Document Type : Research Paper

Authors

• Ali khodaei ¹
• Omid Shokri ²

¹ Department of Psychology, Faculty of Educational Sciences and Psychology, Payam Noor University,

² Assistant Professor, Educational Psychology Dept., Faculty of Psychology and Educational Sciences, Shahid Beheshti University, Tehran, Iran.

Abstract

The rapid adoption of ChatGPT in academic settings has raised concerns about its impact on learning and research. This study aimed to investigate the factor structure and psychometric properties of the Persian version of the ChatGPT Usage Scale in postgraduate students. In this descriptive survey study, 350 students (179 females and 171 males) from Shahid Beheshti University were selected using convenience sampling method and responded to the ChatGPT Usage Scale (Nemt-allah et al., 2024), the Self-Directed Learning Scale (Lounsbury & Gibson, 2006) and the Brief Regulation of Motivation Scale (Kim et al., 2018). The results of the confirmatory factor analysis statistical method empirically supported the three-factor first-order structure of the ChatGPT Usage Scale, which includes, academ‌ic writing aid, academic task support and reliance and trust. Also, the correlation between the ChatGPT Usage Scale with Self-Directed Learning ability and Regulation of Motivation in postgraduate students supported the criterion validity of the scale. The internal consistency coefficients of the ChatGPT usage subscales were obtained using Cronbach's alpha method and were acceptable. Overall, the results of this study, by replicating the three-factor structure of the ChatGPT Usage Scale in Iranian university students, while defending the validity of its underlying theoretical perspective, provided convincing evidence in support of the acceptability of the technical, validity, and reliability characteristics of the Persian version of the scale.

Keywords

• Psychometric properties
• validation
• ChatGPT Usage Scale
• university students