[R-sig-ME] p-value for glmer with quasipoisson family

[CL Pressland]

Please check the archive - there is a lot written on here explaining why 
lme4 does not give p values, unless you're using a MCMC approach which is 
better able to estimate the parameters.

In particular see: 
https://stat.ethz.ch/pipermail/r-sig-mixed-models/2009q3/002833.html

--On 29 September 2009 15:09 +0200 Raldo Kruger <raldo.kruger at gmail.com> 
wrote:

> Hi all,
>
> How come there is no p-value displayed in the summary of the results
> of glmer when using the quasipoisson family? Is there a way of
> extracting the p-values from the results?
>
> Thanks,
> Raldo
>
>> summary(ex5m_r)
> Generalized linear mixed model fit by the Laplace approximation
> Formula: Plants ~ Treat + Year + (1 | Site)
>    Data: ex5m
>   AIC  BIC logLik deviance
>  4685 4719  -2336     4671
> Random effects:
>  Groups   Name        Variance Std.Dev.
>  Site     (Intercept)  3.550   1.8841
>  Residual             12.957   3.5996
> Number of obs: 928, groups: Site, 8
>
> Fixed effects:
>             Estimate Std. Error t value
> (Intercept)  1.02627    0.68130   1.506
> TreatGr     -0.38019    0.17078  -2.226
> TreatK      -0.84274    0.19838  -4.248
> TreatN       0.04644    0.15215   0.305
> Yeartwo      0.66848    0.12931   5.170
>
> Correlation of Fixed Effects:
>         (Intr) TretGr TreatK TreatN
> TreatGr -0.102
> TreatK  -0.088  0.350
> TreatN  -0.114  0.456  0.392
> Yeartwo -0.125  0.000  0.000  0.000
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models