Amir -
I don't know why you have been told it is not possible to do
repeated-measures ANOVAs in SPM. It is.
One way is to perform a separate SPM model for each ANOVA effect, where
the data are contrasts for that effect (see technical note by Henson &
Penny, 2003; or chapter in SPM book; or numerous emails on the SPM
helplist). This emulates the conventional way of partitioning the error
for each ANOVA effect.
Alternatively you can enter all conditions into one SPM model and test
appropriate contrasts there. This is the pooled error approach (though
some people have questioned this approach on the helplist).
In both cases however, for any repeated measure factor with >2 levels,
you should put the "independence" option to "no".
Rik
Amir H Javadi wrote:
> Hello :-)
>
> My question is about running a repeated measure of ANOVA?
> As I've been told it is not possible using SPM. But I suspect if it is possible using the "independent" option in the 2nd level analysis.
> Let me explain it more. I have 2 x 3 x 3 design for condition x time x
> frequency and it is a within subject analysis. I wonder if I can run a
> repeated measure of ANOVA by setting the "independence" option to "no"
> for the "condition" factor. Should I also set "independence" option to
> "no" for the other two factors, "time" and "frequency"? Very many thanks.
>
> Have a good time :-)
> Amir
>
--
-------------------------------------------------------
Dr Richard Henson
MRC Cognition & Brain Sciences Unit
15 Chaucer Road
Cambridge
CB2 7EF, UK
Office: +44 (0)1223 355 294 x522
Mob: +44 (0)794 1377 345
Fax: +44 (0)1223 359 062
http://www.mrc-cbu.cam.ac.uk/people/rik.henson/personal
-------------------------------------------------------
