Abstract
In this article, the authors present a statistically efficient, cost-effective way of collecting a probability sample in the presence of certain field burden constraints: restrictions on the maximum number of participants that can be sampled within a given institution and a small population size relative to the sample size dictated by the study aims. The authors suggest the use of disproportionate stratified random sampling as an alternative to two-stage sampling under these circumstances and illustrate how to account, via weighting, for the participants’ differential probabilities of inclusion. They describe their approach with respect to a study of impoverished women, for which this sampling scheme was quite effective.
Get full access to this article
View all access options for this article.