Hi all,
I have a question about VBM design and contrast. I've three groups, and I'd
like to controll for example age. I am interested in group differences.
group1-group2; group1-group3; ....
1, My design matrix for 6 people looks like:
Group group1 group2 group3 age1 age2 age3
1 1 0 0 31 0 0
1 1 0 0 32 0 0
2 0 1 0 0 25 0
2 0 1 0 0 42 0
3 0 0 1 0 0 30
3 0 0 1 0 0 33
Contrast:
1 -1 0 0 0 0
0 1 -1 0 0 0
and so on
Do I need to demean our nuisance covariates (age1,age2,age3)? I think not,
but I'm not sure. Could somebody prove or confute it?
2, If I didn't like to use a separate covariates for each group:
Group group1 group2 group3 age
1 1 0 0 31
1 1 0 0 32
1 0 1 0 25
1 0 1 0 42
1 0 0 1 30
1 0 0 1 33
Contrast:
1 -1 0 0
0 1 -1 0
Do I need to demean age in this case?
Could somebody help me what is the difference between method 1-2, and
correct them if they are not accurate?
I'd appreciate an early reply.
Thank you,
Gabor
|