Using the two column model specification option for count and binomial models does not work with averageObs :
gm1 <- glmer(cbind(incidence, size - incidence) ~ period + (1 | herd),
data = cbpp, family = binomial)
averageObs(gm1)
This is because the last line of the function requires parsing the names of the model frame, but thesew ill not line up with the original data because of the formula defining the Y variable.