Here's my problem. DV: Working Memory (WM) performance IdV:Group (1 vs. 2) IdV: Gender (male vs. female)
When I do a normal one way ANOVA with IdV: group and DV: WM_performance, the significance of the F-Test ist .007.
When I do the same one way ANOVA but add gender as a covariate, the significance of the F-Test is different because now WM_performance is equal over gender. So far so good!
If I now make a two way ANOVA with the factors group and gender, the main effect of group changes again! My question is why? Is it because now not only gender but also the interaction (gender * group) is held constant?