Hi
(The attachments are too big to be posted, so I have to add links and post it again to the forum, really sorry if you keep seeing this)
I guess variance between groups is likely to be different.
So instead of using this model: http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/GLM#Two-Group_Difference_Adjusted_for_Covariate
we had better use this model:
http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/FEAT/UserGuide#Unpaired_Two-Group_Difference_.28Two-Sample_Unpaired_T-Test.29
But just want to make sure "how to add age as a covariate"
My guess is that we demean across all subjects and not within group, then we create 2 more EVs, one for group 1 age, and another for group 2 age. As for the contrasts, we just put 0 in these 2 new EVs???
Thanks so much!!!
|