Self-Evaluation of Thai Adult Learners in English Writing Practice

This research primarily aims to investigate how the Thai adult learners self-evaluate their English writing performance in relation to the teacher’s evaluation and what the students and the teacher experience from the evaluation practice. The participants were 32 Thai graduate students and a Thai English language teacher of an EFL (English as a foreign language) writing course. The data were collected through student’s selfevaluation and teacher’s evaluation forms, writing tasks, and individual interviews. The qualitative data were considered together with the quantitative data in order to gain the whole picture of the issue under investigation. The findings revealed that the adult learners (or students) were able to self-evaluate their writing performance. The goals set out in the course syllabus, concerning the students’ participation in writing process and evaluating their writing, were practical and the students were well aware of what their performance levels were. Having access to students’ self-evaluations also gave the teacher more useful information for judging their learning achievement. This research highlights the importance of students’ self-evaluation as a real and valid source of information for developing teaching EFL writing, and promoting the students to be autonomous and lifelong learners. Implications are drawn regarding the independent learning goals and recommendations for future research.

This research was initiated by the author's concern about the reform movement in English language teaching (ELT) in the Thai university context and the interest in the Thai adult learners' self-evaluation of their writing practice.Curriculum and syllabus goals in the Thai educational system encourage students to work more independently and take responsibility for their own learning, particularly the students in higher education (OECD/ UNESCO, 2016).A review of the literature on evaluation (e.g., Nicol & Macfarlane-Dick 2006, Rust et al. 2003, Sambell et al. 2013) also raises the researcher's awareness of the importance of understanding the students' own evaluation of their EFL writing performance as well as for the elaboration of evaluation procedures.Nevertheless, in Thailand, both students and teachers seem to have little previous experience of evaluation procedures in the language classroom as the evaluation has traditionally been the teachers' exclusive privilege and obligation.Besides, there has been no research carried out on the conditions that govern adult learners' participation in evaluation.Much of the previous research on formative evaluation and summative evaluation in language learning has focused on young learners learning a foreign language (Brown & Harris 2013, Lee 2017).
According to Brown and Harris (2016), self-evaluation practices need to be explored in order to support the implementation of adult learners' lifelong learning skill which is important for their language learning process.The focus on writing in the teaching of English as a foreign language was chosen because English is the language most Thai students learn, and writing has become more important in foreign language teaching at higher education level.Accordingly, as the role of writing in English as foreign language learning increases, the students' ability to self-evaluate their writing skills is also important because selfevaluation could help reflect individual students' understanding and possibility of improvement.This study will provide valuable insights into Thai adult learners' selfevaluation in their English writing practice with regard to various aspects.As such, the research findings could, in turn, help improve the evaluation practice and bring about benefits to English language teaching and learning in this context.Moreover, it is expected that other useful implications can be formulated and would be applicable for the teaching and learning of other courses in university settings in Thailand.
This research, therefore, aims to contribute to an understanding of whether the use of self-evaluation in the writing course of English as a foreign language in university can help develop lifelong language learning skills and further development of broader and fairer evaluation practices.The study centres on how the adult learners (or "students" hereafter in this study) self-evaluate their writing ability in general and their writing performance task based on the grading criteria of English writing compared with the evaluation of their teacher.The ultimate aim is to find out what the students and the teacher have experienced from the self-evaluation practice in the writing class.The results could be useful for further developing the students' English writing ability.
Three main research questions are developed based on the objectives of the study: 1) How do the students self-evaluate their English writing performance compared with the evaluation of their teacher?2) Based on the evaluation criteria, how do the students' self-evaluate their English writing performance compared with the evaluation of their teacher?3) What do the students and the teacher experience from the use of students' selfevaluation of English writing performance to further develop the students' writing ability?

LITERATURE REVIEW SELF-EVALUATION OF LANGUAGE LEARNING
Language learning and evaluation are common categories of applied linguistics.Selfevaluation, according to Pennycook (2001), is alternative evaluation which is to a large degree based on the critical applied linguistics approach and can be applied to language evaluation.Based on the theory of critical language pedagogy, McGroarty (1998) problematised what could be served as standard indication of learning for teachers, parents, and policy makers when learners are to be evaluated on their learning goals and activities.In response to this, many scholars in education evaluation (e.g., Deeley 2015, Lynch 2001, Scott et al. 2016) gave a view that an additional approach to exploring individual language ability based on the critical perspective can be implemented as an important element for language research in evaluation.Through evaluation, Taras (2002) remarked that self-evaluation helps learners to be empowered rather than suppressed; it encourages learners to think and learn more.However, the complication of the nature of self-evaluation requires learners to describe in the evaluation of their own performance.It is believed that this can occur in and through a discourse related to both summative and formative evaluation and create knowledge of individual learners.Lifelong self-evaluation practices should be seen as a means to learning outcomes and learners need a teacher's support to develop their self-regulating techniques.As part of a broader discourse, self-evaluation represents a shift in evaluation practice which helps learners learn to learn.

