A focus of Dr. Marschke’s research is the relation between innovation and technological change and the science, technology, engineering, and mathematics, or STEM, workforce. It is a robust STEM workforce that drives the improvements in economic productivity that drive a rising standard of living in developed economies. But to many school-aged persons the personal rewards of a STEM career appear meager and uncertain compared to the costs and rigors of the required graduate education and postgraduate training. Indeed, STEM graduate education and training in the U.S. has come to be seen by many education experts as a highly exploitative system that enrolls more students and postdocs than there are STEM jobs to fill. Data for evaluating proposals to reform this system, or more fundamentally, for evaluating and quantifying the long-term impact of current university-based STEM education and training on the young people who undertake it and on the larger society are lacking, however.
This NSF grant funds the construction of a new data platform that enables tracking the careers and measuring the labor market outcomes of STEM PhD graduates and postdocs. Dr. Marschke will lead a team of students and collaborators that will use detailed human resources data from a set of prominent U.S. research universities to develop and validate machine learning algorithms to identify STEM PhDs and postdocs in the U.S. Census’s Longitudinal Employer-Household Dynamics (LEHD) database, a longitudinal matched employee-employer database containing most U.S. workers and their employment spells. The end product will be a new and comprehensive panel data set of PhDs and postdocs along with their demographic, employment and employer information.
In addition, the grant funds research using the new data that will: (1) document STEM career and hiring patterns and flows of STEM graduates between universities and different sectors of the economy; (2) estimate the private returns to education and training for STEM PhDs and postdocs, including by subgroup (e.g., women and minorities); and (3) formulate and estimate new models of STEM labor demand based on state-of-the-art econometric methods and innovative identification strategies that are only possible with the new longitudinal data.
The data and research generated by Dr. Marschke’s team will produce a feedback mechanism for educators and policymakers on the outcomes of STEM graduates as well as on the impact of STEM training on productivity in the economy, thus enhancing the tools available to conduct STEM workforce development research. The project will produce data and models of STEM workforce demand to assist policy-makers in deploying scarce educational resources and for forecasting far enough in advance (the STEM PhD and postdoc pipeline is decades long) to more effectively formulate and implement policies to head off STEM gluts and shortages.
Return to Newsletter home