Dear Ariadna,
Not necessarily, the difference will be the same than the differences
between models 2 and 3 here:
http://mumford.fmripower.org/mean_centering/
If you have a covariate of no interest, that you want to control for
when testing for the main effect of group then it is likely that you
want to leave "Interactions" to "none".
Best regards,
Guillaume.
On 27/05/2019 21:33, Ariadna Albajara Saenz wrote:
> Dear SPM experts,
>
> In a two-sample t-test (2 groups) or a one-way anova (with 3 groups), is it necessary to choose "Interaction with factor1" when adding a covariate of no interest?
>
> Thank you in advance,
>
> Ariadna
>
--
Guillaume Flandin, PhD
Wellcome Centre for Human Neuroimaging
UCL Queen Square Institute of Neurology
London WC1N 3BG
|