Low-intensity community supervision for low-risk offenders: a randomized, controlled trial

The Philadelphia Low-Intensity Community Supervision Experiment provides evidence on the effects of lowering the intensity of community supervision with low-risk offenders in an urban, US county community corrections agency. Using a random forests forecasting model for serious crime based on Berk et al. Journal of the Royal Statistical Society, Series A, 172(Part 1), 191-211, 2009, 1,559 low-risk offenders were identified and randomly assigned to either standard or reduced frequency of mandatory office visits. Treatment as assigned was substantially delivered at 4.5 probation visits per year versus 2.4, for as long as offenders remained on active probation or parole. In a one-year follow-up for all cases, outcomes examined were the prevalence, frequency, seriousness and time-to-failure of arrests for new crimes committed after random assignment was implemented. No significant differences (p=.05) in outcomes were found between standard and low-intensity groups. Non-significant differences for offense seriousness favored the low-intensity group. We conclude that lower-intensity supervision at the tested level of dosage can allow fewer officers to supervise low-risk offenders in the community without evidence of increased volume or seriousness of crime.