John List has been chief economist for UBER and for Lyft. He's also run experiments for United, the United Way, and the Chicago schools. In his new book, The Voltage Effect, he talks about a number of these experiments but what really comes through is how much fun he is having.
Now, I can imagine a lot of ways of doing this--a traditional approach, for example, would be to look at spending decisions, credit card reports, criminal records, marriage lengths, number of children. One can imagine a lot of potential correlations to be found using big data. That would be pretty clever but here's what List and Levitt did:
Moral of the story: if you ever find a wallet in Chicago, return it. You never know who is watching.