Hi Josselin,
If you have two groups (2 controls and 3 patients here) and are looking at the differences between the two with an additional covariate, what I mean by "model the mean" is something like:
1 0 -4
1 0 2
0 1 -2
0 1 0
0 1 4
(then there is no need of -D in randomise)
instead of:
1.2 -4
1.2 2
-0.8 -2
-0.8 0
-0.8 4
(in which case you need -D in randomise)
Cheers,
Gwenaelle
> De: Josselin Houenou <[log in to unmask]>
> Objet: Re: [FSL] Re : [FSL] Covariables and demean
> À: [log in to unmask]
> Date: Samedi 12 mars 2011, 17h04
> Hi Gwenaelle,
>
> I have a quite trivial question about your answer. Let's
> assume all the covariables have been demeaned. You say that
> "run randomise with the -D option if you have not modelled
> the mean for each group". What do you mean by "model the
> mean for each group" ?
>
> Thanks,
> Josselin
--------------------------------------------------------------------
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
--------------------------------------------------------------------
|