Identifying Thresholds of Substance Use and Their Relationship to Criminal Involvement

Mark Asbridge, University of Toronto

This paper examines the general question of the relationship between substance use and criminal involvement. In order to address this question a sample of adolescents between the ages of 13 and 22 are analyzed, looking specifically at their various patterns and subpatterns of substance use and criminal involvement. These subpatterns of substance use situate individuals within particular thresholds of behaviour. The term "threshold" is used to denote distinct frequencies and intensities of substance using and criminal behaviour. Thresholds range from low (individuals who do not use substances at all, associate with very few individuals who do, and hold attitudes strongly against substance use), moderate (individuals who are involved in substance use in which casual experimentation is the norm) and high threshold substance users (Individuals who are frequent users of a wide array of harmful substances, who associate with other user, who have family whom are involved in substance use, and who have positive attitudes towards substance use). Threshold levels, beyond acting as a typology for current behaviour, represents a cut-off for identifying individuals who are likely to be involved in high levels of substance use or criminal involvement later in the life course. Research suggests that individual tendencies towards high involvement in substance using behaviours are influenced by a number of factors, including: age of onset, parental and peer behaviours, history of similar behaviour patterns, attitudes and opinions in favour of such behaviours, and so forth. These factors will be used to examine how thresholds of substance use behaviour relate to various types of criminal involvement. Are particular substance use thresholds associated with distinct forms of criminal involvement? Are high threshold substance users also involved in the most extreme forms of crime and deviance? And does the intensity of substance use hold any relevance in the explanation of involvement in crime and the type criminal behaviour? Data for this project comes from the "Toronto Area Youth Survey", a survey administered to over 3,500 high school students and street youth in the city of Toronto between the Fall of 1998 and the Spring of 2000. Thresholds of substance use are constructed using Factor analysis. Meanwhile a multinomial logit regression is employed to regress substance use threshold, as well as a set of additional predictor variables, on various forms of criminal involvement.

(Return to Program Resources)

Updated 05/20/2006