An Analysis of Oral Peer Feedback Comments on EFL Student Writing

The purpose of this research is to explore what behaviors Korean EFL students exhibit during peer feedback in a university writing class. Fifteen students voluntarily participated in the study and the peer feedback recordings of eighteen pairs were analyzed in terms of the focus of feedback and the types of negotiation generated during oral peer feedback. The students commented on content (159), language issues (154), and organization (103) as well as writing styles (13) and others (15). They focused not only on the global aspects of writing but also on local aspects. The number of comments regarding global aspects was similar across pairs whereas the number of comments on grammar varied. In addition, a total of 18 types of negotiation were found across four pairs of students and the top four frequent negotiation types consisted of phatic comments (50), opinions (49), explanations of problems (42), and suggestions (40). The reviewers generated about 70 percent of the total comments and the nature of oral peer feedback was mostly oneway communication. The age and gender of the pairs also seemed to affect their interaction patterns. Some implications were suggested based on the findings

http://dx.doi.org/10.17936/pkelt.2018.30.4.7 December 31, 2018

An Analysis of Oral Peer Feedback Comments on EFL

Student Writing

Jiwon Paek

(Daegu University)

Paek, Jiwon (2018) An analysis of oral peer feedback comments on EFL student writing

English Language Teaching, 30(4), 121-138

The purpose of this research is to explore what behaviors Korean EFL students exhibit during peer feedback in a university writing class Fifteen students voluntarily participated in the study and the peer feedback recordings of eighteen pairs were analyzed in terms of the focus of feedback and the types of negotiation generated during oral peer feedback The students commented on content (159), language issues (154), and organization (103) as well as writing styles (13) and others (15) They focused not only on the global aspects of writing but also on local aspects The number of comments regarding global aspects was similar across pairs whereas the number of comments on grammar varied In addition, a total of 18 types of negotiation were found across four pairs of students and the top four frequent negotiation types consisted of phatic comments (50), opinions (49), explanations of problems (42), and suggestions (40) The reviewers generated about 70 percent of the total comments and the nature of oral peer feedback was mostly one-way communication The age and gender of the pairs also seemed to affect their interaction patterns Some implications were suggested based on the findings (191 words)

Key words: oral peer feedback, focus of peer feedback, negotiation types, EFL writing, academic

English

I INTRODUCTION

Feedback in writing classrooms is crucial for learners to revise their drafts and ultimately improve their writing ability in both L1 and L2 contexts Teachers have traditionally been considered the prominent feedback providers, but peer feedback has increasingly been adopted in ESL and EFL writing classrooms A substantial amount of research on peer feedback has been done (Berg, 1999; Mendonça & Johnson, 1994; Tusi & Ng, 2000; Villamil

& De Guerrero, 1996; Zhang, 1995)

There are a number of reasons to use peer feedback in L2 writing classrooms Peer

 This research was supported by Daegu University Research Grant, 2016

reviewers have been shown to provide useful feedback and research shows that student writers produced more revisions in response to their peer comments (Kim, 2009; Cho, 2005) than simply revising by themselves Peer feedback also fosters communicative behavior through peer collaboration and scaffolding (Villamil & De Guerrero, 1996) Student responses were more specific than teacher feedback (Caulk, 1994) and peer reviews also enhance students’ ability of “re-viewing their writing with the eyes of another” (Zamel, 1982,

p 206)

Studies on L2 writing have focused on the effects of peer feedback (Berg, 1999; Cho, 2005; Kim, 2009; Min, 2006) and perceptions about peer feedback (Hislop & Stracke, 2017; Yi, 2010) Peer reviewers were found to provide useful feedback for text revision to peer writers and L2 learners in general perceived peer feedback as a valid source though they had different degrees of preferences regarding peer feedback and teacher feedback A couple of studies have investigated student interaction during peer feedback (Choe & Yu, 2012; Mendonça & Johnson, 1994; Villamil & De Guerrero, 1996; Lockhart & Ng, 1995) However, the number

of studies analyzing peer interaction and peer talk during peer feedback is not sufficient enough to understand what shapes peer feedback, what aspects peer reviewers focus on, and how learners interact with their peers Moreover, the actual peer interaction in EFL Korean classrooms has not been fully explored and as a result, there is a gap in peer interaction literature

