Dear All:
Correlation of PET images (N=8) with behavioral data shows a cluster of
size k=174 (p corrected=0.05) and Tmax=17.4 at a height threshold p
=0.005. Plot of data (averaging all cluster) using Matlab, r = 0.98
(p=0.0001) is too good to be true.
Since I have 5 residual d.f. (N-3; 1 behavioral data +1 image data + 1
Ancova global =3), I am trying (see
http://www.mailbase.ac.uk/lists/spm/2000-06/0043.html) partial
correlation r(C|X_0) = sqrt(ess/(ResSS-ess)); ResSS=residual sum of
squares for the full model; ESS=extra sum of squares due to the
covariate.
Examining F-contrast create ess.img and hdr. ResMS (estimated residual
variance image) .img and hdr is created prior to running the F-contrast.
However, I CANNOT find ResSS.
1) Was ResMS mispelled as ResSS?
2) Is my approach correct?
Thanks for any help
Badreddine Bencherif,MD
Department of Radiology
Division of Nuclear Medicine
Johns Hopkins University School of Medicine
601 North Caroline Street / JHOC 4230
Baltimore, MD 21287-0855 USA
Voice: (410) 614-2787 Pager: (410) 283-2050
Fax: (410) 614-1977
|