Administrative Data Pipeline to Aid Wide Variety of Social Science Research

FOR IMMEDIATE RELEASE — In an effort to enhance our understanding of the impact of research funding, UW–Madison has signed a data-sharing agreement with the Institute for Research on Innovation and Sciences (IRIS) at the University of Michigan, which builds on and expands UMETRICS (Universities: Measuring the Impacts of Research on Innovation, Competitiveness, and Science), an initiative begun by universities belonging to the Committee on Institutional Cooperation (CIC) in March 2013. UMETRICS contains longitudinal data on researchers, graduate students, and postdoctoral fellows being supported on federal grants and the vendors providing goods and services to support these research grants.

"UMETRICS is a new way to generate data to quantify the impact of funding that supports university research beyond just the publications and citations of funded researchers," says Brent Hueth, a UW–Madison associate professor of agricultural and applied economics. Hueth, who has spearheaded the collaboration with IRIS, adds that a key motivation is to demonstrate the economic impact of federal research funding. IRIS data in a coded or de-identified form will be accessible to researchers at member institutions.

Existing studies often focus on individual grants, Hueth adds. "It's important to understand the full network of how scientific funding works and how it affects the economy. We want to look at the broader networks of scientific researchers and projects."

The new data source has many possible applications. Hueth says the journal Science has already published a study based on the IRIS data that portrayed the benefits of a university degree on the earnings of graduates. In the future, the data can be used to assess productivity in terms of spinoffs and jobs. The focus of a study could easily be narrowed to particular states, disciplines or institutions.

For his own research, Hueth says, "We're interested in the key determinants of productive research teams. Is it the individual scientist, the institution, or, for example, being close to a large urban area, where there are spillovers to nearby business?"

In March, UW–Madison's Education and Social/Behavioral Science Institutional Review Board approved an "umbrella protocol" to provide UW–Madison researchers access to de-identified IRIS data. The UW–Madison Wisconsin Research Data Center will provide access to data. Individual UW–Madison researchers will need to submit an IRB protocol to use the data.

Each month, UW–Madison's RSP administrative team will upload via a secure server "transactional" data on sponsored research, with records of receipts, on-campus dispersals, and off-campus purchases, to IRIS. The IRIS agreement contains provisions to ensure researcher privacy, Hueth says. Data will be uploaded via an encrypted channel, and held in a secure data enclave.

The IRIS consortium "is one example of the emerging big data revolution in social science," Hueth adds. "Digital data lowers the cost of doing research. And the shrinking availability of public funds for research has heightened the need for advocacy within the science community. We have to be more proactive, to make the case that what we do matters."