Differences Between Groups

Tests for comparing groups on central tendency, variance, or proportions

Most research questions in clinical and biomedical work are about differences: does treatment A outperform treatment B; does a biomarker differ between healthy and diseased patients; does a surgical technique shorten recovery. This section covers the full toolkit for such comparisons.

Subsections

  • Central tendency – t-tests, ANOVA family, and their non-parametric analogues (Mann-Whitney, Wilcoxon, Kruskal-Wallis, Friedman, sign test).
  • Proportions and frequencies – binomial tests and chi-squared goodness-of-fit.
  • Variances – F-test, chi-squared variance test, Levene’s test.

Choosing the right test

Three questions usually narrow the choice to a single test:

  1. Scale level of the outcome: metric, ordinal, or nominal.
  2. Group structure: two groups or more; independent or paired / repeated.
  3. Normality and variance homogeneity: only parametric tests require these.

The interactive wizard walks through this triage in plain English and returns the recommended test with a link to the full method page.

Structure inspired by the University of Zurich Methodenberatung (methodenberatung.uzh.ch). All text, examples, R code, and reporting sentences are independently authored in English.