The results suggest that a low-cost public program can reap big benefits. But they also support arguments that employment alone cannot resolve poverty-related ills.
Study author Sara Heller, an assistant criminology professor at the University of Pennsylvania, said the research offers one possible way to prevent violence among disadvantaged youngsters, and it's a strategy that other cities should consider. "It means adolescence isn't too late" to change destructive behaviour, Heller said.
The study was released Thursday by the journal Science.
Working with the University of Chicago Crime Lab and police data, Heller studied the city's eight-week program the year it launched, 2012, and followed participants for up to about a year after their jobs ended. They were compared with a control group of teens not involved in the program.
About 1,630 kids were involved, all from violence-prone Chicago schools. Most were black and from low-income families. They were randomly assigned to the jobs program or the control group.
Kids in the program worked 25 hours a week as camp counsellors, community garden assistants, office clerks and similar jobs. They earned minimum wage — $8.25 an hour, or an average of $1,400 for the summer, received bus passes if needed, and one daily meal. The cost to the city was $3,000 per kid.
Participants all had job mentors, adults who kept in close contact and helped them work out transportation troubles, conflicts with supervisors or family struggles that could have kept them from going to work and being successful on the job.
About half the jobs participants also got lessons from counsellors in managing social and emotional difficulties, and setting goals. This extra help didn't affect the rate of violent-crime arrests, which was about the same as the jobs-only group.
In the end, there were 43 per cent fewer violent-crime arrests among the jobs group during the follow-up than among the control group.
Adjusting for factors involved in the study design, violent-crime arrests among the jobs group totalled 37, or an average of about 5 arrests per 100 participants. By contrast, violent-crime arrests totalled 82 in the control group, or about 9 per 100 kids. The crimes were mostly assaults.
There was no difference between the two groups in arrests for crimes involving property or drugs, although rates were low.
Heller said previous research also found employment has little effect on non-violent crime. Reasons are unclear but violence often results from "interpersonal conflicts," which Chicago's program aims to help kids overcome, she said.
Patrick Owens says he had run-ins with the law and wasn't doing well in school until he got involved in the summer program two years ago. Now 19 and a freshman at Central State University in Ohio, he says job mentors helped set him straight.
They offered motivation, and the program "gave me something to look forward to," Owens said. "It made you want to work harder to just do something positive for yourself."
Robert Apel, a Rutgers University expert in the link between employment and crime, said the study results "are certainly encouraging, and worth close scrutiny."
Previous evidence on summer jobs programs is scant but suggests they don't have promising results, partly because they tend to offer more menial tasks than in the Chicago program, he said.
But Apel said what made the Chicago jobs program a success might have had little to do with the work, and may have been the mentoring.
Dr. M. Denise Dowd, an ER physician and injury prevention researcher at Children's Mercy Hospitals in Kansas City, said the mentors likely provided crucial guidance to keep kids on the job and regulate disruptive behaviour. Knowing there's an adult who cares can give at-risk kids resilience to deal with adversity, Dowd said.
Journal Science: http://www.sciencemag.org
AP Medical Writer Lindsey Tanner can be reached at http://www.twitter.com/LindseyTannerSuggest a correction