Work out mann whitney u test instantly with clear inputs, formula shown and shareable results.
The Mann-Whitney U test ranks all observations together and compares the rank sums. It tests whether one group tends to give larger values than the other, without assuming normality.
U statistic
U₁ = R₁ − n₁(n₁ + 1)/2, U = min(U₁, U₂)
Normal approximation
z = (U − n₁n₂/2) / √(n₁n₂(n₁ + n₂ + 1)/12)
The rank sum of the first group is 31, giving U = 16 and a z of about 0.3651 — no significant difference.
When the data is ordinal, heavily skewed, or the samples are too small to trust normality.