Utilizing Occupational Employment Statistics (OES) Survey to Improve Small Domain Estimation (SDE) in the Occupational Requirements Survey (ORS)

• Activity Number: 352 - Small Area Estimation, Analysis of Complex Sample Survey Data, and New Advances for Health Surveys
• Type: Contributed
• Date/Time: Thursday, August 12, 2021 : 10:00 AM to 11:50 AM
• Sponsor: Survey Research Methods Section
• Author(s): Xingyou Zhang* and Erin McNulty and Ellen Galantucci and Patrick Kim and Joan Coleman and Tom Kelly
• Companies: Bureau Of Labor Statistics and Bureau of Labor Statistics and Bureau of Labor Statistics and Bureau of Labor Statistics and Bureau of Labor Statistics and Bureau of Labor Statistics
• Keywords: small domain estimation; multilevel statistical models; Standard Occupational Classification (SOC); Occupational Requirements Survey (ORS); Occupational Employment Statistics (OES) survey; establishment survey

Abstract

The Occupational Requirements Survey (ORS) is an establishment-based survey and provides job-related information about the physical demands; environmental conditions; education, training, and experience; as well as cognitive and mental requirements in the U.S. economy. However, more than 50% of estimates for 834 detailed occupations (6-digit 2018 Standard Occupational Classification (SOC) system) under the ORS sampling frame are not publishable because of lack of sample or small sample sizes. In order to improve ORS estimates, we explore and develop a multilevel Small Domain Estimation (SDE) approach that utilizes the Occupational Employment Statistics (OES) survey, a semiannual survey designed to produce estimates of employment and wages for specific occupations, with an annual sample size of nearly 400,000. This approach includes three basic steps: first, multilevel statistical models are constructed with ORS data; second, the fitted multilevel models are applied to OES survey data and make predictions on the ORS outcomes; third, OES survey data are used to produce ORS outcome estimates for detailed occupations. The strengths and limits of this SDE approach will be discussed.