Many offenders are released conditionally to communities in lieu of jails or prisons because, for them, the benefits of sustained social ties and community-based treatment are thought to outweigh any of those brought about by incarceration. There is reason for caution, though, as their release to some extent jeopardizes public safety. Available research, for instance, convincingly suggests a sizeable fraction of offenders enters probation yet fails to comply with release conditions. This steepens the already uphill challenges offender management and reintegration facing supervision agencies.

The underlying goal of this study is the development and validation of an instrument for informing immediate, risk-anticipated security and treatment assignments among community supervised offenders in the District of Columbia. The study examines whether probationers in the population test positive, provide a bogus specimen, or fail to appear for any drug testing event as well as whether and, if so, how often they test positive for each of the seven substances (viz., alcohol, methadone, amphetamine, cocaine, marijuana, opiates, and phencyclidine) screened by the Court Services and Offender Supervision Agency for the District of Columbia. It also examines whether offenders are ultimately convicted given an arrest for a new crime. Analyses also center on how often supervision- and drug-related violations occur as well as the probabilities and rates of ultimately terminating unsuccessfully.

These processes are estimated among a random sample of approximately 200 probationers having terminated their community sentences during the interval beginning on January 1, 2004, and ending on December 31, 2004. From well over 200 theoretically plausible predictors, this study identified a very small set that provide the agency with advance notice of the most challenging groups of offenders. This set of characteristics includes (a) the age at the time of assessment, (b) the expected length of supervision, (c) the number of substances ever used, (d) whether the probationer had ever used opiates or phencyclidine, (e) the number of weapons-related convictions, (f) the SFS-98 score, (g) the recommended sentence, (h) the impression of recidivism risk on the supervising CSO, and (i) local rates of arrests for drug-related and public order crimes.