VIETNAM NATIONAL UNIVERSITY – HO CHI MINH CITY UNIVERSITY OF SOCIAL SCIENCES & HUMANITIES FACULTY OF ENGLISH LINGUISTICS & LITERATURE THE IMPACTS OF EFL WRITING PORTFOLIOS ON STUDENTS’
INTRODUCTION
The general context of the study
Since the 1990s, English has been recognized as the most favored foreign language in Vietnam (Dang, 2012, p.9) To enhance Vietnamese students' English proficiency, various teaching practices have evolved Initially, the grammar-translation approach dominated, focusing on English grammar rules and rote learning (Dang, 2012) Subsequently, the communicative teaching approach was introduced to encourage interaction among students and between students and teachers, thereby facilitating the teaching of English macro skills For example, the Tieng Anh book series for grades 6 to 12 incorporates this new program in secondary and high schools.
3 comprised four English skills (i.e., listening, speaking, reading and writing skills), together with grammar, pronunciation and vocabulary sections
Despite years of English education, many university students lack confidence in their writing skills, struggling to compose formal emails or structured paragraphs This gap in writing proficiency can be linked to several key factors, highlighting the need for more meaningful interactions in English classes to enhance both spoken and written communication.
Despite being a crucial part of the national English curriculum, writing lessons in Vietnam are often neglected Many secondary and high school teachers tend to avoid teaching writing skills due to the significant time required for students to complete assignments and for teachers to provide feedback Consequently, classroom time is predominantly focused on grammar exercises and vocabulary memorization Additionally, students' limited language proficiency and lack of exposure to the target language hinder their writing skill development.
The disconnect between teaching and assessment in writing is evident, as teachers are encouraged to use a process-oriented approach, yet summative assessments predominantly follow a product approach, occurring only twice per course For example, the 2018 high school graduation English exam focused solely on comprehensive reading, neglecting essential writing skills such as sentence construction, opinion expression, and idea summarization (Cameron, 2018) Additionally, there is a lack of a consistent and effective feedback policy to guide students in improving their writing after essay tests, if such feedback is provided at all (Le, 2018).
Given the difficulties above, it seems really hard for learners to master writing skills thoroughly, even for students of high level of English Despite the importance of
Mastering writing skills presents significant challenges for learners, with persistent issues in teaching, learning, and assessment A key factor contributing to students' struggles is the lack of autonomy, which is considered the primary reason for their failures in learning (N.D, 2011, as cited in Phung, 2016) The researcher emphasizes that without autonomy, students are unlikely to succeed in their learning endeavors, regardless of the favorable conditions provided.
Developing learner autonomy is crucial for training an independent and qualified workforce in Vietnam, as acknowledged by various Vietnamese educators and researchers (Dang, 2012; Le, 2013; Nguyen, 2009) However, a recent study by lecturers at Can Tho University (Nguyen, Chung, Truong, Pham, 2014) indicates that learner autonomy remains a relatively new concept for many Vietnamese teachers and learners.
In Vietnam, the exploration of learner autonomy in writing practice, particularly among young learners, remains under-researched This article will delve into the specific context of the study, focusing on the development of writing portfolios and the English language program at the center.
The specific context of the study
This study focuses on enhancing English teaching and learning outcomes by promoting learner autonomy and improving students' writing skills It implements a formative assessment tool in English classes at a local language center, specifically the Regional Training Center (RETRAC) of the Southeast Asian Ministers of Education Organization (SEAMEO) RETRAC provides English language courses for students of various ages and proficiency levels, including a Master's program.
The research conducted at the site aims to benefit young learners and teachers at SEAMEO, contributing to the overall development of Ho Chi Minh City The researcher's familiarity with the program's administration, curriculum, and teaching environment facilitated the study, which involved 60 teenage learners from four different classes.
Since July 2017, the SINT-B program has been focused on teaching English to young learners According to Coerll (2006), students have a preference for smaller class sizes, ideally between 10 to 20 students, with the optimal number being 19 or fewer.
Participants engaged with the final five units of the intermediate-level Solutions coursebook (2nd edition, Falla & Davies, 2012, Oxford University Press) The English program emphasizes integrated skills, with foreign teachers focusing on Vocabulary, Culture, Speaking, and Listening, while Vietnamese teachers handle Grammar, Reading, Writing Student performance is evaluated through a midterm written test and a final assessment that includes both oral and written components, along with an individual presentation project, all designed by the Division of Foreign Studies Both Vietnamese and foreign instructors hold at least a TESOL certificate, with some possessing master's degrees.
The youngster program consists of eight consecutive courses across four stages, covering four English levels from elementary to upper-intermediate Upon completion, students can opt for IELTS preparation or General English courses in the adult program The learning outcome of this program aligns with a B2-C1 level according to the CEFR.
Despite having an intermediate level of English proficiency, many teenage students struggle significantly with writing skills, particularly in using appropriate language and organizing their ideas Observations indicate that these students often lack the ability to effectively manage their self-study, including planning, monitoring, and evaluating their writing, as well as recognizing their strengths and weaknesses The traditional methods of teaching writing have proven inadequate in addressing these challenges Therefore, it is crucial to explore more effective approaches that can enhance learner autonomy and improve writing skills.
For the vital aforementioned issues, the researchers herself found the passion, the necessity and the urgency to conduct this empirical study in order to make this situation
6 better The current study attempted to use writing portfolios to enhance students’ autonomous learning capabilities and their writing achievement
The researcher opted for traditional writing portfolios over e-portfolios due to the significant context of the study At the time, classrooms were equipped with only one computer connected to a TV screen, a whiteboard, and individual student chairs, with limited access to a lab room Many teenage students lacked the ability to use computers freely at home for writing assignments, and training them on an e-portfolio platform would require additional time and computer skills Consequently, paper-based portfolios were more suitable, allowing students to include their classroom writing without the need for extensive re-composition on a computer Additionally, as they had no prior experience with portfolios for developing writing skills, starting with folder-and-paper portfolios was an effective approach.
This study proposes the use of writing portfolios as an effective tool to enhance teenage learner autonomy and improve overall writing achievement The researcher specifically investigates how the implementation of writing portfolios fosters various dimensions of learner autonomy Additionally, the study examines the impact of writing portfolios on students' overall success in developing their writing skills.
The aims of this study can be addressed through the following research questions and sub-questions 1 :
(1) To what extent do writing portfolios improve learner autonomy dimensions? Sub-questions:
1 For all null and alternative hypotheses regarding Research Questions 1 and 2, see Appendix 1, pp 105-106
Workplace practices (WP) significantly promote student involvement in their learning, enhance their use of English, and foster their ability to reflect on their educational experiences.
(2) What are the impacts of writing portfolios on students’ overall writing achievement?
With those aims accomplished by answering the two research questions, the current study has its own significance as follow
The study offers a comprehensive literature review on learner autonomy and writing portfolios, highlighting a gap in empirical research regarding portfolio implementation in L2 contexts like Vietnam While procedural guides exist for L1 writing teachers, there is limited investigation into how portfolio use affects students' learning and writing skills in L2 settings, particularly in terms of enhancing learner autonomy and improving writing outcomes.
Previous studies on students' writing skills in portfolio-based classrooms have been limited in sample size and duration, highlighting the need for more quantitative research (Tran & Duong, 2018) While there has been significant research on learner autonomy and portfolios, no study in Vietnam has specifically focused on the effects of writing portfolios on young learners' writing achievement and the dimensions of learner autonomy addressed in this research.
This study aims to bridge the gap in both national and international literature regarding the use of writing portfolios to improve learner autonomy and writing achievement Additionally, it provides a detailed overview of writing portfolio development, offering valuable insights for teachers in the studied context and similar English teaching environments in Vietnam, enhancing their understanding of the teaching, learning, and assessment processes within portfolio classrooms.
The SEAMEO learners can revitalize their English writing skills with this tool, enhancing their mastery and transforming outdated methods to alleviate boredom in learning By prioritizing students' needs in portfolio assessment, the implementation of writing portfolios can be refined to optimize their writing abilities and foster autonomous learning behaviors.
The study highlights the potential of writing portfolios as a formative assessment tool in educational settings, emphasizing their underutilization By discussing the advantages and disadvantages of this portfolio design and the associated teaching methodologies, the research aims to pave the way for the successful implementation of a portfolio model in the future.
This study investigates the promotion of learner autonomy and the immediate impact of writing portfolio implementation on students' writing skills over a five-month course It focuses solely on the enhancement of learner autonomy dimensions and students' achievements in EFL writing, excluding other macro skills and students' attitudes towards writing portfolios The findings are based on the author's reflections and experiences with writing portfolio construction Additionally, the research was conducted in a Vietnamese EFL language center, limiting the generalizability of the results to similar contexts.
The overall structure of the study takes the form of five themed chapters, including the introductory chapter, namely Introduction, as presented here
LITERATURE REVIEW
METHODOLOGY
Questionnaire
A designed questionnaire was utilized to address the first research question regarding learner autonomy Oppenheim (1992) highlights two key advantages of using questionnaires for data collection: they allow for the simultaneous gathering of consistent, standard, and precise data from all respondents, and they enable participants to express their opinions more easily and naturally However, potential drawbacks include the risk of students misunderstanding questions, which can lead to inaccurate responses, particularly when the questionnaire is in a foreign language To mitigate these issues and streamline the process for respondents, the researcher provided a Vietnamese version of the questionnaire and conducted a pilot test with other students to ensure clarity and straightforwardness.
The questionnaire was designed to explore how writing portfolios enhance learner autonomy dimensions, addressing the first research objective It consists of a formal standardized format divided into two parts to test the research hypotheses related to the initial question Adapted from Phung's (2016) 51-item Likert-scale questionnaire on speaking e-portfolios and learner autonomy, this study integrates three fundamental aspects to assess the impact effectively.
50 operationalized pedagogical principles and five writing rubric criteria in the questionnaire Here follows the description of the questionnaire
Part 1 was designed to obtain data about students’ experiences in using portfolios in learning English
Part 2 was to explore students’ perceptions towards learner autonomy 54 items following the Likert-scale with 5 indexes indicating students’ level of agreement on each item (1 – strongly disagree, 2 – disagree, 3 – no opinion, 4 – agree, 5 – strongly agree) were comprised in this part Ten learner autonomy sub-dimensions accounting for three broad learner autonomy dimensions, namely Learner Involvement, Appropriate Target Language Use, and Learner Reflection, were the criteria to categorize those 54 questionnaire items
The writing assignments in the course significantly contributed to my autonomous learning behaviors In addition to the two reasons proposed by Oppenheim (1992), the questionnaire design was influenced by the need for participants to receive consistent stimuli through standardized wording and question order (FCDR, n.d., para.6) Furthermore, utilizing a multiple-choice response format enhances the speed at which respondents can complete the questionnaire, ultimately saving students' time (FCDR, n.d.).
The questionnaire assessed three key dimensions of learner autonomy: Learner Involvement, Appropriate Target Language Use, and Learner Reflection, which together encompass ten sub-dimensions The Learner Involvement dimension includes five sub-dimensions: Inventory the task, Prepare for my performance, Check my performance while writing, Control and modify my performance while writing, and Evaluate my performance after writing The Appropriate Target Language Use dimension consists of three sub-dimensions: Increase spoken English use, Increase written English use, and Increase the use of English as a language of thought Finally, the Learner Reflection dimension includes two sub-dimensions: Reflect on my learning and Reflect on what I learnt from the task.
51 presents the ten sub-dimensions accounting for three learner autonomy dimensions and the equivalent questionnaire items
Table 3.4 Learner autonomy dimensions and sub-dimensions operationalized in the study and the equivalent questionnaire items
Learner Autonomy No of Items Items
SD1 Inventory the task 4 1 2 3 7 SD2 Prepare for my performance 4 4 5 6 8 SD3 Check my performance while writing 6 9 10 11
12 13 14 SD4 Control and modify my performance while writing 6 15 16 17
18 19 20 SD5 Evaluate my performance after writing 6 21 22 23
SD6 Increase spoken English use 3 30 31 32 SD7 Increase written English use 9 27 28 29
36 37 38 SD8 Increase the use of English as a language of thought 6 39 40 41
SD9 Reflect on my learning 5 45 46 47
48 49 SD10 Reflect on what I learnt from the task 5 50 51 52
Writing pre-test, post-test
The study's pre-test and post-test consisted of two parts featuring different writing types The pre-test's first part required students to write an informal letter, which they had previously learned in unit 5 of the SINT-A course In contrast, the post-test's first part involved writing a formal letter Both tests included a second part focused on opinion essays, taught in units 6 and 9 of the SINT-B course Each test was allocated sixty minutes, with Part 2 valued at twice the weight of Part 1, suggesting students spend approximately forty minutes on Part 2 and twenty minutes on Part 1 The pre-test and post-test materials are available in Appendices 9 and 10.
The selection of these writing types was based on three key reasons Firstly, they represented two of the five writing types covered in the course Secondly, choosing only one type would have limited the researcher’s ability to evaluate students' writing skills across different formats Lastly, the writing topics were closely related to everyday life, making it easier for students to generate ideas Notably, all writing assignments, including pre-tests and post-tests, were consistent for both groups.
To reduce students' workload and alleviate stress that may hinder participation, tests were scheduled on days with shorter lesson content, such as during workbook corrections in reading lessons or at the end of the course for final test evaluations Additionally, each test included guidelines to help students structure their writing more effectively.
The writing tests were designed and validated according to the course objectives focused on writing skills As a result, the outcomes of these tests not only addressed the second research question but also provided valuable insights for future improvements in teachers' writing instruction.
Validity and reliability
The questionnaire was designed by aligning three dimensions of learner autonomy, as outlined by Little (1999 & 2010), with students' metacognitive strategies for writing skills, based on Goh & Burns's framework for productive speaking skills (2012) The researcher meticulously evaluated both content validity and face validity.
The questionnaire utilized in this study is primarily adapted from Phung's (2016) established design, which emphasizes speaking e-portfolios and learner autonomy Phung ensured the content and face validity of the questionnaire through multiple steps, including a thorough review by a senior lecturer in English with a Master's degree in TESOL from New Zealand, conducted three years ago This process involved an in-depth discussion with the reviewer regarding the operationalized aspects of the questionnaire.
The researcher reviewed 53 terms, providing valuable suggestions for modifying item wording and removing irrelevant items from the scales Additionally, the face validity of the questionnaire was ensured by carefully examining its format, layout, font, size, and wording of the statements.
Before distributing the questionnaire to students, the researcher sought final advice from her supervisor and translated it into Vietnamese The revised questionnaire was piloted with six students from a similar class, leading to final adjustments based on their feedback to ensure content and face validity The completed Vietnamese version of the questionnaire is included in Appendix 8.
To ensure the validity of the writing pre-test and post-test in measuring students' writing skills, the researcher sought advice from her supervisor and the program manager Adjustments were made to align the tests with course objectives and content while ensuring they were feasible for students within the given time Additionally, the test developers thoroughly discussed criteria such as relevant topics, appropriate wording, required functional language, and writing rubrics.
The reliability of the questionnaire was assessed by calculating Cronbach’s Alpha score for the scales related to learner autonomy sub-dimensions To enhance internal consistency, the item deletion method was employed, removing one or more items from each scale.
To mitigate bias, writing test measurements were evaluated by a second scorer, emphasizing the importance of consistency between raters This was crucial for assessing how portfolios can improve students' writing skills, necessitating inter-rater analyses The Kappa statistic was employed to determine the inter-rater reliability of both the pre-test and post-test The results indicated a significant level of agreement between the two scorers, who utilized the same writing rubrics.
The scoring rubric for midterm and final tests is holistic, while the writing rubrics employed by the two raters are holistic for Part 1 and analytic for Part 2 (refer to Appendices 12A and 12B for details) The writing rubric for assessing students' opinion essays was adapted from EnglishZone’s verified writing rubric, which is used in the same coursebook level EnglishZone, part of Hoa Sen University (HSU), has since switched to a different coursebook The researcher validated the writing rubric criteria with an experienced HSU lecturer Additionally, the holistic rubric for the letter part was adopted from SEAMEO.
The two test compilers initially employed a holistic scoring scale to evaluate writings from various classes, followed by a comparison using an analytic rubric to identify any significant score differences Based on insights from this pilot phase, the researcher refined the analytic rubrics and sought the supervisor's approval before implementing both scoring scales in the study Consequently, these two writing rubrics demonstrated adequate reliability for pre-test and post-test evaluations, as supported by Pham’s study.
2015, analytic rubrics have various advantages He suggested that analytic scoring scale made it easier, faster and even more objective for the rater to mark writing papers (Pham, 2015)
The marking scheme for opinion essays includes five criteria: organization, content, sentence variations, vocabulary, and mechanics, while the letter writing assessment uses a holistic rubric that evaluates performance on a scale of 5, 10, 15, and 20 points The distinction between these two marking schemes arises from their differing descriptions for the same criteria, reflecting variations in format, organization of ideas, and the level of formality in vocabulary usage specific to letters.
To increase the reliability of measures for further statistical analyses, the researcher who is also the first rater initially trained the second rater how to use the writing rubrics
In cases where writing papers received significantly different scores—specifically, a discrepancy of two or more points out of twenty—from the initial two raters, a third rater was brought in to reassess those papers using the same rubrics The final score was determined by averaging the score from the third rater with the score from the first two raters that was closest to it Additionally, discussions among the raters were held when necessary to reach a consensus Notably, eight out of 120 pre-test and post-test papers required re-evaluation.
This section outlines the data collection procedure for the study, detailing how quantitative data was obtained through questionnaires and pre-tests and post-tests to address the two research questions.
Questionnaire
To collect data from the questionnaire, an English version was initially created and subsequently translated into Vietnamese for better comprehension by the students The questionnaire was piloted with a group of students from the same course as the experimental and control groups to gather feedback on potentially confusing items Necessary adjustments were made before the official distribution In the nineteenth week of the course, students from both groups were given thirty minutes after the final written test to complete the questionnaire.
56 complete the Vietnamese version of the questionnaire under the guidance of the teacher
Researchers were encouraged to seek clarifications on any unclear items during the data collection process Once gathered, the questionnaire data were meticulously coded and input into SPSS software version 23.0 for comprehensive statistical analysis.
Writing pre-test, post-test
The test developers agreed on the design of students' writing pre-test and post-test, ensuring content and format validity aligned with the course's learning objectives The pre-test featured an informal letter, while the post-test required a formal letter, with both tests including opinion essays The pre-test was administered before the first writing lesson in the third week, allowing students sixty minutes to complete it under teacher supervision, emphasizing creativity without the use of dictionaries or electronic devices On the final day of the course, students took the post-test, which was also timed at one hour The collected papers were photocopied for assessment, with raters using consistent rubrics for both tests The scoring scale ranged from zero to four across five criteria: organization, content, vocabulary, sentence variation, and mechanics, totaling twenty points, which mirrored the writing portion of their midterm and final tests.
After having marked, the pre-test and post-test scores of both groups of students were subsequently inputted into SPSS software version 23.0 for further statistical examination
This study employed a quantitative method to explore two research problems The application of quantitative statistical methods to the questionnaire provided essential evidence regarding the impact of writing portfolios on learner autonomy Additionally, the analysis of pre-test and post-test results demonstrated how the use of writing portfolios influenced students' achievement in writing skills The data analysis followed a structured procedure as detailed below.
Statistical methods for the questionnaire
Data analyses were conducted following a pre-established plan, beginning with a descriptive analysis to assess the normal distribution of questionnaire data The internal consistency of ten scales, representing equivalent sub-dimensions, was evaluated using Cronbach’s Alpha, confirming the reliability of the questionnaire for further analysis Subsequently, independent samples t-tests were performed to identify statistically significant differences between the two groups, specifically comparing the ratings of autonomous learning behaviors between EG and CG students These statistical tests aimed to evaluate the impact of writing portfolios on various dimensions of learner autonomy.
Statistical methods for students’ writing pre-test and post-test
Statistical significance was assessed through analysis of variance and t-tests as needed The researcher utilized Microsoft Excel 2016 to tally the final scores of students for both the pre-test and post-test, noting that Part 2 carries double the weight.
The inter-rater reliability of students' writing pre-test and post-test was assessed using the Kappa test, ensuring the reliability of the data for further statistical analyses Two independent samples t-tests were conducted to compare the writing results of experimental group (EG) and control group (CG) students before and after the intervention The first t-test evaluated differences in writing ability between the two groups prior to the writing portfolio intervention, while the second t-test analyzed the post-test scores to determine any statistically significant differences in writing achievement following the implementation Additionally, a paired samples t-test was performed to compare the means of EG students' pre-test and post-test scores, identifying any significant differences in their writing achievement as a result of the treatment.
This chapter outlines the study's methodology, beginning with a reiteration of the research questions and hypotheses It details the research design, study context, participant selection, and sampling procedures The rationale for instrument selection and development is also discussed, followed by a comprehensive explanation of data collection and analysis frameworks These elements aim to enhance the study's reliability and validity The subsequent chapter will provide further insights into the statistical procedures applied to the quantitative data and a discussion of the results.
RESULTS AND DISCUSSION
Analysis of data
This section presents an analysis of data from the questionnaire and students' writing pre-test and post-test scores, organized into five sub-sections The first sub-section details the distribution of the questionnaire data The second sub-section evaluates the reliability of the questionnaire The third sub-section focuses on the inter-rater reliability of the students' writing scores from both the pre-test and post-test.
(4) the extent to which learner autonomy dimensions were promoted by WP And the last sub-section explains (5) the impact of WP on students’ writing achievement
4.1.1 Distribution of the questionnaire data
The questionnaire is divided into two key sections: the first part gathers information about students' experiences with writing portfolios in their English learning, while the second part includes 54 items on a five-point Likert Scale that assess students' ratings of their autonomous learning behaviors.
The Normal Distribution Test was utilized to assess whether the questionnaire data were normally distributed, as outlined by Ghasemi and Zahediasl (2002) To evaluate the distribution, skewness and kurtosis values of students' responses to each questionnaire item were calculated The results, detailed in Appendix 15, indicated that the skewness and kurtosis values ranged from -2 to +2 for all items George and Mallery (2010) suggest that values of asymmetry and kurtosis within this range are deemed acceptable.
60 order to prove normal univariate distribution Table 4.1 is an extract of descriptive analysis for 54 questionnaire items
Table 4.1 An extract of skewness and kurtosis values of 54 questionnaire items regarding learner autonomy dimensions
Statistic Std Error Preview requirement of the task
The descriptive analysis of skewness and kurtosis indicates that a normal distribution was reasonably achieved for the total sample of 54 items Therefore, the data from these questionnaire items is suitable for further analysis.
To assess the internal consistency of the questionnaire scale, Cronbach’s Alpha coefficients were calculated The inter-item correlation among test items directly influences the Cronbach’s Alpha, with higher correlations indicating better internal consistency when all items measure the same dimensions To enhance the reliability of the scale, the item deleted method is recommended Test results suggest that eliminating certain items can help achieve the highest possible Cronbach’s Alpha for the scales related to learner autonomy sub-dimensions.
In Chapter 3, data from 54 items in Part 2 of the questionnaire were analyzed to investigate how three dimensions of learner autonomy—Learner Involvement, Appropriate Target Language Use, and Learner Reflection—were influenced by WP.
The article is divided into three sections representing equivalent dimensions The first section, Learner Involvement dimension, includes five scales: Inventory the assignment, Prepare for my performance, Check my performance while writing, Control and modify my performance, and Evaluate my performance after writing The second section focuses on Appropriate English Use dimension, comprising three scales: Increase spoken English use, Increase written English use, and Increase the use of English as a language of thought Finally, the Learner Reflection dimension in the third section contains two scales: Reflect on my learning process and Reflect on what I learnt from the assignment A reliability test was conducted for these scales sequentially.
According to the summary below, Table 4.2, the Cronbach’s Alpha values of ten aforementioned scales accounting for three learner autonomy dimensions ranged from .62 (SD2) to 931 (SD10)
Table 4.2 Cronbach’s Alpha coefficients for ten investigated learner autonomy sub- dimensions
Learner Autonomy No of items
Items Alpha Dimensions Sub-dimensions
SD1 Inventory the task 4 1 2 3 7 854 SD2 Prepare for my performance 4 4 5 6 8 620
SD3 Check my performance while writing
SD4 Control and modify my writing 6 15 16 17
SD5 Evaluate my performance after writing
SD6 Increase spoken English use 3 30 31 32 775
SD7 Increase written English use
SD8 Increase the use of English as a language of thought
SD9 Reflect on my learning process 5 45 46 47
SD10 Reflect on what I learnt from the task
Utilizing the item deletion method to remove irrelevant items significantly enhances the reliability of each scale This process involves eliminating inappropriate items until the scale achieves its maximum possible Alpha coefficient It is essential to retain at least three items in each scale to ensure the construct's completeness The details of the deleted items and the highest possible Alpha for each scale are presented in Table 4.3.
Table 4.3 Summary of items deleted and highest possible Alpha for each scale
Learner Autonomy No of items
Items remained Dimensions Sub-dimensions
SD2 Prepare for my performance
SD3 Check my performance while writing
SD4 Control and modify my writing
18 19 20 SD5 Evaluate my performance after writing
SD6 Increase spoken English use
SD7 Increase written English use
36 37 38 SD8 Increase the use of English as a language of thought
SD9 Reflect on my learning process
48 49 SD10 Reflect on what I learnt from the task
Bynner and Stribley (1979) assert that the Alpha of a scale must exceed 67 to ensure internal consistency The analysis reveals that the highest Alpha values, determined through the item deleted method, ranged from 728 (SD2) to 931 (SD10), indicating that all ten scales are internally consistent Therefore, the data gathered from these scales are deemed sufficiently reliable for further analysis The next step is to evaluate how these ten sub-dimensions contribute to the promotion of three dimensions of learner autonomy.
WP, independent samples t-tests were performed on the data obtained from these scales The description of its procedure is presented in the later part, Section 4.1.4
4.1.3 Inter-rater reliability of EG and CG students’ scores rated by two scorers in writing pre-test and post-test
This section presents the statistical reliability test of students' writing results from their pre-test and post-test, along with the methodology used to assess the impact of Writing Practice (WP) on their writing achievement As detailed in Chapter 3, section 3.8.2, the students' performance in both tests was evaluated by two scorers, making it essential to analyze the level of agreement between them to ensure the reliability of the scoring To evaluate the consistency of the scorers' assessments for both the control and experimental groups, inter-rater reliability analyses were conducted using the Kappa statistic The results of this Kappa statistical analysis are displayed in Table 4.4.
Table 4.4 Inter-rater reliability of EG and CG students’ scores rated by two scorers in writing pre-test and post-test
Approx Sig EG’s writing pre-test
N of Valid Cases 30 a Not assuming the null hypothesis b Using the asymptotic standard error assuming the null hypothesis
The inter-rater reliability of scores for both experimental group (EG) and control group (CG) students in writing pre-tests and post-tests demonstrated strong agreement, with Kappa values of 0.79 for EG pre-test, 0.77 for CG pre-test, 0.84 for EG post-test, and 0.77 for CG post-test, all significant at p < 0.05 According to Landis and Koch (1977), a Kappa value above 0.7 is considered favorable, indicating reliable scoring among raters.
65 by most statisticians as a good level of agreement Therefore, EG and CG students’ scores in these two writing tests were considered reliable enough for further analyses
This study aimed to assess the effect of WP on students' writing skills by analyzing and comparing the scores of Experimental Group (EG) and Control Group (CG) students An independent samples t-test was conducted on the pre-test scores of both groups to identify any significant differences in their writing abilities prior to the intervention, revealing no statistically significant difference A paired samples t-test was then applied to the EG students' pre-test and post-test scores to determine if there was a significant improvement in their writing skills Subsequently, another independent samples t-test was performed to compare the post-test scores of EG and CG students, aiming to establish whether the EG students outperformed their counterparts Detailed procedures are provided in section 4.1.5.
4.1.4 Research question 1: The extent to which WP promoted learner autonomy dimensions
This research defines learner autonomy through three key dimensions: Learner Involvement (D1), Appropriate Target Language Use (D2), and Learner Reflection (D3) To assess the primary research objective regarding the extent to which WP fosters learner autonomy, it is essential to evaluate its impact.
WP on each of these three dimensions of learner autonomy was essential
The extent to which WP enhanced learner involvement
To assess the impact of WP on learner involvement dimensions, the engagement levels of both experimental group (EG) and control group (CG) students were evaluated The Learner Involvement dimension (D1) comprises the first five scales out of a total of ten Independent samples t-tests were conducted to compare the mean scores of these scales: Inventory the assignment (SD1), Prepare for my performance (SD2), Check my performance while writing (SD3), Control and modify my performance (SD4), and Evaluate my performance after writing (SD5) between the EG and CG students.
A statistically significant difference was observed between the Experimental Group (EG) and Control Group (CG) students across four sub-dimensions of Learner Involvement: Inventory the assignment (SD1), Prepare for my performance (SD2), Check my performance while writing (SD3), and Evaluate my performance after writing (SD5), with a t-value of t(58) = 12.04.
= 11.40, t(58) = 15.49, t(58) = 9.17, and p < 0.05, respectively) Nevertheless, no statistical difference was identified in EG and CG students’ perception towards Control and modify my performance sub-dimension (SD4) (t(58) = 23.02, and p = 0.07 > 0.05)
Table 4.5 Comparison of EG and CG students’ involvement in learning process
Levene’s Test for Equality of Variances t-test for Equality of Means
95% Confidence Interval of the Difference Lower Upper
Furthermore, differences in the mean scores between the two groups were discovered
Discussion of results
This section discusses the study's findings, emphasizing how learner autonomy dimensions—such as Learner Involvement, Appropriate Target Language Use, and Learner Reflection—were enhanced by WP It also leads into a subsequent section that examines the influence of WP on students' writing achievement.
4.2.1 The extent to which learner autonomy dimensions were promoted by WP
The research question explores how well WP promotes learner autonomy, as evidenced by student engagement, their use of English, and reflective practices Statistical findings indicate a significant relationship among these three factors.
77 abovementioned learners autonomy dimensions were enhanced in the course with WP implementation This has been confirmed in an earlier study by Martines and Rubio
In 2009, research indicated that utilizing portfolios in evaluations effectively fosters autonomous learning habits However, while the use of portfolios enhanced specific sub-dimensions of learner autonomy, it did not significantly impact other sub-dimensions within the same learner autonomy framework.
In the context of Learner Involvement, WP significantly influenced students' engagement in the learning process, which is essential for developing learner autonomy The learning process consists of three stages: planning, monitoring, and evaluating During the planning stage, WP aids students in previewing assignment requirements, preparing relevant vocabulary, grammatical structures, and key ideas for their writing It also helps them set goals for their tasks In the monitoring stage, WP enhances students' ability to review their work for logical organization, content development, vocabulary, sentence variation, and mechanics However, there was no significant change in students' ability to control and modify their performance, despite a higher mean score for the experimental group compared to the control group In the evaluating stage, WP notably improved students' ability to assess their performance post-writing.
