The Attitudes Towards Artificial Intelligence Questionnaire (AAIQ) for Medical Students: Psychometric Properties of the Turkish Version

Background: The rapid integration of artificial intelligence (AI) into healthcare practice has resulted in new demands on medical education systems. This study aimed to adapt the Attitudes Towards Artificial Intelligence Questionnaire (AAIQ), originally developed for health profession students, into Turkish and examine its psychometric properties among Turkish medical students. Method: Through convenience sampling, 238 medical students (56.7% female; M = 22.32, SD = 2.58) from various state and foundation universities participated. The scale adaptation process followed a standardized translation and back-translation protocol. Confirmatory factor analysis (CFA) was conducted to examine construct validity, while convergent validity was tested via correlations with digital nativeness and individual innovativeness measures. Internal consistency was evaluated using Cronbach’s alpha, composite reliability (CR), and average variance extracted (AVE). Results: CFA supported the AAIQ's original unidimensional structure, which demonstrated good model fit indices: χ²(35) = 42.51, χ²/df = 1.21, SRMR = .06, RMSEA = .07, TLI = .99, and CFI = .99. Reliability analyses indicated high internal consistency, with α = .88, CR = .95, and AVE = .67. Notably, 81.1% of participants expressed support for the integration of AI into medical education, although only 21.0% had received formal training related to AI. Conclusion: The Turkish version of the AAIQ shows strong validity and reliability, making it a robust tool for evaluating medical students' attitudes toward artificial intelligence.