yes
Cameron Ellis wrote:
Thank you for this information Doug. This is exactly what I was hoping to hear. So to clarify, if I want to contrast two groups on the correlation between thickness and another measure (variable x), I should use dods.
By default, qdec uses "DODS" - different offset, different slope. This means that each class gets its own offset (as with DOSS), but it also gets its own slope, so 0 0 +1 -1 will test the difference between slopes., and 0 0 +1 +1 will test for the slope averaged across the classes. With DOSS, there is only one slope for each continuous variable regardless of the number of classes, and 0 0 1 will test it. I don't think that +1 -1 +1 tests anything meaningful.
doug
Cameron Ellis wrote:
Hi, I have been running analyses in qdec and glmfit and comparing the results. I am puzzled by the results of one analysis that compares two groups. When I loaded a table.dat file in qdec and selected the variables of interest, an fsgd file containing only those variables was generated in the analysis, no problem so far. The problem is that the contrast.mat file contains more "identifiers" than variables I selected.
The generated fsgd file looks something like this: Class a Class b Variables x Input subj1 a 1 Input subj2 b 4 Input subj3 a 2 Input subj4 b 1
and the generated contrast.mat file looks like this: 0 0 +1 -1
I thought that to generate a contrast for these two groups of the correlations between thickness and variable x I shouldn't the contrast.mat file that looks like this: +1 -1 1
Thanks for any insight, Cameron _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
In order to help us help you, please follow the steps in: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting