Testing the specifications of parametric models using anchoring vignettes

A.H.O. van Soest, H. Vonkova

Research output: Contribution to journalArticleScientificpeer-review

Abstract

Comparing assessments on a subjective scale across countries or socio-economic groups is often hampered by differences in response scales across groups. Anchoring vignettes help to correct for such differences, either in parametric models (the compound hierarchical ordered probit (CHOPIT) model and extensions) or non-parametrically, comparing rankings of vignette ratings and self-assessments across groups. We construct specification tests of parametric models, comparing non-parametric rankings with rankings by using the parametric estimates. Applied to six domains of health, the test always rejects the standard CHOPIT model, but an extended CHOPIT model performs better. This implies a need for more flexible (parametric or semiparametric) models than the standard CHOPIT model.
Original languageEnglish
Pages (from-to)115-133
JournalJournal of the Royal Statistical Society, Series A
Volume177
Issue number1
DOIs
Publication statusPublished - Jan 2014

Fingerprint

Probit Model
Parametric Model
Specification
Ranking
Testing
Nonparametric Model
ranking
Self-assessment
Specification Test
Semiparametric Model
Health
Economics
Group
Imply
self-assessment
Parametric model
Ordered probit model
Vignettes
Anchoring
Estimate

Keywords

  • Compound hierarchical ordered probit model
  • reporting bias
  • self-assessed health
  • specification testing

Cite this

@article{282cf3f065e04ed38f358df9d42fb4c3,
title = "Testing the specifications of parametric models using anchoring vignettes",
abstract = "Comparing assessments on a subjective scale across countries or socio-economic groups is often hampered by differences in response scales across groups. Anchoring vignettes help to correct for such differences, either in parametric models (the compound hierarchical ordered probit (CHOPIT) model and extensions) or non-parametrically, comparing rankings of vignette ratings and self-assessments across groups. We construct specification tests of parametric models, comparing non-parametric rankings with rankings by using the parametric estimates. Applied to six domains of health, the test always rejects the standard CHOPIT model, but an extended CHOPIT model performs better. This implies a need for more flexible (parametric or semiparametric) models than the standard CHOPIT model.",
keywords = "Compound hierarchical ordered probit model, reporting bias, self-assessed health, specification testing",
author = "{van Soest}, A.H.O. and H. Vonkova",
year = "2014",
month = "1",
doi = "10.1111/j.1467-985X.2012.12000.x",
language = "English",
volume = "177",
pages = "115--133",
journal = "Journal of the Royal Statistical Society, Series A",
issn = "0964-1998",
publisher = "Wiley-Blackwell",
number = "1",

}

Testing the specifications of parametric models using anchoring vignettes. / van Soest, A.H.O.; Vonkova, H.

In: Journal of the Royal Statistical Society, Series A, Vol. 177, No. 1, 01.2014, p. 115-133.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Testing the specifications of parametric models using anchoring vignettes

AU - van Soest, A.H.O.

AU - Vonkova, H.

PY - 2014/1

Y1 - 2014/1

N2 - Comparing assessments on a subjective scale across countries or socio-economic groups is often hampered by differences in response scales across groups. Anchoring vignettes help to correct for such differences, either in parametric models (the compound hierarchical ordered probit (CHOPIT) model and extensions) or non-parametrically, comparing rankings of vignette ratings and self-assessments across groups. We construct specification tests of parametric models, comparing non-parametric rankings with rankings by using the parametric estimates. Applied to six domains of health, the test always rejects the standard CHOPIT model, but an extended CHOPIT model performs better. This implies a need for more flexible (parametric or semiparametric) models than the standard CHOPIT model.

AB - Comparing assessments on a subjective scale across countries or socio-economic groups is often hampered by differences in response scales across groups. Anchoring vignettes help to correct for such differences, either in parametric models (the compound hierarchical ordered probit (CHOPIT) model and extensions) or non-parametrically, comparing rankings of vignette ratings and self-assessments across groups. We construct specification tests of parametric models, comparing non-parametric rankings with rankings by using the parametric estimates. Applied to six domains of health, the test always rejects the standard CHOPIT model, but an extended CHOPIT model performs better. This implies a need for more flexible (parametric or semiparametric) models than the standard CHOPIT model.

KW - Compound hierarchical ordered probit model

KW - reporting bias

KW - self-assessed health

KW - specification testing

U2 - 10.1111/j.1467-985X.2012.12000.x

DO - 10.1111/j.1467-985X.2012.12000.x

M3 - Article

VL - 177

SP - 115

EP - 133

JO - Journal of the Royal Statistical Society, Series A

JF - Journal of the Royal Statistical Society, Series A

SN - 0964-1998

IS - 1

ER -