Creating Trusted Independent Data About the Impact of Research
The Institute for Research on Innovation and Science (IRIS) is a consortium of universities anchored by an IRB-approved data repository hosted at the University of Michigan’s Institute for Social Research. IRIS was founded in 2015 with support from the Alfred P. Sloan and Ewing Marion Kauffman foundations.
IRIS collects record level administrative data from its member institutions and uses those data to produce secondary de-identified research data that can support research and reporting that will improve our ability to understand, explain and improve the public value of research.
The Need For Better Data
Universities play a central role in performing research, but they also have unique capacities for scholarship that seek to understand the research process, document its results, improve the use of resources, identify fruitful collaborations, advance the education and training of students and early career researchers, and demonstrate the social as well as economic impact of discovery and learning.
Nevertheless, much current research on science and innovation is based on hand curated, artisanal, data efforts. In this environment, researchers often have few incentives to document and distribute their data for use by others. Even when data producers are willing and able to share, the underlying data designed to answer specific questions are often difficult to use for other purposes, limiting possibilities for use.
The IRIS Architecture
Our data architecture combines administrative data derived from University Human Resources, Sponsored Projects, and Procurement systems with data on scientific outputs including publications, patents and dissertations as well as information about the scientific content of federal grants to form the core of an extensible dataset that can be linked at the individual, establishment, and grant level to many additional data sources. Our partnership with the U.S. Census Bureau’s Innovation Measurement Initiative (IMI) enables integration of IRIS research expenditures data with Census economic and demographic data including the Longitudinal Employer Household Dynamics (LEHD) dataset and Longitudinal Business Database (LBD). Linked IRIS-Census data supports research and reporting that provides unique insights into the careers of research trained individuals, their impact on employers and industries.
Making the Study of Research about People
The core outcome of interest for science funders is the creation, transmission, and adoption of ideas and discoveries. Our analytic framework is centered on individual researchers and teams that are embedded in several types of social networks. Science funding works in part by enabling collaboration networks to exist and expand. Research investments enable investigators to hire the people and procure the goods and services necessary to conduct scientific work. That work trains people and produces knowledge. It also creates and sustains the collaboration networks that are the proximate source of innovation. As people move from individual campuses to jobs in many sectors, they carry ideas and skills learned through research with them. The application of these ideas and skills in new settings is what drives social and economic impact. The IRIS data infrastructure is designed to capture these movements and networks in a context that enables rigorous analysis of the joint production and eventual effects of new knowledge and skilled people.
Accessing IRIS Data
IRIS seeks to make the data it produces as broadly and easily available within responsible concerns for privacy and confidentiality. De-identified secondary data produced by IRIS is available through a virtual data enclave. Annual data releases include improvements to both data and documentation. Census linked data are available for research use through the Federal Statistical Research Data Center (FSRDC) system, administered by the U.S. Census Bureau. For more information about our current data release and procedures for accessing it, please see our Research Data page.
HISTORY OF IRIS
The UMETRICS project compellingly demonstrated that universities can leverage and link their own administrative data to better understand their research portfolios, document their varied economic impact, and generate new scholarly findings.
Since 2013, some 63 researchers have accessed UMETRICS data under strict confidentiality protections to pursue research projects that analyze the career outcomes of graduate students, the dynamics and consequences of an aging scientific workforce, the optimal structure of scientific research teams, the contributions of academic discoveries to food safety and security, and entrepreneurship by faculty, staff, and students.
Research projects sparked by UMETRICS were supported by six grants totaling nearly $10 million from three federal science agencies. These projects produced peer-reviewed articles in Science and Research Policy, as well as aggregate and campus-level reports that participating institutions used effectively in outreach to boards of regents and trustees, and to state and federal legislators.
In addition, UMETRICS project team members were invited to present testimony to the Senate Appropriations Committee, and commissioned papers to panels convened by the National Academy of Sciences and the National Academy of Engineering.
The success of UMETRICS in its first two years is a testament to the power of a university-led collaborative endeavor. The Institute for Research on Innovation and Science (IRIS) has continued the progress modeled by the UMETRICS initiative.
WANT TO LEARN MORE?
JOIN IRIS MAILING LIST
HOW IRIS WORKS
Click the image to view larger.
- about 4 weeks agoRead a review of "Big Data and Social Science," edited by Foster, Ghani, Jarmin, Kreuter, & Lane at https://t.co/EtLWCXugMg
- about 1 month agoIRIS Data Summit Invitation - https://t.co/SLXzy7dt0P
- about 1 month agoJason Owen-Smith & Census colleagues on how to access UMETRICS data for research in Census RDCs, IRIS VDE: https://t.co/3mCTcNAgqZ