Initiated by the Committee on Applied and Theoretical Statistics of the National Research Council, the 1997 International Record Linkage Workshop and Exposition was held on March 20–21, 1997 in Arlington, Virginia. More than 200 people were in attendance, and, because the facilities were limited, another 200 interested individuals had to be turned away.
The workshop had two main goals: first, we wanted to celebrate Howard Newcombe's pioneering practical work on computerized record linkage, which began in the 1950s—see, e.g., Newcombe, Kennedy, Axford, and James (1959), “Automatic Linkage of Vital Records,” which appeared in Science—and the theoretical underpinnings of his work, which were formalized in the 1960s by Ivan Fellegi and Alan Sunter in their classic 1969 paper “A Theory for Record Linkage,” published in the Journal of the American Statistical Association. Second, we wanted to broadly update the methodological and technological developments in record linkage research and their applications since the March 1985 Workshop on Exact Matching Methodologies, held in Washington, D.C. The proceedings from that earlier conference, Record Linkage Techniques— 1985, have been widely cited; but much new work has been done since then.
Readers are fortunate to find in the current volume recent papers by two of the pioneers in record linkage research—Fellegi and Newcombe —as well as the work of many others who are exploring various aspects of exact matching techniques. Some of the new areas of related research reflect increased privacy concerns due to record linkage; the growing interest in record linkage as a means for more efficient use of scarce statistical resources; the heightened importance of linkage technology for such policy areas as health care reform; issues related to the physical security of data; and measurement of the risk of nondisclosure and reidentification in public-use microdata files.
The format for Chapters 1 through 10 of this volume essentially follows that of the 1997 workshop agenda —with a section (Chapter 11) added to highlight key contributions made to the literature since the publication of Record Linkage Techniques—1985. In those few cases where a paper was not available, the conference workshop abstract is provided. Chapter 12, a tutorial presented in the form of slides, includes a glossary of terms on pages 477–479. The report concludes with a section (Chapter 13) describing software demonstrated at the workshop. The appendix lists attendees who participated in the March 1997 workshop and accompanying software expositions.
Copy Preparation and Reviews
The contents of the papers included here are the responsibility of the authors. With the exception of selected previously published papers, which were simply reproduced (with permission) as is, all of the papers in this volume underwent only a limited editorial review. Since this did not constitute a formal referee process, authors were also encouraged to obtain their own technical review. Corrections and changes were either made by the authors themselves or cleared through them by the editors. Final layout of the papers was done by the editors, Wendy Alvey and Bettye Jamerson. Minor changes of a cosmetic nature were considered the prerogative of the editors.
A limited independent review was conducted by the National Research Council.
Obtaining Record Linkage Techniques—1985
Unfortunately, those interested in obtaining the companion volume, Record Linkage Techniques—1985, will be hard pressed to find a copy in print. They are few and far between. Instead of reprinting that volume for the March 1997 workshop, the Federal Committee on Statistical Methodology made an electronic copy available (it is a very large pdf file) on its home page, currently being maintained by the Bureau of Transportation Statistics. To access the electronic version, go to <http://www.fedstats.gov>; click on Policy; then click on Statistical Policy Working Papers and go down to Other Multi-Agency Papers.
Other Sources of Related Information
In addition to Record Linkage Techniques—1985, a number of other sources of information may be of interest to record linkage researchers. In particular, the Federal Committee on Statistical Methodology's Statistical Policy Working Paper (SPWP) series includes three especially pertinent reports:
Report on Exact and Statistical Matching Techniques, SPWP No. 5;
Report on Statistical Disclosure and Disclosure Avoidance Techniques, SPWP No. 2; and
Report on Statistical Disclosure Limitation Methodology, SPWP No. 22.
All three of these reports can be accessed electronically at the URL address shown above. A limited number of hard copies may still be available from the Statistical Policy Office, Office of Information and Regulatory Affairs, Office of Management and Budget, Room 10201 NEOB, 725 17th Street, NW, Washington, DC 20503.
Sources of Support
The March 1997 Workshop was made possible with financial and other support from the Bureau of the Census, Ernst and Young LLP, the Federal Committee on Statistical Methodology, the Internal Revenue Service, the National Agricultural Statistics Service, the National Cancer Institute, the National Center for Health Statistics, the National Science Foundation, Statistics Canada, and the Washington Statistical Society.
Fritz Scheuren and William Winkler
The Program Committee would like to thank the Committee on Applied and Theoretical Statistics for undertaking this workshop project. The Program Committee especially acknowledges the support received for this project from the Bureau of the Census, Ernst and Young, LLP, the Federal Committee on Statistical Methodology, the Internal Revenue Service, the National Agricultural Statistics Service, the National Cancer Institute, the National Center for Health Statistics, the National Science Foundation, Statistics Canada, and the Washington Statistical Society.
In addition, the co-chairs would like to thank John Tucker and Ruth O'Brien, of the National Research Council, for their guidance; Carol McDaniel, at Census, for overseeing the arrangements for the workshop; Wendy Alvey, at the Internal Revenue Service, and Bettye Jamerson, of Jamie Productions, for compiling this compendium; and Nancy Kirkendall, at the Office of Management and Budget.
Thanks are also extended to Herman Chernoff, Harvard University; Duncan Luce, University of California at Irvine; Larry Brown, University of Pennsylvania; and a reviewer who wished to remain anonymous, for their review of this volume on behalf of the National Research Council.