Exploring the effectiveness of a DDL-GenAI teaching mode in students’ EFL writing classroom

This study developed and evaluated an innovative teaching mode integrating Data-Driven Learning (DDL) with Generative Artificial Intelligence (GenAI) to enhance EFL student expository writing. Employing a quasi-experimental design, 75 second-year Chinese university students were randomly assigned to one of three instructional modes: an integrated DDL-GenAI class, a DDL class, and a GenAI class, over a three-week intervention. Data were triangulated through students’ writing, questionnaires, reflective journals, and semi-structured interviews. Results indicate that the DDL-GenAI mode significantly improved students’ writing performance, with notable improvement in their knowledge of structure, content quality, efficiency, and classroom engagement. Participants reported positive perceptions of the DDL-GenAI mode. They noted that this mode enriched their learning experience, improved their writing skills and language use, and helped them to balance technological application with critical thinking. However, students expressed less satisfaction with tool speed and awareness of advanced functions; and challenges such as the complexity of tool integration and initial technological frustration were observed. The findings demonstrate both the practical feasibility and the effectiveness of the DDL-GenAI teaching mode in EFL writing instruction, and also provide insights for promoting and improving this DDL-GenAI integration in EFL education.