Dear Michaël,
The contrast you wrote is a parametric contrast testing for a linear
effect over the four levels of factor B (averaged over factor A). The
F-contrast testing for the main effect of B is:
1 -1 0 0 1 -1 0 0
0 1 -1 0 0 1 -1 0
0 0 1 -1 0 0 1 -1
and the t/F-contrast for the main effect of A is:
1 1 1 1 -1 -1 -1 -1
You can get them created automatically with:
c = spm_make_contrasts([2 4]);
Best regards,
Guillaume.
On 21/11/17 13:50, Mouthon Michaël wrote:
> Dear colleges,
> I have a flexible factorial design with Factor A (2 levels) and Factor B (4 levels). I have modeled only the interaction terms (which mean that the design matrix columns are: A1B1, A1B2, A1B3, A1B4, A2B1, A2B2, A2B3, A2B4, subjects....).
>
> Which will be the contrast vector to study the main effect of FactorB ?
>
> The F-contrast : -2 -1 1 2 -2 -1 1 2 zeros(1,# of subjects) is good or not ?
>
> Thank you very much
>
> Michaël
>
--
Guillaume Flandin, PhD
Wellcome Trust Centre for Neuroimaging
University College London
12 Queen Square
London WC1N 3BG
|