skip navigation

PUBLICATIONS

Register for Latest Research

Stay Informed
Register with NCJRS to receive NCJRS's biweekly e-newsletter JUSTINFO and additional periodic emails from NCJRS and the NCJRS federal sponsors that highlight the latest research published or sponsored by the Office of Justice Programs.

NCJRS Abstract

The document referenced below is part of the NCJRS Virtual Library collection. To conduct further searches of the collection, visit the Virtual Library. See the Obtain Documents page for direction on how to access resources online, via mail, through interlibrary loans, or in a local library.

 

NCJ Number: 50564 Find in a Library
Title: MULTIDIMENSIONAL CONTINGENCY TABLE ANALYSIS OF PAROLE OUTCOME-NEW METHODS AND OLD PROBLEMS IN CRIMINOLOGICAL PREDICTION
Journal: JOURNAL OF RESEARCH IN CRIME AND DELINQUENCY  Volume:15  Issue:2  Dated:(JULY 1978)  Pages:172-193
Author(s): D J VAN ALSTYNE; M R GOTTFREDSON
Corporate Author: National Council on Crime and Delinquency
United States of America
Date Published: 1978
Page Count: 22
Sponsoring Agency: National Council on Crime and Delinquency
Newark, NJ 07102
Type: Statistics
Format: Article
Language: English
Country: United States of America
Annotation: THE LOG-LINEAR METHOD OF PREDICTING PAROLE SUCCESS IS COMPARED TO THE BURGESS METHOD OF PREDICTION IN CONSTRUCTION AND CROSS-VALIDATION SAMPLES, AND A REPLICATION STUDY OF THE LOG-LINEAR MODEL IS PERFORMED.
Abstract: THE LOG-LINEAR TECHNIQUE PROVIDES A MEANS OF EXAMINING ALL POSSIBLE COMPLEXITIES OF A MULTIPLE CONTINGENCY TABLE AND ISOLATING THE MOST IMPORTANT ASSOCIATIONS AND INTERACTIONS. IN PAROLE PREDICTION, THE METHOD PROVIDES ESTIMATES OF THE EFFECTS OF VARIABLES ACTING ALONE OR IN CONJUNCTION WITH OTHER VARIABLES, AND A MEANS OF INDIRECTLY TESTING HYPOTHESIZED RELATIONSHIPS AMONG VARIABLES BY SETTING UP COMPETING MODELS AND COMPARING THEM WITH OBSERVED EVENTS. A SAMPLE OF 5,587 OHIO PAROLEES IS EMPLOYED IN A VALIDATION STUDY COMPARING THE LOG-LINEAR TECHNIQUE WITH THE BURGESS TECHNIQUE (SAID TO BE THE LEAST SOPHISTICATED STATISTICAL METHOD USED IN CONSTRUCTING PREDICTION INSTRUMENTS), AND IN A REPLICATION STUDY OF THE LOG-LINEAR MODEL. IN THE VALIDATION STUDY, THE LOG-LINEAR TECHNIQUE ACHIEVED ABOUT THE SAME PREDICTIVE EFFICIENCY AS THE BURGESS METHOD FOR BOTH CONSTRUCTION AND VALIDATION SAMPLES. THE HIERARCHICAL MODEL THAT EMERGED FROM THE LOG-LINEAR ANALYSIS OF THE CONSTRUCTION SAMPLE DID NOT REPLICATE ON THE VALIDATION SAMPLE. IN LIGHT OF OTHER STUDIES, THE FINDINGS INDICATE THAT IMPROVING STATISTICAL TECHNIQUES MAY NOT BE THE BEST MEANS OF IMPROVING PREDICTIVE EFFICIENCY. IMPLICATIONS OF THE RESULTS FROM PREDICTION RESEARCH AND THEORY TESTING INCLUDE THE INDICATION THAT REPLICATION OF HIERARCHICAL MODELS CANNOT BE ASSUMED BUT MUST BE STUDIED EMPIRICALLY. SUPPORTING DATA AND A LIST OF REFERENCES ARE INCLUDED. (AUTHOR ABSTRACT MODIFIED--LKM)
Index Term(s): Mathematical modeling; Parole; Parolees; Prediction; Statistical analysis; Success factors
To cite this abstract, use the following link:
http://www.ncjrs.gov/App/publications/abstract.aspx?ID=50564

*A link to the full-text document is provided whenever possible. For documents not available online, a link to the publisher's website is provided. Tell us how you use the NCJRS Library and Abstracts Database - send us your feedback.