Please use this identifier to cite or link to this item: http://hdl.handle.net/10419/18340
Authors: 
Jenkins, Stephen P.
Lynn, Peter
Jäckle, Annette
Sala, Emanuela
Year of Publication: 
2005
Series/Report no.: 
DIW Discussion Papers 489
Abstract: 
Linkages of household survey responses with administrative data may be based on unique individual identifiers or on survey respondent characteristics. The benefits gained from using unique identifiers need to be assessed in the light of potential problems such as non-response and measurement error. We report on a study that linked survey responses to UK government agency records on benefits and tax credits in five different ways. One matched on a respondent- supplied National Insurance Number and the other four used different combinations of sex, name, address, and date of birth. As many linkages were made using matches on sex, date of birth, and post-code, or on sex, date of birth, first name and family name, as were made using matches on self-reported National Insurance Number, and the former were also relatively accurate when assessed in terms of false positive and false negative rates. The five independent matching exercises also shed light on the potential returns from hierarchical and pooled matching.
Subjects: 
record linkage
matching
National Insurance number
measurement error
Document Type: 
Working Paper

Files in This Item:
File
Size
325.76 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.