method: Principal Components Analysis. a. When components are correlated, sums of squared loadings cannot be added to obtain a total variance.

______Appendix 4B Page 287 Table 9: Loadings of the 32 items on the four factors

Structure Matrix Component Students who succeed best with learning English 1 2 3 4 37: use time effectively .788 49: study things which were not from their class .719 32: make their schedule so they will have enough time to study E .714 62: know the method which suits them best and use it .702 48: try to study English regularly even with limited time .684 38: notice their mistakes & use that information to help them do better .666 30: are determined to achieve the target which they set for their E study .664 6: make good use of materials & resources when studying E at home .649 36: reflect on what they learn and look for something important .642 43: check to make sure that they understood the lesson .626 56: have plans about how to learn English .774 55: think English is important for their future .717 58: are aware of their studies .709 47: do things actively .700 45: want to be good English learners .693 61: find information about English by themselves .668 19: try to improve their weaknesses .636 60: practice English with people outside class .634 25: think about their progress in learning English .568 15: know their good points and weaknesses .544 31: like to study with computers .691 42: go online as a way of learning English .617 33: will do a search on the Internet if they have a question about E .589 28: pay attention when someone is speaking in English .541 46: pay more attention when they see an English website .507 27: consider a lot about study/work before doing it .483 54: try to complete the things they have decided to do .478 4: want to communicate with foreigners in E .729 9: try to find as many ways as they can to improve their E .652 7: look for opportunities to use E as much as possible .618 12: want to study in an English-speaking environment if they have a chance .613 41: want to find a job where only E is used in the future .483 Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalization.

______Appendix 4B Page 288 Table 10: Items to be removed after validation process with factor loadings

Structure Matrix Component Students who succeed best with learning English 1 2 3 4 Factor 1 37: use time effectively .788 49: study things which were not from their class .719 32: make their schedule so they will have enough time to study E .714 62: know the method which suits them best and use it .702 48: try to study English regularly even with limited time .684 38: notice their mistakes & use that info to help them do better .666 30: are determined to achieve the target which they set for E study .664 6: make good use of materials & res when studying E at home .649 36: reflect on what they learn and look for something important .642 43: check to make sure that they understood the lesson .626 Factor 2 56: have plans about how to learn English .774 55: think English is important for their future .717 58: are aware of their studies .709 47: do things actively .700 45: want to be good English learners .693 61: find information about English by themselves .668 19: try to improve their weaknesses .636 60: practice English with people outside class .634 25: think about their progress in learning English .568 15: know their good points and weaknesses .544 Factor 3 31: like to study with computers .691 42: go online as a way of learning English .617 33: will do a search on the Internet if they have a question about E .589 28: pay attention when someone is speaking in English .541 46: pay more attention when they see an English website .507 27: consider a lot about study/work before doing it .483 54: try to complete the things they have decided to do .477 .478 Factor 4 4: want to communicate with foreigners in E .729 9: try to find as many ways as they can to improve their E .652 7: look for opportunities to use E as much as possible .577 .618 12: want to study in an English-speaking envi if having a chance .510 .613 41: want to find a job where only E is used in the future .483 Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalization. Factor loadings sorted by size Loadings smaller than .40 were removed for clarity Shading: Items suggested to be removed by the expert validation process

______Appendix 4B Page 289 Table 11: Total variance explained when 25 items were included

Total Variance Explained Rotation Sums Initial Eigenvalues Extraction Sums of Squared Loadings of Squared Component a Loadings Total % of Variance Cumulative % Total % of Variance Cumulative % Total 1 6.699 26.794 26.794 6.699 26.794 26.794 5.594 2 2.397 9.588 36.382 2.397 9.588 36.382 3.911 3 2.101 8.405 44.787 2.101 8.405 44.787 2.466 4 1.355 5.420 50.207 1.355 5.420 50.207 3.212 5 1.011 4.045 54.252 6 .946 3.783 58.035 7 .839 3.355 61.390 8 .774 3.096 64.485 9 .755 3.019 67.505 10 .716 2.866 70.370 11 .685 2.740 73.110

12 .639 2.555 75.665 13 .624 2.494 78.160 14 .598 2.392 80.552 dimension 15 .562 2.248 82.800 16 .531 2.123 84.924 17 .524 2.097 87.021 18 .489 1.956 88.977 19 .452 1.808 90.784 20 .435 1.742 92.526 21 .415 1.660 94.186 22 .408 1.634 95.819 23 .358 1.430 97.249 24 .356 1.423 98.673 25 .332 1.327 100.000 Extraction Method: Principal Components Analysis. a. When components are correlated, sums of squared loadings cannot be added to obtain a total variance.

