Teacher-Vetted AI Feedback in Primary EFL Writing: A Classroom Cross-Over Study
Main Article Content
Abstract
The increased integration of artificial intelligence (AI) technologies in the English as a Foreign Language (EFL) context has led to significant changes in the learning and teaching paradigm. Automated writing evaluation (AWE) technologies support L2 writers by giving them instantaneous corrective feedback at the grammar, vocabulary, and sentence level. However, questions have been raised about the contextual accuracy of the automated writing feedback and its pedagogical suitability for younger learners who require scaffolding and guidance from their teachers. The current study assesses the impact of vetted teacher AI feedback compared to AI only (no human input) and teacher only feedback on primary EFL learners’ writing performance.
The 6-week classroom cross-over study involved 60 primary aged learners, between the ages of 10 to 12, who completed 3 feedback conditions (AI only, teacher only, and teacher vetted AI feedback) for 2 weeks each. Quantitative and qualitative data were gathered from writing pre- and post-tests, rubric scoring of the essays, and a learner reflection journal. Quantitative results from an exploratory repeated measures design revealed that the human-in-the-loop condition (teacher vetted AI feedback) made a statistically the greatest improvement in accuracy, lexical diversity, and task completion. Learners also showed a higher level of motivation, self-confidence, and feedback acceptance when the teacher vetted, added a personal touch, and explained the AI-generated writing suggestions. Qualitative data demonstrated that human involvement enhanced learners’ trust in automated writing feedback, thus making it more pedagogically meaningful.
The study results support the advantages of a human–machine partnership, in which teacher-vetted AI feedback is more time-efficient and instructionally effective without jeopardizing the pedagogical values manual feedback provides. The study findings also revealed the potential of AI technologies’ inclusion in primary EFL writing classrooms as a complementary pedagogical tool rather than replacing the teachers. It can help develop feedback literacy, learner agency, and sustainable digital disposition in young writers