Hierarchical Bayesian Analysis of Arrest Rates

• NCJ Number: 174172
• Journal: Journal of the American Statistical Association Volume: 93 Issue: 444 Dated: December 1998 Pages: 1260-1270
• Author(s): J Cohen; D Nagin; G Wallstrom; L Wasserman
• Date Published: 1998
• Length: 11 pages
• Grant Number(s): 91-IJ-CX-K015; DMS-9005858; DMS-9357646; RO1-CA54852-01
• Agencies: NIJ-Sponsored

Annotation

This article describes a Bayesian hierarchical model as the basis for calibrating the crimes avoided by incarceration of individuals convicted of drug offenses compared to those convicted of nondrug offenses.

Abstract

Two methods for constructing reference priors for hierarchical models both led to the same prior in the final mode. The study used Markov chain Monte Carlo methods to fit the model to data from a random sample of arrest records of all felons convicted of drug trafficking, drug possession, robbery, or burglary in Los Angeles County in 1986 and 1990. The study focused on comparative analysis because one of the opportunity costs of incarcerating drug offenders is reduced capacity to incarcerate individuals convicted of other offenses. The value of this formal analysis, as opposed to a simpler analysis that does not use the formal machinery of a Bayesian hierarchical model, is to provide interval estimates that account for the uncertainty due to the random effects. Figures, tables, references