Differences Between the Childhood Autism Rating Scale and the Social Responsiveness Scale in Assessing Symptoms of Children with Autistic Spectrum Disorder.

The team compared two rating scales used for autism. One is the Childhood Autism Rating Scale (CARS). The other is the Social Responsiveness Scale, second edition (SRS-2).

They gave both tools to the same group of children with autism. Then they checked how often the scores matched.

The scales agreed on 92.7 percent of cases. That sounds high, but the kappa value was only fair. The two scores were only weakly correlated.

Agreement disappeared in children who had poor verbal understanding. When kids could not follow spoken language well, the scales told different stories.

Levin et al. (2014) saw the same patchy agreement. They compared CARS, SRS, and ADOS in preschoolers. Only about half the children got the same severity label across tools.

Rogers et al. (2017) and Emerson et al. (2013) show the SRS family can be trusted when used alone. Their good psychometrics do not contradict Chen et al. (2018); they simply show single-tool reliability, not cross-tool overlap.

Together the papers say: each tool works on its own, yet they measure slightly different slices of autism. Expect splits, especially in kids with language delays.

Before you pick one score for a report, give both CARS and SRS-2. If the numbers clash, look at the child’s verbal comprehension. A child who struggles to understand language may score high on CARS but modest on SRS-2. Use both tools and explain the gap in plain language to parents and payers.