Dear Mike,
your contrasts look fine assuming that the first 3 columns are the group factors. Additionally, you can also use in a first step a contrast for the F-test to test whether there are any differences between the correlations of the 3 groups:
[zeros(n,n) eye(n)-1/n]
where n is the number of groups. This results in:
0 0 0 0.6667 -0.3333 -0.3333
0 0 0 -0.3333 0.6667 -0.3333
0 0 0 -0.3333 -0.3333 0.6667
Best,
Christian
On Mon, 22 May 2017 13:21:59 +0100, Mike <[log in to unmask]> wrote:
>Hi,
>
>I have 3 groups (group 1-3) with n=10 in each group, and I want to know brain activity showing different between-group correlations with age during a task period. I select Full Factorial, 1 factor ("group"), 3 levels, then specify cells for the 3 groups by order. In Covariates, I input:
>
>Covariate 1: 10 age numbers of group 1, then 20 zeros
>Covariate 2: 10 zeros, then 10 age numbers of group 2, then 10 zeros
>Covariate 3: 20 zeros, then 10 age numbers of group 3
>
>In contrast setting, for example,
>
>Correlation (group 1>2): 0 0 0 1 -1 0
>Correlation (group 2>3): 0 0 0 0 1 -1
>
>Am I correct?
>
>Thanks
>Mike
|