LEVERAGING CHATGPT IN EFL WRITING ASSESSMENT: A SYSTEMATIC LITERATURE REVIEW

Abstract

This systematic literature review investigates the role and effectiveness of ChatGPT in assessing English as a Foreign Language (EFL) learners’ writing performance. Guided by the PRISMA 2020 framework, 26 peer-reviewed empirical and conceptual studies published between 2022 and 2025 were analyzed. The findings reveal that ChatGPT has been leveraged across multiple assessment functions, including automated writing evaluation, corrective feedback, peer review simulation, rubric-based scoring, and prewriting support. ChatGPT demonstrates significant promise in formative assessment contexts, offering feedback comparable in quantity and breadth to human teachers, particularly on surface-level writing features such as grammar, organization, and vocabulary. However, its performance in summative assessment remains limited due to inconsistencies with human judgment and difficulties in evaluating higher-order writing skills, such as argumentation and style. Furthermore, while ChatGPT fosters learner engagement and confidence, concerns persist regarding equity, fairness, and the authenticity of feedback for diverse learner populations. The study underscores the need for rubric standardization, pedagogical integration, and longitudinal research to ensure ethical and effective implementation. Ultimately, ChatGPT is most impactful when used to complement, rather than replace, human expertise in writing assessment.