Shail Butani, Mark Crankshaw, Darrell Greene and Vinod Kapani "Comparing Injury Data from Administrative and Survey Sources: Methodological Issues"

The Bureau of Labor Statistics' (BLS) Job Opening and Labor Turnover Survey (JOLTS) is an establishment survey that seeks to measure dynamic trends in the US labor market. Each month the survey publishes the number of job openings, hires, and separations at the national, regional, and major super-sector industry level of detail. The Survey also publishes establishment size class estimates as an experimental series. After consultation with some of the key data users, it was found JOLTS users prefer firm size (proxy for size of employer) estimates rather than establishment size estimates. The purpose of this paper is to provide details of the methodology used to develop JOLTS firm size estimates. The process begins with the assignment of firm size to the frame and sample; development of birth/death models and independent population controls for employment at the industry by firm size level; and application of alignment and seasonal adjustments at the industry by firm size level. Finally, a comparison is made of the sum of the firm size class estimates for each industry to the published industry estimates which are independently produced.