Abstract Details
Activity Number:
|
339
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, August 5, 2014 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Government Statistics Section
|
Abstract #311272
|
View Presentation
|
Title:
|
Using Record Linkage to Create Big Data? How Good Is It?
|
Author(s):
|
K. Bradley Paxton*+
|
Companies:
|
ADI
|
Keywords:
|
Record Linkage ;
Data Quality ;
Census ;
Testing
|
Abstract:
|
Record Linkage is being increasingly used to create new "big data" collections or improve already existing data. If, for example, a Census data set is correctly linked to a Tax data set, a new or improved data set may result; if, however, the linkage is done incorrectly, then the data set may be made worse. It is important to know how good your record linkage system is performing to optimize it and get the best results for your investment. Record linkage systems, however, are difficult to test.
This paper presents two proven methods for testing record linkage systems: a synthetic data GAMUT (Great Automated Model Universe for Test) for use in the development phase and Production Data Quality (PDQ) in the production phase. These methods were used successfully in the 2010 Census for evaluating forms processing but they are extensible to record linkage and also apply to any IT classification system.
Both methods are cost-effective because they use automation to replace a lot of human effort. Together, the two methods provide "cradle-to-grave" testing and improvement capability. Future work includes trading-off precision and recall to achieve maximum record linkage system value.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2014 program
|
2014 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Professional Development program, please contact the Education Department.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Copyright © American Statistical Association.