Previous research (Lam, 2011; Tran, 2013) indicates that portfolio writing assignments enhance students' ability to monitor their progress and improve self-assessment and self-evaluation skills Additionally, studies on various types of language portfolios have shown that they significantly promote students' responsibility in planning, managing, and monitoring their learning (Mansvelder-Longayrou, Beijaard, Verloop, & Vermunt, 2007; Yildirim, 2013) Similar findings are reported in studies by Gửksu & Genỗ (n.d.), further supporting these conclusions.
78 portfolios could help students to get the learning objectives, self-assess their own language skills, and participate more in the learning process
Several factors may explain these findings The specific guidelines for self-reflection and peer-reflection on writing assignments likely provided students with clear instructions, enhancing their understanding of assignment requirements and improving their writing preparation Additionally, students in the experimental group (EG) were required to document all planning and evaluation notes in their writing portfolios (WP), increasing their engagement in these critical stages The WP's design, which made all students' work visible, fostered a sense of ownership and encouraged collaboration, as students were motivated to review peers' writing and receive feedback on their own This collaborative environment was further reinforced by the mandatory self-reflection and peer-reflection scheme, which significantly increased student engagement in their learning process.
The findings from the monitoring stage contrast with those of Phung (2016) and Hussein & Youself (2016) Phung's research indicated that speaking e-portfolios (SEP) significantly enhanced students' ability to manage and adjust their speech; however, it did not improve their capacity to self-monitor during delivery This discrepancy may arise from the fact that students have more time to review their written work Additionally, Hussein & Youself's study also provided relevant insights.
While writing portfolios (WP) aided students in monitoring their performance in technical report writing, challenges remained in overcoming writing difficulties Despite students acknowledging improved recognition of their mistakes post-course, they struggled to enhance their writing skills, particularly in organizing ideas and developing content This difficulty may stem from the limited course duration and the complexity of the cognitive strategies required for effective writing Consequently, low intermediate-level teenage students need more time and practice to refine these essential skills.
Regarding the second dimension – Appropriate Target Language Use, as reported above, the research findings indicated that WP had a significant effect on maximizing
79 students’ use of written English, which could derive from the constant requirement for using written English to complete portfolio assignments During the course, students in
EG were assigned mandatory writing tasks, which included editing initial drafts, creating planning notes, and engaging in self-reflection, peer-reflection, and class conferences This aligns with findings from previous studies (İlke & Selen, 2010; Muslimi, 2015; Hussein & Youself, 2016) that demonstrate the effectiveness of such activities in enhancing students' written English proficiency and overall writing skills.
Despite the implementation of this method, there was no notable improvement in students' spoken English skills or their ability to think in English In other words, students in the experimental group did not show significant advancements in using spoken English or engaging in internal dialogue in English.
Students often regard speaking as a secondary task in completing writing assignments, leading to a lack of attention to changes in their spoken English As a result, the impact of spoken English on writing practice appears minimal This notion is supported by Phung’s research (2016), which indicates a limited effect of increased spoken and written English use Similarly, İlke & Selen (2010) found that their Learning Portfolio (LP) project, which incorporated both process writing and oral presentations, did not significantly improve students' oral English usage or speaking skills.
The findings align with Phung's research, which highlights the enhancement of students' ability to think in English through the use of SEP (2016) This study emphasizes that students often rely on their mother tongue for cognitive processes.
Many language learners find it unnatural to formulate inner thoughts in a foreign language due to their inborn cognitive habits To help students think in English, a combination of cognitive training and English proficiency development is necessary However, achieving both simultaneously within the limited timeframe of this study poses significant challenges.
Regarding the third dimension – Learner Reflection, as aforementioned, the analysis reported that using WP applied in EG had a significant effect on fostering students’
Students who engaged in writing processes (WP) demonstrated a greater ability to reflect on their learning compared to those who practiced writing without WP implementation This aligns with findings from previous studies (Little, 2004; Jesus, 2009; Lam, 2011; Yildirim, 2013), suggesting that the cycle of assignment submission, peer reflection, self-reflection, and conference reflection enhances learning outcomes Reflective notes for each WP assignment help students evaluate their work, identifying strengths and areas for improvement Additionally, reflection guidelines support students in analyzing the process, content, and results of their learning Jesus (2009) noted that European Language Portfolios (ELP) significantly increase student awareness of the language learning process and foster learner autonomy.
The findings indicate that writing prompts (WP) significantly enhance various dimensions of learner autonomy Specifically, WP aids students in the planning, monitoring, and evaluating stages of writing, although it does not improve their ability to control and modify their writing in real-time Additionally, while WP promotes the use of written English, it does not enhance spoken English skills or encourage thinking in English rather than the native language Furthermore, WP positively impacts students' capacity for reflection on both the learning content and process.