*(sex: male, female)*X 2

*(age: young, old)*X 3

*(time1, time2, time3)*a mixed analysis of variance, with

*time*as the repeated measures variable/dependent variable.

I found a significant main effect for gender, but I want to find where the gender differences on the time 1, 2, 3 variables are.

I've been told I need to run 3 separate univariate ANOVAs and to change the alpha level to p=.02 using the Bonferroni criterion (.05/3=.02 alpha level). Is this correct? If so, how would I go about doing such an analysis? Also what happens if my "homogeneity of variances" assumption is violated in the univariate ANOVAs?