______Appendix 4B Page 290 Table 12: Factor analysis of the 25 items on learner autonomy perceptions

Structure Matrix Component Cumulative % Students who succeed best with learning English 1 2 3 4 of Variance Factor 1: Monitoring learning processes 26.794 37: use time effectively .797 32: make schedule so they’ll have engh time to study E .726 49: study things which were not from their class .725 62: know the method which suits them best and use it .712 48: try to study English regularly even with limited time .684 6: make good use of materials & res when study E at home .666 38: notice mistakes & use that info to help them do better .657 36: reflect on what they learn and look for sth important .643 43: check to make sure that they understood the lesson .627 Factor 2: Goal-setting and evaluating learning 36.382 55: think English is important for their future .728 58: are aware of their studies .727 45: want to be good English learners .714 19: try to improve their weaknesses .668 60: practice English with people outside class .618 25: think about their progress in learning English .614 15: know their good points and weaknesses .564 Factor 3: Using ICTs in learning 44.787 31: like to study with computers .757 42: go online as a way of learning English .692 33: will do a search on internet if they have a question abt E .664 46: pay more attention when they see an English website .520 Factor 4: Initiating learning opportunities 50.207 4: want to communicate with foreigners in E -.714 9: try to find as many ways as they can to improve their E -.678 12: want to study in an E-speaking envi if having a chance -.643 7: look for opportunities to use E as much as possible -.597 41: want to find a job where only E is used in the future -.524 Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalization. Factor loadings sorted by size. Other weaker loadings of each item were removed for clarity.

______Appendix 4B Page 291 Appendix 4C

INTERNAL CONSISTENCY TESTS OF THE FOUR DIMENSION SCALES

Item-Total Statistics Corrected Cronbach's Perceptions of Learner Autonomy Item-Total Alpha if Correlation Item Deleted

Dimension 1: Monitoring learning processes

(9 items, α = .874 )

37: use time effectively .703 .852 32: make their schedule so they will have enough time to study E .643 .857 49: study things which were not from their class .619 .860 62: know the method which suits them best and use it .613 .860 48: try to study English regularly even with limited time .621 .860 6: make good use of materials & resources when studying E at home .554 .866 38: notice their mistakes & use that information to help them do better .610 .861 36: reflect on what they learn and look for something important .600 .861 43: check to make sure that they understood the lesson .551 .866

Dimension 2: Goal-setting and evaluating learning

(7 items, α = .788)

55: think English is important for their future .551 .755 58: are aware of their studies .579 .749 45: want to be good English learners .544 .757 19: try to improve their weaknesses .541 .756 60: practice English with people outside class .477 .768 25: think about their progress in learning English .489 .766 15: know their good points and weaknesses .441 .778

Dimension 3: Using ICTs in learning

(4 items, α = .659)

31: like to study with computers .473 .568 42: go online as a way of learning English .478 .564 33: will do a search on the Internet if they have a question about E .446 .587 46: pay more attention when they see an English website .362 .643

Dimension 4: Initiating learning opportunities

(5 items, α = .683)

4: want to communicate with foreigners in E .484 .613 9: try to find as many ways as they can to improve their E .510 .602 12: want to study in an English-speaking environ. if having a chance .458 .624 7: look for opportunities to use E as much as possible .467 .620 41: want to find a job where only E is used in the future .286 .698

______Appendix 4C Page 292 Appendix 4D EFFECT OF THE THREE GENERAL FACTORS ON PERCEPTIONS OF LEARNER AUTONOMY

Tests of Between-Subjects Effects Type III Sum Mean Source Dependent Variable df F Sig. of Squares Square Corrected D1: Monitoring learning processes 28.142a 33 .853 1.606 .019 Model D2: Goal-setting & evaluating learning 73.752b 33 2.235 6.474 .000 D3: Using ICTs in learning 38.126c 33 1.155 2.161 .000 D4: Initiating learning opportunities 19.836d 33 .601 1.551 .028 Intercept D1: Monitoring learning processes 1011.420 1 1011.420 1905.216 .000 D2: Goal-setting & evaluating learning 1333.373 1 1333.373 3862.643 .000 D3: Using ICTs in learning 991.165 1 991.165 1854.015 .000 D4: Initiating learning opportunities 1270.172 1 1270.172 3277.750 .000 Gender D1: Monitoring learning processes .544 1 .544 1.026 .312 D2: Goal-setting & evaluating learning .556 1 .556 1.611 .205 D3: Using ICTs in learning 1.011 1 1.011 1.891 .170 D4: Initiating learning opportunities .003 1 .003 .006 .936 Living place D1: Monitoring learning processes 2.863 3 .954 1.797 .147 D2: Goal-setting & evaluating learning 26.791 3 8.930 25.870 .000 D3: Using ICTs in learning 1.032 3 .344 .643 .587 D4: Initiating learning opportunities 1.185 3 .395 1.019 .384 Comp. D1: Monitoring learning processes 2.390 4 .597 1.126 .344 proficiency D2: Goal-setting & evaluating learning 1.965 4 .491 1.423 .225 D3: Using ICTs in learning 2.683 4 .671 1.255 .287 D4: Initiating learning opportunities .336 4 .084 .217 .929 Gender * D1: Monitoring learning processes 1.368 3 .456 .859 .462 Living place D2: Goal-setting & evaluating learning 1.164 3 .388 1.124 .339 D3: Using ICTs in learning 1.978 3 .659 1.233 .297 D4: Initiating learning opportunities .867 3 .289 .746 .525 Gender * D1: Monitoring learning processes 2.889 4 .722 1.360 .247 Comp. D2: Goal-setting & evaluating learning 3.519 4 .880 2.548 .039 proficiency D3: Using ICTs in learning 3.638 4 .909 1.701 .148 D4: Initiating learning opportunities 3.828 4 .957 2.469 .044 Living place * D1: Monitoring learning processes 5.869 10 .587 1.106 .356 Comp. D2: Goal-setting & evaluating learning 2.711 10 .271 .785 .643 proficiency D3: Using ICTs in learning 5.511 10 .551 1.031 .416 D4: Initiating learning opportunities 3.364 10 .336 .868 .563 Gender * D1: Monitoring learning processes 2.873 8 .359 .677 .712 Living place * D2: Goal-setting & evaluating learning 5.370 8 .671 1.944 .052 Comp. D3: Using ICTs in learning 6.743 8 .843 1.577 .129 proficiency D4: Initiating learning opportunities 2.730 8 .341 .881 .532 Error D1: Monitoring learning processes 280.299 528 .531 D2: Goal-setting & evaluating learning 182.264 528 .345 D3: Using ICTs in learning 282.271 528 .535 D4: Initiating learning opportunities 204.607 528 .388 Total D1: Monitoring learning processes 8365.222 562 D2: Goal-setting & evaluating learning 11042.755 562 D3: Using ICTs in learning 7889.625 562 D4: Initiating learning opportunities 10059.320 562 Corrected D1: Monitoring learning processes 308.441 561 Total D2: Goal-setting & evaluating learning 256.016 561 D3: Using ICTs in learning 320.397 561 D4: Initiating learning opportunities 224.443 561 a. R Squared = .091 (Adjusted R Squared = .034) d. R Squared = .088 (Adjusted R Squared = .031) b. R Squared = .288 (Adjusted R Squared = .244) Shading: significant level of smaller than .05 c. R Squared = .119 (Adjusted R Squared = .064) ______Appendix 4D Page 293 Appendix 4E

DIMENSIONS OF LEARNER AUTONOMY AND OTHER FACTORS

Table 1: Descriptive means of males and females across the four dimensions

Group Statistics Gender N Mean SD Std. Error Mean D1. Monitoring learning processes Female 118 4.1008 .51994 .04786 Male 29 4.0651 .60751 .11281 D2. Goal-setting & evaluating learning Female 118 4.7542 .32826 .03022 Male 29 4.5320 .54918 .10198 D3. Using ICTs in learning Female 118 3.6674 .63385 .05835 Male 29 3.6552 .64922 .12056 D4. Initiating learning opportunities Female 118 4.3610 .50452 .04644 Male 29 4.1586 .67481 .12531

Table 2: Descriptive statistics on levels of perception and post-test performance

Paired Samples Statistics Std. Error Mean N SD Mean D1. Monitoring learning processes Perception 4.094 147 .53630 .04423 Post-test Performance 3.200 147 .61959 .05110 D2. Goal-setting & evaluating learning Perception 4.710 147 .38996 .03216 Post -test Performance 4.346 147 .55402 .04569 D3. Using ICTs in learning Perception 3.665 147 .63468 .05235 Post -test Performance 3.622 147 .73871 .06093 D4. Initiating learning opportunities Perception 4.321 147 .54575 .04501 Post -test Performance 4.167 147 .59267 .04888

Table 3 : Statistics on the levels of Dimension 3 by computer proficiency

Descriptive Statistics Computer Mean SD N Proficiency Very bad 2.542 .78129 6 Bad 3.146 .71873 12 Average 3.497 .81802 93 Pre-test Performance Good 3.960 .67422 31 Excellent 3.950 .41079 5 Dimension 3: Using Total 3.543 .82313 147 ICTs in learning Very bad 2.958 .45871 6 Bad 3.521 .67805 12 Average 3.573 .74690 93 Post-test Performance Good 3.911 .68175 31 Excellent 3.800 .79844 5 Total 3.622 .73871 147

______Appendix 4E Page 294 Appendix 4F

TWO EXCERPTS OF THE INTERVIEW TRANSCRIPTS

Interview transcript with Student 2

Consent form explanation I: Interviewer S: Student

I Do you have any foreign friend? S No, I only have Vietnamese friends and then they study abroad. I So do you often speak English with your friends out of the class time? S Yes, I do. We have a group of four. We meet every Friday and Saturday evenings in a friend’s house. A topic is set a week earlier. During the meeting, each of us gradually has a presentation about that topic. Each person has about 10-15 minutes. We practice speaking skills in general and public speaking skills in particular. I Where is this idea from? S From me and another close friend. I study English, my friend studies Finance and Banking but she likes to study English. Because the English courses of the Finance and Banking are narrow, we together organize this study group. I How much time do you speak English in such a meeting? S When we first meet, we speak Vietnamese. After that, we speak English during the presentations and comments. We finally finish with speaking Vietnamese. So we often speak English for around 1-1.5 hour in each meeting. I Do you speak English to any foreigner? S Yes, I do, about less than ten times. I How did that happen? S In some situations, I started. For example, I went to a coffee shop where there are foreigners, and I then came and talked to them. In some other situations, I went to Ben Thanh market. When I saw them, I approached them and talked to them. Sometimes foreigners are also invited to class. I When you come to a coffee shop, there should be many people there. How do you choose someone to talk to? S Psychologically, EFL students always want to speak to native English speakers. Therefore, in the coffee shop, I can only talk to those who are not surrounded by students. I do not have many choices at all. I Can you talk a bit about that coffee shop? S That coffee shop is special. It is designed to create an environment for students to talk with foreigners. However, drinks are expensive. The foreigners coming to that place are those who agree to speak English with others. So they are very friendly. I How about in Ben Thanh market? S Some are very friendly when I come and talk to them. I just say I am a student and want to do some research, for example. It is just an excuse for a talk. Some are willing to help and they sometimes talk a lot, about different things. However, some others show that they do not want to talk when I approach them. I Do you join any English Speaking Club? S I am currently a member of an English Speaking Club in the Labour and Culture House. My duties include writing two sections: funny stories and something to learn (expressions

