Abstract
A number of applications, including claims made under Federal social welfare programs, require retrospective sampling over multiple time periods. A common characteristic of such samples is that population members could appear in multiple time periods. When this occurs, and when the marginal cost of obtaining multi-period information is minimum for a member appearing in the sample of the period being actively sampled, the progressive random sampling method developed by the authors earlier [2] can be applied. This paper enhances the progressive random sampling method by combining it with stratification schemes; the resultant Stratified Progressive Random Sampling (SPRS) technique is shown to provide significant improvement over traditional sampling techniques whenever stratification is appropriate.