The first one will test for a difference between group offset/intercept accounting for age, gender, episodes, and residuals. By setting a regressor contrast element to 0, one accounts for that variable because the variable appears in the model that is fit to the data (and so its effects already removed from other variables). One more thing: I would recommend incorporating gender in the class structure not as a continuous variable. So you would have 4 classes: PatentientsMale, PatientsFemale, HCMale, HCFemale then use contrast [0.5 0.5 -0.5 -0.5 0 0 0 0 0 ...]

On 5/9/2024 3:26 PM, Liliana Wu wrote:

External Email - Use Caution

Hello!

I am having trouble with creating my contrast files. I did go over FsfdExamples (MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be https://surfer.nmr.mgh.harvard.edu/fswiki/FsgdExamples). However, I am unsure how should I create my contrast files if I want do a 2 group comparison while controlling 4 covariates:

Class:

Patients, healthy control

variables are:

Age,gender,episodes,residuals,

Here is how I set up my contrast files:

1 -1 0 0 0 0 0 0 0 0 - Contrast1 (measure group differences)

1 -1 0.125 0.125 0.125 0.125 0.125 0.125 0.125 0.125 – Contrast 2 (measure group differences while accounting for covariates)

For some reason I think I am setting up those contrast files wrong… Any help is appreciated.

Best,

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