Anchoring vignettes and cross-country comparability: an empirical assessment of self-reported mobility.

Research output: Chapter in Book/Report/Conference proceedingChapter

Author(s)

Department/unit(s)

Publication details

Title of host publicationHealth Econometrics
DatePublished - 2018
Pages145-174
PublisherEmerald
Place of PublicationEngland
EditorsBadi Baltagi, Francesco Moscone
Volume294
Original languageEnglish
ISBN (Electronic)978-1-78714-541-2
ISBN (Print)978-1-78714-542-9

Abstract

Anchoring vignettes have become a popular method to adjust self-assessed data for systematic differences in reporting behaviour to aid comparability, for example, of cross-country analyses. The method relies on the two fundamental assumptions of response consistency and vignette equivalence. Evidence on the validity of these assumptions is equivocal. This chapter considers the utility of the vignette approach by considering how successful the method is in moving self-assessed reports of health mobility towards objective counterparts. We draw on data from the Survey of Health, Ageing and Retirement in Europe (SHARE) and undertake pairwise country comparisons of cumulative distributions of self-reports, their objective counterparts and vignette adjusted reports. Comparison of distributions is based on tests for stochastic dominance. Multiple cross-country comparisons are undertaken to assess the consistency of results across contexts and settings. Both non-parametric and parametric approaches to vignette adjustment are considered. In general, we find the anchoring vignette methodology poorly reconciles self-reported data with objective counterparts.

Discover related content

Find related publications, people, projects, datasets and more using interactive charts.

View graph of relations