Research output: Contribution to Journal/Magazine › Journal article › peer-review
Research output: Contribution to Journal/Magazine › Journal article › peer-review
}
TY - JOUR
T1 - Multiple cohort data, delinquent generations and criminal careers.
AU - Francis, Brian J.
AU - Ackerley, Elizabeth
AU - Soothill, Keith
N1 - RAE_import_type : Journal article RAE_uoa_type : Statistics and Operational Research
PY - 2004/5
Y1 - 2004/5
N2 - This article investigates the effect of birth cohort membership on the risk of criminal conviction. The data used are taken from the England andWales Offenders Index and include official conviction histories of six birth cohorts of offenders. Particular problems related to the analysis of official conviction data are discussed, including the need to take account of temporal changes in the age-crime curve caused by social change and criminal policy and the need to be aware of identification issues. A Poisson log-linear model approach is taken. Including data on previous convictions can help explain heterogeneity. An interaction of previous convictions with yearwas found to be important, indicating that additional previous convictions have a greater effect on the conviction rate for more recent years. There is a weak cohort effect for males and a very weak effect for females, after controlling for age, year, previous convictions, and interactions between them.
AB - This article investigates the effect of birth cohort membership on the risk of criminal conviction. The data used are taken from the England andWales Offenders Index and include official conviction histories of six birth cohorts of offenders. Particular problems related to the analysis of official conviction data are discussed, including the need to take account of temporal changes in the age-crime curve caused by social change and criminal policy and the need to be aware of identification issues. A Poisson log-linear model approach is taken. Including data on previous convictions can help explain heterogeneity. An interaction of previous convictions with yearwas found to be important, indicating that additional previous convictions have a greater effect on the conviction rate for more recent years. There is a weak cohort effect for males and a very weak effect for females, after controlling for age, year, previous convictions, and interactions between them.
KW - age-crime curve • age-period-cohort models • crime rates • conviction data
U2 - 10.1177/1043986204263768
DO - 10.1177/1043986204263768
M3 - Journal article
VL - 20
SP - 103
EP - 126
JO - Journal of Contemporary Criminal Justice
JF - Journal of Contemporary Criminal Justice
SN - 1552-5406
IS - 2
ER -