The Impacts of Multiple Feedbacks (ChatGPT, Peer and Teacher) on English Composition Revision

This study aims to explore the impacts of ChatGPT, peer, and teacher feedback on English writing revisions. Sixty-seven junior English majors participated in six rounds of writing and feedback activities throughout an academic year. The results reveal that in terms of the number of revisions, ChatGPT and teacher feedback accounted for 85.1%, while peer feedback accounted for 14.9%. For the types of revisions, ChatGPT dominated surface-level revisions, and teacher and peer feedback were mainly responsible for meaning-level revisions. In terms of the effectiveness of revisions, ChatGPT had the highest success rate (82%), and peer feedback had the lowest (56.1%). The study demonstrates that multiple feedback has a positive effect. Teachers should leverage the advantages of each feedback method, such as guiding students to use ChatGPT, enhancing the effectiveness of peer feedback, and strengthening the guiding role of teacher feedback. However, this study lacks a control group. Future research can conduct comparative studies between experimental and control groups to further explore.