[R-sig-ME] Variance explained by random factor

[Renwick, A. R.]

I am currently trying to run a lmer model with poisson distrubution.  I tested the model with a model without the random effect and it inferred that I should include the random effect:

ma1<-glm(RoundedOverlap~sess+breedfem,family=poisson,data=Male)
mixed<-lmer(RoundedOverlap~sess+breedfem+sess:breedfem+(1|Site),family=poisson,data=Male)

#test to see if sig difference between glm and glmm
as.numeric(2*(logLik(mixed)-logLik(ma)))
#99.16136
pchisq(99.16136,1,lower=FALSE)
#2.327441e-23  so should use a GLMM

However,the model output that I get states that the variance explained by the random factor is 0:


Generalized linear mixed model fit by the Laplace approximation
Formula: RoundedOverlap ~ sess + breedfem + sess:breedfem + (1 | Site)
   Data: Male
   AIC   BIC logLik deviance
 109.9 127.2 -45.93    91.86
Random effects:
 Groups Name        Variance Std.Dev.
 Site   (Intercept)  0        0
Number of obs: 51, groups: Site, 14

I would really appreciate if somebody could help me understand why the variance is 0.
Many thanks,
Anna




Anna Renwick
Zoology Building
School of Biological Sciences
University of Aberdeen
Aberdeen
AB24 2TZ


The University of Aberdeen is a charity registered in Scotland, No SC013683.