One of the more promising areas of inquiry at the intersection of education and crime reduction is "soft skills." In education, it might be traits like persistence, in the research of James Heckman; in crime-reduction, it might be "non-cognitive" skills like self-control, as in the promising results from the cognitive-behavioral-therapy approach applied through the Becoming a Man initiative. They're two sides of the same coin—explicitly teaching life skills that kids are expected to pick up implicitly as they grow up.

A recent study, reminiscent of the Becoming a Man study—and by the same lead author, the University of Pennsylvania's Sara Heller—brings very positive findings: a 43 percent reduction in violent crime among disadvantaged high school youth over a 16-month span, well after the skills program ended. And it suggests that there are different, equally effective routes to imparting those skills.

Heller's study, recently published in Science, examined One Summer Plus, a summer-jobs program open to students in high-violence Chicago public high schools. On average, the kids were 16 going on 17, with a C average, and having missed 29 days of school. Twenty-two percent had been arrested. In short, not lost kids, but "on the cusp," to use Heller's words.

They were put through a straightforward summer jobs program: students got paid minimum wage to work engaging jobs—camp counselor, aldermanic assistant, community-garden work—with the assistance of a job mentor. In other words, it's not just digging a hole to fill it back in.

"Youth are smart," Heller says. "They know when you're making work for them just for the sake of doing work. And you must imagine that that's a lot less rewarding, and making you feel a lot less responsible and proud of what you're doing, than if you're really acting as a camp counselor and a role model for younger youth." The students worked part-time for eight weeks over the summer.

But the study was given an interesting twist. Some of the students split their jobs with a "social-emotional learning" curriculum, along the lines of the Becoming a Man program. In essence, the study set up something of a competition, between a straightforward summer-job program and one mixed with explicit instruction in life and emotional skills.

And they both worked. Among the treatment group, violent-crime arrests fell by 3.95 arrests per 100 youth, and the difference between the two treatment groups was almost nil. Plus the effects were strongest five to 11 months after the program ended, suggesting a lasting effect, at least in the medium-term—and that the results were not merely the result of keeping them off the streets for a couple months.

"The fact that those two versions of the program have about the same effect, suggests to me that there may be interchangeable strategies for having this kind of impact on youths for teaching them the skills they need to avoid violent crime," Heller says. "Which is very exciting, because if you look at the rigorous studies that are really convincing on how we reduce violence, we don't have very many answers."

The idea is that working a job, with the aid of a mentor, imparts similar skills as a curriculum teaching self-control and conflict resolution, for reasons that should be obvious to anyone who's held a job, or at least managed to keep one.

"I heard a lot in talking to the program providers, and the employers, that even the jobs themselves are trying to develop some of those same skills that the social-emotional learning is doing," Heller says. "Talking to one of the employers, he told me that the biggest problem he has with young employees is how defensive they are. They show up, they say 'you have to wear closed-toes shoes to work,' and they snap at you, because they're insulted, they don't know how to take constructive criticism or separate that. They feel very defensive about that."

Defensiveness and the inability to step back from a conflict are hardly unique to kids in some of the city's most dangerous neighborhoods. But the ramifications are much worse. "You could imagine that transferring over to not just interactions with employers, but also interactions with peers that would otherwise turn into throwing a punch," Heller says.

One Summer Plus, however, had no effect on other types of crime: property crime, drug crime, and other arrests. The effects were limited to violent crime, an interesting result given how substantial the effects were on that category. But it makes sense in the context of social-emotional/cognitive education, which is focused on personal interactions.

"The bulk of that violent-crime category is assault," Heller says. "Assault happens, often, just because an argument over something stupid blows up. The fact that you see this effect on violent crime, but not things that don't necessarily involve conflicts with people—breaking into a house, or carrying marijuana on you—that's part of what informs my thought about the mechanisms about the program."

With the cooperation of the city, Heller is continuing to delve into the mechanisms that make One Summer Plus work. She's also studying CBT treatment in the realm of juvenile detention—kids farther over the cusp than the One Summer Plus group.

"I do hope that other cities will look at this as an example," Heller says, "and do experiments of their own, to be able to answer exactly questions like that: how much does the type of job matter? How much does the job mentor matter? How much do the peers matter? There's a lot of unanswered questions that I hope other cities will work to answer."