UsefulVerse
Language · English

Free to use, no account

Mann–Whitney U Test Calculator

Paste two independent numeric samples to compare their ranks. The tool reports both U statistics, a two-sided p-value, tie handling, method and signed rank-biserial effect.

Enter 2–500 finite numbers per sample (up to ±1,000,000,000,000). Separate with commas, semicolons, spaces or new lines; use a dot for decimal places. Samples must be independent.

This page runs a two-sided test only. It uses an exact permutation distribution for tie-free samples with at most 40 observations in total; otherwise it uses a tie-corrected normal approximation with continuity correction. All-equal pooled values return p=1.

What does the Mann–Whitney U test compare?

It compares the rank distributions of two independent samples. It is not automatically a test of medians unless additional distribution-shape assumptions hold.

Frequently asked questions

What does the Mann–Whitney U test compare?

It compares the rank distributions of two independent samples. It is not automatically a test of medians unless additional distribution-shape assumptions hold.

Are my sample values uploaded?

No. Ranking and the test are calculated in this browser. Sample values are not sent to a server or saved by this tool.

Is Mann–Whitney the same as the Wilcoxon signed-rank test?

No. Mann–Whitney (Wilcoxon rank-sum) compares two independent groups. The signed-rank test is for paired or matched observations.

Does Mann–Whitney always test whether medians differ?

No. It tests rank distributions. Interpreting it as a median comparison requires additional assumptions, including similarly shaped distributions.

How does the calculator handle ties?

It assigns average ranks to equal values and uses a tie-corrected normal approximation. Its exact permutation method is used only when pooled observations have no ties and the total sample size is at most 40.