Statistics for Six Sigma Green Belts with Minitab and JMP

7.6. Wilcoxon Rank Sum Test for The Difference Between Two Medians

In Section 7.3, you used the t test for the difference between the means of samples taken from two independent populations. If sample sizes are small and you cannot or do not want to assume that the data in each group are selected from normally distributed populations, you can use a nonparametric test (a test with relatively mild assumptions) called the Wilcoxon rank sum test for finding the difference between two medians. The Wilcoxon rank sum test is almost as powerful as the pooled-variance and separate-variance t tests under conditions appropriate to these tests and is likely to be more powerful when the assumptions of those t tests are not met. In addition, you can use the ...