The Incorporation of Facebook-Based Peer Comments Into Writing Revisions: A Framework for Social-Network Peer Commentaries

This study determined the degree to which students used peer feedback from Facebook to revise their writing papers and explored the students’ perspectives on a framework of using a social network for peer commentaries. The study enrolled two intact groups, 40 students in the control group and 32 students in the experimental group, at a university in Ho Chi Minh City, Vietnam. The student’s initial and final drafts, peer reviews, and semi-structured interviews were used to collect data. The results indicated that 45% of changes were made in response to peer reviews, while 55% were made solely on the basis of the student’s own judgments. Additionally, the overall number of revisions made by students was greater than the number of revisions caused by comments at higher levels such as “Clause,”“Sentence,” and “Paragraph.” Numerous previous studies refuted the current study’s results, raising questions for researchers/instructors considering using Facebook as a forum for their writing classes because the students become more responsible for developing their writing skills.


Introduction
Peer feedback, peer comment, or peer review, in general, constitutes an important aspect of improving writing (Pham, 2021a(Pham, , 2021b;;Pham, Huyen, et al., 2020b).It is an effective pedagogical tool to encourage learners to facilitate further writing development.By engaging in peer feedback and receiving comments from other peers, students can be exposed to broader readers' views than just those of their teachers.According to Liu and Hansen (2002), employing peer feedback activities in the writing classroom brings the students to the roles of teachers for each other, and they take responsibility for their own learning.
In addition, when learners read their peers' writing papers to provide comments to help improve each other's writing quality (Pham, 2021b), they attempt to be critical readers; the students feel confident in their ability to comment on their peers' work (Berg et al., 2003).Vu et al. (2022) stated that peer feedback activities helped increase online interactions between students and teachers and helped them become more confident in the learning process.Min (2008) states that peer feedback allows reviewers more time to form explicit comments in English and allows writers to reconsider their peers' comments as often as they want when revising their drafts.Peer feedback can influence the revisions by the students' writers (Pham, 2021b).Pham and Usaha (2016) found that the total number of revisions made by the student writers was larger than the total feedback delivered by peers, contributing to improving students' writing quality.
On social networking sites such as Facebook, posts and comments are displayed in chronological order, making it convenient for students to read and comment on each other's drafts.In order to enhance the students' engagement in peer feedback activities, researchers around the world employ media or social networks in their writing classrooms, and Facebook is not an exception.Living in an era of industrial 4.0, Facebook offers students a myriad of opportunities for students to practice and enhance their writing skills.First, concerning collaboration, Facebook offers excellent potential for collaboration among learners since the learning platform on Facebook can engage learners in creating knowledge and participating in the learning process (Freishtat & Sandlin, 2010).Also, Facebook promotes the level of interaction both between the teacher and students and between students, making it possible for teachers to discuss assignments with their students outside of the classroom (Yunus et al., 2012).Similarly, Tran (2019) also claimed that Facebook is a tool to increase the interaction between students and teachers.It helps the teachers connect their students for assignments and improve students' performance (Chugh & Ruhi, 2018).
Facebook will benefit language classrooms by facilitating communication and interaction.Facebook, according to O ¨zdemir (2017), is the simplest method to communicate with others and could be used to improve intercultural communication effectiveness.O ¨zdemir discovered that the Facebook discussion group performed significantly better than the in-class discussion group.Moreover, according to Cerda`and Planas (2011), using Facebook in the classroom is advantageous because it encourages student interaction, online discussions, and group work, particularly among students who are hesitant to participate in face-to-face activities.Wang and Va´squez (2014) state that Facebook could be used to help L2 learners enhance their writing performance by allowing students to practice writing outside of the classroom.In addition, Bo¨rekci and Aydin (2020) suggest that EFL instructors establish a positive environment for Facebook interaction to support their student's academic and personal development.Consequently, it is an excellent instrument for students to provide peer feedback.Online peer remark activities on Facebook provide an interactive environment for EFL students to enhance their writing by exchanging peer comments and revising drafts.
Second, the students can have a broader audience when they post their writing on a social networking site like Facebook fosters their sense of audience.According to Yunus et al. (2012), students are likely to take accountability for their words since they have a larger audience when posting their drafts on Facebook.Also, by commenting on a peer's draft on Facebook, students can reach a wider audience, encouraging them to put more effort into writing high-quality comments.This also motivates shy students to actively participate in peer comment activities (Yunus et al., 2012).Third, the fact that students can access a variety of drafts, peer comments, and teacher comments on Facebook enables them to receive enough input in order to develop their writing skills (Yunus et al., 2012).Therefore, Tran and Van Nguyen (2020) state that flipped classroom with Facebook will engender students' engagement in the learning process.
In short, using Facebook in blended learning can facilitate students' writing process in higher education by supporting collaborative learning, creating motivation, and providing great input.However, engagements in learning activities in this kind of social media, such as discussion, interaction, or communication, seemed not enough if the researchers failed to investigate the language production, in this study, writing revision, that Facebook might facilitate (Pham, 2016).The purpose of the current study would fill in this gap.

Literature Review
Numerous studies have been performed on qualified peer comments and their effect on the revision and writing quality of ESL/EFL students.Also, students' perceptions of the use of Facebook peer feedback were also explored.Regarding incorporating peer comments into revision, Min (2006) experimented to determine which essay types yielded better results.All participants were sophomores.When training students, results showed that they incorporated a lot of advice into their revisions, accounting for 77%.After participating in peer feedback, the researchers found it beneficial to their revisions and were willing to incorporate it into their program (Min, 2006).
Although it is claimed that peer feedback is the case of a blind leading a blind and that peer feedback quality is doubtful due to students' limited knowledge and competence in generating comments, a number of studies have found evidence that peer comments are, to some extent reliable and beneficial in EFL teaching and learning contexts.Pham (2021a) designed a model of commentaries to train students to be effective reviewers in every teaching section.After learning the teachers' model of commentary, the study found that the students tended to produce better peer feedback to trigger revisions.Remarkably, the students claimed no statistically different effects between the lecturer's feedback and peer feedback on the student's writing revisions.In order words, the students' feedback is as beneficial to the student writers as the teacher's feedback.Hence, quality peer feedback should be addressed in other studies.
As the implementation of peer feedback is still controversial, several studies on the impacts of peer feedback on revisions have been done to work out whether peer comments are beneficial in EFL teaching and learning contexts by examining their impacts on revision and writing performance.Pham and Usaha (2016) employed a blogging platform to help students provide feedback to help each other improve their writing quality.The study also investigated the ratios that the students incorporated peer feedback via the blog into their writing revision.The study found that the total revisions that the students made were more than the total feedback provided by peers.The study also pointed out that the writers' revisions at lower levels, such as word or phrase, didn't need much help from their group members, whereas those at higher levels, such as sentences or paragraphs, needed much help from peers.
Recently, as part of the current project, Pham, Phung, et al. (2020a) investigated whether Facebook peer comments affected students' writing quality and whether they were more effective than the traditional peer comment mode.Seventy-two native Vietnamese students from two distinct curricula at HCMC University of Science participated in the study.The only difference between the two groups was that one received peer feedback on papers in person and the other via the Facebook social network.The pre-and post-tests were assessed by inter-raters and analyzed using t-tests in the SPSS software.Both traditional and Facebook peer comments significantly impacted students' writing quality, but Facebook peer comments were more effective than traditional peer comments.However, this investigation failed to explain why the student papers were distinct.The current study should be expanded to include an in-depth analysis of students' writing revisions and qualitative data to examine students' opinions after receiving feedback on Facebook.
After determining that few studies had conducted indepth investigations of incorporating peer feedback into revision, Pham, Huyen, et al. (2020b) conducted a study at a university in Ho Chi Minh City, Vietnam, to fill this gap.The study included 92 English-major students from two academic writing classes.In 11-week courses, they were taught how to provide feedback to their peers.According to the study's findings, the majority of peer feedback was revision-oriented; thus, the majority of the feedback was incorporated into the students' writing revision.Their writing products become of higher quality as a result.This study did not investigate the students' perspectives on peer feedback activities.
Due to its popularity and its high potential for being applied in teaching English writing, researchers have conducted many studies exploring the effects of Facebook as a platform for peer comments on improving students' writing quality.Researchers in Asia, such as Malaysia, Thailand, Taiwan, and Vietnam, conducted most previous studies on the application of Facebook to teaching English writing.These studies mainly focused on whether exchanging peer comments on Facebook affects students' writing quality, students' attitudes, and the extent to which students incorporate peer comments into revisions.
Concerning students' views on using Facebook as a tool to assist them in improving their writing performances, several studies in Asia, specifically in Malaysia, Thailand, Taiwan, Japan, and Vietnam, have been undertaken to explore the students' perceptions.Yusof et al. (2012) examined the benefits of using Facebook Notes for guided peer feedback throughout the writing process.The research comprised 20 Malaysian students.The students were taught how to use Facebook peer commenting.The students were encouraged to comment on Facebook.Students' Facebook posts and interview responses were analyzed.The study found that the students in a mixed-ability class gained positive attitudes toward using Facebook as a tool in the writing classroom.They regarded the Facebook peer comments as a way to strengthen their pre-writing and self-editing abilities.They liked the activities because they are usually online.This research focuses on peer comments throughout the academic writing preparation stage.The study took 4 weeks.These flaws may have skewed the study's conclusion.Wichadee (2013) investigated the impact of Facebookbased peer comments on students' writing ability, attitudes toward using Facebook to provide peer comments and the extent to which comments were incorporated into revisions.Thirty EFL freshmen from a private Thai university took part in the study.The study's findings revealed that Facebook-based peer comments significantly improved the revised drafts.The findings also revealed that students' peer comments were primarily focused on content rather than grammatical errors.Most frequently, spelling, tenses, and content comments were incorporated into revisions.The interview analysis revealed that students have a positive attitude toward using Facebook to deliver peer comments.Similarly, Sukhwartnarueput and Wasanasomsithi (2012) also found that the students enjoyed interacting with friends and professors through Facebook.Razak and Saeed (2015) studied peer writing revisions in a Facebook group to determine what kind of contributions students' peer feedback had on their written compositions outside of the classroom.Fourteen international students wrote a paragraph, responded to the other students' drafts, and made revisions to their own based on peer feedback on Facebook.According to the research, the students added, substituted, permuted, and consolidated most of the changes.The activities also showed that students felt an interactive, well-established, and educated community based on shared interests and participation.
Dizon (2016) conducted a quasi-experimental study to compare the effects of writing on Facebook versus writing on paper and pencil on improving L2 writing fluency, lexical richness, and grammatical accuracy among university students in Japan.The results of the data analysis revealed that the experimental group made more Pham and Phung significant gains in terms of writing fluency.Neither group made significant progress in lexical richness or grammatical accuracy.Because this study compared the effects of Facebook-based teacher comments versus penand-paper teacher comments on students' writing performance, peer comments were not included.
To find out more about the advantages of utilizing Facebook Groups to teach writing to Vietnamese high school students, Nguyen and Dao (2015) involved 55 11th graders from a high school in Ho Chi Minh City, Vietnam.The experimental group used Facebook to share assignments and communicate with classmates and lecturers.Pre-and post-test surveys collected data.Instructors and classmates were asked to participate in English conversations and post homework on Facebook.The study found that using a Facebook group improved students' writing skills and attitudes toward using Facebook for educational purposes.Likewise, Phuong and Phuong (2016) in Vietnam also claimed that students favor using Facebook for peer comment exchange.

Research Gaps
Concerning the previous research's emphasis, although there are several studies on computer-mediated peer comments and the impact of Facebook on students' writing quality (Jiang & Ribeiro, 2017), there are few studies on how Facebook peer comments affect non-English majors' revision.The majority of the previous study on peer criticism on Facebook has been conducted using quantitative or qualitative methodologies.In the context of research in Vietnam, though the results about peer feedback in technology environments had possible effects on students' writing revisions (Pham, 2021b(Pham, , 2021a(Pham, , 2023;;Pham & Usaha, 2016;Pham et al., 2020aPham et al., , 2020b)), few studies have explored students' perspectives about this area.Also, few studies have compared the impact of Facebook peer comments on students' work to that of other kinds of peer evaluation.Ferris (2003) advised addressing the contradictory results of peer comments.Numerous research on this area lacks data gathering and analytic triangulation.
If students are well instructed in the peer feedback process, they will help their peers in revision sessions and enhance their writing performance.Studies on the impact of peer feedback on revisions and writing performance have yielded both positive and negative results due to different contexts and ways peer feedback is carried out.In the local context, students' writing skills were seen as a lack of motive in writing, and they had to struggle with writing (Nguyen & Nguyen, 2022;Pham, 2021aPham, , 2023;;Tran, 2021).They had limit vocabulary and ideas to develop their writing skills (Ngo & Tran, 2021).The present study is initiated by the positive findings from previous research about the levels of revisions made thanks to peer feedback.The study is conducted in an attempt to further those studies by working out the ratios and the levels of revisions triggered by peers to examine the effects of peer feedback on revisions in a Vietnamese learning context.Additionally, the students' perceptions of these activities are examined.

Research Methodology
The sample for this study was drawn using convenience sampling from HCMC University of Science (HCMC US), one of the member units of Ho Chi Minh City Vietnam National University, one of the largest universities in Vietnam with seven member units (Vietnam National University HCMC, 2021).The sample included 72 individuals from a population of 1,000 freshmen, all of whom were studying English-2 and belonged to two entire classes between the ages of 18 and 20.The researcher used convenience sampling since she was supposed to teach these particular courses that semester.As a result, the courses selected were based on the researcher's planned teaching obligations, enabling simple access for the study.Due to their availability at any given moment, the two chosen classes attended English-2 concurrently, enabling the researcher to gather data from both courses.Their pre-intermediate English proficiency corresponded to the Common European Framework of Reference for Languages level A2 (CEFR).They had been instructed to write a descriptive paragraph in English-1 but not in email writing at the time of this research.They were expected to spend a total of 60 hr in their English-2 course over the course of a 15-week semester honing their four English skills.In terms of writing abilities, students were to be taught how to compose an email in response to a written request.
The researcher randomly assigned one English-2 class with biology majors to the control group and another class with information technology majors to the experimental group.Both classes had an equal probability of being randomly assigned to the control or experimental group.Participants were not compelled to join either the control or experimental groups.They may choose to participate in either group: if they were disinterested in the medium used to exchange writing and peer comments, they could request permission from the instructor to join the other group.Individuals were sorted into four groups in both the control and experimental groups, although the technique for grouping participants differed between the control and experimental groups.In the former, participants were permitted to pick their own groupmates in order to maintain close closeness while exchanging drafts and comments.In the latter, the researcher randomly divided participants into groups since they would be exchanging remarks over Facebook.
This was a quasi-experimental study with pre-and post-test control groups (Cohen et al., 2007;Creswell, 2009).This is a common quasi-experimental design in which the experimental and control groups are chosen arbitrarily.Two groups are administered a pre-and posttest, but only the experimental group receives the treatment.This research design aided in examining the impact of Facebook-based peer comment activities on students' writing quality and presented a causal analysis of the variables.A pre-test and a post-test were administered at the beginning and conclusion of the course, respectively, to ensure that all participants were at the same level prior to treatment and to evaluate their efficacy.The independent variables are peer remark activities via Facebook versus paper-and-pen, and the dependent variable is students' English writing quality.Figure 1 presents the research design of the current study.

Assigning Students to Peer Commenting Groups
The size of peer commenting groups was carefully monitored in this study.A group's size, say Liu and Hansen (2002, p. 62), ''may affect any group task.''Students' levels of skill and maturity also have a role.Many studies have discussed the pros and cons of big peer commenting groups.A large group benefited both professors and students.On the one hand, larger classes are easier to manage (Liu & Hansen, 2002).However, large group members can gain significant advantages from their peers.In a larger group, students must read and comment on more drafts, allowing them to learn more from their peers.More groups have a wider variety of talents, knowledge, experience, and skills, making them better for various activities (Liu & Hansen, 2002).A bigger peer commenting group also allows for greater social engagement and meaningful negotiation.
Large groupings have many advantages but also disadvantages.First, both individuals offering and receiving peer critiques may perceive the disadvantages of large groups.According to Lin and Warden (1998), larger group peer comments were less helpful because members were more inclined to make superficial remarks with no explanation or rationale.Giving peer feedback on too many pieces of work may be draining for many students.Second, instructors struggle to ensure equitable involvement in big groups.In fact, larger groups have less equal Pham and Phung member participation (Liu & Hansen, 2002).Third, large gatherings may cause problems.Wheelan and McKeage (1993) claim that bigger groups suffer from disputes and clique formation because members are less able to comprehend one another.Also, physical distance complicates close family relationships.
Taking into account the benefits and drawbacks of large groups, the researcher divided the students into two groups of four (Pham, 2023).However, the larger minimum group size of three members ensured that each group operated efficiently, even if one or more members dropped out of the course or ceased participating in peer commenting activities.In addition, throughout the semester, students remained in the same peer commenting groups to continue learning about their peers.
The number of students in each peer commenting group was the same in both the control and experimental groups, but the researcher allocated students differently in each group.In the control group, students formed peer commenting groups at their own discretion.This ensured that all students got along in their groups and could easily gather after class for peer feedback.The teacher created a Facebook group for the experimental group and made sure everyone joined.The group was set as private so that only the students in the group could see the entries each other.Students were then placed in peer commenting groups based on alphabetical order in the class roster.Using an asynchronous platform like Facebook, students in these peer commenting groups were less likely to be affected by proximity or personality differences.

The Pedagogical Rationale for Utilizing Facebook as a Platform in an EFL Writing Class
Facebook provides several possibilities for kids to improve their writing abilities.First, in terms of collaboration, Facebook is a great place for learners to collaborate and create knowledge (Freishtat & Sandlin, 2009).Also, Facebook encourages interaction between teachers and students, allowing teachers to discuss assignments with students outside of the classroom (Yunus et al., 2012).Pham (2016) claims that Facebook cooperation helps students produce language.Online peer comment activities on Facebook allow EFL students to improve their writing skills by exchanging peer feedback and revising their own drafts.Second, Facebook affects student motivation (Yunus et al., 2012).The fact that students can reach a larger audience by posting their work on Facebook fosters their sense of audience.Students who post their drafts on Facebook have a larger audience, according to Yunus et al. (2012).With Facebook comments, students may reach a larger audience, which pushes them to write better comments.This encourages hesitant students to peer comment (Yunus et al., 2012).Third, students may access a variety of drafts, peer comments, and instructor comments on Facebook, allowing them to improve their writing abilities (Yunus et al., 2012).Students can view and comment on others' drafts on social networking sites like Facebook since postings and comments are chronologically ordered.In summary, using Facebook in blended learning may help students write better by encouraging cooperation, motivating them, and offering plentiful feedback.

The Writing Cycle of the Training
The researchers utilized Pham, Phung, et al.'s (2020a) writing assignments for each writing assignment.Each writing cycle lasted two weeks.The instructor began by introducing both classes to the writing topic.In addition, she introduced the students to the topic and its associated terminology.In addition, she instructed the students to compose their responses to the emails, the genre of the writing assignments.The students were then instructed to complete their first English writing assignment on a paper sheet during class.Then, they were instructed to revise their manuscripts before distributing them to colleagues for feedback.Students in the experimental group were then required to post their English writing on the group's Facebook page for their peers to comment on.For the instructor to know who commented on which posts, the students used their identities in the Facebook group.The preponderance of posts and remarks were composed in the English language.The students in the control group were instructed to write their comments on the paper and then meet in person to share them.Students in both courses were then instructed to revise their manuscripts and either submit them to the instructor or post them on the Facebook page if they were in the experimental group.The instructor reviewed and commented on the manuscripts of the students in the control group, rectifying errors and making written notes.The instructor commented on the Facebook postings of the experimental group by composing her remarks directly beneath the students' posts in the ''Comment Section.''Both students were then required to revise their manuscripts and compose their final essays.
This cycle was repeated three times, corresponding to the course's three writing assignments.

Revision Analysis Rubric
The researchers gathered students' drafts and critiques from the experimental group for content analysis.The researchers analyzed the experimental group's first and second versions and peer comments to determine the extent to which students incorporated peer feedback into their revisions.To analyze the content of the experimental group's drafts and comments, the researchers repeatedly copied students' writing and comments from the Facebook Group and saved them in a Word document in case students deleted their posts and comments or even removed their Facebook account from the Facebook Group.
The amount to which peer comments were absorbed into revisions was determined by a content analysis of students' first and second versions and peer comments.Robson and Kieran (2016, p. 349) describe content analysis as ''the quantitative examination of what is included in a document.''Cohen et al. (2007) define content analysis as a three-step method.The first phase comprises segmenting the text into analysis units, coding, and classifying the analysis units.The second phase is comparison, which entails comparing categories.The last stage is a conclusion, during which the researchers draw a theoretical conclusion from the text.
The researchers employed a revision analysis rubric to make steps 1 and 2 of the study more methodical and manageable.The rubric for revising was adapted from Pham's (2014, p.77) ''Coding approach for textual revision.''He developed this coding system to ascertain the percentage of students that incorporate peer feedback from blogs into revisions.This criterion aided the researchers in identifying whether the changes made in their second versions were totally due to peer comments, somewhat due to peer comments, or not due to peer comments at all.Eight types of modifications were made: punctuation, spelling, grammar, expression, phrase, clause, sentence, and paragraph.

Feature ''Compare Two Versions of a Document'' of Microsoft Word
While content analysis can be a taxing and timeconsuming process, computerization can help alleviate the burden of the task and bring tremendous benefits (Robson & Kieran, 2016, p. 256).For example, with the Pham and Phung help of computer software, researchers can easily manipulate and display text in various ways.Therefore, in the current study, in order to compare the revision made by the student writers after receiving Facebook-based peer comments with their first drafts to answer research question 2, the researchers made use of the function ''Compare two versions of a document'' of Microsoft Word together with the revision analysis rubric.Figure 2 presents the Facebook group interface.
After collecting students' drafts and comments and saving them in the form of Microsoft word documents, the researchers used the function ''Compare two versions of a document'' of Microsoft Word by choosing ''Review''!''Compare''!''Comparetwo versions of a document'' on the Menu Bar.The reason why the researchers made use of this function to help her to analyze students' drafts and comments to answer research question 2 was that it makes the coding process less laborious since it highlights every single change, including formatting, inserting, deleting, etc. that a writer made on a document.Figure 3 presents the function of ''compare'' two versions of documents in Microsoft Word.

Data Collection and Analysis
The manuscripts, first and second drafts, and semistructured interviews of all students were compiled for analysis.To answer the first research question, the researchers contrasted the writing performances of the two groups to determine the extent to which peer comments influenced the students' revisions of their writing.
The IIG Vietnam (2017)'s 10-band scoring rubric (see Appendix) was used to grade students' pre-test and posttest papers in this investigation.Two of the raters in this study were colleagues of the researcher, who taught English at HCMC University of Science.Before participating in this study, both were instructed by IIG Vietnam on how to utilize the grading rubric.The correlation coefficient between the control group's pre-test and post-test scores were r = .729,p = .000,r = .868,p = .000;and that of the experimental group's were r = .808,p = .000;and r = .840,p = .000.Second, the researchers used the function ''Compare two versions of a document'' of Microsoft Word to compare students' first and second drafts for each writing assignment.All the changes, including insertions, deletions, moves, and formatting, were highlighted on Microsoft Word.Then, the researchers used the Revision analysis rubric to categorize the changes made in the second drafts into eight corresponding levels in the rubric.
Next, the researchers carefully compared the peer comments and the changes made to the second drafts to see whether the changes were made based on peer comments, partly based on peer comments, or not based on peer comments.If a change was made exactly following the suggestion in a peer comment, that change was coded as ''based on the comment''; if a peer comment triggered a change, but it did not adopt the suggestion in the peer comment, it was coded as ''partly based on the comment''; if a change was made without any suggestion from peer comments, it was coded as ''not based on the comment'' (Pham & Usaha, 2016)  Finally, the researchers added the total number of changes made to the second draft and the total number of each type of change to report the results in a summary table.The researchers also used descriptive statistics to represent the percentages of each type of change made to the second draft.
To respond to the second research question, qualitative data collected from the semi-structured interviews were analyzed to answer this research question.Sixteen students from the experimental group were invited to participate in the semi-structured interviews.The interviews took place during the during second-half till the end of the course to explore in-depth information from the participants.Since the interviews were conducted in Vietnamese, the researchers laboriously translated the interview scripts from Vietnamese into English.First, the researchers read the transcripts carefully a few times in order to pick up the thread running through these interviews and highlight appropriate ideas.Then, the researchers read the transcripts again and labeled highlighted ideas with codes.The translated transcripts were also informed to the interviewees to make sure all the meanings of their opinions remained or if they wished to change some information.Next, she crumbled these codes into twelve themes.All these steps were done on the computer using the word processor Microsoft Word.

Results/Findings
Research question 1: Do Facebook peer comments impact students' writing performances?If yes, to what extent do the students incorporate Facebook-based peer feedback into their revisions?
The main purpose of the first research question was to investigate the ratios of incorporating Facebook-based peer comments into students' writing revision.In order to respond to it, first, comparisons of the student's writing outcomes were made to investigate whether the peer comments on the two modes impacted the students writing revision.Then, only comparisons of the first and second drafts of the students in the experimental groups were analyzed.Forty written papers of the control group in the pre-test were compared to 32 papers from the experimental group.
In order to compare the mean scores of the two groups, tests of normality were run to see if the scores of the students in the pre-and post-tests were normally distributed.If the null hypothesis was not rejected, parametric tests would be employed.On the contrary, the non-parametric tests would be used.A Shapiro-Wilk's test (p ..05) for the pre-tests of the two groups indicates that the pre-test scores were approximately normally distributed for both the control and experimental group, with a Skewness of 20.372 (SE = 0.374), and a Kurtosis of 20.642 (SE = 0.733) for the control group, and a Skewness of 20.287 (SE = 0.414), a Kurtosis of 21.135 (SE = 0.809) for the pre-test of the experimental group.This indicates that the comparison of the pre-tests between the two groups could be run with a parametric test.However, Shapiro-Wilk's test (p \ .05)for the posttest of the control group indicates that the post-test scores of the control group were not approximately normally distributed, with a Skewness of 20.795 (SE = 0.374), and a Kurtosis of 1.520 (SE = 0.733) for the control group.In this case, the Mann-Whitney U Test of the non-parametric test was run to compare the results of the post-tests of the two groups.Table 1 presents the comparisons of the student's writing performances in the pre-tests between the two groups.
The results of the independent sample T-test are shown in Table 1.The mean score on the pre-test for the control group is M = 5.730 (SD = 0.872), while the mean score on the pre-test for the experimental group is M = 5.903 (SD = 0.621).The control group's average pre-test score (M = 5.730; SD = 0.872) is lower than the experimental group's (M = 5.903; SD = 0.621).The mean difference between the two groups' pre-test scores is MD = 20.173.However, the independent sample t-test analysis indicates that the pre-test scores of the two groups are not significantly different (p = .329).Table 2 compares the students' writing performances in the post-tests between the two groups.As a result, there is no statistically significant difference in students' writing quality in the control and experimental groups prior to treatment.This conclusion was made to facilitate any further conclusions concerning the impacts of peer commenting on Facebook and penand-paper peer commenting activities.
As mentioned earlier, the scores of the post-test of the control group were not approximately normally indicate that there is a significant difference in post-test scores between the control and experimental groups.The p-value of .015 is below the commonly employed significance threshold of 0.05.Therefore, the observed difference between the control and experimental groups is statistically significant.In other words, the data analysis results indicate that Facebook-based peer comment activities were more effective at improving students' writing quality than paper-and-pen peer comment activities.
The following section presents the analysis of the ratios of incorporated Facebook peer comments into the writing revision.
The first and second versions of 72 responding emails were analyzed to answer this research question.With the help of the feature ''Compare two versions of a document'' of Microsoft Word, the researchers counted any single change from the first version to the second version of 72 responding emails and categorized the changes into categories in the Revision analysis rubric.The total number of changes from the first to the second was 1,368.
Table 3 illustrates the specific incorporation number according to different linguistic unit levels and whether the students' writers incorporated the peer comments into their revision.
As can be seen from Table 3, there were a total of 1368 changes, of which 274 changes (20%) were made based on comments (incorporation), 340 (24.9%) were partly based on comments (partly incorporation), and 754 (55.1%) were not based on comments.On average, 15 changes were made on each draft (Mean = 14.87).In other words, there are more changes in the second drafts made by the student writers themselves than the changes based on peer comments and those partly based on peer comments.
The most frequent level of revision was ''Word'' (n = 336; 24.6%), in which changes based on comments accounted for 22.9% (n = 77), changes partly based on comments accounted for 13.1% (n = 44), and 64% of the changes were not based on peer comments.It is noticeable that the total number of changes triggered by comments was far lower than the number of changes made by the student writers themselves, with 121 changes compared with 215, respectively.
The second most frequent level of revision was ''Grammar'' (n = 229, 16.7%), in which 22.7% of the changes (n = 52) were based on comments, 37.1% of the changes (n = 85) were partly based on comments, and 40.2% of the changes (n = 92) was not based on comments.Unlike the ''Word'' level at which the student writers made the majority of changes, students made most of the grammar changes in their second drafts with the help of peer comments.The changes triggered by peer comments (n = 137) outnumbered those made by the student writers themselves (n = 92).
The third most frequent level of revision was ''Phrase'' (n = 212; 15.5%), in which 14.6% of the changes (n = 31) were based on comments, 35.8% (n = 76) of the changes were partly based on comments, and 49.6% of the changes (n = 105) were not based on comments.It is evident that at the ''Phrase'' level, nearly half of the changes were triggered by peer comments, and the student writers themselves made the other half.It is also apparent from Table 3 that the number of changes related to the two levels, ''Punctuation'' and ''Spelling,'' ranked the lowest among the eight levels reported.On the contrary, comments on the ''Word'' level and ''Grammar'' levels seemed to influence revisions the most.It is worth noticing that at macro levels such as ''Clause,'' ''Sentence,'' and ''Paragraph,'' the number of changes made by the student writers themselves exceeded the number of changes triggered by comments.This indicates that after obtaining the Facebook-based peer commentary activities, the students took more responsibility for their studies and paid more attention to the improvement of their writing products.
Research question 2: What are the students' perspectives on using Facebook-based peer commentary activities in studying writing in terms of preferences, utility, and impacts?
To address this question, qualitative data from semistructured interviews with all experimental group participants were analyzed.The themes of the interviews included (1) preferences for using Facebook for commentary activities in the writing classroom, (2) the utility of Facebook-based peer comments, and (3) the impact of Facebook-based peer comments on students' writing revisions.

Students' Preferences for Using Facebook for Commentary Activities
First, most of the experimental group students were interested in connecting and discussing their writing with their friends via Facebook.They preferred reading and commenting on their peers' drafts via Facebook to via paper and pens because ''Giving comments over Facebook using laptops or smartphones was much more convenient than writing up a comment on a piece of paper.''I think conducting these activities made the course more interesting than the traditional course.I did not meet any difficulties; in contrast, I found the course less burdensome because I could post my writing and comment on my peers' writing any time before the deadline.
The students also liked their friends to read and comment on their writing via Facebook.Some students particularly stated that they liked Facebook-based peer comment activities because they helped them spend meaningful time on Facebook rather than spending their time on another social network.In other words, Facebook-based peer comments were more time-efficient and could help more peers; the students' handwriting was illegible.
When a student posts his writing on Facebook, his writing mistakes can be pointed out right away by his peers and corrected in a minute.In contrast, if a student handed in his writing to his teacher, those mistakes were usually not corrected right away since teachers tended to correct only mistakes made by a majority of students.

The Utility of Facebook-Based Peer Comments
Second, regarding students' perception of the usefulness of Facebook-based peer comment activities, the majority of the students perceived the usefulness of Facebookbased peer comment activities.They agreed that giving peer comments via Facebook was convenient, and exchanging peer comments enabled them to study English writing in a more relaxing environment.
Every time a post or a comment was posted on the Facebook group, I received a notification of it from Facebook.This notification reminded me of the written assignment, which was helpful because I sometimes forget home assignments.At no time did I forget to do the writing assignments, thanks to notifications from Facebook.Some students found that exchanging comments on Facebook was motivating since they knew for sure that their friends would read and provide comments on their drafts thanks to the notifications of Facebook.
I was a bit motivated to post my writing on Facebook because I knew my friends would read my posts, and I often waited in suspense for the comments.These activities motivated me to work harder on my English writing.
Also, the students claimed Facebook-based peer comment activities made them spend more time practicing writing and helped them better remember the structure of a responding email.One of the reasons they spent more time practicing writing must have been that they were required to give comments in English, not in their mother tongue, so they had to think a lot before writing.This seemed to make the learning occur, ''on commenting on my peers' writing, I had to struggle to write comprehensible comments, which helped me improve my writing skills.'' The Impact of Facebook-Based Peer Comments on Students' Writing Revisions Furthermore, students strongly agreed that, as a result of Facebook-based peer comment practices, they found that their writing contained several errors that they were unable to point out, and they learned different writing styles and ideas from their friends while also avoiding mistakes created by their peers.As a result, these exercises supported them in coming up with new ideas for Pham and Phung revising their own writing and developing their writing skills.
Reading my friends' writing taught me new sentence structures to convey my ideas.I could also notice the contexts in which my friends used some structures.I knew these structures, but it never occurred to me that I could use these structures in such contexts.Thanks to reading my peers' writing, I learned the contexts in which I could use the sentence structures that I knew to express my ideas more clearly.
It was evident that not only giving comments and reading posts on the Facebook group helped, but peer comments that students received from their peers played a key role in the success of the activities.
It is said that ''outsiders see more than insiders.''When my peers read my writing, they assessed my writing from different angles; therefore, they could assess my writing impartially, and they could point out many mistakes in my writing.When my friend proofread my writing, they could easily point out my mistakes that I could hardly realize since when I wrote, I focused more on expressing myself in my writing than the English language rules.Third, when asked about the impact of Facebookbased peer feedback on writing revision, students decided that sharing their writing on Facebook for their friends to read and comment on made them think more about their writing content.This face-saving element stimulated students' interest in their writing, ''I tried to write coherent and cohesive emails in order to receive positive comments from my peers.''They agreed that peer comments were helpful to their revisions and that peer comments via Facebook Community supported them in reorganizing the ideas in their emails more logically.These students also reported that after each revision based on peer input, their sentence structures became more varied, and they improved their vocabulary, grammar, and spelling, saying that ''realizing errors and redrafting an email several times helps me improve my emails.'' The activities were incredibly helpful to me.In a few first emails that I wrote, I wrote very simple functional sentences.But when reading my peers' emails, I found that they used advanced sentence structures, which I knew but rarely used; I learned the high-level sentence structures and styles from my peers to improve my writing skills as a result.
Apart from positive evaluation of Facebook-based peer comment activities from most students, a few students showed indifference to these activities since their peers did not participate effectively in the activities or doubted the accuracy of their peers' comments because some students did not take it seriously.''These activities enabled peers to correct each other's writing.However, some peers did not do the activities properly, which created a lot of difficulties for other peers.''In this situation, a student admitted that ''the problem was that I was too lazy to take part in the activities fully.''It is suggested that the platform was just a platform.The students should take responsibility for their learning to take full advantage of it.
From the above findings, the current study invented a framework in the Figure 4 as follows:

Discussion
Research question 1 investigated whether Facebook peer comments impact students' writing performances and how much they incorporate Facebook-based peer feedback into their revisions.The results of the current study indicated that 20% of the students' second drafts' changes were based on comments, 24.9% were made partly based on comments, and 55.1% were not based on comments.This means that less than half of the changes were triggered by peer comments (only 44.9%).In other words, in the total revision, 45% of the peer comments were incorporated into writing revisions, and the writers' own decisions were revised by 55%.These results were roughly similar to these of the studies by Liu and Sadler (2003), Wichadee (2013), and Pham and Usaha (2016).Liu and Sadler (2003) concluded that 41% of e-comments led to revisions.Similarly, Liou and Peng (2009) found that peer comments in asynchronous computer-mediated peer responses triggered 48% of changes.These results are slightly different from Pham and Usaha's (2016) conclusion that peer comments triggered 39% of revisions.These findings are encouraging as they highlight peer comments' role in making EFL writing learners more autonomous.The student writers were not fully reliant on their peers when revising their drafts.
The results of the current study confirm the view held by many researchers (Berg, 1999;Liu & Hansen, 2002;Pham & Nguyen, 2020;Pham et al., 2022;Tran & Pham, 2023) that students, particularly those who have received training in peer comments, are able to give useful comments in their peers' drafts.Vo (2022) found the students were hppy to welcome both peer and teacher feedback.Also, the fact that peer comments in the current study stimulated only 44.9% of changes in the second drafts is in line with Tuzi (2004)'s conclusion that electronic peer comments were not the main incentive for revisions.Yet, they play an important role in the revision process.Thus, it can be concluded from these findings that the students are held accountable for their own products.Still, the impact of peer comments on revisions is undeniable.Not all peer comments are useful or revision-oriented, but the acts of reading and commenting on their peers' drafts and reading comments on their own drafts might have caused the student writers to reflect on their drafts critically and have second thoughts about the points they were not sure in his or her drafts.Whether peer comments are valuable or not, they motivate student writers to proofread their own drafts again and again to improve their drafts.
Regarding the levels of revision, the results also showed that, of the eight levels of revision investigated, the ''Word'' level was the most frequent level of revision (24.6%), which corroborated the findings of previous studies by Tuzi (2004) and Pham and Usaha (2016), who discovered that the ''Word'' level was the most frequent revisions.Interestingly, at the ''Word'' level, students made the majority of the modifications (64%) on their own, although only 22.9% of revisions at the ''Word'' level were based on peer feedback, and 13.1% of revisions were partially based on comments.The ''Grammar'' level was the second most frequently updated level (=229, 16.7%).In contrast to the ''Word'' level, peer comments caused more than half of the changes (59.8%) at the grammar level.As a result, it can be deduced that students often need peer assistance to develop their drafts' grammar.When it came to revisions at higher levels, such as ''Clause,'' ''Sentence,'' and ''Paragraph,'' the number of changes made by the student writers themselves outnumbered the number of changes caused by comments.This current study's result was inventory as it contradicted Pham and Usaha's (2016) finding that student writers needed less support from peers at higher levels.This is partly to confirm that Facebook aided students in improving their performance (Chugh & Ruhi, 2018).
The highest level of revision, ''paragraph,'' was the third least frequent level of revision, after ''punctuation'' and ''spelling'' levels.This result is different from the findings of Tuzi (2004), Min (2006), and Pham and Usaha (2016), who found that the ''paragraph'' level was among the four most frequent levels of revision.This can be attributed to the difference in the students' proficiency levels and the kinds of writing tasks.Concerning the students' proficiency levels, the participants of the current study were non-English majors taking a pre-intermediate level general English course, while those participating in the studies mentioned above were English majors who were taking an English writing course.Concerning the kinds of writing tasks, the participants in the current study were asked to write a responding email.This nonacademic genre was a short piece of writing and contained far fewer paragraphs than the genres that the participants in the other studies were asked to write.Research shows that novice writers have a tendency to make surface revisions (Liu & Hansen, 2002).

Pham and Phung
The current study's findings also showed that the number of changes related to the two levels, ''Punctuation'' and ''Spelling,'' ranked lowest among the eight levels published, accounting for 6.7% and 6.5% of the total changes made in the second drafts, respectively.This result appears to be consistent with the conclusions reached by Liu and Sadler (2003), Tuzi (2004), Pham andUsaha (2016), andHo (2015) that at lower levels of revision, such as ''spelling'' and ''punctuation,'' student writers can identify and correct their mistakes on their own using Microsoft Word's spelling check feature, and thus peer comments have an effect at higher levels of revision.The current study confirmed the findings of Pham et al. (2020a), who found that Facebook peer comments improved EFL students' English writing outcomes.
Research question 2 investigated the students' perspectives on using Facebook-based peer comments in writing classrooms.Data analysis from the semistructured interviews indicates that the students expressed a highly positive attitude toward Facebookbased peer comment activities (M = 4.15).This finding is in line with that of previous studies on the use of computer-mediated peer comments in teaching writing conducted by Liu and Sadler (2003), Pham andUsaha (2016), andHo (2015).Liu and Sadler (2003) claimed that giving peer comments on Microsoft Word and exchanging peer comments via MOO were more appealing effectively than exchanging peer comments via the traditional mode.Similarly, Pham (2014) found that students expressed highly positive attitudes toward the use of blogbased peer comments in revisions.More specifically, students' positive attitudes toward the introduction of Facebook-based peer comments in the course support the findings of previous studies on Facebook-based peer comments by Phuong and Phuong (2016), Suthiwartnarueput and Wasanasomsithi (2012), Wichadee (2013), Yusof et al. (2012).Thus, it can be inferred from the results that Facebook-based peer comment activities are workable, and teachers can consider applying these to their writing classes since students express a highly favorable attitude toward the activities.
With regards to student's evaluation of the usefulness of Facebook-based peer comment activities, the students appreciated the usefulness of Facebook-based peer comment activities.This finding supports the ideas suggested by many researchers (Chugh & Ruhi, 2018) about the usefulness of computer-mediated peer comments.This finding is also consistent with Yunus et al.'s (2012) claim that Facebook promotes interaction between the teacher and students and between students.
Regarding students' evaluation of the effects of Facebook-based peer comments on their writing revision, they agreed that they help improve their writing quality.This finding confirms the findings of previous studies by Yusof et al. (2012), Suthiwartnarueput and Wasanasomsithi (2012), Wichadee (2013), and Phuong and Phuong (2016).The finding confirms Pham's (2016) idea that students' collaboration via Facebook helps language production.This finding aligns with Yunus et al.'s (2012) view that Facebook enables students to get enough input by accessing multiple drafts, peer comments, and teacher comments to develop their writing skills.
However, the present study's result about the students' highly positive attitudes toward Facebook-based peer comments seems to contradict the findings reported by Xu (2007), Ho and Savignon (2007).Xu (2007) asserted in his study that the students did not express any overt preference for either commenting using the functions ''track changes'' and ''add comment'' of Microsoft Word or paper-and-pen peer comments.In contrast, in Ho and Savignon's (2007) research, they concluded that despite many valuable and convenient features of Microsoft Word, such as ''Track Changes'' and ''Spelling and Grammar Checks,'' they favored face-to-face peer comments.
Still, it is true from the results of this study that Facebook-based peer comments did not receive all student preferences.This can be explained by the view of Liu and Hansen (2002) that there exists uncertainty about peer comments, such as they can be of low quality since students tend to focus too much on surface structure and lack investment in peer comments or their knowledge of English is limited.Nonetheless, the number of students who did not express a highly favorable attitude toward Facebook-based peer comments is far lower than the number of those who did.Hence, writing teachers need to have Plan B if some of their students are not interested in these activities when implementing these activities.

Conclusion
Regarding the extent to which Facebook-based peer comments are incorporated into revisions, interestingly, results show that only 44.9% of the changes in the student's second drafts were triggered by peer comments, which 20% of the changes were based on comments, and 24.9% of the changes were partly based on comments.The current study found that 55% of the changes were not based on comments.The student writers made revisions to their papers by their own decision.This indicates that the students took high responsibility for their learning process to improve the quality of their writing products.Thus, it can be concluded that Facebook-based peer comment activities encourage students to revise their drafts and take more responsibility for their own writing.
With regards to the level of revision, the most frequent level of revision was ''Word,'' followed by ''Grammar,'' ''Phrase,'' and ''Sentence.''However, it is worth noticing that at higher levels such as ''Clause,'' ''Sentence,'' and ''Paragraph,'' the number of changes made by the student writers themselves exceeded the number of changes triggered by comments.After obtaining the Facebook-based peer commentary activities, the results reveal that the students took more responsibility for their studies and paid more attention to improving their writing products.This finding contradicted Pham and Usaha's (2016) finding that the students needed less help from peers, whereas at higher levels, the students needed more help from peers.In other words, Facebook-based peer comments, a social network, might play an important role in motivating the students to care more about their writing quality when they post an entry.
Regarding students' evaluation of the effects of Facebook-based peer comments, the students claimed that it is motivated to conduct peer comments on Facebook.The students spent more time on the commentary activities because they had to pay lots of attention to do this activity.The students also revealed that employing Facebook to learn writing helped them use time efficiently on this social network.Therefore, writing teachers can consider implementing Facebook-based peer comment activities in their writing classes since students generally express a highly favorable attitude to these activities.
The following practical conclusions for L2 writing instructors are drawn from this research.To begin, more than half of the adjustments made to the second version after peer comments were made by student authors, indicating that negative remarks do not always matter.These awful remarks may be scrutinized later by writing professors.The beauty of peer comments is that they encourage student writers to modify and enhance their drafts, implying that student writers take more responsibility for their studies and develop greater autonomy.Instructors, particularly those teaching English to nonmajors, should be aware of the value of peer comment activities and include them in their classes to help students develop their autonomy.Finally, the present study's discovery that students have a favorable attitude toward Facebook-based peer comment activities offers writing professors permission to include these activities in their classes.This is because when the students have a positive attitude about an activity, they are more likely to be helpful and eager to participate.These results assist writing instructors in determining the most appropriate platform for peer comment activities in their writing classes.If the students have access to current technology, such as the Internet and computers, peer comment activities on Facebook assist them in using accessible technologies to improve their writing.On the other hand, if the students are technophobes and prefer the conventional approach, peer comment activities using paper and ink are viable substitutes.
While the present research contributed scientific evidence to the literature on EFL peer comments, it did have several limitations.To begin, the sample for this research was not drawn randomly; as a result, applicability to the study's population is restricted.Additionally, since the sample size was limited, the findings of this research cannot be extrapolated to the whole population studied.Second, one may claim that submitting a draft on Facebook required students to use word processors to verify their grammar and spelling, contributing to the students' writing quality development.
The sections that follow provide further research ideas.Future researchers should employ random sampling to increase the possibility of generalizing results.Second, in order to fully compare the consequences of both models, future research should analyze how peer comments are absorbed into subsequent versions.The effect of Facebook-based peer comment activities is assessed only when students publish their drafts to a closed Facebook Group and exchange asynchronous peer remarks as ''comments'' beneath each post.More studies might be done on synchronous Facebook-based peer comments and other Facebook features to determine whether they enhance peer commenting.

Figure 3 .
Figure 3.The layout of the function ''compare'' two versions of a document in Microsoft Word.

Table 1 .
Comparison of the Pre-Tests* Between the Control and Experimental Groups.Test was run to compare the scores of the post-tests of the two groups.Table2reveals that the total number of observations was 72 (n = 72; MD = 6.8).The Mann-Whitney U statistic of 853.000 indicates that the sum of ranks for one group exceeds that of the other group.The z-score of 2.43 indicates that the disparity between the two categories is substantial.The Mann-Whitney U test results

Table 2 .
Comparison of the Post-Tests Between the Control and Experimental Groups.

Table 3 .
Revisions From First Versions to Second Versions.