Dear Luca,
At 21:53 04/05/99 +0200, Luca Finelli wrote:
| I found the following old mail in the SPM mailing list. There Dr. Andrew
| Holmes described a method ('interaction-of-contrasts' analysis) for
| doing special between-group interactions. I applied this method to my
| data and I would be happy if I could cite a reference addressing this
| problem.
The technique described in email:
http://www.mailbase.ac.uk/lists/spm/1998-02/0043.html
applies only for the case of one scan per subject per condition, and makes
the (strong) assumption of spherical variance. With only two conditions pre
group in the model (i.e. only the conditions to be contrasted ), giving
only two scans per subject, variance sphericity is assured.
This design is a standard split-plot design, with subjects as "plots", but
assumming no within-plot correlation of residuals (variance sphericity).
See (for example) BJWiner, DRBrown & KMMichels (1991) Statistical
Principles in Experimental Design (3rd Ed.), McGraw Hill, New York.
With only two conditions per group, the test is essentially a two-sample
t-test on the within-subject between condition differences - sort of a
two-sample paired t-test!
Hope this helps,
-andrew
+- Dr Andrew Holmes ------------------ mailto:[log in to unmask] -+
| ___ __ ___ Robertson Centre for Biostatistics |
| ( ,) / _)( ,) Boyd Orr Building, |
| ) \( (_ ) ,\ University Avenue, |
| (_)\_)\__)(___/ Glasgow. G12 8QQ Scotland, UK. |
+----------------------------------------- http://www.rcb.gla.ac.uk/ -+
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|