Dear Haris,
Flexible factorial is not recommended for group comparison, as has
correct df / error term for within-subject and not between-subject
analysis. What you could do is an interaction comparing differences in
conditions between groups. With Design of G1C1 G1C2 G2C1 G2C2 it would
be 1 -1 -1 1.
Best regards,
Jacek
W dniu 24.10.2018 o 11:36, Charis Styliadis pisze:
> Hi to all, I just realized that I have written it wrong.
>
> I have estimated a two-sample t-test, not a paired one.
>
> For the flexible factorial design, I have tried to setup the contrasts that seemed logical to me but I always get an invalid contrast error
>
> Thanks
--
Laboratory of Brain Imaging (LOBI)
Neurobiology Center
The Nencki Institute of Experimental Biology
Pasteur 3, 02-093 Warsaw, Poland
Tel.: +48 22 5892 551
mail: [log in to unmask]
website: http://lobi.nencki.gov.pl/
|