Redesign of the National Compensation Survey

Yoel Izsak, Lawrence R. Ernst, Chester H. Ponikowski, Steven P. Paben, and Jason Tehonica


Among the key products produced by the National Compensation Survey (NCS), which is conducted by the Bureau of Labor Statistics, are locality wage surveys. Consequently, a sample redesign of the NCS is necessary for each decade following the release of new metropolitan area definitions. NCS uses a rotating panel design, with three stages of selection, namely: geographic areas; establishments, and occupations. We discuss some of the sample redesign issues that we are studying, including impact on between area variances of (1) single outside Core Based Statistical Areas (CBSAs) counties versus multi-county clusters outside of CBSAs, (2) micropolitan areas included in strata with metropolitan areas, or with strata of counties outside of CBSAs, or separately, and (3) alternative stratification variables. Additional issues are: number of sample establishments, number of sample areas, and allocation of sample areas among certainty metropolitan areas, noncertainty metropolitan areas, and the remaining areas; use of an overlap maximization procedure to select sample areas; sampling and variance estimation during the transition. Preliminary results are given.