A great deal is known about the "risk factors" that characterize people who are involved with crime, violence and drug use, including individual-level attributes such as gender, age and race, family-level characteristics such as poverty or a history of violence, and community-level attributes such as high rates of residential instability.
Much less is known about the most cost-effective policy strategy for reducing crime and violence. Just because family poverty is correlated with delinquency does not mean that providing poor families with more income will reduce delinquency. It is possible that whatever factors allow parents to succeed in the labor market might also affect their ability to be successful parents as well, so that the correlation between poverty and delinquency might simply reflect the influence of other factors that influence both outcomes. Moreover addressing persistent problems of poverty and social inequality will be a daunting, long-term task. The fact that so many people who grow up in disadvantaged circumstances do not become involved with crime or violence suggests the existence of a range of other proximate risk (or protective) factors that could also be productive targets for policy intervention.
In what follows we provide a brief review of the data sources that are available for learning more about patterns of crime, violence and risky behavior in the U.S., and then provide a brief summary of what is currently known about criminal justice, education and social policy levers capable of reducing anti-social or risky behavior. Ours is necessarily a very selective review - we focus on providing short, accessible summaries of some of the major intervention approaches that have been tried, with a special focus on interventions that have been rigorously evaluated using randomized experiments, or "natural" experiments where receipt of the intervention is affected for some people or places and not others due to changes in state or local laws or other means. We hasten to add that this summary reflects the views of the Crime Lab directors; on a topic this complicated, there will inevitably be disagreement even within our team of Crime Lab affiliates.