The Reliability of the Locality Pay Estimates

Joan Coleman and Penny James


In support of the Federal Employees Pay Comparability Act of 1990, the Bureau of Labor Statistics was required to develop a program of white collar occupational wage surveys in private industry and state and local governments. To accomplish this goal, the Area Wage Survey and White Collar Pay (WCP) survey programs were integrated. This new program produced wage estimates for thirty-two selected metropolitan areas and estimates for national products. The national estimates were based on these thirty-two areas and other areas which were selected by stratified random sampling with probability proportionate to size. To determine the reliability, relative standard errors (RSEs) for the wage estimates were computed for the entire program and compared to RSEs of similar past surveys. The RSEs for the thirty-two metropolitan areas were compared to RSEs of surveys from the Area Wage program; while the RSEs for the national products were compared to RSEs from the WCP survey. Based upon the comparison of RSEs, the new combined program was determined to be as reliable as the previous programs.