We introduce a method for estimating overdispersion in Poisson models
for vital rates, We assume smoothness conditions on the counts to obta
in pointwise variance estimates that we combine to obtain an estimate
of the overdispersion parameter. We create confidence intervals about
the observed rates using this estimate and an approximation based on t
he gamma distribution. The advantage of this method is that the estima
tes of the superpopulation rates do not depend on the smoothness assum
ption, yet when this assumption is met we obtain approximately unbiase
d estimates of the overdispersion parameter. Thus, we may calculate co
nfidence intervals for vital rates under an overdispersed Poisson mode
l without making parametric assumptions on the mean rates. (C) 1997 by
John Wiley & Sons, Ltd.