Abstract
In statistical ecology, the number of shared species is a standard measure of similarity between two communities. Assume that a multinomial sample is drawn from each of the two target communities. Each observation (individual) in the sample is classified to species identity, and the frequency for each observed species is recorded. This paper uses the concept of sample coverage to estimate the number of species in common to the two communities (the number of shared species). The result generalizes Chao and Lee (1992) to a two-community situation. Simulation results are reported to examine the performance of the proposed estimators. Bird census data collected from April 1994 to March 1995 in Ke-Yar and Chung-Kang estuaries in Taiwan are used to illustrate the estimation procedure.