Guideline-Based Justice: Prediction and Racial Minorities

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Interpretable classification models for recidivism prediction (2015, background): “Some studies have shown that race is important for accurate recidivism prediction (Petersilia and Turner 1987, Berk 2009).”
Actuarial Sentencing: An “Unsettled” Proposition (2013, background): “Petersilia and Turner ( 1987 ); the authors found that omitting factors that are correlated with race from a model to predict recidivism reduced the accuracy of the model by 5-12 percentage points (cited in Oleson, 2011 ).”
Statistical Risk Assessment: Old Problems and New Applications (2006, background): “Concern over the legitimacy of specific predictor items, or even of entire risk assessment scales, is not new (Fisher & Kadane, 1983; Goldkamp, 1987; Goldkamp, Gottfredson, Jones, & Weiland, 1995; S. D. Gottfredson & Gottfredson, 1986; Petersilia & Turner, 1987; Tonry, 1987).”
The Role of Offender Risk Assessment: A Policy Maker Guide (2010, unknown): “Even with large data sets and advanced analytical techniques, the best models are usually able to predict recidivism with about 70% accuracy ( Petersilia & Turner, 1987 )—provided it is completed by trained staff.”
A Multiple Models Approach to Assessing Recidivism Risk (2002, background): “The risk factors in this study include 29 measures typically used by criminal justice organizations to make recidivism predictions (Petersilia & Turner, 1987).”