ABSTRACT

Nonparametric tests are useful in the analysis of data that do not meet the conditions required for parametric tests, for example, if researchers are working with small sample sizes or ordinal / rank data, or if the assumption of normally distributed data may not be justified. This chapter presents non-parametric alternatives to t-tests, namely the Mann-Whitney U test, which is analogous to the independent-samples t-test, followed by the Wilcoxon Signed-rank test, which is analogous to the paired-samples t-test.