On Tue, 22 Apr 2008 20:07:31 +0100, Brian Essex <[log in to unmask]>
wrote:
>I have done an f contrast at the individual subject level and now
>would like to look at the effect of this contrast across subjects (2nd
>level/random effects). What is the correct way to do this? I cannot find
an
>F test in the 2nd level toolbox.
>
>Basically, what statistical test can I use to implement this and how can I
>then define my contrast in the contrast manager?
You don't take F test images to the second level (group level). You only
take con images.
How easy it is to do F tests (or ANOVAs) at the group level depends on
whether you're taking more than one con image per subject to the group
level. If you do take more than one, it's a repeated measures design, and
while it can be done, it's a lot more complicated than if there's just one
image per subject.
Best thing to do is search this list for advice, and also consult the
manuscript "ANOVAs and SPM," by Henson and Penny:
http://www.fil.ion.ucl.ac.uk/~wpenny/publications/rik_anova.pdf
>Thanks,
>Brian
>========================================================================
|