Page 40 - IJPS-11-6

International Journal of
Population Studies Gender gaps in reporting limitations

focusing on hypothetical situations rather than personal consisted of an off-year mail survey, which included a brief
experiences, they reduce social desirability bias and set of questions about respondents’ personal health and
encourage more candid responses. For example, vignettes disability status, along with a series of anchoring vignettes
can help mitigate justification bias, where respondents (Baguley, et al., 2022; Yin & Heiland, 2022).
exaggerate their health limitations to rationalize being out SHARE is a large-scale cross-national panel study of
of the labor force. This application has been particularly individuals from multiple European countries aged 50 or
useful in labor economics and health research, where older (Börsch-Supan & Jürges, 2005; Servais, 2010). It is
accurate self-reports are critical for informed policymaking sponsored by the European Commission, the U.S. National
(Yin & Heiland, 2022).
Institute on Aging, and the German Federal Ministry of
Beyond reducing bias, the vignette approach provides Education and Research (http://www.share-project.org/).
a unique lens for exploring psychological and contextual SHARE was intentionally modeled after the U.S. HRS
factors that shape attitudes and behaviors. It enables and follows a harmonized structure across participating
researchers to capture the interplay between individual countries to facilitate cross-national research. A unique
characteristics and external influences, offering a more feature of the SHARE data is that the self-reported work
nuanced understanding of subjective assessments. Recent disability question, along with a series of vignettes, was
studies employing vignettes have revealed how cultural administered to the same respondents twice – first in
background, education, and socioeconomic status 2004 and again in 2006 – making it panel in nature. Eight
influence self-reported health and well-being, providing SHARE countries (Belgium, France, Germany, Greece,
critical insights into reporting heterogeneity across Italy, the Netherlands, Spain, and Sweden) agreed to
populations (Baguley et al., 2022; Kapteyn et al., 2009; Xu include both the self-assessment and vignette questions on
& Xie, 2016). work limitations in each of these two waves. The SHARE
By improving the validity and comparability of vignettes are a subset of the HRS vignettes. The names of
subjective measures, the vignette approach has become an the vignette characters are changed to common names
invaluable tool in survey-based research. Its applications in the country where the survey was conducted. The
extend beyond health to fields such as education, political translation procedures and translation assessments are
science, and consumer behavior, demonstrating its provided by Harkness (2005).
versatility and theoretical relevance in addressing the
limitations of self-reported data. 2.2. Measurements
2.2.1. Vignette questions
2. Data and methods
The work disability vignettes in both HRS and SHARE
2.1. Data sources describe hypothetical work limitations in three domains:
Our analyses used data from the Health and Retirement Pain, cardiovascular health, and depression. In each
Study (HRS) and the Survey of Health, Ageing and domain, several distinct vignettes are used to describe
Retirement in Europe (SHARE). HRS is an ongoing the conditions of a hypothetical person. An example of a
nationally representative biennial panel study of the vignette is the following: Lori has had heart problems in
U.S. population over the age of 50. It is sponsored by the past and she has been told to watch her cholesterol level.
the National Institute on Aging and the Social Security Sometimes, if she feels stressed at work, she feels pain in her
Administration and has been conducted by the University chest and occasionally in her arms. Following each vignette
of Michigan since 1992 (http://hrsonline.isr.umich.edu/). description, the respondent is asked: “How much is s/he
The information collected includes health, socioeconomic limited in the kind or amount of work s/he could do?”
status, and social program participation. Although the The answer follows a five-point scale: (1) None; (2) Mild;
HRS is a panel dataset, the vignette questions included (3) Moderate; (4) Severe; and (5) Extreme/Cannot do any
in the 2004 and 2007 waves that this study relied on were work. The last response category is “Cannot do any work”
administered as separate and independent experiments. in the HRS 2004 wave, a bit different from the HRS 2007
To account for the fact that only a subset of respondents wave and the SHARE, where the last response category is
participated in these vignette studies, we used the sampling “Extreme.” Preceding the vignette questions, respondents
weights provided by HRS for each wave in all our analyses. are asked about their own work limitations: “Do you have
In 2004, the vignettes were included in a leave-behind any impairment or health problem that limits the kind
questionnaire completed by a subsample of respondents or amount of work you can do?” with the same answer
following their core interview; this questionnaire featured categories. The list of vignettes is provided in the Appendix.
a series of work disability vignettes. The 2007 wave In addition, two tables are included: Table A1 summarizes

Volume 11 Issue 6 (2025) 34 https://doi.org/10.36922/ijps.1969