Mann-Whitney U Test calculator

Compare two independent groups with skewed or ordinal data. Perform a Mann-Whitney U test online with rank-biserial correlation.

What it does

Compares two independent groups through ranks instead of raw values, so a single extreme value cannot dominate the result.

Cautions

• If group distributions differ in shape, a significant result means values in one group "tend to be higher/lower" — it is NOT specifically evidence about medians.
• With heavy ties (e.g. a 5-point Likert item) the normal approximation degrades — use the exact/permutation p-value.

Alternatives

• Welch's t — normality is fine and n is adequate — gives a more precise estimate of the mean difference

How to read the output

U statistic

Counts how often a value from one group exceeds a value from the other, across every possible pairing. This does not compare medians. It compares whether values in one group tend to exceed the other's. The two coincide only when the distributions have the same shape, so a significant result with differently-shaped distributions means 'these differ', not 'the median differs by X'.

The group means shown alongside

Printed for description only, so you can see which way the difference runs. The test itself never used them. Do not report the difference in means as the result of this test — it was not tested. Report the direction, the effect size, and a median or quartile summary of each group.

p-value

How surprising a difference at least this large would be IF the groups really were identical on this measure. A small p means the data would be surprising under that assumption. It is not the probability that the result is due to chance, and not the probability that the groups are the same — those read the conditional backwards. It also answers 'would this be surprising?', never 'is this big enough to matter?': with a large sample a trivial difference returns a small p, and the result looks identical to an important one.

How this calculator is validated · Which statistical test should I use?