NCJ Number
85050
Date Published
1980
Length
224 pages
Annotation
The thesis develops and applies a mathematical (probabilistic) model to estimate the changes in the number of crimes committed in a society as a function of the level of imprisonment.
Abstract
The model is limited to the incapacitative aspect of imprisonment. It builds upon the work of Avi-Itzhak and Shinnar (1973) and Shinnar and Shinnar (1975) with the following additions: (1) a heterogeneous crime rate analysis is developed in which offenders have different underlying crime and arrest rates, and (2) the model is applied to estimate changes in the size of the prison population as well as the number of crimes. The analysis is also based upon an empirical study of criminal histories of 5,000 offenders. The empirical data are used for analyzing the criminal justice system and also offender behavior, and to estimate the conditional sanction probabilities (e.g., the probability of conviction given arrest and imprisonment given conviction) and the mean time served. Analysis indicates that the length of time an offender has been in a criminal career does not help in predicting future direction. The study concludes that crime levels are relatively insensitive to the size of the prison population. At present levels of imprisonment, about 11 percent of the crime is estimated to be averted in comparison to the benchmark level of no imprisonment. Tables, figures, and references are provided. (Author abstract modified)