Model Results: Multiverse Analysis of a Social-Emotional Learning Intervention

Secondary analysis of data from a cluster randomized control trial involving 62 schools and 2,425 children. The dataset contains results from a multiverse framework applying ten pre-registered CACE model specifications to assess a universal social-emotional learning intervention's efficacy on student relational outcomes. Authored by Annie O'Brien and last updated on 2026-06-02.

Use Cases

• Analyzing the sensitivity of causal effect estimates based on different definitions of compliance (fidelity, dosage, quality, responsiveness, reach)
• Comparing intervention effect sizes and statistical significance across pre-registered model specifications
• Investigating the relationship between implementation variability and student outcomes like bullying, peer support, and loneliness

Strengths

• Dataset is based on a substantial trial involving 62 schools and 2,425 children
• Analysis employs a pre-registered multiverse framework with ten theoretically plausible model specifications
• Results include specific effect sizes, confidence intervals, and entropy values for model comparisons

Limitations

• Row count is unknown, which may limit suitability assessment
• Column-level documentation is absent; field semantics must be inferred after download
• The dataset is very small at 44.8 KB, indicating limited scope

Provenance

Source

figshare

Collection Method

Secondary analysis of data from a cluster randomized control trial of the 'Passport' social-emotional learning intervention.

Freshness

Last updated 2026-06-02 17:41:57; freshness should be verified