Hi Panos, are you asking whether theunbalanced design will leadto one slope being different than another just because the numbers are different? That is not a problem. You can simply follow the instructions on the wiki. doug
On 05/31/2013 11:26 AM, Fotiadis, Panagiotis wrote:
Hey Doug,
Sorry for the overwhelming amount of questions. I wanted to confirm something. Suppose you want to do group analysis between two groups say a healthy cohort vs a diseased cohort. Each cohort has ~40 and ~30 subjects each (different people), and you want to compare how for instance the cortical thickness decreases over age in the one group versus the other. Do you still set the contrast vector as [0 0 1 -1] ([intercept 1 intercept 2 slope 1 slope 2]) and follow the analysis described in https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/GroupAnalysis or do you go on a different way since there are different subjects instead of one with 40 (or 30) different longitudinal data for each cohort? Don't we need to correct the age slopes for the fact that we have 40 (and 30) different subjects in each cohort?
Thanks, Panos _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer