Bias reduction for multiple-recapture estimators of closed population size.

A bias correction for multiple-recapture sampling is proposed. The correction is similar to Chapman's (1951, University of California Publication in Statistics 1, 131-160) correction of the Lincoln-Petersen estimator of population size. Small-sample properties of the modified population size estimator are examined using Monte Carlo simulation. Substantial reductions in estimator bias and variance are obtained with the modified estimator.