Abstract
The present study aimed to evaluate the reliability and validity of the Childhood Autism Rating Scale: Second Edition (CARS2) in diagnosing individuals with autism in Iran. A mixed-method approach was used and 313 participants were recruited, with an age range of 2-32 years, for CARS2-Standard Form (ST) and 218 individuals aged 6-25 years for CARS2-High Functioning (HF). The participants were recruited from daycare centers, schools, and clinics with different developmental trajectories: autism, intellectual disabilities, and neurotypical development. All participants with autism and intellectual disabilities had been clinically diagnosed previously. In addition, the CARS2-Questionnaire of Parent Concerns (QPC) was used to gather qualitative data on 30 randomly selected parents and the perspectives of the 20 test administrators were also collected. The CARS2 had high internal consistency, inter-rater reliability, and test-retest reliability. Factor analyses revealed a one-factor structure for CARS2-ST and a three-factor structure for CARS2-HF. When adjustments were made to the cut-off points, the discriminant analyses indicated that CARS2 effectively differentiated between those with autism and typical development but less so with persons who had intellectual disabilities. The qualitative data analysis and the extracted themes suggest that the CARS2-QPC is a valid tool for collecting autism-related information from parents. Our findings suggest that the CARS2 is broadly a reliable and valid instrument for diagnosing autism spectrum in Iran in the absence of more extensive assessments.