<< Exclusion of Evidence for Private Privacy Violations | Main | News Scan >>

Rehabilitation, Evidence-Based Practices, and the Selection Problem

Sasha Volokh has a couple of posts (here and here) at the Volokh Conspiracy on the question of whether faith-based prisons actually "work," i.e. produce a lower recidivism rate than the same prisoners would have had in regular prisons.  The difficulty in answering that question is the "selection bias" problem, as indicated by the title of the second post, "What if faith-based prison programs just attract better prisoners?"

"Evidence-based practices" has become a buzzword in corrections, but for the reasons Volokh points out, much of the "evidence" is nearly worthless.

After yesterday's introduction to the topic, today I'll talk about how the self-selection problem makes any evaluation of faith-based programs with regular programs problematic. I'll illustrate with some of the most problematic studies, which show the self-selection problem in its most naked form. I'll then show some of the better studies, which control for certain important variables, but I'll explain why even those are inadequate to solve the self-selection problem.
The problem is not by any means limited to faith-based programs.  It permeates the whole field.  As long as the "treatment group" and the "control group" are selected in a way that makes them different in their attitude toward going straight, the study is essentially garbage.  Random selection and large sample sizes are necessary to valid studies, but random selection is a tough sell.  Can we really assign people to rehabilitation programs in a lottery, denying the guy who wants it and assigning the one who doesn't give a damn?

Leave a comment

Monthly Archives