By all accounts, Rayid Ghani's data work for President Obama's reelection campaign was brilliant and unprecedented. Ghani probably could have written a ticket to work at any company in the world, or simply collected speaking fees for a few years telling companies how to harness the power of data like the campaign did.
But instead, Ghani headed to the University of Chicago to bring sophisticated data analysis to difficult social problems. Working with Computation Institute and the Harris School of Public Policy, Ghani will serve as the chief data scientist for the Urban Center for Computation and Data.
Before the campaign, Ghani said that he found it difficult to use his data skills for social good. There were plenty of corporate jobs that wanted people who could do analytics, but not many non-profits. "The reason I got on the campaign is that I was trying to connect things I cared about with what I was good at. I wanted to use analytics and data for social problems," he said. "But when the campaign was done, I was back in the same place."
Crunching numbers at Google, Facebook, or in finance, data scientists feel they are doing something important because their analyses can change millions of peoples lives or send millions of dollars caroming through the markets. "Improving search results or optimizing ad clicks, they are not improving lives but they are still having an impact," Ghani said. But they're not doing much for the non-corporate world.
Ghani hopes that his work at the University of Chicago can give students a way out of the impact/do-gooding conundrum. Towards that end, he's running the Eric and Wendy Schmidt Data Science for Social Good Fellowship, which is pairing up 40 computer science fellows from around the country with non-profits who have difficult data problems.
What kinds of problems are we talking about? "One of the problems that we're looking at is college admissions," Ghani told me. "You've got students with very high potential who are at risk of not applying to college or who apply to much worse colleges than they could get into... You want to be able to look at students and see who is at risk of this behavior." Perhaps there are early indicators of this kind of behavior lurking within an otherwise excellent academic record. If the students can be identified, they can be channeled into programs that help them.
It may not be as exciting as the Obama campaign, but it might go a small way to solving the problem so memorably identified by early Facebook employee, Jeff Hammerbacher. "The best minds of my generation are thinking about how to make people click ads," he said. "That sucks."