Assessment & Research

A comparison and evaluation of three commonly used autism scales.

Sevin et al. (1991) · Journal of autism and developmental disorders 1991
★ The Verdict

ABC, RLRS, and CARS each carry different strengths—check the tables and later updates before you choose.

✓ Read this if BCBAs who screen or diagnose autism in clinics or schools.
✗ Skip if RBTs whose role ends after data collection.

01Research in Context

01

What this study did

The authors lined up three popular autism checklists: ABC, RLRS, and CARS.

They looked at how each scale matched DSM-III-R and how reliable each felt.

No new kids were treated; the team just compared the forms head-to-head.

02

What they found

The paper gives raw numbers for reliability and DSM fit, but it does not crown a winner.

In plain words, each tool has strong and weak spots; you need the tables to pick one.

03

How this fits with other research

Gerhardt et al. (1991) ran a direct replication the same year and showed the ABC total score alone can sort kids with and without autism 91% of the time.

Dawkins et al. (2016) later superseded the CARS results by proving the newer CARS-2 still lines up with DSM-5.

Matson et al. (1994) extended the CARS work by finding which factor best spots autism and tracks change after treatment.

Lan et al. (2025) keeps the tradition alive, pitting ABC against a newer screener and again showing ABC’s edge.

04

Why it matters

If you test for autism, do not grab a form out of habit. Check the latest psychometric update first. Use ABC for quick screening, CARS-2 for DSM-5 alignment, and always peek at factor scores if you plan to measure growth.

FREE CEUs

Get CEUs on This Topic — Free

The ABA Clubhouse has 60+ on-demand CEUs including ethics, supervision, and clinical topics like this one. Plus a new live CEU every Wednesday.

60+ on-demand CEUs (ethics, supervision, general)
New live CEU every Wednesday
Community of 500+ BCBAs
100% free to join
Join The ABA Clubhouse — Free →
→ Action — try this Monday

Pull the ABC score first for fast screening, then confirm with CARS-2 if DSM-5 criteria matter.

02At a glance

Intervention
not applicable
Design
other
Sample size
24
Population
autism spectrum disorder
Finding
not reported

03Original abstract

Reliability and validity of three commonly used autism scales, the Autism Behavior Checklist (Krug, Arick, & Almond, 1980), the Real Life Rating Scale (Freeman, Ritvo, Yokota, & Ritvo, 1986), and the Childhood Autism Rating Scale (Schopler, Reichler, & Renner, 1988), were investigated. Data analyses were based on completed protocols for 24 children or adolescents who met DSM-III-R criteria for pervasive developmental disorders. First, to replicate previous findings, interrater reliability of each of the two direct observational scales was assessed. Second, correlations between pairs of the three scales were calculated. Third, diagnostic classifications based on autism scale cutoff scores were compared to classifications based on DSM-III-R criteria. Fourth, relationships between autism scale scores and adaptive behavior scores were investigated. Results and implications for the use of these scales in the assessment of autistic behaviors are discussed.

Journal of autism and developmental disorders, 1991 · doi:10.1007/BF02206868