Initial Efforts to Extend the EZS Noise Methodology to the Quarterly Census of Employment and Wages

• Activity Number: 455
• Type: Topic Contributed
• Date/Time: Wednesday, August 3, 2011 : 10:30 AM to 12:20 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Michael Buso*+
• Companies: Bureau of Labor Statistics
• Keywords: tabular data ; disclosure ; noise ; random noise

Abstract

The Bureau of Labor Statistics (BLS) Quarterly Census of Employment and Wages (QCEW) program has been using primary and complementary cell suppression techniques as its disclosure limitation methodology. The employment and wage levels of over 9 million establishments are collected under confidentiality pledges and other provisions that obligate BLS to protect the identities and characteristics of over 9 million establishments whose data are assembled into a multi-dimensional array of over 3.5 million populated data cells. The techniques have been providing an unsatisfactory balance between the protection provided to respondents and the usefulness of the data for economic and policy analysis uses, as over 2.1 million cells are suppressed. BLS is seeking to extend the basic random noise methods of Evans, Zayatz, and Slanta (EZS), to the characteristics of the QCEW program. BLS also is considering whether several of the protective mechanisms of the original EZS paper can be relaxed while still providing reasonable protection to respondents.