The Effects Of Peer Feedback On Composition Revision Of Non-English Majors Of Different Writing Proficiency Levels

In China, English learning is an essential part of the college curriculum. Among the four basic skills, listening, speaking, reading and writing, writing and its development has been a big headache and kept troubling college teachers and students. To improve the situation, practitioners and researchers have been trying very hard to find effective approaches. One of the approaches is the process approach. In the process approach, feedback and revision are two central elements.There are various kinds of feedback, among which teacher feedback and peer feedback have aroused the most attention. However, feedback on writing, especially peer feedback, has not been well-researched or practiced in China. Although there have been a number of studies aboard, no unified conclusions have been drawn.Considering this phenomenon, the present study attempted to probe into the effects of peer feedback on composition revision of non-English majors of different writing proficiency levels. The research questions are as follows:1) How effective would peer feedback activities be for non-English majors in revising their compositions? And in which writing dimensions (organization, content, grammar, vocabulary and mechanics)?2) Would peer feedback activities be effective for all levels of writing proficiency in all the writing dimensions investigated?To answer the research questions, the author conducted an 11-week empirical study on one class of 40 subjects. In the first week, pretest, pretraining the subjects and grouping the subjects according to their writing proficiency levels were carried out. Then in the following six weeks, the subjects were to finish three compositions and peer feedback would be carried out. In the eighth week, posttest, questionnaire and interviews would be conducted.The findings are yielded as follows:1) Peer feedback activities were effective for the non-English majors in revising their compositions, because 74% of peer feedback suggestions received were accepted and incorporated into their second drafts and the average score of all the subjects' writing performances in the posttest increased by 1.5 compared with that of pretest. Moreover, through the analysis of the questionnaire and interviews, it could be concluded that students held giving and reading peer feedback were helpful for them to revise their compositions. But comparatively speaking, giving peer feedback was more effective.As for the writing dimensions researched, there was improvement in all the five writing dimensions, but mainly in the dimensions of grammar, vocabulary and mechanics.96.7% of peer feedback suggestions received and 97.5% of peer feedback suggestions incorporated focused on the surface changes. Compared with the pretest, there was improvement in all the writing dimensions in the posttest, especially in the dimensions of mechanics, vocabulary and grammar. Moreover, as the questionnaire and interviews revealed, students also claimed that the peer feedback suggestions received were mainly restricted to the surface problems and reading peer feedback was mostly helpful for them to revise writing on the surface level.(2) Peer feedback activities were effective for all levels of writing proficiency in revising their compositions. But still slight differences existed among different writing proficiency levels. While revising, the upper-intermediate students received the least amount of feedback while they incorporated the most. At the same time, the comparing of the scores of pretest and posttest showed that the lower-intermediate students improved the most. As for giving and reading peer feedback, all of the students held that the former was more effective for them to revise.As for the writing dimensions, the students of different levels all received and incorporated the most feedback on grammar, mechanics and vocabulary in the descending order. The scores of the two tests demonstrated that the students of middle-intermediate writing proficiency level and upper-intermediate writing proficiency level have made the most improvement in the writing dimension of mechanics. And the lower-intermediate students improved the most in the dimension of vocabulary, then in organization. Through interviews, the upper-intermediate ones held that giving feedback enabled them to enrich the content of their compositions while the ones of lower-intermediate level claimed they got to know what to write through reading. What's more, in the interviews, students of all levels pointed out the benefits gained from reading peer feedback were mainly restricted to surface problems, for due to their limited ability, peer feedback received chiefly focused on the writing dimensions of surface level.The above findings have several implications for EFL college instruction and learning in China. First, peer feedback activities should be implemented in classes and desirably should be complemented together with teacher feedback, which should focus more on the writing dimensions of content and organization. Second, compositions should be exchanged among more groups, enabling students to read more of their peers'compositions.