I'm running a logistic regression model in the form: lmer(response~1+(1|site),family=binomial,REML = FALSE)

Normally I would calculate the ICC from the intercept and residual variances, but the summary of the model doesn't include residual variance. How do I calculate this?