Taking on Social Problems with Analytics

By Sean Thornton • July 23, 2013

This summer, thirty-six young data scientists have descended upon Chicago to partake in a first-of-its kind program that leverages data science as a mechanism to solve social problems.

The Data Science for Social Good Fellowship (DSSG), a University of Chicago program sponsored by Google chairman and wife Eric and Wendy Schmidt, is partnering with multiple Chicago governments and NPOs to innovatively take on real-world problems these organizations face. DSSG aims to not only make a positive social impact on Chicago going forward, to but to help create a culture of social conscientiousness among data scientists.

Developing the Program

DSSG would not be possible without Rayid Ghani, former chief data scientist of the Obama 2012 campaign, who recently joined the University of Chicago in April. A month prior, he and Schmidt, who worked together on the Obama campaign, developed the idea for the program. Soon, Ghani was at the helm as program director, with support from U of C’s Computation Institute and Harris School of Public Policy. Together with key staff members Juan-Pablo Velez and Matt Gee, DSSG was ready to accept applications for its inaugural launch in June.

While Ghani’s team had hoped for a strong turnout, the response far exceeded expectations. 550 graduate and undergraduate students from six continents applied, giving Ghani, Velez and Gee the difficult but welcome challenge of choosing which applicants would be the best fit for DSSG.

In doing so, the methodology of Ghani’s team favored a diversity of experiences among Fellows. The program’s Fellows come from a collection of backgrounds, including computer science, statistics, social sciences and other domains—as Velez’s DSSG blog observes, a data scientist is not an “all-knowing, do-everything data rock star.” Accordingly, Fellows have been tasked to work in teams that combine their varied skill sets with a wide range of partners.

Partnerships and Projects: The City of Chicago and Beyond

During development, DSSG’s other major undertaking was forging partnerships for projects. While Ghani, Velez and Gee contacted a wide range of government agencies and nonprofits, the challenge was to find organizations that fit the Fellowship’s criteria.

When asked, Ghani stated that he and his team were firstly searching for projects that addressed actual social problems that Fellows could address. Prospective partners also needed to be able to provide real workable data to be analyzed, and needed to be mindful of DSSG’s time frame as a summer fellowship. Fortunately, many organizations were on board with DSSG—some even coming to Ghani and his team before being approached.

DSSG’s Chicago government partnerships include the Department of Innovation and Technology (DoIT), the Chicago Transit Authority (CTA), the Chicago Department of Transportation (CDOT), and the Chicago Police Department (CPD). For DoIT, Fellows are analyzing 311 call data to develop prediction models, which contribute to and enhance the city’s efforts to make smarter decisions on issues such as graffiti, potholes, and other areas. The work is also being done in partnership with Chapin Hall, a policy research center at U of C.

DSSG also has multiple transit-oriented partnerships with Chicago. They include a group assisting the CTA by developing transit planning tools that predict the implications of service changes on bus and train routes, which would further help the agency monitor wait times and overcrowding. In another project with CDOT, Fellows are aiding the launch of the City’s brand-new Divvy bike-sharing program by predicting when bike stations will be empty or full.

For the CPD, Fellows are analyzing department crime data to develop new algorithms that detect emerging crime patterns. These efforts are being done in partnership with the University of Chicago’s Crime Lab, a research group launched in 2008 that develops new and innovative violence reduction approaches using insights from basic science.

DSSG is also partnering with Cook County and the recently established Cook County Land Bank Authority (CCLBA) to build an open-source analytics tool that can assist CCLBA in policy decisions about selecting properties to acquire and develop. The tool, being built in partnership with the Institute for Housing Studies at DePaul University, will serve as a useful guide to help the County stem the effects of increased foreclosed properties following the Great Recession.

DSSG’s nonprofit partnerships, which are distributed both locally and nationally, include the Environmental Defense Fund, Lawrence Berkeley National Laboratory, North Shore University Health System, Nurse-Family Partnership, Ushahidi, and the Qatar Computing Research Institute. The public school system of Mesa, Arizona, is a partner as well.

Bringing Data Science and Social Issues Together

Ghani states that in the program’s first year, they have already learned a lot for the future. For one, numerous partnerships with a diverse set of organizations have been established, and word has been spreading for organizations to be ready and willing to produce meaningful projects and associated data in future years.

Generally speaking, many governments and nonprofits have also never fully realized the value of their data. By serving as a pro-bono partner, DSSG’s presence creates a helpful exercise for governments and non-profits to reevaluate what they’re capable of doing with their current resources.

DSSG also helps establish Chicago as a growing center for data analytics. In the near future, Ghani sees DSSG potentially serving as a catalyst for a national network of talent. He has already received inquiries from other cities and universities who are looking to Chicago’s DSSG model, and would be more willing to collaborate rather than compete with them.

At the heart of DSSG, however, is positive change—not done just by short-term projects, but by helping change the way data scientists view their role in the world. “We want to bring these students together with mentors and problems and organizations so they can not only solve important problems this summer, but start thinking about them more broadly,” Ghani said in an April interview with Chicago’s Computation Institute. “That’s the goal: to solve some concrete problems in the summer but also, long-term, create a culture where a lot of these data scientists are thinking about social problems.”

About the Author

Sean Thornton

Sean Thornton is a Program Advisor for the Ash Center's Civic Analytics Network and writer for Data-Smart City Solutions.  Based in Chicago and working in partnership with the city's Department of Innovation and Technology, Sean holds joint Masters’ degrees from the University of Chicago, in Public Policy and Social Service Administration. His work has spanned the city's public, philanthropic, and nonprofit sectors.