AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 1 BLOG-BASED PEER RESPONSE FOR EFL WRITING: A CASE STUDY IN VIET NAM Pham Vu Phi Ho phamvuphiho@gmail.com Nong Lam University, Viet Nam Siriluck Usaha siriluck@sut.ac.th Suranaree University of Technology, Thailand Abstract Electronic peer response or E-peer response has recently been introduced to help L2 student writers improve their writing. However, few studies have been conducted to see how the new medium of the blog platform in the world of the Internet users (bloggers) today can be applied effectively in a real EFL context. Twelve 2nd year English major students taking a 15-week academic writing course at a Vietnamese university in Ho Chi Minh City, Vietnam, participated in this study. The students used blogs to post their essays, were trained in a 6-step procedure of peer response; then they provided and received comments two times from their peers on the first and the second drafts for revisions. The data were collected during 5 weeks on cause/effect essays including drafts 1 – 3, students‟ comments, semi-structured interviews, and learning journals. The quantitative analysis showed that the 4 most common types of comments generated via the blog-based peer response were “clarification”, “suggestion/ advice”, “explanation”, and “alteration”. The study also found that the comments did affect the students writing quality based on both the qualitative and quantitative analyses. Finally, most students expressed positive attitudes toward using blog-based peer response in the writing classroom. The study highlights a relatively new form of technology that can be used to help EFL students become better writers. Keywords: Blog, peer response, Blog-based peer response, Computer Mediated 1. Introduction With the emergence of the process approach to writing in the late 1960s and early 1970s, writing has been seen as a developmental process of inquiring, discovering, and problem solving rather than a single action resulting in a finished product (Wennerstorm, 2006). It is a result of employing strategies to manage the composing process of gradually developing a text. It involves a number of activities such as setting goals, generating ideas, organizing information, setting appropriate language, making a draft, reading and reviewing it, then revising and editing it (Hedge, 2002; Hyland, 2003, Oshima & Hogue, 2006). The writing process approach views the writing activity as dynamic, nonlinear, and recursive. The writing process approach encourages student writers to “engage in brainstorming activities, outlining, drafting (focusing on meaning), rewriting (focusing on organization and meaning), and editing (focusing on style and grammar)” (Liu & Hansen, 2005: 3). The process approach gave greater attention to teacher-student encounters around texts and encouraged teachers to support writers through multiple drafts by providing feedback and suggesting revisions during the process of writing rather than at the end of it (Hyland & Hyland, 2006). In short, the main purpose of the process approach is to empower and to motivate the student writer through the gradual discovery and development of a unique authorial “voice” (Wennerstrom, 2006; Reid, 1995) and emphasizes the writer as an independent producer of text (Hyland, 2003). mailto:phamvuphiho@gmail.com mailto:siriluck@sut.ac.th AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 2 To help L2 students become independent writers, peer response, also known as peer feedback, peer review or peer editing, has been introduced at the revision stage of the process approach to bring the students to work together to provide response on one another‟s writing in both written and oral formats through active engagement with one another‟s progress over multiple drafts (Liu & Hansen, 2005). It is a learner-centered, application of the social-cultural theory in the learning process in terms of collaborative learning (Hyland & Hyland, 2006). Students can learn from one another when they provide and receive response from their peers/instructor. In addition, peer response helps student writers with the sense of multiple readers (Liu & Hansen, 2005). Nunan (1993:100) asserts that “if we want to ensure that our ESL writing classes prepare students for their life outside the classroom, we must give them opportunities to experience collaborative writing,” adding that collaborative writing is essentially a social process through which writers look for areas of shared understanding in “real-world” contexts, writing is not a solitary enterprise; it is a social act. More recently, Hyland and Hyland (2006) indicate that peer response has been seen as a key element of students‟ growing control over writing skills in genre-oriented approaches, where social cultural theories of scaffold instruction and learning as a social practice are important. Also, from a socio-cognitive perspective, peer response can be seen as formative developmental process that gives writers the opportunities to discuss their texts and discover others‟ interpretations of them. Besides, instructors are now very conscious of the potential of peer response which helps create a supportive teaching environment for modeling ideas about good writing and developing the ways students talk about writing, especially for mediating the relationship between students‟ wider cultural and social worlds and their growing familiarity with new literacy practices. As technology develops and computer facilities become widely available, the role of the computer in delivering and mediating feedback has become a focus for research. Some researchers claim that the technological developments can empower students in the learning process and make writing classes more collaborative. To cite Warschauer et al. (1996), computer-mediated communication (CMC) allows students to take more active and autonomous roles when seeking feedback since they can ask questions whenever they wish and take the initiative in discussions. Student conferencing makes discussions more “student-centered”, fosters a sense of communicating, encourages a sense of group knowledge, and increases student participation. Also, CMC may empower disadvantaged and less able students to be equal in participation with those students who often dominate the discussions and may be particularly motivating L2 students in the learning activities (Warschauer, 2002). Some researchers also claim that CMC can lead to better writing products and more focused and better quality peer response (Braine, 1997; Sullivan & Pratt, 1996). In conclusion, computer-mediated communication peer response has been able to make peer response more effective to L2 student writers. 2. Literature review 2.1 Writing quality in CMC environment Sullivan and Pratt (1996) implemented a qualitative and quantitative research to examine students' attitudes towards writing with computers, writing apprehension, and writing quality. Thirty-eight students whose native language is Spanish from the University of Puerto Rico at Mayaguez, participated in the study. The researchers made use of the Daedalus Computer Program developed by The Daedalus Group Inc. The process methodology emphasized multiple drafts, peer and teacher responses to drafts, and dialogue learning logs. The results showed that writing environment had no effects on attitudes toward writing with computers or writing apprehension. Also, writing quality did improve in the computer- assisted classroom. Braine (1997) conducted a study to compare ESL student writing in two contexts: a networked computer class and a traditional lecture-style class. The aim was to see which setting promoted better writing, more improvement in writing, and more peer and teacher feedback. Sixty-nine students wrote three major assignments on the same topic during the academic quarter. The process approach was followed AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 3 throughout the course, with small group discussions, peer reviews, teacher feedback, occasional teacher- student conferences, and multiple drafts of essays. The study found that holistic scores for first drafts indicated writing quality in networked classes was moderately better than in traditional classes. However, final drafts in traditional classes showed a slightly higher mean improvement than in networked classes. In 2001, Braine was interested in comparing a local-area network (LAN) and traditional classes writing to explore which context produced better writing and more improvement in writing. Eighty-seven Chinese undergraduates participated in the study. The peer reviews in both the LAN and the traditional classes were allocated 100 minutes. The study found that although first drafts in LAN classes were qualitatively higher than in traditional classes, final drafts in traditional classes were of a higher quality. Furthermore, final drafts in traditional classes showed greater improvement. 2.2 Comments and Effects of E-Peer Response on revision Rodriguez (2003) was interested in computer-mediated peer response and its impact on revision. Twelve students participated in the study. The students wrote their first draft individually on a word processor in the language lab, and sent it to a peer by e-mail. The students then read one of the essays written by a peer, and wrote a 150- to 200-word feedback paper for the writer. The feedback comments were sent as attachments through e-mail. The participants read the feedback on their own work, revised their first draft, and sent the modified version (second draft) to the instructor. Results indicated that students used reacting, advising and announcing language functions when providing feedback, and focused mostly on content. Liu and Sadler (2003) investigated to see if there were some differences between the effect and affect of peer review on L2 writing in electronic and traditional modes. Forty-eight students participated in the study. The two groups followed the same basic syllabus, but the activities for the computer-enhanced group were performed on computers, for example, using Microsoft Word for in-class writing assignments, using the commenting features in Word when writing journal assignments, and using a MOO for class discussions. The findings showed that the overall number of comments, the percentage of revision- oriented comments, and the overall number of revisions made by the technology-enhanced group were larger than those by the traditional group. However, the study found that face-to-face interaction resulted in a more positive response with more focused feedback and more questions and interactions among peers. Tuzi (2004) investigated the impact of e-feedback on the revisions of L2 writers in an academic writing course. Twenty L2 writing students participated in the study. In the learning process, the L2 writers would post their essays onto the writing website with their user accounts, and they could also provide comments to the other authors as well. The students had approximately 10 days to read any e-feedback they received from their peers, teachers, and website visitors, and revise their papers before the final drafts were due. The students could also obtain oral or written feedback from their peers and assistance from visiting the writing center. The study found that students preferred oral feedback to e-feedback. However, e-feedback had a greater impact on revision than oral feedback. Additionally, e-feedback affected L2 writers‟ revisions at a higher structural level. In order to prepare students to be better peer reviewers, Stanley (1992) examined the effect of training students to become better peer evaluators in university ESL freshman composition classes. The training sessions lasted 7 hours during the first 4 weeks of a 15-week semester. During the training sessions, the students engaged in role-play, analyzing the genre of peers‟ essays and discovering rules for effective communication. Results showed that the four response categories that most often produced revision among the coached group were pointing remarks, advising remarks, collaborating, and questioning. Students who received coaching looked at each other‟s writing more closely and offered writers more specific guidelines for revision than did the uncoached group. AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 4 Min (2005) conducted a study aimed at training students to become successful peer reviewers. Participants in the study were 18 EFL sophomore students in the researcher‟s composition class at a large university in southern Taiwan. They were introduced to the practice of peer review at the beginning of the fall semester. The training session was composed of two phases: 4 hours in class and 30-minute teacher – student conference. The researcher required the students to compose the last essay at home and bring their drafts to class for peer review. The study found that the numbers of comments and words produced post- training were significantly higher than those prior to training. In addition, the students were able to provide a greater amount of feedback on the global features after training. They all pointed out that the four-step procedure helped them become better reviewers, although following the four steps was both time- and energy-consuming. 2.3 Literature gap in employing CMC for peer response Recently, many researchers have conducted studies which employed quite a few tools of technology. Li (1999); Rodrigez (2003); Lightfoot (2006) made use of e-mail to assess students‟ writing and interactions. Sullivan and Pratt (1996); Braine (1997 and 2001) employed LAN (Local Area Networked) to compare the two modes of peer responses. Jones et al. (2006) used the MaxQDA software for the on-line tutoring modes either asynchronous or synchronous. Sadler and Liu (2003) took advantage of MOO (Multi-user Domains Object-Oriented) systems to compare the effectiveness of traditional peer response versus electronic peer response. However, very few studies, if any, have examined the use of blogs, which are quite popular among Internet users today, to investigate asynchronous peer responses to see if it helps students improve their writing quality. In addition, no research has been conducted in a Vietnamese context to apply blog-based peer responses in writing classes. The advantage of using blogs for writing classes is that when students publish their writing online, they will take more care of their writing quality, “since the report stage is public, students will naturally want to be accurate” (Willis, 1996 in Sangarun, 2003: 26). According to the blog search site Technorati (2006), there were over 35.3 Million blogs in the world and there appeared to be approximately 75,000 new blogs every day. Gartner (2007) predicted that there would be nearly 100 million bloggers in 2007. In Vietnam, according to Vietnamnet (2007), there were more than one million Vietnamese using the blog services of Yahoo! 360o, at least in one sort of blog. If other kinds of blogs are accounted for, the number could be far higher. The online publications VNexpress (2006) and Tuoitre (2007) stated that many Vietnamese people, especially students have become bloggers to connect to their classmates, share information about their studies, and post their assignments on their blogs. Thanhnien (2005) also described many students‟ writing on their blogs related to their learning activities and their assignments, specially their reflections about their everyday learning activities. Therefore, the current researcher has chosen to incorporate the use of blog-based peer response into this study to help students enhance their writing quality. 2.4 Rationale for the study In general, writing instruction in Vietnam has traditionally been based on finished products that focus on form over meaning and the finished text rather than on the process in which writing takes place (Tran, 2000; Nguyen, 2002; Nguyen, 2004; Lien, 2006). Tran (2000) claimed that in writing courses, when assigning a whole composition, the instructor asked the students to write on a topic only once. There exists always the pressure of time and little chance for improvement after the students produced their first drafts. In other words, students have had only one chance to complete their writing assignments with no opportunities to revise. Also, the time provided for their writing is limited. In addition, the student writers have only one audience to write for the instructor himself. Consequently, the writing process does not occur in real life. Nguyen (2004) investigated factors leading to success in the use of group work in foreign language classes. She found that group work practice was still unsatisfactory for two main reasons. The first was part of the students' characteristics - their low proficiency levels, lack of AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 5 motivation, and preference for the traditional learning style. The other was related to part of the instructors' implementation of group work, the English program, and the teaching materials. Lien's survey (2006) showed that the learners‟ concept about learning English was product-based; they perceived a lack of grammar knowledge that made learning English difficult. Luu (2006) and Tran (2006) posited that the current educational system of compositions has brought student writers to learning by rote with no chances for creative writing. Helping Vietnamese L2 students to write more effectively in English through the process approach, which is believed to enable student writers to become independent, self-editors, is by no means easy. Three research questions are addressed in this study: (1) What types of comments are most frequently produced by the students during the peer response activities? (2) Do the comments affect the students‟ writing quality? (3) What attitudes do students express on the use of blog-based peer response? 3. Methodology 3.1 Participants Participants selected in the study were 12 sophomores majoring in English in the instructor/researcher class. There were 9 females and 3 males, and their ages were ranked from 20 to 22 with the mean at 20.67 (fig. 2). They all were native speakers of Vietnamese and had passed the National College Entrance Exam (English Major) administered by the Ministry of Education and Training in Vietnam before being admitted to the Faculty of Foreign Languages of a Public University in Ho Chi Minh City, Vietnam. Their English proficiency, as measured by the Test of English as a Foreign Language (TOEFL) exam, ranged from 420 to 507. None of the participants had received any training in peer response via the blog prior to the study. 3.2 Setting The students of this study were English majors and the period of this study took place while they were in the second semester of the second year. They all got involved in two previous semesters of Academic Writing in which they had learned how to write paragraphs of Narration and Description. In addition, they had learned three genres of essays: Logical Division of Ideas, Supporting an Opinion, and Comparison- Contrast essays as the normal curriculum of the Faculty of Foreign Languages. The focus of this classroom was to develop students‟ writing skills in Cause/Effect and Chronological Order/ Process essays. This course met twice a week for 15 weeks. The instructor/researcher adopted and modified the “writing cycle” (Tsui & Ng, 2000) in designing his class of Academic Writing. The writing cycle was described as follows: Topic selection, brainstorming, writing the first draft, posting that draft on the blog, giving and receiving peer comments, revising writing the second draft, second round of peer comments, further revision for writing a third draft, teacher comments (as a normal activity of the writing process), then final revision to write the fourth draft. However, in order to observe the effects of peer comments, only the drafts 1 -3 were counted for data analysis. AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 6 The writing cycle Figure 1 Working in the computer lab, the students were required to search for information to support for their essays. Four topics, two at the paragraph level (reviewed as the normal curriculum of the Faculty) and two at the essay level, were written during the course of the semester – “How to write a good paragraph”; “The causes of traffic accidents in Vietnam”; “The benefits of living in a big city”; and “How to maintain good relationship with friends”. Each topic could be revised up to three times, two after receiving peer comments and the other after the instructor comments. 