This study sought to explore the nature of Korean EFL learners’ interaction during peer feedback using a qualitative approach to analyze the discourse of peer feedback It will provide explanations about the focus of Korean learners’ peer feedback and characteristics

of their interaction because Korean EFL learners could be different from learners in ESL context in terms of their culture and learning experience This study will offer guidelines for teachers to implement peer feedback in Korean writing classrooms and what precautions they need to take for peer feedback to be truly beneficial to Korean learners Under the purpose, the study will address the following research questions:

1) What aspects do students focus on during oral peer feedback?

2) What characteristics does peer feedback interaction show?

II PREVIOUS STUDIES

Several studies reported the positive effects of peer feedback in EFL writing classrooms Tusi and Ng (2000) summarized the benefits First, peer feedback comments are more informative than teacher feedback Peer feedback raises awareness of strengths and weaknesses in writing and critical thinking when they read each other’s draft From affective

perspectives, the help of supportive peers can lower learners’ apprehension and enhance their attitudes toward writing Learners are encouraged to take more responsibility for their writing and revision

The findings of Min (2006) also support peer feedback benefits overall improvements of revised drafts when peer review training takes place through in-class modeling and teacher-reviewer conference After training, the number of comments produced by teacher-reviewers and the number of comments incorporated into revision were significantly higher and the quality of revised drafts was enhanced

Cho (2005) examined types of feedback that students provided to their peers The participants were more concerned with local aspects than global aspects and they did pay attention to both general feedback and specific feedback unlike the worries of some writing teachers that general feedback is not helpful for learners to revise their drafts Regarding an improvement of revised drafts, a significant improvement was only found in language but not in content and organization

There is a substantial body of research on college students However, little research has focused on high school students Kurihara (2016) explored how peer reviews affected writing abilities of Japanese high school students with achievement below the national average The findings indicated that peer feedback contributed to an improvement in organization and cohesion Peer feedback led students to question the validity of feedback before they decided whether to incorporate it or not into their revision The process might have had the students internalize writing knowledge in such areas

Lee (2016) explored how college students perceived peer review experience and what factors affected the review process The students completed four peer review sessions in pairs for one semester and 26 participants submitted reflective journals on their experiences Most participants commented peer review activities as being beneficial, valuable, and meaningful and their anxiety levels had more decreased at the end of the semester The study suggests the number of exposures to peer review and the creation of trusting class atmosphere as well

as clear guidance are crucial to success of peer review activities

Cho (2011) also explored how learners perceived peer feedback, what kinds of feedback learners gave to their peers and how giving peer feedback influenced their revision She found giving peer feedback not only helped learners develop critical views in writing but also apply them to their own writing Learners’ perceptions on peer feedback affected not only the types

of feedback but also the frequencies of feedback When the learner perceived peer feedback

of no use, the types and the frequency of feedback were highly limited compared with the learners who perceived it beneficial The most frequent types of feedback were explanations about content and about opinions regardless of their perceptions

Based on the previous studies, this study aimed to explore how Korean EFL learners interact and what they focus on during oral peer feedback The nature of their peer feedback

will offer how writing teachers should implement peer feedback in Korean writing classrooms and what they should consider in order for peer feedback to be beneficial to Korean EFL writers

III RESEARCH METHOD

1 Research Context

This research was conducted in one of two writing classes opened for English education majors in a natural setting without manipulation The class consisted of twenty-five students, who were mostly freshmen and some sophomores None of them had taken an English writing class nor experienced peer feedback in a secondary or a tertiary school The instructor was a female American and had worked as an assistant professor in the department for three years, but she had been teaching English for six years Thus, the class was taught in English and the students had to use English in class, except during peer feedback sessions They were allowed to use Korean only during peer feedback sessions in order to eliminate any chances that their limited English proficiency might produce misunderstandings and deteriorate the quality of peer feedback