______Appendix 4F Page 295 and proverbs) in each discussion topic. I often go online, look for information, synthesize it, and present it in the topic paper. I How did you become a member of the club? S It is an interesting story. On that day, I came to join the club as a normal participant, but it was the time that people were rehearsing for the birthday ceremony of the club. Therefore, the chairman of the club invited me to be a member. He said if I would like to be a regular member of the club, I could start from that event. He asked me to write an essay and submitted to him. Later I did that and became the club member since the birthday event. I How about before that? How did you know about the club? S I knew about it from a school mate. She told me about the club when we were on a bus. After that, I came to the club with another friend. The one who told me about the club did not come any more. When we came, the club was in shortage of human resources for that special event. So I talked directly to the chairman and wrote an essay after that. He also asked me to introduce myself, and he introduced me to other members. I then talked to them to learn about the club. The next three Sundays, I came and took part in a rehearsal of some songs and a play for the event. I What are the factors that enable you to take part in this activity regularly? S Because I think that my English is not good and it is a very good environment for me to practice. I can talk to different people and improve my communication skills. The members who prepare the discussion topics also receive some advantages. They will work as a MC once a month, so I can also have opportunities to improve my public speaking skills. I You have been with the club for a while. What are the factors that you think prevent people from taking part in the club regularly? S I think it is the repetition and it does not have anything creative. Some programs are not well-prepared and they seem to be routine. When people come to the program and only see the repetition, they would like to have something new. In addition, I realize that sometimes the club services do not match with the participants’ needs. They primarily come to the club for speaking opportunities. However, sometimes the MC talks too much and does not give the participants enough space to talk. I How about learning in class? Before coming to class, do you often read the textbooks in advance? S I sometimes read the textbooks but I do not pay much attention to them. I mainly look for other books to read. I do not know if I am too subjective but I think that I have already known about the lessons in the grammar textbook, for example. So I do not read it at all. For the speaking course, if the teacher asks me to read a lesson in advance, I’ll do it. Otherwise, I just read other references. I often read English novels, English magazines, or proverbs, expressions, and slang on the internet. I Are those materials useful for your school courses or are they different from your courses? S I only learn according to my favorites, so it depends a lot on my interests. Learning in such a way can also help me increase my vocabulary repertoire. Some words from those materials will be seen again later. From the magazines and newspapers, I can also learn sentence structures that are useful for my writing skills. I Can you tell me why you choose those materials to learn? S Because the language is familiar with me. The content is not very important. When I read something and understand it, I’ll continue to read it. I Do you have a general plan for your learning? S I used to make a general learning plan but it was not practical so I only make daily plan.

______Appendix 4F Page 296 When I have a general plan, although I want to do it but I may want to hang out or my family wants me to do something else, or other courses from the general program comes to the examination, or even some courses of the English program comes to the examination, I’ll have to learn for these examinations. As a result, I forget my plan. Later, I only make daily plan so I can finish it within a day. A general plan includes a long list and maybe I cannot finish it. I So do you ever change your daily plan at all? S Last semester, I changed my plan very often, but this semester I’ll need to reconsider my learning performance and the importance of each learning plan. For example, last semester, when a plan was changed, I was attracted by that change and forgot my initial plan. However, this semester, I’ll give priorities to the plans which can help me achieve my objectives. I When you fail to implement a learning plan, what do you do with it? S I try to finish half of it right after that or do it the day after. Sometimes I have to give up some plans because so many remain unsolved and I cannot do them all. I What is the main reason that prevents you from implementing a learning plan? S Perhaps friends. For example, my friends sometimes call me and we talk for long. As a result, I cannot do what I plan to do. In some other cases, friends invite me to go out for a drink or something. So will you continue to make learning plans? S It depends on the outcome of the previous plan. If I make plans and I see steady progress, I’ll continue to do that. Otherwise, I’ll stop doing it. However, I continue to make plans for these days because the examination is coming. I According to you, to what extent does a learning plan affect your learning outcome? S I learnt many things from the last semester. I know that without a learning plan, my study is affected by different factors, and that directly has an influence on my major. I Do you often check or evaluate your learning performance? S Yes, I do. I often use short multiple choice tests, or ask my friends or teachers to see if I make any progress because I think they are more objective. I How do you ask them? S If it is a friend, I’ll speak to him/her in English for a while and ask her about my performance, for example, my speaking skills or my vocabulary. She/he will then have comments. I do not talk to teachers very often because there is always a distance between students and teachers. I In class, do you often compare your English language proficiency with your friends’? S Yes, I do. I listen to my friends talking in class and I’ll see if they make any grammatical mistake and if they are fluent. I then compare their proficiency with mine. If I see a friend who is better than me, I’ll ask myself how he/she can do that and when I can be that good. I When a teacher returns a home assignment, do you care about your friends’ scores? S Yes, I do. I often ask two friends next to me about their scores. After that, I’ll borrow their assignments and check with mine. I If you see that your friends’ assignment is as good as yours, but you get a lower mark, what will you do? S I just see it carefully and will not take the two assignments to the teacher because I know that I am not very good at writing. I have read a lot of essays from my friends. Sometimes they make several grammatical mistakes but their ways of expressing ideas are clear and logical. Although I have fewer grammatical mistakes in my writing, I often translate the message from Vietnamese to English. Therefore, I read my friends’ writing for reference. I can learn from them to minimize my mistakes but I never take the assignments to the