3.3 Procedure Before the class started, the participants took a paper TOEFL test for estimating their English proficiency. The results of the test were used for grouping the participants into three different groups of four each. The students were grouped by mixed ability and shared ability levels (Richards & Lockhart, 2000). In order to mix the students‟ levels of proficiency, two students who obtained highest scores were mixed with the two lowest proficiency students (group 1); two second highest proficiency students were mixed with two Topic selection (instructor and students) Brainstorming Writing first draft Peer comments (focus on content & organization + peer_editing sheet) Writing the second draft Peer comments (focus on forms and content & organization + peer_editing sheet) Third draft Teacher comments (focus on content & organization and forms) Fourth (final) draft AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 7 second lowest proficiency students (group 3), and finally, the four medium proficiency students were put together (group 2). After the instructor/researcher determined the three groups, the group members selected a monitor “to get the ball rolling” for each group. This was shown in the figure 2 below: Participants‟ information Group Nicknames Sex Role Ages TOEFL Scores 1 Thienphuoc (S1) Male 21 420 Sunflower_forever (S2) Female 21 507 Starmeomeo (S3) Female Monitor 20 503 Thanh_le1608 (S4) Female 20 403 2 Bluesun (S5) Female Monitor 22 430 Haquyen (S6) Female 20 443 Bebi (S7) Female 20 443 Trumthong (S8) Male 20 490 3 Hoatrongda (S9) Female 20 493 Lanphuongkt1 (S10) Female Monitor 21 427 Nhoc_HT (S11) Male 19 501 Bocaukhongduathu (S12) Female 22 423 Figure 2 The 15-week Academic Writing course was sequenced as follows. Each week, the students had a class meeting for 3 hours in the Computer Lab (compulsory), and about other 3 hours in the Lab (optional) for providing and receiving comments. As stated in the curriculum of the Faculty of Foreign Languages, the first 4 weeks were spent reviewing writing at the paragraph level; other 5 weeks were used for learning to write a Cause/Effect essay; 1 week for the midterm test; and the last 5 weeks were for learning to write a Chronological Order/Process essay. During the first week of the course, students visited the website http://360.yahoo.com to sign up for an account (fig. 3) for their own weblog or blog (if someone who already had an email account with Yahoo, he/she was not required to sign up). The students were trained to set their blogs for selected friends only (group members) so that only the designated group members could read and provide comments on their writing entries. They were trained to be familiar with the appropriate computer usage, accessing the Internet, and how to provide comments online. Two assignments of paragraph writing were done via the blogs without any specific peer response training. The purpose was to help students become accustomed to the required computer techniques. AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 8 Sign up for 360.yahoo.com Figure 3 3.4 Peer response training The peer response training took place during the third writing cycle, after the first draft of the Cause/Effect essay posted on the blogs, and consisted of two phases: in-class training and one-on-one student - teacher conferences. The in-class training lasted 3 hours during the third writing cycle of Cause/Effect essay. The Peer-edit sheet (Oshima & Hogue, 2006) provision was used as guidelines to help students read and provide comments (see Appendix A). The one-on-one student - teacher conferences lasted from 15 to 20 minutes beyond normal class meetings after the first round of commentaries on the first drafts of the Cause/Effect essay. 3.4.1 In-class training The in-class training, based on 11 guidelines of Berg (1999), started after the first post of the third writing cycle. During the in-class training, the instructor/researcher first helped the students understand the importance of peer response in the writing process; then he helped the students provide comments on some essays composed by former students based on a 6-step procedure as follows: 1. Evaluation: (Stanley, 1992; Tuzi, 2004) The peers valuated some parts of the writer, or some sentences or phrases or some ideas. The evaluation could be positive or negative. However, in some cases, praise was used to reduce the tension because some students might not feel comfortable in critiquing other‟s writing for fear that the writer might not receive their criticism as well (Liu & Hansen, 2005). Therefore, positive evaluations were encouraged. Students could evaluate the writing: Generally: “This is really good”, “I like this paragraph” Specifically: “This is a great thesis statement”, “This is a not clear thesis statement” 2. Clarification: (Stanley, 1992; Zhu, 2001; Min, 2005) Peer readers identified or located a particular problem in order to help the writer realize and revise his/her essay. They could point to: AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 9 specific ideas: “Where you say… what do you mean?”, “Could you explain your thesis statement in more details?” particular word choices: “What do you mean by …?”, cohesive gaps: “You say „…‟ How does this sentence connect to the one before?” unity of the paragraphs: “Do you think this sentence or phrase is united to the main idea of this paragraph?”, “Do you think this sentence „………….‟ directly explain or prove the main idea?” 3. Alteration: (Tuzi, 2004; Liu & Hansen, 2005) The peers provided comments in an imperative tone instead of advice. Ex: “I try to break the door down → tried to break.” or “Change your thesis into X” 4. Suggestion/ advice: (Zhu, 2001; Tuzi, 2004; Min, 2005) Peer readers suggested ways to change the words, content, and organization of essays. The advice could be general or specific: Specific advice: “Your thesis statement should be explained more clearly”, “You might include an example/fact/statistic here”, “You should change this transition signal „……….” to”……..” to show the contrast idea. General advice: “You should introduce your introduction paragraph in the form of a funnel, or historical background, or surprising statistics, or dramatic story”, “You need more ideas on this paper”, “Write more reasons to support your opinion”. 5. Explanation: (Zhu, 2001; Min, 2005; Tseng & Tsai, 2007) Peer readers explained why they thought a given term, idea, or organization was unclear or problematic, which should or should not be used in the essay. This step included the advice and clearer information for the problems. Ex: „You should change "Despite … into although" (Despite + N/N phrase, although + clause)', „I think you should reduce these two sentences because they talk about the conveniences of computer, not about the good education in a big city. 6. Confirmation: (Zhu, 2001) Peer readers tried to confirm the information of a particular feature either for revision or non- revision. However, there was no suggestion for revision. In the case of questioning, the peer readers might not be sure about a particular feature for revision; then they asked the writer to pay thought again on a specific feature to see if he/she needed to change. Ex: “Your essay has a thesis statement, and topic sentences”, “will wait for you or wait for you?” 3.4.2 One-on-one student – teacher conferences The one-on-one student-teacher conferences, out of normal class meetings, were to help students learn from their own commentary experience if necessary for improvement of the quality of the comments. After the first round commenting on the cause/effect essay, the instructor/researcher had close looks at the comments of each students and held one-on-one student-teacher conferences for 15 to 20 minutes. Some good comments of other students were also shown as models. In addition, student-teacher conferences were to check students‟ comprehension of instruction and feedback (Min, 2005). Furthermore, during the student – teacher conferences, the instructor/researcher helped address students‟ problems concerning such things as unclear comments from their peers as well as explain some problems that the group members did not know for sure about some specific grammatical structures or particular ideas. For example, one reader commented on Starmeomeo‟s essay about a grammatical point of tense used in a conditional sentence “So if we have ability and certificate, many career positions wait for us”. The reader suggested revision by pointing out: “Is this a conditional sentence? Many career positions will wait for AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 10 you or wait for you?” In this case, both the reader and the writer were not sure about the accuracy of the grammar; so the writer sought help from the instructor. The instructor/researcher helped her to clarify what kind of idea she really wanted to express and confirmed this grammatical point. 3.5 Inter-rater training The trained inter-raters were two of the researcher‟s colleagues who have worked in the same university. Two of them had taught EFL writing for approximately 4 and 5 years, and during the period of this study, they were enrolled in MA programs at foreign universities located in Ho Chi Minh City. The training included 2 phases: Rating the students‟ essays and coding the comments. 3.5.1 Rating students’ essay training The researcher and the two trained inter-raters met two times before the course of Academic Writing began. First, the inter-raters were given the Analytic Scoring Rubric (see Appendix B) which was explained by the researcher (adapted and modified from Oshima and Hogue, 2006). Second, five hard copies of the essays of former students were given to the inter-raters for marking independently. The essays were assigned scores from 1 to 10 scores. Half-point increments were also considered (e.g. 1.5; 2.5; 3.5; 4.5…, 9.5). If the discrepancy is less than or equal to 1 point between two inter-raters, the two readers met to reach an agreement. If the two readers could not reach the agreement, and the difference in this case is small (≤ 1 point), an average was calculated between the two readers‟ scores. The researcher also adopted the method of Berg (1999) in dealing with discrepancy in scoring. A discrepancy in scoring was considered to exist when scores were more than 1 point apart, in which case a third rater (the researcher) would read that essay using the same scoring rubric. An average was then calculated based on the third reader‟s score and one of the two raters‟ whose score was closer to the third reader. Third, the researcher compared the scores from those two raters to check the consistency. In the first round of scoring, only two papers were consistent. Other two were discrepant by 0.5 and 1.0. The researcher asked the inter-raters to compare their rubrics point-by-point to find out the discrepancy and discuss the issues. They then reached their agreements on those papers. Afterward, 5 different essays were given to the two inter-raters for scoring. This time, only one paper was discrepant at a score difference of 0.5. This training lasted for 3 hours. The second round of inter-rater training took place one week before the essays of the participants were given. The purpose was to make sure that the scoring of the two trained inter-raters was still in correlation. The procedure of the training was the same as the first round, except the time was shorter, just 2 hours, due to the familiarity of the scoring process. The researcher used the analytic scoring scheme for the following reasons. First, analytic schemes are seen as far more useful tool for training and standardization of new examiners, as inexperienced raters can more easily understand and apply the criteria than holistic scales (Weir, 1990; Weigle, 2002). Hence, it was proper for the two trained inter-raters in this study. Second, analytic scoring scales are devised in an attempt to make the assessment more objective because they encourage the raters to be more explicit about their impressions (Weir, 1990). Third, analytic scoring schemes are particularly useful for L2 student writers as well because they clearly and distinctly address the different aspects of their writing abilities. Then, the student writers will be more aware of and thus focus more on the problem areas (Weigle, 2002). Finally, analytic scoring scales can be more reliable than holistic scoring in terms of additional items added to a discrete-point test, so a scoring scheme in which multiple scores are given to each script tends to improve reliability (Hughes, 2003; Weir, 1990). 3.5.2 Coding comment training Apart from the scoring rubric training, the researcher and the two inter-raters met for a 3½-hour for coding comment training. First, two copies of the coding scheme (see Appendix C) were given to the coders and explained carefully by the researcher. Since the comments were sometimes complex, AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 11 comprehensive discussions occurred during the training. After one hour of discussion, six sets of comments from two essays by former students were given to the coders. Coders were given 3 sets each. The two coders worked independently to code those comments. They marked types of comments (6 types) and areas of comments (global and local areas) into the “Grid for comment analyzing” (fig. 4). The most inconsistent types they made were the distraction of “suggestion/advice and alteration”. For example the first coder considered the comment “this sentence should be changed into passive voice” as the type of “suggestion/advice” while the other found it as “alteration”. Then the researcher made a compromise that it was in the “alteration” type and if the comment included the subject such as “you” before the hedging devices would be seen as “suggestion/advice”. The researcher and the coders also tried to make sure that a complex comment on one issue should be counted as 1 nature of comment, either revision- oriented or non revision-oriented (described in details in data analysis). 3.6 Data collection procedure After the in-class peer response training, the students were required to provide comments on the first drafts of the cause/effect essays of their group members via the blogs. The students could make it in the computer lab of the university or at an Internet Café or at home if they had Internet. The students were given 4 days for peer comments and 3 days for revisions of each draft. Then the instructor/researcher collected drafts 1 to 3, and peers‟ comments via the blogs for analysis. The average number of words produced for the first draft and third drafts were 451 and 588, respectively. The two inter-raters carefully read the drafts to assess the essay quality. The inter- raters also identified the types of comments made by the student writers. In addition, the researcher triangulated “methods” through writing journals after the revision stage, and semi-structured interviews at the end of the course to obtain in-depth information from the students. 3.7 Semi-structure interviews Semi-structured interviews were conducted around two weeks after the training from May 6th to May 20th, 2008. In order to select better respondents, the researcher adopted the methods of Oppenheim (1999) for the interviews. Three key students (monitors) of the three groups and a rich informant were selected. Each student was asked about 10 structured questions (See Appendix D) and some other questions emerged based on the responses during the conversations. The semi-structured interviews were tape-recorded and each lasted for 20 to 30 minutes. 3.8 Data analysis Both quantitative and qualitative methods were used. Regarding the quantitative analysis, the researcher compared the number of types of comments, the scores of the first drafts and the third drafts. Regarding the qualitative analysis, the researcher explored the students‟ attitudes on this training as described in their writing journals after they revised their essays. The purpose of the writing journal was to provide students a space to reflect their opinions on the blog-based peer response activities. The students were encouraged to express their own honest opinions about the training. Also, the data from the semi-structure interviews after the training were also accounted for analysis. 3.9 Coding procedure For the quantitative analysis, two inter-raters coded the comments blindly (without students‟ names on the papers) based on the grid for data analyzing (fig. 4). A rubric-coding scheme (see Appendix C) was provided to the inter-raters as guidelines for their coding the comments. They first tallied the number of written comments including types of comments via the blogs from drafts 1 to drafts 2. In terms of types of comments, seven types of comments (6 in-class training and 1 (statement) was not) were coded in terms of evaluation, clarification, alteration, suggestion/ advice, explanation, confirmation (as described in the in-class training) and statement were counted. AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 12 The category of “statement” was not counted in the in-class training because this kind of comment was neither of a revision-oriented nor non-revision oriented nature. Instead, they were of a social solidarity nature. For example, after commenting on an essay, thienphuoc wrote “This is just my opinion, I hope it will help you a lot” or sunflower said “These are some points I give you. I hope they help you much”. The reliability of each type of comments reached from 0.93 to 0.98. Grid for comment analyzing Figure 4 (adapted and modified from Liu & Hansen, 2005) Examples of applying coding scheme for comment analyzing: Example: I think this part is off topic.// You are talking about “chances for education”, why do you talk about transportation?// Suppose that you live in a big city, for example in district 1, but your school is at Thu Duc district, do you think that you spend a lot of time or a little?// In this comment, there was only one meaningful unit (nature of comment) but three types of comments. “I think this part is off topic” was coded as global (area), clarification – unity Nickname: ………………………… Draft …………. GRID FOR COMMENT ANALYSING Types of comments No. of types Global Areas Local Areas No. of areas Revision- oriented Non- revision- oriented Revision- oriented Non- revision- oriented Evaluation Generally Specifically Clarification Specific ideas Particular word choices, phrases, or sentences Cohesive Unity Alteration Suggestion/ advice Generally Specifically Explanation confirmation Statement Total AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 13 (type), and revision-oriented (nature); “You are talking about “chances for education”, why do you talk about transportation?” was coded as clarification – specific of idea (type); “Suppose that you live in a big city, for example in district 1, but your school is at Thu Duc district, do you think that you spend a lot of time or a little?” was coded as explanation (type). If two comments were addressed to one issue, only one of them was counted as a comment. Obviously, the better comment was considered while the other was out of concern. In the following examples, the second comment was counted in the coding scheme because it at least pointed out the light for revision. Comment 1: In the sentence: “For me, who have been living in a big city all my life, living in a suburb also …” What does it mean? Comment 2: In the sentence: "For me, who has been living in a city all my life, living in a suburb..." is not right in grammar and not logical. You should rewrite it such as "For me, a person who has been living in a city all my life, considers that living in a suburb..." 3.10 Rating procedures The same rating procedures were made as in the rating training process. Drafts 1 and drafts 3 were rated by analytic scoring rubric (see Appendix B). This was done after names of students and other identifiers, nicknames, were removed from all papers. All papers looked similar because they were laser printed. The papers were rated by the two inter-raters using a 10-point scoring guide. The essays were rated from 1 to 10. Half-point increments were also considered (e.g. 1.5; 2.5; 3.5; 4.5…, 9.5) to be more precise. This numerical scoring was applied to match the numerical scoring system of the university in particular, and of the scoring system of the whole country in general. If the discrepancy was less than or equal to 1 point of the two trained inter-raters, a discussion between the two readers was held to reach an agreement. There was no discrepancy higher than 1 point in this rating procedure of this case study. The correlations of the two inter-raters‟ scores were significant at the 0.01 level (2-tailed). 4. Findings The findings would be discussed according to three research questions for this study. Research Question One: What types of comments (evaluation, clarification, alteration, suggestion/advice, explanation, and conformation, and statement) are most frequently produced by the students during the peer response activities? AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 14 Table 1 Types of comments (draft 1) Types of comments that students received from their peers during the peer response session 1 evaluation clarification alteration suggestion/ advice explanatio n confirmatio n statement students Mean = 3.58; SD = 2.31 Mean = 10.75; SD = 5.07 Mean = 3.08; SD = 2.28 Mean = 10.67; SD = 4.52 Mean = 3.67; SD = 3.34 Mean = 2.67; SD = 2.54 Mean = 2.33; SD = 1.67 S1 5 17 3 18 13 2 2 S2 2 10 2 8 1 2 5 S3 2 22 3 15 6 4 2 S4 2 16 7 18 2 10 1 S5 5 6 3 5 3 2 5 S6 1 8 0 7 4 4 3 S7 9 8 7 14 3 2 4 S8 2 5 1 8 5 2 0 S9 4 10 3 7 1 1 3 S10 3 9 4 10 1 1 1 S11 6 7 0 11 3 1 1 S12 2 11 4 7 2 1 1 Total 43 129 37 128 44 32 28 Of 12 first drafts (drafts 1) of 12 students, there were totally 441 comments delivered over 7 types of comments (table 1). The mean of each essay receiving comments was 36.75. The total number of “clarification” comments was 129 (29.25%), followed by 128 of “suggestion/ advice” (29.02%). “Explanation” type was in the third place of 44 (9.97%). Next to “explanation” was “evaluation” type of 43 (9.75%). “Alteration” and “confirmation” types were of 37 (8.39%) and 32 (7.25%). Finally, “statement” type was last with 28 (6.35%). AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 15 Types of comments Types of comments 0 20 40 60 80 100 120 140 ev al uat io n cl ar ifi ca tio n al te ra tio n su gg est io n/ ad vi ce ex pl ana tio n co nf irm at io n st at em en t Series1 Figure 5 The most frequent types of comments generated during peer response activities of drafts 1 were “clarification” (Mean = 10.75) and “suggestion/advice” (Mean = 10.67) in each draft. “Explanation” (Mean = 3.67) was third most common type. The mean of “evaluation” was 3.58, indicating that students tried to keep the harmony in commenting. The fifth rank of comments went to “alteration” of which the mean was 3.08, followed by “Confirmation” at 2.67. A part from those six types of common comments, “statement” which was not categorized because it was neither revision oriented or non-revision oriented (nature of comments), was provided by readers (Mean = 2.33). This showed the interactions of students during the peer response activities. The findings suggested that the most common types of comments favored by students during the peer response activities were “clarification” and “suggestion/ advice”. The discrepancy between those two was nearly equal (129 vs. 128). “Explanation” was also favored by students in the first drafts. AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 16 Table 2 Types of comments (draft 2) Types of comments that students received from their peers during the peer response session 2 Students evaluation clarification alteration suggestion / advice explanation confirmation statement Mean = 2.92; SD = 2.07 Mean = 9.67; SD = 4.74 Mean = 4.08; SD = 3.85 Mean = 9.33; SD = 3.89 Mean = 3.67; SD = 3.11 Mean = 3.17; SD = 2.52 Mean = 2.50; SD = 1.62 S1 2 6 4 11 10 0 4 S2 4 16 13 16 8 2 5 S3 2 16 0 14 5 1 5 S4 1 18 8 12 6 7 2 S5 2 7 3 8 2 5 3 S6 4 11 5 6 3 6 1 S7 8 4 8 3 4 2 4 S8 1 8 1 6 3 7 1 S9 2 7 3 8 0 1 1 S10 1 9 1 13 1 1 1 S11 3 4 1 9 2 2 2 S12 5 10 2 6 0 4 1 Total 35 116 49 112 44 38 30 The total number of comments on draft 2 appeared to be slightly fewer than that of drafts 1 (424 vs. 441) (table 2). The mean number comments that each essay received was 35.33. The most common type was toward “clarification” at 116 (27.36%). The mean of each essay receiving “clarification” comments was 9.67. This was followed by “suggestion/ advice” with a total of 112 (26.42%) and a mean of 9.33. “Alteration” was 49 (11.56%; Mean = 4.08) which was higher than that compared with the first drafts (8.39%), and ranked the third common type after “clarification” and “suggestion/ advice” while it was in sixth in drafts 1. “Explanation” was the same amount as in the first drafts of 44 (Mean = 3.67), but it was higher in percentage (10.38% vs. 9.97%) and favored as the fourth common type in drafts 2 while it was in the third in drafts 1. The fifth common type of comments went to “confirmation” of 38 (8.96%; Mean = 3.17). The sixth ranking type of comments “evaluation” included 35 comments (8.25%; Mean = 2.92) while it ranked quite higher in the first drafts of 43 (9.75%). It appears that the students seemed to prefer more specific comments than general evaluative statements. “Statement” type was least in the commentary activities (7.08% = Mean = 2.50). Compared to the first drafts, the findings indicated that students were still relying primarily on “clarification” and “suggestion/ advice” types during the peer response sessions. “Alteration” was found to be favored more in the second drafts if the same issues were repeated from the first to the second drafts. AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 17 Types of comments Types of comments 0 20 40 60 80 100 120 140 ev al ua tio n cl ar ifi ca tio n al te ra tio n su gg es tio n/ ad vi ce ex pl an at io n co nf irm at io n st at em en t Series1 Figure 6 Research Question Two: Do the comments affect the students‟ writing quality? Table 3 Pre-test – Post_test Mean S.D Std. error mean t df Sig. (two-tailed Pair 1 Pretest - Posttest -1.292 0.4981 0.1438 -8.983 11 .000** a. Pretest refers to the first drafts b. Posttest refers to the third drafts *P < 0.05 **P < 0.01 As described in the rating procedures, two trained inter-raters independently rated students‟ essays blindly. The inter-rater correlation was significant at the 0.01 level (2-tailed) making the inter-rater reliability of the first drafts 0.73. Also, the correlation of drafts 3 was significant at the 0.01 level (2- tailed) with a reliability of inter-raters at 0.86. Discrepancies were discussed between the two inter-raters for the first draft because there were no discrepancies higher than 1 point, the researcher did not serve as the third rater. The mean score of the 12 first drafts (pretest) was 5.91 and the third drafts (posttest) was 7.21 (table 7). No essay scored was less than 5 on the 10-point scale. Matched Paired test showed that the improvement in mean scores of the pretest and posttest was statistical significant (P < 0.01). The findings indicated that the peer response via the blogs helped student writers enhance their writing quality through their revisions after receiving comments. AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 18 Research Question Three: What attitudes did students express on the use of blog-based peer response? The attitudes of students on blog-based peer response were described qualitatively based on the semi- structured interviews (see Appendix D) after the training and the writing journals after the revisions. The researcher transcribed the interviews from tape recorders and then translated the interview transcripts and the writing journals into English. Additionally, the researcher asked two other Vietnamese, one has attended the PhD and the other attended the MA program, to do the back translation after the texts were categorized for verification of the translation. Most students expressed positive attitudes toward applying blogs to the peer response activities though some of them had some initial difficulties with the technology. Time saving and conveniences of blog- based peer response were also reported. R So far you learn how to post writing on the blog, provide comments on peers‟ essays, and revise your writing. What do you think about it? S3 I now like to apply it more because at first I was not familiar with it, so I thought it was complicated, I did not like it. However, after getting used to it, I found that my writing improved. Before taking this course, I also studied in group with some of my friends, sitting together to discuss and correct mistakes for one another‟s essay, but we just did it on papers, not on blog. That way consumed a lot of time because we had to spend the whole morning for gathering, waste of time. Working via the blog was more convenient because I posted my writing on the blog; then dated with friends to be online. I did not need to travel from Bien Hoa to Thu Duc (around 20 kms from her home to the University). S5 I have enjoyed accessing the Internet so far. I also have had a blog for a long time, but I did not know what to use it for, or to take advantage of its full usefulness. Thanks to this course, I found the blog useful to post my writing for my group members to read and comment; so I realize it is good.” S10 In my opinion, it is very good because my friends could provide comments any time. When my friends are in good mood, they can provide better comments on my essay.” In the writing journal entries, students stated that they had gained high positive attitudes on the usefulness of blog-based peer response in their writing classroom. In addition, the harmony among the group members was also reported in their entries. Students also confirmed that they learned from one another when reading their peers‟ essays. S11 Learning to provide comments via the blog is rather interesting and it is very useful to the Vietnamese [students]. It also helps not only me but also my friends to improve the writing styles of each person… S6 To me, this is a new learning approach and it should be applied widely to all other subjects. … This is a new method of learning which is worthy for my effort and pursuit… S12 To me, this way of learning is interesting and relaxing... I learn a lot due to the enthusiastic comments from my peers in my group. I also learn new things from reading my peers‟ essays… S2 I found that this way of learning is very interesting and useful. It helps me get closer to my friends. My peers commented on my writing every well… I appreciated my friends a lot and I hope that we still develop to help improve our writing skills better and better. S9 ... Now the person who taught me writing was not only the teacher, but also my friends in my group and I myself… During the peer response sessions, the students sometimes debated to defend their stance over some features if they had some disagreements on the issues. However, the discussions were helpful to one AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 19 another to improve knowledge of writing. In this vein, Hyland and Hyland (2006) stated that from a socio-cognitive perspective, peer review can be seen as a formative developmental process that gives writers the opportunities to discuss their texts and discover others‟ interpretations of them. In addition, Warschauer, et al. (1996) claimed that computer-mediated communication (CMC) allowed students to take a more active and autonomous role when seeking feedback since they can raise questions whenever they want to and take the initiative in discussion S3 I really like to have discussions, discussing, not arguing. Once I debated seriously with one of my friends, because when she commented, I thought I was right. I was afraid that I might be wrong if I revised based on her suggestion. Then I went to the classroom to debate with her. We used our old books (former materials) to check. We also looked for some other materials [related to that point]… I finally found I was wrong. Afterward, I agreed with her on her subsequent comments, so there was no longer any serious debating. S10 At that time, my concept was conservative. I thought my writing was clear in expression. I did not know how to add more ideas [as my peer commented]… then we met in the in the classroom to discuss about the problem. Despite the positive response of students using the blog for peer response activities, there were some restrictions which should be taken iton consideration such as Internet expense and the time consuming. R Is there any inconvenience when you post your essay via a blog? S3 There are also some inconveniences because my friends do not have the Internet connection at home, so they need to go to the Internet café to discuss. In fact, I often respond via the blog then we discuss the issues in the classroom. S1 I think the blog is useful to respond because it‟s easy to read. I can copy and correct it easily. I don‟t spend a lot of time. It is easy to observe my writing and comments. However it‟s not convenient if I must go to the net-shop (Internet café) out my house (assuming that I don‟t have computer and connect internet at home). It also costs a lot of money. S6 This method consumed rather much time and money as well… S12 … The first thing to mention is that this learning approach consumes much time… 5. Discussion 5.1 Types of comments generated via blog-based peer response The first research question of this case study was to investigate which types of comments were most frequently produced by the students during the peer response sessions. A six-step procedure was taught to students before they performed peer response. Quantitative analysis revealed that through two drafts, the students were able to produce most on “clarification” and “suggestion/ advice” types. The means of those two types of comments were much higher than any of the others. The findings seemed to echo (1) Stanley‟s (1992) findings that pointing remarks (clarification) and advising remarks (suggestion/ advice) were favored by students during the peer response sessions; (2) Zhu‟s (2001) that non-native speakers employed “announcing” and “questioning” (clarification) most frequently; (3) and Rodriguez‟s (2003) that students used “advising” (suggestion/ advice) and “announcing” (clarification) when providing feedback. However, these findings differ from Liu and Sadler‟s (2003) in which “clarification” and “suggestion” were generated in fewer numbers than “alteration” and “evaluation” in the Technology Enhanced Group. The findings of this study, based on the coding data of this case study, indicated that student writers helped their peers clarify the problems first; then provided suggestions for revision orientations. Therefore, the correlation of these two functions showed that the students conducted their comments comprehensively. In other words, the students got quite involved in the blog-based peer response activities. Besides, “explanation” and “alteration” were also considered. “Explanation” which was favored (third in drafts 1) fourth in drafts 2, the number of comments stayed the same (44), indicated that the students took their responsibilities in their comments in order to help one another revise their AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 20 essays. In terms of interaction, the students produced more engagement in the “explanation” type because it included a suggestion and an explaining statement. The type of “evaluation” in drafts 1 seemed to be rather high (9.75%), and positioned fourth after “explanation”. This finding seemed to correlate with Tuzi‟s (2004) and Liu and Sadler‟s (2003) findings that students were more comfortable writing praise comments. In this case study, “evaluation” ranked fourth most common type in drafts 1 while it was second most commonly used type for the L2 writers in Tuzi‟s (2004) and Liu and Sadler‟s (2003). However, “evaluation” in this study decreased remarkably in drafts 2 (8.25%) and took the last position in the table of types of comments (table 2). It showed that the students in the second drafts focused more on the quality of their comments, not just as perfunctory work. This finding supports Sullivan and Pratt‟s (1996) and Liu and Sadler‟s (2003) conclusions that students in the computer-assisted classroom demonstrated more focus on their written comments. 5.2 Comments affect the students’ writing quality The researcher employed the triangulation method of trained inter-raters, semi-structured interviews, and writing journals to ascertain validity of the findings. This case study adopted the analytic scoring rubric of Oshima and Hogue (2006) and employed two trained inter-raters as discussed in the rating procedures. Although this analytic scoring rubric is generally not popularly used by researchers, it was objective and proper for inexperienced raters (Weir, 1990; Weigle, 2002). The findings of this study showed that the writing quality of students, compared the means of pretest and posttest, did improve and was statistical significant (P<0.01) by the Paired t-test. The results bolstered the conclusions reached by Sullivan and Pratt (1996) and Braine (1997; 2001) that the writing quality did improve in the computer-assisted classroom from the first to the final drafts. There were some additional reasons indicating the improvements of the writing quality. First, the most common comments types were “clarification”, “suggestion/ advice”, “explanation” and “alteration” which were considered as specific and provided lights for revisions. Second, as the students reported in the semi-structured interviews and their reflections in the writing journals, the process showed that the group members enthusiastically helped one another improve their writing revisions by providing effective comments. R Could I have known that your group members in your group provided comments on your essays enthusiastically to help you revise your writing? S3 Yes. My group included 4 members. Three of them enthusiastically commented on my essays for example thienphuoc... The other one was not unenthusiastic, but I found that her comments were not as qualified as thienphuoc‟s whose comments not only related to grammar, spellings… but also to content. He did very well and I found it helpful. S5 At first, the comments in drafts 1 seemed not very good. But with drafts 2, I found that everyone tried to do his/her best. S11 … My peers in my group commented carefully on my writing which helped me revise my later draft much better than the previous one… S6 I really like to receive comments from my peers in my group. Every time I read the comments from you, though the comments were more negative than positive… S12 ... I learned a lot due to the enthusiastic comments from my peers in my group. I also learned new things from reading my peers‟ essays… Third, the students themselves indicated that they believed their writing quality improved in content, organization and grammar/structure issues from drafts 1 to drafts 3 after blog-based peer response sessions. Their stated beliefs might be the basis for motivation to get them involved in the learning process. R In your point of view, do you think that the quality of your writing improve from draft 1 to draft 3 thanks to the comments of your peers via the blog? AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 21 S3 I especially found the improvement from draft 1 to draft 3. I myself knew that my draft 1 was not good in cohesion. Then thienphuoc showed me [on his comments]. I started to revise some mistakes from draft 2 because my writing lacked supporting ideas. I knew how to state the [main] idea, but I did not know how to support it; so I wrote it to expect my friends to provide comments to help me revise… [my writing now is] good in both content and organization… I found my writing from draft 1 to draft 3 is far more improved‟ S5 So far my writing is much better compared with draft 1. At that time, I wrote as an impulsive act: wrote what I thought. But now I know how to organize my ideas from Introduction to body and the conclusion. S10 I feel that I am able to write faster than before. I know how to organize the ideas, which idea does not support the main idea. That is the biggest mistake that I had since last semester. S1 The fact that after receiving comments from my peer, I recognize that I make a lot of mistakes in grammar and content. I must write 2 or 3 times. Rewriting reduces mistakes. Now my writing is more coherent and smooth. Clearly, thanks to my peer‟s comments, my writing skills are improved. The final reason might come from the positive attitudes in using the blog-based peer response as described in the third research question in the findings and interpretation (No.4). This reason was supported by Hyland and Hyland‟s (2006) contention that electronic feedback through peer response increases student writing output, enhances student motivation, provides a nonthreatening environment, makes papers more readily available for sharing, and allows instructors greater opportunity to monitor peer response. In addition, DÖrnyei (2001) claimed that cognitive approaches place the focus on the individual‟s thought, beliefs, and interpretational processes that are transformed into action. Therefore, when the students preferred this one specific kind of media (blog), they took part in their learning process. 5.3 Students’ attitudes on the use of blog-based peer response Students expressed positive attitudes based on the semi-structured interviews and writing journals. The findings of this study contradicted those of DiGiovanni and Nagaswami (2001) and Tuzi (2004) who concluded that the students preferred traditional peer response to e-peer response. It was worthy knowing the students‟ attitudes on e-peer response in order to get the most collaboration in the learning process. In a “student-centered” approach, the students are considered as the central subjects in the teaching/learning process. There should be high agreement from both the instructor and the students to gain effective results in the learning. Pedagogically, when the instructor obtains a high degree of agreement from his or her students, he/ she would have better chance of successful student collaboration in the classroom activities; hence, the success of teaching would follow. There were three possible reasons for the positive attitudes expressed by students using blog-based peer response. First, the students experienced working in a technological environment. One student writer, bocaukhongduathu (S12), said, “I found it really wonderful: learning via the computer. I feel that I like it! The ways of commenting online to another‟s, my friends correct mistakes [in my writing] for me, I feel like it…”; Another student, bebi (S7) said, “Actually, I feel very comfortable when learning in such a way, not tense at all… Since I have joined this course, I could learn more things, especially about computer.” The other student, Nhoc_HT, (S11) said, “This is the first time I did my writing via the blog and received the comments from you. I am very happy…” Second, many students said that they found it beneficial to receive comments from their peers to revise their later drafts. Lanphuongkt (S11), said, “my peers in my group commented carefully on my writing which helped me revise my later draft much better than the previous one. I appreciated my friends a lot and I hope that we still develop to help improve our writing skills better and better.” She also compared what she could not do in the past and what she could do after this course of Academic writing in her interviews as she replied, “I know how to organize my ideas more easily, knowing which idea do not support the main idea. That‟s the biggest mistake I had since last course.” This finding was supported by Warschauer‟s (2002) that student discussion activities AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 