Dear SPM/VBM/Statistic experts
We have done some t-test on VBM data. Because the two groups already differ
significantly in their global GM volumes we included an interaction between
the covariate total grey matter volume (TGMV) and the factor group.
Is this reasonable?
Are there any theoretical or data-driven guidelines that help to decide
whether one should include such an interaction or model TGMV as a "simple
covariate" only (i.e., as only one regressor)?
Whereas the results derived from a design with and a design without such an
interaction were almost identical, the kind of centering dramatically
changed the results (overall mean vs. factor 1 mean). Here againg:
Are there any theoretical or data-driven guidelines that help to decide
whether one should centering at all, and if yes, which centering method is
the most appropriate?
Does anybody made similar observations?
Thanks in advance
Best regards
Juergen
-----------------------------------------------------------
Juergen Haenggi
Ph.D. (Dr. des.)
Division of Neuropsychology
Institute of Psychology
University of Zurich
Binzmuehlestrasse 14, PO Box 25
8050 Zurich, Switzerland
0041 44 635 73 97 (phone office)
0041 76 445 86 84 (phone mobile)
0041 44 635 74 09 (fax office)
BIN 4.D.04 (office room number)
[log in to unmask] (email)
http://www.psychologie.uzh.ch/neuropsy/ (website)
http://www.juergenhaenggi.ch (private website)
-----------------------------------------------------------
------ End of Forwarded Message
|