[R] help with glmmPQL

[Douglas Bates]

John Fox wrote:
> Dear Doug,
> 
> I assume that in the absence of a suitable anova() method, there's nothing
> wrong with testing the difference between two GLMM objects nested in their
> random effects by manually comparing the log-likelihoods as printed or
> returned by logLik(). Is that correct?
> 
> Regards,
>  John
> 

Yes.

I should have thought of that myself.  At this time of year those in 
Canada where they celebrate Thanksgiving in October are obviously 
thinking more clearly than those in the United States whose minds are 
fogged by too much turkey and televised football games.