Hi Patricia, I am not an authority on this, but Tom is, so I would take his word for it ;-) As he said, if you're not dying for degrees of freedom, then the first approach might be the best one, as then the other C & D groups' variance will not influence your results when comparing A & B... If you've got significant results already, you don't need the uncorrected p-values. Cheers, Gwenaelle De: Patricia Pires <[log in to unmask]> Objet: Re: [FSL] Re : [FSL] Glm À: [log in to unmask] Date: Jeudi 25 novembre 2010, 12h03 Hello Gwenaelle, Thank you again...I chose the ANOVA because of some people told me it was more correct than t-test, but if you tell me it's ok for my model i'll choose 2 sample ttest unpaired, then. :) I have signifficant results with corrected option in randomise, is necessary run -x option in this case? Bests regards, Patricia. 2010/11/25 Gwenaëlle DOUAUD <[log in to unmask]" target="_blank">[log in to unmask]> Hi Patricia, it looks ok to me, though I am not sure why you would choose to follow the second model rather than the first one based on the previous answer, especially if you're only looking at the contrasts between the healthy group and every other disease group... You might want to add the -x option in randomise to output the uncorrected p-values. Cheers, Gwenaelle -------------------------------------------------------------------- Gwenaëlle Douaud, PhD FMRIB Centre, University of Oxford John Radcliffe Hospital, Headington OX3 9DU Oxford UK Tel: +44 (0) 1865 222 523 Fax: +44 (0) 1865 222 717 www.fmrib.ox.ac.uk/~douaud --------------------------------------------------------------------