Hi Eric, that all looks good. Because you have the age covariate in the
model, your [-1 1 0] contrast is already "controlling" for age. To "not
control for age" would simply mean leaving that covariate out.
Note though that a statisticians might say that another definition of
controlling for age would be that you made sure that the two groups were
age-matched (which you also have here, at least as far as mean age goes).
Cheers, Steve.
On Mon, 1 Nov 2004, Eric Claus wrote:
> Hi!
> I am trying to do a covariate analysis to control for differences that may
> exist between two groups. I have looked at earlier messages in the list,
> but there was still one thing that was unclear - what does the contrast
> look like?
>
> I have my higher level model set up in the following manner. I have two
> groups (each with 12 subjects), so I used 2 EVs to capture these groups,
> and then set my 3rd EV to age, demeaning it.
>
> Group EV1 EV2 EV3(age)
> 1 1 0 3
> 1 1 0 1
> 1 1 0 0
> 1 1 0 -1
> 1 1 0 -3
> 1 0 1 -1
> 1 0 1 -.5
> 1 0 1 0
> 1 0 1 .5
> 1 0 1 1
>
>
>
> If I wanted to know just the group differences, my contrast would look
> like -1 +1 0, right? If I wanted to look at group differences, controlling
> for age, what is the contrast?
>
> Thanks,
> Eric
>
Stephen M. Smith DPhil
Associate Director, FMRIB and Analysis Research Coordinator
Oxford University Centre for Functional MRI of the Brain
John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
|