Hey Doug,
I have run a group analysis on my data with two different ways. (The input is cortical thickness.) In the first analysis, I specified 4 groups (Diseased_Male, Diseased_Female, Healthy_Male, and Healthy_Female) whereas in the second, just two groups (Diseased, Healthy). One of the comparisons that I did was the age slope between Diseased and Healthy. Therefore in the first analysis, I set the contrast vector to be [0 0 0 0 0.5 0.5 -0.5 -0.5] and in the second analysis I set it to [0 0 1 -1]. However, the results that came up were somewhat different. For instance there were clusters in the second analysis that were significant but were not significant in the first analysis.
Shouldn't the results be the same?
Thank you,
Panos