Variance Estimation for the Occupational Requirements Survey

Bradley D. Rhein and Chester H. Ponikowski

Abstract

The Occupational Requirements Survey (ORS) is an establishment survey conducted by the Bureau of Labor Statistics (BLS) for the Social Security Administration (SSA). The survey collects information on the vocational preparation and the cognitive and physical requirements of occupations in the U.S. economy, as well as the environmental conditions in which those occupations are performed. Due to a desire to maximize the number of units collected in the survey, the 2016 sample consists of two independent, multi-stage samples, each with its own set of variance strata. The calculation of the estimated variances for means, percentiles, and percentages involves a modified version of the Balanced Repeated Replication (BRR) method called Fay’s BRR method. The paper describes the formation of the replicate half-samples based on the assignment of the variance strata, reflecting the sampling variability of the two complex samples. Finally, the paper addresses the easy adaptation of the variance procedure when ORS transitions to the use of a single complex sample.