Home > Publications . Search All . Browse All . Country . Browse PSC Pubs . PSC Report Series

PSC In The News

RSS Feed icon

Frey and colleagues outline 10 trends showing scale of America's demographic transitions

Starr says surveys intended to predict recidivism assign higher risk to poor

Prescott and colleagues find incidence of noncompetes in U.S. labor force varies by job, state, worker education


PAA 2015 Annual Meeting: Preliminary program and list of UM participants

ISR addition wins LEED Gold Certification

PSC Fall 2014 Newsletter now available

Martha Bailey and Nicolas Duquette win Cole Prize for article on War on Poverty

Next Brown Bag

Mon, March 9
Luigi Pistaferri, Consumption Inequality and Family Labor Supply

Hongwei Xu photo

Assessing the Effectiveness of Anchoring Vignettes in Bias Reduction for Socioeconomic Disparities in Self-Rated Health among Chinese Adults

Publication Abstract

Download PDF versionXu, Hongwei, and Yu Xie. 2014. "Assessing the Effectiveness of Anchoring Vignettes in Bias Reduction for Socioeconomic Disparities in Self-Rated Health among Chinese Adults." PSC Research Report No. 14-820. May 2014.

This study investigates how reporting heterogeneity may bias socioeconomic and demographic disparities in self-rated health, a widely used health indicator, and how such bias can be adjusted by anchoring vignettes among Chinese adults. Drawing data from the 2012 wave of the China Family Panel Studies, we find strong evidence of systematically different cut-points applied by people of varying groups to rate their overall health status. In many cases, such cut-point shifts are not parallel in that the effect of certain group characteristic on the shift is stronger at certain level than another. We find that the resulting bias of measuring group differentials in self-rated health can be too substantial to be ignored. We further demonstrate that anchoring vignettes prove to be an effective survey instrument and statistical tool in obtaining bias-adjusted estimates of health disparities. We also find it sufficient to administer vignettes to only a small subsample (20-30% of the full sample) in order to adjust reporting heterogeneity in the full sample. Using single vignette can be as effective as using more in terms of anchoring, but the results are sensitive to the choice of vignette design. Our findings suggest that future research using self-rated health should guard against reporting heterogeneity and employ adjustment techniques such as anchoring vignettes whenever appropriate.

Country of focus: China.

Browse | Search : All Pubs | Next