Phase‐in of the Redesigned National Compensation Survey Area Sample

Jason Tehonica, Lawrence R. Ernst, and Chester H. Ponikowski


The sample design of the National Compensation Survey (NCS), conducted by the Bureau of Labor Statistics, is an area-based design. A sample redesign of the NCS is necessary each decade to conform with new area definitions set by OMB. Key products of the NCS include the Employment Cost Index (ECI), national benefit incidence and provision estimates, locality wage surveys, and area-based salary data provided to the President's Pay Agent. The NCS must adopt the new area definitions and transition to a new area sample as quickly as possible while maintaining its key products and workload levels. The transition is complex and lasts several years because the NCS locality products are generated using a rotating panel design with approximately 1/5 of the establishment sample replaced annually. In the transition, locality products will be generated from a mix of old and new area designs. This paper discusses the phase-in of the new area sample and issues associated with the transition, including potential bias issues associated with implementing a new area sample, impact on variance estimates, additional weighting complexities, and impacts on national wage estimates and the ECI.