Wednesday, August 09, 2006

What Works and How do we know it works

The National Association of Sentencing Commissions (NASC) 2006 meeting concluded yesterday in the City of Brotherly Love with an interesting visit to Eastern State Penitentiary. This prison began as a Quaker-inspired vision of strict but humane incarceration designed to instill a penitent attitude among its charges, but ended rather sadly, with far more examples of inhumane treatment before eventually closing in 1971.

NASC members, themselves in a penitent mood, then retreated to a nearby tavern for a period of reflection.

The NASC theme, Keystone of Sentencing: Balancing Fairness and Costs, was actually dominated by a related but distinct topic, Evidence-Based Policy in the area of sentencing and corrections. We will no doubt spend several postings on this important topic, and welcome your comments. Perhaps a few introductory comments and definitions are in order.

Evidence-based policy analysis attempts to answer the question of “what works” in policy and program design through some form of data-driven analysis. Perhaps the best known and most successful example of evidence-based policy analysis is the Cochrane Collaboration in health care. It is widely recognized as the single best source of evidence on the effectiveness of health care and medical treatments. It got this reputation by maintaining a fierce reliance on strict standards of what constitutes “evidence.” Meaning multiple experimental trials, and not just by the interested parties such as drug manufacturers.

Which, of course, brings us to the world of sentencing and corrections policy making, where we largely ignore the standards of policy evidence, and what stands for evidence is either 1) a personal testimonial by a respected person or organization or 2) a politically-motivated claim which may simply ignore the need for evidence and relies on ideology with a thin veneer of selective facts.

But many folks are aiming at a higher standard in the service of discovering “what works” in sentencing and corrections policy. It is probably fair to say that correctional programs and policy are further along in this area (See this useful publication on community corrections, for example).

That brings us to sentencing policy, and one particular panel at the recent NASC conference. Steve Chanenson of Villanova U., Marc Miller of U. of Arizona, and Ronald Wright of Wake Forest U., led a panel discussion on Monday entitled “Sentencing Information Exchange.” Their intention, as I understand it, is twofold. First, they believe that current efforts at web-based sentencing information systems such as those underway in Scotland and Oregon may allow practitioners (judges, defense attorneys, prosecutors, etc.) better decision tools, and these information systems should be better known so that the innovations may be scrutinized and diffused elsewhere, as appropriate. Second, they are advocating cooperation among diverse jurisdictions to share data on their sentencing practices, using a common format for charges, prior record, etc., as a means to begin the process of scientific discovery regarding what works in sentencing policy (For example, what forms of sentencing guidelines, if any, work to reduce crime or reduce unwarranted disparity). Needless to say, I find this initiative to be long overdue and very exciting.

Great start!

No comments: