Hi Carlos,
Thanks. I wanted to confirm that you indeed have a "heterogeneous
slopes" model, which you do. Note that you didn't center your scores at
0 prior to splitting them into group-specific slope EVs, which will
impact any contrast involving your group mean EVs (EVs 1 and 2), but is
irrelevant otherwise. See Jeanette Mumford's web page on mean centering
of continuous covariates if you have any questions in that regard.
Given your design matrix and contrasts, you are indeed testing for a
difference in "slope" of the scores between groups, but in the context
of controlling for the other scores that you have included in the model
at the same time. To the extent that those scores are correlated (which
I suspect they are), this can be difficult to interpret. For better
interpretability, if you are interested in the possibility that these
scores have a differential relationship with FA/RD across groups, then
you might want to consider a model in which you test for slope
differences between groups using only one set of scores at a time.
cheers,
-MH
On Wed, 2011-09-07 at 16:36 +0100, Carlos Faraco wrote:
> I mean here is the mat file. Sorry, the listserv has been rejecting my snapshot attachments, this one is in txt format.
>
> Thanks,
>
> Carlos
|