Perform a Wilcoxon signed-rank test for paired non-parametric data.
The Wilcoxon signed-rank test replaces the paired t-test when differences are not normally distributed. It uses ranks of the absolute differences, so outliers have limited influence. It has about 95 per cent of the power of a paired t-test when normality does hold, which makes it a low-cost default for small samples.
Wilcoxon Test
E(W) = n(n+1)/4; SD(W) = √(n(n+1)(2n+1)/24); z = (W − E(W)) ÷ SD(W)
E(W) = n(n+1)/4; SD(W) = √(n(n+1)(2n+1)/24); z = (W − E(W)) ÷ SD(W) The Wilcoxon signed-rank test replaces the paired t-test when differences are not normally distributed. It uses ranks of the absolute differences, so outliers have limited influence.
It has about 95 per cent of the power of a paired t-test when normality does hold, which makes it a low-cost default for small samples.
This calculator takes 3 inputs: Sum of the smaller signed ranks, Number of non-zero pairs, Significance level. The pre-filled defaults are a realistic starting point — replace them with figures from your own environment for a result you can act on.