Greetings! I am currently working on the Behavioral Risk Factor Surveillance System (BRFSS) and need to do ANOVA of age (continuous) by subpopulations while maintaining the weighting/strata scheme in BRFSS. Currently, I am using the following:
proc surveymeans data=xxxx.xxxxxxx mean df t clm;
domain group*categor*race_cat / dfadj;
Is there a way to modify this test to do an anova or some other type of method that can be employed to maintain the weighting scheme and test for a significant in age based on the group/category/race combo?