What is the permutation test equivalent for a mixed ANOVA?

Hi Mike and friends, I’m wondering (is there a way?) how I could run the permutation test if I want to test a within-subject effect (Condition A vs Condition B), a between-subject effect (Group A vs. Group B), and their interaction at the same time?

Hi Chao. There is something called a permutation ANOVA. To be honest, I’ve never really looked into it, but I assume it works the same way as a regular permutation test: shuffle the data across factors/levels N times to build distributions of H0 F-values.

Thank you for the information.

You can also check the factorial permutation-based ANOVA.

This paper is associated with a tool called FMUT, which can be useful if you have your data in EEGLAB format.

Thank you for the resources!