Social and nonsocial factors in the Childhood Autism Rating Scale.
Break the CARS total into its five factor scales to make cleaner differential calls between autism and PDD-NOS.
01Research in Context
What this study did
The team ran a factor analysis on the Childhood Autism Rating Scale. They wanted to see if the single total score hides smaller, clearer parts.
Kids with autism, kids with PDD-NOS, and non-autistic kids all took the CARS. The math pulled out five tight sub-scales.
What they found
Five factors emerged: social, emotional, sensory, thinking, and body-movement items. Each factor formed its own mini-scale.
When the researchers scored kids with these new mini-scales, autism, PDD-NOS, and non-autistic groups split apart more cleanly than with the old total score.
How this fits with other research
Mazur et al. (1992) had already shown the full CARS lines up well with DSM-III-R. The 1999 paper keeps that validity but adds sharper blades inside the same tool.
Saemundsen et al. (2003) later saw CARS catch more cases than the ADI-R. The five-factor view explains why: CARS spreads items across domains, so it does not miss kids who look different on only one or two factors.
Lancioni et al. (2006) extended the idea to toddlers. They found CARS still agreed with ADOS-G, even when toddlers had few repetitive behaviors. The factor structure likely helps CARS stay sensitive at young ages.
Why it matters
You can now score the five CARS factors instead of relying on one lump sum. If a child scores high on sensory and low on social, you can flag sensory intervention needs early and dig deeper into social skills. The sharper split also helps you explain to parents why a child lands in autism versus PDD-NOS in plain, factor-based language.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Pull last month’s CARS protocols and re-score them with the five-factor keys to see if any kids shift diagnostic buckets.
02At a glance
03Original abstract
The Childhood Autism Rating Scale (CARS) was factor analyzed to determine if distinct and independent "subgroups" of symptoms could be derived, which would be consistent with the current multidimensional theory and nosology for autism. To address this issue, the CARS was factor analyzed for a sample of 90 children with diagnoses of either autism or PDDNOS, based on DSM-III-R diagnostic criteria. Five factors emerged: Social Communication, Emotional Reactivity, Social Orienting, Cognitive and Behavioral Consistency, and Odd Sensory Exploration. Factor-based scales were created. These factor-based scales were demonstrated to distinguish subjects with autism from subjects with PDDNOS and nonautistic subjects. Factor-based scores were examined to determine the degree to which they were associated with individual differences (such as age, IQ, gender, history of regression, and history of abnormal EEGs) among children with pervasive developmental disorders (PDDs). The application of these distinct and independent factors may have important clinical and research implications. The generation of factor-based scales may provide information on the nature of the individual differences that are thought to be present among children with autism. Additionally, the use of factor-based scale scores may increase the sensitivity of the CARS for identifying younger and/or higher functioning individuals within the PDD spectrum.
Journal of autism and developmental disorders, 1999 · doi:10.1023/a:1022111419409