Hi Henk,
If you are running separately each of the t-tests, yes, you can use -1
to reverse a contrast. For F-tests, it should make no difference (these
are squared).
About demeaning, if you are using the factor effects model, the grand
mean is included, then you don't demean the other EVs.
All the best,
Anderson
Am 18.02.14 09:07, schrieb H van Steenbergen:
> Dear FSL-ers,
>
> I have a follow-up question on this. I am also using an unbalanced design and I would like to investigate interaction between a couple of factors with a different number of levels. I am using the "Factor effects model" template mentioned here: http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/GLM
>
> Is it still allowed to use -1 and +1 here? Or should I demean the individual EVs? And what's the exact rationale for doing this?
>
> Thanks a lot for your help in advance.
>
> Best,
>
> Henk van Steenbergen
|