Generative AI-assisted feedback and EFL writing: a study on proficiency, revision frequency and writing quality

Abstract

This study investigated the impact of generative AI-assisted writing feedback, specifically using Grammarly, on English as a Foreign Language (EFL) learners’ writing proficiency, revision practices, and writing quality. Sixty postgraduate EFL students were randomly assigned to either an experimental group using Grammarly as an AI-enhanced feedback system or a control group receiving traditional instruction. Pre- and post-tests assessed overall writing proficiency, and semi-structured interviews were conducted with 25 students from the experimental group to gather qualitative data. Quantitative results indicated that the experimental group achieved significantly higher post-test scores than the control group. Additionally, positive correlations were found between the use of AI features, including generative AI, grammar, and vocabulary tools, and increased revision frequency, as well as improvements in content, organization, and cohesion. Thematic analysis of the qualitative data revealed that AI-assisted feedback promoted student engagement by reducing frustration, boosting confidence, and facilitating a greater sense of accomplishment. Participants emphasized the need to strategically integrate AI tools with traditional instruction, to foster critical thinking, and for AI tools to prioritize natural language fluency as well as to provide detailed feedback beyond basic grammar and mechanics. These results suggest that generative AI-assisted writing feedback, when used with pedagogical awareness, can effectively enhance EFL learners’ writing skills and support active participation in the writing process. The study also highlights that critical thinking skills are essential and that students should avoid overreliance on AI tools.