Hey all-
This may be a simple question to answer, but in an analysis looking at say cortical thickness between Group 1 and Group 2 with say 3 continues variables (age, anxiety, IQ) there would be a contrast matrix of 1 -1 0 0 0 . How does free surfer correctly remove the effects of the continues variables?
It includes them in the model as regressors, so, for example, the Group 1 mean (beta) is computed simultaneously with the slope. Think of a bunch of dots that form a line. You fit it to y = x*m + y0. y0 is the intercept (ie, expected y at x=0). In your case you have two y0's (one for each group) and three m's (one for each covariate). By using [1 -1 0 0 0], you are specifying that the intercepts be compared. doug
On 06/12/2012 02:37 PM, mdkruepke@uwalumni.com wrote:
Hey all-
This may be a simple question to answer, but in an analysis looking at say cortical thickness between Group 1 and Group 2 with say 3 continues variables (age, anxiety, IQ) there would be a contrast matrix of 1 -1 0 0 0 . How does free surfer correctly remove the effects of the continues variables?
-- Michael D. Kruepke PhD - University of Illinois at Urbana-Champaign BA - Psych - University of Wisconsin-Madison mdkruepke@gmail.com mailto:mdkruepke@gmail.com (262)-483-7449
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
freesurfer@nmr.mgh.harvard.edu
-
Douglas N Greve -
mdkruepke＠uwalumni.com