Dear Freesurfer experts,
I'm performing a group analysis in Qdec correlating a biological marker and the CTh, and I would like to covariate the gender and age effect, i.e. ideally get rid of both effects and isolate the biological marker effect. So, I understand that the age effect is (at least partially) removed when I select it as a Nuisance variable in the Design Tab. I've treated the Gender both as a continous and as a discrete variable. When treating it as a continous variable, I can select it as a Nuisance Factor in the design tab, so I understand that the Gender effect is theoretically removed. However, the gender is not actually a continous variable. So, when treating the Gender as a discrete variable, I can't select it in the nuisance factors. The only option I have is to select it as a Discrete (Fixed Factors) in the Design tab and then select the Option "Does the correlation between thickness and the biological parameter, accounting for GENDER, differ from zero?". Is that one the correct option in order to covariate (get rid off) the gender effect? In other words, is this the correct option to treat a discrete variable as a "Nuisance Factor"?
Any comments would be greatly appreciated,
Thanks in advance for your attention,
Eduard