Dear FSL users,
I want to compare three groups which corresponds to 3 stages of a disease. I don't really understand how can I do:
1) Is it better to do a t-test or an ANOVA?
2) I don't really understand how to build the design. I would like to do this:
EV1 (stage A) EV2 (stage B) EV3 (stage C)
1 0 0
1 0 0
1 0 0
0 1 0
0 1 0
0 1 0
0 0 1
0 0 1
0 0 1
and I would like to contrast each group with the others:
A vs B -1 1 0
A vs B 1 -1 0
B vs C 0 1 -1
B vs C 0 -1 1
A vs C 1 0 -1
A vs C -1 0 1
3) I have another problem: I have a lot of covariables: age, sexe (binary value), alcohol addiction (binary value), drug addiction (binary value) and presence of treatment (binary value). I demean age for each group or for the entire sample?
4) Have I to change the column "group" given by defaukt by the GLM GUI?
Thanks for all,
C.
|