Dear SPM-experts,
I would like to run a repeated measures ANOVA in SPM8 with ng=2 groups,
nc=2 conditions and 30 subjects (n). Here, the interaction term should
be the following t-contrast (acc. to the Gläscher & Gitelman tutorial),
which works perfectly fine:
zeros(1,n) zeros(1,ng) zeros(1,nc) 1 -1 -1 1
Now, there is a third group, again with 2 conditions (see image for
design matrix). For the interaction term I feel that I need an F-test
(with post-hoc t-tests) but I am not sure how this would look like. Is
the following correct...
... for the f-contrast (and if yes, why are there no cluster statistics?)...
zeros(1,n) zeros(1,ng) zeros(1,nc) 1 -1 -1 1 0 0
zeros(1,n) zeros(1,ng) zeros(1,nc) 0 0 1 -1 -1 1
...and the 3 separate t-contrasts?
zeros(1,n) zeros(1,ng) zeros(1,nc) 1 -1 -1 1 0 0
zeros(1,n) zeros(1,ng) zeros(1,nc) 0 0 1 -1 -1 1
zeros(1,n) zeros(1,ng) zeros(1,nc) 1 -1 0 0 -1 1
Thank you in advance for your help,
Andreas.
--
Andreas Hahn, MSc.
Functional, Molecular & Translational Neuroimaging
Department of Psychiatry and Psychotherapy
Medical University of Vienna, Austria
Phone: +43 664 8001662002
Email: [log in to unmask]
Web: http://www.meduniwien.ac.at/neuroimaging/
|