CPS launches SUNGEO project to assist in merging data across different scales
June 12, 2024
Contact: Jon Meerdink ([email protected])
ANN ARBOR — The Institute for Social Research’s Center for Political Studies (CPS) has launched a new project to address a common challenge for social researchers: misalignment that arises when data are collected at varying levels of scale.
The Subnational Geospatial Data Archive (SUNGEO) went live in its beta form on April 17, 2024. It uses a variety of tools to integrate geospatial data, allowing researchers and analysts to evaluate data relationships from a variety of sources, scales, and geographical contexts.
Having misaligned data is a key hurdle for many researchers across disciplines. Co-principal investigators Yuri Zhukov and Ken Kollman, along with their coauthors, outlined the nature of the problem in a March 2023 paper on the issue.
“Researchers studying legislative elections in the United States might observe data for variables at the electoral district level (e.g., campaign strategies) and at the county level (e.g., crime). To understand how, for example, local crime influences campaign strategies, one must integrate two datasets, using measured values at the county level to estimate levels of crime in each legislative district,” the team wrote. “Statistically, this represents a change-of-support (CoS) problem: making inferences about a variable at one geographic support (destination units) using measurements from a different support (source units). Changes of support entail information loss, potentially leading to consequential measurement error and biased estimation.”
SUNGEO offers technical solutions to address this common issue. According to CPS director Kollman, SUNGEO’s versatility should allow researchers to find ways to rectify misalignment problems.
“This new infrastructure gives a brand new set of tools for researchers to merge these data sets using a variety of methods,” he said .”Especially with the software underneath, people can try out different methods for merging data to see if their scientific results are consistent.”
In addition to Kollman and Zhukov, the project team includes Julia Lippman, Christopher Fariss, Robert Franzese, Allen Hicken, Brian Min, Anne Pitcher, Yuki Shiraito, and Walter Mebane — all of the University of Michigan — as well as David Backer of the University of Maryland.
SUNGEO offers data on demography, politics, climate, violence, public health, weather, and terrain at a variety of geographic scales. The tool is currently available in beta format at sungeo.org. Its 2.0 iteration is currently in development. The project was funded by a grant from the National Science Foundation.