Dear FSL users,
I have posted this question before, yet I still have problems with understanding it.
I'm doing an analysis with three groups, let's say A, B and C. Over these three groups I want to do an F-test.
My matrix looks like this:
A 1 0 0
B 0 1 0
C 0 0 1
A>B 1 -1 0
A>C 1 0 -1
B>A -1 1 0
B>C 0 1 -1
C>A -1 0 1
C>B 0 -1 1
I proceeded my analysis with the suggestion of dr. Jenkinson that I should do the F-test over A>B and A>C, but does this mean that I am not allowed to do the F-test over A, B and C? And since the t-contrasts are post-hoc tests, I do not have to perform corrections. But over A, B and C a t-test is performed and not a post-hoc. Does this mean that I have to do a multiple comparissons correction over these, like a Bonferroni?
I’m also wondering about the correction that needs to be performed on the number of networks that I look at. So let’s say I have 4 networks of interest and for each network I have the before mentioned 9 t-contrasts. Does this mean I need to perform a Bonferroni with a number of 4x9 = 36? Or do I only need a number of 4x3= 12, since my t-contrasts 4 to 9 are already corrected (viewed as post-hoc test) and only 1 to 3 lack this correction?
Thanks in advance,
Sanne
|