A Comparison of Teacher-directed and Author- directed Peer Review in a Japanese University EFL Class

Peer review is widely-used in EFL writing classes, and it has been found that students hold generally positive attitudes towards both providing and receiving peer feedback. However, while it is common for teachers to utilize peer response worksheets, little research has looked at the impact that the specific style of these sheets can have on how language learners conduct peer review and their attitudes towards it. In this study, participants first used a generic peer response sheet designed by the teacher (teacher-directed peer review), then an individualized peer response sheet that each learner had created, focusing on specific points that they wanted advice on (authordirected peer review). Surveys were conducted after each peer review session, and the data revealed that the two peer review styles prompted peer reviewers to give different types of feedback, with a greater focus on surface level issues during author-directed peer review. Furthermore, although all students agreed that both styles had been useful, 60% stated a preference for the author-directed style.


Introduction
The use of peer review in writing classes gained popularity with the rise to prevalence of the process approach to writing in the 1970s and 80s. With regard to English language teaching, much early research into peer review was carried out in ESL classes, particularly in North America (see, for example, Manglesdorf, 1992; 2. Literature Review

The Benefits of Peer Review
The immediate short-term goal of incorporating peer review activities into L2 writing classes is to provide students with an additional source of feedback which can help them make revisions improving the quality of their written compositions. Over and above the utility of the feedback they receive, it has been proposed that learners benefit from peer review in various other ways, both as writers and as language learners more broadly.
Firstly, in additional to the improvements generated by the feedback they receive on their own writing, through reviewing texts students can learn from both the successes and the mistakes of their peers (Hu, 2005;Zhao, 2014), with the former serving as models to emulate and the latter as pitfalls to avoid, thus helping to raise the reviewer"s awareness of the strengths and weaknesses of their own writing (Tsui & Ng, 2000).
Next, studies have shown that by undertaking peer review learners can develop a greater awareness of audience than if they are only writing for the eyes of a teacher (Hu, 2005;Mangelsdorf, 1992;Mendonca & Johnson, 1994;Tsui & Ng, 2000). This in turn can encourage students to devote more care and attention to their writing, as the knowledge that your work will be viewed by your peers can generate a different, and perhaps stronger, motivation than if a teacher will be the sole reader.
In other ways too, when the two are used in conjunction, peer review can offer benefits complementing teacher feedback. Compared to teacher feedback, peer review comments can be easier to understand (Zhao, 2014), particularly if learners are able to give and receive comments in their L1 (Williams, 2018). Peer review has also been found more likely than teacher feedback to promote collaborative dialogue and negotiation of meaning (Hu, 2005;Mendonca & Johnson, 1994;Rollinson, 2005;Tsui & Ng, 2000), both for the practical reason that peers are more likely to have time to discuss comments in detail, and the affective reason that the inherent power relations of teacher and student may make students more comfortable questioning and discussing feedback provided by their peers (Zhao, 2014).
Finally, beyond improving individual pieces of writing, peer review has also been credited with helping to develop students as effective language learners in the long term. By critically reading and appraising the work of classmates, students become better able to approach their own writing in the same way (Rollinson, 2005), and peer review can thus help to promote learner autonomy (Tsui & Ng, 2000;Yang, Badger & Yu, 2006).

Student Attitudes towards Peer Review
Research on attitudes to peer review has uncovered students" perceptions of potential problems in the activity, particularly those relating to Asian students studying in ESL classes. Mangelsdorf (1992) found that of six students who expressed a negative view of peer review, five were from countries in Asia, leading her to suggest that "the peer review task may be resisted by students not familiar with a collaborative, student-centred environment" (p. 280). Similarly, a series of studies focusing mainly on the attitudes of Asian, in particular Chinese, ESL students towards peer review concluded that they tended to be reluctant to offer critical feedback and were less likely to value feedback received from peers (Carson & Nelson, 1994;Carson & Nelson, 1996;Nelson & Carson, 1998;Nelson & Murphy, 1993). However, as Carson and Nelson rightly note in their 1996 paper, the key issue here may not necessarily be the cultural beliefs of Asian students per se, but the interaction of those beliefs with a mixed nationality student group. Hu (2005), for example, in a study of an ESL writing class in Singapore comprising entirely Chinese students, found that all 20 students involved believed peer review to have been useful in improving their writing skills.
Similarly, recent EFL research also indicates that in this context students hold broadly positive attitudes towards doing peer review in writing classes, with few of the affective concerns noted by some ESL studies being apparent. Wakabayashi (2008) found that all 25 of the Japanese university students in her study agreed that peer review had been effective, with 21 of the 25 expressing a desire to do it in future classes. Hirose (2008), also working in Japan, found that students enjoyed both giving and receiving peer review comments, and believed that their peers were capable of giving useful feedback. Similar results were obtained in other Japanese studies, with both Coomber and Silver (2010) and Morgan, Fuisting and White (2014) finding that while students doubted themselves as reviewers, they had few affective concerns about providing feedback, and had confidence that peer review could help them improve their writing. In a study carried out in China, Wang (2014) noted that only six of 53 students mentioned interpersonal concerns in a post peer review questionnaire, while a larger proportion focused on more practical concerns such as their partner"s English ability.

Implementation of Peer Review
While various studies have thus found students" attitudes to be largely positive, it is important to note that, as with most classroom activities, there are endless permutations as to exactly how peer review is carried out in practice. No two teachers are likely to carry out peer review in exactly the same way, and even an individual teacher will probably vary their peer review procedures according to course goals, class size and level, type of writing to be reviewed, students" previous experiences, and countless other factors. Bearing this in mind, it seems that decisions made by the teacher as to how peer review is conducted could potentially have a major impact on both how successful it is and how students feel about it. Yet this is rarely mentioned in the research literature, with general claims made for peer review often decoupled from the details of exactly how the peer review in specific pieces of research was carried out.
As one of 19 proposed guidelines for implementing peer review, Hansen and Liu (2005) suggest that teachers "create purposeful and appropriate peer response sheets for a given task, genre and purpose" (p. 33). This piece of advice seems to have been widely followed by teachers and researchers, with numerous investigations of peer review having been carried out using some kind of teacher generated peer response sheet (see, for example, Hirose, 2008 and2012;Hu, 2005;Mangelsdorf, 1992;Min, 2005;Sengupta, 1998;Taferner, 2008;Williams, 2018;Yang, Badger & Yu, 2006). However, the format of these peer response sheets can vary widely in terms of the number and complexity of questions that peer reviewers are requested to address. That used by Hirose (2008, p. 552), for example, gives four simple and broad instructions: Min"s (2005, p. 307) peer response sheet, in contrast, has six complex and specific points for peer reviewers to deal with, the third of which is shown here as an example: "Now read the following two or three sentences. Did the writer write according to your expectation(s)? If not, what did the writer write instead? Do you think that writer was sidetracked? Go back to the bridge (second sentence). Did the author choose a word that is not the controlling idea to develop? Did the author talk about an idea more general than or in contrast to the controlling idea? If none of these applies, reread the topic sentence to make sure that you understand the writer"s intention." Clearly, as Hansen and Liu (2005) suggest, a peer response sheet needs to be appropriate for use with the specific task, and of course with the specific students, for which it was created: like most classroom activities, peer review is not a onesize-fits-all exercise. And while peer response sheets are widespread, they are not ubiquitous. Rather than having them answer a list of pre-determined questions, it is also possible to give students a freer hand when conducting peer review and have them write comments directly on a peer"s draft, deciding for themselves on what they consider the most salient feedback to provide (examples of this approach include Coomber & Silver, 2010;Mendonca & Johnson, 1994;Rahimi, 2013;Zhao, 2014). Whether or not a peer response sheet is used, and if so what form it takes, would seem to be a crucial contextual variable in determining how students experience peer review, and it is thus rather surprising that few, if any, studies to date seem to have investigated the effect that the presence or form of this type of worksheet has on the outcomes of peer review. Charles (1990), writing about teacher feedback, puts forward a method through which this could be made more interactive, proposing a "system of self-monitoring, whereby students annotate their drafts with comments or queries on their problem areas, before handing their texts in to the teacher" (p. 286). By allowing students to exert an influence over the type of feedback they receive, this idea certainly offers the merit of transferring more responsibility and control to the learner. Charles notes that this method also "allows the writer to raise issues that would not normally be commented on by an editor" (p. 289), something which could perhaps be even more relevant if the technique were applied to peer review. In concluding her article, Charles suggests several potential benefits to the self-monitoring approach, which can be summarized as follows:

Applying 'Self-monitoring' to Peer Review
(i) writers become active participants in the feedback process, not passive recipients.
(ii) writers are more receptive to using comments they have asked for. (iii) writers feel responsibility for their own writing and better able to analyze it. (iv) the editor has the sense that they are actually helping the writer.
(v) there is a greater focus on the students" own writing aims.
Peer response sheets produced by the teacher, although widely used, can take control of the review process away from the students. On the other hand, it is clear that most students will benefit from some structure when doing peer review, as even with training only those at a higher level of higher English proficiency are likely to be able to provide effective feedback without some indication of which points to focus on. Thus, adapting Charles" self-monitoring technique by having individual students create a peer response sheet specific to their own composition, rather than using a generic sheet created by the teacher for the whole class, could be an effective way to provide support and guidance for reviewers while at the same time giving greater control over the peer review process to learners. Although the self-monitoring technique can be effective in improving the implementation of teacher feedback, it appears that little, if any, research has been conducted into whether it could be similarly applied to peer review.
Thus, the current paper reports on an attempt to adapt self-monitoring to peer review, comparing peer review using a teacher-created response sheet with peer review using a response sheet created by the author of the composition being reviewed. These two different styles will hereafter be referred to as "teacher-directed peer review" and "author-directed peer review" respectively. The following research questions are addressed: 1) Did the style of peer response sheet affect the feedback that students gave? 2) Did the students prefer teacher-directed peer review or author-directed peer review?

Participants
The research was carried out with a class of 26 first-year students in a Japanese university. On the basis of a placement test conducted at the start of the semester, these students were in the top 20% of their cohort with regard to English ability; however, they varied considerably in terms of their previous experiences of English.
Of the 26 students, 16 had experience of studying abroad, mainly in English speaking countries, for periods ranging from three weeks to four years. In terms of their self-reported experiences of writing in English, some had written essays of over 1000 words, while others had never written over 300. Eleven students had no prior experience of doing peer review, while eleven had done it before in English, one in Japanese, and three in both languages.

Peer Review Schedule
Over the course of a 15-week semester, students wrote one essay of at least 1000 words on a topic of their own choice, including citations and references. They submitted five drafts of the paper: the first and third were peer-reviewed, the second and fourth received teacher feedback, and the final draft was for evaluation.
In week 5 of the course, students reviewed the introduction to their partner"s essay. This was intended as a peer review training session, and made use of a teachercreated peer response sheet. Students were given 15 minutes to read their partner"s introduction and complete the response sheet, and then 5 minutes to discuss their writing. In week 6, using a similar but more detailed peer response sheet (Appendix 1), students followed the same procedure over 30 and 10 minutes respectively, this time reviewing their partner"s first full draft. The final peer review was carried out in week 11, using the handout in Appendix 2. Students were given 20 minutes to read and review their own drafts and write five questions about which they would like advice from a peer (they had also been asked to think about this for homework). As with the previous session, they then had 30 and 10 minutes for written and oral response respectively. For each peer review, students worked with a different partner.

Data Collection and Analysis
A questionnaire survey was administered three times during the semester (Appendix 3). Before each survey, it was emphasized to students that their responses were completely unconnected to their evaluation in the course, and that they should feel free to answer all questions honestly. In week 4, before doing any peer review, students completed a preliminary survey consisting of four background questions and ten Likert scale statements (Q1-Q10) investigating their attitudes to peer review in general. The statements were based on the survey developed by Coomber and Silver (2010), and a six-point Likert scale was used in order to discourage fencesitting. The second survey was completed after students had done peer review of their first drafts using the response sheet provided by the teacher (teacher-directed peer review). This survey consisted of the same ten Likert scale items, plus three open ended questions (Q11-Q13). In the same way, after doing peer review using the response sheets consisting of their own questions (author-directed peer review), students completed the third survey. This was the same as the second survey, but with one further open-ended question (Q14). This paper focuses on the data obtained from the four open-ended questions. Students" responses to these questions were coded and categorized according to the themes that emerged from the data. In the case that a student made more than one point within their comment, each point was coded and categorized individually.

Research question 1: Did the style of peer response sheet affect the feedback that students gave?
The first open response question in the survey (Q11) asked students how they benefited as the recipient of peer feedback on their compositions, with answers summarized in Table 1. As can be seen, in both sessions, students reported receiving peer advice on a wide range of issues. Over a third of respondents mentioned that in the teacher-directed peer review, their partner had suggested they add more detail to their essay. This is perhaps unsurprising, as questions 2 and 7 on this response sheet directly relate to the level of detail of the essay (see Appendix 1). Likewise, several questions on the sheet focus on essay structure, the second most common answer. After the authordirected peer review session, only half as many students mention getting advice to add further detail to their essays. Two explanations seem likely. Firstly, it is possible that by the time they did this peer review, students had already added more details in response to either peer or teacher comments, and thus did not need further advice. Alternatively, when students created their response sheets, it may be the case that it was easier for them to focus on what was in their essay, rather than what was not: logically, it seems more likely that a writer will be more aware of potential problems with something they have written, rather than something they have not written.
One criticism of peer review sometimes heard is that students may focus excessively on surface issues (Connor & Asenavage, 1994;Mangelsdorf & Schlumberger, 1992). However, only one student reported that their partner had given them advice on grammar during the teacher-directed peer review session (although it is possible that some of the comments referring to unspecified "mistakes" may also have related to grammatical points). It has been suggested that feedback on early drafts is better targeted at global issues of content and structure (Zamel, 1985), so it is encouraging that the participants seem to have done exactly that. During the second, authordirected, peer review, more advice on local issues (grammar and citations) was reported, and less on global issues. Again, two reasons seem possible: either many global issues may have already been dealt with by this point, or, when asked to create their own response sheets, students may be more inclined to focus on the local.
Much research demonstrates that doing peer review can be of benefit not only to the person receiving advice, but also to the peer reviewer him or herself (Yu & Lee, 2016). Question 12 investigated whether these students felt they benefited by reading their classmates" compositions, with results presented in Table 2. Responses to question 12 also indicate that students had a greater focus on global issues, such as structure and clarity, during the teacher-directed peer review. Of the nine comments relating to essay structure, seven included explicit terminology that we had used in the classroom, for example: "My partner had really well-constructed essay with amazing hook and I felt I had to follow it." "My topic sentences were so bad but I could improve because my partner's topic sentences were so good and I could get idea from my partner." In contrast, only five students mentioned essay structure in relation to the authordirected peer review, and only two of these used explicit terminology. This perhaps gives an indication of the degree to which using a peer response sheet created by a teacher can serve to control and circumscribe the ways in which students respond to their peers" writing. While this is not necessarily a bad thing, it is something that teachers should be aware of when deciding how to implement peer review in their classes.
In contrast, after the author-directed peer review, over twice as many students wrote comments related to grammar and vocabulary in response to question 12. These comments can be divided into two categories: those which note that finding mistakes in their peers" work helped identify similar mistakes in their own, and those which suggest that the strong points of the paper they reviewed inspired them to revisit their own essay: "When I read my partner's essay I found grammar mistake. Actually, I did same mistake. I could find it."

"She used idioms and transition words which I didn't use. That was pretty good I think. I think that I don't have a lot of vocabulary like other students, so their essays always become my motivation."
Interestingly, very few of the author-created peer response sheets included specific questions about grammar or vocabulary, with most focusing on either content, use of citations, or essay structure, and in this respect, they were not so dissimilar to the teacher-created sheet. Thus, it seems that the additional focus on grammar and vocabulary in the second peer review did not result from students" particular desire for feedback on this. Rather, it may be that when doing peer review, students feel less constrained using a response sheet created by their partner, as opposed to the teacher, thus allowing them greater freedom to think about all elements of the writing instead of focusing narrowly on the response sheet itself.
In summary, from the many detailed answers given to questions 11 and 12, it seems that the two styles of peer review did generate different types of advice. However, it should be noted that it cannot be stated with certainty that this is an intrinsic feature of the two styles: it may also be due to either the design of the specific response sheets, or the order in which the peer review sessions were conducted.

Research question 2: Did the students prefer teacher-directed peer review or author-directed peer review?
Question 13 investigated students" perceptions of their own effectiveness as a reviewer. A summary of answers to this question is presented in Table 3. Although research has shown that students can lack confidence in their abilities as reviewers (for example, Coomber & Silver, 2010;Hirose, 2008;Sengupta, 1998), after both peer reviews a clear majority of students felt that their feedback had been useful for their partner. However, while 59% of students who completed the survey after teacher-directed peer review stated that they felt this way, this rose to 88% after the author-directed session. Two explanations seem possible: firstly, it may be the case that questions asked by a peer are more consistent with students" level of understanding than those written by the teacher; alternatively, students could, quite logically, feel greater confidence that their advice had been useful if they had answered specific questions asked by their partner, rather than generic ones set by the teacher.
After a brief recap of the two types of peer review, question 14 asked participants to state directly which style they preferred. All 25 students who completed this survey expressed a clear preference for one or the other, and all stated at least one reason for that preference. Ten students preferred the teacher-directed peer review, and fifteen the author-directed peer review, with their reasons summarized in Tables 4  and 5.

Table 4.
Reasons that students preferred teacher-directed peer review.

Reason
Number of students mentioning this The teacher has more knowledge about essay writing 5 You can get advice about points you haven"t noticed by yourself 3 You can focus on the rules and conventions of essay writing 3 You can ask your friends about specific points outside class time 2 Table 5.
Reasons that students preferred author-directed peer review.

Reason
Number of students mentioning this You can focus on specific things you want to know 10 You can get more detailed advice 7 Having specific questions makes it easier for the reviewer 4 Making your own decisions is a good thing in itself 3 You can review your own paper while writing the questions 2 Half of the students who preferred the teacher-directed style mentioned the superior knowledge of the teacher as one of their reasons, exemplified in the following comment: "I prefer peer review for first draft. Because the points are decided by teacher who knows more about essay than me. Therefore the viewpoint must be clear and accurate, so maybe I can get more useful advices from my partner if she/he gives me advice based on the viewpoints." In itself, this reason is perhaps predictable, and reflects the findings of other research (Sengupta, 1998;Zhang, 1995) that students often ascribe high value to teacher input. What is somewhat more surprising is that all five students giving this reason had studied abroad for at least ten months, and were relatively proficient and confident speakers of English. In their response, one stated that "I don't have enough knowledge about good essay but I believe teachers do". This may be connected to the second reason given in Table 4, that using a peer response sheet created by the teacher allows you to get advice about problems you hadn"t noticed. Clearly, this is less likely when the author creates the sheet: as one of the participants noted, "it is difficult to find points that need check"; moreover, even learners of English of an advanced level may struggle to fully understand the rules and conventions of academic prose. Thus, the fact that students view the teacher as more knowledgeable about essay writing does not necessarily suggest a lack of confidence in their own abilities, but perhaps a realistic appraisal of the relative levels of expertise of teacher and learner in the field of academic writing.
The two most common reasons for preferring author-directed peer review, that you can get advice about your specific areas of concern, and that the advice received is more detailed, are closely connected. While a response sheet created by a teacher is for use by the whole class, and thus necessarily generic, one created by the author of the work to be reviewed offers the possibility of getting peer feedback tailored to the particular concerns that author has about that essay, something which was clearly appreciated by many students, as indicated in the following comments: The other reasons offered by respondents also highlight important benefits of author-directed peer review which may not apply to the more commonly-used teacher-directed format. Four students pointed out that, as well as generating more precisely targeted feedback, this method made reviewing papers easier and more efficient, with one noting that: Self-directed review and revision is a crucial means by which second language learners can both improve their compositions and enhance their ability as writers in the longer term, and it is something that students are more likely to engage in when in some way prompted to do so (Coomber, 2016). Over and above the advice they can receive from a peer, a key advantage of author-directed peer review may thus be that it requires students to carefully re-read their work.

Further Discussion
Increased autonomy and reduced reliance on the teacher is a key advantage claimed for using peer review in writing classes, as it can make students less dependent on teacher input and allow them greater control over what feedback to incorporate into future drafts and what to ignore (Hyland, 2000). However, Mendonca and Johnson (1994) make the important but rarely addressed point that the use of a teacher generated response sheet during peer review may have the unintended effect of transferring control of what should be a student-centred activity back to the teacher. As Hyland (2000, p. 51) notes, peer response sheets "can take away the spontaneity from the response situation and may be viewed by the responders as just another worksheet to fill in", and students may end up spending more time answering questions set by the teacher than addressing the concerns of their peers. Given the relationship between autonomy and motivation (Ryan & Deci, 2017), the fact that 60% of participants in this study preferred author-directed peer review may thus be linked to the increased autonomy it offers: through creating their own peer response sheets, L2 writers retain greater control over the process of feedback and revision. Similarly, when completing an author created response sheet, peer reviewers can be more confident that their advice will be both relevant and appreciated, which may in turn satisfy the psychological need for feelings of competence (Ryan & Deci, 2017).
Although Hansen and Liu (2005) made detailed suggestions as to how peer response sheets and grading rubrics could be developed by students themselves, this seems to be a suggestion that teachers have rarely implemented, and there are few examples of other studies investigating the potential of author-directed peer review. Ruegg (2018) adopted author-directed peer review, however this in itself was not the focus of her study, which compared the gains in self-efficacy made through peer and teacher feedback respectively, rather than comparing different modes of peer review. Xiang (2004) investigated the self-monitoring technique proposed by Charles (1990), but did so within the context of teacher feedback, not peer review, finding that most students displayed positive attitudes towards self-monitoring. Within the context of peer review, this result is reflected in the present study, in which all students considered author-directed peer review to have been useful. While more research is clearly needed, the present study is thus significant in that it is one of the first to make direct comparisons between peer review carried out using author created and teacher created response sheets.
However, although the results suggest that the use of author created response sheets can be an effective way to conduct peer review, certain limitations of this research should be pointed out. Firstly, the small sample size means that these results cannot be generalized beyond the participants in this study: further research is needed to determine how learners in different contexts and with different levels of English proficiency would respond to author-directed peer review. Secondly, and connected to this point, individual pairings of students may have impacted upon participants" perceptions of the success of each peer review: it is quite possible that a student who preferred the author-directed format did so simply because they had a more helpful and competent partner. In a study of this size, such individual variations may have skewed results in a way that could be avoided in a larger scale investigation. Lastly, it should be acknowledged that the order in which peer review sessions were carried out may have had an impact on both the type of feedback students gave and the participants" evaluations of the two formats, and a similar study making use of a counter-balanced research design would thus be valuable.
As noted previously, little research appears to have been carried out into authordirected peer review prior to this study, and several further questions appear worthy of investigation: When students create their own peer response sheets, are they able to write meaningful questions, and how easy or difficult is this for them? What type of comments tend to be given by reviewers? Are they more likely to offer praise, suggestions, or criticism than when using a teacher created response sheet? How do rates of uptake compare between author-and teacher-directed peer review, and what proportion of resulting revisions succeed in improving the composition? Further research into these, and other issues, would be welcome.

Conclusion
In summary, students agreed unanimously that both teacher-directed and authordirected peer review had been useful, both in terms of the feedback they received and the knowledge they gained as a reviewer. However, the data obtained from survey responses appear to suggest that students use and respond to a peer response sheet in different ways if it has been created by the author of the composition they are reviewing. When using the teacher created response sheet, students largely focused on global issues of content and essay structure, whereas the author created response sheets prompted a more balanced mix of feedback on both global and local areas. In terms of overall preference, 60% of students opted for the author-directed format, with two-thirds of these students mentioning the possibility of getting more specific and tailored feedback as one of the reasons. Students also seem to value the additional autonomy that this style of peer review offers, both allowing them to determine the type of feedback they want to receive, and offering greater flexibility when reviewing the work of their peers.
As Saito and Fujita (2004) note, the fact that students may express a preference for one type of classroom activity over another should not be taken to mean that they do not find value in both, and the participants in this study appreciated the utility of both types of peer review, with one explicitly noting in her response to question 14 that "using both style is the best to do peer review". Clearly, even with relatively proficient students, a response sheet created by a teacher can target problems that learners are less likely to identify by themselves; on the other hand, in addition to allowing them to focus on problems of their choice, by requiring them to re-read and think more deeply about their own work, asking students to create their own peer response sheet can aid their long-term development as autonomous language learners. Min (2006, p. 119) asks the valid question "why do the majority of peer comments fail to be utilized in students" subsequent revisions?" While there may be various answers, one possible reason is a mismatch between the concerns of the writer and those of the reader: in other words, the feedback given by a peer reviewer may not always address the questions that a writer has about their own text. If peer review is to be a genuinely collaborative interaction between learners, then perhaps we need to ensure that writers have the opportunity to influence the type of feedback they receive from their peers. Furthermore, in many contexts, particularly in Asia, cultural and interpersonal issues such as power relations and the concept of face can "exert negative influence on students" engagement with or participation in peer review" (Yu & Lee, 2016, p. 477); as mentioned by one participant in this study, students may be concerned both about being criticized by peers and having to be critical of them. Thus, one of the greatest potential benefits of the author-directed format could perhaps be that it allows peer review to be re-conceptualized as students asking their peers for help with problems they choose.