You are almost there. In the first step, you would test for an interaction between age and diagnosis in the DODS model with:
0 0 0 0 1 1 -1 -1
If there is no interaction, then you can move to the DOSS model (which includes age) with contrast
0.5 0.5 -0.5 0.5 0
You need that last 0 to account for the age regressor

On 10/29/2019 3:50 PM, Georgina Angelopoulou wrote:

        External Email - Use Caution        

Dear FreeSurfer Developers, 
 
I am trying to design a matrix in order to run a group comparison analysis and I would like to verify if I am following the right steps.
I have read a lot of things in the forum, but I got confused regarding how to treat nuisance factors. 
I have two groups (patients and controls) and I want to test if there is a significant difference in cortical thickness regressing out
the effects of age and sex. 
My fsgd file is ControlMales
                ControlFemales
                PatientMales 
                PatientFemales
 
Again, based on what I have read so far, in order to use the age as a nuisance variable, I first tested the age slopes using the
row values of age (meaning without centering them), is that correct? 
 
Since there is no significant difference in age slopes, that means that now I can run the analysis without including the
age at all as a nuisance variable in the model? (what you usually refer as DOSS?) with the contrast matrix 
0.5 0.5 -0.5 0.5 (meaning that I only control for the effect of sex)?
 
Thank you in advance,
Georgina

_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer