Working Paper: NBER ID: w23019
Authors: Wanying Chang; Wei Cheng; Julia Lane; Bruce Weinberg
Abstract: Funding of research is critically important because it affects the flow of new, doctorally qualified scientists into the workforce. This paper provides new insights into how survey data can be combined with administrative records to examine the ways in which funding affects workforce decisions. We show that NSF supports more graduate students per dollar spent than other federal agencies. Not surprisingly, NIH heavily supports biology, health, and psychology PhDs, while NSF heavily supports PhDs in engineering, the physical sciences, mathematics, and computer science. Federal funding overall and by agency is related to who does research – a larger share of doctoral recipients supported by NIH are women (50%), African American (2.6%) and Hispanic (4.2%), compared to NSF, the Department of Defense (DOD) or the Department of Energy (DOE). Finally, federal funding is highly correlated with the pipeline of researchers going into different fields, particularly R&D fields, and the decision to pursue postdoctoral fellowships.
Keywords: federal funding; doctoral workforce; research; administrative records; survey data
JEL Codes: I23; J24; O3
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Federal funding (I22) | Doctoral recipients' career outcomes (Y40) |
| Federal funding (I22) | Likelihood of pursuing postdoctoral positions (I23) |
| Federal funding (I22) | Workforce decisions (J29) |
| Federal funding (I22) | Demographic composition of doctoral recipients (J79) |
| Intensity of funding (I22) | Decision to pursue postdocs (M51) |