February 15, 2019, John Wetzel - National Institute of Justice
It sounds embarrassingly simple: Stop doing things that don’t work and do more of the things that do. Yet we blindly pour millions of dollars into corrections programs and practices without demanding improved outcomes like reduced violence or recidivism. When you consider the stakes of corrections work and the sheer volume of the lives we impact, I see basing decisions and policies on research as the only means to go about responsible practice.
At the Pennsylvania Department of Corrections, research has become part of the fiber of who we are as an organization. In 2014 we began a long-term partnership with New York University’s BetaGov, which supports practitioner-led research in the public sector. This partnership has been key for scaling research within our organization. With BetaGov, we have completed and published more than three dozen trials, with hundreds more studies still in progress.
Promoting Trial, Celebrating Error The only way to improve outcomes is to try something, measure it, and scale it if evaluation shows that it is promising. If it’s not effective, we figure out why, adjust, and try again. A study that shows absent or negative effects isn’t a failure — we learn as much from these studies as we do from studies that show what does work. When we saw mixed outcomes from the research trial of an in-prison swift, certain, and fair (SCF) program, it forced us to look at the policy and design to understand why. We’re now adapting a more consistent policy that builds on strengths in areas of the program that showed positive outcomes.
In Pennsylvania we still struggle with drugs coming into our facilities, and we have tested various ways to curb this problem. We experimented with sending letters to family members to warn them of the consequences of bringing drugs into our facilities. It sounded like a good idea and I thought it would work. But when we measured outcomes, the data told us that the letters didn’t have a positive impact on drugs in our facilities — we actually saw a slight increase in drugs after this trial. Given this result, we won’t waste money on that program. Along a similar vein of experimenting, we tried a new approach to drug interdiction that involved phone monitoring and led to the biggest drug bust in the history of the Pennsylvania Department of Corrections.
Keeping an Open Mind In Pennsylvania, we’ve been willing to study everything. We did a study on aromatherapy in restrictive housing units. I thought this study was ridiculous, but the data told us that aromatherapy does reduce violent incidents, and it’s cheap, so this is something we’re taking to scale. Another study with results that surprised me involved providing more colorful bed linens for inmates. We seem to enjoy especially good outcomes in trials where we’ve introduced additional choice to inmates.
Many of the ideas for studies we carry out come from our line staff. I spent nine years of my career as a corrections officer, but sitting at the top of an 18,000-person agency, it can be difficult to see the on-the-ground implications of policies and protocols. In conducting research, it’s been important to find and implement a mechanism to consistently get input from line staff in a way that is meaningful and measureable. If there’s a complicated problem that we have to solve as an organization, we involve as many staff as we can in the solution. Directors of departments of correction come and go, but ideas with agency-wide buy-in will stay.