# anova type III test for a GLMM

I am fitting a glmer model in the lme4 R package. I'm looking for an anova table with p-value shown therein, but I cannot find any package that fits it. Is it possible to do it in R?

The model I am fitting is of the form:

model1<-glmer(dmn~period*teethTreated+(1|fullName),
family="poisson",
data=subset(dataset,
group=='Four times a year'),
control=glmerControl(optimizer="bobyqa"))


If you're willing to settle for Wald tests this should work:

library(lme4)
library(car)
gm1 <- glmer(cbind(incidence, size - incidence) ~ period + (1 | herd),
data = cbpp, family = binomial)
Anova(gm1,type="III")


However, note (from ?Anova) that:

The designations "type-II" and "type-III" are borrowed from SAS, but the definitions used here do not correspond precisely to those employed by SAS. Type-II tests are calculated according to the principle of marginality, testing each term after all others, except ignoring the term's higher-order relatives; so-called type-III tests violate marginality, testing each term in the model after all of the others. This definition of Type-II tests corresponds to the tests produced by SAS for analysis-of-variance models, where all of the predictors are factors, but not more generally (i.e., when there are quantitative predictors). Be very careful in formulating the model for type-III tests, or the hypotheses tested will not make sense.

I would check your results very carefully to make sure they make sense!

Alternatively, you can use afex::mixed to get analogous tables via likelihood ratio test or parametric bootstrap; the latter is the most accurate, but also the slowest by far.

See ?pvalues in the lme4 package for more general discussion of p-value computation in the context of GLMMs.