Mann-Whitney U (Wilcoxon rank-sum) — when and why?
mediumAnswer
- Non-parametric test comparing two independent groups.
- Combines all data, ranks it, compares rank sums between groups.
- Doesn't assume normality — good for skewed, ordinal, or small-sample data.
- Effectively tests whether one distribution is shifted vs the other.
- Less powerful than t-test when normality holds; more robust when it doesn't.
- Not exactly a 'test of medians' — it's a stochastic-dominance test.
Check yourself — multiple choice
- Requires normality
- Rank-sum test on two independent groups; no normality assumption; robust for skewed / ordinal / small samples
- Only for classification
- Same as chi-square
Mann-Whitney U: rank-based two-sample test; robust to skew / non-normality.
#hypothesis-testing#non-parametric
Practise Statistics Fundamentals
215 interview questions in this topic.