Contents
- 1 In what situation is it appropriate to run post hoc tests?
- 2 What is the appropriate post hoc procedure in the event of a significant Kruskal Wallis test?
- 3 Is the Wilcoxon post hoc test appropriate for Friedman?
- 4 Why do we use post hoc Wilcoxon signed rank?
- 5 What does a p value mean for the Wilcoxon test?
In what situation is it appropriate to run post hoc tests?
Because post hoc tests are run to confirm where the differences occurred between groups, they should only be run when you have a shown an overall statistically significant difference in group means (i.e., a statistically significant one-way ANOVA result).
What is the appropriate post hoc procedure in the event of a significant Kruskal Wallis test?
If the Kruskal–Wallis test is significant, a post-hoc analysis can be performed to determine which groups differ from each other group. Probably the most popular post-hoc test for the Kruskal–Wallis test is the Dunn test. The Dunn test can be conducted with the dunnTest function in the FSA package.
What does a post-hoc test do?
Post hoc tests allow researchers to locate those specific differences and are calculated only if the omnibus F test is significant. If the overall F test is nonsignificant, then there is no need for the researcher to explore for any specific differences.
Is the Wilcoxon post hoc test appropriate for Friedman?
Yes, it will probably be helpful to conduct post-hoc testing. One approach appropriate for Friedman is Conover test. Another approach is pairwise sign tests with a p-value correction for multiple tests. The Wilcoxon test you mention may be the paired rank sum test.
Why do we use post hoc Wilcoxon signed rank?
That is, if you conduct multiple pairwise comparisons, the probability of falsely rejecting the NULL of in at least one of these tests increases as the number of pairwise comparisons increases. This is the multiple testing problem that is typically introduced in a stat class to motivate Analysis of variance.
Is the Wilcoxon test like the paired rank sum test?
The Wilcoxon test you mention may be the paired rank sum test. This makes sense as a post-hoc test, although my understanding is that Friedman is more like an extended sign test, leaving the paired rank sum test somewhat of a mismatch for Friedman.
What does a p value mean for the Wilcoxon test?
Whether exact or approximate, p-values do not tell us anything about how different these distributions are. For the Wilcoxon test, a p-value is the probability of getting a test statistic as large or larger assuming both distributions are the same.