The Effect of Dialogue Journal Writing on EFL Learners ' Descriptive Writing Performance : A Quantitative Study

This study sought to evaluate the effect of dialogue journal writing on writing performance as well as its different subcomponents, namely content, organization, vocabulary, language use, and mechanics (Following Polio, 2013). Participants were 84 EFL intermediate learners who were selected based on their performance on Oxford Quick Placement Test (2004) and divided randomly into experimental and control groups. While the participants in the control group took part in descriptive writing pre and post-tests only, their counterparts in experimental group were asked to write 3 journals a week for about 6 months in the period between the preand post-tests. The instructor of the experimental group provided feedback to each journal entry mostly on its content and message to which the participants replied in a dialogic manner. Results of independent sample t-test located a significant difference between the experimental and control group regarding the overall writing performance, as well as the sub-components of content, organization, and vocabulary in the post-test. However, the obtained results did not reveal a significant effect of dialogue journal writing on language use and mechanics of writing performance. The results which promise implications for writing instructors, curriculum developers, and material designers are fully discussed.


Introduction
Dialogue journal writing is a writing activity by which language learners can make a bond of written communication with their teachers as well as practice different aspects of the target language constantly.This characteristic can make dialogue journal writing a learner-centered instrument to enhance learners' proficiency in different aspects of language.In addition, dialogue journal writing is a research instrument to find about teachers' ideas on teacher education (Baily, 1990), learners' responsibility for their learning (Porter, Goldstein, Leatherman, & Conrad, 1990), learners' strategies (Halbach, 2000), learners' thoughts and ideas toward EFL learning (Myers, 2001), and to impact learners' reflection (Hashemi & Mirzaei, 2015).Despite such various lines of research on dialogue journal writing, very few researchers have investigated its effect on the EFL overall writing performance in general and the writing components in particular.This study aims to bridge this gap via investigating the effect of writing dialogue journals, as an outside classroom activity, both on the EFL learners' overall descriptive writing performance and its constituent components.

Literature Review
2.1 The nature of writing Linguists like Saussure in the 20th century proposed that writing is just "the spoken word alone […] and not the internal system of language" (De Saussure, 1968, p. 24. quoted in Harklau, 2002).But this view on writing has changed when, from one end, researchers recognized the newly developed field of native English speaker composition, and from the other, teachers realized the needs of L2 learners to write in the academic environment (Reid, 2001).However, the writing needs are not confide to this as language learners need to write for both cognitive demands and communicative functions (Weigle, 2002).The processes of writing and language learning are closely interrelated.Raimes (1983) provides three ways through which writing makes contribution to the learning process.First, writing "reinforces the grammatical structures, idioms and vocabulary that we have been teaching our students" (Ibid,p. 3).In other words, to master writing, one needs to be proficient in vocabulary, grammar and writing mechanics (Bitchener, 2008).Second, writing enables learners to take risks and go beyond what they have learned.Third, writing makes it possible for learners to become involved with language.While writing, learners "discover a real need for finding the right word and the right sentence" (Raimes, 1983, p. 3).Defining writing in the context of language teaching and learning is not a simple task since "as researchers in both first and second language writing have pointed out, the uses to which writing is put by different people in different situations are so varied that no single definition can cover all situations" (Weigle, 2002, p. 3).This perspective resulted in different views on the nature of writing skill during the history of language teaching from focusing on form, writer, reader, and content (Raimes, 1991) to what the writer does and how the writer actually does it, including guided writing, Flourishing Creativity & Literacy free writing, process writing, and academic writing (Silva & Matsuda, 2002).

