A simple and versatile capture-recapture method for "one-sample" data, with applications

Capture-recapture techniques are widely used to estimate the size of hard to count human populations. Applications typically focus on the overlap between two or more samples, but some studies involve "one-sample" data on the number of times each of a set of individuals is captured or encountered. After briefly reviewing general capture-recapture methods, we will present a simple and technically accessible method for estimating population size with sparse "one-sample" capture-recapture data. The method allows for entries into and exits from the population and for a difference between probabilities of initial and subsequent contacts.

Our substantive motivation is to estimate the prevalence of male clients of prostitute women. We will discuss the importance of such research and report results from applying our method to data from 13 metropolitan communities in North America. We will conclude with an overview of other areas of application and possible extensions of the method.