[R-sig-ME] glmer and poisson distributions and P values

[Emma Stone]

Dear all,

I am trying to run a glmer model with count data,with one random factor. 
When I run it with more than one predictor all is well, but when there is 
only one predictor in the model I get this warning message, can anyone 
explain this at all?

Warning message: In mer_finalize(ans) : false convergence (8)

Also, for other models in the canidate set I am looking at interactions 
between predictors and I want to get the P values for these interactions, 
any ideas how I can do this from a glmer model?

Thanks

Emma


--