A comparative analysis of language choices in written peer feedback provided by high- and low-proficiency Japanese EFL students

Abstract

This study reports on part of a larger research project that investigates the role of students’ first language in learning English as a foreign language (EFL) in the context of Japanese higher education. Despite the prevalent use of peer review in EFL writing classes, students’ use of different languages and the effects of English proficiencies on collaborative interaction as mediated in written peer feedback remain underresearched. To add to the rather sparse literature, this study comparatively analysed peer feedback provided by high- and low-proficiency Japanese EFL students and conducted stimulated recall interviews to determine the factors affecting their language choices. The text analysis presented a stark contrast in students’ language choices stemming mainly from a difference in the type of feedback provided by each group: ‘corrective feedback’ versus ‘complimentary overall comment’. The stimulated recall interview accounts suggested that this polarization is attributable mostly to the proficiency gap between peer dyads and the affective need to contribute to the peer-review process. Inconsistent with the reports of previous studies that peer review benefits all students involved, the advanced-level participants voiced serious doubts about its effectiveness.

Keywords

Japanese EFL students L2 writing language choice peer review written peer feedback

I Introduction

Peer review is widely practiced as a learning activity in tertiary second language (L2) writing classes. To perform peer review, students read other students’ writing and make suggestions for improvement in the context of reciprocal reviewing. This enables them to assume an active role in their own learning (Liu & Sadler, 2003; Nelson & Carson, 1998), changing their position in L2 learning from a peripheral to a more centered one (Zheng, 2012). The focus of this collaborative activity is on writing as a vehicle for communication in an ongoing process rather than a formal product, and thus supports the shift of emphasis from a product to a process in writing instruction (DiPardo & Freedman, 1988).

Sociocultural theory – a theoretical framework developed by Vygotsky (1978, 1986) that emphasizes the role of social interaction in L2 learning and development (Lantolf & Thorne, 2006) – has been applied to various fields of L2 studies, including the sociocognitive dimensions of peer review (Villamil & de Guerrero, 1996). According to the Vygotskyan perspective, cognitive functions are formed on the basis of interactions with more skilled individuals. However, recent studies have challenged this view, suggesting that scaffolding occurs among peers and not just with guidance from the more knowledgeable other (e.g. de Guerrero & Villamil, 2000; Tsui & Ng, 2000). Barnard et al. (2015) and Zhu (2001) have shown that engaging L2 students in peer review contributes positively to their L2 learning, with peer-to-peer collaboration constituting a form of mutual scaffolding. Although L2 students are likely to be less familiar with conventions of academic writing than their teachers, peer feedback has been found to complement teacher feedback, each focusing on different aspects of writing (Yang et al., 2006). As Villamil and de Guerrero (1996) noted in their examination of social interaction in the L2 writing classroom, dyadic peer review offers an opportunity for bilateral participation and enables both peers to teach and learn how to revise.

Despite the prevalent use of peer review in L2 writing classes, students’ language choices in written peer feedback and the factors underlying them remain underresearched. Among a few studies investigating the topic, the earliest was conducted by Yu and Lee (2014). They explored the effects of first language (L1) use by analysing the written comments generated by Chinese learners of English as a foreign language (EFL) and showed that L1 and L2 are used for different purposes, such as L1 use to address issues concerning content and organization and L2 use to provide corrective feedback. Pointing out the absence of personal, socializing comments provided in L1, Williams (2018) added to this finding and suggested that EFL students subconsciously distinguish between roles of L1 and L2. Studies by Yu (2016) and Yeh (2018), in which the participants were required to use either L1 or L2 in a given condition, similarly reported that allowing L1 use helps learners to mediate cognitive resources in working memory, facilitates ability to produce more specific comments on content, and contributes to meaning negotiation and mutual scaffolding. Taken as a whole, L1 use was shown to constitute a major peer feedback strategy that improves EFL students’ understanding of the texts and facilitates group interaction (Yu & Lee, 2016a).

More recently, Kim and Chang (2022) examined feedback comments produced by Japanese EFL students, who were allowed to switch between L1 and L2 as necessary, and conducted stimulated recall interviews to determine the factors affecting their language use. Analyses of the interview data indicated that the participants’ language choices were influenced by the interaction between inter- and intrapersonal factors, such as the (relative) L2 proficiency of the peer reviewer and the ease of mitigating negative feedback in L1. These findings were supported in Kim’s (2023a, 2023b) subsequent studies, in which the participants were found to have attenuated negativity in their critical but constructive comments by using mitigation strategies in L1 at the discourse, syntactic, lexical, and paralinguistic levels.

Although peer-review activities may benefit all students regardless of their L2 proficiencies, it cannot be denied that the degree of engagement or the contribution to the communicative task is susceptible to the L2 proficiencies of the parties involved (K. Hyland & Hyland, 2006). Analysing how Japanese university students’ L2 proficiency modulates the peer feedback process, Allen and Mills (2016) showed that high-proficiency reviewers made a significantly larger number of suggestions and that their suggestions were incorporated more often than those from lower-proficiency reviewers. The number of suggestions was highest when high-proficiency reviewers reviewed texts written by low-proficiency writers. Zhou et al. (2022) similarly reported that while high-proficiency learners tended to view themselves as (pro)active agents in the feedback process, low-proficiency learners indicated unreadiness to engage in peer feedback, which stemmed mainly from their lack of confidence in providing useful feedback on global issues. Inconsistent with the findings of these two studies, Chinese EFL students of lower proficiency in Wu’s (2019) study tended to make more suggestions and clarification requests than their stronger counterparts. This discrepancy could be attributed to multiple factors, such as participation variables (e.g. whether contributions to interaction require little or extensive negotiation), participant variables (e.g. whether learners are of the same gender), or affective variables (e.g. whether learners are willing to communicate) (for details, see Robinson, 2007). As Allen and Katayama (2016) suggested, it is equally possible that the students’ perceptions of their peers’ and their own ability exerted a significant influence on the quantity and type of feedback.

The studies outlined above provide meaningful observations on the impact of ‘L1 use’ and ‘student’s L2 proficiency’ in the peer-review process and outcomes. However, they examined only one of these two independent variables, and an interaction between them – i.e. the patterns of peer interaction of different L2 proficiency groups in terms of their language use – has not been addressed sufficiently. Considering the pivotal role of language as a cognitive instrument that mediates collaborative interaction (de Guerrero & Villamil, 1994), the interrelatedness of language choices in an interactive discourse during peer review and the effect of the proficiency gap among peers due to inevitable group formation of learners of differing L2 proficiencies in authentic L2 classroom settings (Connor & Asenavage, 1994; Watanabe & Swain, 2007) needs more scholarly attention.

To broaden our understanding of the cognitive and social functions of L1 in collaborative learning activities (DiCamilla & Antón, 2012) and the role of L2 proficiency in the process, a multiphase research project is ongoing in the context of Japanese higher education that combines ethnographic data with data from other sources, such as written peer feedback and subsequent drafts, for triangulation. This article reports on part of the first phase of the project that explores the effect of L2 proficiency on the use of L1 and L2 in the peer-review process and the factors that affect such language use. Specifically, this study seeks to answer the following research questions:

• Research question 1: When given a choice between L1 and L2, to what extent do high- and low-proficiency Japanese EFL students use L1 to mediate their written peer feedback?

• Research question 2: What types of written feedback are given in L1 and L2 by high- and low-proficiency Japanese EFL students?

• Research question 3: What factors influence high- and low-proficiency Japanese EFL students’ language choices when providing written peer feedback?

• Research question 4: How do high- and low-proficiency Japanese EFL students perceive the use of L1 in written peer feedback?

II Methods

1 Participants and setting

This study was conducted in ongoing English writing classes at a private Japanese university, and data were collected over two years. Of the 55 Japanese EFL students who participated in the first phase of the research project, data collected from six students are presented in this study. The 55 students could be roughly divided into four L2 proficiency groups: advanced (AC) (n = 3), intermediate (n = 23), low-intermediate (n = 26), and low (LW) (n = 3). The six participants were from the AC and LW groups.

The students were recruited from three sessions of English Writing 5 and one session of English Writing 6 that were taught by the same instructor. These classes constituted a two-course sequence that was compulsory for international studies majors in their third year (aged 20–21 years). Each class met once a week for 100 min over 14 weeks. Of the larger number of students enrolled, those who missed a peer-review activity or took the prerequisite course – English Writing 4 or 5 – from a different teacher were excluded to ensure that their familiarity with peer review (Hansen & Liu, 2005), feedback training (Hu, 2005; Rahimi, 2013; Zhang & Yu, 2022), or teacher feedback practices (Yu & Hu, 2017a; Yu & Lee, 2014) did not affect their performance. Since the participants had engaged in peer review in the prerequisite course, they were familiar with peer-review activities and their rationale when they participated in the study. The instructor frequently used proofreading symbols when providing teacher feedback, so the students were also familiar with those symbols.

All members of the AC group, which comprised two female students and one male student, came from different sessions. Of the three LW group members (all male students), two came from the same session. Examination of peer-review group formation confirmed that the two LW group members from the same session had never been grouped together. That is, the AC and LW groups reviewed essays written by relatively lower- and higher-level peers, respectively. Peer-review groups were formed randomly in an authentic classroom environment, not in a purposefully manipulated condition in which the advanced-proficiency participants were selected and paired with the weakest counterparts to heighten the impact of proficiency in peer interactions. In the following sections, the participants are referred to as a combination of their proficiency group and rank, with AC-1 and LW-3 representing the strongest and the weakest.

2 Measuring participants’ L2 proficiency

The proficiency of the students was measured on the basis of their final exam essays for the prerequisite course that they took in the immediately preceding semester. Using Jacobs et al.’s (1981) Scoring Profile – an analytic scoring rubric that assigns four broad ability bands in the five sub-domains of writing ability (i.e. content, organization, vocabulary, language use, and mechanics) – two raters (the author and a college writing instructor with 12 years of experience) scored the students’ final exams independently. The Jacobs et al.’s Scoring Profile was chosen because it is known as one of the most reliable rubrics for rating writing skills (Y.-W. Lee et al., 2008), particularly those of L2 writers who may have uneven profiles of performance across different aspects of writing (Hamp-Lyons, 1991; Weigle, 2002). Before actual scoring, the raters met for practice scorings to ensure inter-rater consistency. They reviewed four sample student essays and scored them using the Jacobs et al.’s Scoring Profile. The inter-rater reliability was tested, and the correlation coefficient was fairly high (r = .90). The students’ proficiency levels were determined by averaging the total scores given by the raters: advanced (⩾90), intermediate (70–89), low-intermediate (50–69), and low (⩽49).

3 Procedures and data collection

Throughout the 14-week semester, the students were given three major assignments of writing a five-paragraph essay in different genres (400–450 words) and participated in peer review in class when their first drafts were due (Weeks 4, 8, and 12). One week prior to the first peer review, they received a 50-min training that was designed drawing mostly on Hu’s (2005) peer feedback training activities. Using the ‘Peer Review Checklist’ (see Appendix A), the instructor first explained what features to look for during peer review – i.e. content, organization, vocabulary, language use, and mechanics, which correspond with the five sub-domains in the Jacobs et al.’s Scoring Profile, according to which all of their assignments and the final exam essays were graded. Then, the instructor demonstrated an essay example written the previous semester by another student and explained how peer review should work. Following the demonstration and instruction, the students practiced providing feedback on a draft that the instructor provided (Reddy et al., 2021) and discussed their feedback in groups of three (or four if they could not form groups of three).

For each of the actual peer-review activities, the students prepared two copies of their assignments. The instructor randomly assigned them to groups of three to provide broader perspectives and writing styles than dyads do (Ferris & Hedgcock, 2014) and help the writer attain a second opinion, by which the inter-rater reliability of feedback could be strengthened (M.-K. Lee, 2015). Each group member independently read and wrote comments on the other two group members’ essays for approximately 30 min each. Then, the two peer reviewers jointly provided oral feedback to each writer for 10 min, for a total of 30 min. (Analysis of the oral feedback is not in the scope of this study.)

When the task was complete, the two peer-reviewed drafts were collected from each participant. In total, 36 peer-reviewed drafts were collected (six drafts per participant) when three rounds of peer reviews were complete at the end of the semester. Although the instructor encouraged the students to use L2 during peer review, they were assigned with learner agency – ‘the capacity to establish personal goals, set up conditions, and choose the means that best suit their motives or needs in learning’ (Villamil & de Guerrero, 2006, p. 28) – so that their choice of and switching between L1 and L2 during the process could be accounted for. In the following sections, the sources of the written feedback data will be specified as ‘R-1’, ‘R-2’, and ‘R-3’ for those derived from the first, second, and third round of peer review, respectively.

To yield a rich body of data about factors contributing to Japanese EFL students’ language choices in their written feedback and their perceptions of peer review, stimulated recall interviews were conducted at the end of the semester (for sample questions, see Appendix B). Looking at the peer-review comments they provided and received, the interviewees were asked to recall and reflect on their peer-review process. Each of the individual interviews lasted for approximately 50 to 80 min. Based on the observation that presenting events can differ greatly depending on the language used (Richards, 2009), the interviews were conducted in the interviewees’ L1, Japanese.

4 Data analyses

To answer research questions 1 and 2, written feedback on the 36 peer-reviewed drafts generated by the AC and LW groups from the three rounds of peer-review activities was examined. Feedback points were first divided into ‘corrective feedback’ and ‘feedback commentary’. Corrective feedback comprised both direct corrections of and suggestions for correcting grammatical, lexical, and mechanical errors (Van Beuningen et al., 2012). Formatting issues, such as indentation and line changes, were additionally incorporated into the classification scheme. Feedback commentaries that are defined as ‘in-text comments [provided] in the form of annotations on students’ work’ (Derham et al., 2022, p. 896) encompassed F. Hyland and Hyland’s (2001) three broad feedback functions of praise, suggestion, and criticism.

Following the coding framework in Kim and Chang (2022), feedback commentaries were sorted into five focus categories: language use; content; organization; administration, which concerns meeting the task requirements; and overall comment for comments that did not fall into the aforementioned categories. The feedback commentaries were investigated in terms of the number of issues (categories) they concerned and were counted as a single feedback point or multiple points accordingly, regardless of the number of sentences. The coded feedback points were then further categorized according to language use: L1, L2, L1 and L2 mixed within a sentence, and proofreading symbols.

Using the pre-established codes described above, the author deductively analysed the corrective feedback and feedback commentary data. To increase the validity of the analysis, the coding process was repeated three times at intervals of approximately two weeks. Percentage agreements with previous coding ranged between 97.5 and 100.0, indicating excellent intra-rater reliability.

To address research questions 3 and 4, the transcribed recordings of the stimulated recall interviews were analysed. Following conventions of thematic analysis (Braun & Clarke, 2006), the interview transcripts were read iteratively and recursively to capture the salient themes in the data. The analysis took an inductive approach without using a pre-existing coding scheme.

III Results and discussion

1 High- and low-proficiency students’ language choices in written peer feedback

The analysis of the written peer feedback data yielded 561 feedback points, of which 488 (87%) were made by the AC group. The remaining 73 (13%) were made by the LW group, which roughly corresponded to a mere 15% of the total amount of feedback provided by the AC group. This substantiates the findings of Allen and Mills (2016) that reviewer proficiency strongly influences the number of peer-review comments, with the difference being most apparent when higher-proficiency reviewers are paired with lower-proficiency writers. Table 1 summarizes language use in these feedback points provided by the AC and LW groups.

Table 1.

Language use in written peer feedback.

	Corrective feedback		Feedback commentary		Total
	AC	LW	AC	LW	AC	LW
L1	4 (1%)	8 (11%)	17 (4%)	24 (33%)	21 (4%)	32 (44%)
L2	299 (61%)	21 (29%)	6 (1%)	2 (3%)	305 (63%)	23 (31%)
L1 and L2	7 (1%)	–	1 (0%)	–	8 (2%)	–
Symbols	154 (32%)	18 (24%)	–	–	154 (32%)	18 (25%)
Total	464 (95%)	47 (64%)	24 (5%)	26 (36%)	488 (100%)	73 (100%)

Notes. Proficiency groups: AC = advanced; LW = low. L1 = first language; L2 = second language. Percentages are rounded to the nearest whole number.

Of the 488 (100%) feedback points made by the AC group, 464 (95%) were corrective feedback and 24 (5%) were feedback commentaries. Among them, 21 (4%) were delivered in L1, 305 (63%) in L2, 8 (2%) in both L1 and L2, and 154 (32%) in proofreading symbols. The use of both L1 and L2 was made exclusively at the intrasentential level to reference part of the text that needed revision (see Excerpt 1), to illustrate examples for revision (see Excerpt 2), or to use terms learned in English (see Excerpt 3).

(1) good は形動詞を修飾するときは副 (AC-1, R-1)

(‘good’ is an adjective. When modifying a verb, [you need an] adverb)

(2) In conclusionなどがないのでつけた方がいいです。(AC-2, R-2)

(Because there is none [no transition word] like ‘In conclusion’, it would be better to add one.)

(3) topic sentenceと全く同じ文は X

パラフレーズを使わないとだめ (LW-1, R-1)

(You cannot use exactly the same sentence as the topic sentence

You should paraphrase)

Of the 73 (100%) feedback points made by the LW group, 47 (64%) were corrective feedback and 26 (36%) were feedback commentaries. Among them, 32 (44%) were delivered in L1, 23 (32%) in L2, and 18 (25%) in proofreading symbols. Unlike the AC group, the LW group did not mix L1 and L2 within a sentence regardless of the feedback type. Other noticeable differences were that the ratio of L1 use was noticeably greater in the LW group (44%) than in the AC group (4%) and that the percentage of L2 feedback provided by the AC group (63%) was approximately double that of the LW group (31%).

In terms of the feedback type, the AC group was shown to be preoccupied with microlevel features (e.g. grammar, vocabulary, and mechanics), exhibiting a similar pattern of the writing consultants in Chen’s (2010) study, who mainly served as proofreaders of local errors during writing consultation sessions. Similarly, Allen and Mills (2016) reported that peers with higher proficiency tended to focus on meaning-preserving and language-related issues rather than content-related ones.

2 Focus of written peer feedback and language use

Table 2 shows the results of the corrective feedback analysis. Of the 464 (100%) corrective feedback points generated by the AC group, 238 (51%) concerned grammar, 119 (26%) language use, 52 (11%) mechanics, and 55 (12%) formatting. Of the 47 (100%) corrective feedback points generated by the LW group, 12 (26%) were on grammar, 5 (11%) on language use, 17 (36%) on mechanics, and 13 (28%) on formatting.

Table 2.

Language use in corrective feedback.

	Grammatical		Lexical		Mechanical		Formatting		Total
	AC	LW	AC	LW	AC	LW	AC	LW	AC	LW
L1	–	–	–	–	–	–	4 (1%)	8 (17%)	4 (1%)	8 (17%)
L2	175 (38%)	4 (9%)	102 (22%)	5 (11%)	5 (1%)	12 (26%)	17 (4%)	–	299 (64%)	21 (45%)
L1 and L2	2 (0%)	–	3 (1%)	–	–	–	2 (0%)	–	7 (2%)	–
Symbols	61 (13%)	8 (17%)	14 (3%)	–	47 (10%)	5 (11%)	32 (7%)	5 (11%)	154 (33%)	18 (38%)
Total	238 (51%)	12 (26%)	119 (26%)	5 (11%)	52 (11%)	17 (36%)	55 (12%)	13 (28%)	464 (100%)	47 (100%)

Notes. Proficiency groups: AC = advanced; LW = low. L1 = first language; L2 = second language. Percentages are rounded to the nearest whole number.

Not surprisingly, the vast majority of corrective feedback was provided in L2 (or proofreading symbols), constituting 97% and 83% of the corrective feedback points provided by the AC and LW groups, respectively. L1 alone or in combination with L2 was used only sparingly when the participants presented their thinking in the form of a suggestive question (see Excerpt 4), a suggestive statement (see Excerpt 5), or a direction for clear formulation of the identified problem (see Excerpts 6). (Texts from the peer-reviewed drafts are set in italics, and boldface is added to the problematic parts that the reviewer identified.)

(4) As far as I can , adults and children can play smartphone games . . .

knowの方がいいかも? (AC-2, R-3)

(Maybe it’d be better to use ‘know’?)

(5) First of all, it is an alarm that I think is a function that everyone uses .

I think everyone uses とかに変えてtopic の後にもってくる方がいいと思う. . . (AC-3, R-3)

(I think it’d be better to change [the underlined sentence] to ‘I think everyone uses’ or the like and to place it next to the topic sentence . . .)

(6) 改行はしない。(LW-2, R-1)

(Do not start on a new line [until the paragraph ends].)

Table 3 outlines the results of the feedback commentary analysis. The AC group generated 24 (100%) feedback commentaries, of which 1 (4%) related to language use, 9 (38%) to content, 7 (29%) to organization, and 5 (21%) to administration, and 2 (8%) was an overall comment. The LW group generated 26 (100%) feedback commentaries, of which 5 (19%) were on content, 2 (8%) on organization, and 1 (4%) on administration, and 18 (69%) were overall comments. No comments were made on language use.

Table 3.

Language use in feedback commentaries.

	Language use		Content		Organization		Administration		Overall comment		Total
	AC	LW	AC	LW	AC	LW	AC	LW	AC	LW	AC	LW
L1	1 (4%)	–	8 (33%)	5 (19%)	2 (8%)	2 (8%)	5 (21%)	1 (4%)	1 (4%)	16 (62%)	17 (71%)	24 (92%)
L2	–	–	1 (4%)	–	4 (17%)	–	–	–	1 (4%)	2 (8%)	6 (25%)	2 (8%)
L1 and L2	–	–	–	–	1 (4%)	–	–	–	–	–	1 (4%)	–
Total	1 (4%)	–	9 (38%)	5 (19%)	7 (29%)	2 (8%)	5 (21%)	1 (4%)	2 (8%)	18 (69%)	24 (100%)	26 (100%)

Notes. Proficiency groups: AC = advanced; LW = low. L1 = first language; L2 = second language. Percentages are rounded to the nearest whole number.

While both groups opted mostly for L2 rather than L1 when providing corrective feedback, the reverse was true for feedback commentaries, particularly for the LW group. Excerpts 7, 8, and 9 below show how the AC group used L1 and L2 in their feedback commentaries. All these commentaries were written within the body of the text beside the part the peer reviewer was commenting on and thus were quite specific.

(7) topic sentence の keyword を明確に (AC-1, R-3)

(Make the keyword in this topic sentence explicit)

(8) 文章が長くてわかりずらいです。(AC-3, R-1)

(Because this sentence is too long, it is difficult to understand.)

(9) If we tried and looked it up by ourselves, we can understand it in easier English.

So that things could be established among us.

なんかちょっと分かりづらいよね~ (AC-2, R-2)

(Somewhat, a little bit difficult to understand~)

Excerpts 10 and 11 below are from the LW group. Both were provided as a final comment in the bottom margin of the essay and were rather vague and superficial compared with those provided by the AC group.

(10) 短く内容がまとまっていて、とても良いエッセイでした。(LW-2, R-3)

(Because the contents are concisely organized, this is a very good essay.)

(11) 自分の意見ではっきりしていて、読んでいて分かりやすかった!!! (LW-1, R-2)

(Because you clearly expressed your opinion, it was easy to read and understand your essay!!!)

Interestingly, all LW group participants left overall comments mostly in the form of a complimentary remark. While only three such comments were made by the AC group – one in the form of praise of the organization of the essay (see Excerpt 12), which was counted as a feedback commentary on organization rather than an overall comment, and the other two in the form of an overall comment accompanied by a constructive suggestion on language use (see Excerpt 13) and content (see Excerpt 14) – the LW group provided 18 overall comments, constituting 69% of all feedback commentaries generated by this group. These comments were written almost exclusively in L1. Although LW-3 attempted to provide an overall comment in L2, he ended up erroneously translating the L1 comment he wrote on another paper he reviewed (see Excerpt 15) into L2 (see Excerpt 16).

(12) good hook! (AC-1, R-2)

(13) I like your writing. It would be much better if you make underlined sentences more simple. Thank you!

(14) [In response to an essay about the pros and cons of smartphone use in the classroom]

2つ目のアイデア面白いと思いました。しかし最後の文だけスマホの悪い影響について書いているので、最後の文の内容は良い面について書いた方がいいと思います。とても分かりやすくて良い文でした。(AC-2, R-3)

(I thought the second idea is very interesting. However, since only the last sentence deals with the negative influence of the smartphone [in a paragraph that asserts its benefits], I think it’d be better to change it to something positive. Your essay was very easy to understand and well written.)

(15) これはとても上手です! (LW-3, R-3)

(This is very good!)

(16) This is very well (LW-3, R-3)

In addition to complimentary remarks, a different type of overall comment was identified in the LW group data: self-reflection. These appeared as end comments in the bottom margin, and two of the three LW group participants left such comments, as exemplified below.

(17) とても良いエッセイでした。今後課題でエッセイを書く機会があれば、このエッセイを参考にさせていただきます。英語力の差を痛感し、とても情けない気分になりました。反省します。自らの英語能力の低さを知るいい機会です。今後、このようなことが起きぬよう精一杯頑張ります。ありがとうございます。 (LW-1, R-3)

(This was a very good essay. If I have a chance to write an essay as an assignment in the future, I will use this essay as a reference. I became keenly aware of the English proficiency gap [between you and me], which made me feel pathetic about myself. I reflect on myself. It is a good chance to realize my low English level. I will do my best to prevent this from happening in the future. Thank you.)

(18) [In response to an essay about whether students need to be allowed to use their L1 in L2 courses]

自分の考えと比較して、また違った意見、考えを知ることができ改めて言語の大切さを感じさせられました。(LW-2, R-2)

(Reading your paper made me realize the importance of language once again because I could compare my own opinion with [others’] different opinions and thoughts.)

Leaving a self-reflective comment might otherwise be viewed as an act of self-evaluation or metamemory. According to Hesse et al.’s (2015) framework of collaborative problem solving, metamemory refers to ‘knowledge about oneself’ (p. 46), a social processing skill required for any successful collaborative problem-solving groups. In addition, these comments exemplify one of the documented positive learning-by-reviewing effects in that commenting on peer drafts enhances reviewers’ awareness of their own compositions, which in turn helps them develop their own writing skills (e.g. Berggren, 2015; Lundstrom & Baker, 2009).

3 Factors operating on polarized feedback type

The comparative analysis of the feedback commentaries provided by the AC and LW groups showed that while the AC group left mostly corrective feedback (constituting 95% of all peer feedback they provided), well over one-third of the feedback provided by the LW group comprised feedback commentaries. Among these comments, approximately 70% were overall comments that were devoid of any mention of specific components of the reviewed work. Rather, they either took the form of encouraging observations on the whole, rather than a specific portion, of the paper they reviewed or were self-reflective in nature.

The thematic analysis of the stimulated recall interview accounts suggested that high- and low-proficiency Japanese EFL students’ language choices are mediated to a great extent by the type of feedback they provide. The AC group tended to focus on making line-by-line error corrections in L2, and they rarely left overall comments. The following two interview excerpts hint at the reasons underlying such feedback practice:

How can I not correct these errors when they are so obviously wrong? It’s better to correct them myself than making suggestions about how to do the job. They might make changes for the worse. If they still don’t understand my corrections, they can ask me face to face [during the oral discussion time]. Because it’s a writing class, if your grammar is poor, your readers get confused or misunderstand your point. No matter how interesting your ideas are, they are useless unless you convey them in a grammatical, I mean, in an understandable manner. (AC-1)

If you wish to receive only general comments and think those comments can help improve your English writing, you are naïve. . . . I didn’t have enough time to attend to other aspects of writing, like content. First and foremost, I had to focus on grammar because that can at least make incomprehensible sentences comprehensible. It’s a basic thing that when you compose a piece of writing, you have to make sure that it’s understandable, right? . . . Sometimes I could sense [during verbal communication with peer group members] that they don’t really understand my explanation, so I made it a rule to write down corrections first and explain verbally later. (AC-2)

Compared with their stronger counterparts, the LW group was attentive to leaving complimentary feedback commentaries in L1 or in single-word L2 transliterations, such as ‘ファイト!’ (/faito/). (Japanese people use a transliterated English ‘Fight!’ as encouragement, meaning ‘Way to go!’) The LW group participants articulated their reasons for leaving such comments as follows:

Because there often is a huge gap between me and my peer group members in terms of English ability, I tried to help but I really couldn’t do so. I know the purpose of doing peer review, but my English is not good enough to give feedback on other’s work. It makes me feel like ‘I’m always a receiver’. It makes me feel bad. So I left some nice words about the essay [in Japanese] because I really wanted to do something . . . anything. (LW-3)

It was painful to sit next to your group members who were vigorously reading your essay, leaving comments here, there, and everywhere. Although I pretended to be okay with the situation, I wasn’t, really. I had to look up the dictionary [to finish reading the essay, let alone giving feedback]. I was so confused. The peer-review activity was not good for me, although I don’t deny that I got a lot of help from my group members. I had to do something in return, so I wrote a note. (LW-1)

These observations indicate that low-proficiency L2 learners opt to leave feedback commentaries in L1 to fulfill their affective need to contribute to the peer-review process. Unlike these two participants, LW-2 explained that he left complimentary comments because of the affective motivation to encourage others, although this might be subordinate to L2 learners’ affective need as a whole:

If there’s an obvious mistake in the essay, I correct it and then write an encouraging message. If my partner does something correctly, I write an encouraging message that he or she deserves. I think that’s very important. Who wants to receive harsh comments from peers? None in Japan, I believe. Even if my partner’s essay is not perfect, I can still find a certain part that deserves praise, and mentioning that must definitely be motivating. (LW-2)

A number of studies (e.g. Crossman & Kite, 2012; Lundstrom & Baker, 2009; Topping, 2009; Yu & Lee, 2016b) maintain that peer reviewers benefit from the activity as much as, if not more than, those who receive the feedback because they can improve their own writing by ‘enhancing their reader awareness . . . and engaging in self-reflection and regulation as a result of commenting on their peers’ writing’ (Yu & Hu, 2017b, p. 187). However, these claims were not supported in this study, at least when judged from the AC group’s remarks about the experience of commenting on weaker peers’ work. AC-1 and AC-2 voiced their honest feelings as follows:

Some classmates’ English ability is good, but I haven’t been paired with them. When I am grouped with weaker students, I feel I’m not getting anything from them, while I’m giving everything to them. It’s kind of unfair. When weaker classmates review my essays, they only give very superficial comments like ‘good’ or simple, plain comments. What I really want is constructive criticism. As ironic as it might sound, those words of encouragement are discouraging. (AC-1)

Because we’re learning how to write English essays, you don’t just need to correct vocabulary or spelling mistakes, but you also have to [help peers] formulate correct sentences. I find it very rewarding to correct my classmates’ English essays. The thing is, I don’t think you can improve your English ability if you only get those nice words from your reviewers, although their kindness is heartwarming. Umm, it’s a difficult thing to say, but at least, that was not the case for me. (AC-2)

The effect of reinforcing feedback (i.e. positive or supportive comments) on enhancing task performance is inconclusive. According to F. Hyland (2000) and Tseng and Tsai (2007), for instance, peers reported that they considered praise comments to be more useful than critical comments because they may lead to improved writing quality through motivational effects. However, the AC group participants in this study indicated frustration with such comments, which lack meaningful cognitive information (Hattie & Timperley, 2007) and thus have little effect on improving actual writing performance (Cho & MacArthur, 2010, 2011). They apparently perceived detecting problems and offering solutions as more important than motivating peers through praise. This observation accords with that of Cao et al. (2019), in which the attitude of Chinese EFL students who placed little value on receiving peer feedback was shown to have stemmed from their experience of receiving mostly complimentary comments but no useful advice for making substantive revisions.

4 Perceptions of L1 use in written peer feedback

Regarding the reasons for using L1 in their written peer feedback, the AC and LW groups both identified L2 proficiency as a determining factor, although whose L2 proficiency they referred to was directly opposite. To the interview question concerning their use of L1 in providing feedback, the LW group typically responded as follows:

Because I am not good at English. I use Japanese, only Japanese, in my daily life because I am Japanese. The most important thing in peer review is to communicate with others, isn’t it? If you can’t use English well, then you have to think of an alternative way. If my English ability were better, I might give English comments, maybe. But as of now, I don’t think I’m ready. (LW-1)

I’m afraid that I may make mistakes in English, which will confuse my partner and cause inconvenience. If my partner is my teacher, I might attempt to leave comments in English because I can learn from her. I mean, she will teach me how to correct my mistakes [in my feedback]. However, I don’t want to run the risk [of confusing others] with my classmates. I really hate that feeling [of causing inconvenience]. Even if I pull it off and give English feedback by painstakingly translating what I want to say, my partner will painstakingly translate it back into Japanese. Why can’t we just use Japanese in the first place? (LW-2)

As the above interview excerpts illustrate, the LW group participants’ recourse to L1 pertains mostly to their own limited L2 proficiency. For the AC group, the opposite was the case, with peers’ limited English proficiency shaping their decision to use L1. In the three sets of interview data, which depicted almost the same view among AC group participants, one word was used repeatedly, clearly representing their position: the verb omoiyaru (思いやる, ‘to be considerate of’) in its conjugated forms or noun form, omoiyari (思いやり). Omoiyari, which does not simply refer to sharing feelings with others but entails anticipating their needs and putting a plan in place, is the cultural concept that underpins all aspects of daily life in Japan, emphasizing the importance of community and helping others (Longhurst, 2020).

What’s the point of using English if your partner doesn’t understand what you say? It’s meaningless. A waste of time, even. We need to be considerate of others before making [language] choices. Our job is to negotiate how to improve our writing, so it’s better to use Japanese with weak peers. Depending on your partner’s English level, sometimes you have no choice [but to use Japanese]. (AC-3)

Of the two overall comments left by the AC group, constituting a scant 8% of all feedback commentaries they provided, one was written in L1 (refer to Excerpt 14 in the previous section) on a draft written by one of the students (not included in this study) whose English proficiency level was low-intermediate. When asked about such language use, AC-2 responded as follows based on omoiyari:

In consideration of my partner who is not very good at English, I sometimes use Japanese to give a general comment at the end. I think it’s okay to use Japanese because it’s just an impression of the whole essay, not a specific suggestion about how to correct a certain part. . . . If my partner doesn’t understand what I mean [in my feedback written in English], he or she can check my general comments written in Japanese as a way to double-check. (AC-2)

A review of the relevant literature yields inconsistent results regarding L2 proficiency being a major contributing factor in the amount of L1 use. However, the personal accounts of the LW group about having to resort to L1 to offset their limited L2 skills support the facilitative role of L1 in task completion (e.g. García Mayo & Hidalgo, 2017; Zhao, 2010), while the accounts of the AC group align with the finding of Moore (2013) that higher-proficiency students use more L1 when paired with lower-proficiency peers than when paired with peers with similar levels of proficiency.

When asked about their perceptions of receiving peer feedback provided in L1 versus L2, members of both groups appeared indifferent to the reviewers’ language choices. Although such leniency is not surprising for the AC group, the LW group responded similarly, as illustrated below.

Both Japanese and English comments are okay. I don’t care. Japanese comments are a lot easier to understand. That’s for sure. But I can check [the meaning of English comments using] the Internet or dictionaries if I don’t understand them. That’s also a kind of English learning. Perhaps what’s more important is that I receive [direct] corrective feedback. I’d like my partners to point out exactly which parts are wrong and tell me how to correct the mistakes. If they give me long explanations [in the form of indirect corrective feedback], sometimes I don’t understand them. (LW-2)

I prefer to receive feedback in Japanese because my partner might be able to write only simple comments in English. So I prefer Japanese for overall comments. However, because it’s an English class, I think we should use English as much as possible. Otherwise, we really don’t have any chance to use English. In that respect, I think English [feedback] is also okay. When I receive English comments, it takes some time before I can understand them clearly, but that’s only because my English ability is weak. (LW-3)

Although the interview question did not directly involve the type of feedback he preferred to receive, LW-2 touched on the issue and expressed a clear preference for explicit corrective feedback as the latter parts of his interview excerpt illustrate above. This observation echoes the findings of relevant studies such as those by Kim (2019) and Kim and Lan (2021), who reported that EFL students with low proficiency tend to see the need for error correction as their main impetus for seeking peer feedback and thus do not solicit content-related comments from reviewers.

IV Conclusions and implications

This study examined the language use of advanced- and low-proficiency Japanese EFL students in the peer-review process as mediated in written peer feedback and the factors responsible for such practice. The text analysis showed that the AC group tended to take a ‘prescriptive stance’ (Mangelsdorf & Schlumberger, 1992, p. 247) when they worked on texts written by peers lacking grammar skills, thereby directing their attention to microlevel issues of form over those of the meaning or structure of the text. These observations were supported by the thematic trends that emerged during the stimulated recall interviews. Because these students provided form-focused corrective feedback mostly by means of correcting individual errors themselves, their feedback was given mostly in L2.

The interview accounts of the AC group resonated with Vygotsky’s general theoretical claim of the importance of the learner’s zone of proximal development (ZPD). In a peer-review context, if the ZPD of the peer writers is much lower than that of the reviewer, they may not benefit fully from the feedback, as they lack the capacity to incorporate the comments into a subsequent draft. This remark accords with that of Allen and Mills (2016), who suggested that learning may vary when the dyad’s ZPDs differ and that this is most pronounced when the levels are considerably different. A case study by Hamp-Lyons (2006) similarly reported that an L2 learner who lacked the capacity to incorporate her teacher’s feedback failed to benefit from the feedback that could have scaffolded learning.

In contrast, the essays reviewed by the LW group only randomly bore markings or comments within the body of the text. Instead, these students often left overall impression statements in L1 in a gentle, encouraging tone in the bottom margin, although their stronger counterparts perceived these comments as ‘superficial, or otherwise unhelpful’ (Bean, 2011, p. 295). While feedback commentaries accounted for only 5% of the total feedback points provided by the AC group, they constituted 36% of those provided by the LW group. In line with Kim (2023a), who explored actual and perceived needs for L1 use in understanding the dynamics of peer review groups in the Japanese higher education context, the interview data from the LW group exemplified the social functions that L1 serves in helping to externalize inner speech (Vygotsky, 1986) in ways that help construct a social space that facilitates the completion of the task by enabling learners to achieve a shared perspective on the task (Antón & DiCamilla, 1999). The self-reflective comments left by the LW group, in particular, reaffirmed the benefit that lower-proficiency writers are reported to gain when working with higher-proficiency peers, as they get to have the opportunity to read better-written essays in a similar genre, which in turn serves as a meditational tool that can increase learning (Allen & Mills, 2016).

Nevertheless, the benefits of peer review are inherently asymmetrical, and there was an obvious disparity between what mixed-proficiency dyads could offer to each other in terms of both the quantity and quality of peer feedback (Patchan & Schunn, 2015). Although quantitative comparison of feedback points provided by the AC and LW groups could not be made because each participant reviewed different texts, a stark contrast between the groups in their language choices for providing peer feedback is noteworthy. As shown in the previous section, only 4% of the total feedback points made by the AC group were given in L1, while L1 was used in as much as 44% of the total feedback points provided by the LW group. The close examination of the feedback data showed that this contrast stemmed from the difference in the main type of feedback provided by the AC group (i.e. corrective feedback) and the LW group (i.e. complimentary overall comments) driven by the practical need to improve peers’ writing accuracy for the former and an affective need to contribute to the process for the latter.

However, these results do not rule out the possible influence of other factors. As previous research has indicated, a myriad of elements operate on the way an individual interacts with others in peer-review activities. Studies explaining pattern variations in dyadic interaction within the framework of activity theory have shown that participants’ motives can shape their patterns of interaction (e.g. Storch, 2004; Yu & Lee, 2015; Zhu & Mitchell, 2012). Apart from L2 proficiency, intrinsic factors such as the participants’ peer feedback stances (Yu & Lee, 2015), affective states (Sato, 2017), or social relationships between the interactants (Storch, 2002; Storch & Aldosari, 2013; Yu & Hu, 2017a) were not taken into account, which also restricts the generalizability of the findings. Not only these factors but also a possible interaction between (one of) them and L2 proficiency should be examined in future studies.

Despite these limitations, its exploratory nature allows some implications to be drawn from this study. The findings suggest that L2 learners make flexible language choices that are influenced by factors such as peer writers’ and their own L2 proficiency (Kim & Chang, 2022; Yu & Lee, 2014) or the affective need to contribute to the process. These empirically identified factors that are responsible for L1 use, in particular, constitute a valuable source of information for L2 teachers in English as a second language (ESL) settings, who could directly address the constraints that their students are likely to encounter when conducting peer review in L2 by providing targeted support or mediate them by introducing coping strategies. This also applies to those teaching in EFL settings if they (intend to) conduct peer-review sessions exclusively in English because of personal beliefs or institutional language policies.

A clear consensus about the optimal proficiency pairing or the configuration of peer reviewers (Hu, 2005) for peer interactions has not yet been reached. Even if such a consensus did exist, it is unfeasible to form an entire class into such dyads/groups in an authentic classroom context. However, the findings indicate that pairing students at opposite extremes of L2 proficiency can be highly problematic for both parties. To facilitate smooth, productive peer interactions, teachers might at least need to ensure that the strongest and weakest writers in a class are not paired together. It is equally essential that teachers address the issue of the marginalization of low-proficiency L2 students by guiding the whole class to accept that ‘collaboration rather than correction is the goal of the writing group’ (Brammer & Rees, 2007, p. 81). By further exploring the challenges and frustrations that L2 students are likely to experience, L2 writing teachers in ESL and EFL settings alike will be able to target these issues beforehand and provide their students with scaffolding to facilitate their performance and engagement in the L2 context (García & Li, 2014).

Footnotes

Appendix A

Appendix B

Funding

The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Grants-in-Aid for Scientific Research (18K00854, 22K00747) from the Japan Society for the Promotion of Science.

ORCID iD

Sugene Kim

References

Allen

Katayama

. (2016). Relative second language proficiency and the giving and receiving of written peer feedback. System, 56, 96–106.

Allen

Mills

. (2016). The impact of second language proficiency in dyadic peer feedback. Language Teaching Research, 20, 498–513.

Antón

DiCamilla

F.J

. (1999). Socio-cognitive functions of L1 collaborative interaction in the L2 classroom. Modern Language Journal, 83, 233–247.

Barnard

de Luca

. (2015). First-year undergraduate students’ perceptions of lecturer and peer feedback: A New Zealand action research project. Studies in Higher Education, 40, 933–944.

Bean

J.C

. (2011). Engaging ideas: The professor’s guide to integrating writing, critical thinking, and active learning in the classroom. 2nd edition. Jossey-Bass.

Berggren

. (2015). Learning from giving feedback: A study of secondary-level students. ELT Journal, 69, 58–70.

Brammer

Rees

. (2007). Peer review from the students’ perspective: Invaluable or invalid? Composition Studies, 35, 71–85.

Braun

Clarke

. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3, 77–101.

Cao

Huang

. (2019). A qualitative inquiry into undergraduates’ learning from giving and receiving peer feedback in L2 writing: Insights from a case study. Studies in Educational Evaluation, 63, 102–112.

10.

Chen

C.W.-y

. (2010). Graduate students’ self-reported perspectives regarding peer feedback and feedback from writing consultants. Asia Pacific Education Review, 11, 151–158.

11.

Cho

MacArthur

. (2010). Student revision with peer and expert reviewing. Learning and Instruction, 20, 328–338.

12.

Cho

MacArthur

. (2011). Learning by reviewing. Journal of Educational Psychology, 103, 73–84.

13.

Connor

Asenavage

. (1994). Peer response groups in ESL writing classes: How much impact on revision? Journal of Second Language Writing, 3, 257–276.

14.

Crossman

J.M.

Kite

S.L

. (2012). Facilitating improved writing among students through directed peer review. Active Learning in Higher Education, 13, 219–229.

15.

de Guerrero

M.C.M.

Villamil

O.S

. (1994). Social-cognitive dimensions of interaction in L2 peer revision. Modern Language Journal, 78, 484–496.

16.

de Guerrero

M.C.M.

Villamil

O.S

. (2000). Activating the ZPD: Mutual scaffolding in L2 peer revision. Modern Language Journal, 84, 51–68.

17.

Derham

Balloo

Winstone

. (2022). The focus, function and framing of feedback information: Linguistic and content analysis of in-text feedback comments. Assessment and Evaluation in Higher Education, 47, 896–909.

18.

DiCamilla

F.J.

Antón

. (2012). Functions of L1 in the collaborative interaction of beginning and advanced second language learners. International Journal of Applied Linguistics, 22, 160–188.

19.

DiPardo

Freedman

S.W

. (1988). Peer response groups in the writing classroom: Theoretic foundations and new directions. Review of Educational Research, 58, 119–149.

20.

Ferris

D.R.

Hedgcock

J.S

. (2014). Teaching L2 composition: Purpose, process, and practice. 3rd edition. Routledge.

21.

García

. (2014). Translanguaging: Language, bilingualism and education. Palgrave Macmillan.

22.

García Mayo

M.P.

Hidalgo

M.Á

. (2017). L1 use among young EFL mainstream and CLIL learners in task-supported interaction. System, 67, 132–145.

23.

Hamp-Lyons

. (1991). Scoring procedures for ESL contexts. In Hamp-Lyons

(Ed.), Assessing second language writing in academic contexts (pp. 241–276). Ablex.

24.

Hamp-Lyons

. (2006). Feedback in portfolio-based writing courses. In Hyland

Hyland

(Eds.), Feedback in second language writing: Contexts and issues (pp. 140–161). Cambridge University Press.

25.

Hansen

J.G.

Liu

. (2005). Guiding principles for effective peer response. ELT Journal, 59, 31–38.

26.

Hattie

Timperley

. (2007). The power of feedback. Review of Educational Research, 77, 81–112.

27.

Hesse

Care

Buder

Sassenberg

Griffin

. (2015). A framework for teachable collaborative problem solving skills. In Griffin

Care

(Eds.), Assessment and teaching of 21st century skills: Methods and approach (pp. 37–56). Springer.

28.

. (2005). Using peer review with Chinese ESL student writers. Language Teaching Research, 9, 321–342.

29.

Hyland

. (2000). ESL writers and Feedback: Giving more autonomy to students. Language Teaching Research, 4, 33–54.

30.

Hyland

. (2001). Sugaring the pill: Praise and criticism in written feedback. Journal of Second Language Writing, 10, 185–212.

31.

Hyland

. (2006). Feedback on second language students’ writing. Language Teaching, 39, 83–101.

32.

Jacobs

H.L.

Zingraf

S.A.

Wormuth

D.R.

Hartfiel

V.F.

Hughey

J.B

. (1981). Testing ESL composition: A practical approach. Newbury House.

33.

Kim

. (2019). Japanese student writers’ perspectives on anonymous peer review. ELT Journal, 73, 296–305.

34.

Kim

. (2023a). ‘Who expresses an honest opinion as it stands in Japan?’: Cultural issues and Japanese L2 students’ experiences of face-to-face and anonymous peer review. Higher Education Research & Development, 42, 322–335.

35.

Kim

. (2023b). ‘Sorry, I don’t good English’: Japanese L2 students’ written peer feedback in the face-to-face and anonymous review modes. Journal of Writing Research. Available at: https://www.jowr.org/pkp/ojs/index.php/jowr/article/view/941 (accessed August 2023).

36.

Kim

Chang

C.-H

. (2022). Japanese L2 learners’ translanguaging practice in written peer feedback. International Journal of Bilingual Education and Bilingualism, 25, 1363–1376.

37.

Kim

Lan

. (2021). L2 writers’ perspectives on face-to-face and anonymous peer review: Voices from China. Porta Linguarum, 35, 149–164.

38.

Lantolf

J.P.

Thorne

S.L

. (2006). Sociocultural theory and the genesis of second language development. Oxford University Press.

39.

Lee

M.-K

. (2015). Peer feedback in second language writing: Investigating junior secondary students’ perspectives on inter-feedback and intra-feedback. System, 55, 1–10.

40.

Lee

Y.-W.

Gentile

Kantor

. (2008). Analytic scoring of TOEFL^® CBT essays: Scores from humans and e-rater^®. ETS Research Report Series, 2008, i–71.

41.

Liu

Sadler

R.W

. (2003). The effect and affect of peer review in electronic versus traditional modes on L2 writing. Journal of English for Academic Purposes, 2, 193–227.

42.

Longhurst

E.N

. (2020). Omoiyari: The Japanese art of compassion. HarperCollins.

43.

Lundstrom

Baker

. (2009). To give is better than to receive: The benefits of peer review to the reviewer’s own writing. Journal of Second Language Writing, 18, 30–43.

44.

Mangelsdorf

Schlumberger

. (1992). ESL student response stances in a peer-review task. Journal of Second Language Writing, 1, 235–254.

45.

Moore

P.J

. (2013). An emergent perspective on the use of the first language in the English-as-a-foreign-language classroom. Modern Language Journal, 97, 239–253.

46.

Nelson

G.L.

Carson

J.G

. (1998). ESL students’ perceptions of effectiveness in peer response groups. Journal of Second Language Writing, 7, 113–131.

47.

Patchan

M.M.

Schunn

C.D

. (2015). Understanding the benefits of providing peer feedback: How students respond to peers’ texts of varying quality. Instructional Science, 43, 591–614.

48.

Rahimi

. (2013). Is training students’ reviewers worth its while? A study of how training influences the quality of students’ feedback and writing. Language Teaching Research, 17, 67–89.

49.

Reddy

Harland

Wass

Wald

. (2021). Student peer review as a process of knowledge creation through dialogue. Higher Education Research & Development, 40, 825–837.

50.

Richards

. (2009). Trends in qualitative research in language teaching since 2000. Language Teaching, 42, 147–180.

51.

Robinson

. (2007). Criteria for classifying and sequencing pedagogic tasks. In García Mayo

M.P.

(Ed.), Investigating tasks in formal language learning (pp. 7–26). Multilingual Matters.

52.

Sato

. (2017). Interaction mindsets, interactional behaviors, and L2 development: An affective-social-cognitive model. Language Learning, 67, 249–283.

53.

Storch

. (2002). Patterns of interaction in ESL pair work. Language Learning, 52, 119–158.

54.

Storch

. (2004). Using activity theory to explain differences in patterns of dyadic interactions in an ESL class. Canadian Modern Language Review, 60, 457–480.

55.

Storch

Aldosari

. (2013). Pairing learners in pair work activity. Language Teaching Research, 17, 31–48.

56.

Topping

K.J

. (2009). Peer assessment. Theory into Practice, 48, 20–27.

57.

Tseng

S.-C.

Tsai

C.-C

. (2007). On-line peer assessment and the role of the peer feedback: A study of high school computer course. Computers and Education, 49, 1161–1174.

58.

Tsui

A.B.M

. (2000). Do secondary L2 writers benefit from peer comments? Journal of Second Language Writing, 9, 147–170.

59.

Van Beuningen

C.G.

De Jong

N.H.

Kuiken

. (2012). Evidence on the effectiveness of comprehensive error correction in second language writing. Language Learning, 62, 1–41.

60.

Villamil

O.S.

de Guerrero

M.C.M

. (1996). Peer revision in the L2 classroom: Social-cognitive activities, mediating strategies, and aspects of social behaviour. Jour nal of Second Language Writing, 5, 51–75.

61.

Villamil

O.S.

de Guerrero

M.C.M

. (2006). Sociocultural theory: A framework for understanding socio-cognitive dimensions of peer feedback. In Hyland

Hyland

(Eds.), Feedback in second language writing: Contexts and issues (pp. 23–41). Cambridge University Press.

62.

Vygotsky

L.S

. (1978). Mind in society: The development of higher psychological processes. Harvard University Press.

63.

Vygotsky

L.S

. (1986). Thought and language. Revised edition. MIT Press.

64.

Watanabe

Swain

. (2007). Effects of proficiency differences and patterns of pair interaction on second language learning: Collaborative dialogue between adult ESL learners. Language Teaching Research, 11, 121–142.

65.

Weigle

S.C

. (2002). Assessing writing. Cambridge University Press.

66.

Williams

. (2018). Language input effects on L2 composition peer review feedback. TESOL International Journal, 13, 33–46.

67.

. (2019). Lower English proficiency means poorer feedback performance? A mixed-methods study. Assessing Writing, 41, 14–24.

68.

Yang

Badger

. (2006). A comparative study of peer and teacher feedback in a Chinese EFL writing class. Journal of Second Language Writing, 15, 179–200.

69.

Yeh

C.-C

. (2018). L1 versus L2 use in peer review of L2 writing in English. Asian EFL Journal, 20, 124–147.

70.

. (2016). An exploratory study on the role of L1 use in peer written feedback of L2 writing. Porta Linguarum, 25, 135–146.

71.

. (2017a). Understanding university students’ peer feedback practices in EFL writing: Insights from a case study. Assessing Writing, 33, 25–35.

72.

. (2017b). Can higher-proficiency L2 learners benefit from working with lower-proficiency partners in peer feedback? Teaching in Higher Education, 22, 178–192.

73.

Lee

. (2014). An analysis of Chinese EFL students’ use of first and second language in peer feedback of L2 writing. System, 47, 28–38.

74.

Lee

. (2015). Understanding EFL students’ participation in group peer feedback of L2 writing: A case study from an activity theory perspective. Language Teaching Research, 19, 572–593.

75.

Lee

. (2016a). Exploring Chinese students’ strategy use in a cooperative peer feedback writing group. System, 58, 1–11.

76.

Lee

. (2016b). Understanding the role of learners with low English language proficiency in peer feedback of second language writing. TESOL Quarterly, 50, 483–494.

77.

Zhang

. (2022). Training student writers in conducting peer feedback in L2 writing: A meaning-making perspective. Applied Linguistics Review, 13, 439–460.

78.

Zhao

. (2010). Investigating learners’ use and understanding of peer and teacher feedback on writing: A comparative study in a Chinese English writing classroom. Assessing Writing, 15, 3–17.

79.

Zheng

. (2012). Understanding the learning process of peer feedback activity: An ethnographic study of exploratory practice. Language Teaching Research, 16, 109–126.

80.

Zhou

Liu

Jiang

. (2022). Examining the role of writing proficiency in students’ feedback literacy development. Applied Linguistics Review. Advance online publication. http://doi.org/10.1515/applirev-2021-0133

81.

Zhu

. (2001). Interaction and feedback in mixed peer response groups. Journal of Second Language Writing, 10, 251–276.

82.

Zhu

Mitchell

D.A

. (2012). Participation in peer response as activity: An examination of peer response stances from an activity theory perspective. TESOL Quarterly, 46, 362–386.