Evaluating the utility of linked administrative data for nonresponse bias adjustment in a piggyback longitudinal survey

Tobias J. M. Büttner, Joseph W. Sakshaug, Basha Vicari

Research output: Contribution to journalArticlepeer-review

Abstract

Nearly all panel surveys suffer from unit nonresponse and the risk of nonresponse bias. Just asthe analytic value of panel surveys increase with their length, so does cumulative attrition,which can adversely affect the representativeness of the resulting survey estimates. Auxiliarydata can be useful for monitoring and adjusting for attrition bias, but traditional auxiliarysources have known limitations. We investigate the utility of linked-administrative data toadjust for attrition bias in a standard piggyback longitudinal design, where respondents from apreceding general population cross-sectional survey, which included a data linkage request,were recruited for a subsequent longitudinal survey. Using the linked-administrative datafrom the preceding survey, we estimate attrition biases for the first eight study waves of thelongitudinal survey and investigate whether an augmented weighting scheme thatincorporates the linked-administrative data reduces attrition biases. We find that adding theadministrative information to the weighting scheme generally leads to a modest reduction inattrition bias compared to a standard weighting procedure and, in some cases, reducesvariation in the point estimates. We conclude with a discussion of these results and remark onthe practical implications of incorporating linked-administrative data in piggybacklongitudinal designs.
Original languageEnglish
Pages (from-to)837-864
Number of pages28
JournalJournal of Official Statistics
Volume37
Issue number4
DOIs
Publication statusPublished - 26 Dec 2021

Keywords

  • attrition
  • auxiliary data
  • between-wave events
  • panel survey
  • weighting

Fingerprint

Dive into the research topics of 'Evaluating the utility of linked administrative data for nonresponse bias adjustment in a piggyback longitudinal survey'. Together they form a unique fingerprint.

Cite this