Skip to content
Prev 1675 / 20628 Next

Likelihood Ratio tests and fixed effects with LMER

On Tue, 16 Dec 2008, David Grimardias wrote:

            
Yes, though there has been some discussion on this list about the adequacy of
the chi-square approximation for the LRTS in these models.

You should also compare the results to those from simple models ignoring
the fact that females may appear more than once.  They should be broadly
similar.  How large is the variance due to the random effects anyway?
REML is not done for GLMMs IIRC.

David Duffy.