U.S. flag

An official website of the United States government

Skip Header


Estimation of the Percent of Unique Population Elements on a Microdata File using the Sample

Written by:
RR91-08

Abstract

National statistical agencies publicly release information about a nation's population that has been collected under a pledge of confidentiality. A population element which has a unique combination of characteristics and is represented in a sample of microdata where those characteristics appear as categorical variables is at risk of disclosure. An intruder could match the element's unique combination of variables on the microdata to the same combination of variables on some other data base containing identifiers and thus link the element to its microdata record. The percent of unique population elements on a microdata file can be regarded as one component of a measure of disclosure risk. In this paper, two methods of estimating the percent of unique population elements on a sample microdata file using information from that sample are presented and evaluated. A third method of estimation was discussed by Willenborg, Mokken, and Pannekoek (1990) and is reviewed here.

Related Information


Page Last Revised - October 28, 2021
Is this page helpful?
Thumbs Up Image Yes Thumbs Down Image No
NO THANKS
255 characters maximum 255 characters maximum reached
Thank you for your feedback.
Comments or suggestions?

Top

Back to Header