> Wouldn't the main effect calculated by running ttest++ without group in the model only be correct if the group and age were
> perfectly orthogonal (which they are not in my case)?
By the main effect of age, I assume that you mean the age effect among all the subjects. If that's the case, it has nothing to do with the orthogonality between group and age.
> it has been stuck on slice 19 for 3 hours
That is strange indeed. Are you running on a server? You many want to check out the status of the jobs.
Gang