The following Stata output tested: . bysort gender: swilk age -> gender = 1 Shapiro-Wilk W test for normal data Variable Obs W V Z Prob>z age 78 0.96936 2.060 1.581 0.05691 gender = 2 Shapiro-Wilk W test for normal data Variable Obs W V Z Prob>z age 281 0. 89780 20.549 7.073 -> gender = 3 Shapiro-Wilk W test for normal data Variable Obs W V Prob>z age 116 0.97843 2.017 1.569 0. 05832 O a. The independence of age by each level of gender O b. The independence of gender by each level of age O c. The assumption for homogeneity of variance of gender by each level of age O d. The assumption for normality of gender by each level of age O e. The assumption for normality of age by each level of gender Of. The assumption for homogeneity of variance of age by each level of gender