An impact assessment of machine learning risk forecasts on parole board decisions and recidivism

The Pennsylvania Board of Probation and Parole has begun using machine learning forecasts to help inform parole release decisions. In this paper, we evaluate the impact of the forecasts on those decisions and subsequent recidivism.

Saved in:  
Bibliographic Details
Main Author: Berk, Richard (Author)
Format: Electronic Article
Language:English
Published: 2017
In: Journal of experimental criminology
Year: 2017, Volume: 13, Issue: 2, Pages: 193-216
Online Access: Volltext (Resolving-System)
Journals Online & Print:
Drawer...
Check availability: HBZ Gateway
Keywords:
Forecasting
Multinomial logistic regression
Parole·Machine learning
Recidivism
Regression discontinuity design
Description
Summary:The Pennsylvania Board of Probation and Parole has begun using machine learning forecasts to help inform parole release decisions. In this paper, we evaluate the impact of the forecasts on those decisions and subsequent recidivism.
ISSN:1572-8315
DOI:10.1007/s11292-017-9286-2