Child Evidence

Social-skills interventions for autistic children and youth produce sm · Child Evidence

special education / social-skills interventions for autism

Social-skills interventions for autistic children and youth produce small-to-moderate gains on rated and tested social skills, but improvements often reflect social KNOWLEDGE more than real-world social PERFORMANCE, generalization and maintenance across settings are weak, and effect sizes shrink under more rigorous designs and independent informants.

For the Court

Contested. Social-skills interventions yield small-to-moderate gains on rated/tested skills, but it is debated whether they change real-world social performance, given weak generalization and informant effects.

What This Claim Does NOT Say

Does not claim social-skills interventions reliably change real-world social behavior; gains often reflect knowledge rather than performance and may not generalize to school.
Effect sizes vary widely by informant (large self/parent-report, near-null teacher-report) and shrink in rigorous designs.
Does not claim a specific program is established as effective; heterogeneity, low fidelity, and publication bias limit conclusions.
Does not imply social-skills deficits should be 'corrected' to a neurotypical norm; the rating concerns measured outcomes, not goals.

Scope — Where It Holds

A meta-analysis of 19 RCTs found medium overall effects but attributable to social knowledge (g = 1.15), not performance (g = 0.28), with limited generalization (Gates et al., 2017); an earlier meta-analysis of single-subject studies found interventions "minimally effective" (Bellini et al., 2007); an umbrella review rated the evidence only "suggestive" with few results surviving low-bias sensitivity analysis (Gosling et al., 2022); fidelity is often low (Afsharnejad et al., 2023).

Sources & Verification

Seminal

Gates, J. A., Kang, E., & Lerner, M. D (2017). Efficacy of group social skills interventions for youth with autism spectrum disorder: A systematic review and meta-analysis. Clinical Psychology Review.
Meta-analysis
Group social-skills interventions showed medium overall effects (g = 0.51), but the self-report effect was wholly attributable to gains in social KNOWLEDGE (g = 1.15), not social PERFORMANCE (g = 0.28), and may not generalize to school or self-reported behavior.
Bearing on this claim: RCT meta-analysis: gains reflect social KNOWLEDGE (g=1.15) not PERFORMANCE (g=0.28); weak generalization.
doi.org/10.1016/j.cpr.2017.01.006

Supporting

Wolstencroft, J., Robinson, L., Srinivasan, R., et al (2018). A Systematic Review of Group Social Skills Interventions, and Meta-analysis of Outcomes, for Children with High Functioning ASD. Journal of Autism and Developmental Disorders.

Precedent Bridge

Where social-communication goals are part of FAPE under Endrew F., the weak generalization of social-skills interventions cautions that IEP services target real-world performance, not just social knowledge.580 U.S. 386 (2017)

Limits & Counter-considerations

Effects depend heavily on informant (large self/parent-report, near-null teacher-report); generalization and maintenance are weak; fidelity and publication bias limit conclusions; and outcomes are often context-bound.

Full Dossier

Last reviewed June 28, 2026