The measure of success in both these cases has been whether or not program graduates
re-offend within two or three years (depending on the study). In the case of “scared
straight” programs, this seems reasonable. Re-offences later than that may have little to
do with the program failing and more with other, new circumstances. There is, however,
one significant weakness with the measure. A re-arrest does not necessarily mean an
offense has been committed. A known juvenile delinquent, for example, may be arrested
simply as being a “usual suspect.” But it is hard to find a better measure. Self-reports of
criminality are notoriously unreliable, although useful when comparing two groups who
are asked the same questions (they are therefore useful in evaluating the success of
properly designed studies). Nor do all crimes lead to convictions, and, especially in the
case of juvenile crime, most of the parties are concerned to prevent the case getting that
far. Convictions are therefore a less-than-satisfactory measure of recidivism. Re-arrests,
therefore, may present the best generally available measure of the success or failure of
rehabilitation programs. In the case of boot camps, however, there are other factors to
bear in mind.
The avowed purpose of the er offenders, around two-thirds re-offending12.