Writing difficulties
L2 learners are always complaining about the problems and difficulties of writing which are caused by some special features of this productive skill.For the first thing, second language writers should normally write about what is assigned to them and not about their desired topics (Byrne, 1991).This, consequently, puts the non-native writer in a difficult situation that they "may not have enough ideas to write down or, even worse, have nothing to say" (Tho, 2000, p. 36).Matsuda (1997) mentioned that the writer's task is not as simple as creating a representation of reality; rather he or she should negotiate his or her own view on the issue with those of the readers.This means that the writer should have a good stock of knowledge of the rhetorical situation, i.e. elements comprising the context of writing, and genre knowledge, i.e. the knowledge that helps the reader to understand and respond to the rhetorical situation.The role of reader has been mentioned as another source of difficulty in dealing with writing.According to Silva and Matsuda (2002), the writer should bear in mind for what type of audience he or she is writing; whether they are real audience or imaginary ones invoked by the text.This gains much significance if the audience is an error corrector, especially grammatical errors, which results in writing unwillingness or even apprehension.
Adding to what was discussed above, learner differences were viewed by Reid (2001) to function as an obstacle in writing.He claimed that "L2 writers differ in the sequence of writing behaviors, the constraint they face in their preponderance and types of evidence, and their knowledge of the expectations of the NES [Native English Speaker] audience" (Ibid,p.30).This variation makes the writing a complex and at the same time a challenging task.The aforementioned features of writing makes language learners not at ease and uncomfortable with writing, specifically when someone else intends to read the written product.The question that arises here is that how should we overcome these difficulties?As Lagan (2000, p. 14) suggests "it makes sense that the more you practice writing, the better you will write".In other words, language learners need to write as frequently and regularly as possible to make it a "routine writing workout" (Uduma, 2011, p.59).This can be signified if the topic is meaningful and enjoyable for the writer.A good means to this end might be journal and/or dialogue journal writing which, as a type of process writing, reacts against the more traditional product writing.

From journal to dialogue journal writing: A different view on writing skill
Bailey defined journal writing as "a first-person account of a language learning or teaching experience, documented through regular, candid entries in a personal journal" (Bailey, 1990, p. 215).In fact, a journal is a daily written record of personal thoughts and feelings which seems to be important to the writer which is not set for publication (Curtis & Baily, 2009).In addition, recently researchers view journal as a writing activity where writers can freely write about their topic of interest without the fear of evaluation (Barjesteh, Vaseghi, & Gholami, 2011).A similar view on journals is hold by Crème (2008) who argued that student learning journals "can be seen as a hybrid genre of writing positioned between 'life narrative' and the 'university essay'" (p.49) [emphasis added].He continued that learner journals can also be regarded as transitional writing, formally and functionally, which offer transitional learning space that enables creative activity, fosters autonomy, and promotes learners to take responsibility for their own learning (Hamp-Lyons & Heasly, 2006;Marefat, 2002).In this regard, Barnett (2005) nominate journal writing a 'critical space' and considers it as an activity that provides students the guidance, setting, and time needed to explore ideas and expand their critical thinking ability.This safe space can allow students to improve learners' enquiry, knowledge and wisdom which have been damaged as a result of cooperation with others (Rowland, 2000).Therefore, a satisfactory space is the one that is safe and at the same time challenging which encourages learners to take risks so that they feel engaged yet free to write.Proposing this space through journal writing might decrease the L2 writing apprehension discussed above which can prevents learners from writing in different genres.Journals can also be used in writing courses as a tool to link writing and learning (Porter et al., 1990).Confirming this claim, Allison (1998) stated that when second language writers are free to choose their own topic, they are deeply focused on language learning.This was signified by Marefat (2002) referring to journal writing as an "exercise in selfreflection about learning itself" (p.104).Dialogue journal writing, however, goes much further than journal writing and is viewed as "an informal written conversation between the students and the teacher" (Larrotta, 2008, p. 21) which contains the three equally significant elements of "(a) the written communication itself, (b) the dialogic conversation, and (c) the responsive relationship" (Staton, 1991, p. xvii).Deeply rooted in Vygotsky's constructivist learning theory, Halliday and Hassan's (1989) learning as a social process, and Swain's (1995) comprehensive output hypothesis, dialogue journal writing serves as a social and cognitive activity whereby language learning is mediated by language use.According to Jones (1991), such approach to writing prepares the ground for learners to willingly express their thoughts and ideas through real written dialogue which encourages them to unconsciously investigate and acquire the accurate use of grammar, vocabulary, and spelling of words, as well as increases their writing fluency (Holmes & Moulton, 1997;Larrotta, 2008;Orem, 2001;Peng, 2007).Such development occurs not as a result of explicitly correcting students' mistakes but due to models of the correct structures provided by the teacher in the course of interaction (Yoshihara, 2008).Peng (2007) suggested that dialogue journal writing can be beneficial in enhancing learners' abilities in transferring their grammatical and lexical knowledge into real and purposeful language use and can develop confidence in writing.Stating it differently, dialogue journal writing brings the chance of authentic fluent writing (Kose, 2005).Additionally, dialogue journal writing was claimed as an appropriate means which might develop learners' writing ability regarding the writing content and confidence (Peng, 2007).

Empirical Findings
Journal writing has been found to play the role of data collection technique in language learning studies.Myers (2001), motivated by Baily's (1990) comment on re-reading the journal entry to obtain maximum benefit, explored the extent to which exchanging of and reflecting on each other's journals, hence building a written dialogue, can help learners recognize their language learning objectives.He concluded that such reflections based on Jamesian 'stream of thought' can increase learners' perceptions about strengths and weaknesses in terms of vocabulary, learning strategies, thinking skills and writing creativity.Another line of dialogue journal study focuses on the usefulness of journals as a writing tool in mastering writing recount text.Through using experimental design, it was revealed that inserting dialogue journal writing activity can enhance learners' performance in writing recount texts (Hidayat, 2011).While the studies mentioned above focused on dialogue journal writing as a research and writing tool, some studies tried to view journal writing as a classroom writing activity.A pioneering study in this regard was conducted by Marefat (2002) in which she asked 80 Persian speaking undergraduate English major students who were taking their writing course to spend 5-10 minutes at the end of each session and write their comments, reactions and feelings about the session.Content analysis of the journals revealed the areas of interest and difficulty which could be led to syllabus revision.
Marefat's study prompted other researchers to consider dialogue journal writing as a beneficial activity for writing courses.Wafa, Syafei, and Riyono (2010), implementing experimental design, concluded that dialogue journal writing was an effective activity to develop students' writing.Similar results were found by Tuan (2010) who investigated the possibility of simplifying writing complexities by engaging learners in writing journals.He found that post-test writing scores gained by learners who kept journals for thirteen weeks increased by 24.67%, while the writing scores of learners with no such writing activity increased by only 7.32% compared to their performance in pre-test.More specifically, the writing speed, measured by "the number of words produced within a limited length of time" (p.84), of the participants who wrote dialogue journals improved significantly in comparison with those who did not.In addition, the thirteen week of journal writing practice brought about a noticeable decrease in the average number of mistakes made by the participants (64.46%) though such change was not observed in the participants who did not keep journals (29.70%).Dialogue journal writing has been compared to other approaches to writing instruction by different writing scholars.Foroutan, Noordin, and Hamzah (2013a) compared email dialogue journal writing with its paper-and-pencil counterpart in enhancing writing performance.Using quasi-experimental design, they concluded that email journal writing outperformed the paper-and-pencil version not only in overall writing but also the language use aspect.However, they reported no significance difference between the two groups regarding content, organization, vocabulary, and mechanics.A great disadvantage of their study is lack of inter-rater reliability index of writing test scores which can make their findings unreliable.In another study, Foroutan, Noordin, and Hamzah (2013b) compared the effect of dialogue journal writing with task-based writing on EFL learners' writing skills and its components.Using experimental method, they found significant improvement between pre-and post-tests in terms of content, vocabulary, organization, and language use, while no significant development was observed in overall writing performance.When the mean scores of each subcategory of writing performance were compared, it was found that task-based writing outperformed dialogue journal writing regarding organization and language use, while dialogue journal writing could result in better performance in terms of content and vocabulary.It should be mentioned that such comparisons could be done more accurately and with a more reliability if Solomon design (with two experimental and one control group) was utilized.This way, the researchers would be able to check if either of the methods were effective in improving writing performance.Hemmati and Soltanpour (2012) found a greater gain of grammatical accuracy in writing as well as overall writing performance in participants who were exposed to reflective learning portfolios (RLP) than those who wrote dialogue journals during a fourteen-session treatment.The authors discussed that RLP necessitates intentional reflection scaffolding by a collaborator while dialogue journal writing requires incidental learning in which the instructor "responses without referring explicitly to the errors" (Ibid, p. 22).Therefore, "the greater efficacy of RLP in this study could have been due to the greater efficacy of intentional learning over incidental one" (Ibid, p. 23).

