I received this interesting note from Dr. Greg Little (see yesterday’s post) explaining his research methods in more detail and discussing his findings:
Overall you present a good summary. But I can answer your questions. The study’s subjects all applied for entry into a drug treatment program (MRT) operated by the Shelby County Correction Center in Memphis, TN from 1986-1991. All were felons serving from 1 to 6 years. The control group was formed from a smaller number of individuals who were randomly excluded because of limited treatment slots. The treated subjects were randomly selected to enter…after all the subjects were placed into a pool of eligibles.
So both the study group and the control group were people who had applied to take part in a drug treatment program. That solves the problem of self-selection, in a way, making the data on the effect of the treatment more reliable, for the main difference between the two groups would be the treatment program, and only the treatment program.
It makes me wonder about the recidivism rates for offenders who didn’t try to apply for the drug treatment program, though (not that you can get a recidivism rate much higher than 94%). Were they simply not substance abusers? Were they excluded because of behavioral issues such as violence? Additionally, felons serving more than six years were excluded from the study, so we don’t know the recidivism rates for them. Undoubtedly, members of that group include the sorts of violent criminals whose propensity for recidivism is most worrying. And offenders serving less than a year weren’t counted either.
None of this is to say that the study isn’t valuable, nor that the researchers here are misrepresenting their findings. But it’s important to be aware of the difference between what a study proves and what it cannot prove. Too often, the media ignores this difference. And when the research is conducted by activist organizations with anti-incarceration agendas (not the case here), like the Pew Foundation, or the Sentencing Project, the claims they make are often extremely unreliable. At best.
Dr. Little continues:
There were no differences between the treated and control groups. There have been about a dozen prior published studies in peer-reviewed journals on these groups covering their time periods from 1 to 10 years after release. We were interested in what honestly happens to these people after 20 years of release. The local government, which we are not affiliated with, supplied the data.
You are correct that the authors (I am the senior author) are engaged in starting programs that reduce recidivism. We all make our living in criminal justice, we are all long-term professionals, and I have been in the field since 1975. All rearrests, only with minor traffic charges excluded, were collected as were all reincarcerations. The criminal justice system has always supplied misleading statistics, and that’s something we have battled for decades and have included such ethically-challenged issues in our textbooks and articles. There is a difference between what could be called “accurate” and what is “true” or “honest,” and we wanted to present a true and completely honest picture of what happens after 20 years. The data were, quite frankly, highly disappointing, but also somewhat encouraging. The real point is that there is a proportion of offenders that will return after their release no matter what we do. Right now, reducing those rearrested from 94% to 81% after 20 years is the best anyone has found. Reducing the reincarceration rate (which is rearrest, conviction, plus new sentence) is from 82% to 61%, also the best ever found. It means even using the best treatment known currently, 81% will be rearrested and 61% will still be reincarcerated. Without using that method, 94% are rearrested and 81% are reincarcerated.
The link to the original full article can be found here:
As I wrote yesterday, I don’t oppose realistic rehabilitation efforts (who would, really?). What I object to is using substance abuse as an excuse for crime, which results in untold numbers of offenders escaping punishment simply because they say they’re helpless addicts. And that doesn’t do anyone, including them, any good at all. Nor does it help to romanticize criminals, or encourage them to believe that they are victims of society, as so many rehabilitation programs do. Changing Lives Through Literature, for example, seems less about “rehabilitating” offenders than convincing them that their own convictions were unjust (see here and here).
Unfortunately, such anti-incarceration activists (who are currently in force in the Justice Department, in academic departments, and, of course, in the rehabilitation industry) never change their tune, no matter the evidence presented about the inevitability of re-offending. Their first line of defense is claiming that recidivism rates are not nearly as high as many believe. But hand them a 94% re-arrest rate, and they will say it’s proof that prison doesn’t work. If we never incarcerate anyone, the line goes, then there will be less crime (thank goodness they’re not in charge of the laws of gravity).
A few years ago, I ran into a former co-worker who attributed his ability to kick a cocaine habit to a long sentence behind bars. He never would have stuck with drug treatment, he told me, if he had not been incarcerated. Then he listed other co-workers we knew who died young. He considered himself lucky. The so-called drug war, and stiff sentencing, doesn’t get enough credit for saving lives.
What do we do with a 94% re-arrest rate? There’s no one good answer. But one thing we definitely should not do is keep pretending that all that crime doesn’t really exist.