Using Poisson Class Regression to Analyze Count Data in Correctional and Forensic Psychology: A Relatively Old Solution to a Relatively New Problem

• NCJ Number: 221034
• Journal: Criminal Justice and Behavior Volume: 34 Issue: 12 Dated: December 2007 Pages: 1659-1674
• Author(s): Glenn D. Walters
• Date Published: December 2007
• Length: 16 pages
• Publication Format: Article
• Publication Type: Measurement/Evaluation Device
• Language: English
• Country: United States of America

Annotation

This study reanalyzed data from two previously published studies in an attempt to illustrate how count data could be analyzed with Poisson class procedures and compared the results to those obtained with binomial and linear regression.

Abstract

Count data present a formidable challenge to researchers in the correctional and forensic psychology fields. The benchmark model for count data is the Poisson distribution, and the standard statistical procedure for analyzing count data is Poisson regression. However, highly restrictive assumptions lead to frequent misspecification of the Poisson model. Alternate approaches, such as negative binomial regression, zero modified procedures, and truncated and censored models are consequently required to handle count data in many social science contexts. Two empirical examples from correctional and forensic psychology are provided to illustrate the importance of replacing ordinary least squares regression with Poisson class procedures in situations when count data are analyzed. Figures, tables, references