Stemming inequality? Employment and pay of female and minority scientists and engineers

We presented an earlier version of this paper at the annual meeting of the Association for Public Policy Analysis and Management, Washington, DC, November 10, 2007. We are grateful to the U.S. Office of Personnel Management for providing the 1% sample of the Central Personnel Data File and to the Andrew Young School of Policy Studies, Georgia State University, for funding purchase of the data. Lakshmi Pandey and Chester S. Galloway provided invaluable assistance in preparing the 2000 Census 5% Public Use Microdata Sample. Julia Melkers, Mary Frank Fox, and Paula Stephan provided insightful comments on an earlier version of this paper.

Abstract

When white men overwhelmingly dominated the science, technology, engineering, and mathematics (STEM) work force, the high pay in STEM occupations was a major source of gender and race inequality in the U.S. economy. As women, Blacks, and Latinos increasingly study STEM fields, new possibilities for achieving pay equality are opening. We test whether the reality matches the promise using two large data sets. Analysis of a five percent Public Use Microdata Sample (PUMS) of the 2000 Census and the 2001–06 American Community Surveys shows that women and minorities earn more, relative to comparable white men, in STEM than in non-STEM fields. This general pattern persists in analysis of a 1% sample of federal personnel records, which include better measures of work experience and education. Thus, federal efforts to increase the representativeness of the STEM workforce should increase pay equality in the economy by moving women and minorities into traditionally high-paying fields.

Key words:

• Discrimination
• Earnings of scientists and engineers
• Women and minorities in science

Notes

We presented an earlier version of this paper at the annual meeting of the Association for Public Policy Analysis and Management, Washington, DC, November 10, 2007. We are grateful to the U.S. Office of Personnel Management for providing the 1% sample of the Central Personnel Data File and to the Andrew Young School of Policy Studies, Georgia State University, for funding purchase of the data. Lakshmi Pandey and Chester S. Galloway provided invaluable assistance in preparing the 2000 Census 5% Public Use Microdata Sample. Julia Melkers, Mary Frank Fox, and Paula Stephan provided insightful comments on an earlier version of this paper.

1 Tel.: [phone].

2 Census data classifies all occupations in the economy into 23 major groups. The other 19 occupations include management, legal, business and finance, healthcare practitioner, protective service, and sales and related occupations, etc. Throughout the paper, we refer these occupation groups as “non-STEM occupations” or “other occupations” and compare employees in STEM occupations with those in non-STEM occupations.

3 Hispanic women had smaller disparities in legal occupations than in the physical sciences. For Asian women, the pay disparity in law was smaller than those in the physical or biological sciences and comparable to those in computers and engineering.