Dear SPMers;
I am using SPM to extract brain gray matter volume (GMW) and white
matter volume (WMW), and want to use them to predict certain cognitive
function (COG).
I want to compare two models: one using GMW only and the other using WMW
only as predictor. Although it is generally suggested that the model
with higher R-square is better, I want to know whether there is a way to
test whether the difference of R-square between two models is
significant or not since in our study the two values are close.
I guess the conventional F-test, suitable for comparison of full and
reduced models, may not be applied in the circumstance. Please correct
me if I am wrong.
Any suggestion and comment are welcome and more than appreciated!
Regards,
Deqiang Qiu
Department of Diagnostic Radiology
The University of Hong Kong
Tel: (+852) 2855 3307
Fax: (+852) 2855 1652
|