L2 writers’ perspectives on face-to-face and anonymous peer review: Voices from China

To validate a recent study conducted in the Japanese context that contradicted the previously held view that Asian students are culturally predisposed to be reluctant peer reviewers, this study examined Chinese EFL learners’ experiences with and perspectives on peer review in a writing classroom. Fifty-seven college students were asked to perform peer review in face-to-face and anonymous modes to examine whether anonymising the process encourages them to offer more constructive criticism. Data were collected from a Likert-based survey, interviews, and the instructor’s field notes. Statistical analysis of the survey data only partially supported the findings of the study replicated, indicating Chinese students’ preference for both face-to-face and anonymous review modes. Although most of the participants exhibited resistance to peer review during the initial stage, they began to appreciate the benefits of the activity, particularly the verbal discussion component of the face-to-face mode, while some of them refused to give negative feedback on their peers’ work. Regardless of the peer-review mode, the participants remained overly conscious of their feedback accuracy and continued to seek confirmation and oversight from an authority figure (the teacher) rather than taking initiative.


BACkgRound
Peer review has long been integrated into writing curricula as an integral part of the composition process in both English as a second and English as a foreign language (ESL and EFL, respectively) contexts. Copious studies have supported its benefits, reporting that engaging students in peer review encourages cooperative learning (Tsui & Ng, 2000) and improves both content-and language-related aspects of student writing (Berg, 1999). Studies have also indicated that second-language (L2) writers learn to identify similar problems in their own essays through an interactive process of reviewing their peers' work (Grabe & Kaplan, 1996).
Frequently, Asian students are collectively described as group oriented and obedient, and their overriding concern for maintaining positive group relationships (Carson & Nelson, 1996) has been considered a disadvantage that prevents these students from fully benefitting from peer review. After interpreting survey results in which five of the six respondents who had completely negative perceptions of peer review were from Asia, Mangelsdorf (1992) suggested that 'the peer review task may be resisted by students not familiar with a collaborative, student-centred environment ' (p. 280). Likewise, Chinese learners' abstention from making negative comments has been explained in terms of their culture, in which verbal negotiations or expressions of one's inner thoughts are less valued than preserving group harmony (Allaei & Connor, 1990). Pointing out such harmony-maintenance strategies, Zander (1983) noted that for those from collectivist cultures, the primary goal of group work seems to lie in maintaining group harmony, leading group members to eschew behaviours that might 'ruffle the composure of those assembled' (p. 7). Hyland (2000) similarly showed in her study of Asian ESL students at a New Zealand university that many of the participants felt 'uncomfortable with the peer response situation' (p. 52) and completed written peer-feedback tasks perfunctorily because it was against their cultural norms to criticise peers' work.
To cope with Asian learners' alleged reluctance to critique others' work, attention has been directed towards introducing anonymity into the peer-feedback process. Hosack's (2003) study, for example, found that Japanese students favour the anonymous mode over the traditional face-to-face format. It was reported that the participants felt less worried about offending the writer when providing feedback anonymously, which in turn resulted in more useful feedback. His study, however, was conducted on returnee students who had lived overseas prior to entering university, limiting the generalisability of the findings to all Japanese students. A few years later, Coomber and Silver (2010) examined the same issue with a larger Japanese student body, exploring whether depersonalising the process ameliorates Japanese EFL learners' reluctance to provide critical comments. Contrary to expectations, the results showed that the participants favoured the face-to-face and anonymous review modes almost equally.
More recently, Kim (2019) analysed learners' perspectives on peer review, also in the Japanese EFL context. The survey results indicated that the participants as a whole did not have a marked preference for either peer-feedback mode, though there was a slight tendency to prefer the face-to-face mode. Participants' interview accounts, however, revealed a preference for the face-to-face mode because of the ease of mitigating negative feedback and of communicative collaboration through verbal communication. Drawing upon such findings, she suggested that L2 student writers' preference for a certain review mode interacts closely with various factors, such as self-assessed target language competence and learning style.
The previous studies outlined above have reported meaningful observations of Asian learners' perceptions and practices of peer review. However, the cultural mediation of peer feedback in the L2 writing classroom remains severely under-researched despite the fact that culture is a useful construct to account for L2 students' attitudes towards and participation in peer feedback (Hu, 2019). To bridge this research gap, the authors replicated Kim's (2019) study and examined (1) L2 learners' experiences with and perspectives on peer review in a writing classroom and (2) whether and how anonymising the process affects their peer-feedback performance by collecting data from China -a country where intragroup harmony and avoidance of overt conflict in interpersonal relationships lie at the core of the cultural values (Hu & Grove, 1991).

Research setting and participants
A homogenous group of 57 Chinese EFL students participated in this study. The students were engineering majors in their first year attending a public university in Guangdong Province, China. The gender ratio of male to female students was approximately 6:1. They were from two compulsory, integrated-skills English classes offered to intermediate learners of English who had English scores on the National College Entrance Examination ranging between 79 and 110 (out of 150), roughly corresponding to levels B1 to B2. Responses to a language background survey conducted at the beginning of the semester showed that the participants had received approximately 11 years of formal EFL instruction at the time of the experiment. None of them had lived overseas or had engaged in peer-review activities before. The class met twice per week for 80 minutes each session. Throughout the 16-week course, four weeks (hereafter referred to as the 'Writing Weeks') were designed to involve writing components. The school policy requires all students from the same department to take the same classes for at least eight hours per day except on weekends. Therefore, by the time the participants engaged in the first peer-review task in week 5, they were close friends with one another.

Procedure
Following the curriculum planned by the department, the participants wrote four 5-paragraph essays on given topics during the Writing Weeks. At the beginning of the semester, the participants were informed that they would engage in peer-feedback activities and that they could volunteer to be interviewed about their peer-review experiences. The participants were also informed and gave written consent that these interviews would be analysed for research and teaching-improvement purposes.
The first writing assignment was announced one week before the Writing Weeks, and the peer-review sessions were held on the days when the first drafts were due. During the four peer-review sessions, one class first engaged in two face-to-face sessions followed by two anonymous sessions, and the reverse order was used in the other class. Although the participants were encouraged to use English during the peer-review activities, they were allowed to switch to their L1 -Mandarin Chinese-as necessary. For each of the four peer-review sessions, the students brought two copies of their assignments.
Immediately before the Writing Weeks, the students received focused instruction regarding peer review (Hansen & Liu, 2005) for two consecutive class sessions (hereafter referred to as the 'practice sessions'). Guided by a structured peer-response sheet that included the features to look for while reviewing (see Appendix A), the students practised writing comments on two essays selected by the instructor from student essay samples from the previous year.
During the actual face-to-face review sessions, the instructor randomly divided the students into groups of three. The group members exchanged their assignment copies and wrote comments on their two peers' papers for 40 minutes (20 minutes per essay). Then, the reviewers provided oral feedback to each writer (group member) for ten minutes for a total of 30 minutes. After the completion of the peer-review session, the reviewers returned the drafts to the writers.
During the anonymous review sessions, the students used pseudonyms as their writer/ reviewer identities; this information was sent to the instructor via email a day before the class meeting at the latest. On the anonymous peer-review days, the instructor randomly distributed half of the collected assignments to the class, ensuring that none of the students received their own papers. Those who received the same copy were paired and sat together. For the first 20 minutes, the students read and commented on the anonymous writer's essay independently. Then, the students shared feedback with their partners for ten minutes. After the first round was completed, the instructor distributed the remaining half of the assignments, and the students repeated the procedure with the same partner. After both rounds were finished, the instructor collected the reviewed drafts and returned them to the pseudonymous writers.
Regardless of the mode, all participants received feedback from two peer reviewers for all of their writing assignments, after which they produced revised drafts by the following class meeting. The only difference between the two peer-review modes was that the reviewers could have verbal discussions directly with the writer in the face-to-face mode, whereas they could share their feedback only with the other reviewer in the anonymous mode.

Data collection and analysis
Drawing primarily upon Kim (2019), the 22-item Likert-based questionnaire (see Appendix B) was administered after all four peer-review sessions were completed. The survey items explored the participants' attitudes towards writing in English; perception of the importance of writing successive drafts; competence in the target language; level of trust towards peer reviewers; perception of the value of peer review in the writing process; and preference for the face-to-face or anonymous mode as a reviewer and a writer, respectively. The students specified their level of agreement or disagreement on an even-point scale ranging from 0 ('Strongly disagree') to 5 points ('Strongly agree'), with no neutral 'Neither agree nor disagree' option. An exploratory factor analysis was conducted on the survey results, with the participants' English proficiency data included. Principal axis factoring with oblimin rotation was employed, applying the Kaiser normalisation for rotating factors.
In addition, interview data were collected from 59% of the participants (n=32), who volunteered to be interviewed about their peer-review experiences. To decrease response bias, the interviews were conducted by a professor colleague in the interviewees' L1, and all of the interview sessions were audio recorded with the interviewees' prior consent. The following interview questions were asked: 1. What did you enjoy most about the peer-review process? 2. What did you not enjoy about the peer-review process? 3. Which peer-review mode -face-to-face or anonymous-did you prefer as a writer and as a reviewer? Why?
The transcribed recordings were translated into English, and thematic analysis was performed on the translated interview data to identify salient themes related to Chinese students' perspectives on peer review and their reasons for favouring a certain review mode. To supplement the survey and the interview data, the instructor took field notes during the observation of the participants' peer-review performance to record both (1) descriptive information, such as behaviours, events, and conversations observed, and (2) reflective information, such as the instructor's thoughts, questions, and concerns (Schwandt, 2015).

FindingS And diSCuSSion
The factor analysis revealed five distinct dimensions with eigenvalues above 1.0. They accounted for approximately 72% of the variance, and all the items in this analysis had primary loadings over 0.5. The factor structure is summarised in Table 1, and each factor is named reflecting the characteristic/construct it represents. Note. The asterisk symbol (*) represents a negatively loaded item.
Both factors 2 and 3 concerned the preference for the face-to-face mode, constituting approximately 32% of the variance. Inconsistent with the findings of Kim (2019), a factor was identified that indicates Chinese L2 learners' preference for anonymous peer review (factor 1), amounting to approximately 18% of the variance. Factors 4 and 5 were related to the participants' positive and negative perceptions of peer-review activities, respectively, each representing approximately 10% of the variance. The factor analysis results are discussed further in the subsections below based on the participants' interview responses and the observations made in the instructor's field notes.

Resistance to peer review in the initial stage
One marked tendency displayed during the practice sessions was the participants' strong resistance to the peer-review activity to the point that the smooth progress of the Writing Weeks seemed unpromising. The participants' lack of active learning experiences during their primary and secondary education in China (Ho & Crookall, 1995), coupled with concern about the difficulties inherent in group work, seemed to hinder their willingness to review others' work, which Chinese students generally consider the teacher's job, not the learner's. Most of the participants appeared puzzled by the unexpected role switch. The following statements during the first practice session starkly illustrate the overall atmosphere on this day: I don't like to see my essays corrected everywhere by my classmates. There might be different ways of writing, and are we knowledgeable enough to know them all? I don't think that we are qualified.
Why do we have to comment on each other's work when our teacher can do a much better job? Besides, we are very much afraid of making mistakes [by giving wrong feedback], and doing our assignments [writing essays] is quite a heavy amount of work already. I cannot say I am confident in giving advice. I have never done that before.

Preference for the face-to-face mode
Despite the distrust and resistance that the participants exhibited when introduced to the peer-review process, the instructor could sense a lively interest among the participants once the Writing Weeks started. They provided extensive feedback and helped peer writers brainstorm ideas, and some of them appeared to genuinely enjoy the activity. One student even volunteered to write a second draft and send it to the instructor later that day before it was officially announced that they were to revise their first drafts. These observations were supported by the factor analysis results (factors 2 and 3) and interview responses, with the majority of the interviewees reporting that they had enjoyed the peer-review activitiesparticularly the face-to-face mode-for reasons outlined in the subsections below.

Benefits of verbal communication
Approximately two-thirds of the interviewees (67%) expressed a preference for the faceto-face mode. The most commonly cited reason for this preference was that the reviewers could confirm directly with the writer whether they had properly interpreted a questionable phrase/sentence, allowing them to provide useful suggestions. Considering that most of the participants appeared overly concerned about the accuracy of their feedback (Wen & Clément, 2003), it is unsurprising that they preferred the face-to-face mode, in which they could be 'more' correct.
I particularly liked the verbal interaction in the face-to-face mode because it was easier to give feedback. There was this one phrase I couldn't understand [in my group member's essay]. When I was about to ask the writer, the other reviewer told me about the very phrase I was wrestling with and asked for my opinion about her guess about the meaning. I could suddenly understand the meaning, which the writer confirmed to be correct.
Understandably, students who preferred the face-to-face mode complained about the lack of communication with the writer in the anonymous mode. As reviewers, they could not fully understand the text, which led to frequent text appropriation by confusing the writer's purpose in writing the text with their 'own purposes in commenting' (Sommers, 1982, p. 149).
I do not think they [the reviewers] ever understood me [my writing] in the anonymous mode. They changed my essay in a weird way, and I sighed reading the comments... The blind nature of the mode encouraged me [as a reviewer] to seek a more authoritative explanation from the teacher, and I don't think that's what peer review is for.
One of my anonymous reviewers wrongly corrected a few sentences that were grammatically correct. Frankly, I was annoyed. If I cannot argue back [to the negative comments by the reviewers], they won't know I could be right here and there. It's unfair.
Indeed, since there was no way for the anonymous reviewers to confirm the intended meanings of confusing parts of the texts, they continually asked the instructor to translate the text into Chinese or confirm their guesses, although he repeatedly refused. The students found it burdensome to play the reviewer role in the anonymous mode because it was emotionally demanding to tolerate their uncertainty about the correctness of the feedback that they provided. I enjoyed the peer-review activity in general. However, I found it not at all pleasant to have to guess the meaning of others' writing again and again. There were too many ambiguous sentences.
One of the interviewees recounted that he had to stop providing feedback halfway through the anonymous peer review because he could not make sense of the overall flow of the text. He explained why he had to withdraw and terminate the task before the time had expired: I could not understand the essay. After a certain point, I wasn't even sure whether it was [written in] English at all. I caught a glimpse of my partner's paper. Blank. He seemed to be struggling as much as I was. There was nothing I could do. I didn't even want to underline anymore because I didn't want to break the entirety of it [being totally incomprehensible]!

Peer feedback beyond error correction
Approximately half of the interviewees reported that the verbal interaction with their group members helped them to acquire a new perspective about the topic at hand. The participants were observed vigorously negotiating meaning and working in collaboration with their peer reviewers on the content of their essays (factor 4) (Villamil & de Guerrero, 2006). Further, approximately one-third of the interviewees, both as a reviewer and a reviewee, expressed that negative feedback was less needed during the face-to-face peer review because they could work collaboratively to negotiate meaning and make corrections during the oral discussion session (Hyland, 2000). I thought peer review was only about catching grammar mistakes. However, I learnt a lot more from the discussions with my reviewers. They taught me that I need to write from my reader's perspective because failing to do so can lead to more serious problems than grammar errors. I never knew we could help each other this much.
Through the face-to-face discussions, I could identify the parts of my essay that made sense only to me. I worked mostly on these parts in my revision, which was half completed during the group discussion. . . . Me and the other reviewer laughed so hard reading our group member's essay -he expressed almost everything in Chinglish! [Through peer-review activities] I finally got to understand the direction we need to progress towards-we should make our writing simple, and we should never assume that others know what we know.
While the more advanced students appeared to attend equally to the language and content aspects of their writing, those who self-identified as low-proficiency writers in the language background survey tended to focus primarily on the grammatical aspects. These participants were conservative in making content-related comments and did not seem to welcome such comments from others. I was embarrassed when my reviewers said I needed to strengthen my argument. Unlike grammar corrections, such [content-related] comments were difficult to implement because there are no correct answers. I couldn't change much after all [in terms of the content in the revised draft].

Reluctance to provide negative feedback
In addition to the participants' increased interest in and generally positive attitudes towards peer review, certain affective factors -not always culture-related-seemed to have modulated the peer-feedback process, albeit to varying extents across the participants. For instance, presumably because of the indirect nature of Chinese interactions, some students were stubbornly resistant to providing negative comments face to face (factor 1), while other students appeared to be quite casual about giving and accepting criticism. Overall, the participants' foreign language anxiety seems to have been one of the major issues affecting their peer-review performance, as routinely cited in L2 studies on a range of cultural groups, including Japanese (Dwyer & Heller-Murphy, 1996) and Turkish (Savaşçı, 2014) L2 learners.
Regardless of the peer-review mode, a tendency was observed among some reviewers to place question marks even on obvious grammar mistakes and to avoid confronting their peers about more subjective aspects of writing, such as the logical flow of thoughts. When these reviewers engaged in the face-to-face peer review, the conversations within the group naturally avoided such areas. The reason underlying this propensity to 'beat around the bush' might be that students wanted to save either their peers' faces or their own face in case their theories were rejected. It could equally be attributed to anxiety about giving incorrect feedback.
In the anonymous mode, the students' concerns about saving others' faces seemed to have been somewhat alleviated, and growing levels of criticism were observed among the reviewers. In particular, when the criticism was acknowledged by the other reviewer, the students' ability to offer criticism seemed to gain momentum. However, in the absence of interaction with the writer, the increased ease of offering criticism and the validation of the reviewers' opinions did not always translate into increased feedback accuracy.
Since there were only two of us during the anonymous peer review, the other reviewer was the only person we could turn to. Most likely because of that, there was huge overlap in the feedback [in both of my essays reviewed anonymously] from the two reviewers. What was funny was that they sometimes made identical wrong comments. When your essays are reviewed by two low-level [anonymous] reviewers, the result can be disastrous.
However, approximately 20% of the interviewees reported that they did not feel or perform differently in the different review modes. While most of them barely managed to articulate the cause of this propensity and opted to answer with 'I don't know' or a similar comment, one interviewer replied, With my name, without my name [revealed as a reviewer identity], it's the same me. That never changes. What's important is that I fulfil my duty.

Refusal to claim authority
Another observation made by the instructor was that regardless of the review mode, most peer reviewers refused to claim authority (Carson & Nelson, 1996). Unless they were completely sure of the accuracy of their feedback (such as placing an '(e)s' after a verb in the simple present for a third-person singular noun), they sought confirmation from the teacher. Throughout the activities, the instructor was continuously called upon to answer questions such as, 'What do you think this should be changed to?' or 'Do you think this part is correct?'. After receiving an answer from the instructor, the students often responded contentedly, 'Oh, I thought so too'.
In other words, the students were not asking questions; they were asking for confirmation that the feedback that they were about to provide was correct. Most of the participants seemed to hold a view that teachers are the repository of knowledge and input, a status that cannot be challenged. This code of conduct that emphasises status and hierarchy -or obedience to authority-explains the participants' initial resistance to the activity. Since Chinese students tend to submit to authority in the process of learning (Wen & Clément, 2003), involving the teacher in the peer-feedback process by seeking constant oversight is deemed to have helped the students to overcome their discomfort with their roles as reviewers.
At one point, my opinion [as a reviewer] differed from the other reviewer's. I knew I was right, but I didn't want to confront him. It would make both of us very, very uncomfortable. I lied that we'd better ask the teacher because I was unsure [whose opinion was correct]. When the teacher confirmed that I was correct, I felt so content.

Lingering resistance to peer review among low-level learners
The students with low English proficiency appeared to be fairly concerned that their limited language ability would adversely affect the effectiveness of peer review (factor 5), echoing the findings of previous studies on peer review with L2 learners in general, not just with Asian students (Allen & Mills, 2016;Lundstrom & Baker, 2009;Yu & Hu, 2017). During the interviews, four interviewees (13%) responded that they preferred the anonymous mode. They expressed their honest reservations about the limited contributions that they could make: As a student who is below average [in English skills], I feel sorry for my classmates. They gave me genuine and persuasive advice, which was very helpful in generating a revision. Regrettably, I could not do the same for them. Even when I spotted problems, I refrained from giving feedback because I wanted to keep a low profile. That felt like the right thing to do.
It was painful to sit next to my group members and say nothing. To be honest, I could barely finish reading the paper on some occasions, let alone give comments. I felt useless. I think the teacher should give feedback, not us.

ConCluSionS And impliCATionS
Because saving face is such a strong motivating force in China, it was expected that anonymous peer review would help to reduce Chinese students' reluctance to provide con-structive feedback to their peers. This theory was confirmed only marginally. As discussed previously, the factors underpinning the preference for the face-to-face mode constituted approximately 32% of the variance, while the one for the anonymous mode accounted for approximately 18% of the variance. These statistical results were supported by the thematic trends observed during the interview sessions, with 67% of the interviewees stating that they preferred the face-to-face mode to the anonymous mode. Equally unexpectedly, the participants were concerned about saving face by playing their newly assumed roles successfully (i.e., by providing correct feedback) rather than by avoiding offering criticism, as initially assumed.
The Chinese participants who exhibited a preference for the face-to-face mode cited similar reasons as their Japanese counterparts described in Kim (2019): Close verbal interaction with their peers was helpful in improving their writing, and it was easier to perform the reviewer's role. In addition, the fact that the 13% of the interviewees who reported favouring the anonymous mode were among the weakest of all of the participants lends support to the claim that sociocultural attributes might not be the sole factor influencing Asian students' reluctance to provide negative feedback.
Most of the stereotypes of Chinese L2 learners identified in the relevant literature were generally confirmed in this study. The students' lack of confidence in claiming authority, for instance, echoes Kumaravadivelu's (2004) view that such behaviour is inconsistent with hierarchical ideals, which represent an underlying set of values in China. Similarly, the participants' refusal to make error corrections independently or to confront their peers to resolve disagreements could be explained in light of the group orientation of Chinese culture, in which the avoidance of face-threatening acts is regarded as virtuous in any collective endeavour (Allaei & Connor, 1990).
Throughout the peer-review process, the instructor could sense intragroup dynamics in which the members found it extremely difficult to initiate a conversation (Keaten et al., 2000) for reasons such as 'nobody else said anything yet' or 'I was unsure whether my opinion was correct'. As is well illustrated by one interviewee's confession, 'I feared becoming a laughingstock by giving incorrect feedback', it seems that most of the Chinese EFL learners' obsessive fear of making mistakes (Littlewood, 2004;Tsui, 1996) continued to haunt them even in the anonymous mode. In particular, the students dreaded precipitating disunity or conflict within the group by making direct error corrections, which they tactfully substituted with question marks. Given that indirect feedback can sometimes foster language learning by requiring learners to be more active in their responses to indicated errors (Ferris et al., 2013), indirectness in Chinese culture seemed to have inadvertently helped the participants to reinforce their grammatical knowledge and encouraged autonomous learning behaviour.
The themes that emerged from this study might be construed in terms of vulnerability within the group (Carson & Nelson, 1996). However, the direct conversations with the participants suggested another possibility: Ironically, the lack of an authority figure (the teacher) might have been a major obstacle to taking the initiative in the supposedly student-centred activity. As shown in the previous section, the participants frequently sought oversight from the teacher before proceeding to the next part of the task. Although this propensity might have resulted from Chinese L2 students' enthusiasm for grammaticality (Li & Liu, 2011) or lack of confidence (Tsui, 1996), it is also believed to be rooted in the Chinese tradition of seeing oneself as a part of a relational hierarchy, establishing the general context within which Chinese students' stubborn refusal to assert autonomy (out of respect for authority) can be understood (Ho & Crookall, 1995). As the instructor's field notes described, The tension in the classroom is almost palpable. The students seem to panic when they are asked to enter the realm of what is considered 'teacher power', even when they know that it's part of the task.
These observations bear important pedagogical implications because an understanding of the challenges faced by students during the peer-review process could help instructors to address these issues more effectively. As suggested in Hu (2019), it is essential to create a milieu where cultural influences potentially in tension with peer feedback are negotiated proactively and productively by adopting pedagogical approaches that recognise a cultural sensitivity of a particular L2 learner group and by equipping students with 'effective strategies for addressing their culturally rooted concerns' (p. 59). To this end, writing teachers need to help students understand 'what the purpose of peer review is' and 'what it is not'. As attested in both Hu (2005) and Littlewood (2009), Chinese students not only respond well to peer review when the procedure and its purpose are well explained but also welcome interactive learning if it is implemented in supportive ways. If peer review is mistakenly approached as a proofreading exercise rather than a collaborative learning opportunity involving brainstorming and identifying one's strengths and weaknesses (Topping, 2009), students are unable to fully benefit due to their distrust of their peers' ability to provide 'quality feedback' (Hyland & Hyland, 2006a, p. 91). Therefore, thorough preparation for peer review is necessary, which in turn could help students understand the type of help to expect from their peers. It might be equally helpful for teachers to model communicative and interpersonal strategies that can convey critical feedback in a way that does not strain positive group relations (Berg, 1999;Hu, 2005Hu, , 2019. In addition, teacher feedback should follow the peer-review process (Rollinson, 2005). When students understand that peers and teachers can act as complementary sources of feedback, they become less overcautious about providing feedback and view the activity much more favourably.
Overall, it is difficult to determine whether the findings of this study confirm the hypothesis that the authors attempted to verify. While some findings supported the initial assumption that face matters in Chinese EFL students' peer-review interactions, part of the observed phenomenon (i.e., a preference for the face-to-face mode) and the reasons underlying this preference deviated from expectations. Moreover, it must be acknowledged that factors -such as the level of familiarity within the peer-response group, the level of understanding of the topic (Davies, 2006), the L2 proficiency of the learners themselves and the relative L2 proficiency of peer writers/reviewers (Allen & Mills, 2016)-that could have a bearing on learners' preference for and performance in a certain review mode were not controlled in this study. Since this was the participants' first encounter with peer review, their (un) readiness to participate in classroom interaction (Brammer & Rees, 2007) must have also influenced their initial reluctance to participate in the activity and likely had a lasting influence on their response to it. Future studies might design research considering these factors and possible interactions among them. Given that 'human cognition and learning is rooted in cultural environments and influenced by the discursive practices of their social groups' (Hyland & Hyland, 2006b, p. 12), the effects of anonymity on L2 learners' peer-review performance need to be examined with participants from non-collectivist cultures to further verify whether some students' perceived difficulty in providing negative feedback stems from sociocultural attributes. As Porte (2012) notes, replication studies are essential to confirm the outcomes of L2 learning and teaching studies because 'other variants can address the robustness or generalizability of a study by the introduction of further variables or contexts' (p. 6). Although we are yet to address an array of factors to extend our understanding of the influence of the sociocultural factors on peer interactions, one thing seems clear: It might be a hasty generalisation to describe Asian students collectively as reluctant peer reviewers, and Chinese students are no exception.