|
Activity Number:
|
285
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, August 5, 2003 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Government Statistics
|
| Abstract - #301842 |
|
Title:
|
The Feasibility of Aggregating Small Blocks for Census 2010
|
|
Author(s):
|
Philip M. Steel*+
|
|
Companies:
|
U.S. Census Bureau
|
|
Address:
|
5457 Nebraska Ave., NW, Washington, DC, 20015-1349,
|
|
Keywords:
|
confidentiality ; census geography ; aggregation
|
|
Abstract:
|
Since the basic Census geographic unit is not directly tied to population density, a "block" may contain anywhere from one household to several hundred. The tabulation of short-form census items pose little or no disclosure risk for blocks with many households. On the other hand, the tabulation of a block with a single household could reveal several particulars of its census response and a geographic identification that is as specific as an address. The Census Bureau must apply disclosure avoidance techniques to hide the responses of individual households and persons in its publications. These techniques force a trade-off between effectiveness of protection and data quality. Aggregation is the natural solution to this kind of problem. We hope to answer some of the following questions: How much aggregation would be needed to affect that trade-off? What algorithms are available and what are their strengths and weakness? What effect would aggregation have on the flexibility of data products? What software is available to implement an aggregation? What features in Census 2010 processing would facilitate or interfere with implementing an aggregation scheme?
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2003 program |
