Research Reports

Recent Work on the Microdata Analysis System at the Census Bureau

Jason Lucero, Lisa Singh, and Laura Zayatz

KEY WORDS: Disclosure Avoidance, Confidentiality, Data Dissemination, Remote Access, Differencing Attack, Sub-sampling

ABSTRACT

The U.S. Census Bureau collects its survey and census data under Title 13 of the U.S. Code which promises confidentiality to its respondents. The agency also has the responsibility of releasing data for the purpose of statistical analysis. In common with most national statistical institutes, our goal is to release as much high quality information as possible without violating the pledge of confidentiality. This paper discusses a Microdata Analysis System (MAS) that is under development at the Census Bureau. The system is designed to allow data users to perform various statistical analyses (for example, regressions, table generations, generation of correlation coefficients) of survey and census data without seeing or downloading the actual underlying confidential microdata. This paper begins with an overview of the Microdata Analysis System and discusses the confidentiality rules currently implemented in the system. The remainder of the paper will focus on a recently evaluated Drop q Rule and a cutpoint generation program. These are used to protect the confidentiality of results generated from the system while still maintaining data quality and utility. We will then conclude with a brief discussion about future work on the MAS.

CITATION: Lucero, Jason, Lisa Singh, and Laura Zayatz. (2009). Recent Work on the Microdata Analysis System at the Census Bureau. Statistical Research Division Research Report Series (Statistics #2009-09). U.S. Census Bureau. Available online at <http://www.census.gov/srd/papers/pdf/rss2009-09.pdf>.

Source: U.S. Census Bureau, Statistical Research Division

Published online: November 16, 2009
Last revised: November 2, 2009