www.cmm.bristol.ac.uk/forum

Hi everyone, as far as i know, it is impossible to do lr-test, so how can i try out a logistic two level-model that is prefered to logistic with runmlwin? Thanks a lot for your help.

The recommendation for comparing these types of models is usually to use a Wald test, or if you are estimating via MCMC to compare the DIC statistics.