small-n-stats-guardrails

What problem does it solve?

This Skill helps researchers apply rigorous small-sample statistical guardrails for N=34 studies to prevent overstated claims, ensure adherence to the prespecified statistical analysis plan (SAP), and document necessary caveats alongside reported p-values and effect sizes.

Core Features & Use Cases

• Checklist Enforcement: Validates that test selection matches endpoint type, enforces Holm correction, and reports post-correction p-values.
• Effect Size & Uncertainty: Requires reporting of effect sizes with bootstrap confidence intervals (≥2000 replicates) and runs leave-one-out sensitivity checks.
• Suppression & Reporting Rules: Suppresses unsupported predictive or interaction claims, flags small subgroup analyses as exploratory, and inserts small-N caveats into manuscript language.
• Use case: Run this guardrail before finalizing a manuscript results section to generate a checklist report detailing passed and failed checks with remediation guidance.

Quick Start

Run the small-n-stats-guardrails skill with your results, group sizes, p-values, and effect sizes to produce a checklist report and conditional approval status.