Should We Trust Regression to Measure Social Disparities?

Christopher Wu

No 4znu5, SocArXiv from Center for Open Science

Abstract: When investigating socioeconomic disparities in exposure to social ills, it is common to use (semi-)parametric regression, and interpret the parameter estimates as measures of inequality. However, disparity researchers often know little about the data generating process; therefore, not only are models misspecified, but it becomes difficult to even know which model is `better'. Under misspecification, common interpretations of regression coefficients can break down. Without a `right' model, model selection can be arbitrary, leading to credibility concerns as estimates are often sensitive to model selection. Predictive model selection can attenuate the measure of inequality if irrelevant predictors explain away the inequality. Examples are presented to illustrate key pitfalls.

Date: 2022-07-20
New Economics Papers: this item is included in nep-hea
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://osf.io/download/62d7940427b746063a0ac0bf/

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:osf:socarx:4znu5

DOI: 10.31219/osf.io/4znu5

Access Statistics for this paper

More papers in SocArXiv from Center for Open Science
Bibliographic data for series maintained by OSF ().