______Appendix 4F Page 297 teacher. I only do it when I am sure that it is totally wrong because I need the teacher to confirm my understanding. I cannot just say my essay is also good. I Do you often evaluate your proficiency by taking tests such as TOEFL? S I do not do those tests but I do other ways to evaluate my learning. For example, when I started to study in the university, I came to an English Speaking Club. At that time, I mainly learnt English grammar from high school only. I could speak even much worse than a seventh grade student. He could speak English very fluently, while I could only introduce myself and listen to the others quietly. However, when I started in the university, I had opportunities to practice with friends. After four or five months, I realized that I could speak what I had in mind. I then think I can practice to speak better. I So do you often record your speaking for evaluation? S My teacher talked about that but I have not done it. I only talk to friends and ask them to evaluate my skills. I Before submitting an assignment, do you often check it again carefully? S I often check it, except when I am sure that it was already done and checked carefully. I How much does such a check help you? S It does help. I can often realize mistakes in tenses or word uses. I Do you think that these sorts of mistakes are really important? S I think it is useful for me because I may be in a hurry when doing an assignment and make mistakes. Sometimes, I cannot pay adequate attention to some details and that can become a bad habit. So I think checking the assignment again is always useful. I Regarding the LMS, what do you often do with it? S I often get there to place postings. If anyone posts mp3 files, I’ll download it and listen. Otherwise, I’ll listen to those postings directly from the internet. I Is there any posting or topic on the LMS that motivates or encourages you to look for further information? S Yes. The VOA program is an example. After listening to a posting of a friend, I find it rather easy to understand and good for my listening skill improvement. So I search for that site and look for more listening recordings. Similarly, there are music clips posted on the LMS but only some background information is provided. After listening to such clips, I look for the lyrics for further learning. I Do you often place comments on the postings of other friends? S Yes, I do. I like doing that. If I read a posting and find it suitable for me, I’ll comment on it. For example, after listening to a song, if I find it interesting and meaningful, I’ll place a comment, expressing my thanks to the author of the posting and some of my feelings, even though I have not read the lyric of the song yet. Similarly, if the topic of a posting is similar to what I am think about and suitable for my understanding, I’ll make a comment. Now, I do not remember exactly about the authors of those postings. I When you come to a section in the LMS, there are a lot of threads, how do you choose a thread to read? S I look at the titles first to see if there is anything interesting. After that, I look at the number of comments. I often think that a topic that receives many comments should be somehow interesting. So I base on these two criteria for my selection. I Do you pay attention to the author of the thread? S A little, normally I prefer to read the postings of my close friends first. I Do you think that I spend more time on this course because it has the LMS component? S Yes, I think so, but it is worth spending more time on it. I Do you think that you are encouraged/motivated to go online more frequently when the LMS is introduced to the course?

______Appendix 4F Page 298 S Yes, I do. There was a period of time when I go to the LMS and internet every day because the LMS has very interesting clips and listening exercises posted by other friends. I download those files and become more motivated for my listening practice. Normally, I turn on a long lecture and listening to it. Now I can listen to different materials, posted by different friends. So it becomes more motivating and interesting. I During your participation in the LMS activities, do you often compare your English language competence with others? S Yes, I do. I often pay attention to their ways of expressing ideas, grammar, and then content. I Do you often read the postings that you placed before again? Yes, I do. I read them again to see what I wrote before. I read all of my postings, including those that receive comments from others and those that do not get any comment because they are all mine. I read them again to see if I was fluent in my writing, if there was any mistake, or how my language use was. I Do you think that the LMS helps you understand more about your teachers and friends? S Yes, I think so. As far as I see, the personalities of my friends in normal life are often consistently reflected in the LMS postings. Those with a sense of humor often place funny postings in English; those who are gentle in real life often post sweet songs. Therefore, I can partly understand their favorite and music preference. It also happens with the teacher. For example, my teacher posts a message about zodiac, and I am very curious. I keep asking myself where she gets that resource. She places two postings, the first one without the subtitle and the second with the subtitle. I intend to ask her to learn more about such a learning resource. I After you place a posting, do you want your teacher to comment on it? S Yes, I do. I want my teacher to comment on my grammar and word use. I want my teacher to clearly point out the mistakes rather than giving a general comment. I think it is difficult for the teacher to particularly pay attention to students in class but the LMS is a convenient environment for teachers to share with students. I The comments on your posting in the LMS will be archived, and many others can read them. What do you think about that? S I may feel a little embarrassed, but it can make me remember longer, and I can learn from that. That’s fine. I It seems that you go online quite often. If now you are given a computer connected to internet, what do you do first? S I log in Yahoo Messenger, come to Google and my blog. I What do you do with those programs? S I log in YM to check mail and sometimes buzz a friend for a chat. I If you have something interesting and would like to share it with the others, how are going to do it? S I write it in my diary if I am tired. If I am not tired, I write it on my computer. After that, I’ll share it with my friends via phone. I do not have a habit of blogging such stories. I use blog to share some thoughts with friends sometimes only. I do not think that it is the place for me to really talk about myself. I How many people are there in your YM friend list that you have never met in real life? S Nobody. I have met all of them in real life. I Do you know them first and add their nicknames or do you add their nicknames first and meet them in real life? S I all meet them first and then add their nicknames, no other way round. I How about the friend list of your blog?

______Appendix 4F Page 299 S The friend list in blog is different. When I surf web, I can go around and if I see any impressive names or funny avatars, I’ll come and read about them. If they are nice, I’ll add them to my friend list. In my friend list, there is only one person whom my friend introduces to me. I That means you add other people? S Yes, that’s right. I Is there any situation that someone adds you first? S Yes, there is. I What do you do then? S I feel happy and will accept the friend invitation right away, without looking at their blog in advance. I When you chat with a friend in YM, do you expect a prompt response, given that your friends can be invisible, busy, or available but away from the computer? S I always expect a prompt response. Technical problems can sometimes be a problem. So I often wait for around 10 minutes, if I do not see any response, I’ll buzz him/her. If he/she does not answer, I’ll sign out of YM. Only when I have something urgent and wait for response, I’ll make it invisible. Otherwise, I just turn it off. I What do you often write in your blog? S Often about my thoughts about life and I often expect comments, sharing my thoughts such as a comment or a direct talk about that. I Does this means that you blog because you want others to share with you? S First, blogging is to release stress. Second, it is to get comments from friends. I Thank you very much.

______Appendix 4F Page 300 Interview transcript with Student 10

Consent form explanation I: Interviewer S: Student

I Do you have any foreign friend? S Yes. I have one, but I have never met him. I only voice chat with him. I How did you get acquainted with him? S A high school friend of mine introduced me to that foreign friend (giving nicknames). After that, because I like to study English and want to improve my speaking and listening skills, I start to chat with that foreign friend and we have very nice chats. That’s how we get to know about each other. I So do you contact him regularly? S Because he does not go online often, so we chat with each other around once a week, for around 30 minutes. I Apart from this friend, do you speak English with other friends outside of class time? S Rarely. I only do that when I am very excited at doing it. I can ask a friend besides me to speak English and the talk just goes. However, I do not do it regularly. Such a conversation only lasts for about 20-30 minutes, and I only do that with my close friend. I rarely speak English with others. I So you make friends with any other foreigner? S No, foreigners often use a lot of abbreviations/short forms in chat, but my foreign friend is very special. He writes every word in full form and his speaking is also easy to understand. I Do you often look for and talk to any foreigner that you come across in the street to improve your listening –speaking skills? S No, but I take part in an English Speaking Club in the Youth House a few times. One time, a friend and I went across the Youth House and saw a notice of an English Speaking Club. So we came in and joined it. When I first came in, I saw that others were very confident, talking happily. So I felt a bit shy and could not talk much. I just listened to them and answered their questions. I How does that English Speaking Club organized? S They have groups and when I come in, I often look around to join the group that does not talk much because I think they are similar to me. I Why don’t you take part in the Club regularly? S Because this place is far from my house and I have a part-time job. I Do you think that taking part in the Club is useful? S Because I do not join it often so I am not sure about its effectiveness. I Do you often read English news or buy English books? S Yes, I often buy books like TOEFL, TOEIC an English magazines like Special English. Although I do not buy them regularly, I buy a lot at a time when I go to the bookshop. I am a bit lazy to go to the bookshop regularly because such magazines are only sold in a few places. I Do you often read English newspapers on the internet? S I do not read newspapers on the internet but I listen to radio on CNN. I Do you often prepare your lessons before going to class? S The textbook of the course is very boring, so I am not interested in reading it. I only read other books. Only when teachers ask me to do something in the book, I do it. Otherwise, I

______Appendix 4F Page 301 do not look at the lessons in advance. In addition, I do not think that reading the lesson in advance is useful. I only feel bored when doing so. I read other books instead. I choose those with interesting lessons to buy. I Do you think that reading books other than your course textbook is useful for your learning at school? S I think yes. For example, in an examination, I saw a familiar test that I used to read somewhere. Therefore, I could do that examination very well. I Do you often make a general plan for your EFL learning? S I never make any learning plan. I only try my best to study. I So, do you make any weekly plan? S I sometimes make a learning plan in mind only. Sometimes I work according to my plan, sometimes I cannot finish my plan. I Why? S Because of different reasons, both subjective and objective. For example, on a Thursday, I planned to read that article, but I came home late and became tired. So I just fell asleep. On the following day, I did not read it either, time passed by, and I finally could not make it. I Do you think that learning with a plan is better? S In general, a learning plan can create a motivation for me to learning better. Without a learning plan, things just go on as usual, sometimes aimless, without any particular learning objectives. I But you do not often make a learning plan. Why? S Perhaps it is my habit. I just study what I like. However, when the examination comes, I make careful plans. Normally, I don’t. I So will you make learning plans in the future? S I am not sure, because I am an unplanned person. I Referring back to the books that you mentioned earlier, it seems that they only have practice tests. They are not similar to the course textbooks which are designed to improve language skills. Why so? S Those books are normally introduced by the others, and I do the practices tests in there to improve my skills. I When you do those practice tests, do you look at your overall score? S Yes for sure. I want to see my level of competency and progress. I Apart from that, do you try any other way to know about your language competence, for example, asking your friends to evaluate your language skills? S Very often I ask my friends to evaluate my pronunciation skills. For example, in the pronunciation class, I often ask my nearby friends to check my pronunciation. I also look at my achievement in the mid-term and final exams. I also ask other friends whose language competence is similar to mine to evaluate my skills. I Do you record your reading? S Yes, I do. I do that very often because I like doing so. After recording my reading, I listen to it again and try to correct my mistakes and record again. I often record two or three times for a paragraph. I do that until I think that it is fine. I Do you compare those recordings? S No, I keep those recordings but I do not compare them. I If you come to a relative’s house and you see and English test, what will you do? S As a habit, I’ll have a look at it. Probably I’ll get a pencil and try to answer every question. I Why do you do so?

______Appendix 4F Page 302 S Because it is a test, I am curious enough to try it as a habit. Because I also tutor a few high school students, I want to try the test to see how it is. I If you have an assignment, before submitting it, do you often look at it very carefully? S I often check spelling mistakes only before submitting. I Do you think that reviewing an assignment like that is useful? S I think so, but this is also my habit, I do not often review it carefully. I Regarding the course LMS, what do you often do with it? S I often get there to place postings. I When you come to the LMS, what do you often do first? S I come to my class section and come to the audio/video sharing section. I Do you think that because this course has an online component, you become spending more time on the course? S I think so. When I come to the internet, I always come to the LMS first. That’s my habit. I When you read postings in the LMS, do you see anything that encourages/motivates you to look for further details? S In general, I just read those postings only. I do not look for related information. I In your virtual class section, have you seen any source or link that leads you to further reading? S If they are songs, I’ll look for the lyrics of those songs and other songs of those singers. I Do you look for the biography of those singers? S Rarely I When you come to a section in the LMS, you may often see a long list of posting threads. So how do you select a thread to read? S I look at the number of comments made for each thread. I Does it mean that you will come to the threads with more comments first? S Yes. That’s right. I If you do so, you are never the first one who replies to a topic thread. S Sometimes I am also the first one who comments on a topic starter. However, if a thread can receive many comments, I think it should be interesting. So I come and read it I Do you often pay attention to the thread title or the thread author? S No, the author name does not draw my attention at all. I only pay attention to the number of comments on a thread and the thread title. I When coming to the LMS, you know that your friends often place postings. Does that motivate you to come to the LMS more often? S There are two reasons for my coming to the LMS: I want to learn more and it is the requirement of my teacher. She asks us to place at least one audio/video posting a week, but I often end up with placing several postings, partly because I know that my friends also post there and partly because I often go to YouTube and see interesting clips, so I post them to the LMS. I When you read a posting in the LMS, what detail do you particularly pay attention to? S Content of the postings. I do not often care about spelling mistakes or word use at all. I When you are in class or online, do you often evaluate your friends’ competence or compare your competence with theirs? S I do that very often. I When your teacher returns an assignment, for example, do you often compare your score with the others? S Yes, I do. I If you think that your assignment is as good as your friend’s but gets lower marks, do you take that to your teacher?

______Appendix 4F Page 303 S If the difference is big, I’ll ask the teacher. If it is only one mark different, I won’t. I Why do you want to compare your competence with the others? S Generally because I want to know about my progress. For example, if I see that my score is lower than my friends’, I need to try harder next time. I So when you go to the LMS, do you often compare the postings over there with your competence? S Yes, I do but not often. I basically look at the content of the postings only. I So do you think that the LMS helps you understand your friends’ competence better? S Yes, it does, but it is not everything. For example, I have a classmate who talks every little in class, but when she posts very interesting messages with good recourses on the LMS. I When placing a posting on the LMS, do you want your teacher to comment on it? S Yes, I do but my teacher makes very few comments. I Why do you want so? S Because my teacher is more experienced. I What kind of comments do you expect from her? S For example, about spelling, grammar, content… or she can talk about what she thinks about my posting. I When you log in the LMS, do you often read the postings that you previously placed? S Yes, but I do not read my previous postings. When I see new comments on my postings, I’ll come and ready the comments. If a posting of mine receives no comment, I won’t come back to it. I Have you ever read your postings again to see if there is any mistake, for example? S Almost never. I It seems that you go online quite often. So what do you do when you go online? S I log in Yahoo Messenger first. I then go to the LMS during the last semester. Now I go to some news sites such as DanTri. I When you go to YM, what do you do? S I first check my offline messages and then email. I Do you blog? S Yes, I do, on Yahoo. I also have accounts in other sites such as Facebook, twitter. I Do you often check all of these sites? S Yes, I do. When a new message comes, I log in and check it. I Which blog do you use most? S I used to use Yahoo but now it is closed so I now use Facebook. I Where are the friends in your YM friend list and blogs from? S They are those who used to study with me, those who study with me now, those from my friends’ connections, and those from some chat rooms. I How many people in your friend list have you never met? S About 35 percent of them I So how did you get to know about them? S Mainly from my friends’ introduction, from blog, and from YM or blog list. I That means if someone adds you, you’ll agree. So do you add anyone first? S Right. I only add a new friend if my friend introduces him/her to me. I So when you surf webs and blogs, do you add anyone? S Yes, I do. If I see an interesting blog, I’ll add that person. I If now you go home and see something interesting. You would like to share it with your friends. How will you do it? S I’ll put it on my blog. That’s my habit. When I have something fun or sad, I always put it

