External Email - Use Caution
Dear Statistics Experts
we performed a multivariate analysis of covariance (MANCOVA) between two groups using fractional anisotropy (FA), axial diffusivity (AD), radial diffusivity (RD), and mean diffusivity (MD) of the inferior fronto-occipital fasciculus as dependent variables.
Because the two groups differ also in global (across the whole white matter) FA, AD, RD and MD values, we used all these four measures simultaneously as covariates of no interest.
However, some authors challenge the usage of all four global measures simultaneously as covariates of no interest and denoted this procedure an inappropriate “overcorrection”.
Is this procedure statistically problematic?
Is it more appropriate to first regress out the global effect of each single measure and then run the multivariate analysis just on the residuals performing an MANOVA instead of a MANCOVA model?
Any suggestions is highly appreciated.
Thanks in advance Best regards Jürgen Hänggi
----------------------------------------------------------------------------------------------------- University of Zurich Dr. Jürgen Hänggi, Ph.D. Department of Psychology Division Neuropsychology 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)
j.haenggi@psychologie.uzh.ch (email) http://www.psychologie.uzh.ch/neuropsy/ (website) http://www.juergenhaenggi.ch (private website)
This email (and any attachment/s) contains confidential and/or privileged information. If you are not the intended recipient (or have received this email in error) please notify the sender immediately and destroy this email. Any unauthorised copying, disclosure or distribution of the material in this email is strictly forbidden. -----------------------------------------------------------------------------------------------------