Using intelligence to predict subsequent contacts with the criminal justice system for sex offences

This article examines the relationship between IQ and subsequent criminal justice system contacts for sexual offences, and the extent to which the relationship is linear, as past research utilizing linear strategies documents that the relationship is weak. Data were extracted from a secondary dataset of prisoners who completed IQ tests, and whose criminal records were prospectively followed up for 20 years, resulting in 194 persons with justice system contact for a sexual offense. Criminal career dimensions examined included: age at offense, time of contact with the justice system, mean number of sexual convictions, whether someone was a recidivist and whether someone was convicted or arrested. Intelligence was assessed utilizing the General Aptitude Battery, consisting of eight scales. Linear models showed slightly less accuracy and significance than recursive partitioning modeling which predicted criminal careers for rape. Collectively, the results suggest that Finger dexterity, Motor aptitude and Spatial aptitude emerge as weak yet consistent predictors of rape across a variety of different criminal career rape outcomes, and that the IQ crime relationship is not linear.