The Second of Three Ideas that could Transform Corrections...Keeping Score

June 16, 2009 Madrid
If you have time, I recommend reading the first 3 short essays as background for this essay. But I will try to make it self-contained for the busy reader.

We've already discussed the enormous costs of running the correctional system and the relatively poor success that we get for our money. This article deals with an idea for increasing the effectiveness of our prisons in terms of actual rehabilitation.

Keeping Score - you can't get a result if you don't measure for it. Actual rehabilitation in prisons is such a hopeless subject that the professionals don't really want to discuss it. Wardens running our prisons are concerned with 1) staying inside the budget; 2) keeping staff and inmates from getting hurt; and 3) staying out of the newspapers. Changing the inmates values and goals is not a part of the plan - it is considered impossible, or nearly so.

It must be impossible; after all, didn't we make the effort back in the 70's and 80's when we threw our most powerful behavior altering tool, psychology, into the prison system. And psychology simply didn't make the difference that we had hoped for. So with shrinking budgets and ever more new prisons to build, we fell back to mere "warehousing" of prisoners - just holding them off the streets so that they couldn't commit crimes.

Once "warehousing" became the dominant operating mode, then other changes emerged from that (or perhaps just returned to an earlier mode used in the days before psychology was tried). To prevent staff injuries, the relationship between staff and inmates became more formal and much more separate. As staff became more and more established as guards rather than counselors, then the two distinct cultures within the prison became hardened - each with a negative definition of the other.

In the staff culture, the older guards trained the newer ones to be wary of the convicts and promoted the notion that "there is nothing that can be done for these people". Because staff were no longer involved in counseling, their educational requirements were reduced along with their pay (given the budget crises). As a result, the staff culture became a bit less idealistic. Being a guard was a job, not a calling.

In the inmate culture, the guards just became the unsmiling face of the non-Outlaw culture. As Outlaws will do, the inmates hoarded every injustice as proof of the inherent antagonism of the non-Outlaw culture. The result has been prisons where the Outlaws primarily relate to other Outlaws. Healthy interaction between staff and Outlaws became the exception.

Now let's get to the heart of the second idea: we don't really measure effectively the success and failure rate of each of our institutions. At the national level, we collect data on individual criminals and their arrests. This is done by the Bureau of Justice Statistics. There has been an effort to continually improve this record keeping, especially since the data has been computerized. It would not be all that hard to add a few new fields to the database and track the institutions where the criminal has spent time.

The reason for doing this would be to be able to rank all the correctional institutions in the nation in order of their success in releasing inmates who remain free of arrests. If such a list were published every six months, each of the wardens could see the comparative effectiveness of their own institutions. Let's imagine that there is some prison in Huntsville, Alabama that happens to be Number One in effectiveness. The other wardens would have to ask, "What are they doing that we aren't doing?" That is a question that isn't being asked today!

I assert that effectiveness in treating Outlaws will turn out to be based on the quality of non-Outlaw relationships available to the Outlaw as well as the availability of socialization to teach the Outlaws how to function in the non-Outlaw society. But once we start keeping accurate measurements of success, we may all be surprised at what works. Until we have the measurements, we are really just guessing.

The beauty of this idea is that it doesn't really cost very much. It doesn't add new bureaucracy. We already collect data - there would be a little clerical effort involved in reporting what inmates are in what institutions, but the amount of time required is fairly trivial. This is the heart of this second idea.

Now the correctional professionals reading this will at this point be eager to point out that any given inmate will probably spend time in a variety of institutions during his criminal career. It can legitimately be questioned how responsibility for the criminal's success or failure could be fairly apportioned between the various institutions.

Further, there are a variety of types of institutions ranging from open forestry camps to maximum security prisons. The forestry camps get relatively low-risk inmates, perhaps convicted of lesser crimes. The maximum security prisons get the members of prison gangs, violent and unsocialized individuals, and inmates convicted of serious crimes. How can a maximum security prison be forced to compete fairly with a forestry camp? The maximum security prison would expect a much higher failure rate given the pathology of it's inmates.

These two problems illustrate the difficulty of fairly comparing the effectiveness of different institutions. Difficult, but not a sufficient reason to abandon the idea. I assert that an enterprising graduate student in criminal statistics might very well take on as a PhD. project the development of an algorithm that would apportion responsibilty to each of the institutions that an inmate passes through. And it would clearly be necessary to establish classes of institutions so that maximum security prisons are fairly compared with other maximum security prisons and not with forestry camps. The algorithms used could be continually improved over time, but of course would need to be insulated to ensure that they are not subject to political pressures.

I remember as an undergraduate in Sociology, the Western Electric Study in which it was discovered that the mere act of measurement alters the behavior of an institution. How can we go on throwing 80 BILLION dollars a year at a system for which we have no measurement of effectiveness?