Generative AI and Essay Writing: Impacts of Automated Feedback on Revision Performance and Engagement

This study investigates the impact of feedback generated by large language models (LLMs) on improving the essay-writing skills of first-year university students in Hong Kong. Specifically, it examines how generative AI supports students in revising their essays, enhances engagement with writing tasks, and influences their emotional responses during the revision process. The study followed a randomized controlled trial design, with one group of students receiving AI-generated feedback on their essay drafts while a control group did not. A mixed-methods approach was used to evaluate the feedback's effectiveness, combining statistical analysis of essay grades with student surveys and interviews. Quantitative results demonstrated that students who received AI feedback achieved significant improvements in essay quality, while qualitative findings revealed higher levels of engagement, increased motivation, and mixed emotional responses to the feedback process. These findings highlight the potential of generative AI as a tool for enhancing essay revision performance and fostering student engagement in higher education. However, further research is needed to explore its long-term impacts and applicability across diverse educational contexts.