Tests that do not require assumptions about the populations
Many of the statistical algorithms outlined in preceding chapters are classified as parametric tests because specific assumptions must be made, such as a normally distributed population or equal variances for two populations. If these assumptions are invalid, the results garnered from the statistical tests are invalid as well.
Nonparametric tests are not restricted by such assumptions, and they are relatively easy to perform. However, they tend to be less precise than parametric tests and require more compelling evidence to reject the null hypothesis.
This chapter will investigate a series of nonparametric tests that can be performed on one or more populations, ...