Multiple Cohort Data, Delinquent Generations, and Criminal Careers

• NCJ Number: 205843
• Journal: Journal of Contemporary Criminal Justice Volume: 20 Issue: 2 Dated: May 2004 Pages: 103-126
• Author(s): Brian Francis; Keith Soothill; Elizabeth Ackerly
• Date Published: May 2004
• Length: 24 pages
• Grant Number(s): R000223035
• Sponsoring Agency: British Economic and Social Research Council
• Publication Format: Article
• Publication Type: Report (Study/Research)
• Language: English
• Country: United States of America

Annotation

Based on data from the England and Wales Offenders Index, including official conviction histories of six birth cohorts of offenders, this study examined whether cohort or generational effects can be identified in official conviction history data once allowance has been made for age and year.

Abstract

The Offenders Index of England and Wales is a court-based database of all "standard list" (all serious offenses and many of the less serious offenses that are triable at magistrates courts only or in either magistrates or crown courts) criminal convictions in the two countries from 1963 to the present. Criminal convictions are recorded for all offenders ages 10 or older. In discussing methodology, the article addresses problems related to the analysis of official conviction data, including the need to consider temporal changes in the age-crime curve caused by social change and criminal policy, as well as the need to take into account identification issues. A Poisson log-linear model was selected for the analysis, since it allows for the increasing variability in convictions as the population size increases. The study found strong interactions between previous convictions and age, indicating that the number of previous convictions became less important as age increased. This indicates that an alternative variable might be the rate of previous offending over the course of the criminal career rather than the number of previous convictions; however, the interactions between previous convictions and year indicates that the number of previous convictions became more important with year, i.e., every previous conviction increased the rate of current convictions more in recent years than in earlier years. This result was unexpected. This is apparently an artifact of the workings of the criminal justice system rather than a change in the relationship between previous offending and current offending. Although the study identified small cohort effects in both the male and female data, the nature of these effects could not be identified. 4 tables, 4 figures, and 31 references