Hi! I hope my title makes sense and my problem also makes sense! I'm comparing lsmeans of groups of people who meet exercise minute guidelines (variable=guideline_levels, and the groupings are none, insufficient, adequate, and high). And I want to look at their cardiovascular score (iCVH). With the guideline_levels, I want to group compare means by APO (adverse pregnancy outcomes). FYI: The other variables in my code are age, time between visits, race, insurance status, and smoking status. These are just additional variables for my model. Below is a graph I have, but it's not what I'm going for. I would like to have the guideline_levels together. proc glm data= exercise.analyticsample;
class guideline_levels (ref='None') APO crace Insurance Smoking;
model iCVH_v5 = guideline_levels age_at_v1 time_bw_cv_ped crace Insurance Smoking APO guideline_levels*APO;
lsmeans guideline_levels*APO / cl obsmargins;
run;
... View more