Hi,
 
I'm working on a study that has a baseline and two follow-up time points for two separate groups (three time points each). However, the n is different for each group and time point.
 
The group numbers are never even. One group has a few more subjects than the other at baseline, same number return at immediate follow up visit from baseline, but less subjects return for the last time point. 

I can compute within group differences without a problem, but I'm trying to compute between group differences, and I'm getting an error (using V17.f). 
 
Do you have any suggestion?
 
 Thank you