Conventional statistical tests are usually referred to as “Parametric tests”. In a lot of medical articles, parametric tests are used more frequently than nonparametric tests, because the majority of medical researchers know the former, especially that statistical software packages strongly recommend parametric tests. Parametric tests necessitate an essential assumption; assumption of normality which means that the sample distribution is normally distributed. Nevertheless, parametric tests can be misleading when this assumption is not satisfied. In this case, nonparametric tests are the available alternative methods, because they do not need the normality assumption. Nonparametric tests are the statistical methods built on signs and ranks. This article tackles the basic concepts and practical use of nonparametric tests for the guide to the appropriate use.