checking-test-assumptions

What problem does it solve?

This Skill prevents invalid statistical conclusions by checking whether the assumptions behind a chosen test are actually met before you interpret the result.

Core Features & Use Cases

• Test-specific gating: Maps common tests to the correct assumptions, such as normality of differences for paired t-tests, per-group normality and equal variance for two-sample t-tests, and expected cell counts for chi-squared.
• Evidence-based verdicts: Reports pass or fail for each assumption with the relevant statistic, p-value, and consequence if the assumption fails.
• Recommended alternatives: Directs users to the right fallback, such as Mann-Whitney, Welch's t-test, Fisher's exact test, or robust regression, when a check fails.
• Use case: A researcher asks whether to trust a t-test, ANOVA, or regression output, and the Skill returns the exact diagnostic checklist needed to decide the next step.

Quick Start

Check the assumptions for my planned two-sample t-test and tell me whether I should proceed with pooled t, switch to Welch's t, or use a non-parametric alternative.