Huahui Zhao for Bera2007
Who takes the floor: peer feedback or teacher feedback?
A comparison study of peer and teacher feedback for the development of Chinese English majors' EFL writing proficiency
Huahui Zhao University of Bristol
Paper presented at the British Educational Research Association Annual Conference, Institute of Education, University of London, 5-8 September 2007
Abstract
The increasing use of peer feedback in the ESL/EFL writing instruction has given rise to the research on peer feedback in the ESL/EFL writing contexts. The research literature suggests that peer feedback plays a crucial and irreplaceable role in the development of ESL/EFL students’ writing quality and proficiency.
The current study -part of a wider research project- explored the provision and effectiveness of peer feedback for the development of Chinese English majors’ writing quality, by taking teacher feedback as the comparison baseline. One writing tutor and 18 sophomore English majors in a large-scale university in China participated in this longitudinal study (i.e. February 2006- June 2006). The analysis of peer and teacher feedback along with their use in students’ revised drafts suggested similarities and differences between these two types of feedback. Specifically, the tutor provided more feedback instances than the peer assessors on the same assignments, although the focus and type of peer and teacher feedback were rather similar. The use percentages of feedback in students’ revisions indicated that teacher feedback exerted greater effects than peer feedback on students’ writing. This study concludes with the possible recommendation for the use of peer feedback in EFL writing classes in Chinese or other similar educational contexts.
Literature review
It has been a long tradition that the agents in developing and assessing student writing to be writing tutors rather than student writers themselves. In the last two decades, peer feedback has been increasingly used in ESL/EFL writing instructions, which gives rise to the research on the provision and effectiveness of peer feedback for the development of EFL/ESL learners’ writing quality and ability. In the research literature on peer feedback, the following three issues were addressed:
(1) students’ choice between peer and teacher feedback;
(2) the provision and effectiveness of peer feedback relative to teacher feedback;
(3) the implementation of peer feedback
These three lines of enquiry are elucidated below.
Ferris (2003) identified two benefits of the line of enquiry into students’ opinions about different types of feedback on their writing for language educators: (1) help us to be aware of what students may think and how they may react to pedagogical practices; and (2) assist us in a perceiving way in which our philosophies and practices and even our specific feedback techniques might be misunderstood by students (pg.93). She further pointed out that these two benefits could induce us to explain our decisions to the students and thus improve student motivation, help them understand instructors better and promote communication between teachers and students. Based upon these merits of asking students’ voice in peer feedback, a number of studies have been conducted in various contexts and the findings are consistent. Zhang (1995), for instance, asked this question to 81 Asian ESL college students who had full exposure to teacher-, peer- and self-feedback:
Giving a choice between teacher feedback and non-teacher feedback-that is feedback by peers or yourself-before you write your final version, which will you choose?
(Zhang 1995, pg. 215)
Students’ responses suggested that 76 (94%) of students preferred teacher feedback to peer- or self-feedback. Zhang, therefore, concluded that ESL learners unequivocally favoured teacher feedback over peer feedback. He, however, cautioned that his study was designed to examine the relative appeal of the three types of feedback and therefore “it should not be misinterpreted to mean that peer feedback is detrimental to ESL writing or resented among ESL learners. It may well be that all three types of feedback are beneficial, although with varying degrees of appeal (pg.219)”. His assertion was supported by the studies of Carson and Nelson (1996) and Nelson and Carson (1998) of 6 ESL college students (including 3 from China), in which all participants prioritised teacher feedback to peer feedback given that teachers had richer experiences in writing and writing feedback than peers and thus they could provide higher quality feedback.
The either/or choice between teacher and peer feedback in Zhang’s question was a concern for Jacobs and his colleagues (1998), who claimed that peer and teacher feedback could co-exist, and thus forcing students to make a choice between these two types of feedback might be misleading. They, then, asked 121 Chinese ESL college students (from Hong Kong and Taiwan), who had previously experienced teacher and peer feedback in their writing courses, to write in response to two options which focused exclusively on whether they liked or disliked peer feedback:
1. I prefer to have feedback from other students as one type of feedback on my writing.
2. I prefer not to have feedback from other students on writing.
(Jacobs et al. 1998, pg.311)
The results showed that 112 out of the 121 participants expressed their preference for having peer feedback as a form of feedback because of its benefits in spotting problems they themselves might have missed and the value of reading peers’ writing. In this way, Jacobs and his colleagues argued to take a middle path on the issue of feedback: to combine teacher-, peer- and self-directed feedback in a judicious way via a well-planned implementation process of integrating peer feedback into writing instructions, sharing teachers’ own writing experiences, providing peer review sheets, training students in how to work with peers, and using peer feedback in different writing phases.
In a response to Jacobs and his colleagues’1998 study, Zhang (1999) defended his either/or choice between teacher and non-teacher feedback, arguing that:
l The options used in the Jacobs et al (1998) questionnaire did not allow students to spell out their preferred type of feedback; therefore, their results could not be used to refute his 1995 finding;
l Studies which did not use the either/or choice, such as Nelson and Carson (1998), produced similar results to his 1995 finding. In this sense, their criticism of his question as misleading could not support their criticism of his finding;
l That students would like to have peer feedback as a type of writing feedback did not mean they preferred peer feedback over teacher feedback;
l Jacobs et al (1998)’s finding, that the majority of students expected to have teacher feedback to assist peer feedback, indicated that students preferred teacher feedback over peer feedback. This actually suggested the convergence of the findings in these two studies.
With reference to these arguments, Zhang (1999) once again asserted that peer feedback was less popular than teacher feedback among ESL student writers. A similar claim was made by Yang et al’s (2006) study of 12 Chinese EFL college students, who fully confirmed the value of teacher feedback for the improvement of their final writings whereas held reservations about the value of peer feedback. The most salient reason reported by learners for their preference for teacher feedback was that teachers were more experienced and knowledgeable and thus more trustworthy in providing valid written feedback than students who might provide “incorrect” (Yang et al. 2006, pg.189) feedback. In this sense, we can observe that the empirical studies commonly suggested that ESL/EFL learners prioritised teacher feedback to peer feedback for their writing.
In terms of the provision and effectiveness of peer feedback were explored, it seems to be a convention of using teacher feedback as the baseline for comparison. Yang et al’s (2006) study of twelve Chinese college EFL learners in a Chinese university reported that the teacher provided more written feedback instances (N=235) than the peer reviewers (N=225) on the same students’ writing assignments. Caulk (1994) observed that the peer assessors focused more on content like ordinary readers whereas she, as the writing tutor, paid more attention to forms and clarity given their importance for academic writing. In accord with this, Yang et al (2006) also noted that peer feedback was more meaning-oriented, whereas teacher feedback was balanced on both surface (e.g. grammar, vocabulary and sentence structure) and meaning-oriented (e.g. content) changes. In view of the type of feedback, Caulk (1994) observed that teacher feedback was more generalised without supplying revision strategies (i.e. indirect feedback) whereas peer feedback was more specified via providing specific revisions strategies (i.e. direct feedback).
As for the effects of peer feedback in students’ revisions, Paulus (1999) suggested that teacher feedback exerted more effects on students’ revisions than peer feedback. He found that the 12 undergraduate international students in his pre-freshmen composition class were far more likely to incorporate teacher feedback (87%) than peer feedback (51%) in their revisions. As with Paulus’ participants, the 12 Chinese college EFL learners in Yang et al’s 2006 study used a larger amount of teacher feedback (90%) than peer feedback (76%) in their revisions. We may see, accordingly, that teacher feedback exerted more effects on learners’ revisions than peer feedback by college ESL/EFL learners. This, on one hand, explains learners’ preference for teacher feedback and, on the other hand, suggests the possible impact of learners’ affective disadvantage over peer feedback on their use of it.
This finding makes it particularly interesting to consider the role of peer feedback relative to teacher feedback. Hu’s (2005) study, based upon his observation of three writing groups of Chinese ESL learners in Singapore, found that students made many valid suggestions which he, as the writing tutor, had not provided. This suggested that peer feedback could play a complementary role to teacher feedback. In line with this, Tsui and Ng (2000), through asking secondary ESL learners about the role played by peer and teacher feedback, reported the following value of peer feedback which teacher feedback could not engender:
l Enhance a sense of audience;
l Raise awareness of own problems through reading peers’ writings;
l Encourage collaborative learning where students can clarify and convey their meaning;
l Foster the ownership of text so that they could rely less on teachers and be more confident in themselves with their writing.
