I am in the process of testing my variances to see if I meet the assumptions of a two way ANOVA. While doing this, I was wondering exactly what I have to test.
For example, I used the Fligner Test to test the variances of the interaction: fligner.test(Pros_Width~interaction(Fouling,Sex),data=Pros_Fouling)
.
Would I also need to test my independent variables by themselves such as :
fligner.test(Pros_Width~Sex)
fligner.test(Pros_Width~Fouling)
And if any of these 3 tests were significant, then I would not meet the assumptions of a two way ANOVA? Should I transform if not, or try to find a non-parametric version?