← Back to Statistics
Non-Parametric Tests
Mann-Whitney U Test
Rank-based comparison of two independent groups.
Calculation engine
Client-side JavaScript; no external statistics package is loaded.
Results are calculated in the browser and are not uploaded.
Client-side JavaScript; no external statistics package is loaded.
Results are calculated in the browser and are not uploaded.
Method overview
When to use
Use for two independent groups when data are ordinal, skewed, or normality is doubtful.
Input requirements
Enter two numeric vectors. Ties are handled using average ranks.
Key assumptions
- Two independent samples.
- Outcome is at least ordinal.
- The test compares rank distributions; interpretation as a median test needs similar distribution shapes.
Null hypothesis
H₀: the two groups have the same distribution of ranks.
Method used
Mann-Whitney U statistic with normal approximation.
References
- Mann HB, Whitney DR. On a test of whether one of two random variables is stochastically larger. Annals of Mathematical Statistics, 1947.
Data input and results
Enter two independent groups. The test uses ranks and is suitable for ordinal or skewed data.
No calculation has been performed.