Hi - in general we try to dissuade people from using ANOVA when possible,
as we find it much more natural to think in terms of direct modelling of
effects using the GLM. So my temptation would be to have a separate EV for
each of the 6 cells and directly create the relevant contrasts. Would this
give you what you want?
Cheers, Steve.
On Tue, 15 Mar 2005, Cornelius Werner wrote:
> Hello list,
>
> I performed a study with a 2x3 factorial design:
>
> A1 A2 A3
> B1 B2 B3
>
> Apart from the main effect B-A, I'd be interested in the interaction.
> I calculated the following interactions on the first level:
>
> (B1-A1)-(B2-A2)
> AND
> (B2-A2)-(B3-A3)
>
> Now I'd like to put both contrasts into an F-Test with [1 0; 0 1]
> (should equal an ANOVA with partitioned error terms).
>
> Can you give me some hints on how to actually do it in FSL?
>
> Thanks as always,
> Cornelius Werner
>
>
> --
> Cornelius Werner
> Institute of Medicine (IME)
> Cognitive Neurology Group
> Research Centre Juelich
> 52425 Juelich
> Germany
>
> Phone: +49-(0)2461-61-8609
> Mail: [log in to unmask]
>
--
Stephen M. Smith DPhil
Associate Director, FMRIB and Analysis Research Coordinator
Oxford University Centre for Functional MRI of the Brain
John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
|