Full Citation

Title: Should We Trust Occupational Income Scores?

Citation Type: Miscellaneous

Publication Year: 2021

ISBN:

ISSN:

DOI:

NSFID:

PMCID:

PMID:

Abstract: Historical studies of labor market outcomes frequently suffer from a lack of data on individual income. The occupational income score (OCCSCORE) is often used as an alternative measure of labor market outcomes, particularly in studies of the U.S. prior to 1950. While researchers have acknowledged that this approach introduces measurement error, no effort has been made to quantify its impact on inferences. Using modern Census data, we find that the use of OCCSCORE biases results towards zero and can frequently result in statistically significant coefficients of the wrong sign. We show that a simple adjustment to OCCSCORE can substantially reduce this bias. We illustrate our results using the 1915 Iowa State Census, a rare source of pre-1950 earnings data. Using OCCSCORE in this context yields an attenuated wage gap for blacks and a statistically significant wage gap of the wrong sign for women; our adjusted OCCSCORE eliminates almost all of this bias. We also examine how bias due to the use of OCCSCORE affects estimates of intergenerational mobility using linked data from the 1850-1910 Censuses.

Url: https://ar5iv.labs.arxiv.org/html/1704.08299

User Submitted?: No

Authors: Saavedra, Martin

Publisher: Department of Economics, Oberlin College; Tate Twinam Interdisciplinary Arts and Sciences, University of Washington Bothell

Data Collections: IPUMS USA

Topics: Labor Force and Occupational Structure, Methodology and Data Collection, Poverty and Welfare

Countries: