About | HeinOnline Law Journal Library | HeinOnline Law Journal Library | HeinOnline

35 Canadian J. Criminology 31 (1993)
Re-Analyses of J.P. Rushton's Crime Data

handle is hein.journals/cjccj35 and id is 35 raw text is: Re-analyses of J.P Rushton's crime data
Zack Z. Cernovsky (1)
Department of Psychology
University of Western Ontario
London, Ontario
Larry C. Litman
Forensic Services
St. Thomas Psychiatric Hospital
St. Thomas, Ontario
Rkeemment, Rushton affirmait qu 'il existe des diff~rences importantes dans les taux de
criminalitg entre les groupes raciaux. Ses mthodes d'analyse statistique (l'analyse de
variance), cependant, ne fait pas ressortir la grandeur de ces difflrences. Les auteurs
ont rO-analys  les donnes de Rushton et dmontrent qu 'il n 'y a pas de rapports fermes
entre les facteurs (le coifficient de corr~lation moyen itait .24, ce qui laisse supposer
que moins de 6% de la variance est commune). Les tendances gtaientfaibles et
contradictoires (70.8% des co~fficients sont oufaibles ou sans importance). Les
donnges de Rushton semblent indiquer qu 'en se basant sur le facteur racial comme
prdicteur de criminalitM il en r~sulterait un taux tr~s  lev  (99.9%o) defaux-positifs.
Rushton recently reported significant racial differences in crime rates. His statistical
method (the analysis of variance), however, does not assess the size of these
significant trends. A re-analysis of his data shows the absence of strong relationships
(the average correlation coefficient was .24, suggesting that less than 6% of the variance
is shared). The trends were mostly weak and inconsistent (70.8% of the underlying
coefficients are classified as low or nonsignificant). Rushton's own data suggest that
relying on race as a predictor of crime in individual cases would result in an absurdly
high rate (99.90o) offalse positives.
In a recent article in this Journal, Rushton (1990) presented his statistical
analyses to document his view that crime frequencies follow his model of racial
differences in behaviour. Rushton reports that he collected his data (see Table
1) from international criminal police archives, calculated a one-way analysis of
variance (ANOVA), and concluded that the races differ significantly in crime
Canadian Journal of Criminology                31 to 36
Revue canadienne de criminologie    January/janvier 1993

What Is HeinOnline?

HeinOnline is a subscription-based resource containing nearly 3,000 academic and legal journals from inception; complete coverage of government documents such as U.S. Statutes at Large, U.S. Code, Federal Register, Code of Federal Regulations, U.S. Reports, and much more. Documents are image-based, fully searchable PDFs with the authority of print combined with the accessibility of a user-friendly and powerful database. For more information, request a quote or trial for your organization below.

Short-term subscription options include 24 hours, 48 hours, or 1 week to HeinOnline with pricing starting as low as $29.95

Contact us for annual subscription options:

Already a HeinOnline Subscriber?

profiles profiles most