R Statistics Cookbook

There's more...

Similarly to the paired t-test, the Mann-Whitney-Wilcoxon test can be used with paired data. In this case, we will reuse the same example we used for paired t-tests. We first load the data, and we then call wilcox.test with paired=TRUE, as shown in the following code:

data = read.csv("./pre_post_employee.csv")wilcox.test(data$post_bonus, data$pre_bonus,paired=TRUE) 

In this case, we reject the null hypothesis and we conclude that it is not equally likely for a sample taken at random from the pre_bonus group to be larger or smaller than one taken at random from the post_bonus group as follows: