Posts Tagged ‘research’

In a Diversion and Rehabilitation course I’m currently teaching, we recently compared and contrasted Martinson’s “Nothing Works” doctrine with the more recent “What Works” approach that’s steadily gaining traction in the criminal justice system. It was incredibly rewarding to see students grasping the power and promise of research-based crime reduction approaches, as opposed to the outdated notions of mass incarceration and the just deserts philosophy that had predominated over the last couple of decades.


Coincidentally, I also read Dylan Matthews’ post yesterday that summarized twelve strategies currently being used to reduce crime and violence around the country.  It’s a must-read for anyone who’s interested in the diversity of ways we can effectively deal with crime in our communities. Conspicuously absent from the list are the typical calls for  longer prison sentences and harsher treatment of offenders that had been offered up as the best approach in the past. Instead, systemic and data-driven interventions, such as lead abatement, early education, hot-spot policing, and providing training and support to parents are emphasized.

Further research on each of these is certainly warranted, but they all hold the promise for a more peaceful, safer society. It’s time to move the balance of our crime reduction efforts more toward prevention, which is cheaper, safer, and simply better for society as a whole.

Because of their complexity, evaluating criminal justice interventions is a particularly challenging endeavor. Attempting to isolate and control for all of the variables involved is a real barrier, and the reported outcomes tend to suffer as a result. This is just as true for court-based interventions as it is for other any other area of the system. Single-site evaluations of a particular approach, such as a specialty court, can easily be weakened by threats to validity caused by variations in implementation.


A recent study released by the Center for Court Innovation about the effects of New York’s Domestic Violence courts took a different approach that overcomes some of the threats to validity that can limit the applicability of results. Instead of focusing on a single case, researchers examined outcomes across a system of 24 Domestic Violence courts.  They evaluated a matched sample of offenders, half who participated in a traditional court, and half who participated in a Domestic Violence court in the same system.

The results are subtle and nuanced, and provide some useful insights for jurisdictions looking to implement such a court.  Here are the authors’ conclusion:

This study of New York’s domestic violence courts demonstrated a modest positive impact on
recidivism among convicted offenders, though not among all defendants. The study did not
detect a significant overall impact on conviction rates or incarceration sentences, although the
domestic violence courts produced significantly more punitive outcomes (higher conviction and
incarceration rates) for male offenders. Consistent with previous research, the study suggests that
not all domestic violence courts seek the same goals, follow the same policy model, or achieve
the same impacts. This study also found that those domestic violence courts that prioritize
deterrence and that both prioritize and implement specific policies to sanction offender
noncompliance, while also addressing the needs of victims, are most effective in reducing
recidivism. Knowing that modest recidivism reductions are possible can set the stage for future
research and development on promising practices that offer the prospect of maximizing the
benefits of these specialized courts (emphasis added).

As in other criminal justice interventions, implementation appeared to be key to improving outcomes. It’s not just the presence of a Domestic Violence court that makes a difference, it’s the specific policies and practices, as well as the philosophical orientation of the participants that leads to positive results. In other words, the details matter.

Are you aware of any other outcome studies like the above and, if so, how do the results compare?