[R-sig-ME] z transform versus random intercept

Daniel Ezra Johnson danielezrajohnson at gmail.com
Thu Dec 3 17:09:49 CET 2009


Is another, perhaps less important, difference that the z-scaling
approach does "handle" (or remove) heteroscedasticity between subjects
- at least in some fashion - whereas an lmer() model with random
intercept would assume that each subject's observations have the same
variance?

DEJ




More information about the R-sig-mixed-models mailing list