This study investigated the issue of cross-situational consistency by studying the relationship between risky behaviour in traffic and criminal behaviour. The analyses were based on a random sample of 1531 persons involved in accidents. The data were coming from two independent registration systems of the police: the accident registration system and a national data-base on offending. Descriptions of the accidents by the police were used to identify individuals who displayed risky traffic behaviour, evidence of offending was based on registered police contacts. The authors hypothesized that drivers who had been judged to cause the accident by displaying some form of risky behaviour might have been more often involved in criminal activity than those who were judged to be passive accident victims. Log-linear analyses revealed that persons who displayed risky traffic behaviour directly preceding the accident had an odds of 2.6 for having a police record for violent crime; these odds were 2.5 for vandalism, and 1.5 for property crime, and 5.3 for having been involved in traffic crime. It is concluded that these findings support the cross-situational consistency thesis. The results were consistent with the idea of a common factor underlying risky behaviour in traffic and criminal behaviour. This underlying trait may represent a general disregard for the long term adverse consequences of one's actions and could be labelled risk-taking, impulsiveness or self-control. This study extends over previous research as there is (i) no selection bias in the data, (ii) no self-report bias in the measures used, (iii) the design controls for exposure, (iv) were able to examine different types of crime and (v) to control for alcohol use as a possible intervening variable. (A)
Abstract