A Multiple Models Approach to Assessing Recidivism Risk

By Eric Silver, Lynette Chow-Martin. 2002

How researchers see this paper

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

1. Cynthia Rudin
2. Berk Ustun
3. Jiaming Zeng
4. Yuan-yuan Liu
5. Tong Wang
6. A. Piehl
7. Min Yang
8. S. Bushway

Citations in the corpus

• Interpretable classification models for recidivism prediction (2015, methodology): “Stalans et al. (2004) also found that decision trees are better than logistic regression for prediction of violent recidivism, and Silver and Chow-Martin (2002) used trees as a meta-learner over multiple existing recidivism models.”
• Forecasts of Violence to Inform Sentencing Decisions (2013, methodology): “Other approaches, such as the ‘‘Chi-Squared Automatic Interaction Detector’’ (CHAID), can be formally very different despite the use of tree-structure visualizations and some earlier applications of the sort emphasized in this paper (Silver and Chow-Martin 2002; Banks et al. 2004).”
• A Comparison of Logistic Regression, Classification and Regression Tree, and Neural Networks Models in Predicting Violent Re-Offending (2011, methodology): “The Iterative Classification Tree (ICT) approach was developed by Steadman et al. (Banks et al. 2004; Monahan et al. 2000, 2005, 2006; Silver and Chow-Martin 2002; Silver et al. 2000; Steadman et al. 2000).”
• Is It Who You Know, or How Many That Counts? Criminal Networks and Cost Avoidance in a Sample of Young Offenders (2010, background): “The outcome of the partitioning process is the creation of more or less homogeneous groups with regard to the dependent variable (Silver & Chow-Martin, 2002).”
• Identifying Three Types of Violent Offenders and Predicting Violent Recidivism While on Probation: A Classification Tree Analysis (2004, background): “Studies have found that Iterative Classification Tree (ICT; Monahan et al., 2000; Steadman et al., 2000) and standard CTA (Silver et al., 2000; Silver & Chow-Martin, 2002 ) compared to logistic regression has higher accuracy in classifying offender populations into low- and high-risk groups on viol”
• The Inextricable Link Between Age and Criminal History in Sentencing (2007, background): “Furthermore, sentencing on group characteristics may lead to fewer errors than leaving discretion in the hands of individuals (Silver & Chow-Martin, 2002).”