Hi,
Check our wiki pages for this:
http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/GLM/Faq
and the linked page by Jeanette Mumford.
In general you should model the slope and intercept separately, and so you should demean your data and include a column of ones (the intercept) at the higher level, though the details depend on the type of design you are running. See our GLM wiki page for more details.
All the best,
Mark
On 28 May 2014, at 13:09, SUBSCRIBE FSL Anonymous <[log in to unmask]> wrote:
> Hi,
>
> I'm doing a correlation between FA values and some psychological data with FSL and I'm quite new in this kind of analysis, so my question is: should I demeaned all the data? even age and gender (using them as covariates)? What's exactly the difference between demeaned or not in the result's interpretation?
> I have also read something about add a column of "1" as EV in the GLM Setup to demand the regression, what about that?
>
> Thank you so much in advance
|