EVALUATION OF EFL WRITING
The evaluation of EFL writing is integral to the process of effective teaching and learning of writing (Jones 2002, Regala-Flores & Yin 2015).When conducted in a dynamic manner, writing evaluation can support teaching, both theoretically and virtually.In this way, the evaluation alternatives that are best suited to the development goals help students to identify where they are, in terms of their writing development (Herrington & Curtis 2003).This, in turn, enables greater understanding of what the students need to support their learning.
The use of self-evaluation of writing is very important as it lets students practice their skills on the underlying process.This is the way to bridge the students' path to independent learning.Taras (2002) affirmed that evaluation and feedback can be formative when they are initially separated from grading.Students should be allowed to develop their own judgment before receiving feedback from other evaluators.In addition, formative evaluation and summative evaluation may not be as different as commonly proposed.Rather, a summative test can be used as a positive part of the learning process.However, there might be some problems that students are not well aware of embedded in different evaluation procedures (Rea-Dickins & Germaine 2013).
There has been little research on students' self-evaluation of task-based performance.In a small-scaled study of students on an independent learning program, Brown and Harris (2014) used annotated texts produced by EFL students to reflect oriented criteria, such as content and sociolinguistic appropriateness, text structure, organisation and coherence for students evaluating their own performance by comparing the annotated texts with their own.The study reveals that this method is reliable and useful for the students to self-evaluate their writing and to learn specific language skills.In doing so, the students gained perspective on their own specific tasks in relation to the other students' writing tasks, perceived the learning of different language skills, and were able to identify and correct their own language mistakes.More importantly, from the text annotating practice, the students became more aware of certain areas which could help them to see their writing problems.Some students were frustrated because they could not self-correct their language.To this point, Brown and Harris (ibid.)confirmed that the students can develop growing awareness of how to evaluate their writing and what to consider when writing.Therefore, not only can self-evaluation be a tool for evaluation, but also a learning tool for the students.

STUDENTS' SELF-EVALUATION AND TEACHER'S EVALUATION
Students' self-evaluation is considered to be able to play a key role in formative evaluation.It is believed to have the potential to promote learning, raise learners' awareness, underpin learners' learning autonomy in a lifelong perspective, and to be conducive to self-regulating learning processes and needs analysis.Nicol and Macfarlane-Dick (2006) emphasised that self-evaluation is an essential part of formative evaluation as it can lead to the expected goal of language learning, the evidence of learning progress, and the information for understanding the gap between students' and teacher's evaluation.Importantly, selfevaluation must be apprehended by students in order to improve their learning.Excluding students in the evaluation process may lead to a variation between test scores and actual communicative ability of the students.
The meta-analysis of Falchikov (2005) comparing students' self-evaluation and teacher's evaluation indicates that involving students in self-evaluation is beneficial for their learning improvement.Students' self-evaluation is dependent on teacher's feedback.Students can work with teacher's feedback while the summative grade is withheld.The only critical students' responses reported in Taras' study (2002), for example, were concerned with the self-evaluation of grades.Some students felt that they had neither the experience nor knowledge of the evaluation.Students' positive reactions were that feedback and selfevaluation helped them to focus more on the evaluation criteria.In a subsequent investigation, Taras (2003) indicated that teacher's feedback allowed the students a high level of independence to consider their language errors, to understand evaluation procedure including criteria and feedback, and to realise what their strengths and weaknesses were before receiving grades.From these two studies, it can be noted that self-evaluation without teacher's feedback cannot help students to be aware of all their errors.Students' selfevaluation with teacher's feedback is an efficient way of helping students to overcome unrealistic expectations and focus on their achievement rather than on the input required to produce their writing tasks.

DEVELOPING WRITING ABILITY THROUGH STUDENTS' SELF-EVALUATION AND TEACHER'S FEEDBACK
According to Taras (2002), students usually know the importance a teacher gives to a task by how much time is assigned and how much emphasis is put on it.If students are to selfevaluate from an informed position, they need to take feedback on board.Through feedback, students have an opportunity to understand what positive qualities their writing tasks have, or what needs to be worked on more.In this way, students can be developed towards autonomous learning.When feedback is given before the writing task is graded, student's reflective ability can also be developed.Students will learn how to evaluate their writing tasks realistically and how to have a sense of control of the learning situation.Upon the completion of the evaluation process, students should be able to become both responsible and self-sufficient in learning and not dependent on teacher guidance.In addition, from the study of Trumbull and Lash (2013), many teachers believed that grades, comments, and words of praise or even blame are adequate, when in actual fact students want information specifically linked to their writing performance and guidance on what they should do to improve.As stated by Ross (2006), the most effective feedback will draw the students' attention on their progress in mastering the required task.This helps not only to enhance selfefficacy and encourage effort attribution, but also to reduce the focus on comparison with peers.Hence, useful information about the task performance should be provided consistently with more detailed feedback to support the students to work through misconceptions or other weaknesses in their performance.Praise should also be used sparingly, and should be taskspecific.For Sadler (2013), the fairest step should let students revise and resubmit work for other evaluations after their self-evaluation.Taras (2003) also emphasised that students need to understand that they can develop their writing ability through effort.Mistakes are an inevitable part of learning, and they can control their own learning.To make use of the feedback obtained, students must be able to produce better-quality work.In this regard, the introduction of self-evaluation methods can potentially strengthen the link between feedback and learning (Bennett 2011).To open up feedback for improving language performance, Sadler (2013) proposed that a transition from feedback to self-monitoring needs three conditions to be satisfied.Students should realise what quality they wish for; that is, they should understand the criteria set through descriptive statements and/or exemplars.Then, originality and creativity should be developed through the understanding of the transcendence of normal boundaries.It is important that students need to be familiar with the discipline or genre to go beyond it.Students themselves should be able to choose appropriate strategies to bring their performance close to the goal or to selfevaluate their writing tasks.Besides, a teacher should help students to develop self-evaluation skills of their own work during the process of learning production (Rea-Dickins & Germaine 2013).

METHODOLOGY RESEARCH SETTING AND SAMPLING
The research setting of this study is a graduate school of a Thai public university.The sampling method used is a non-probability sample, deriving from the research targeting a specific group in the knowledge that it does not statistically represent the wider population (Cresswell 2013).Thus, the sampling method follows the purposive sampling technique.The sample consists of 32 Thai EFL students and one Thai English teacher.During the period of this study, the students were studying for a Master's degree in English and were commonly considered to be high achievers in English.They were participating in an EFL writing class -English Writing II -as a requirement of the degree programme.Table 1 below shows demographic information of the student participants.As for the teacher participant, with the education background in English language teaching, he has 18-year experience of teaching English for postgraduate students.In particular, he has expertise in the English writing process and approach.His level of English is of high proficiency based on the researcher's observation and he obtains a high level of students' evaluation on his teaching performance.

RESEARCH INSTRUMENTS
Based on the study aims, closed and open-ended questionnaires, semi-structured interviews, and English writing tasks were developed.The researcher selected methods that were logical and appropriate to the research objectives and questions.Triangulation of methods was used to increase methodological validity (O'Leary 2004).The importance of trust in data collection and analysis was also emphasised when trying to ascertain validity.Considering the main focus of the study, validating student's rating against those of teacher's rating and having these as a standard was a researcher's concern for validity.It was not necessarily the case that teacher's grading was reliable or a valid indicator of student's writing performance as several kinds of bias could be apparent when grading.If students were able to evaluate their own performance, measure more or less in accordance with the teacher's evaluation, there would be the question of what was actually being evaluated and how the criteria or the rubrics for evaluating writing performance were understood on both sides.
In order to obtain the reliability of the questionnaires, the questionnaires (the students' self-evaluation and the teacher evaluation forms) were administered with six EFL students who enrolled in other writing courses and with two teachers of those courses.Data obtained from the questionnaires were analysed to compute the Alpha coefficient to measure the reliability of the questionnaires.The questionnaires or the evaluation forms indicated a Cronbach's alpha of .97(higher than .80).
The reliability problem of the specific text analysis of the students' writing tasks had to do with the ambiguity of word meanings and variable language category definitions.What should be defined or categorised as grammar or sentence structure was to some extent not clear-cut.In this regard, the issue of reliability was concerned and the writing tasks were also considered by the researcher and another two experts in teaching EFL writing as stability could be determined when the same content was coded more than once by the same coder.The linguistic analyses gave the same results when re-assessed by the researcher and the three experts in TESOL.Hence, the interrater reliability was in all cases found to be satisfactory.
In terms of interrater reliability of the interview data, categorisation of the data was undertaken and the findings were interpreted into themes.To ensure reliability, the researcher summarised, categorised, and interpreted the interview data conducted with nine students and one teacher into themes.Then, the researcher asked the three experts to check if her judgement or measurement on students' and teacher's excerpts corroborated the researcher's interpretation.The experts were asked to check her answers on the interrater checklist.The degree of interrater reliability could be expressed either as a reliability coefficient or as a simple percentage of agreement between the two data sets.In this study, the percentage of agreement was applied.
To ensure content validity of the questionnaire, the researcher followed validation processes suggested by Fraenkel and Wallen (2000).Thus, the first version of the questionnaires was validated by the three experts to judge the congruence between the questionnaire statements and the evaluation practice constructs.The obtained data were utilised to calculate the Index of Item Objective Congruence (IOC) of each item of the questionnaire statements.Based on the experts' comments, some of the question items which could be redundant and obtained unacceptable content validity were deleted.When the confidence of content validity was .91 (higher than .80),all the statement and question items were acceptable (Gorard & Taylor 2004) and were used in the pilot study and the main study.
To ensure the trustworthiness of the study, the validity was checked by interpretive validity.This strategy was to obtain the participants' feedback or member checking, which involved discussing the findings with the participants (Dörnyei 2007).Besides, the reliability was obtained by interrater reliability.O' Leary (2004) contended that the researcher needed to be concerned with the reliability from two independent observers or raters.The interpretive validity was performed when the qualitative findings were analysed and interpreted.Discussions of the findings were held between the researcher and five students and the only one teacher who participated in the interviews.It was found that all the student and teacher participants agreed with the data interpretation carried out by the researcher and confirmed that the findings and interpretations were their real opinions.
Having reviewed previous literature, research questions, and theoretical rationale, the researcher agreed that there would be two questionnaires of student's self-evaluation of their writing performance (Writing Tasks 1 and 2) and another two questionnaires of teacher's evaluation of the students' writing performance.The student/teacher questionnaires for Writing Task 1 consists of three parts: 1) student/teacher's demographic information; 2) student's self-evaluation/teacher's evaluation of writing performance; and 3) an open question asking for additional comments and suggestions on the writing task.Another questionnaire for Writing Task 2 consists of four parts.The first three parts are the same with those used for Writing Task 1.There is an additional section, part 4, which contains question items exploring student/teacher satisfaction with specific writing skills.
Both students' and teacher's evaluation of writing performance are the 4-point rating scale which was developed based on the rubric for evaluating writing performance adapted from Rubrics for Assessing Academic Language and Writing of RMIT (Study and Learning Centre 2015) and the student's expected writing performance stated in the English Writing II course aim.The rubric focuses on five important dimensions of writing skills: 1) writing purpose and relevant content; 2) organisation of ideas and paragraphing; 3) research for writing (supporting evidence, paraphrasing, and referencing); 4) linguistic accuracy (sentence, grammar, vocabulary, spelling, and punctuation; and 5) clarity and style.
Two different writing tasks (Task 1 and Task 2), naturally expository and argumentative, were given to the students in order that the teacher was able to see how well each student could fulfill the course aim and how each student could improve his/her writing performance.The topics of the two writing tasks were created by reflecting the specific course goals for the writing course at graduate level.
The questionnaire data were triangulated with the semi-structured interview data to help convince that the study is valid and reliable.To understand how the students and the teacher experienced working with self-evaluation practice in the EFL writing class, nine students and one teacher were interviewed individually after the students' two writing tasks were completed and the teacher already gave his feedback to each student.

DATA COLLECTION PROCEDURE
Writing Task 1 was assigned in the 4th week of the course.The students had one hour for completing the writing task and another fifteen minutes for giving responses to the selfevaluation form.A week later, the writing tasks together with the teacher's feedback on the teacher's evaluation form were given back to the students and each student was to discover and correct the error him/herself.Subsequently, the students revised their Writing Task 1 and returned the task to the teacher.In the following week, the teacher returned the writing task with scores and feedback based on the writing criteria behind the scores given.In this way, the process was recursive and generative, with the students considering their task, evaluating it, reacting to the teacher's feedback and then moving on for further practice.Figure 1 below shows the data collection procedure.
The teacher assigns Writing Task 1 in the 4th week of the writing course.
Each student completes Writing Task 1.

The teacher evaluates
Writing Task 1 with scores and feedback.
Each student self-evaluates Writing Task 1.
Teacher's feedback and student's self-evaluation are given back to each student.
Each student considers, evaluates, and reacts to the teacher's feedback and their self-evaluation.
The teacher runs writing class regularly and assign Writing Task 2 in the 7th week.
Each student completes Writing Task 2.
The teacher evaluates Writing Task 2 with scores and feedback.
Each student self-evaluates Writing Task 2.
Teacher's feedback and student's self-evaluation are given back to each student.
Each student considers, evaluate, and reacts to the teacher's feedback and their self-evaluation.
The researcher considers and compares the teacher's and student's evaluation on the writing tasks.
The researcher interviews nine selected students and the teacher by focusing on the evaluation practice, the feedback, and the writing performance.In the 7th week of the English Writing II course, the teacher reminded the students about the course aim and informed them the evaluation criteria of EFL writing before assigning the students Writing Task 2. Once completing the writing task, the students were asked to self-evaluate their performance and to indicate how satisfied they were with their specific writing skills.Then, the teacher evaluated each student's second writing task by completing the teacher's evaluation form of student's writing performance.After that, the data obtained from both students and teacher evaluating the two writing tasks were compared.
Nine students were selected for individual interviews to further explore what the students experienced from the self-evaluation practice of their EFL writing performance.The students were interviewed with the benefit of hindsight after they had completed and selfevaluated their two writing tasks.Later, the teacher was interviewed to elicit his views on experience of using the students' self-evaluation practice in the EFL writing course.

DATA ANALYSES
The quantitative data relating to demographic information agreement with the statements concerning the student's self-rating and the teacher's rating were coded numerically and transferred to the mainframe computer for processing and tabulation.Descriptive statistics of individual responses to the evaluation form were determined.The total frequency and percentages of each item were calculated and the mean scores and standard deviations of the variables were compared.Students' and teacher's responses to the satisfaction with the writing skills in the last section of the evaluation form for Writing Task 2 were calculated into percentages.The t-test was used to compare the values of the means from the student's self-evaluation and the teacher's evaluation.Students' responses to the open questions about their views on writing performance of the two tasks were simply prescribed into categories and generated a frequency tally of the range of feedback as a preliminary to coding classification.
Qualitative data from the interviews and the open-ended question in the questionnaires were content analysed by using the interpretative analysis method.

RESEARCH FINDINGS EVALUATION OF STUDENTS' WRITING PERFORMANCE
After completing Writing Task 1, in response to the first research question, all the students (N = 32) were liable to positively evaluate their writing performance as "good" and were fairly confident in their writing ability (M = 2.45; SD = .60).Noticeably, as shown in Table 2, the teacher rated the students' writing performance as "satisfactory" (M = 2.14; SD = .58)and most of the students overrated their performance task in relation to the evaluation of the teacher.In response to the second research question, most of the students (25 out of 32) evaluated their Writing Task 2 as "good" performance (M = 3.06).However, as shown in Table 3, all the students tended to evaluate themselves fairly higher than what the teacher did the second time.As shown in Table 4, a t-test of the difference between the mean scores of the students' self-evaluation and the teacher's evaluation indicated that the obtained difference was not significant (t = 1.512; p = .102).On average, students' self-evaluation of their two writing tasks corresponded fairly well with the evaluation of the teacher; the mean scores of the students' self-evaluation were 2.75 (SD = .72)and the mean scores of the teacher's evaluation for both tasks were 2.70 (SD = .87).The students' self-evaluations of their writing tasks were fairly accurate, with a slight overestimation on the first and the second tasks.The latter was given after the teacher's feedback, the self-evaluation practice along with other writing assignments in class during the four-week writing skill practice, and the revision of the writing Task 1 and other writing assignments.All of these may, therefore, indicate a positive influence on the writing Task 2. The tendency to overrate on the part of some students could be an indication of a lack of deeper understanding of the evaluation criteria and what the criteria actually stand for.
When evaluating their writing Task 2, the students also indicated their satisfaction with specific writing skills.In Table 5, three specific writing skill categories ranked as the most satisfactory were paragraphing (73%), punctuation (66%) and spelling (58%).Grammar was the skill that was indicated the least "satisfied" by only 37% of the students.Considering the students' rating of their writing performance, it was remarkable that most of the students evaluated their linguistic accuracy as "satisfactory" (M = 2.16; SD = .52)which was the writing aspects that the students felt least content with.As shown in Figure 2, the students underestimated their specific writing skills in comparison with the teacher's satisfaction.The largest difference between the teacher and the students' satisfaction with the specific skills appeared in grammar and the smallest in sentence structure.In response to the third research question, the interview questions focused on the students' and the teacher's views and experiences of the use of students' self-evaluation of English writing performance to further develop the students' writing ability.It was apparent that the students considered self-evaluation could help facilitate their English learning in general.
Self-evaluation was one of the most important things they could learn.As one student (S5) explicated: "If I could evaluate myself in a correct way, I could see what I'm learning and could improve better.Self-evaluation could be considered as an important skill for me when learning language skills as it is good for helping me to be self-critical." An individual interview with the teacher elicited his general view on experience of teaching English writing as a foreign language, his comments and reflections on the students' self-evaluation practice.Basically, the teacher was satisfied with the students' self-evaluation and the way in which he taught English writing for the adult learners.The students were well motivated because they realised the importance of English for their graduate study and future careers.As he explained: "Luckily, I've got motivated students.When they are allowed to be involved in the evaluation practice, they will be more enthusiastic to learn.However, they might not be able to focus on everything that should be taken into consideration." Interestingly, almost all of the students commented that student involvement in evaluation of their own skills should start earlier and this should be implemented as a continuous practice.Six students also postulated that a teacher should train students how to self-evaluate their English ability in general and their writing performance in particular.One of the students (S6) emphasised that the accuracy of self-evaluation depended on the degree of self-criticality and self-confidence each student had.As she elaborated: Through the self-evaluation practice, their teacher let them think for themselves more than the other teachers did before.All the students admitted that they were generally positive towards the self-evaluation practice and felt that it was an important skill to develop early.Self-evaluation helped enhance their responsibility and critical skills.This practice was useful because it made them more aware of their own strengths and weaknesses in writing, as well as overall difficulties in learning English as a foreign language.The students admitted that they could give details and could develop their thoughts better when writing through the self-evaluation practice.The self-evaluation could help them learn how to work differently, particularly after knowing the evaluation criteria.As one student (S7) who really appreciated the self-evaluation practice expressed: From the individual interview with the teacher, his general view on experience of teaching English writing as a foreign language, his comments and reflections on the students' self-evaluation practice showed that he was satisfied with the students' self-evaluation and the way in which he taught English writing for the adult learners.Apparently, the students were well motivated because they realised the importance of English for their graduate study and future careers.As he elaborated: "Luckily, I've got motivated students.When they are allowed to be involved in the evaluation practice, they will be more enthusiastic to learn.However, they might not be able to focus on everything that should be taken into consideration." To this point, the students commented that they would probably either overestimate or underestimate their writing performance if they were not trained how to self-evaluate their writing skills.Inevitably, they may give themselves higher scores than they deserved.In order to self-evaluate writing skills accurately, almost all of the students agreed that they needed to know much more about the English writing skills and better understand the evaluation criteria.They still needed to have a teacher or an objective reader to give them contributing feedback.
Considering the benefits of self-evaluation, all the students experienced the practice as a matter of evaluating their own performance in relation to what they wanted to communicate in writing and how they could improve.This practice helped them to see and learn from their mistakes so that few mistakes could be made next time.Most of them agreed that self-evaluation of the writing task was enjoyable as it gave them feelings of independence and of being in control.It forced them to think and learn on their own.If they continued working in this manner, self-evaluation would become so automatised that they would be able to revise their work in accordance with the evaluation criteria.As one of them (S7) confirmed, "Surely, if I were able to evaluate what I've done, I would see how my writing became much better." From the teacher's view, it was important that the adult learners, unlike younger learners, showed good responsibility for their study.They were able to participate in the writing class, plan their writing tasks, ask for clarifications, look up words or grammar details by themselves, and hand in the tasks on time.Discussing the students' self-evaluation, the teacher described that it was when students evaluated their level of knowledge in relation to their learning and the teacher's teaching goals.This practice reflected the way in which the students were able to think independently about what and how they learned to become better learners.The teacher also affirmed that most of the students were able to evaluate their EFL writing performance reasonably in the approved manner.They had a good sense of their own proficiency in learning English writing as a foreign language and had an objective picture of their own writing performance.They knew the areas where they needed improvement.
In addition, the teacher noticed that the value of self-evaluation was inherent in the reflections it initiated among the students in their learning process.This was an important step in the students' development towards becoming autonomous learners.As he explained: "I think that many students know what they have to do when writing.They have good background knowledge of the writing skills which could be helpful for self-evaluating their writing performance." On the other hand, some students were unable to evaluate themselves impartially.For this reason, the teacher commented that it was necessary to work with self-evaluation over a period of time and there were plenty opportunities for the students to perform self-evaluation properly.Nevertheless, the teacher was satisfied with the way in which the students selfevaluated their writing performance.Students' self-evaluation made it easier for them to reach estimated goals.They told the teacher that through self-evaluation, they had learned to reflect on their own learning English as a foreign language and English writing skills.This led them to become more strategic and more successful in their English learning.The students also appreciated working with the criteria of evaluating writing, where they could make objective evaluations and then relate them to their own writing tasks.
Considering the impact that the students' self-evaluation had on his own teaching and other valuation practices, the teacher revealed that he tried to help students think and as a consequence to learn more independently.As he emphasized: "I hope that the students could think more on the evaluation in relation to the syllabus in order to express themselves about what they think they have done well and what they need to improve." Overall, the students had several suggestions for implementing a more formative type of evaluation approach in learning English as a foreign language, including the students' selfevaluation used in this study.More emphasis on working with criteria, time for relevant feedback and revision, special time for teacher-student discussion, peer evaluation, and selfevaluation were also suggested as possible ways to involve students more in their own learning evaluation.For the teacher, the influence of self-evaluation that he and his students had experienced caused him to reconsider his feedback on each student's writing task.Standardised writing exercises, an evaluation guide and a writing guide along with the selfevaluation practice in several steps of his teaching should be used continually.In doing so, the students could have more focus and opportunity to reflect on and be responsible for achieving their goals.

DISCUSSION
The results of the study show that students' ability in evaluating their own writing ability, using a teacher's grades as a criterion, is reasonably accurate.The students were able to evaluate their language proficiency level, which in turn meant that they were in a position to take responsibility for the planning of what they needed to learn and for the evaluation of their writing tasks.This interpretation is in line with previous research carried out by Lew et al. (2010) and Polio and Shea (2014) reporting that students were fairly good at evaluating their learning English as a foreign language.
Another explanation for the different outcomes in correlations is the possibility that the teacher's evaluation focused more on the students' formal language skills than the evaluation criteria.Reasons why the students and the teacher made different evaluations of the students' writing performance may be that the students had unrealistic views of their own proficiency and real indications of ability they received outside school which were not emphasised in the classroom.This is in agreement with the study of Maclellen (2001) concluding that the writing tasks may correspond to course requirements, but the interpretation of how these were transferred to real-life expectations and experiences was difficult to make.Thus, the closer relationship between the students' and the teacher's evaluation on the writing Task 2 could be an indication that the aims of the second writing task to apprehend broader writing performance, as described in the course outline, were easier for both students and a teacher to understand and relate their evaluation to the writing task.For the writing Task 1, it was dependent on more certain circumstances and instructions and was related to specific task expectations that students may not be aware of.Students' ability to self-evaluate their writing performance, therefore, seemed to be dependent on the task type and the situation at hand.As Andrade and Du (2007) and Mistar (2011) suggested, it is necessary that students understand both the evaluation criteria and the reasons behind the self-evaluation practice.
Besides, the issue of overrating and underrating of language skills has been the attention of many researchers in the area of language self-evaluation (e.g., Rea-Dickins & Germaine 2013, Shermis & Burstein 2013, Oscarson 2014).Students who overrated their language proficiency may believe that they were in control of things they really did not grasp, and thus did not take in skills that they in reality needed to learn.Students who underrated their competence may possibly apply themselves to work on areas they actually already mastered and failed to challenge themselves.The making of reliable and realistic selfevaluations is, for this reason, an important part of the students being able to focus correctly, and learn efficiently, by not spending either too much or too little time on certain language issues.In case of lifelong language learning, without any supports of a teacher, it is recommended that this skill be developed by language learners themselves.
Comparing with the interview data, it is noticeable that the more mistakes of language skills the students considered they had made, the more often they noted that they could develop.Corresponding to the research of Borg (2003), the fact that the students and the teacher in this study tended to focus more on grammar and sentence structures could be reflected on what they perceived as essential for English language learning.However, this appears contrastive to the view of the current educational policy which concerns more on communicative language teaching.In relation to this, one may wonder why these adult learners, who should have developed their language skills further, did not focus on the communication skills that were also relevant and essential for their language development in general.Hence, it is crucial that the use of language for communication, made either through written of spoken language, should be specific and grammatically correct so as not to cause errors and misunderstandings.
Regarding the results of the students' self-evaluation of their particular writing skills, their different satisfaction with the skills could reflect on different underlying attitudes on writing skills.To "be satisfied", therefore, did not indicate that the students believed that what they wrote was correct.Certainly, according to Andrade and Boulay (2003), it could be an expression used when the students needed to self-defense their English skill performance.When it came to language production, the students may not be able to improve even if they were in themselves correct or satisfactory.
By and large, the students became better at understanding what writing skills they needed to improve through reflecting on their writing tasks and course expectations.Expressing their experience of self-evaluating EFL writing as a process of directing their learning, many students were responsive to the course goals that their EFL writing practice should be in congruent with the concept of lifelong learning.This is in line with what Kaplan et al. (2013) emphasised that self-evaluation made students think and consequently learn more.In this regard, as supported by many researchers in the area of evaluation of English language teaching and learning (e.g., Rea-Dickins & Germaine 2013, Richards & Rodgers 2014, Sambell et al. 2013), the EFL students in the context of this study need to be trained how to self-evaluate themselves in various EFL writing situations for better awareness of language proficiency.
The evaluation method used in the writing task was generally considered to be a good way to create students' awareness of their own EFL writing competence.One aspect that the students especially responded to the grading or evaluation criteria and the level of language expected at the end of the course included the insight that results attained were neither about the self nor about the amount of effort put into a task.All of these are aspects that Boud (2000) also found the most important for effective language learning.Hence, the discourse shown in the course syllabus could be easily understood in such a way that the goals in EFL are inherently understood by the students themselves, rather than something which the school or even the teacher has responsibility to inform their students what they need to do.Working with and discussing the grading or evaluation criteria with peers and the teacher in order to grasp their meaning and what was expected of them should be of more concern.In so doing, the students find it possible to have a realistic understanding of how well they could achieve the goal.
Last but not least, the application of self-evaluation in the context of this study supports a reflective attitude, allied to enthusiasm and willingness to learn, and in this way, developing an intellectual responsibility to the self (Bloxham & Boyd 2007, Brown & Harris 2016).In this regard, self-evaluation in EFL writing can be a way to reach self-regulation and strengthen lifelong language learning attitudes if it becomes part of everyday classroom practice.The chances are that it can be a means to promote more comprehensive and fairer evaluation if the students are well trained and encouraged to self-evaluate their writing tasks continuously as a formative evaluation.This can also be carried out in combination with a summative evaluation (Taras 2010).

CONCLUSION AND IMPLICATIONS
By and large, the students or adult learners in this study were able to self-evaluate their EFL writing.The goals specified in the course syllabus, pertaining to students' involvement in developing and evaluating their EFL writing, were practical and the students were well aware of what their performance levels were.
The teacher needed students' supports in conducting evaluations.Having access to students' self-evaluations gave the teacher more useful information for judging their learning achievement.The students' self-evaluation could be considered as authentic and appropriate information for further development of EFL writing practice.A shared understanding of the implications of evaluations and feedback on the writing tasks should be built up in the classroom as well as in discussion between the teacher and individual students.
The students' evaluation of their language skills, such as grammar and sentence structure, could be the reflection of the ways in which the teaching and learning situations were emphasised.It was important that the students should comprehend what the practical use and the purpose of each language skill.Once the students were taught to self-evaluate their writing, and to use grading criteria in comprehensive ways, they would be able to improve their writing performance and have deeper awareness of their achievement levels.As a result, they were well equipped for other steps of EFL learning.
Both the students and the teacher shared positive experiences of integrating independent learning goals in the form of students' self-evaluation.It was apparent that the self-evaluation of EFL writing, in this context, could help the students to self-evaluate their writing tasks and to motivate them to improve their EFL writing skills.As well, the selfevaluation encouraged the students to replicate and to learn more through reflections as it made discernable much of what was unnoticed in the learning process of EFL writing.
The implications of this study for teaching and learning EFL writing in graduate level contexts represent the importance of introducing self-evaluation practices in relation to students' proficiency of English and EFL writing.Grades of individual tasks throughout the term can be used for formative purposes, and summative evaluations should be considered according to the course goals.Having grades on different writing tasks collected all the way through a writing course and evaluated at the end can be a constraining factor to learning.
Future research studies in the area of EFL learning could be carried out by focusing on different evaluation results of various forms of writing tasks to see if the results of this current study can be replicated, or if any recurring patterns can be emerged.The adult learners' motivations of their self-evaluations of their writing tasks and to what they might have attributed their self-evaluated grades in the present study may be analysed further.Besides, positive aspects of specific writing skills of adult learners or other groups of students could be of interest to other researchers in the field.

FIGURE 1 .
FIGURE 1. Flow Chart of the Data Collection Procedure

FIGURE 2 .
FIGURE 2. Students' Satisfaction with Specific Writing Skills in Comparison with Teacher's Satisfaction STUDENTS' AND TEACHER'S VIEWS ON EXPERIENCE OF USING STUDENTS' SELF-EVALUATION TO FURTHER DEVELOP THE STUDENTS' WRITING ABILITY

"
I think it is so good.Asking myself what I am satisfied with and what I should improve help me realise what I should be concerned when learning and trying to improve my writing skills."

TABLE 1 .
Demographic Information of the Student Participants

TABLE 2 .
Students' Self-evaluation of their Writing Performance (Task 1) and Teacher's Evaluation

TABLE 3 .
Students' Self-evaluation of their Writing Performance (Task 2) and Teacher's Evaluation

TABLE 4 .
Difference between Students' Self-evaluation and Teacher's Evaluation of Writing Performance

TABLE 5 .
Students' Satisfaction with their Writing Skills If it turns out that the teacher gives me better feedback on my writing task, I would feel accomplished and work harder to improve my writing skills.On the other hand, I would become less motivated to improve my writing if the teacher evaluates me with lower ratings."