Psychometric Properties of the Persian Version of the Comprehensive Autism Trait Inventory (CATI)

Karim  Abdolmohamadi; Asgar  Alimohamadi; Jafar Samari  Safa

doi:10.18502/ijps.v20i4.19687

Karim Abdolmohamadi Department of Psychology, Azarbaijan Shahid Madani University, Tabriz, Iran.
Asgar Alimohamadi Department of Psychology and Education of Exceptional Children, Allameh Tabataba’i University, Tehran, Iran.
Jafar Samari Safa Department of Counselling, University of Hormozgan, Bandar Abas, Iran.

DOI: https://doi.org/10.18502/ijps.v20i4.19687

Keywords: Autism; Factor Analysis; Psychological Tests; Surveys and Questionnaires; Validity and Reliability

Abstract

Objective: The present study aimed to culturally adapt and psychometrically evaluate the Comprehensive Autism Trait Inventory (CATI) for use within the Iranian society, thereby facilitating cross-cultural research on autistic traits.

Method: This cross-sectional survey included 1,013 Iranian adults (70.1% female), aged 15 to 50 years (M = 24.91, SD = 5.47), recruited through convenience sampling. The translation of the CATI into Persian was developed based on a carefully implemented forward and backward translation procedure through bilingual experts. Discrepancies were resolved by a panel of psychologists and psychometricians to ensure semantic and conceptual equivalence. The broad autism phenotype questionnaire (BAPQ), a tool designed to measure traits associated with the broad autism phenotype, was also administered to measure convergent validity. In order to assess internal consistency, Cronbach’s alpha was calculated, while construct validity was evaluated through confirmatory factor analysis (CFA) and Pearson correlations with BAPQ scores.

Results: CFA confirmed that all items aligned meaningfully with their intended factors, supporting the persian CATI’s six-factor structure and indicating a good model fit (CFI = 0.93; RMSEA = 0.047; χ² / df = 3.25; GFI = 0.90; SRMR = 0.059). The average variance extracted (AVE) was 0.45 for both the components and the overall scale. Correlations between CATI subscales and the BAPQ ranged from 0.03 to 0.40, demonstrating acceptable convergent and divergent validity. The reliability of the six-factor model was verified by Cronbach’s alpha (0.84) and McDonald’s omega (0.82).

Conclusion: The Persian version of the CATI shows sound psychometric properties for evaluating autistic traits in the Iranian population. These findings support utility of CATI in research settings and highlight the importance of culturally adapted assessment tools for improving diagnosis and intervention across diverse contexts.