Data science for the public good is a major driver for research at the Social and Decision Analytics Laboratory (SDAL) in the Biocomplexity Institute of Virginia Tech. 

But from May to August, Data Science for the Public Good (DSPG) takes a more formal stance as a program rooted in the lab in the National Capital Region. The program gives students an opportunity to work on teams with SDAL faculty and post docs to determine how data generated in the community can be leveraged to improve quality of life.

While the six graduate students, recruited as Fellows from universities across the country, and 12 undergrads from Virginia Tech have some background in data analytics, the majority do not. The students come from fields like criminology, psychology, sociology, political science, urban planning, economics, computer science, electrical and mechanical engineering, and statistics. 

“The team science research model we use benefits tremendously from diversity of both experience and disciplines,” said Sallie Keller, professor and director of the Social and Decision Analytics Laboratory.

“What our junior colleagues have in common is that they have grown up in a different world than I and other more senior researchers have,” Keller said. “Their approach to technology and data is fearless and a real advantage, particularly in the area of data science.”

Wayne Strickland, executive director of the Roanoke Valley-Allegheny Regional Commission, concurred with Keller. “We were very impressed with the students’ work. They had ways of analyzing data that we had not even thought about before,” he said.

A DSPG team helped the commission look at workforce development in the region and how best to attract new talent.

“We have 55,000 college students in our area and instead of having them go elsewhere, we want to keep them here,” Strickland said. “At the same time, though, we want to retain our current workforce population.”

In considering what attracts people to stay or come to an area, the team expanded its research to include health-related issues like food insecurity, primary care providers, and obesity, as well as natural assets like air quality and greenways.

“They provided us with indices and maps that will help with our goals and strategies,” said Strickland, “but our research collaboration is nowhere near over. It will be continuing.”

Strickland was one of two keynote speakers at a symposium on August 9 that concluded the DSPG summer program. About 150 academics and policymakers attended the event which included a session where the students presented posters about their research.

The other keynote speaker was Catherine Woteki, former undersecretary for Research, Education and Economics and chief scientist for the U.S. Department of Agriculture, who lauded the DSPG program as “a wonderful example of how communities can use data to solve a wide variety of problems.”

“We need more resources, training, and programs like DSPG both nationally and internationally,” Woteki said.

Three of the 13 DSPG projects this summer were from Arlington County, a frequent collaborator with SDAL.

“The best part of collaborating with partners like SDAL is that we can point to an outside organization to provide validation and that is invaluable to us,” said Master Police Officer Dimitrios Mastoras of the Arlington County Police Department (ACPD).

The DSPG team working with the police department evaluated data from the ACPD’s Arlington Restaurant Initiative that focuses on training restaurant personnel on best practices to reduce the risk of alcohol-related disorders. An interactive dashboard built by the team plotted data that showed a decrease in arrests.

The next step, said Mastoras, is to develop methods to estimate reduction in crime solely due to the Arlington Restaurant Initiative since other factors also play a role in crime reduction. This information can further be useful in helping calculate cost savings as a result of the initiative.

Alexa Nosal, a junior majoring in sociology, criminology, and psychology, said the DSPG program met all of her expectations. “I was attracted to the program by its interdisciplinary aspect and it was great to be involved in something like the Arlington Police Department project that is helping to keep the community safer.”

Luke Kim, a junior majoring in computational modeling and data analytics, said it was the first time he had been exposed to an academic environment outside the classroom and cited the networking opportunities afforded through the program as being particularly valuable to him.

“I was learning and discovering new things every day by diving into community data. I now look at things like air quality and traffic conditions in a totally different light,” said Kim, who worked on a team to help the Virginia Cooperative Extension better understand its geographical scope and better allocate its resources.

“The transformation of students like Kim over the summer is pronounced,” said Aaron Schroeder, senior research scientist and co-director of the DSPG program. “Leaving the classroom — where problems and solutions are generally well-defined — and reaching out to discover and define data they did not even know existed has an effect on them.”

In addition to the Roanoke Valley-Alleghany Regional Commission; Arlington County Police Department and the Virginia Cooperative Extension, other sponsors of the DPSG program include: Virginia Tech Center for Peace Studies and Violence Prevention; Arlington Community Progress Network; Virginia Tech Office of the Executive Vice President and Provost; National Center for Science and Engineering Statistics at the National Science Foundation; Arlington Department of Human Services; U.S. Army Research Institute for Behavioral and Social Science Research; Virginia Department of Social Services; United Methodist Foster Services; Virginia Tech Biocomplexity Institute; the United States Department of Agriculture National Institute of Food and Agriculture; Virginia Tech Data and Decisions Area; Virginia Tech Office of the Vice President in the National Capital Region, and Sage Publishing.

“As the DSPG program marked its fourth year, we were very pleased to have five new project sponsors,” said Gizem Korkmaz, research assistant professor at SDAL, co-director of the DSPG program, and faculty on several of the project teams. “We had 16 sponsors, some new, and  some returning. We are proud that our returning sponsors continue to ask our assistance in solving the many problems they face and also to have new sponsors.”


Note: The students enrolled in this summer’s DSPG program include: Ph.D. students Bayoan Santiago, an economics major at Claremont Graduate University; Emily Sheen, a statistics major at Penn State; and Davon Woodard, a major in planning, governance, and globalization/urban computing at Virginia Tech. Master’s students Fangfang Lee, a statistic major at New York University; Daniel Liden, a statistics major at the University of Minnesota-Twin Cities; and Devika Nair, a data analytics major at American University. Virginia Tech Honors College undergraduates Hannah Brinkley and Keren Chen, juniors majoring in computer science; Eirik Iversen, Megan Grondine and Cory Kim, seniors majoring in computational modeling and data analytics; Luke Kim, a junior majoring in computational modeling and data analytics; Douglas Mattingly, a junior majoring in mechanical engineering; Kelsey McMahon, a junior majoring in sociology and criminology; Armaghan Mosavian, a junior majoring in international public policy and public and urban affairs; Alexa Nosal, a junior majoring in sociology, criminology, and psychology; Raghav Sawhney, a senior majoring in  computational modeling and data analytics; and Weerasu (Jay) Warajuntano, a senior majoring in electrical engineering.

Written by Barbara Micale

Published by Tiffany Trent, September 18, 2018
Tags: Data Science for the Public Good  Education and Outreach