Dear FSL list experts,

This is a question about a TBSS/randomise analysis.

I ran correlation between scalars (e.g FA) and FSIQ in two groups: a clinical group and controls. The clinical group has a significant correlation between FA and IQ whereas the control group does not.

I would like to check and see if these correlations are significantly different between groups (obviously the groups are much larger than this example).

Would this model answer this question?:

EV1   EV2   EV3   EV4
1 0   IQ 0
1 0 IQ 0
0 1 0 IQ
0 1 0 IQ

contrast: 0, 0 -1 1 and 0,0, 1,-1

Also - should the IQs be demeaned?

Thank you,

Naama