Friedman's ANOVA

Question 1

What are the key features of this test?

Answer 1

Non-parametric alternative to one-way repeated measures ANOVA, testing for differences between two or more conditions and using ordinal level data

Question 2

How do we conduct this test?

Answer 2

Much as we do for the Wilcoxin signed rank, we need to rank our data - scores are ranked and then ranks are summed for each condition
Use chi-squared table to assess significance

Question 3

How is post hoc analysis conducted?

Answer 3

Multiple Wilcoxin signed rank tests
In order to reduce risk of type 1 error we want to use a Bonferroni correction i.e. critical value (.05) divided by number of groups to get a more appropriate critical value