[R-sig-ME] Index-terms confusion

Petar Milin pmilin at ff.uns.ac.rs
Thu Dec 23 14:42:14 CET 2010


Hello all!
I noticed that many users, here and otherwise, use term "hierarchical 
models" to refer to models that are nested. However, in other places, 
one can find "hierarchical models" to refer to the "multilevel models". 
There, one determine direct/indirect relations of some predictors with 
the criterion, aiming in particular to understand mediator effects 
(significance of changes in betas of one variable when another intervene 
etc.). Recently, I realized that some of my students are quite confused 
with this terminology. How would you explain commonalities and 
differences between the two? I know about nlme plus multilevel packages 
for multilevel regression 
(http://cran.r-project.org/doc/contrib/Bliese_Multilevel.pdf). Any other 
recommendation? How about statistics that should inform about changes in 
betas? Some advices for practitioners?

Thanks,
Petar




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