Assessing the validity and reliability of the Persian version of the artificial intelligence literacy questionnaire of Carlos et al.

Objective: Artificial intelligence literacy is a set of competencies that enable individuals to critically evaluate artificial intelligence technologies, communicate and collaborate effectively with them, and utilize them as tools at home and in the workplace. Parallel to the growing importance of artificial intelligence literacy, various validation studies have been conducted on measurement tools, culminating in newly proposed assessment instruments. Therefore, this study was conducted to adapt, validate, and assess the reliability of the artificial intelligence literacy questionnaire developed by Carlos et al.(2023).
Methodology: This quantitative study was designed to describe the psychometric properties of the artificial intelligence literacy questionnaire. The target population comprised students at Shahid Chamran University of Ahvaz, who were selected using a sample size determined by the number of questionnaire items. Face and content validity indices were evaluated, and confirmatory factor analysis was employed to assess construct validity. Item analysis was performed using a polytomous item response model, and Cronbach's alpha coefficient was calculated to determine reliability.
Findings: The instrument is structured around an eight-factor model comprising: (1) recognition and understanding of artificial intelligence, (2) use and application of artificial intelligence, (3) evaluation and creation of artificial intelligence, (4) artificial intelligence ethics, (5) artificial intelligence problem-solving, (6) artificial intelligence learning, (7) artificial intelligence persuasion literacy, and (8) emotion modulation with artificial intelligence. These eight factors load onto two broader components: "artificial intelligence literacy" and "artificial intelligence self-management." The finalized instrument consists of 30 items. Face validity confirmed the appropriateness of the items for Iranian students, and content validity demonstrated a logical relationship between items and factors. Confirmatory factor analysis indicated that the eight-factor model provides a robust structure for measuring artificial intelligence literacy, exhibiting a good fit with the empirical data. Furthermore, the items demonstrated good item discrimination across varying levels of respondent attitudes within the polytomous model. Reliability analyses confirmed acceptable internal consistency for both individual factors and the questionnaire as a whole.
Conclusion: The results indicate that this questionnaire effectively evaluates artificial intelligence-related skills and attitudes, serving as a useful instrument for researchers, educators, and professionals across diverse fields. Emphasizing psychological skills, the tool enhances the understanding and effective utilization of artificial intelligence while facilitating the development of targeted training programs. Because this study represents an initial effort to establish the reliability of this questionnaire, direct comparisons with prior research are limited. Furthermore, as no prior assessments have specifically measured the artificial intelligence literacy of students at Shahid Chamran University of Ahvaz, this study provides a foundational framework for future, more extensive investigations in this domain.
Originality/Value: A validated Persian-language instrument for measuring artificial intelligence literacy has been largely absent, highlighting a significant gap in the literature. This adapted questionnaire serves as an effective indicator for assessing artificial intelligence literacy and is recommended for evaluating literacy levels across various populations nationwide.