Online Program Home
My Program

Abstract Details

Activity Number: 122
Type: Contributed
Date/Time: Monday, August 1, 2016 : 8:30 AM to 10:20 AM
Sponsor: Government Statistics Section
Abstract #319496
Title: Using Scores to Identify Small Cells in Tables for Disclosure Deidentification
Author(s): Stephen Cohen* and Linette Scott and Joshua Borton
Companies: NORC at the University of Chicago and California Department of Health Care Services and NORC at the University of Chicago
Keywords: Confidentiality ; De-identification ; Cell suppression ; Scores

The usual way to protect intruders from identifying an individual's data in a table is to suppress all small cells (primary suppression) and secondary (complementary) cells that must be suppressed to protect the primary suppression. An organization typically uses fixed suppression cut offs for all cells, typical between 3 and 10 or higher, for a particular data product. While this technique is sufficient to de-identify a vulnerable cell, often it might result is over suppression. In this paper we explore the use of scores to rate a cell for potential de-identification that are currently in use in some State health departments. The scores we will illustrate in this paper are based on demographic characteristics such as gender, race, time- reporting period (e.g. weekly, monthly, quarterly, etc.), residence geography, etc. We will illustrate how applying the scores to different tables can identify cells with disclosure risk and result in different outcomes for the very same small cell when put into context of other available data. The rationale behind the scores will also be discussed.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2016 program

Copyright © American Statistical Association