Using Point and Intervalized Data in Occupational Employment Statistics Survey Estimates

• Activity Number: 89
• Type: Contributed
• Date/Time: Sunday, August 1, 2010 : 4:00 PM to 5:50 PM
• Sponsor: Section on Survey Research Methods
• Author(s): David Salvatore Piccone*+ and Teresa Elizabeth Hesley
• Companies: Bureau of Labor Statistics and Bureau of Labor Statistics
• Keywords: Establishment survey ; estimation ; Occupational Employment Statistics ; wage point data

Abstract

The Occupational Employment Statistics (OES) survey conducted by the U.S. Bureau of Labor Statistics produces detailed occupational employment and wage estimates. To reduce response burden, the OES form asks respondents to report the number of employees by wage interval. Many of OES's larger respondents, such as the Federal Government, report data electronically, providing wage rates for each of their employees, or "point data", which OES then reclassifies into wage intervals for estimation. A mean wage rate is calculated for each interval from a secondary source and given to all employees within the interval. Capturing and integrating point-data wage distributions into estimation could increase the accuracy of estimates. This paper describes the research and implementation of mean and percentile wage estimators, along with variances estimators, using available point data wage data.