Hi Donald,
Thanks for you reply. Sorry there was an error: In (3), if I did not adjust data for the F contrast, there would still be correlation.
I'm confused when should I adjust and when I don't have to adjust. In PPI/gPPI, we have to adjust to account for null space. Why no adjustment here? Also, I noticed that extracted eigenvalues with adjustment and without adjustment are highly correlated with each other. So what's the relationship between them?
Thanks again
Mark
|