Research Purpose
Despite copious studies on dialogue journal writing in different areas of language pedagogy, their role in writing performance, as a productive language skill, has remained a less researched issue, specifically in an EFL context.On the other hand, as Mlynarczyk (2013. P. 37) claimed "practice, attitude, and individual tutoring" are three most important elements to develop EFL learners' writing performance which are focal to dialogue journal writing in that "[dialogue] journals give students extensive writing practice, the opportunity to express and perhaps to change their attitudes toward writing, and the chance to develop a personal relationship with the teacher" (p.36).This gains significance when journal writing is used as an outside classroom activity.Therefore, referring to the above-mentioned, the present study was an attempt to explore this claim through investigating the effect of dialogue journal writing, as an out of class activity for writing courses, in EFL learners' descriptive writing performance and its sub-components, namely content, organization, vocabulary, language use and mechanics of writing.To this end, the following research questions were formulated: 1. Does dialogue journal writing have any effect on content aspect of Iranian EFL learners' descriptive writing performance?2. Does dialogue journal writing have any effect on organization aspect of Iranian EFL learners' descriptive writing performance?
3. Does dialogue journal writing have any effect on vocabulary of Iranian EFL learners' descriptive writing performance?4. Does dialogue journal writing have any effect on language use aspect of Iranian EFL learners' descriptive writing performance? 5. Does dialogue journal writing have any effect on mechanics aspect of Iranian EFL learners' descriptive writing performance?6.Does dialogue journal writing have any effect on Iranian EFL learners' overall descriptive writing performance?
Based on the research questions above, the corresponding null hypotheses were formulated and probed in this study.

Participants
A sample of 84 intermediate Iranian EFL learners was selected based on the results of the Oxford Quick Placement Test ( 2004).The participants were high school students who were native speakers of Persian and whose ages ranged from 17 to 22 years.In addition to their weekly mainstream English education at public schools, the participants have been enrolling in an evening language institute in Iran for about 5 years.Normally, there is no systematic emphasis on writing skill in Iranian English programs at institutes.However, the learners should submit at least 8 writing tasks during each semester, as the requirements of their course book, i.e.Top Notch 3 by Saslow and Ascher (2006).The rational for selecting this sample was due to the nature of the study which required participants to have a reasonably good command of English.
In addition, two EFL instructors participated in this study as teachers of experimental and control groups and also as the raters of the participants' descriptive essays.

Dialogue journals
Each participant was asked to write at least three journals a week during the process of the study, i.e. six months.They were expected to write their reactions to the English class events, the teacher, the class atmosphere, and other students.
Since it is possible that the participants find writing about class events tedious (Marefat, 2002), they were given the opportunity to write about whatever topic they found enjoyable or interesting as well as the uninteresting or anything they want to remember for personal reasons in their daily life, either inside or outside the class.The participants were free to write their questions, suggestions and also their problems.In this way, the participants would select their topic and style which can decrease their writing apprehension.To produce the dialogic effect, the instructors commented on each journal entry to which the participants replied in a conversational manner.Following Taagart and Wilson (2005), the instructors did not correct the participants' errors and mistakes in terms of grammar, punctuation and spelling.However, the instructors write back in the learners' journals and provide feedbacks on the content or message of their journal entries.As was suggested by Young and Crow (1992), the instructors were asked to control their word counts not to exceed that of their students in order to prevent submerging the students' voice.

Writing tests
Two writing tests, one as the pre-test and the other as the post-test, were developed for the purpose of this study.In preand post-tests, the participants were asked to describe their best teachers and a life-changing experience, respectively (see Appendix A).In order to guide the participants to write about similar aspects of the given descriptive essay topics, a list of questions and elements were provided for them.The participants were asked to write about 300 words on each task.The purpose of the pre-test writing test-tasks was to specify whether or not the participants were homogeneous regarding their writing performance.4.2.4Analytic writing scoring scale Despite the fact that use of either holistic or analytic writing scheme is common in scoring writing performance, an analytic writing scale was utilized in the present study to rate the participants' descriptive essays.As Weigle (2002, p. 114) stated, analytic scales provide "more detailed information about a test taker's performance in different aspects of writing" and therefore it is "preferred over holistic schemes by many writing specialists" (Ibid., p. 115).Among different analytic rubrics, Jacobs, Zinkgraf, Wormuth, Hartfiel, and Hugley's (1981) rating scale was selected to score the descriptive writing essays of the participants.This scale is based on the five fundamental components of writing performance, including organization, content, language use, vocabulary and mechanics and has been reported as "one of the best known and most widely used analytic scales used in ESL" (Weigle, 2002, p. 115).The present study, however, benefited from an adapted version of this scale which was recently revised by Polio (2013).The updated scale contains the same five categories as were in the original one, but includes modified descriptors of each category.Moreover, opposite to the original scale, the modified one assigned equal weights to all the categories (See Appendix B).Previous studies reported that the revised scale is more reliable and valid than the original one (Connor-Linton & Polio, 2014;Polio, 2013).

Procedure 4.3.1 Data collection
The design of the present study is true experimental.Initially, in order to ensure the least difference among the proficiency level of the participants, Oxford Quick Placement Test ( 2004) was administered to 147 EFL learners.Based on the results of this test, 84 EFL learners at the intermediate proficiency level were selected.Then, the selected participants were randomly divided into two groups, to function as experimental and control (There were three classes of 14 students in each group).Following this, both groups were given the pre-test on descriptive writing ability before any treatment was applied to the experimental group.While participants of the control group took part in the pre/posttests of descriptive writing only, the ones in the experimental group were given the pre-test first, and then for their treatment they were asked to write at least three journals a week for about six months as was explained above.After the treatment, they were given the post-test of descriptive writing test task.4.3.2Scoring the essays and data analysis In order to ensure about the inert-rater reliability of the writing tests, the two raters scored the writing papers based on the same scoring scheme after being trained in using the scale.Then, Pearson Product-Moment correlation coefficient procedure was applied to the writing scores.Tables 1 and 2 represents the results yielded for inter-rater reliability of the writing pre and post-tests which shows significant and strong coefficients between the raters in overall writing and all the sub-categories of the scoring scale.At the end of the course of the study, to compare the mean scores of experimental and control groups in pre-and posttests regarding overall writing performance as well as each sub-category of the scale, the obtained data were analyzed using independent sample t-test.All the statistical tests were run using SPSS version 23.Note: 1 = rater one; 2 = rater two; **p ˂ .01.

Results
As indicated earlier, to make sure about the homogeneity of the participants with respect to their writing performance, they were subject to a pretest on writing prior to any kind of treatment to be applied to the experimental group.The descriptive statistics in Table 3 presents a general profile of their achievements both in overall writing performance and the sub-categories of the writing scale.In order to make sure that there was no significant difference among the participants in control and experimental groups regarding the writing ability before application of the treatment, the independent-samples t-test was run.As Table 4 shows, results revealed no significant difference between the two groups regarding overall writing performance and all the components of the writing scale.Thus, both groups were homogenous regarding their writing ability.

3.72925
Finding that there was no significant difference between the groups, the experimentation (assigning journal writing) was manipulated to the experimental group for six months and then a writing post-test was administered.Table 5 profiles the participants' achievements in the writing post-test.In order to probe the null hypotheses predicting no significant effect of journal writing on the EFL learners' overall writing performance and writing sub-categories, independent sample t-test was applied to the post-test writing scores.
Table 6 provides summary of the results of independent sample t-test for the means of post-test writing scores.As Table 5 represents, the mean difference between the control and experimental groups in the post-test was statistically significant in overall writing performance (t = -.492,p<.05), content (t = -1.590,p<.05), organization, (t = -.994,p<.05) and vocabulary (t = -.997,p<.05), but not in language use (t = -.366,p>.05), and mechanics (t = -.726,p>.05).In other words, the null hypotheses was rejected only for overall writing, content, organization and vocabulary whereas the ones for language use and mechanics were retained.
The magnitude of the differences in the means, i.e. the effect size for the overall writing (eta squared = .66),content (eta squared = .16),organization (eta squared = .21),and vocabulary (eta squared = .24)proved to be high based on Cohen's (1988) ratings (as cited in Pallant, 2001, p. 175).This means that about 66 percent of the overall writing performance, 16 percent of the content, 21 percent of the organization, and 24 percent of vocabulary in the writing posttest was explained by dialogue journal writing practice.

Discussion
In the present study, the effect of writing dialogue journals on the improvement of EFL writing performance was investigated.More specifically, the improvement of writing sub-components as a result of dialogue journal writing practice was scrutinized.Results indicated a significant difference between the experimental and control group in terms of the overall writing performance, as well as the sub-components of content, organization, and vocabulary in the posttest.However, no significant effect of dialogue journal writing on language use and mechanics of writing performance was observed.The obtained results supports the claims made by Lagan (2000) and Uduma (2011) which states the importance of regular writing practice in enhancing the quality of writing performance.Results also confirms findings of previous studies which indicated that writing dialogue journal entries can improve writing performance (Holmes & Moulton, 1997;Larrotta, 2008;Orem, 2001;Peng, 2007;Tuan, 2010;Wafa et al., 2010).This might be traced back to the three fundamental features of dialogue journal writing, namely the freedom in choosing the content to write about, writing a lot about those interested topics in a stress-free environment, and the dialogic individual feedback by the teacher (Mlynarczyk, 2013).Results of the present study regarding the observed improvement of content in learners' writings are in line with Foroutan et al. (2013b), Kose (2005), and Peng (2007) who found a significant effect of dialogue journal writing on the content quality of learners' writing performance.This observed effect in the present study can be discussed with regard to the nature of dialogue journal writing which calls for writing about one's topics of interest.According to Den-Bolton (2013, p. 6), "dialogue journals are the perfect antidote to writer's block as they are a place for students to write for fun, experimentation, and communication".In other words, learners can write without any mental struggle and with a variety of ideas on the topics about which they have more information than the assigned ones which are of little relevance to their personal lives.In addition, exchanging dialogue journals provides learners with enough time to reflect on their ideas, form them in an accurate and meaningful frame and cross-check the conflicting ones.These unique characteristics of dialogue journal writing might be the cause of enhancing the quality of content in writing performance.It was previously mentioned that dialogue journal writing can be an appropriate practice to transfer learners' lexical knowledge into real and purposeful language use (Peng, 2007).The current study, revealing dialogue journal writing significantly affected the vocabulary sub-component of writing performance, found empirical evidence in support of this claim.Additionally, since the model language provided by the teacher as feedbacks to the journals focused more on suitability of vocabulary and not on grammatical mistakes, learners might become aware of their lexical mistakes and problems and avoid repeating them in their future writings.In other words, as Larrotta (2008) stated, learners acquire new vocabulary through the practice of dialogue journal writing without even realizing it.This was also supported by Foroutan et al. (2013b).The non-significant effect of dialogue journal writing on language use and mechanics is in good agreement with Puengpipattrakul (2009) who found a mismatch between learners' ideas about the positive effect of journal writing on their grammatical accuracy and their actual improvement of the use of grammar as a result of practice in journal writing.A possible explanation for this can be that learners do not notice these features while writing their journals.Stating it differently, since the focus of dialogue journals were mostly the transfer of message content using appropriate words, mistakes in grammar, spelling and punctuation were tolerated and therefore learners do not attempt to imitate teachers' responses in terms of use of language and mechanics of writing in their journals and future writings.However, the present results are in contrast with Abdolmanafi Rokni and Seifi (2013) who reported significant positive effect of journal writing on EFL learners' grammatical knowledge.The reason for such contradiction is that in Abdolmanafi Rokni and Seifi's study the grammatical knowledge was checked using grammatical tests of explicit rules while this is done through assessing use of language and grammatical features in the act of writing.

Conclusion
The present study provided empirical evidence for the effect of dialogue journal writing on writing performance, in general, and on the writing subcomponents of content, organization, and vocabulary, in particular, while no significant effect was revealed for language use and mechanics.The findings signals the importance of dialogue journals in EFL writing instruction in that it offers an orderly regular writing practice which can lead L2 writers to make connections to what they are writing about through writing on their topic of interest.Such a repeated practice, based on the findings of the present study, can result in the improvement of learners' writing performance.Taking the results of the present study into account, a number of implications seems possible for the practice of teaching writing.Writing instructors are recommended to include dialogue journal writing as an appropriate outsideclass practice to help learners promote their use of written language conventions, especially use of learnt vocabulary in a meaningful context.Another implication is for material developers and curriculum designers to put increasing emphasis on dialogue journals in writing instruction materials and writing course syllabi to help learners write in a stress-free context.A life changing experience Every day we encounter different events in our lives that shape our experience.Some of those events become special for us since they change our lives in one way or another due to their unique features.Describe one of those experiences that changed your life or will change it in future in at least 250 words.You can use the following questions to know what you need to describe.
 What is that event? When and where did it happen? What exactly happened in that event?(Details)  Who were present in that event? What did you learn from that event? Why do you think that event changed/will your life?

Appendix B. Analytic writing scale
The instructors were both PhD candidates of Applied Linguistics and had at least five years of teaching experience, specifically in teaching writing courses and scoring essays.4.2 Instruments4.2.1 Oxford Quick Placement Test (2004)The purpose of administering this test was to determine the language proficiency of the participants and select intermediate EFL learners.It consists of 60 items which was developed by Oxford University Press and University of Cambridge Local Examinations Syndicate.The validity of the test has been proven in 20 countries with around 6000 students and its reliability has reached 0.90(Geranpayeh, 2003).

Table 1 .
Inter-rater reliability coefficients between the two raters (pre-test)

Table 4 .
Group statistics and independent samples t-test (pre-test) t

Table 5 .
Descriptive statistics for the writing post-test

•
Who this teacher is and what does he teach?• How long you have known him/her?• What qualities does he/she have?• What does he look like?(Appearance) • How is his/her personality?• Why do you think s/he is your best teacher?