From: Annika on
Hi!

I'm quite at the beginning using GLIMMIX and I have some problems
selecting the best model.
As I read using the -2logL or the AIC is not adaequate because that
are only "pseudo" values.
But what is the adaequte tool to select the best model?

Some say that if it's possible to use MIXED one should use that
instead of GLIMMIX. But how do I know if I can use it? I've got two
levels (like classrooms and pupils) and a binomial distribution of my
dependent variable.

Can anyone help?
Annika