Examining the Factors Influencing Students’ Intention to Use ChatGPT as a Virtual Assistant for Academic Learning

Abstract

Generative artificial intelligence (AI), such as ChatGPT, enables interactive natural language conversations tailored to the specific needs and questions of users in real time, providing on-demand support. This pilot study, conducted among students (n=100) at a Croatian university, analyzes the factors that influence students’ intention to continue using ChatGPT as a virtual assistant for academic learning. Spearman correlation results show that factors such as perceived compatibility, relevance to academic learning, perceived usefulness, quality of results, hedonic motivation, trust, social influence, and computer anxiety are statistically significantly (p<0.01) associated with the intention to use ChatGPT among users (n=77), while a statistically significant weak association (p<0.05) was confirmed with the perception of ChatGPT's ease of use. The results of this pilot study reveal that the intention to use ChatGPT differs not only between users and non-users but also among users, depending on the intensity of use of the tool as a virtual assistant for academic learning. No significant differences were found between users and non-users on factors such as trust, social influence, compatibility, technological issues, and perceived ease of use. Furthermore, regular users reported significantly higher scores on constructs such as perceived usefulness, relevance to academic learning, hedonic motivation, and intention to use ChatGPT compared to occasional users. The findings of this study can inform the development of strategies for implementing generative AI in educational settings and encourage further research on its adoption and acceptance in higher education, ultimately fostering a more personalized and engaging teaching and learning experience.