I have a data set on animal species diversity at 40 study sites from3 sub areas. The data comprise about 60 environmental variables. I am interested in the effect of each variable on the species diversity. I computed univariable GLMMs (negative binomial) of the form Species_Diversity ~ scale(Environmental_Variable_Xi) + (1|Sub_Area) in R using the lme4-package. It works fine, but I was a little bit confused because all models have very similar estimate standard errors of roughly 0.053.

The environmental variables contain very different parameters such as temperature, precipitation, vegetation cover, elevation, ... and most are not (strongly) intercorrelated. So I wonder if someone can explain where these very similar estimate errors stem from. On a side note: the estimates themselves vary widely. It almos seems like the estimate errors are somehow "fixed" - perhaps by the number of study sites in my models?



Browse other questions tagged or ask your own question.