NCJ Number
105055
Journal
Canadian Journal of Criminology Volume: 29 Issue: 2 Dated: (April 1987) Pages: 171-183
Date Published
1987
Length
13 pages
Annotation
This paper argues for and proposes a longitudinal and experimental research design intended to advance knowledge about the causes, prevention, and treatment of crime in Canada.
Abstract
To test theories and draw implications about crime prevention and treatment, crime causes and correlates must be distinguished. This is impossible in cross-sectional and correlational research, which currently dominates criminological research. Longitudinal research, on the other hand, provides information about time ordering, which can show that some correlates, because they do not predict crime, are not causes. Longitudinal research has been especially useful in providing knowledge about the natural history of criminal careers, the cumulative prevalence of different types of offending, ages of onset and termination, career lengths, the extent of specialization as opposed to versatility in offending, escalation in offense seriousness, and the extent to which some life events are stepping stones to others. The best way of establishing the effects of events or interventions is through randomized experiments. The time is ripe to combine longitudinal and experimental research methods in a new way and launch one or more major Canadian longitudinal studies containing planned experimental interventions. The primary aim of such a study should be to provide information about the development of offending from birth to the mid-twenties, and the first priority should be to provide information on those crimes which the public perceives as particularly threatening and on the most frequent and serious offenders. 38 references.