Optimization in the Design of a Large~Scale

State Sample, Proceedings of the American Statistical Association, Section on Survey Research Methods, 1980.

Analytic FocusOptimization in the Design of a Large~Scale State Sample

Project details

One of the most important questions concerning the recently conducted 1980 census is “how well did it work, how many people were missed?”.  A large-scale state and substate sample, entitled the Post Enumeration Survey (PES), was planned to answer these questions. This survey would, in part,  produce estimates of undercount (the number of persons missed) for each of the fifty states and the District of Columbia with a specified coefficient of variation, as well as estimates of undercount for thirty-two selected cities, and for the entire metropolitan area in which these cities lay, called Standard Metropolitan Statistical Areas (SMAS’s).  The thirty-two selected cities were chosen based on two criteria.  First, a size criteria was chosen; all those cities whose population exceeded five hundred thousand in either 1970 or 1976 were included.  Second, a minority population criteria was used; all those cities whose 1970 census population exceed two hundred fifty thousand and whose 1970 minority population exceeded forty percent of the total population for that city were included.

-Schwartz, Sidney H., Cowan, Charles D., and Sausman, Kenneth R.

Back to Publications

Related Projects