The course adopted a process writing approach and the students completed three writing assignments with multiple paragraphs in the comparative, cause and effect, and argumentative genres They were required to submit three drafts in each genre and received peer feedback from two peers on their first drafts and teacher feedback in individual conferences on their second drafts The professor provided worksheets which contained slightly different guidelines for different writing genres She explained what to do and what not to do in general and then provided modeling of peer feedback by going through the guidelines with examples before the first peer feedback The professor dealt with both language issues and content and organization though most students reported that teacher feedback had focused mostly on language issues during interviews On the other hand, peer feedback guidelines were centered on the aspects of content and organization Nonetheless, the students were encouraged to provide as much corrective feedback to their peers as possible

Each peer feedback session was mostly done in pairs but in some cases, in small groups The students were given 10 minutes for Writing 1, a comparative essay and 20 minutes for Writing 2 and 3, a cause and effect and an argumentative essay, respectively For the most part, the students partnered with classmates sitting near them For the given time, they exchanged their drafts with their partners and read their partners’ drafts in class After finishing their reviews, they filled out the feedback worksheets for their partners and started

oral feedback Therefore, students could spend more time on oral feedback when both partners reviewed fast When the first session finished, they formed new pairs and went through the same procedures with new partners in the second session Because draft reviews and oral feedback should be done in class, less proficient students had more difficulties compared to their peers with higher proficiencies

2 Participants and Data

Fifteen students voluntarily participated in the study They recorded their feedback sessions and sent the files to me and completed three interviews The first interviews were intended to understand their general perceptions on writing feedback and their previous experience of feedback In contrast, the second and third interviews were focused on their choices of peers, experience, and the content of feedback on Writing 1 and Writing 2 When the participants worked with non-participants, the sessions were recorded only when their partners agreed to be recorded The non-participants were informed that their feedback comments would be used for the study if they agreed to be recorded The participants’ proficiency could be identified as low-intermediate, ranging from Level 2 to Level 4 on the English portion of the Korean university entrance exam

The data collected for this study includes recordings of peer feedback sessions, worksheets1, writing drafts, recordings of individual interviews and fieldnotes of class observations However, only the recordings were analyzed while the other data was used to understand the characteristics of peer feedback and participants’ behaviors and intentions All feedback sessions were recorded, but the recordings from Writing 2 were used for this study because some participants were not sure about what they were supposed to do in the feedback activity on Writing 1 and some of them were very conscious of being recorded The oral feedback comments on Writing 2 were analyzed and the worksheets and interviews were used for the purpose of cross-checking the analysis

The data analysis was conducted using a data-driven approach in order to explore what Korean EFL students focus on and what types of negotiation they engage in during oral peer feedback Therefore, two separate analyses were conducted The recordings of eighteen pairs were transcribed verbatim and analyzed

The purpose of the first analysis was to identify which topic areas the participants focused

1 The worksheet the students used for peer feedback on a cause and effect essay is attached in Appendix

2 Because there was no second analyst in this study, the research personally communicated with an anonymous reviewer

in English education field to minimize possible problems of one analyst

on during peer feedback I started by analyzing the comments in terms of which categories they appeared in on the analytical scoring rubrics (e.g content, organization, and language issues) and grouped the other comments that did not fit into three categories First of all, I read all the comments and segmented them whenever the students shifted topics Some comments were excluded because they were neither related to the texts nor to text revision For example, the excluded comments include comments about feedback procedures, phatic comments for saving face, or comments for emotional support such as encouragement and empathy Then, the similar segments were grouped together The regrouping process continued until I reached the final categories Five areas were discussed by the students: content, organization, language issues, academic writing styles, and overall comments Table

