Adaptive and personalized writing assessment through Generative AI: Enhancing EFL learners’ engagement and competency development

Abstract

The process of writing evaluations and instructions has evolved with the integration of artificial intelligence in education. This research aims to test whether adaptive and personalized assessments through Generative AI can enhance EFL learners’ writing engagement, cognitive and competency development, and explore their perceptions of the use of Generative AI to improve their writing quality. The study employed a mixed-methods research design, with 80 students divided equally into experimental and control groups in an Indonesian educational institution over four weeks. The quantitative findings reveal that generative AI–based adaptive assessment significantly improves EFL learners’ writing outcomes across engagement, cognitive, and competency domains. The qualitative finding demonstrated that EFL learners have positive perceptions of adaptive and personalized assessments, especially those that use generative AI. The use of such assessments is beneficial for student development and engagement, self-regulation, confidence building, and constructive feedback. Using personal feedback and task design continuously in writing instruction across various domains will enable learners to be more actively involved, think more critically, and develop stronger writing skills. The application of adaptive and personalized assessments using Generative AI can be used to optimally support EFL learners’ writing growth and development, making it a transformative teaching tool.