Hi Nick,
Your wording of within- vs. between-maps confused me quite a bit before. Thanks for the clarifications.
> A direct comparison with 3dttest++ with -setA, setB and -paired gives a lot
> of blobs for the within>between just because it has higher mean values.
The interpretation of such a direct comparison is very straightforward. However, I see two potential issues with your scaling scheme:
1) The z-scores scaled by their square root of SOS (sum of squares) follow some sort of t-distribution with a scaling factor? Running t-test on such scaled values could be problematic if you don't have enough number of subjects.
2) The square root of SOS depends on the number of subjects, and is not exactly the standard deviation. How are you going to interpret the result from your scaled method?
Gang
Edited 1 time(s). Last edit at 08/06/2013 06:24PM by Gang.