Thermography for the follow-up of skin and soft tissue infections

The parametric tests are restricted by their assumptions, especially of normal or near-normal distribution. If the data fail to meet these assumptions and the information about their underlying distributions is not known, the predicted parameters, means and the standard deviation, could be invalid. The nonparametric methods of statistical inference are helpful in these situations. Nonparametric tests are the distribution-free alternative to the more powerful parametric test. Like parametric tests, among a dozen nonparametric tests, the test of significance is chosen depending on dependent and independent variables. The chapter will discuss nonparametric tests such as the Mann–Whitney U test, Wilcoxon signed rank-sum test, Kruskal–Wallis test, Jonckheere–Terpstra test, and Friedman and Kendall's W test.