Cyril,
> Thank you for the reply Tom,
>
> -- 'but what is Conjunction Null?' --
Regarding 'Conjunction Null' vs 'Global Null' see the recent SPM list
emails about it (under the heading 'Negative T Values) and the paper
on this web page
http://www.sph.umich.edu/~nichols/Conj
But a more important question is: Don't you get asked 'Conj Null'
'Global Null' as part of results? If not, you don't have the
current spm2_updates.
> >So I belive that the explanation is that nonsphericity has 'colored'
> >your ANOVA design, and that makes the design space of your contrasts
> >non-orthogonal, forcing the creation of new a contrast when you request
> >Global Null conjunction inference.
> >
> >One 'solution' (if this is a problem), is to use the Conjuction Null,
> >which doesn't require independence of test statistics.
> >
> I think you're right .. looking at the result (1>4) and (2>5) should
> give me the non corrected conjunction whereas looking at the new
> contrast should correspond to a conjunction corrected for non
> independant contrasts ? ... Thus this is not a real bug .. however, and
> this message is valid for everyone who use a conjunction in a repeated
> measure ANOVA, the first result displayed by SPM is false.
Well, again, I don't know what you mean by 'false'. I won't say it's
wrong (its a set of orthogonalized contrasts, where lack of orthogonality
is induced by whitening to account for nonsphericity), it's just perhaps
not what you expected. Does this make sense?
> Do you think I can use the result of the corrected conjunction? or
> should I use Conjunction Null ...
> but I don't know what it is :'(
Conjunction null, definately.
Hope all this helps.
-Tom
-- Thomas Nichols -------------------- Department of Biostatistics
http://www.sph.umich.edu/~nichols University of Michigan
[log in to unmask] 1420 Washington Heights
-------------------------------------- Ann Arbor, MI 48109-2029
|