NCJ Number
75952
Journal
International Journal of Comparative and Applied Criminal Justice Volume: 4 Issue: 2 Dated: (Winter 1980) Pages: 203-211
Date Published
1980
Length
9 pages
Annotation
This paper recommends the use of dynamic-analytic techniques to select and verify outcome measures in the criminal justice system and to consider changes and predict their future impact in the context of the total system.
Abstract
Much of crime data is inadequate because data are generally limited to one particular stage of criminal processing, measurements are in the form of summary statistics, the data are specific to one agency, and the unit of count changes with different organizational structures. Thus, the existing sources of data frequently measure only one temporal point and fail to reflect the dynamic aspects of criminal processing. Similarly, many of the analytic techniques used for evaluation are static and do not reflect the dynamic aspect of the system. Data systems such as Standardized Crime Reporting System (SCRS), Attribute Base Crime Reporting System (ABCR), and Prosecution Management Information System (PROMIS) should be used to alleviate these problems because of the transactional nature of the data base. The changes that occur at one stage of the system can be examined and evaluated with respect to changes that are likely to occur at subsequent stages and the amount of time required to process offenders may be determined. These data are also useful for monitoring repeat offenders and for examining differential outcomes that occur at various stages. In conjunction with such data systems there is a necessity to apply dynamic-analytic models. Static techniques may be applied longitudinally at various decision points in the system to determine the differential probabilities of results from that decision point, given the range of causes and preceding system processing. Such techniques might include Multiple Regression and Predictive Attribute Analysis. A merger of such static techniques, preformed at several stages of the system under scrutiny, might be used to form an integrated model which is dynamic in form. In addition to dynamic modeling using static techniques, more traditional models might be useful - such as Markov chains, transition matrices, and other stochastic processes. While these techniques do not represent a new methodology in themselves, their application to transactional data will provide information not available before. A matrix showing joint effects of organizational units and time upon juvenile justice processing is described, footnotes and over 20 references are included.