Xiaojuan Yin, Yanqing Lin

Xiaojuan Yin

25 October, 2024

ChatGPT, End commentary, Middle school students, English writing, Technology Acceptance Model (TAM).

In this study, through an English writing teaching experiment, questionnaires and interviews, the efficacy of corrective feedback of ChatGPT on 46 participants' English writing and participants' acceptance of the end commentary generated by ChatGPT were investigated among junior high school students in an ordinary middle school in Fuzhou, China. It was found that ChatGPT performed well in identifying ten categories of linguistic errors including grammar, spelling, word choice etc.. The accuracy of students' error corrections based on ChatGPT's feedback was as high as 98%, indicating participants have a high acceptance and effective use of the end commentary generated by ChatGPT. Based on the Technology Acceptance Model (TAM), results of questionnaires and interviews revealed that students rated ChatGPT highly in terms of ease of use and usefulness, with scores of 3.86 and 4.0; good attitude, with a score of 4.2; a strong willingness to use it, with a score of 4.1; and high evaluation of its actual use, with a score of 4.0. However, ChatGPT's feedback on sentence logic, writing content and paragraph structure still needs to be improved. The results of the study provide inspiration to English practitioners and researchers on how to effectively use ChatGPT to assist writing instruction, and highlight the importance of enhancing the awareness and education of teachers and students about Generative AI technology.

© 2024, the Authors. Published by ISCCAC

This is an open access article distributed under the CC BY-NC license