In the 2nd option, age and weight (and other variables) are simultaneously fit to the data. Most people refer to this as "regressing out" the effect of one variable when looking at the other. But it is not regressing out age and then looking at the correlation of age with the residual. doug
On 03/27/2013 07:02 AM, Martijn Steenwijk wrote:
Dear all,
I've a question regarding computing correlations with cortical thicknesses on the surface; Suppose I have two variables for each subject: (age and weight) and I want to compute the correlation between cortical thickness and weight, corrected for age.
Option 1: Fsgd contains: Variables weight Contrast file contains: 0 1
Option 2: Fsgd contains: Variables age weight Contrast file contains: 0 0 1
Is it true that the effect of age is regressed out in the second option; and it computes the correlation between CT and weight after correcting for age?
Best, Martijn
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer