On 10/29/2013 01:10 PM, Ashley Shurick wrote:
Hi all,
I'm running analyses in Qdec and want to verify a few things:
- When comparing two groups (HC vs patients) I am including mean
cortical thickness as a covariate, using the following equation:
bh.thickness = (lh.thickness*lh.surfarea + rh.thickness*rh.surfarea) / (lh.surfarea + rh.surfarea)
Is this the best way to calculate global mean thickness?
Yes
- I want to regress out any effects of age, therefore I need to
demean the ages for my groups as a whole, and not perform this calculation on each group separately, correct?
Yes, in which case the test of the difference between groups will be done at an age equal to the mean of the ages.
- I'm including questionnaires as a covariate of interest - do I need
to demean these values as well?
When you do a test on a continuous covariate, that test will be unaffected by demeaning of the covariate. Demeaning will affect the difference between groups if you use a separate covariate for each group. If you have a single covariate across groups, demeaning will have no effect.
doug
Thank you in advance for your help!
Ashley
-- Ashley A. Shurick Ph.D. Candidate Department of Psychology Stanford University
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer