Mann-Whitney U Test Calculator
Enter two groups of numbers (one per line or comma-separated) to calculate the Mann-Whitney U statistic.
Last updated: September 5, 2026
Frequently asked questions
What is the Mann-Whitney U test used for?
Comparing two independent groups to see if they come from the same distribution, without assuming the data is normally distributed - it's the non-parametric alternative to the independent two-sample t-test, based on ranking rather than means.
How is U calculated?
Combine both groups, rank all values together (averaging ranks for ties), sum the ranks for group 1 (R1), then U1 = R1 − n1(n1+1)/2. U2 is found the same way for group 2, or as n1×n2 − U1. The reported U statistic is the smaller of U1 and U2.
How do I know if my result is significant?
For larger samples, compare the z-score to a standard normal critical value (like ±1.96 for a 0.05 two-tailed test). For smaller samples (roughly n1 or n2 under 8-10), the normal approximation is less reliable - look up your exact U statistic against a Mann-Whitney critical value table instead.
When should I use this instead of a t-test?
When your data isn't normally distributed, contains outliers, or is ordinal rather than truly numeric - the Mann-Whitney test only relies on rank order, making it more robust than the t-test to non-normal distributions and extreme values.
What does it mean if the two groups have completely non-overlapping ranks?
That's the most extreme possible result - if every value in one group ranks below every value in the other, U for the lower-ranked group is 0, indicating the strongest possible separation between the groups.
How are tied values handled when ranking?
Tied values are given the average of the ranks they would have occupied - for example, if two values tie for 3rd and 4th place, both receive a rank of 3.5, keeping the total sum of ranks consistent.
Is the Mann-Whitney U test the same as the Wilcoxon rank-sum test?
Yes - they're mathematically equivalent tests with different names; the Wilcoxon rank-sum statistic can be converted directly to the Mann-Whitney U statistic and both compare two independent groups using rank order.