Undercoverage of the elderly institutionalized population: The risk of biased estimates and the potentials of weighting institutionalized population and not the risk of biased estimates: A simulation study using cross‑national survey data
Year: 2019
Type: article
Abstract: In most social surveys, the elderly institutionalized population is not part of the target population because it \nis considered as hard-to-reach and hard-to-interview. The deliberate exclusion of institutionalized elderly \nfrom survey samples might cause bias, like previous studies investigating institutionalized elderly persons \nand their transition to institutions implied. We use a Monte Carlo simulation based on cross-national \nsamples of the Survey of Health, Ageing and Retirement in Europe (SHARE) to test whether the \nnoncoverage and undercoverage of the elderly institutionalized population lead to biased estimates. \nMoreover, we examined to what extent weights could be used to correct for the underrepresentation of the \ninstitutionalized population. Our results show that noncoverage leads to biased estimates in two healthrelated \nvariables. With respect to undercoverage, the precision of all estimates is better, especially if \nweights accounting for the hard-to-survey population are applied.
Author: Schanze Et Al.
Language: English
Subfield: Sociology and Political Science
Field: Social Sciences
Domain: Social Sciences
Open Access status: diamond
OpenAlex ID: W2936086115