A Gentle Introduction to Non-Parametric Analysis
Oftentimes, researchers find themselves in the situation that the most used parametric analysis assumptions cannot be satisfied when they analyze their data, such as normality and equal variances assumptions. A few attempts at data transformation still cannot remedy the assumptions violation. In these situations, there is a different branch of statistical analysis researchers can resort to --- Non-parametric Analysis. This seminar focuses on three classic non-parametric tests for hypothesis testing which have their corresponding parametric counterparts. They are the Mann-Whitney U test, Wilcoxon signed rank test and the Kruskal Wallis test (Kruskal Wallis one-way analysis of variance).