Appendix to Disclosure Limitation in Longitudinal Linked Data
John M. Abowd
Cornell University, U.S. Census Bureau,
CREST, and NBER
Simon D. Woodcock
August 20, 2001
Extracted from Disclosure Limitation in Longitudinal Linked Data, in ConÞdentiality, Dis-
closure and Data Access: Theory and Practical Applications for Statistical Agencies, P. Doyle,
J. Lane, L. Zayatz, and J. Theeuwes (eds.), (Amsterdam: North Holland, 2001), forthcoming.
Reproduced with permission.
The research reported in this paper was partially sponsored by the U.S. Census Bureau, the
National Science Foundation (SES-9978093), and the French Institut National de la Statistique et
des Etudes Economiques (INSEE) in association with the Cornell Restricted Access Data Center.
The views expressed in the paper are those of the authors and not of any of the sponsoring
agencies. The data used in this paper are conÞdential but the authors access is not exclusive. No
public use data sets were released as a part of this research. Restricted access to the French data
was provided to Abowd by INSEE through an agreement with Cornell University. The authors
thank Benoit Dostie, Sam Hawala, Janet Heslop, Paul Massell, Carol Murphree, Philip Steel,
Lars Vilhuber, Marty Wells, Bill Winkler, and Laura Zayatz for helpful comments on earlier
versions of this research.
1 Appendix: Recent Research on Disclosure Limitation
In recent years, statistical agencies have seen an increasing demand for the data they collect, coupled with
increasing concerns about conÞdentiality. This presents new challenges for statistical agencies, who must
balance these concerns. Doing so requires techniques to allow dissemination of data that is both analytically
useful and preserves the conÞdentiality of respondents.
This appendix presents recent research on disclosure limitation methods and concepts. Although our
primary interest is in methods appropriate to longitudinal linked data, this topic has not been well-addressed
in the literature. Hence, w