22 brought them to be more “student-centered” , fostered a sense of community, encouraged a sense of group knowledge, and increased student participation…” Finally, the fact that the students found their writing quality improved from the first to the third drafts might result in positive attitudes towards the learning process, specifically in blog-based peer response activities. Two restrictions reported by the students in the writing journal entries and the semi-structured interviews were Internet expense and time-consuming. The restriction on the Internet expense might be resolved in the future since the Vietnamese government (Ministry of Education) was promoting IT in some selective Universities in 2006- 2007 (H.Mai, 2006), and has implemented, gradually in most of Universities in Vietnam since 2008. However, the restriction about time consuming did not seem to be the problem because the findings showed that the students worked hard to provide comments on their peers‟ essays and were rewarded with good quality revisions. That is, it was worth the time. 6. Conclusion This case study found that the students who took part in the blog-based peer response and obtained a 6- step training procedure employed four most frequent types of comments which were “clarification”, “suggestion/ advice”, “explanation” and “alteration”. It indicated that “clarification” should be followed by specific “suggestion” or “explanation” in order to “set lights” for revisions. Second, these 4 common types of comments did significantly affect the students‟ writing quality in mean scores of the pretest and posttest. Finally, this case study found that students expressed positive attitudes toward blog-based peer response activities. These three findings seemed to mutually support one another. Furthermore, the findings indicated that this popular medium of technology, the blog, could be applied in the Academic writing classroom in Vietnam, particularly at Nong Lam University, Ho Chi Minh City. Although the findings of this study could not be generalized to other contexts due to small number of participants and the lack of a control group, writing instructors/researchers could test the strength of the blog-based peer response and the 6-step training procedure in the training in experimental studies. In spite of variety of blogs provided widely on the Internet, the instructor should select the most popular one used by the community of the local students in order to promote the participation in the learning process. Further research also needs to explore how far the students might incorporate the comments in their revisions, which types of comments are most favored by the writers in the revision stage. AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 23 References Berg, E. C. (1999). The Effects of Trained Peer Response ESL Students‟ Revision Types and Writing Quality. Journal of second language writing, 8 (3), 215 – 241. Braine, G. (1997). Beyond word processing: Networked computers in ESL writing classes. Computers and Composition, 14 (1), 45-58. Braine, G. (2001). A study of English as a foreign language (EFL) writers on a local-area network (LAN) and in traditional classes. Computers and Composition, 18 (3), 275-292. Dörnyei, Z. (2001). Teaching and Researching Motivation. Harlow: Pearson Education Ltd. Gartner (2007). Predicts 2007 Blog. Retrieved on Nov. 21st, 2007 at http://weblog.gartner.com/weblog/weblogIndex.php. Hedge, T. (2002). Teaching and learning in the language classroom. Oxford: Oxford University Press. Hyland, K. (2003). Second Language Writing. Cambridge: Cambridge University Press. Hyland, K. & Hyland, F. (Eds.)(2006). Feedback in second Language writing. New York: Cambridge University Press. Hughes, A. (2003). Testing for Language Teachers. Cambridge: Cambridge University Press. Jones et al. (2006). Interactional dynamics in online and face-to-face peer-tutoring sessions for second language writers. Journal of Second Language Writing, 15 (1), 1-23. Lien, K. (2006). Oct. 12th, 2006. Bay nam hoc o pho thong: Tai sao khong biet noi tieng Anh?. Retrieved Nov.17th,2006,from http://www.tuoitre.com.vn/Tianyon/Index.aspx?ArticleID=166558&ChannelID=13. Lightfoot, M. J. (2006) A comparative analysis of e-mail and face-to-face communication in an educational environment. The Internet and Higher Education, 9 (3), 217-227. Liu, J. & Hansen, G. J. (2005). Peer Response in Second Language Writing Classroom. U.S.A: The University of Michigan Press. Liu, J. & Sadler, R. W (2003). The effect and affect of peer review in electronic versus traditional modes on L2 writing. Journal of English for Academic Purposes, 2 (3), 193-227. Luu, Q. Thanh (2006, August, 12th). De ra Kieu Ay Thi Con Nhieu Tai Tieng. Retrieved on June 3, 2007 from http://www.tuoitre.com.vn/Tianyon/Index.aspx?ArticleID=155515&ChannelID=87 Michaels, B. (2007). Quotations from Writers. Retrieved on June 11, 2007 at http://www.kimskorner4teachertalk.com/writing/general/quotes.html#Qideas. Mai, H. (2006, Nov. 4th). Ket noi giao vien voi tuong lai thong tin. Retrived on June 20th 2007 from http://www.tuoitre.com.vn/Tianyon/Index.aspx?ArticleID=170582&ChannelID=13#top. Min, H. T. (2005). Training students to become successful peer reviewers. System, 33 (2), 293-308. Nguyen, T. K. Thu (2002). Error Feedback in L2 Writing Classes: How explicit does it need to be? Master of TESOL research, University of Social Sciences & Humanities, HCMC. Nguyen, T. T. Ha (2004). The use of group work in foreign language classes: Factors leading to success. Master of TESOL research, University of Social Sciences & Humanities, HCM City, Vietnam. Nunan, D. (Ed.), (1993). Collaborative Language Learning and Teaching. Great Britain: Bell & Bain Ltd. Oppenheim, A. N. (1999). Questionnaire Design, Interviewing and Attitude Measurement. London: Biddles Ltd. Oshima, A. & Hogue, A. (2006). Writing Academic English (4th ed.). New York: Pearson Education, Inc. Richards, C. J. & Lockhart, C. (2000). Reflective teaching in second language classrooms. UK: Cambridge University Press. Reid, J. (1995). Teaching ESL Writing. Upper Saddle River, NJ: Prentice Hall Regents in Carter, R. & Nunan, D. (eds.) (2002). Teaching English to Speakers of Other Languages. Cambridge: Cambridge University Press. Rodriguez, R. R. (2003). Computer-Mediated Peer Response and Its Impact on Revision in The College Spanish Classroom: A Case Study. Doctoral Dissertation, College of Arts & Sciences and College of Education University of South Florida. http://www.sciencedirect.com/science/journal/87554615 http://www.sciencedirect.com/science/journal/87554615 http://www.sciencedirect.com/science?_ob=PublicationURL&_tockey=%23TOC%236537%231997%23999859998%23322086%23FLP%23&_cdi=6537&_pubType=J&view=c&_auth=y&_acct=C000054425&_version=1&_urlVersion=0&_userid=5092761&md5=51a5df1bb801391784080dba76a21873 http://www.sciencedirect.com/science/journal/87554615 http://blog.gartner.com/blog/predicts2007.php http://weblog.gartner.com/weblog/weblogIndex.php http://www.tuoitre.com.vn/Tianyon/Index.aspx?ArticleID=166558&ChannelID=13 http://www.sciencedirect.com/science/journal/10967516 http://www.sciencedirect.com/science?_ob=PublicationURL&_tockey=%23TOC%236554%232006%23999909996%23631596%23FLA%23&_cdi=6554&_pubType=J&view=c&_auth=y&_acct=C000054425&_version=1&_urlVersion=0&_userid=5092761&md5=52325a656761c540dfb5573348c41f9b http://www.sciencedirect.com/science?_ob=PublicationURL&_cdi=7302&_pubType=J&_auth=y&_acct=C000054425&_version=1&_urlVersion=0&_userid=1750281&md5=7b76512033965cc371afddf291583b75 http://www.sciencedirect.com/science?_ob=PublicationURL&_cdi=7302&_pubType=J&_auth=y&_acct=C000054425&_version=1&_urlVersion=0&_userid=1750281&md5=7b76512033965cc371afddf291583b75 http://www.tuoitre.com.vn/Tianyon/Index.aspx?ArticleID=155515&ChannelID=87 http://www.kimskorner4teachertalk.com/writing/general/quotes.html#Qideas http://www.tuoitre.com.vn/Tianyon/Index.aspx?ArticleID=170582&ChannelID=13#top AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 24 Stanley, J. (1992). Coaching student writers to be effective peer evaluators. Journal of Second Language Writing, (1), (3), 217-233. Sullivan, N. & Pratt, E. (1996). A comparative study of two ESL writing environments: a computer- assisted classroom and a traditional oral classroom. System, 29 (4), pp. 491 – 501, 1996. Sangarun, P. (2003). Task-based instruction should be implemented in Thai University English Language classrooms. Suranaree Journal of Social Science, Vol. 10 (1); January-March, 2003. Technorati (2006). State of the Blogosphere, April 2006 Part 1: On Blogosphere Growth. Retrieved on Nov. 21st, 2007 at http://www.sifry.com/alerts/archives/000432.html. Thanhnien (2005). Luu but thoi @. Retried on Nov. 21st, 2007 at http://www.tuoitre.com.vn/Tianyon/Index.aspx?ArticleID=78679&ChannelID=7. Tran, Ngoc Bao (2000). A combination of skills & strategies and genre analysis approaches in designing a course to improve magazine writing. Master of TESOL, University of Social Sciences and Humanities, HCM City, Vietnam. Tran, P. L. Phan (2006, August 26th). Đổi mới cách đào tạo giáo viên văn. Retrieved on June 3, 2007 at http://www.tuoitre.com.vn/Tianyon/Index.aspx?ArticleID=158342&ChannelID=13. Tseng, S. C. & Tsai, C. C. (2007). On-line peer assessment and the role of the peer feedback: A study of high school computer course. Computers & Education, 49 (4), 1161-1174. Tsui, B.M. Amy & Ng, Maria (2000). Do secondary L2 writers benefit from peer comments? Journal of Second Language Writing (9), (2), pp. 147–170. Tuoitre (2006, August 11). Loi gioi thieu: Blog 360 do - Nhung sac mau cuoc song. Retrieved on Nov. 21st, 2007 at http://www3.tuoitre.com.vn/Tusach/Book/ArticleView.aspx?ArticleID=200055&ChannelID=371. Tuoitre (2007). Ban Tron: Day Van, Hoc Van: Toi Chang He Bang Hoang. Retrieved on June 3, 2007 at http://www.tuoitre.com.vn/Tianyon/Index.aspx?ArticleID=156949&ChannelID=13. Tuzi, K. (2004). The impact of e-feedback on the revisions of L2 writers in an academic writing course. Computers and Composition, 21 (2), 217-235. Vietnamnet (2007). Yahoo khong man ma voi thi truong Vietnam. Retrieved on Nov. 22nd, 2007 at http://www.vietnamnet.vn/cntt/2007/11/754558/. Vnexpress (2006, August, 11). Bai Van Dat Diem 10 Chep Tu Bai Van Mau. Retrieved on June 3, 2007 at http://vnexpress.net/Vietnam/Xa-hoi/2006/08/3B9ECF84/. Warschauer, M. (1996). Computer Assisted Language Learning: an Introduction. Retrieved on June 4, 2007 at http://www.ict4lt.org/en/warschauer.htm. Warschauer, M. & Turbee, L. and Roberts, B. (1996). Computer learning networks and student empowerment. System, 24 (1), 1-14. Warschauer, M. (2002). Networking into Academic Discourse. Journal of English for Academic Purposes, 1 (1), 45–58. Weigle, C. Sara (2002). Assessing Writing. Cambridge: Cambridge University Press. Weir, J. Cyril (1990). Communicative Language Testing. UK: Prentice Hall International (UK) Ltd. Wennerstrom, A. (2006). Discourse Analysis in the Language Classroom – Volume 2. Genres of Writing. USA: The University of Michigan Press. Zhu, W. (2001). Interaction and feedback in mixed peer response groups. Journal of Second Language Writing, 10 (4), 251–276. http://www.sciencedirect.com/science/journal/10603743 http://www.sciencedirect.com/science/journal/10603743 http://www.sciencedirect.com/science?_ob=PublicationURL&_tockey=%23TOC%236569%231992%23999989996%23322135%23FLP%23&_cdi=6569&_pubType=J&view=c&_auth=y&_acct=C000054425&_version=1&_urlVersion=0&_userid=5092761&md5=3e22c0fe975c1dcad30f51218f479af9 http://www.sifry.com/alerts/archives/000432.html http://www.tuoitre.com.vn/Tianyon/Index.aspx?ArticleID=78679&ChannelID=7 http://www.tuoitre.com.vn/Tianyon/Index.aspx?ArticleID=158342&ChannelID=13 http://www.sciencedirect.com/science/journal/03601315 http://www.sciencedirect.com/science?_ob=PublicationURL&_tockey=%23TOC%235956%232007%23999509995%23666499%23FLA%23&_cdi=5956&_pubType=J&view=c&_auth=y&_acct=C000054425&_version=1&_urlVersion=0&_userid=1750281&md5=093fc1e192064138ddd4e9c88055b4f9 http://www3.tuoitre.com.vn/Tusach/Book/ArticleView.aspx?ArticleID=200055&ChannelID=371 http://www.tuoitre.com.vn/Tianyon/Index.aspx?ArticleID=156949&ChannelID=13 http://www.sciencedirect.com/science?_ob=PublicationURL&_tockey=%23TOC%236537%232004%23999789997%23503311%23FLA%23&_cdi=6537&_pubType=J&view=c&_auth=y&_acct=C000054425&_version=1&_urlVersion=0&_userid=5092761&md5=2b2b900cef4343c16932692680806564 http://www.vietnamnet.vn/cntt/2007/11/754558/ http://vnexpress.net/Vietnam/Xa-hoi/2006/08/3B9ECF84/ http://www.ict4lt.org/en/warschauer.htm AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 25 Appendix A Peer-Editing Worksheet of Cause/Effect Essays Peer editor: …………………………… Date: ……………………….. 1. What kind of introduction does this essay have (funnel, entertaining story, etc.)? …………………………………………………………………………………. How many sentences does it contain? ………………………………………… Does it capture your interest?  yes  no Where is the thesis statement placed? ………………………………………… 2. How many paragraphs are there in the body? Number: …………….. The topics of the body paragraphs are as follows: 1. ……………………………………. 3. ……………………………….. 2. ……………………………………. 4. ……………………………….. (If there are more or fewer paragraphs, add or delete lines.) 3. What kind of supporting details does the writer use in each body paragraph? 1. ……………………………………. 3. ……………………………….. 2. ……………………………………. 4. ……………………………….. 4. Check each paragraph for unity. Is any sentences unnecessary or “off the topic”?  yes  no If your answer is yes, write a comment about it (them) …………………………………………………………………………………. 5. Check each paragraph for coherence. Does each one flow smoothly from beginning to end?  yes  no What key nouns are repeated? ……………………………………………….. What transition signals can you find? ……………………………………….. ……………………………………………………………………………….. 6. What expressions does the writer use to link paragraphs? If there is none, write none. (If there are more or fewer paragraphs, add or delete lines.) To introduce the first body paragraph …………………………………………. Between paragraphs 2 and 3 …………………………………………………… Between paragraphs 3 and 4 …………………………………………………… Between paragraphs 4 and 5 …………………………………………………… To introduce the conclusion: …………………………………………………... 7. What kind of conclusion does this essay have – a summary of the main points or a paraphrase of the thesis statement? ………………………………………………………………………………….. Does the writer make a final comment?  yes  no What is it? ………………………………………………………………………………….. Is this an effective ending (one that you will remember)?  yes  no 8. In your opinion, what is the best feature of this essay? In other words, what is this writer‟s best writing skill? ………………………………………………………………………………….. AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 26 Appendix B Scoring Rubric Maximum score Actual score Content – 2.5 points The essay fulfills the requirements of the assignment. The essay is interesting to read. The essay shows that the writer used care and thought. 0.5 1.0 1.0 ___ ___ ___ Total 2.5 Organization – 4.5 points The essay follows the outline, and it has an introduction, a body, and a conclusion. Introduction: The introduction ends with the thesis statement. Body Each paragraph of the body discusses a new point and begins with a clear topic sentence. Each paragraph has specific supporting material: facts, examples, quotations, paraphrased or summarized information, etc. Each paragraph has unity. Each paragraph has coherence. Transitions are used to link paragraphs. Conclusion: The conclusion summarizes the main points or paraphrases the thesis statement, begins with a conclusion signal, and leaves the reader with the writer's final thoughts on the topic. 0.5 0.5 0.5 1.0 0.5 0.5 0.5 0.5 ___ ___ ___ ___ ___ ___ ___ ___ Total 4.5 Grammar and Sentence Structure-2.5 points Estimate a grammar and sentence structure score. 2.5 Total 2.5 Mechanics-0.5 points Punctuation: periods, commas, semicolons, quotation marks (0.3), capitalization (0.1), spelling (0.1) Total 0.5 Grand Total 10 Adopted from Oshama & Hogues (2006) AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 27 Appendix C GRID FOR COMMENT ANALYSING Types of comments No. of types Global Areas Local Areas No. of areas Revision-oriented Non-revision- oriented Revision-oriented Non-revision- oriented Evaluation Generally This paragraph is too short to prove your idea. You have a clear writing, I like it. This sentence does not make sense I like this sentence a lot Specifically Your thesis statement is not very good. "You have 3 clear supporting sentences." This word "…" is not clear for your meaning here Your grammar structure of this sentence is nice Clarification Specific ideas I don‟t understand what you mean in “For me, who have been living in a city all my life” No example for this category No example for this category No example for this category Particular word choices, phrases, or sentences No example for this category No example for this category “What do you mean by "night school”? - This sentence is wrong in grammar. No example for this category Cohesive “You say „…‟ How does this sentence connect to the one before?” No example for this category These two sentences need a transition signal. No example for this category Unity I think this part is off topic. You are talking about “chance for education”, why do you talk about transportation? Your supporting ideas are connected to the topic sentence No example for this category No example for this category AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 28 Alteration Your thesis statement should be "Living in a big city brings us many benefits" No example for this category “their” should be changed to “our” No example for this category Suggestion/ advice Generally In each benefit, you should give us more convinced ideas to support your writing. No example for this category I think when you write a paragraph or an essay you should use simple word. No example for this category Suggestion/ advice Specifically you should support for this idea, don‟t talk about another idea. No example for this category I think that you should use an adjective there as “social problems”, and “electronic libraries”. No example for this category Explanation I think these two sentences should be reduced because they talk about the conveniences of computer, not about the good education in a big city. No example for this category You should change "Despite … into Although" (Despite + N/N phrase, Although + clause)'. 'I think "so " is used to connected two clauses.' No example for this category confirmation Are you sure all people prefer living in a big city? You have a thesis statement, and topic sentences. “wait for you or will wait for you?” No example for this category Statement No example for this category This is just my opinion, I hope it will help you a lot. I understand what you mean. No example for this category No example for this category Total AsiaCall Online Journal (ISSN 1936-9859) Vol. 4 No. 1 October 2009, Pham Vu Phi Ho and Siriluck Usaha 29 Appendix D Questions of the Semi-structured and in-depth Interviews I will ask a few questions about the peer response activity in which you gave responses to your peers, and you received comments about your writing as well. In addition, the use of the blog for peer response is also in the consideration. Apart from my guided questions, you can provide more of your ideas if you feel necessary to express your thoughts. Please give me all your thoughts on each question. 1. Do you think that peer response is helpful to you? 2. So far you have learned how to post writing on the blog, provide comments on peers‟ essays, and revise your writing. What do you think about it? 3. Do you expect your friends to comment on content, organization or on grammar, structure, and spelling? 4. Do you think that the quality of your writing has improved through draft 1 to draft 3 thanks to the comments of your peers via the blog? 5. Do you like to post your writing on the blog for your friends to read and comment? 6. Do you learn any thing from your peers when you read and provide comments on your peers‟ essays? 7. Are your peers' comments useful to you for revising your essay? 8. Could I have known that your group members in your group provided comments on your essays enthusiastically to help you revise your writing? 9. What do you focus on when you write your comments? 10. What areas (global or local) do you prefer to provide comments as well as to receive comments from your peers? 11. Do you benefit from giving comments to others? If so, what are the benefits? If not, why not? 12. if you have two options: (1) the teacher comments on your essay only without peer comments; (2) peer comments then the teacher comments, which one do you prefer? 13. Would you like it if there were only peer comments but not teacher comments? Why? 14. Do you usually understand your peers‟ comments and corrections? 15. What do you do if you do not understand your peers‟ comments? 16. Does your teacher/peer give you positive or encouraging comments? 17. Do you feel that your peers‟ comments have helped you to succeed in this course and improve your writing? Why or why not? 18. In what way do you wish that your peers would change or improve their comments? 19. What do you think of the blog-based peer response? 20. Do you often read your peers‟ comments and corrections via the blog? 21. Is it convenient to provide comments on your peers‟ essay via the blog? 22. Is there any inconvenience when you post your essay via a blog? 23. In the future, when you finish this course, will you still make use of your blog to post your writing?