Looks ok - though you should probably change EV2 to have -1 and 1
values to be orthogonal to the grand mean.
Cheers.
On 31 Oct 2008, at 19:23, liang wang wrote:
> Hi FSLers
>
> Following the regression questions, now I have a similar issue.
>
> I want to examine the relationship between FA and BEV, group
> difference and the interaction between group*BEV together while
> removing the effect of age. Is the following regress model ok?
>
> Group EV1(beh, demeaned values) EV2 (group) EV3 (interaction,
> group*bev) EV4 (AGE, demeaned) EV5 (grand mean)
> 1 -1
> 1 -1 -12 1
> 1 -3
> 1 -3 -10 1
> 1 -5
> 1 -5
> -7 1
> 1 3
> 2 6 12 1
> 1 1
> 2 2
> 7 1
> 1 5
> 2 10 10 1
> Having EV5 that model the mean, I don't need to add -D option. Is it
> correct?
>
> --
> Liang Wang, Postdoc
> Department of Psychiatry
> University of British Columbia
> BC Mental Health & Addiction Services
> 938 West 28th Avenue
> Vancouver BC V5Z 4H4
> Telephone: 604-875-2000 (ext. 4735)
> Fax: 604-875-3871
> Email: [log in to unmask]
---------------------------------------------------------------------------
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director, Oxford University FMRIB Centre
FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
---------------------------------------------------------------------------
|