WRITING PROCESSES AND META-COGNITIVE STRATEGIES OF FOUR INDONESIAN TERTIARY LEVEL EFL STUDENTS :A Think Aloud Protocol Based Study.

STATEMENT OF DECLARATION

I hereby certify that this thesis is completely my own work. I am fully aware that I have quoted and paraphrased some statements and ideas from other sources, and they are properly acknowledged in the text.

Bandung, August 2009

Writing Processes and Meta-cognitive Strategies

of Four Indonesian Tertiary Level EFL

Students

(A Think Aloud Protocol Based Study)

By Fatma Helmiyantriyani

Approved by

________________________________

Emi Emilia, M.Ed., Ph.D. (Main Supervisor)

__________________________________ Dr. Didi Suherdi, M.A. (Co-Supervisor)

ABSTRACT

The research reported in this thesis investigated the issues of second/foreign language writing processes and strategies. The purpose of the study was to investigate the processes as well as meta-cognitive strategies applied by four Indonesian EFL tertiary students while they were writing an English argumentative essay. Data for the study consisted of think-aloud reports as well as post-writing questionnaires collected from two higher-scored and two lower-scored writers with different levels of language proficiency (i.e. low intermediate, high intermediate, and advanced). Analyses of both the transcribed protocols and post-writing questionnaires substantiated previous researchers’ findings that all of the participants followed complex, non-linear, recursive processes of writing, in which the sub-processes (i.e. planning, organizing, evaluating, and revising) might appear in the pre-, while-, and post- writing stages. Besides that, the findings also revealed important differences of meta-cognitive strategies used by both pairs of writers, in that the higher-scored writers displayed a larger number of meta-cognitive strategies, in terms of strategies for planning, organizing, evaluating, and revising, than the lower-scored ones.

ACKNOWLEDGEMENT

The writing of this thesis is not an easy task. It needs much effort and time, in order to get a polished thesis. In this occasion, I would like to express my immense gratitude to all those persons who have given their valuable support and assistance, so that I can complete this thesis. In particular, I am profoundly indebted to my supervisors, Emi Emilia, M.Ed., Ph.D., as my main supervisor, and DR. Didi Suherdi, M.A., as my co-supervisor, for their generously providing time and knowledge in assisting me to complete this thesis. Without them, I can hardly imagine the completion of this research project.

My special thanks are also due to the participants of this study, who have given their time and best effort in dealing with the PTESOL test as well as the writing tasks. My best friend, Lika Widiantoro, who gave me constant support and suggestions during the completion of this thesis, also deserves a special mention.

I also thank many lecturers and friends in the English Education Program of School of Postgraduate Studies, Indonesia University of Education, for their share and support as well as feedback to my study.

My deepest thanks go to my parents, sisters, and brothers, as well as parents, sister, and brother in laws who in different but equal ways have given great contribution to my study, for their sincere love and prayers.

Finally, I want to acknowledge special thanks to my family: to my beloved husband, Agus Nazil Furqon, and to my lovely daughters, Hana and Raisya, for their material and immaterial support to a wife and a mother in completing her study. It is a great pleasure that to them all I dedicate this thesis.

LETTER OF CONTENTS

DECLARATION PAGE i

APPROVAL PAGE ii

ABSTRACT iii

ACKNOWLEDGEMENT iv

LETTER OF CONTENTS vi

Chapter I: INTRODUCTION

1.1. Research Background 1

1.2. Objectives of the Study 8

1.3. Significance of the Study 8

1.4. Scope of the Study 9

1.5. Definitions of the Terms 10

1.6. Thesis Organization 11

Chapter II: LITERATURE REVIEW

2.1. The Values of Writing 13

2.2. The Processes and Strategies of Writing 15

2.2.1. The Pre-writing Stage 21

2.2.2. The Writing/drafting Stage 26

2.2.3. The Post-writing Stage 29 2.3. The Use of Think-aloud Protocols in Writing Process and Language

Learning Strategy Studies 34

2.4. Concluding Remarks 39

Chapter III: RESEARCH METHODOLOGY

3.1. Research Questions 41

3.2. Research Design 41

3.3. Participants of the Study 43

3.4. Research Instruments 45

3.4.1. Writing prompt 45

3.4.2. Post-writing questionnaire 46

3.5. Data Collection Place, Time, and Procedure 47

3.5.1. Think-aloud protocols 48

3.5.2. Post-writing questionnaire 50

3.6. Data Analysis Procedures 51

3.6.1. The Scoring of the Writing Task 51

3.6.2. Think-aloud protocols 60

3.6.3. Post-writing questionnaire 62

Chapter IV: DATA PRESENTATION AND ANALYSES

4.1. Findings and Discussions 64

4.1.1. Data from the Think-aloud Protocols 64

4.1.1.1. Processes of Writing 65

4.1.1.2. Meta-cognitive Strategies 74 4.1.2. Data from the Post-writing Questionnaire 101

4.1.2.1. Strategies Used in the Pre-writing Stage 101

4.1.2.2. Strategies Used in the Writing/drafting Stage 104

4.1.2.3. Strategies Used in the Post-writing Stage 108

4.2. Concluding Remarks 109

Chapter V: CONCLUSIONS 5.1. Conclusions of the Study 110

5.2. Limitations of the Study 113

5.3. Recommendations 114

REFERENCES 115

APPENDICES 128

Appendix I : Tables and Figures Appendix II : Outlines of the Essays Appendix III : The Participants’ Essays

Appendix IV : Example of the Think-aloud Protocols Transcription Appendix V : Example of the Timeline Charts/Coding Sheets of the

Transcribed Think-aloud Protocols

CHAPTER I INTRODUCTION

This thesis is a report of a think-aloud protocol-based study of foreign language writing processes and strategies. The study was based primarily upon the think aloud protocols of four Indonesian tertiary level EFL students taken while they were writing an English essay. This first chapter of the thesis presents the background of the study, specifies the problems of the study, and describes its significance. The chapter concludes by noting the scope of the study, defining some special terms used, and listing the organizations of the thesis.

1.1. Research Background

Nowadays, English has become one of foreign languages that are important to be acquired in Indonesia, especially if it is viewed from the effect of globalization. As the language of knowledge and information, English has become an important language for international communication in the globalization era. Therefore, it is essential that our next generations be able to communicate fluently in English, both in oral and written modes. Through English mastery, they can compete with people from different parts of the world.

As stated in the introduction of KBK Bahasa Inggris SMA/MA (English Competence Based Curriculum for High School) (Departemen Pendidikan Nasional, 2003), English writing proficiency has become one of the language teaching elements

emphasized in the national curriculum. Furthermore, it is also stated that university students are targeted to acquire English writing in various genres in both informational and epistemic levels. In informational level, students are expected to be able to access knowledge with their language proficiency; meanwhile, the epistemic level requires students to be able to express their knowledge into the target language (Wells, 1987 cited in Departemen Pendidikan Nasional, 2003). In the same vein, Emi Emilia (1998: 47) in one of her articles, entitled “Pendekatan Process Dalam Pengajaran Menulis (Process Approach in the Teaching of Writing)”, also argues that writing plays an important role in accelerating learning processes, building character developments, and establishing empowerments of the writers. The statements above indicate that mastering English writing proficiency has become essential for Indonesian tertiary students to help them become more successful in their study both nationally and internationally.

However, writing an English essay in a foreign language setting like in Indonesia is often perceived as a daunting task for students. This may happen because English is not used as the medium of communication among the people in daily conversation. For most EFL learners in Indonesia, writing in English tends to be secondary, i.e. it is only done during classroom writing activities with limited exposures of authentic written expressions. In a similar vein, Nunan (1999: 271) states that it is a big challenge for second/foreign language learners to produce “a coherent, fluent, extended piece of writing”, since the rhetorical conventions of texts —i.e. structure, style, and organization— of different languages are often different

from one another. Therefore, they need to apply some efforts (strategies) in order to recognize and manage the differences (Leki, 1991) found in the process of writing an English essay. The challenge in writing an English essay above was also found in the participants’ process of writing in this study. They seemed to have problems in finding the right words that are relevant to their intended meanings as well as organizing their ideas into a coherent, fluent, and extended essay. In order to cope with those problems, they applied some strategies.

As has been noted by Alister Cumming (2001) in his review on the last two decades studies of how people learn to write in a second language, many studies have been conducted to reveal the processes and strategies of second/foreign language writing (see also Dyson & Freedman, 1991). Most of the studies (e.g. Arndt, 1987; Raimes, 1987; Skibniewski, 1988; Hirose & Sasaki, 1994; Sasaki & Hirose, 1996; Victori, 1997; Roca de Larios et al., 1999, 2001, 2008; Sasaki, 2000; Hu & Chen, 2006) base their theories on the cognitive models of English as a first language writing processes and strategies (e.g. Emig, 1971; Stallard, 1974; Perl, 1978, 1979; Pianko, 1979; Sommers, 1980; Flower & Hayes, 1980, 1981; Bereiter & Scardamalia, 1987), with the belief that basically there are similarities between first and second language writing processes and strategies (see Zamel, 1976; Skibniewski & Skibniewska, 1986; Arndt, 1987; Raimes, 1987; Jones & Tetroe, 1987; Fagan & Hayden, 1988; Skibniewski, 1988; Moragne e Silva, 1989 for further analysis on the similarities and differences between first and second language writing processes and strategies). Therefore, in the present study, two basic models of cognitive processes

and strategies of first language writing studies (e.g. Flower & Hayes, 1981; Scardamalia & Bereiter, 1987) were also employed as the basic theories. This decision was taken since both of them were relevant to the topic of this study, i.e. cognitive processes and strategies in writing.

However, in the second/foreign language writing process, as were also found in the participants’ writing processes of the present study, it was found that writers tend to pay much attention to decisions about the right form/structure as well as to searching for appropriate words in the second language (Cumming, 2001: 5). This is in line with Silva (1993)’s statement that compared to first language writing, second language writing is “strategically, rhetorically, and linguistically different in important ways” (p. 669). Those second/foreign language problems may constrain writers’ attention to formulate complex ideas, their capacity to cope with the high knowledge demand situations, and the scope of their planning (Jones & Tetroe, 1987; Fagan & Hayden, 1988; Uzawa & Cumming, 1989; Whalen & Menard, 1995; Roca de Larios et al., 1999; Cumming, 2001). To deal with the constraints, the writers need to apply a set of strategies that can be used to manage the second/foreign language problems in the writing process.

Furthermore, one of meta-cognitive researchers in writing studies, Gerring (1990) suggests that “writing is a highly complex enterprise, one which requires the knowledge and regulation of many cognitive activities” (p.1, cited in Chen, 2003: 1). In this case, the cognitive activities may include many complex and repetitive strategies. Among others are retrieving information from memory, analyzing readers’

needs, determining genre and writing style, brainstorming for ideas related to the topics, planning structure and content before and while writing, organizing ideas, adjusting wording and sentence patterns, reassessing and revising the written text, and confirming the whole text with writer’s goal for the task. Those activities, which were regulated by their meta-cognitive knowledge, were also found in the pre-writing, pre-writing, and post-writing stages of the participants’ processes of writing in the present study. As suggested by several researchers in second/foreign language learning and writing strategies, meta-cognitive strategies are the key to successful language learning (O’ Malley & Chamot, 1990; Chamot, 2004) and are directly responsible for the execution of cognitive activities, such as reading and writing (Wenden, 1991; Mu, 2007). Therefore, more research on meta-cognitive strategies in second/foreign language writing process, as what was concerned in this study, should be conducted.

As far as the researcher is concerned, most studies in second/foreign language writing process have compared the processes and strategies of skilled and unskilled (e.g. Zamel, 1983; Skibniewski & Skibniewska, 1986; Raimes, 1987; Skibniewski, 1988; Sasaki, 2000, etc.) or between good and poor (e.g. Victori, 1997) writers, based on their cognitive style and capacity, writing experience, and personality. However, the present study examined the differences of meta-cognitive strategies used by higher- and lower-scored writers with different levels of English language proficiency. Different with the previous study, the determinations between higher- and lower-scored writers here were made based on the scores of participants’ writing

products. This decision was taken in order to compare the processes and the products of writing strategies used. As stated by some researchers on learning strategies (e.g. Brown & Palinscar, 1982; Brown, 1980; and Wenden, 1991, 1993) cited in Victori (1997: 179), there is direct relation between meta-cognitive strategies and more successful language learning. Thus, the scores of the writing products may become the indicators of success in this study to be related with the use of meta-cognitive strategies. Besides, different writers with different levels of language proficiency may perform different strategic behaviors in their writing processes (see Roca de Larios et

al., 1999; Manchon et al., 2000). Therefore, this thesis focused on investigating the

processes and meta-cognitive strategies used by four Indonesian tertiary level EFL student writers in their process of writing an essay in English as a foreign language based on their scores of the writing products as well as their levels of language proficiency.

Last but not least, basically there were several verbal report methods utilized by previous researchers in their investigations of writing processes and strategies (Cohen, 2001). However, this study only applied concurrent think-aloud protocols or also called “composing aloud” (Emig, 1971; Perl, 1978, 1979, 1984; Raimes, 1985, 1987), as were also done by several researchers in writing process studies (e.g. Flower and Hayes, 1980, 1981; Perl, 1978, 1979, 1984; Arndt, 1987; Raimes, 1987; Whalen & Menard, 1995; Armengol-Castells, 2001; Hu & Chen, 2006; Roca de Larios et al., 2008; etc.) and retrospective questionnaires taken immediately after executing the task.

The use of the two data collecting methods stated above was instigated by some reasons. First, the concurrent think-aloud protocols seemed to have been the most chosen method of data collections in the studies of writing process and strategies. This may happen because the main goal of this kind of study was to describe the patterns of processes and strategies used while writing an English essay. Besides that, this method was concerned as better than the other methods for its richness, authenticity, and less structuring of data than those gained from the others (Pressley & Afflerbach, 1995; Jansen et al., 1996). Moreover, the immediateness of the data collection time, concerning the writers’ use of short term memory, had also become an advantage of this method. Through think-aloud method, “insights which are difficult or even impossible to obtain by other methods” (Hurd, 2007: 4; see also Perl, 1984: 23) could be revealed. For example, the information obtained from think-aloud method may not appear on the draft and may be forgotten by the time an interview or questionnaire takes place. Therefore, based on the considerations above, think-aloud method was chosen as the main method of gaining the data in this study.

Second, as cautioned by Adina Levine and Thea Reves (1998) in their comparative study on the uses of think-aloud protocols versus strategy questionnaires on reading and writing processes and strategies, think-aloud protocols should be applied together with another data collecting methods in order to gain the validity of the data. For this reason, think-aloud protocols in combination with post-writing questionnaires were employed as the methods in identifying the EFL student writers’ processes and strategies in this study.

1.2. Objectives of the Study

Based on the research background stated above, a study entitled “Writing Processes and Meta-cognitive Strategies of Four Indonesian Tertiary EFL Students” was conducted, aiming

1. To identify the processes of writing underwent by the EFL students.

2. To identify the meta-cognitive strategies used by the participants in their processes of writing an English essay.

1.3. Significance of the Study

The results of the study are expected to give contributions to both the theory and practice of second/foreign language writing studies. First, this study may fill the gap of the scant of studies in think-aloud protocol based research on the process and strategy of writing conducted in foreign language context, especially in Indonesia. As has been noticed by Guangwei Hu and Bo Chen (2006), most studies with the aims of revealing the processes and strategies of writing were conducted in western educational settings involving ESL learners. Even though much research has been conducted in EFL settings, most of them were in Chinese (e.g. Wang & Wen, 2002; Hu & Chen, 2006; etc.), Polish (e.g. Skibniewski & Skibniewska, 1986; Skibniewski, 1988; etc.), and Spanish (e.g. Victori, 1997, 1999; Roca de Larios et al., 1999, 2001, 2006, 2008; Armengol-Castells, 2001; Roca de Larios & Murphy, 2001; Manchon, et

al., 2005; etc.) language settings. Therefore, research to uncover the complex patterns

context and language setting is needed in order to add information to the theory of writing in the country.

Second, the information provided from the results of the study is expected to lead to the enhancement of writing teachers/lecturers’ reflections of their current practices. The reflections may enable teachers/lecturers to adapt their approaches in teaching English writing with more considerations to their learners’ awareness of thinking about their writing process and strategies. Moreover, this may also raise the issue of ‘teach-ability’ of meta-cognitive composing strategies in order to empower learners with control of their compositions. This study investigates the processes as well as meta-cognitive strategies used by students in writing an English essay. By having identified the processes and strategies used by the students in the process of writing an essay, English writing teachers can introduce the processes and teach the strategies to them, so that they can help the students adapt to the target discourse community more quickly.

1.4. Scope of the Study

This study focused on a think-aloud protocol-based study of four Indonesian tertiary level EFL student writers taken while they were writing an English argumentative essay. Based on the analyses of participants’ writing products, the participants involved in this study were categorized as having higher and lower scores of writing. Besides, they also have different levels of language proficiency, from low intermediate to advanced (based on Proficiency Test of English for Speakers of Other Languages (hereafter PTESOL) test taken a week before the data collection time).

Previous studies on process and strategies of second/foreign language writing (e.g. Sasaki & Hirose, 1996; Roca de Larios et al., 1999; Manchon et al., 2000; Sasaki, 2004; etc.) have shown that different writers may perform different strategic behavior in their processes of writing a second language essay. Furthermore, those studies have also indicated that the level of language proficiency may influence the processes and strategies used in writing an essay in a second/foreign language. Therefore, this study investigates the processes as well as meta-cognitive strategies used by the participants with higher and lower scores of writing, besides their different levels of language proficiency.

1.5. Definitions of Terms

To avoid misunderstandings, misinterpretations, or ambiguities of several terms used in the study, the researcher defines the terms as follows:

1. Writing processes: the stages gone through by writers as they compose, as suggested by experts in writing process studies (e.g. Emig, 1971; Perl, 1978, 1979, 1981; Pianko, 1979; Stallard, 1979; Hayes & Flower, 1980; Murray, 1980; Sommers, 1980; Murray, 1982; Flower and Hayes, 1981; Zamel, 1983; Bereiter & Scardamalia, 1987; Skibniewski, 1988; Roca de Larios et al., 2001, 2008).

2. Writing strategies: various strategies/methods employed by writers in their composing processes in order to generate ideas, plan, draft, organize, evaluate revise, and to reduce constraints faced while writing (Cummings, 1989, 2001; Victori, 1997; Manchon, 2001; Mu, 2007).

3. Meta-cognitive strategies: strategies applied by second/foreign language learners that entail planning for, monitoring, and/or evaluating the success of a learning activity (O’ Malley & Chamot, 1990: 44), and are directly responsible for the execution of a writing task (Wenden, 1991; Mu, 2007: 9). 4. Think-aloud protocols: “rich data sources” consisting of subjects’ “spoken

thoughts” (Ericsson & Simon, 1980) associated with working on a task.

1.6. Thesis Organization

This thesis consists of five chapters with the organization presented as follows:

Chapter I discusses the introduction dealing with the background of the research, the objectives of the study, the research methodology in brief, the significance of the study, the scope of the study, the definitions of the terms, and the organization of the thesis.

Chapter II discusses the literature review underpinning the values of writing in general, the theories of writing process and strategy based on previous studies in second language writing processes and strategies; as well as the use of think-aloud protocols in the studies of writing processes and strategies.

Chapter III discusses the research methodology in detail dealing with the research questions.

Chapter IV discusses the data presentation, analyses, and findings of the results, based on those obtained from the think-aloud protocols and post-writing questionnaires.

Chapter V discusses the conclusions, limitations and recommendations for the next research.

CHAPTER III

RESEARCH METHODOLOGY

The previous chapter has discussed the literature review underpinning the values of writing in general, as well as the processes and strategies used in writing. This chapter discusses some important aspects related to research methodology, including the research questions, research design, participants of the study, research instruments, data collection place, time, and procedures, and data analysis procedures.

3.1. Research Questions

The study was conducted in order to answer the following questions:

1. What writing processes did the Indonesian tertiary EFL students undergo?

2. What meta-cognitive strategies did the participants apply in their processes of writing an English essay?

3.2. Research Design

In order to answer the questions above, think aloud protocols were used as a main method. As suggested by Cumming (2001) that think aloud protocols are recommended to be used in investigating writing process and strategies, since most of the data in these studies are “mental and self-directed” (p.6). Furthermore, earlier studies in composing processes (such as Emig, 1971; Perl, 1978; Flower & Hayes,

1980, 1981; Raimes, 1985; Arndt, 1987; Bereiter & Scardamalia, 1987; Victori, 1997, 1999; Armengol-Castells, 2001; Mu, 2007; and Roca de Larios et al., 2008) and language learning strategies (e.g. O’Malley & Chamot, 1990) employed the think aloud protocols as their research method. This method was used in the belief that it would give a more accurate picture of participants’ online processing on what they were thinking while writing, so that the processes and strategies used by the subjects could be described.

In addition, this research is also a case study, which is qualitative in nature, observing and analyzing intensely on a “single unit” or ”bounded system” (Smith, 1978 cited in Merriam, 1998: 19, see also Fraenkel and Wallen, 2003: 439) such as an individual program, event, group, activity, or ongoing process and developing “as full an understanding of that case as possible” (Punch, 1998: 150 cited in Silverman, 2005: 126) in a “bounded context” (Miles and Huberman, 1994: 25). In the same vein, Sharan Meriam (1998) suggests that “a case study design is employed to gain an in-depth understanding of the situation and meaning for those involved” (p.19). Here, the interest of the study is more “in the process rather than outcomes” (1998:19). Through qualitative case study, the rich data gained from small number of participants are expected to give thick descriptions of the strategies undergone by the writers while they were writing an English essay. Even though the focus of the present study is qualitative, occurrences of each strategy category, the time spent by each writer, and the length of the essay were counted by using simple counting

method. This was especially done in order to illustrate the results of the data analyses in chapter four.

Following previous case studies on English as first, second, and foreign language writing process and strategies (e.g. Emig, 1971; Perl, 1978; Zamel, 1983; Arndt, 1987; Raimes, 1985, 1987; Victori, 1997, 1999; Armengol-Castells, 2001; and Hu & Chen, 2006) except Flower and Hayes (1981), this study employed multiple data collection methods. Among them were think aloud protocols, post-writing questionnaires, and video-taped observations of the participants in the process of writing. As suggested by Yin (1993) that “multiple sources of evidence – converging from the same set of issues” (p.32) or “multiple data collections and analytic procedures” (Freebody, 2003: 83) should be applied in order to increase the construct validity of the study (Yin, 1993: 39-40). Therefore, the findings obtained from the think aloud protocols were triangulated with the findings obtained from the other data collecting methods used in this study.

3.3. Participants of the Study

There were four participants involved in this case study, consisting of two males (Writer 1 and 2) and two females (Writer 3 and 4). All of them were from the same level of education, i.e. third year university students, taking an English Education program in an education university in Bandung. All participants involved had taken two semesters of English writing course at the university and voluntarily

participated in the study. As stated in chapter one, the participants of the study were classified based on their levels of English language proficiency, as one of the variables.

In order to get more valid results of the participants’ levels of English proficiency, a week before the data collection time, each of them took a PTESOL (Proficiency Test of English for Speakers of Other Languages) test at the language center. This test was taken without any preparations in order to get their real ability of English language proficiency. This procedure is in line with Carson and Kuehn (1992) and Cohen and Brooks-Carson (2001) that standardized test should be used in classifying participants’ language proficiency. As listed by Carson and Kuehn (1992), participants with TOEFL scores of 420-480 could be classified as having a low intermediate level of English proficiency. Those who scored 480-520 were classified as high intermediate, and those who scored 525 and above were considered as advanced. Since PTESOL test has been claimed as similar to TOEFL, in which English structure, reading comprehension, and listening are included (Sukyadi, in private conference through telephone, June 24, 2009), the scores obtained from PTESOL could be equaled to those of TOEFL.

From the PTESOL test, it was found that Writer 1 got the highest score, i.e. 553. Thus, his English language proficiency was classified as advanced. Meanwhile, Writer 2 and 3 got 483 and 500 respectively, indicating a high-intermediate level of English proficiency. Writer 4 got the lowest score on the PTESOL test, i.e. 467,

describing a low-intermediate level of English language proficiency. The descriptions of the participants’ characteristics can be seen more clearly in Table 3.1. (Appendix I)

3.4. Research Instruments

There were two instruments employed in the data collection of this study. Both of them were writing prompt and post-writing questionnaire. All of which will be elaborated in the following sub-titles.

3.4.1. Writing prompt

A topic of a recent issue in English education field was used as the writing prompt in the data collection. The prompt was as follows:

Nowadays, English mastery has been emphasized to the educational

outcomes in Indonesia. This is shown through the regulation of the 2003 System of National Education Act number 20 article 50 subsection 3, which states that “the government and/or the local government[s] will establish at least one educational institution of all levels to be developed into an international standard educational institution”. Since this regulation was released, the government has established numerous international standard schools (Sekolah Bertaraf Internasional = SBI), in which English is used as the medium of instruction for subjects that have global significance, such as science and mathematics. However, there are pro and contradictive opinions towards this program. The proponents state that this program may be successful and may increase students’ proficiency of both English and the content subjects. In contrast, the opponents argue that this program may not be feasible since the teachers and the supporting facilities have not been ready yet. How do you think about this? Do you agree or disagree to the use of English as the medium of instruction in math and science subjects as those practiced in SBI programs? Why? State your stance whether pro or contra and explain your arguments specifically in an approximately 300 words persuasive essay!

The writing prompt with the topic of bilingualism in education above was used in the belief that all participants would be familiar with it, since all of them were

students of English education program at the same university. Besides, bilingualism in education has become a burning issue in Indonesian educational field (Hallett, 2005; Arifin, 2003). As argued by Roca de Larios et al. (2008: 35), topic familiarity is essential to be considered in designing writing tasks since it may influence participants’ degree of involvement. Therefore, the researcher used the topic above as the writing prompt in her study.

3.4.2. Post-writing questionnaire

Besides utilizing the writing prompt, the researcher also made use of post-writing questionnaire as an instrument in order to substantiate the findings gained from the think aloud protocols. Adopted from Sasaki and Hirose (1996: 170-171)’s study, the questionnaire asked the writing processes and strategies undergone by the participants at the three main stages, i.e. pre-writing, writing, and post-writing (see Appendix V). As presented in the questionnaire, six questions were included into three main points, such as first, whether or not the participants did any pre-writing activities. If they did, they should state the things done during the pre-writing stage. Second, the questionnaire asked the participants’ strategy in keeping their flow of writing, for example, whether they wrote the essay based on the outlines that had been made before or directly wrote the essay without making any overall plans. Besides that, this point also asked the participants’ foci of attention (such as grammar, spelling, content, organization, or diction) during writing the essay (measured by a 5-point Likert scale) and whether they had any problems concerning

those aspects as well as their strategies in solving the problems. In the third point of the questionnaire, the participants were asked to state the things they did after writing the essay, whether they did some proof-readings or revisions or not.

3.5. Data Collection Place, Time, and Procedures

In this study, the data were consisted of participants’ audio and video recorded think aloud protocols taken while performing an English persuasive/argumentative writing task. Besides that, participants’ retrospective verbal data were also gained from post-writing questionnaires. The think aloud session was done within a time limit of one hour plus another 30 minutes for revisions conducted in the next week. Total time spent for the think aloud sessions was 90 minutes. The data collection was conducted in a language laboratory in the Language Center (Balai Bahasa) of Indonesia Education University, Bandung. This place was chosen for its availability of advanced recording tools and quiet atmosphere, so that the participants’ voice could be clearly recorded. Since the primary data used in this study was participants’ verbalizations during writing an English essay, clear recordings of their voice were obviously required. The data were collected in two separated times for the essay and revision, started from April 30th to May 14th 2008.

There were two kinds of data collection procedures employed in this thesis. Among them were think aloud protocols and retrospective verbal report in a form of

post-writing questionnaire. Each of which will be discussed in more details in the following sub-titles.

3.5.1. Think aloud protocols

As stated previously that this study applied think aloud protocols as a main method of gaining the data. However, this method has some limitations (see point 2.5 in the previous chapter) that may threaten the validity. Therefore, to anticipate the threats of validity associated with this method, the researcher took some measures in collecting the data, which basically entailed making decisions regarding to three main areas. Among them were the choice of the task, the elicitation of the verbalizations, and the coding of the data (see Roca de Larios et al., 2001 for further information about this issue). The first two measures will be discussed in this sub-title, whereas the latter will be discussed in the data analysis procedures sub-title.

First, regarding to the choice of the task, an argumentative writing mode was chosen in the belief that it would lead the participants to adopt a more knowledge transforming approach (see Bereiter & Scardamalia, 1987), and especially, to apply more problem solving behavior in their process of writing. Here, their cognitive activity would be the focus of attention, so that it would be more available for verbal reports (Ericsson & Simon, 1984 cited in Roca de Larios et al., 2008). It was expected that by using an argumentative writing task, the problems of automaticity

commonly occur in think aloud protocols could be overcome and a more useful and informative protocol data would be obtained.

Second, regarding to the elicitation of verbalizations, before starting the data collection session, the researcher gave carefully selected wordings of instruction in Indonesian. This procedure was taken in order to avoid over- and/or under-estimations and interpretations about the things required from the participants and the type of information the participants should report (Jourdenais, 2001: 356 in Roca de Larios et al., 2008: 35). The instruction translated into English was as follows:

I’d like you to write a composition based on the writing prompt that I am going to give you now. While you are writing your composition, I would like you to SAY ALOUD anything and everything that goes through your mind. You have to do everything that you would normally do when writing a composition. The only difference is that today you are going to do it talking aloud. You may use any language(s) that you normally use when writing. You will have a maximum of 1 hour to complete the task.

Immediately before the participants dealt with the task, all of them took a 20 minute training session in order to familiarize and make them get used to the think aloud procedure. This session includes practicing with a mock composition with the topic “The Importance of English Mastery in the Future” without any demonstrations provided by the researcher in order to avoid influencing the participants’ behavior (see Roca de Larios et al., 2008: 35). As stated by Smith (1994, cited in Roca de Larios et al., 2001:508), researcher is not to provide any modeling of the think-aloud method, since participants may restrict their “thoughts” to those they had seen modeled. Besides that, modeling the think-aloud technique might have influenced

participants’ choice of language. This was avoided, since it was essential to let the participants use the language they would feel more comfortable with for their verbalizations. After all participants have got used to talking whilst writing the mock essay, the data collection session was started.

During the execution of the task, the participants were allowed to use both English and Indonesian. This was done because language choice in think aloud procedure might require participants’ recoding of information. Besides that, any constraints such as poor command of the language being used for reporting may cause information loss or alter the original thinking processes. Therefore, the participants were allowed to make use of any of both languages they felt comfortable with.

Furthermore, in the execution of the task, the participants sometimes forgot to think aloud. In order to remind them to keep talking what they were thinking, the researcher prompted them with questions such as “What are you thinking about?” or “What is in your mind?”, whenever the participants paused for more than five seconds. By taking the two measures above, as suggested by Roca de Larios et al. (2001), the validity of the think aloud method may be increased.

3.5.2. Post-writing questionnaire

Right after the participants finished doing the think aloud procedure, they were asked to fill in a post-writing questionnaire. This was done retrospectively in

nature in order to support the findings gained from the think aloud protocols, so that the researcher could draw a complete description of the participants’ writing process and strategies. Besides that, the findings from this method were also used in order to triangulate the findings from the think aloud protocols as well as observation, so that the validity and reliability of the research may be obtained.

3.6. Data Analysis Procedures

After collecting the data, the researcher analyzed them qualitatively and triangulated them in order to get the conclusion. The procedures of the data analysis will be elaborated in the following subsections.

3.6.1. The Scoring of the Writing Task

In order to compare the strategies used by each participant and the products of the writing process, the final drafts of the essay were collected and scored. To meet the standards of reliability (consistency) and validity (accuracy) issues to the assessment of the writing products, some measures were taken by the researcher. Among them were using standardized holistic scoring rubrics as the guide for scoring, and employing and training another rater to evaluate the essay, so that more objective scores may be obtained. As stated by Cassanave (2001), in order to reach objectivity in writing assessment, the “criteria by which writing samples are assessed” (p.121) should be described and “raters to evaluate essays according to the same criteria” (p.121) should be trained.

First, in order to fulfill the requirements of reliability and validity, a standardized holistic scoring guide was applied. Even though the subjects were given additional 30 minutes to revise their first drafts on the next day, the tasks were timed writings, which essays were essentially first drafts. As stated by Carson and Kuehn (1992: 168) that timed writing essays “do not allow for an evaluation of the full range of writing abilities that might be explored if the writers were given more time and opportunities to revise”. Therefore, in this study the essays were evaluated by using “The 1993-1994 Focused Holistic Scoring Guide for Persuasive/argumentative Composition” published by North Carolina Annual Testing Commission (1994). This scoring guide was chosen for its valid and reliable scorings for persuasive/argumentative essays.

There are two domains included in the scoring guide as the criteria of evaluating; i.e. composing characteristics, which “bear on or influence the effective delivery of the writer’s statement or message” (North Carolina Annual Testing Commission, 1994), and language conventions of written composition. The composing domain consists of four general aspects; i.e. main idea, supporting details, organization, and coherence. First, regarding to the main idea, the essays should imply that the writer has identified the cause that is being promoted through the composition and focused on the subject matter being identified. As stated in the scoring guide, moving away from the subject matter may cause the focus to be unclear and the main idea is weakened. It has been the nature of persuasive mode that

the writer must attempt through the composition to bring about a change in a current condition, whether circumstantial or philosophical (Connor & Lauer, 1988: 138), so that the reader becomes in the same line with the writer’s point of view (Derewianka, 2004: 75). Therefore, the writer should take a position regarding to the subject matter by possibly making use of inductive organizational plan and building an argument through examples or details so that the readers may identify it. A composition would be regarded as having a clear main idea if the writer’s position is clear, whether or not the statement of the position occurs at the beginning of the composition.

Second, in dealing with supporting details, the scoring guide mentions that the essays should provide sufficient and appropriate information (or evidence) to persuade the reader that the change being promoted has merit. Sufficient details are indicated by the use of powerful information load, such as using facts and evidences. This is in line with Sebranek et al. (1999) that “the more types of convincing evidence you offer, the stronger your argument will be” (p. 122). If the writer simply lists undeveloped details or using repetitious paraphrasing of the same point, the composition may be regarded as insufficient. Appropriateness is indicated when the composition can maximally persuade the specified audience by using selective and well-presented information, among others are prediction, statistics, observation, expert testimony, comparison. Therefore, an essay would be regarded as sufficient and appropriate if it uses enough supporting details that can persuade the readers with well-presented information.

Third, a composition is valued as having a good command of organization if it establishes a sense of overall plan in the beginning, development, and ending. Besides that, it should also follow the schematic structure of persuasive/argumentative writing, either exposition (thesis statement (position) – arguments – restatement of the thesis (position)) or discussion (issue – argument for – argument against – conclusion) (see Derewianka, 2004: 70-71; Feez & Joyce, 1998: 138-140, for further discussion about types of persuasive/argumentative writing). Fourth, an essay may be regarded as coherent if the sentences (such as ideas, causes, and/or statements) are logically connected. Coherency results from the effective use of organization, unity, reason, and logic by employing one or more cohesive devices. As identified by Halliday and Hasan (1976, in Gibbons 1991: 83-86), there are five ways of making text cohesive; among them are by using reference, conjunction (connectives), substitution, ellipsis, and lexical cohesion (synonyms). First, reference is words pointing to something in the text. These words may point back to something that has been mentioned or forward to what is going to be said. Reference includes words such as he, she, it, his, hers, him, her, their, them, here, there, the, this, those,

that. For example, like what mentioned in the following text:

A tall black figure was outside Sophie’s house. The figure turned and faced her window, and then he walked on. He came to Mr. Goochey’s house and there he stopped. (adapted from The BFG by Roald Dahl, cited in Gibbons, 1991: 84).

From the text above, it can be seen that the underlined words refer to what have been mentioned in the previous sentences. The use of these words makes the text more coherent.

Second, conjunctions or connectives are words that are used to link and organize ideas. These include words indicating addition, contradiction, cause and effect, and conclusion, such as and, but, because, so, unless, although, if, however,

nevertheless, therefore, etc. Besides that, there are also words that are used to

sequence ideas in time (time connectives), such as and, and then, before, after, later,

the next day, millions of years later, afterwards, etc. By using conjunctions and

connectives, readers may interpret or predict the next idea in the text. The example of the use of conjunctions can be seen in the following text:

He walked all day although he was exhausted. Finally he stopped outside a small hotel in a village. If he went in, someone might recognize him, but unless he stopped to rest he would get no further. So, hoping he would remain unrecognized, he opened the door. (adapted from Gibbons, 1991: 84).

From the example above, it can be seen that the use of conjunctions will make the text sound more fluent and structured, since this helps readers to recognize and interpret the main ideas of the text.

Third, substitution is used whenever something has been mentioned in the previous sentence, in order to avoid the repetitions of the same word. For example, the substitution of the word bike with one in the following sentences:

He was given a new bike for his birthday. His old one was too small for him. (Gibbons, 1991: 85)

The example above indicates that by using substitution, the text can sound more coherent and logical.

Fourth, a text can be coherent by using ellipsis, in which some parts of a sentence with parallel structure are omitted, in order to avoid redundancy. For example, like those happen in the following sentences:

Some cats like cheese but some (cats) don’t (like cheese).

He sat down, (he) stood up and then (he) sat down again. (Gibbons, 1991: 85)

From the example above, we can see that the words in brackets can be omitted in order to make the text more effective and avoid it to sound unnecessarily ‘wordy’.

Finally, in order to avoid repetitions of using the same words in a text, a writer may apply lexical cohesion, in which different words with similar meanings (synonyms) are used. Besides that, the use of lexical cohesion may also indicate links of the sentences. For example, like those used in the following text:

They were now in a country of thick forests and rushing rivers. The giant had definitely slowed down and was now running more normally although normal was a silly word to describe a galloping giant. He leapt over a dozen rivers. He went

rattling through a great forest then down into a valley and up over a range of hills,

and soon he was galloping over a desolate wasteland that was not quite of this earth. (adapted from The BFG by Roald Dahl, in Gibbons, 1991: 85-86).

The example above indicates that many of the content words in the text fall into two groups, i.e. those to do with movement (the bold ones) and those to do with physical aspects of the scene (the italic bold ones). These strings of related words can hold the sentences together.

The composing domain is evaluated by using four scale focused holistic rubrics (see Table 3.2. in Appendix I). Here, the possible highest score of 4, indicating a strong command of persuasive writing, would be assigned to compositions with following criteria: (a) the writer has identified and taken a positive position about the subject matter; (b) the writer has appropriately and effectively presented persuasive information in support of the position having been taken: (c) the organization provides a clear sense of logical progression and overall completeness; and (d) the composition is coherent. In contrast, the possible lowest score of 1, exhibiting a lack command of persuasive writing, would be devoted to compositions that have: (a) indication that the writer has read the prompt and has attempted to respond to it; (b) little or no sense of focus on the subject and little or no clear cause for the position provided; (c) no strategy or sense of control or have control with only one or two unelaborated reasons to support the position.

In addition to the scores above, the essays were also rated for the language conventions used (see Table 3.3. in Appendix I). The written composition language conventions were rated as “+” (plus) or “-“(minus), indicating whether or not the writer has demonstrated a reasonable and acceptable level of proficiency in the

domain, including sentence formation, usage, and mechanics. This means that a composition would be rated as having a “+” for convention if it has complete sentences, correct usages, and skillful use of mechanics. Sentences are considered to be complete if the “T-units and their constituents are internally related and either coordinated with or separated from each other” (North Carolina Annual Testing Commission, 1994). Examples of typical errors in sentence formation are run-on sentences (in which two complete T-units are fused together without any conjunctives and punctuated as one sentence) and sentence fragments (which do not form a complete T-unit but is punctuated as a sentence).

Regarding the usage, a composition would be considered as having a correct usage if it makes use of acceptable and effective selections of words, grammatical forms, and idioms for standard, formal, edited, written English communication. Errors in using pronouns, verbs, subject-verb agreements, and negating constructions in English composition would be rated as “-“ for the conventions. Finally, a “+” would be given to the compositions with adept use of mechanics. This involves the use of conventional representational cues and symbols (such as spelling, punctuation, and capitalization) in constructing language units. Errors in spelling, punctuation, and capitalization would also be scored as “-“.

Furthermore, the length of the final written products (such as the total paragraphs, sentences, and words written by the participants in their essays) was also counted in order to know how the use of certain strategies affect the final written

products of each participant. The scores of the written products, together with the total time spent by each participant, and the length of the final written products, compared with the strategies used by the participants in this study, were used as the data to answer the second research question.

The next issue of reliability and validity is related to the objectivity of the evaluation, whether or not the writing samples were judged objectively. As proposed by White (1994) that “[u]nreliable [i.e. inconsistent] measures are merely subjective impressions by disparate individuals” (p.36) and the evaluations of writing must “stand up to outside scrutiny” (p.30). Therefore, in scoring the participants’ essay, another rater was trained and employed, besides the researcher herself. This outside rater was an English writing teacher, who was also an expert in the field. In order to get stable scoring criteria, the items of the scoring rubrics stated before were previously discussed and agreed to be used as the scoring guide. After that, each rater evaluated each of the essays twice, in order to get rater reliability. The intra-rater scorings were done within a range of two weeks, in order to ensure the consistency of the rater’s scorings. Finally, the scores produced by each rater were then compared and discussed. Any disagreements of the scores were discussed and compromised by both raters, so that final objective scores would be obtained.

3.6.2. Think aloud Protocols

There were several steps taken in the analysis of the think aloud data. First, following Perl (1978, 1984), Raimes (1985), Arndt (1987), Victori (1997, 1999), and Roca de Larios et al. (2008) the recorded think aloud protocols were transcribed verbatim using the standard Indonesian and English writing systems with the inclusion of false starts, hesitations, repetitions, pauses, and paralinguistic features (such as laughter, coughs, etc.). The underlined written text was then distinguished from the processes which generated it and rereading/repetitions were italicized. Finally, the revisions/editings made to the written text were marked in the protocols with double underlining (see Appendix IV).

After transcribing the think aloud protocols, the researcher set up the coding scheme through a recursive process. This process involved a number of passes through the protocols with series of tests and reviews of the data until the final categories were constructed and defined (Kasper, 1998; Smagorinsky, 1994). Once the categories had been settled, the protocol was coded and the results were put on a timeline chart (Perl, 1984) (see Appendix V). In addition, the video recordings were played while coding each of the protocols in order to know exactly what the participants were doing; whether they were writing whilst speaking or writing silently, or they were reading the text.

The categorizations of the coding scheme were based on the preliminary analysis on the think aloud transcription as well as the synthesis of theories of second/foreign language composing process and learning strategies stated in chapter two. The coding schemes used in the study were as follows:

1. Planning strategies: Strategies undertaken when planning and talking out the

next ideas, and explicitly stating his/her objectives for content organization and writing procedures.

a. Planning overall content and ideas (PLid): retrieving ideas, relating new information, making connections among existing ideas and setting general content goals either in the form of notes or verbalization.

b. Planning procedures (PL pr): Planning subsequent actions such as procedures or strategies to be adopted (sub), or planning delayed actions such as postponing an action deliberately (Post)

c. Planning organization (PLor): grouping ideas, deciding on the overall organization of the text, deciding how to sequence ideas and how to structure the text as a whole or parts of it.

d. Planning linguistic text (PLtx): rehearsing or verbalizing several versions of the text to be produced.

2. Monitoring strategies: Strategies undertaken when checking and verifying

progress in the composing process and when identifying oncoming problems. a. Task Monitoring (TM): assessing how the task is progressing; how

successfully the intended meaning is conveyed; tracking the use of how well a strategy is working or whether there is a need for adopting a new one.

b. Self Monitoring (SM): expressing one’s feelings towards the task, becoming aware that one is having problems.

3. Evaluating strategies: Strategies undertaken when reconsidering the written

text, previous goals, planned thoughts, and changes undertaken on the text. a. Evaluating strategies (EV): questioning or evaluating the written text or

planned thoughts.

b. Reviewing strategies (REW): reconsidering previously set goals (g), or reading (R) for the whole text (wl), the previous sentence (s) or paragraph (p)

c. Revising strategies (REV): making changes to the text in order to clarify meaning (such as problems with ideas, word choice, cohesion, coherence, and organization)

d. Editing strategies (ED): making changes to the text to correct the grammar (gr), syntax (syntx), word choice (wc) (not for clarifying meaning), spelling (sp), and punctuation (punc).

After the protocols were coded in the timeline charts, the items of the categories were counted by using a simple count, in order to know the frequency of each category occurrences. Then, based on the timelines and the occurrence frequency of each strategy, the strategies used in the writing process of each participant were analyzed and described.

Besides counting the occurrence of the strategies used, the researcher also measured the time spent by each participant in each of the writing stages in order to illustrate the findings and compare it with the length and score of the final written products, so that the conclusions to the findings could be made.

3.6.3. Post-writing questionnaire

Participants’ answers to the questionnaire were coded in accordance with the coding schemes for the think aloud protocols and triangulated with the findings gained from the think aloud protocols as well as the writing products and observation in order to validate the results of the study.

3.7. Concluding Remarks

This chapter has attempted to explain the research method used in the study in terms of the characteristics of the participants, instruments, data collection place, date, and procedures, and the procedures taken in the data analysis. The next chapter, i.e. chapter four, will present the results obtained from the data analysis and the discussions of the results.

CHAPTER V CONCLUSIONS

Chapter four has displayed and discussed the data analyses from think-aloud protocols and post-writing questionnaires. Chapter five concludes the study by proposing the contribution or consolidation of the reported study to the previous studies in writing processes and strategies (Section 5.1). Besides that, this chapter also warns the readers with the limitations of the study (Section 5.2.) and puts forward the implications of the study to the teaching of writing that may be beneficial for English as a foreign language writing teachers and lecturers (Section 5.3.). Finally, in the same section, some recommendations are also suggested for further think-aloud protocol based studies of writing processes and strategies.

5.1. Conclusions of the Study

This thesis reports the results of a think-aloud protocol-based study of the writing process and meta-cognitive strategies gone through and applied by four Indonesian tertiary EFL students in writing an English essay. The purpose of the study is to discover the processes of writing as well as the way the meta-cognitive strategies used by the participants.

Findings of the study, both from the think aloud protocols as well as the post-writing questionnaire analyses, suggest that all participants in this study underwent complex, non-linear, recursive processes of writing, in which the sub-processes of planning, monitoring, revising/editing, and reviewing may occur repeatedly in any

stages of writing process. This finding is relevant to the findings of previous researchers (e.g. Emig, 1971; Perl, 1978, 1979, 1981; Pianko, 1979; Stallard, 1979; Flower and Hayes, 1980, 1981; Sommers, 1980; and Bereiter & Scardamalia, 1987).

Besides that, it is also shown that all participants, in accordance with the scores of the writing products and their levels of language proficiency, applied all meta-cognitive strategies differently. It is evident that some writers are able to undertake a writing approach that is more conducive to produce better writing products than others with lower or higher levels of language proficiency. Interestingly, lower scored essays were written by participants with advanced and low-intermediate levels of language proficiency. This corresponds to previous researchers (e.g. Cumming, 1989, Whalen & Menard, 1995; Sasaki & Hirose, 1996; Victori, 1997) that there is a lack of interdependence between linguistic and writing competence.

From the analyses of the think-aloud protocols and the post-writing questionnaire in chapter four, it was revealed that the higher-scored writers were found to be more interactive in approaching the writing processes. This was evident in the larger numbers of meta-cognitive strategies (i.e. planning, monitoring, and evaluating strategies) applied at different levels. First, they planned the overall ideas, by generating more ideas and alternatives and setting more numbers of procedural and organizational goals than the lower-scored ones. Second, they constantly and effectively monitored what and how they were doing, by identifying oncoming problems and assessing their texts against their intended meanings. Finally, they

evaluated their essays by applying larger numbers of revision. Besides, they also set a wider range of reviewing and editing objectives, including reviewing and editing the topic knowledge and the content of the essays.

Furthermore, the higher-scored writers were found to have approached the writing processes with more effort than the lower-scored ones. This was indicated by their efforts in getting their intended meanings across, besides having enough knowledge about the topic of writing. These efforts and possessions of enough topic knowledge has led them to the generation of more alternative ideas, more rehearsing attempts, more evaluations of both ideas and word choice, and more revisions on ideas and coherence. As a result, these meta-cognitive strategies seemed to have brought the use of stronger words, more elaborated ideas and more cohesive and coherent paragraphs than those of the lower-scored ones.

The findings are consistent with those obtained in other first, second, and foreign language writing studies, in which the more successful writers were found to be more interactive (e.g. Raimes, 1987), to put more effort into their writing approach (Stallard, 1974), and to engage in a larger range of problem-solving or decision-making strategies (Flower & Hayes, 1980; Cumming, 1990), especially strategies for planning, organizing, evaluating, and revising/editing content. Finally, in terms of learning strategies, the findings of this study are also relevant to those of other research in the field (e.g. Brown, 1980; Brown & Palinscar, 1982 cited in Victori, 1997; and Wenden, 1991, 1993), which suggest that there is direct relationship between the use of meta-cognitive strategies and more successful language learning.

5.2. Limitations of the Study

Some factors in this study may exhibit some limitations, one of which is the size of the sample, i.e. four Indonesian university level EFL students. This small sample may result in a low level of generalization. However, considering the limited time and budgets provided to conduct a research with larger subjects, the researcher decided to conduct a small sample one instead. Moreover, there might be unnoticeable biases on behalf of the researcher in synthesizing the results of the study. The overall synthesis of the findings may be affected by the limitations of the researcher’s knowledge of the subject matter as well as her analytical and expressive abilities. Therefore, a second rater was employed in analyzing as well as evaluating the data and drafts of this thesis, so that more objective results may be obtained.

Besides, the opportunities to demonstrate the subjects’ strategies in composing may be limited, since they were required to write only in one specific writing mode, i.e. argumentative. As argued by some experts (e.g. Bereiter & Scardamalia, 1987; Cumming, 1989; Grabe, 2001; Wang & Wen, 2002; Hu & Chen, 2006), the same writer writing in different modes (e.g. argumentative and descriptive) may apply different processes and strategies. However, the application of the think-aloud procedure may serve as a rich-data source that provide the researcher with loads of data to be analyzed.

Finally, the limited time available had caused the participants only given limited time to do the writing task. A more complete data may be obtained if they were given more time to revise their essays, as that in a more natural setting. The

factors above may result limitations in demonstrating subjects’ overall writing processes and strategies.

5.3. Recommendations

Concerning the limitations of this study, some measures are recommended to be done for further research. Among them are, first, in order to obtain the feasibility of the findings to be generalized, further study should involve more participants, with specific characteristics, such as those with certain level of language proficiency or those with certain experience of English writing.

Second, in order to avoid biases that commonly appear in think-aloud based studies, other procedures of data collection technique, such as interview and questionnaires should be used. This is done as a member-check in order to support or validate the data obtained from the think-aloud protocols.

Finally, more think-aloud protocol-based studies in writing process and strategies should be applied, since studies using this method are still rare, especially in Indonesian context. However, more writing tasks, with different modes of writing (such as argumentative and descriptive or narrative) should be employed in order to demonstrate more process and strategies in writing. By applying the recommendations above, further studies are hoped to be more improved than the present.

REFERENCES

Afflerbach, P. (2000). ‘Verbal Reports and Protocol Analysis’. In M.L. Kamil, P.D. Mosenthal, and R. Barr (Eds.), Handbook of Reading Research, 3, pp.163-179. Mahwah, New Jersey: Lawrence Erlbaum.

Alamargot, D., Favart, M., and Galbraith, D. (2000). ‘Evolution of Ideas in Argumentative Writing: Writing as a Knowledge-constituting Process?’. EARLI

– Writing Conference 2000.

Anderson, N.J. and Vandergrift, L. (1996). ‘Increasing Metacognitive Awareness in the L2 Classroom by Using Think-aloud Protocols and Other Verbal Report Formats’. In R.L. Oxford (Ed.), Language Learning Strategies Around the

World: Cross-cultural Perspectives, pp.3-18. Honolulu, Hawaii: University of

Hawaii Press.

Arifin, Anwar. (2003). Memahami Paradigma Baru Pendidikan Nasional dalam

Undang-Undang SISDIKNAS, POKSI VI FPG DPR RI.

Ardnt, V. (1987). ‘Six Writers in Search of Texts: A Protocol Based Study of L1 and L2 Writing’. ELT Journal, 41, pp.257-267.

Armengol-Castells, L. (2001). ‘Text-generating Strategies of Three Multilingual Writers: A Protocol-based Study’. Language Awareness, 10(2 &3), pp.91-106. Benton, S.L., Kiewra, K.A., Whitfill, J.M., and Dennison, R. (1993). ‘Encoding and

External-Storage Effects on Writing Processes’. Educational Psyhcology, 85(2), pp.267-280.

Bereiter, C., and Scardamalia, M. (1987). The Psychology of Written Composition. Hillsdale, NJ: Lawrence Erlbaum Associates.

Bernardini, S. (1999). ‘Using Think-aloud Protocols to Investigate the Translation Process: Methodological Aspects’. RCEAL, Working Papers in Applied

Linguistics. Cambridge: Cambridge University Press.

Bernardini, S. (2001). ‘Think-aloud Protocols in Translation Research: Achievements, Limits, Future Prospects’. Target, 13(2), pp.241-263.

Block, E. (1986). ‘The Comprehension Strategies of Second Language Readers’.

TESOL Quarterly, 20(3), pp.463-494.

Britton, J. (1978). ‘The Composing Process and the Functions of Writing’. In C. Cooper and L. Odell (Eds.), Research on Composing. Urbana, Illinois: National Council of Teachers of English.

Carson, J.E. and Kuehn, P.A. (1992). ‘Evidence of Transfer and Loss in Developing Second Language Writers’. Language Learning, 42(2), June, 1992, pp. 157-182. Cassanave, C. P. (2004). Controversies in Second Language Writing: Dilemmas and

Decisions in Research and Instruction. USA: The University of Michigan.

Cavalcanti, M. (1982). ‘Using the Unorthodox, Unreasonable, Verbal Protocol Technique: Qualitative Data in Foreign Language Reading Research’. In S. Dingwall, S. Mann, and F. Katamba (Eds.), Methods and Problems in Doing

Applied Linguistic Research, pp. 72-85. UK: University of Lancaster.

Chamot, A.U. (2004). ‘Issues in Language Learning Strategy Research and

Teaching’. Electronic Journal of Foreign Language Teaching, 2004, 1(1), pp. 14-26 [Online]. Available at: http://e-flt.nus.edu.sg/. Retrieved on December, 16th 2008.

Chamot, A.U. (2005). ‘Language Learning Strategy Instruction: Current Issues and Research’. Annual Review of Applied Linguistics, 25, pp.112-130.

Chen, F. (2003). Students’ Perception of English Writing and Their Metacognitive

Knowledge. Thesis. Taiwan: National Yunlin University of Science and

Technology.

Chien, S. (2007). ‘The Role of Writing Strategy Use in Relation to Chinese EFL Students’ Achievement in English Writing: A Cognitive Approach’. In CamLing, 2007, pp.25-31.

Cohen, A.D. (1986). ‘Mentalistic Measures in Reading Strategy Research: Some Recent Findings’. English for Specific Purposes, 5, pp.131-145.

Cohen, A.D. (1987). ‘Studying Learner Strategies: How We Get Information?’. In A. Wenden and J. Rubin (Eds.), Learner Strategies in Language Learning, pp.31-39. Englewood Cliffs, New Jersey: Prentice Hall.

Miles, M.B. & Huberman, A.M. (1994). Qualitative Data Analysis: An Expanded Sourcebook. (2nd edn.). Thousand Oaks, California: SAGE.

Moragne e Silva, M. (1989). ‘A Study of Composing in a First and Second Language’. Texas Papers in Foreign Language Education, 1, pp.132-151. Mu, C. (2007). ‘A Proposal for a Taxonomy of ESL Writing Strategies’. STETS

Language and Communication Review, 6(1), pp.5-13.

Mu, C. and Carrington, S. (2007). ‘An Investigation of Chinese Students’ English Writing Strategies’. TESL-EJ, 11(1).

Murray, D.M. (1978). ‘Internal Revision: A Process of Discovery’. In C. Cooper and L. Odell (Eds.), Research on Composing. Urbana, Illinois: National Council of Teachers of English.

Murray, D.M. (1980). ‘Writing as a Process: How Writing Finds its Own Meaning’. In T. Donovan & B. McClelland (Eds.), Eight Approaches to Teaching

Composition. Urbana, Illinois: National Council of Teachers of English.

Myles, J. (2002). ‘Second Language Writing and Research: The Writing Process and Error Analysis in Student Texts’. TESLJ-EJ, 6(2), September 2002 [Online]. Available at: http://www-writing.berkeley.edu/TESL-EJ/ej22/toc.html. Accessed on February 14th 2008.

North Carolina Annual Testing Commission. (1994). North Carolina Focused Holistic Scoring Guide: The Persuasive/Argumentative Composition Grade. (1993-1994) [Online]. Available at: http://www.ERICDIGEST.com. Accessed on December 16th 2008

Nunan, D. (1999). Second Language Teaching and Learning. Boston: Heinle and Heinle Publishers.

O’Malley, J. & Chamot, A. (1990). Learning Strategies in Second Language Acquisition. Cambridge, England: Cambridge University Press.

Oxford, R. L. (1990). Language Learning Strategies: What Every Teacher Should Know. Harlow: Newbury House.

Perl, S. (1978). Five Writers Writing: Case Studies of the Composing Processes of Unskilled College Writers. Unpublished Doctoral Dissertation. New York

Perl, S. (1979). ‘The Composing Processes of Unskilled College Writers’. Research in the Teaching of English, 23, pp.317-336.

Perl, S. (1981). ‘Creativity and the Composing Process: Making Thought Visible’. In ERIC, ED202025.

Perl, S. (1984). ‘Coding the Composition Process: A Guide for Teachers and Researchers’. In ERIC, ED240609.

Pianko, S. (1979). ‘A Description of the Composing Process of College Freshmen Writers’. Research in the Teaching of English, 23, pp.5-22.

Poulisse, N., Bongaerts, T. and Kellerman, E. (1986). ‘The Use of Retrospective Verbal Reports in the Analysis of Compensatory Strategies’. In C. Faerch and G. Kasper (Eds.), Introspection in Second Language Research, pp.213-229.

Clevedon, UK: Multilingual Matters.

Pressley, M. and Afflerbach, P. (1995). Verbal Protocols of Reading: The Nature of Constructively Responsive Reading. Hillsdale, New Jersey: Lawrence Erlbaum. Qi, D. S. (1998). ‘An Inquiry into Language Switching in Second Language

Composing Processes’. In The Canadian Modern Language Review, 54, pp. 413-435.

Raimes, A. (1985). ‘What Unskilled ESL Students do as They Write: A Classroom Study of Composing’. In TESOL Quarterly, 19, pp. 229-257.

Raimes, A. (1987). ‘Language Proficiency, Writing Ability, and Composing

Strategies: A Study of ESL College Student Writers’. In Language Learning, 37, pp. 439-468.

Ransdell, S.E. (1995). ‘Generating Think-aloud Protocols: Impact on the Narrative Writing of College Students’. American Journal of Psychology, 108, pp.89-98. Riazi, A. (1997). ‘Acquiring Disciplinary Literacy: A Social-Cognitive Analysis of

Text Production and Learning among Iranian Graduate Students of Education’. Journal of Second Language Writing, 6, pp.105-137.

Richards, J.C., Platt, and Platt. (1992). Longman Dictionary of Language Teaching and Applied Linguistics. Essex: Longman.

Roca de Larios, J., Manchon, R.M., and Murphy, L. (2006). ‘Generating Text in Native and Foreign Language Writing: A Temporal Analysis of Problem-solving Formulation Processes. The Modern Language Journal, 90(1), p.100-114. Roca de Larios, J. Manchon, R. M., and Murphy, L. (2008). ‘The Foreign Language

Writer’s Strategic Behavior in the Allocation of Time to Writing Processes’. Journal of Second Language Writing, 17, pp. 30-47 [Online]. Available at: http://www.sciencedirect.com/ciil/Fulltext/Journal_of_second_language_writing_ vol_15_no_1/Vol_17_1_2008/Article_3.pdf. Accessed on May 15th 2008.

Roca de Larios, J., Marin, J., and Murphy, L. (2001). ‘A Temporal Analysis of Formulation Processes in L1 and L2 Writing’. Language Learning, 51(3), September, 2001, pp. 497-538.

Roca de Larios, J. and Murphy, L. (2001). ‘Some Steps Towards a Socio-cognitive Interpretation of Second Language Composition Processes’. International Journal of English Studies. 1(2), 2001, pp.25-45 [Online]. Available at:

http://www.um.es/engphil/ijes. Accessed on January 26th 2008.

Roca de Larios, J., Murphy, L., and Marin, J. (1999). ‘The Use of Restructuring Strategies in EFL Writing: A Study of Spanish Learners of English as a Foreign Language’. Journal of Second Language Writing, 8, pp.13-44.

Sasaki, M. (2000). ‘Toward an Empirical Model of EFL Writing Processes: An Exploratory Study’. Journal of Second language Writing, 9, pp.259-291. Sasaki, M. (2004). ‘A Multiple Data Analysis of the 3.5-year Development of EFL

Student Writers’. Language Learning, 54, pp.525-582.

Sasaki, M. and Hirose, K. (1996). ‘Explanatory Variables for EFL Students’ Expository Writing’. Language Learning, 46(1), March 1996, pp.137-174. Scardamalia, M. & Bereiter, C. (1987). ‘Research on Written Composition’. In C.

Wittrock (Ed.), Handbook of Research on Teaching. pp.778-803. New York: MacMillan.

Sebranek, P., Kemper, D., and Meyer, V. (1999). Write Source 2000: A Guide to Writing, Thinking, and Learning. Wilmington, MA: Houghton Mifflin Company. Shaughnessy, M. (1977). Errors and Expectations. New York: Oxford University

Silva, T. (1993). ‘Toward an Understanding of the Distinct Nature of L2 Writing: The ESL Research and its Implications’. TESOL Quarterly, 27, pp.657-677.

Silva, T., Reichelt, M., Chikuma, Y., Douval-Couetil, N., Mo, R. J., Velez-Rendon, G., and Wood, S. (2006). ‘Second Language Writing Up Close and Personal: Some Success Stories’. In Barbara Kroll, Exploring the Dynamics of second Language Writing, pp.93-114. Cambridge: Cambridge University Press.

Silverman, D. (2005). Doing Qualitative Research: A Practical Handbook. London: SAGE Publications.

Skibniewski, L. (1988). ‘The Writing Processes of Advanced Foreign Language Learners in Their Native and Foreign Languages: Evidence from Thinking Aloud and Behavior Protocols’. Studia Anglica Posnaniensia, 21, pp.177-186.

Skibniewski, L. and Skibniewska, M. (1986). ‘Experimental Study: The Writing Processes of Intermediate/Advanced Foreign Language Learners in Their Native and Foreign Languages’. Studia Anglica Posnaniensia, 19, pp.143-163.

Smagorinsky, P. (1989). ‘The Reliability and Validity of Protocol Analysis’. Written Communication, 6(4), October 1989, pp.463-479.

Smagorinsky, P. (1991). ‘The Reliability and Validity of Protocol Analysis’. Written Communication, 6, pp.463-479.

Smagorinsky, P. (1994). ‘Think-aloud Protocol Analysis: Beyond the Blackbox’. In P. Smagorinsky (Ed.), Speaking About Writing: Reflections on Research

Metodology. Thousand Oaks, California: Sage Publications.

Sommers, N. (1980). ‘Revision Strategies of Student Writers and Experienced Adult Writers’. College Composition and Communication, 31(4), pp.378-388.

Spandel, V. (2009). Creating Writers through 6-Trait Writing. Boston: Pearson. Stallard, K. (1974). ‘An Analysis of the Writing Behavior of Good Student Writers’.

Research in the Teaching of English, 18, pp.206-218.

Stratman, J.F. and Hamp-Lyons, L. (1994). ‘Reactivity in Concurrent Think-aloud Protocols: Issues for Research’. In P. Smagorinsky (Ed.), Speaking about Writing: Reflections on Research Methodology, pp.89-112. Thousand Oaks, California: Sage Publications.

Swain, M. and Lapkin, S. (1995). ‘Problems in Output and the Cognitive Processes They Generate: A Step Towards Second Language Learning’. Applied

Linguistics, 16(3). Oxford: Oxford University Press.

Thorson, H. (2000). ‘Using the Computer to Compare Foreign and Native Language Writing Process: A Statistical and Case Study Approach’. The Modern Language Journal, 2(84), pp.155-169.

TIEW (The Institute for Excellence in Writing). (2001). Teaching Writing: Structure and Style Syllabus and Seminar Workbook. California: The Institute for

Excellence in Writing.

Torrrance, M. & Jeffery, G. (1999). The Cognitive Demands of Writing. Processing Capacity and Working Memory in Text Production. Amsterdam: Amsterdam University Press.

Victori, M.B. (1997). ‘EFL Composing Skills and Strategies: Four Case Studies’. RESLA, 12, pp.163-184.

Victori, M.B. (1999). ‘An Analysis of Written Knowledge in EFL Composing: A Case Study of Two Effective and Two less Effective Writers’. System, 27, pp.537-555.

Walshe, R.D. (1981). Every Child can Write. Rozelle, NSW: PETA.

Webster, J.B. (1994). Blended Structure and Style in Composition. Saskatoon: Blended Sound-Sight Method of Learning.

Wenden, A. (1991). Metacognitive Strategies in L2 Writing: A Case for Task

Knowledge. Proceedings of the Georgetown University Roundtable of Languages and Linguistics. Georgetown: Georgetown University Press.

Wenden, A. (1993). Strategic Interaction and Task Knowledge. Proceedings of the Georgetown University Roundtable of Languages and Linguistics, Georgetown: Georgetown University Press.

Whalen, K. and Menard, N. (1995). ‘L1 and L2 Writers’ Strategic and Linguistic Knowledge: A Model of Multiple-Level Discourse Processing’. Language Learning, 45(3), September 1995, pp.381-418.

White, E. M. (1994). ‘Portfolios as an Assessment Concept’. In L. Black, D. A. Daiker, J. Sommers, and G. Stygall (Eds.), New Directions in Portfolio

Assessment: Reflective Practice, Critical Theory, and Large-scale Scoring, pp. 25-39. Portsmouth, New Hampshire: Boynton/Cook.

Witte, S. P. and Cherry, R.D. (1994). ‘Think-aloud Protocols, Protocol Analysis, and Research Design: An Exploration of the Influence of Writing Tasks on Writing Processes’. In P. Smagorinsky (Ed.), Speaking about Writing. Reflections on Research Methodology. Thousand Oaks, California: Sage Publications.

Yan, G. (2006). ‘A Process Genre Model for Teaching Writing’. English Teaching Forum [Online]. Available at:

http://exchanges.state.gov/forum/vols/vol43/no3/p18.htm#top. Accessed on January 13th 2008.

Yang S. X. (2002). ‘The Difference between Successful Writers and Unsuccessful Writers in Strategy Use’. Foreign Language World, 89(3), pp. 57-64.

Yin, R.K. (1993). Applications of Case Study Research. California: Sage Publication. Zamel, V. (1982). ‘Writing: The Process of Discovering Meaning’. TESOL

Quarterly, 16, pp.195-209.

Zamel, V. (1983). ‘The Composing Processes of Advanced ESL Students: Six Case Studies’. TESOL Quarterly, 17, pp.165-187.