Harma
> Dear SPM-ers
> I would very much appreciate your opinion on the following matter. I am
> analyzing T1 images of two groups using VBM. Often global gray matter is
> used as a nuisance covariate but I cannot do that because my groups differ
> significantly on this covariate, as also commented on in a few other posts
> and in the paper by Miller and Chapman (2001).
I don't understand the problem here - if you run a multiple regression
and each subject and voxel is analyzed such the global is accounted
for; how the fact that it exists a difference in globals affects the
local results in a detrimental way? my understanding is that this is
the opposite; local effects are better understood when accounting for
globals; even more if you have a difference between groups that you
want to remove.
what is the argument in the referenced paper?
> However, it is very common in VBN to use a global measure as a nuisance
> covariate to ‘take out’ or ‘control for’ the effect of global differences in
> gray matter. So maybe another way to approach this would be to create two
> covariates for global gray matter. One for the controls and one for the
> patient group. Each would contain zeros for the other group and their scores
> would be mean corrected. In that way you do ‘correct’ for global differences
> in gray matter, but now I would do it per group. However, I am not sure this
> is sound either, because it would change the groups in different ways.
> Does anyone have an opinion on this matter?
I don't think that splitting the global will change anything since I
assume your regressor of interest are also split with values and 0s
accoring to the groups - one thing that matters though is if you use
zscoreed regressors or not; a usual thing to do is to zscore all your
regressors (not the dummy ones though) so that they all have the same
influence.
Hope this helps
Cyril
--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.
|