Dear Donald,
Thank you very much for your reply. Now it is much clearer to me how to build a contrast.
Still, I have a doubt regarding which is the best contrast to test my working hypothesis.
I want to test how the covariate (behavioral measure) correlates with the brain structure, and if this correlation is different as a function of the group. That is, to test for an interaction between Group and the covariate of interest.
I predict a positive correlation between the brain structure and the behavioral measure in certain regions, and that this correlation would be stronger in group 1.
My aim, thus, is to test the differences between these three contrasts:
0 0 0 1 0 0 > 0 0 0 0 1 0 + 0 0 0 0 0 1
Which contrast should I apply to do it?
Thanks in advance,
Borja
|