I just want to investigate significant correlation with the csf variable correcting for apoe-status, gender, diagnosis and age

below previous correspondence

your classes are right. What are you trying to test with your contrast
matrix?

On 01/20/2016 05:33 AM, Erik Lindberg wrote:
> Dear Freesurfers,
>
> I am running a design in which I want to correct for a number of
> categorical variables: diagnosis, gender and apoe status and one
> continuous variable (age) in a correlation analysis
>
> Did I get this right? (below)
>
> Thanks, Eric
>
> GroupDescriptorFile 1
>
> Title MyTitle
>
> Class AMCI_F_APOE-
>
> Class AMCI_F_APOE+
>
> Class AMCI_MALE_APOE-
>
> Class AMCI_MALE_APOE+
>
> Class CTL_FEMALE_APOE-
>
> Class CTL_FEMALE_APOE+
>
> Class CTL_MALE_APOE-
>
> Class CTL_MALE_APOE+
>
> Class NON_A_MCI_FEMALE_APOE-
>
> Class NON_A_MCI_FEMALE_APOE+
>
> Class NON_A_MCI_MALE_APOE-
>
> Class NON_A_MCI_MALE_APOE+
>
> Class SCI_FEMALE_APOE-
>
> Class SCI_FEMALE_APOE+
>
> Class SCI_MALE_APOE-
>
> Class SCI_MALE_APOE+
>
> Variables AGE_DEMEAN CSF_Ab40_ADx
>
>
> With the contrast matrix: corr.mtx:
>
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1
> 1 1 1 1 1 1 1 1 1 1 1 1 1
>
>
>
> _______________