1 illustrates the topic areas and their explanations

TABLE 1 Topic Areas of Peer Feedback Comments and Explanations

Content All comments related to writing topics, meanings of sentences/words, text coherence and cohesion, and content of supporting det

ails and causes and effects

Organization

Comments on

- the existence of a hook, a thesis statement, topic and concluding sentences, a take-away statement, and a restatement of the thesis statement;

- a logical sequence of ideas;

- body paragraph organization such as existence of a main idea and supporting details

- the length balance among paragraphs;

- transitions to indicate a cause and effect genre

Language

issues

Grammar Comments on sentence structures, tense, agreements, coordinating conjunctions, and word forms Lexis Comments on word choice which are not related to word forms Mechanics Comments on spelling, capitalization, and punctuation

Writing styles Comments on stylistic conventions of academic writings

Others Comments on impressions and overall quality of writing

The second analysis was done to explore what characteristics the peer feedback interaction showed For this purpose, the transcripts of four randomly chosen pairs were analyzed for the types of negotiation that occurred during oral feedback The, the transcript of every fifth pair was analyzed: transcripts of Pair 1, Pair 5, Pair 10, and Pair 15 Though the analysis was data-driven without tightly prefigured categories, it was based on categories of negotiation (Mendonça & Johnson, 1994) and strategies of providing scaffolding to peers during peer reviews (Villamil &

De Guerrero, 1996) None of the coding categories perfectly matched the Korean EFL students’ comments that were generated during peer feedback Therefore, I attempted to generate new descriptive categories that encompassed all types of the negotiations found in these peer reviews After the initial analysis, I analyzed the data according to who initiated each type of negotiation; negotiations were considered to be either writer-generated or reviewer-generated The negotiation types and their definitions are shown in Table 2

TABLE 2 Negotiation Types and Definitions

Questioning

Request for clarification Asking the interlocutor to clarify intended meaning Request for

opinions Asking the interlocutor to provide opinions about suggested alternatives Request for

suggestions Asking the interlocutor to suggest alternatives Comprehension

checks Checking whether the interlocutor understands the initiator’s intention or meaning of sentences or words Confirmation

checks Asking whether the initiator correctly understands the interlocutor’s intention or the text

Explanations

Clarification Offering clarification of meaning or intention Justification Defending choices or decisions made about the text Opinions Offering general opinions about the text, either positive or negative Suggestions Offering alternatives

Explanation of problems Explaining problems in certain parts of the text Explanation of

content Explaining the content of certain parts of the text or the whole text Restatements Restating the previous comments of the initiator or the interlocutor Translation Translating sentences word by word into Korean

Instruction Explaining grammar or conventions

Corrections Correcting language errors

Acceptance Accepting the interlocutor’s explanations or opinions

Phatic comments Commenting for social or emotive purposes and for saving face of the initiator or the interlocutor Comments for task management Commenting on starting, ending, or turn-taking

IV FINDINGS AND DISCUSSION

In three individual interviews per participant, the participants in general evaluated peer feedback as a valuable tool through which they learned how they should review their own drafts and realized what strengths and weaknesses their own drafts had The students’ perceptions of peer feedback changed by the end of the semester and they reported that peer feedback activities were fun and interesting, and they felt that the feedback was more effective in helping to improve their writing than they had initially expected, regardless of the actual effectiveness However, their degree of preferences of peer feedback over teacher feedback and perceived benefits varied

1 Aspects that the Students Focused on

The number of feedback comments was counted to evaluate which aspects the students focused on during oral peer feedback Each pair generated 16 to 42 comments, with an average of 23.83, and with an average time of 13 minutes and 67 seconds per pair Table 3 illustrates how the peer feedback comments were distributed across the comment categories

TABLE 3 Discussed Aspects of Peer Feedback

Pair Cont Org Grammar Language issuesLexis Mech Style Overall Total

17 8 10 13 9 0 0 2 42

* Cont = content, Org = organization, Mech = mechanics

The most frequently addressed aspect during peer feedback was content (159 occurrences), and the second and the third aspects were language issues and organization, with totals of

154 occurrences and 103 occurrences, respectively Though the sum of the comments related

to language issues outnumbered the comments on organization, the number of individual areas of language issues was less than that of the comments on organization About 69 percent

of the total comments centered on content and organization issues which highly influence the overall quality of writing According to Elbow (1981, cited in Cho, 2005), global feedback indicates “the developments of ideas, audience, purpose, and organization writing” whereas local feedback concerns “word usage, grammar, and punctuation” (p 40) Thus, it is certain that the students were more concerned about the global aspects than local aspects in their writing

This tendency is interpreted as heavily influenced by the guidelines of the peer feedback worksheet the students referred to The participants confirmed in interviews that they had reviewed their partner’s draft while they checked items (1) to (12) on the worksheet The transcripts also supported that the guidelines absolutely affected the behaviors of the students because the order of feedback comments reflected the order of the guidelines on the worksheet in most cases

Excerpt from Pair 14

- Reviewer: The hook, the example, this is really good, very fresh (content), and these connecting sentences are good, connecting the hook and the thesis statement (organization), but I think you should mention three [causes] here [in the thesis statement] (organization) Then, the result of this research in

this body paragraph, right?

- Writer: The research about smart phones?

- Reviewer: I like this The use of research result, a reliable source is good, and [but] this explanation

about smart phones is [overlap] (content)

- Writer: [overlap] Long

As the above excerpt indicates, the review started from the engaging hook, connecting sentences, and the main topic and controlling idea in the thesis statement and moved to the body paragraphs, which mirror items (1) to (5) on the peer feedback worksheet

Excerpt from Pair 14

- Reviewer: First of all, I personally like the topic, it was good to understand and emphasize

(content) In the first paragraph [the introduction paragraph], the engaging hook, connecting sentences, and thesis statement, all look good (organization)

The second excerpt is from the same pair but the reviewer is the writer in the first excerpt The second excerpt occurred before the first excerpt in time line In the second case, the reviewer started from the topic of the writing but commented on numbers (1) to (3) at once without specific information When I asked about the reason for this in the third interview, the student mentioned that she thought that she was supposed to check ‘yes’ or ‘no’ about the existence of the engaging hook, connecting sentences, and thesis statement The second case more commonly occurred than the first case during peer feedback

As a result, the behaviors of the participants were affected how they interpreted what they were supposed to do during peer feedback The guidelines alone did not guarantee a high quality of feedback though the professor spent time explaining what the students were expected to do and provided modeling in class These students may have needed more training sessions or some experience of receiving feedback from the professor before they engaged in peer feedback activities

Other than global aspects, the students also generated 154 comments on language issues,

35 percent of the total comments Though there was no specific guideline about language issues on the worksheet, the professor encouraged her students to correct errors by themselves

or with partners Therefore, some students paid attention to language errors and addressed them in peer feedback Most comments on grammar were on tense, agreement, and word forms while the comments on lexis were on collocations and word choices

Pair 16 and Pair 17 generated 20 and 22 comments on language issues, respectively Pair

16 predominately dealt with grammar (17 comments), while Pair 17 commented on both grammar and lexis (13 comments and 9 comments, respectively) Bomi3, the reviewer of both Pair 16 and Pair 17, explicitly mentioned that she focused on language issues when she started her reviews The comments about global aspects from these pairs are similar in number to those of other pairs Pair 16 and Pair 17 spent 18 minutes and 36 seconds and 22 minutes and

22 seconds, respectively, on peer feedback, which was more than the average time of 13 minutes and 67 seconds Bomi was a student with a higher proficiency compared to other students and had confidence in her English abilities In contrast, Pair 13, Pair 14, and Pair 15 generated 3, 1, and 1 comments on language, and the reasons for their lack of comments were because they did not notice language problems in their partner’s drafts according to the interviews The most prominent factor influencing the error correction was the students’ language proficiencies

3 All the names appearing here are pseudonyms