Exploring Iranian university students’ behavioral intention to use ChatGPT for academic purposes

Abstract

Generative artificial intelligence (AI) tools such as ChatGPT are increasingly shaping higher education by providing new forms of interaction, immediate access to information, and personalized support for learners. This study examined Iranian university students’ behavioral intention to use ChatGPT for academic purposes, drawing on the Unified Theory of Acceptance and Use of Technology (UTAUT) and its extensions to investigate the influence of performance expectancy, effort expectancy, social influence, trust, and perceived risks. A cross-sectional survey was administered to undergraduate students majoring in Teaching English to Speakers of Other Languages (TESOL), and Bayesian linear regression with model averaging was employed to determine the relative contribution of each predictor. Results indicated that social influence, including endorsement from peers and instructors, and trust in reliability of ChatGPT and ethical integrity are the strongest drivers of adoption, whereas perceived risks, such as concerns about misinformation or data privacy, exert a modest negative effect. The study advances understanding of AI adoption in higher education by illustrating the interplay of technological, psychological, and social factors in shaping student decision-making. Practical recommendations are provided for educators, administrators, and policymakers to encourage responsible integration of generative AI tools, focusing on trust-building measures, peer-led diffusion strategies, and critical AI literacy training. The research also demonstrates the value of Bayesian analytical approaches for assessing relative predictor importance, offering a methodological contribution to technology acceptance studies.

Ähnliche Arbeiten