Wilcoxon Signed Rank Test

This chapter shows the circumstances where a Wilcoxon signed rank test is used. It presents list of the dataset as an SPSS data file and an MS Excel spreadsheet. The chapter also presents the list of video for detailed instructions on using G*Power to calculate sample sizes for a paired t-test, and SPSS to carry out the Wilcoxon signed rank test. A key part of statistician's discussion will be to compare the effect size seen in their work against the minimum clinically/practically relevant difference. An exact calculation of necessary sample sizes for a study to be analyzed by a Wilcoxon signed rank test is not straightforward. However, a pragmatic approach can be based upon the fact that even with the most unfavorable circumstances, the power of a Wilcoxon signed rank test is 86% of that achieved by applying a paired t-test to the same data.