you can select age, IQ and mean thickness as 'nuisance' variables, and then your group can be the main discrete variable.

as for mean thickness causing a failure, it it probably the case that the values are too close to each other. treating mean thickness as a covariate is probably not necessary anyway.

n.

On Tue, 2011-09-20 at 16:23 -0600, Dalwani, Manish wrote:

Hello Freesurfer’s,

I am using QDEC and I have created and FSGD descriptor file with 2 groups (pts and ctrls) and three continuous variables: age, IQ and mean thickness. I have 2 questions:

 1. I can only choose 2 covariates at one time? How can I choose
    all three covariates in one model in my design?
 2. I tried group and 2 covariates combinations to check if it
    runs ok and found that when I select age and IQ, it
    runs...when I select IQ and mean thickness it runs but it
    fails when I try running with age and mean thickness in the
    model? The error in the terminal says matrix not correct,
    check duplicates..., etc..

What I am doing wrong?

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