Hi Cagri,
Do you have a reason to hypothesize that there is a linear relationship
between the groups? If not, then it doesn't make sense to model the
groups with a linear EV. That said, your model itself is correct,
except contrast [0 1] would capture where the dependent variable
increases as you go from group 1 to group 4 (not from group 4 to group
1). And yes, an intercept term is needed unless you have reason to
believe that the intercept should be forced to be 0.
cheers,
-MH
On Thu, 2012-01-05 at 09:03 +0000, Cagri Yuksel wrote:
> Dear FSL experts,
>
> I am reposting this question with the hope of making it more clear this time.
>
> We have 130 subjects in 4 groups (3 patient groups X,Y,Z and a control group) and we compared the GM volumes between these groups in an ANOVA design.
>
> We were asked by a reviewer to do a 'multiple regression with a group variable (e.g. X=4, Y=3, Z=2, Controls=1)' and, 'an independent estimation of group variables' to examine the continuum of GM abnormalities in this sample.
>
> I was planning to give 4 for group X, 3 for Y, 2 for Z and 1 for controls, and demean these values and enter in EV1.
> Assuming that we have 3 patients in each group the model would look like this:
>
> Groups Ev1 Ev2
> 4 1 1.5
> 4 1 1,5
> 4 1 1,5
> 3 1 0,5
> 3 1 0,5
> 3 1 0,5
> 2 1 -0,5
> 2 1 -0,5
> 2 1 -0,5
> 1 1 -1,5
> 1 1 -1,5
> 1 1 -1,5
>
> (I am not sure if I actually need EV1)
> Contrast A: [ 0 1 ] for the areas with the GM volume increasing as we go from group x to controls (4 to 1)
> Contrast B: [ 0 -1] for the areas with the GM volume increasing as we go from controls to group x (1 to 4)
>
> Do think this is correct ?
>
> If not what are your suggestions ?
>
> Thank you very much
>
> Cagri Yuksel
|