Dear experts,
I have been searching through the web without finding any solid response.
I have a Brodmann Area which has been internally parcellated into different subregions.
I've extracted signal change percent with Marsbar and performed planned comparisons between my two conditions,
in each subregion separately, and the corrected for multiple comparisons.
I would like to add a subregion factor to get an idea of how the omnibus F-test looks like.
Can I then directly compare the signal change percent across the subregions directly?
This is the only "partially positive" response I found from Matthew Brett:
"Percent signal change does of course assume the signal is proportional".
http://sourceforge.net/p/marsbar/mailman/message/31271216/
Any help would be highly appreciated!
Thanks a lot,
Emiliano
|