______Appendix 4F Page 304 on my blog. I What do you expect when you finish a blog entry? S I would like to receive comments which share my feelings. I When your friends come to your entry, do you want them just read it or comment on it? S Either way is fine. Leaving a comment or not is not important. I So blog is for you to share your happiness and sadness. What is the YM for? S To chat with my close friends. I So what is the email for? S Email is used to check the updates from the websites that I registered. I When you go to YM, some of your friends may set their status as being available, some make it invisible… So when you chat with someone, do you expect a prompt response from them? S When I chat with someone, I always expect a prompt response. I If you do not receive a response right away, how do you feel? S A bit annoyed. I If you chat with a friend and you know that he/she is always invisible, how can you know if he/she is online and should expect a prompt response? S There is a website that can check if that person is invisible. I If you check and you know that your friend is invisible and he/she does not respond to you, what do you think? S I think that he/she may not want to talk to me. I In your YM friend list, how many people do you know on the internet first and then see them offline? S About 18 percent I If put this 18 percent together with 35 percent of people that you have never met in real life, it will make up more than 50 percent. That is the number of people that you know from the internet first? S Yes, that’s right. I Thank you very much for participating in the interview.

______Appendix 4F Page 305 Appendix 4G EXAMPLES OF THEMES AND CODING

Mediating Factors of Learner Autonomy Performance in Offline Learning Environment

1. Personal preference

This theme concerns what types of learning activities students like to conduct and why they are interested in those activities.

1.1. Always making learning plans

I made a master learning plan when the semester started… Every two months, I made weekly plans in details. I scheduled to learn listening and reading in the afternoon and evening because my room was usually quiet at that time… If I could not complete a task in a plan, I would try to finish it at the weekend. (Student 9)

1.2. Sometimes making learning plans

I make no plan for my learning because it is me. I learn what I like at a particular time, [not guided or regulated by a plan]. However, if the examinations come, I’ll make a detailed plan… [because] I’ll be more motivated to study… I am not sure if I’ll make learning plans in the future because I am an unplanned person. (Student 10)

1.3. Never making learning plans

I work totally upon my mood. If I just buy an interesting book, I’ll try to finish reading it before preparing for an examination scheduled in the following day… I usually study when I am free and feel like studying. So only a learning mood, [not a learning plan], works for me. (Student 6)

2. Motivation

This theme concerns the sources of factors that motivated students’ EFL learning.

2.1. Personal interests (Intrinsic motivation)

I am interested in literature... and read that kind of books... I rarely pay attention to my classmates’ learning achievements. I don’t care much about that… I sometimes check with my teachers if I do not feel happy with my assignment grades… but I never check with my friends. (Student 6)

2.2. Successes of friends or peers (Extrinsic motivation)

If I meet a person and realize that he/she is better than I, I would ask myself why he/she could be successful like that and when I would be able to be that good… [However,] I would not ask him/her about his/her learning methods... because I don’t want to be just like him/her. (Student 2)

______Appendix 4G Page 306 3. Attitude

This theme concerns students’ attitude towards the learning materials and activities

3.1. Attitude towards textbooks

I think there is nothing to read in [those textbooks]. I just need to improve my four language skills [speaking, listening, reading, and writing]. Attending classes and reading the textbooks do not help me learn better. Really, nothing at all… So it is better to make my own learning plan and follow it. (Student 11)

3.2. Attitude towards learning activities

My goal is to write a complete essay every week... However, if I don’t see it appropriate and I do not make any progress after some time, I’ll modify it... After finishing an essay, I save it and read it again later. It can be like a stranger reading my essays and see where my mistakes are. I can then know about my progress. (Student 1)

I always review my work very carefully before submission. When it is marked and returned, I’ll look at it again and again. I love reading my work although it may not very good. (Student 4)

______Appendix 4G Page 307 Appendix 5A

ALLOCATION OF ITEMS TO THE MODEL OF LEARNER AUTONOMY DEVELOPED FROM THE LITERATURE REVIEW

Benson (2001), Littlewood (1996, 1999), Macaro (1997, 2008), Nunan (1997), & Scharle and Items in the four-dimension model observed in the current study Szabo (2000). See Figure 2.2, page 34 Cognitive Demonstrated Situation processes behaviors management Dimension 1: Monitoring learning processes 37: use time effectively x 32: make their schedule so they’ll have engh time to study E x 49: study things which were not from their class x 62: know the method which suits them best and use it x x 48: try to study English regularly even with limited time x 6: make good use of materials & res when studying E at home x 38: notice mistakes & use that info to help them do better x x 36: reflect on what they learn and look for sthing important x x 43: check to make sure that they understood the lesson x Dimension 2: Goal-setting and evaluating learning 55: think English is important for their future x 58: are aware of their studies x 45: want to be good English learners x 19: try to improve their weaknesses x x 60: practice English with people outside class x 25: think about their progress in learning English x 5: know their good points and weaknesses x Dimension 3: Using ICTs in learning 31: like to study with computers x 42: go online as a way of learning English x 33: will do a search on the Internet if have a question about E x 46: pay more attention when they see an English website x Dimension 4: Initiating learning opportunities 4: want to communicate with foreigners in E x 9: try to find as many ways as they can to improve their E x x 12: want to study in an E-speaking envi. if having a chance x 7: look for opportunities to use E as much as possible x x 41: want to find a job where only E is used in the future x

______Appendix 5A Page 308