Dear all,
 
I have 2 short questions regarding a correlation analysis I'd like to perform with mri_glmfit (surface-based DTI data are correctly preprocessed and concatenated into mgh file, statistics are performed on fsaverage space).
 
1) I'd like to look for correlation between the surface-data and a behavioural measure BEH in one group, correcting for gender and age.
 
I set up the following fsgd:
 
GroupDescriptorFile 1
Class Male
Class Female
Variables BEH Age
Input subj1 Male 0.76 72
Input subj2 Male 0.44 67
Input subj3 Female 0.33 80
....
 
My contrast file is: 0 0 0.5 0.5 0 0 (referring to the FSGD examples Fsgd2G2V, GLM with 2*(2+1) columns).
Is this setup correct? If not, how would I have to set up fsgd and C?
 
 
 
2) In a next step, I'd like to answer the question (now including another group2): Where does group1 (OLD) show increased correlation with BEH compared to group2 (YOUNG), correcting for GENDER and AGE within each group. I wonder because AGE is the relevant difference between both groups.
 
GroupDescriptorFile 1
Class OldMale
Class OldFemale
Class YoungMale
Class YoungFemale
Variables BEH Age
Input subj1 OldMale 0.76 72
Input subj2 OldMale 0.44 67
Input subj3 OldFemale 0.33 80
....
Input subj11 YoungMale 0.80 20
Input subj12 YoungMale 0.55 21
Input subj13 YoungFemale 0.44 31
....
 
My contrast file would look like: 0 0 0 0 0.5 0.5 -0.5 -0.5 0 0 0 0 (referring to GLM with 4*(2+1) columns).
Correct?
 
Thank you very much for your help! I am really not sure whether I did it correctly and whether I can trust my results.
 
Best wishes, Robert

Besuchen Sie uns auf: www.uke.de


Universitätsklinikum Hamburg-Eppendorf; Körperschaft des öffentlichen Rechts; Gerichtsstand: Hamburg
Vorstandsmitglieder: Prof. Dr. Martin Zeitz (Vorsitzender), Prof. Dr. Dr. Uwe Koch-Gromus, Astrid Lurati (komm.), Joachim Prölß, Matthias Waldmann (komm.)


SAVE PAPER - THINK BEFORE PRINTING