They, accordingly, stressed the value of peer feedback and suggested integrating peer feedback into writing instructions.
In view of learners’ affective disadvantage to peer feedback and the smaller use percentages of peer feedback than teacher feedback in students’ revisions, researchers made suggestions on how to develop a more efficient peer feedback, for which, training appeared to be the most frequently recommended approach (e.g. Stanley 1992; Berg 1999; Patri 2002; Tuzi 2004). For example, Rollinson (2005) justified training as engineering the efficiency of peer feedback in terms of the focus, the appropriateness and the collaboration dynamics, saying this:
Without such training, it is more likely that student response will be inappropriate: it may be destructive and tactless (or, conversely, overgenerous and uncritical); it may also tend towards dealing with surface matters rather than issues of meaning and content, or it may be prescriptive and authoritarian rather than collaborative and supportive. (Rollinson 2005, pg.26)
This was ascertained by McGroarty and Zhu (1997) through investigating the effects of trained peer response through a comparison of 169 first language student writers who were divided into two groups, one trained in how to participate in peer response to writing and the other not trained. They obtained a finding similar to Zhu’s (1995) with the same group of students by adopting the same experimental approach: the trained group made significantly more and better comments and held more positive attitudes towards peer feedback than the untrained group. Similarly, in Berg’s (1999) study of 46 ESL student writers, she noted that coached peer responses made more meaning-oriented changes than the untrained ones. In line with this, Hu (2005), in his three year action research on using peer review with Chinese ESL college student writers, identified the following value of training:
These students were much more positive about and developed much better attitudes towards peer review. Their peer comments were far superior both in quantity and quality to those made by the previous students [who were not trained]. All students also made serious efforts to incorporate at least some peer suggestions in each of their revised drafts, and the second drafts improved ostensibly as a result of the revisions. (Hu 2005, pg. 334-335)
This showed that training might lead to students’ more positive attitudes towards, and active performance in, peer feedback. This value was corroborated the findings in Min (2005) with 18 EFL sophomore students in her composition class in Taiwan. The data before and after training revealed that training resulted in significantly more comments on global issues (e.g. idea development and organisation), facilitated writers to elaborate their ideas in more specific and focused way, and helped less-advanced reviewers gain confidence in regarding themselves as competent readers.
Based on the empirical studies on peer feedback above, we can summarise that:
(1) Teachers generated more feedback instances than the peer assessors.
(2) Peer feedback could play a complementary role to teacher feedback.
(3) The focus and type of peer feedback differed from those of teacher feedback.
(4) More teacher feedback was used than peer feedback in students’ revisions.
Research focus and context of this study
This study —part of a wider project— explored the use of peer feedback on writing with Chinese English majors in a large-scale high-rank Chinese university, aiming to examine the following three hypotheses which were developed with reference to the findings on peer feedback above:
Hypothesis 1: The writing tutor provided more feedback instances on students’ writing than the peer assessors; however, peer feedback played a complementary role to teacher feedback in students’ revisions;
Hypothesis 2: The focus and type of teacher feedback and peer feedback were different.
Hypothesis 3: Students used more teacher feedback than peer feedback in their revisions.
Eighteen sophomore English majors along with their writing tutor participated in this study for over four months (i.e. from February till June 2006). These students obtained high English examination marks in the university matriculation examination (on average 120/150) and labelled as intermediate English learners by their writing tutor, who came from United States and had taught English as a Foreign Language for about two years in China. Before this, he was a professor in an American college for over 30 years and could, thus, be considered to be an experienced instructor. The data was collected from writing classes held every Wednesday evening from 6:30pm to 8pm. The classroom was arranged with eight multiple-square desks, facilitating group work. Students were paired by assigning a number from 1-9 and the two students with the same number were in one pair. Eight sequential steps were followed to conduct peer feedback, encompassing: