Testing strategies for preventing violence and crime | MIT News
J-PAL North America funds randomized studies to evaluate crime-prevention programs.
J-PAL North America, a research center at MIT, has announced that it has awarded grants to fund randomized evaluations focused on employing behavioral science insights to prevent crime and violence. The grants were awarded to Anuj Shah and Aurélie Ouss from the University of Chicago and Jennifer Doleac and Benjamin Castleman at the University of Virginia.
“Behavioral science is generating intriguing insights into interventions that can nudge people to make better decisions,” said Quentin Palfrey, J-PAL North America’s executive director. “As policymakers grapple with surging prison populations and high rates of recidivism, rigorous research is needed to evaluate what works to prevent violence and improve our criminal justice system.”
The announced grants, which are made possible by the Robert Wood Johnson Foundation, are designed to advance our understanding of how the principles of behavioral science can be used to improve public policy in the area of criminal justice.
Shah and Ouss will evaluate whether an app designed to lead at-risk youth to participate in safe activities can help them avoid dangerous default situations and behaviors. “We know that many youths do not plan their activities in advance,” Shah and Ouss state. “Instead, they default into habitual activities, many of which are risky and make criminal or violent acts more likely, even when those acts were not the original intention.” The researchers will make an app called ChiPlan available to randomly assigned disadvantaged youth in Chicago and test whether its use leads to changes in arrests, victimization, and program participation with the Chicago Department of Family and Support Services.
Doleac and Castleman will focus on reducing recidivism — the tendency of individuals to become repeat offenders — testing a tablet-based re-entry module created to strengthen inmates’ transition back into society. Before release, the module will help inmates create a personalized transition plan. Post-release, the researchers will provide ongoing information to help keep former inmates on track. As Doleac and Castleman point out, “[the U.S.’s] high recidivism rate signals our failure to help formerly incarcerated individuals build stable lives after prison. This project aims to provide evidence on a highly-scalable, technology-based, behavioral science strategy to improve re-entry outcomes.” Their experiment is modeled on an approach that has proven successful in other contexts, particularly postsecondary education.
These projects will join a growing body of research pioneered by J-PAL affiliates on what works in the fight against crime and violence. Among the most successful tested strategies are cognitive behavioral therapy programs implemented in schools and a juvenile detention center — found to reduce violent crime arrests, dropout rates, and recidivism — and an inmate re-entry program that provided social services and subsidized work — found to reduce the likelihood of re-arrest. J-PAL North America is also working with the Louisville Department of Corrections to design an evaluation of an innovative pay-for-success initiative that provides treatment to individuals with substance abuse disorders immediately upon release from jail.
J-PAL North America is the regional office of the Abdul Latif Jameel Poverty Action Lab, a research center at MIT that seeks to reduce poverty by ensuring that policy is informed by scientific research and works to improve social programs by running randomized controlled trials, disseminating policy lessons, and building the evaluation capacity of governments and non-profits. The Robert Wood Johnson Foundation is committed to creating a continuous cycle of research, evaluation, and learning to address some of the most pressing societal challenges in the United States.Reprinted with permission of MIT News
Share this article: