That contrast does not make sense because it compares intercept, age slope, and mmse slope. I'm not sure what you are trying to do. Do you want an F-test to test whether the groups differ on intercept OR age OR mmse?
On 02/25/2014 03:48 PM, dgw wrote:
Hi,
I have a quick contrast question. If I don't want to regress out one of my variables for a contrast how do i set that up in a contrast file? For example
fsgd: Class a Class b Class c Class d Class e Variables age mmse Input s1 a 23 30 Input s1 b 25 21 Input s1 c 29 25 Input s1 d 26 23 Input s1 e 27 22
Would I have an mtx that looked like this to contrast e from c and d ? 0 0 -.167 -.167 .333 0 0 -.167 -.167 .333 0 0 -.167 -.167 .333
Does this make sense? Would i need to setup different fsgd files?
Thank You, Dan _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer