Measuring and controlling for non‐consent bias in linked survey and administrative data

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

Linking surveys to administrative databases is a common practice in the health and social sciences. Some progress has been made in implementing strategies to reduce the risk of non-consent bias in surveys, either through careful design of the linkage consent request or statistical procedures applied post-data collection by the survey organization or the data analyst. This chapter reviews these strategies for measuring linkage non-consent bias and controlling for its effects. Specifically, the empirical literature is summarized, highlighting examples where attempts have been made to measure and/or control for linkage non-consent bias, noting the methods used and their associated strengths and limitations. The chapter covers methods that aim to maximize the linkage consent rate; methods of statistically adjusting for non-consent bias are covered.
Original languageEnglish
Title of host publicationAdministrative Records for Survey Methodology
EditorsAsaph Young Chun, Michael D. Larsen, Gabriele Durrant, Jerome P. Reiter
PublisherWiley-VCH Verlag
Chapter7
Pages155-178
Number of pages24
ISBN (Electronic)9781119272076
ISBN (Print)9781119272045
DOIs
Publication statusPublished - 6 Apr 2021

Keywords

  • administrative databases
  • linked survey
  • non-consent bias
  • survey analysis
  • survey organization

Fingerprint

Dive into the research topics of 'Measuring and controlling for non‐consent bias in linked survey and administrative data'. Together they form a unique fingerprint.

Cite this