Experimental evidence for the efficacy of generative AI in improving students’ writing skills

Abstract

Purpose Given continued debates on the potentials of newly emerging artificial intelligence (AI) like generative AI (GenAI), this study aims to contribute to corporal studies by investigating the efficacy of GenAI in improving students writing skills. Design/methodology/approach A mixed-methods research design with an experimental approach was used to elicit information from 70 undergraduate students studying at a private university. A writing course was designed and used to elicit information from the participants on the efficacy of using ChatGPT in their writing instruction. Findings After collecting data through experiments and interviews, the result indicates that although ChatGPT may assist students in providing ideas in writing lessons, it may not improve their overall writing skills. Originality/value This study provides empirical evidence limited to the scholarly literature on the role of ChatGPT in improving students’ writing skills. This study adds to scholarly discussions on the potential of ChatGPT which has recently sparked debates in academia.