not sure what your question is. The pcc should have the same sign as the contrast.
On 09/13/2018 12:20 PM, Nicola Toschi wrote:
External Email - Use CautionHi List,
a dimple question about using partial correlation (pcc.mgh) as effect size measure. I am running multivariate regression and setting positive and negative contrasts one variable at a time, e.g. a design matrix with columns like
Intercept Age Gender Fact1 Fact2 ....
and a bunch of contrasts like
0 0 0 1 0
0 0 0 -1 0
Now let's say the second contrast gives me a significant results after MC correction in a certain cluster, what sign should i expect in PCC in that cluster (on average)?
Thank you very much in advance,
Nicola
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer