Association for Public Policy Analysis & Management

Considerable research on academic R&D has focused on the research activity among senior scholars, with recent attention focused on the dynamics of innovation within labs and scientific teams. Less attention, however, has focused on the role of academic institutions in training the science and engineering (S&E) innovative workforce. The population of S&E graduate students, or emerging researchers, defines the next generation of innovators with significant economic potential – a topic of great policy significance. As they progress through their academic apprenticeship, these students are expected to develop and pursue their own research agenda. However, unlike their mentors, emerging researchers have few opportunities to secure external funding support for their research.

The National Science Foundation’s (NSF) Graduate Research Fellowship Program (GRFP) provides a rare opportunity for emerging researchers to obtain a substantial and prestigious award. The grant is unique and large, providing the most promising students with three years of guaranteed funding to pursue their own research agenda with no service obligation to their departments. In 2015, approximately 16,000 applicants submitted proposals, 12.5 percent were awarded competitive funding, and 6.25 percent were deemed near-misses and given formal recognition as honorable mentions due to their exceptional quality.[1]

Given the closeness in quality of GRFP award winning and honorable mention proposals, we have an opportunity to examine whether additional external factors that are unique to an applicant’s graduate university-department influence award assignment. More specifically, in this paper, we ask: are there organizational academic characteristics of the department or university that affect an applicant’s assignment to award versus honorable mention? The GRFP is unique to the rest of the NSF’s programs given the abbreviated proposal and early-stage of professional training for the applicant. Given the increased level of uncertainty that is inherent within the GRFP proposal, we argue that department and university traits influence GRFP assignment in two ways. First, an applicant’s academic institution (either department or university) may directly impact the assignment as the perceived quality of the institution serves as an indicator of the student’s quality and potential for success. Second, institutions also may indirectly impact assignment as they could provide resources and support that improve the quality of the application.

We merge the GRFP database with the National Research Council’s (NRC) data on Research Doctorate Programs to provide detailed information on department characteristics for a representative sample of doctoral departments across the US. Our sample contains 3,129 department-year observations from 199 universities and 41 S&E academic fields. Using these data we estimate a series of models to examine the effect of academic characteristics on GRFP award assignment. At the department level, we estimate the impact of student cohort productivity and diversity, faculty productivity and diversity, department support and size, in addition to university traits on having any GRFP award activity and the share of students in a department that receive an award. Preliminary results suggest that both cohort and faculty productivity are positively associated with award assignment.