[R-sig-ME] Extracting a p-value for a random effect in glmer

[Michel, Eric]

Good Afternoon,
I am responding to a reviewers request to provide a p-value for the random effect I used in a glmer model in the lme4 package. I have found plenty of code to produce p-values for lmer models, but not glmer models. Here is my model structure:

Int.litter<-glmer(litter_size~maternal_mass*repro_status + (1|maternal_id), family=binomial, data=litter).

I'd greatly appreciate any help.



[[alternative HTML version deleted]]

[Philippi, Tom]

Eric--
Ben Bolker has kindly covered that in his useful Mixed Models FAQ:
http://bbolker.github.io/mixedmodels-misc/glmmFAQ.html#testing-significance-of-random-effects


Tom

"To do science is to search for repeated patterns, not simply to accumulate
facts..." --Robert MacArthur 1972, Geographical Ecology
"Statistical methods of analysis are intended to aid the interpretation of
data that are subject to appreciable haphazard variability" --Cox &
Hinkley 1974; Theoretical Statistics

-------------------------------------------
Tom Philippi
Quantitative Ecologist & Data Therapist
Inventory and Monitoring Program
National Park Service

Post by Michel, Eric
Good Afternoon,
I am responding to a reviewers request to provide a p-value for the random
effect I used in a glmer model in the lme4 package. I have found plenty of
code to produce p-values for lmer models, but not glmer models. Here is my
Int.litter<-glmer(litter_size~maternal_mass*repro_status +
(1|maternal_id), family=binomial, data=litter).
I'd greatly appreciate any help.
[[alternative HTML version deleted]]
_______________________________________________
https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models

[[alternative HTML version deleted]]

[Phillip Alday]

Not answering your question at all, but a binomial model doesn't seem
like the right choice for litter_size, which I assume is a count
variable and not binomially distributed, just on account of biology ....

Related to that: why do you want a p-value for your random effects? In
other words, what's your actual inferential question? Are you interested
in whether mothers differ? Do you have multiple litters from each mother?

Phillip

Post by Michel, Eric
Good Afternoon,
Int.litter<-glmer(litter_size~maternal_mass*repro_status + (1|maternal_id), family=binomial, data=litter).
I'd greatly appreciate any help.
[[alternative HTML version deleted]]
_______________________________________________
https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models