Post-Hoc Tests

Statistics · Axiom Academy

Understanding Multiple Comparisons After ANOVA 2 The Multiple Comparison Problem 3 Family-Wise Error Rate (FWER) 1. Tukey's HSD (Honestly Significant Difference) Best for: All pairwise comparisons with equal sample sizes Characteristics: Moderate power, exact FWER control Best for: Small number of planned comparisons Characteristics: Very conservative, simple to compute Best for: Complex comparisons, including non-pairwise contrasts Characteristics: Most conservative, works for all possible contrasts 6 Interpreting Post-Hoc Results Compare these to your significance level (α). If p < α, the difference is significant. If the CI for the difference doesn't include 0, the groups differ significantly. Groups that don't differ significantly are grouped together in tables.

This is the written version of the interactive lesson above. See the full Statistics course.