Intraclass correlation (ICC) and standard error / 95% confidence interval with GLLAMM

[Jason Thompson]

I am interested in calculating a standard error and 95% confidence interval for the intraclass correlation coefficient of a multilevel logit model using the GLLAMM command. I know how to calculate the ICC, but am having difficulty in calculating the SE and 95% CI. I know not to put too much faith in these calculations, due to the skew of the sampling distribution, but I would still like some sort of range with which to gauge difference between ICCs across models. Thanks for the help!