Estimating the risk of extinction or decline requires estimates of the variability in vital rates, such as survival and fecundity. This paper describes a method for estimating variance of survivals and fecundities from census data. The method involves calculating an estimate of the variance in survival and fecundity due to demographic stochasticity and subtracting this estimate from an estimate of total variance. The method is demonstrated by applying it to the data set on acorn woodpeckers by Stacey & Taper (1992), and tested in a series of simulations. The results show that the method can estimate the environmental variance with no bias. The main advantages of this method are its simplicity, its independence of the underlying distribution, and its applicability to estimating the variance of fecundity as well as of survival.