Actually, the wiki page is not quite right. For a simple interaction between
age and weight, you would need to create a new regressor = age*weight, then set
the contrast to [0 0 0 0 0 0 1]. If you want an interaction between group, age,
and weight, then you would need two new regressors, age*weight for group1 and
age*weight for group2, then the contrast will be [0 0 0 0 0 0 1 -1]. I would
suggest demeaning both age and weight prior to computing the product.
On 8/27/2019 10:12 AM, Zhi Li wrote:
External Email - Use Caution
Dear Freesurfer experts,
May I ask how to understand the contrast of interaction between covariates in
the example https://surfer.nmr.mgh.harvard.edu/fswiki/Fsgdf2G2V? The
interaction of group x age x weight is:
[0 0 1 -1 0 0
0 0 0 0 1 -1]
Why the interaction between group and weight should be put in the second row
(what does each row mean in the F-contrast matrix)?
If there are three levels in the factor one (three groups), I guess the
interaction of group x age x weight should be:
[0 0 0 1 -1 0 0 0 0
0 0 0 0 1 -1 0 0 0
0 0 0 0 0 0 1 -1 0
0 0 0 0 0 0 0 1 -1]
Is this correct? If not how should I set this contrast?
In addition, if I would like to know the interaction between age and weight
without group, how should I set the contrast then?
Looking forward to your kind suggestions.
Thanks and best wishes,
Lizhi
_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu<mailto:Freesurfer@nmr.mgh.harvard.edu>
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
