Abstract
Probation and parole decisions have traditionally been based upon case studies, governed by informal norms prescribing base rates, and made with overconfidence in their wisdom. Actuarial tables that statistically identify the best predictors of past violations, then convert information from them into a prediction score, have repeatedly been shown to classify offenders into future risk categories more accurately than do case study prognoses. Although such tables have been available since the 1920s, they were only widely adopted by parole boards during the 1970s with the development of decision guidelines that also take offense severity into account and that prescribe for any current case not a specific penalty but a choice within the range of punishments previously imposed for cases similar in actuarial risk and in offense severity. Application of actuarial prediction to sentencing decisions by use of guidelines analogous to those for parole has been attempted, but in practice thus far has had little impact due to the custom of determining sentences by pretrial plea bargaining.
Get full access to this article
View all access options for this article.