REU Site: Using Data Science Tools to Improve Neighborhoods

This project is funded from the Research Experiences for Undergraduates (REU) Sites program in the Directorate for Social, Behavioral, and Economic (SBE) Sciences. This program will engage students in exploration of economic, environmental, and infrastructure challenges in Dallas, Texas, using publicly available data sources. The eight-week program will begin with a three-week period of lectures in social science theory and research methods, including place-based economics, data collection methodologies and sources, and data ethics.

Simultaneous technical workshops will equip students with skills in coding, data wrangling, and use of mapping software. Each participant will then join a research team focused on measuring variability or mitigating disparities across urban neighborhoods. The objectives of the REU site are to provide students with expertise in data science methods and tools for studying problems affecting small geographies; to increase interest in data science and the likelihood that participating students will pursue graduate studies in a related field, especially for students underrepresented in these fields; to engage students in research that requires interdisciplinary skills and knowledge; and to increase collaboration across disciplinary boundaries for both faculty and students, and between university researchers and community partners.

Examples of research team topics are: (a) assessing the fairness of county polling place locations; (b) describing the change in urban heat island locations and intensity over time, as well as its proximity to neighborhoods; (c) evaluating economic development, workforce development, and affordable housing interventions in small urban areas; and (d) educating and empowering neighborhoods near toxic chemical waste sites using data-driven findings. To address its questions, each team will use specialized methodologies and tools, which the students will be exposed to under the supervision of their mentoring faculty and graduate student team.

The methodologies will vary by problem, but collectively include causal inference, complex survey design and analysis, clustering and prediction, data visualization and storytelling, and human-centered design. The research findings will be shared in a university-wide three-minute thesis contest for all summer undergraduate researchers, and an opportunity to publish a paper in the universities Journal of Undergraduate Research.

Besides the participants’ direct contribution to these studies, the data products they produce by retrieving, wrangling, and curating the public data will make a valuable contribution to a collective data resource about Dallas, which can be shared with local researchers.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.