New methods of census record linking

Ronald A Goeken, Lap Huynh, T. A. Lynch, Rebecca J Vick

Research output: Contribution to journalArticle

33 Scopus citations

Abstract

The Minnesota Population Center (MPC) has released linked data sets through its North Atlantic Population Project and Integrated Public Use Microdata System, making them readily accessible to researchers. Before the availability of complete-count census microdata from the MPC, researchers applied various forms of record-linking software. This article describes the techniques used in the MPC's linking program and briefly compares this technique with those used by other researchers. The key feature of the MPC linking method is the construction of cumulative name-similarity scores, based on approximately 2.5 billion record comparisons; it also uses support vector mechanics to classify potential links. In this article, the authors explain modifications made for the final linked data sets and include a discussion of the role of weighting variables when using linked data.

Original languageEnglish (US)
Pages (from-to)7-14
Number of pages8
JournalHistorical Methods
Volume44
Issue number1
DOIs
StatePublished - Jan 1 2011

    Fingerprint

Keywords

  • census samples
  • complete count
  • historical demography
  • microdata
  • record linkage

Cite this