Inverse Sampling

Inverse sampling is an adaptive sampling technique credited to J. B. S. Haldane's work in the 1940s. Under many study designs, it is desirable to estimate the frequencies of an attribute in a series of populations, each [Page 394]of which is much larger than the sample taken from it so that the population size is assumed to be infinite. However, the probability of the attribute occurring in some of the populations may be so small that under a fixed sample size design, not enough cases of interest are selected to estimate the attribute of interest.

Inverse sampling draws from the negative binomial distribution in that a series of Bernoulli trials are conducted until a predefined r number of successful cases occur. Usually, r is the desired ...