Transaction Search Form: please type in any of the fields below.
Date: April 29, 2024 Mon
Time: 11:18 pm
Time: 11:18 pm
Results for offender classification (pennsylvania)
1 results foundAuthor: Barnes, Geoffrey C. Title: Classifying Adult Probationers by Forecasting Future Offending Summary: Random forest modeling techniques represent an improvement over the methodologies of traditional risk prediction instruments. Random forests allow for the inclusion of a large number of predictors, the use of a variety of different data sources, the expansion of assessments beyond binary outcomes, and taking the costs of different types of forecasting errors into account when constructing a new model. This study explores the application of random forest statistical learning techniques to a criminal risk forecasting system, which is now used to classify adult probationers by the level of risk they pose to the community. The project principally focused upon creating a risk prediction tool within a partnership between University-based researchers and Philadelphia’s Adult Probation and Parole Department (APPD). This report details the model building process, including an explanation of the random forest procedures, and sets out the issues in data management and in policy considerations that are associated with creating a prediction tool. The importance of developing strong researcher-practitioner partnerships, especially with regard to tailoring the prediction tool to real-world concerns, is also considered. The prediction models developed as a part of this project have been used since 2009 to assess all incoming probation and parole cases. Each risk prediction is then used to assign offenders to risk-stratified supervision divisions. This report discusses the development and accuracy of the three generations of models that have been employed to make these predictions, as well as the salient features of each iteration. For the most recent version of the model, the influences of major predictor variables are discussed, with a focus on those that were most powerful in the Philadelphia sample. Additionally, the predictions of the three models are also validated using a sample of cases from 2001, a cohort not used to build any of the prediction models. The long-term offending patterns of these 2001 offenders, with regard to their assessed risk level, are also considered. Finally, suggestions and step-by-step instructions are offered for practitioners seeking to build a similar prediction instrument for use in a wide variety of criminal justice settings. As a matter of policy, criminal sanctions that are encompassed under the umbrella of community corrections have become an increasingly prevalent punishment option. Given resource constraints, as well as concerns about public safety, statistical risk assessment tools – such as the one developed here – have begun to take increasingly prominent roles in determining levels of supervision. This report, and the partnership supporting it, highlights the promises and potential of the methodology. Details: Report to the U.S. National Institute of Justice, 2012. 64p. Source: Internet Resource: Accessed June 27, 2012 at: https://www.ncjrs.gov/pdffiles1/nij/grants/238082.pdf Year: 2012 Country: United States URL: https://www.ncjrs.gov/pdffiles1/nij/grants/238082.pdf Shelf Number: 125408 Keywords: Offender Classification (Pennsylvania)ProbationersRecidivismRisk Management |