Home > Research > Publications & Outputs > Estimation issues and generational changes in m...
View graph of relations

Estimation issues and generational changes in modelling criminal career length

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Estimation issues and generational changes in modelling criminal career length. / Francis, Brian J.; Piquero, Alex R.; Soothill, Keith.

In: Crime and Delinquency, Vol. 53, No. 1, 01.2007, p. 84-105.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

APA

Vancouver

Francis BJ, Piquero AR, Soothill K. Estimation issues and generational changes in modelling criminal career length. Crime and Delinquency. 2007 Jan;53(1):84-105. doi: 10.1177/0011128706294441

Author

Francis, Brian J. ; Piquero, Alex R. ; Soothill, Keith. / Estimation issues and generational changes in modelling criminal career length. In: Crime and Delinquency. 2007 ; Vol. 53, No. 1. pp. 84-105.

Bibtex

@article{408276e9651b41eca2200ab47f46ad2d,
title = "Estimation issues and generational changes in modelling criminal career length",
abstract = "This article seeks to model criminal career length using data from six different birth cohorts born between 1953 and 1978, totaling more than 58,000 males and females from England and Wales. A secondary aim of this article is to consider whether information available at the first court appearance leading to a conviction is associated with the length of a criminal career. Four key findings emerged. First, one can estimate career length from variables available at the first court conviction. Second, the risk of desistance remains constant during a period of 20 to 25 years if the offender does not immediately stop after the first conviction. Third, the most significant variable is age at first conviction. Finally, gender differences and birth cohorts are also important.",
keywords = "career length • criminal careers • cohort effects • Cox model • survival analysis",
author = "Francis, {Brian J.} and Piquero, {Alex R.} and Keith Soothill",
note = "RAE_import_type : Journal article RAE_uoa_type : Statistics and Operational Research",
year = "2007",
month = jan,
doi = "10.1177/0011128706294441",
language = "English",
volume = "53",
pages = "84--105",
journal = "Crime and Delinquency",
issn = "0011-1287",
publisher = "SAGE Publications Inc.",
number = "1",

}

RIS

TY - JOUR

T1 - Estimation issues and generational changes in modelling criminal career length

AU - Francis, Brian J.

AU - Piquero, Alex R.

AU - Soothill, Keith

N1 - RAE_import_type : Journal article RAE_uoa_type : Statistics and Operational Research

PY - 2007/1

Y1 - 2007/1

N2 - This article seeks to model criminal career length using data from six different birth cohorts born between 1953 and 1978, totaling more than 58,000 males and females from England and Wales. A secondary aim of this article is to consider whether information available at the first court appearance leading to a conviction is associated with the length of a criminal career. Four key findings emerged. First, one can estimate career length from variables available at the first court conviction. Second, the risk of desistance remains constant during a period of 20 to 25 years if the offender does not immediately stop after the first conviction. Third, the most significant variable is age at first conviction. Finally, gender differences and birth cohorts are also important.

AB - This article seeks to model criminal career length using data from six different birth cohorts born between 1953 and 1978, totaling more than 58,000 males and females from England and Wales. A secondary aim of this article is to consider whether information available at the first court appearance leading to a conviction is associated with the length of a criminal career. Four key findings emerged. First, one can estimate career length from variables available at the first court conviction. Second, the risk of desistance remains constant during a period of 20 to 25 years if the offender does not immediately stop after the first conviction. Third, the most significant variable is age at first conviction. Finally, gender differences and birth cohorts are also important.

KW - career length • criminal careers • cohort effects • Cox model • survival analysis

U2 - 10.1177/0011128706294441

DO - 10.1177/0011128706294441

M3 - Journal article

VL - 53

SP - 84

EP - 105

JO - Crime and Delinquency

JF - Crime and Delinquency

SN - 0011-1287

IS - 1

ER -