so you want to test whether the cross-sectional slope changes across time? That is unusual, but you can do it by specifying two classes in the FSGD file, one for each time point. Also specify a variable which is your measurement. Then create a contrast 0 0 1 -1. This will test whether the slopes change between "groups", where a "group" is the given time point. This analysis does not make sense to me, but that is how to do it. doug
On 02/18/2016 12:16 PM, amirhossein manzouri wrote:
Hi, We would like to run paired t-test on longitudinal within group data for correlation analysis of one measurement and cortical thickness, not cortical thickness comparison at the respective time point . I found the paired analysis and read that but in fact we need to see if correlation between our measure and CTH differs from time point one to time point 2. So the question will does the thickness and measure1 correlation differs between time point1 and tp2?
Best regards, Amirhossein Manzouri