Nonparametric ANOVA Dashboard

An R Shiny app for comparing data from two or more groups using nonparametric analysis of variance (ANOVA). For uncensored data, comparisons among groups are made using the parametric ANOVA and the nonparametric Kruskal-Wallis (KW) test. These tests determine whether the groups essentially belong to the same underlying population, or whether there are significant differences between the group means (in the case of parametric ANOVA) or the empirical distribution functions (in the case of the KW test). When the KW test is applied to censored data, all observations below the highest reporting limit are assigned to the same value. When ranked, these observations become tied at the lowest rank. Data above the highest reporting limit are ranked using the same method as for censored data. When survival analysis techniques are used, there is no need to recensor the data using the highest reporting limit. Because of this, survival analysis techniques provide greater power in determining differences between groups of data with multiple censoring limits.

This app was created by Charles Holbert. Check out the app here. The User Guide section of the app contains more information.