Alcohol and Crime: Behavioral Aspects - Interpreting Event-based Studies

Information about the co-occurrence of alcohol and criminal events is widely reported, but does not establish a relationship between alcohol and crime. Knowing the proportion of offenders who were drinking at the time of the crime is not meaningful unless we know the proportion of drinkers among people who did not commit crimes; if these proportions are the same, then there is no association of drinking to crime. The National Research Council Panel on the Understanding and Control of Violent Behavior concluded that existing data was not "sufficient to show that alcohol use or intoxication increases the general risk of violence. To test that hypothesis with prevalence data, one would need a benchmark: the fractions of people not involved in violence or crime while drinking—with appropriate adjustments for demographic characteristics of participants, time of day, day of week, and place of occurrence" (Reiss and Roth, p. 184).

Collecting appropriate comparison information is difficult, and the appearance of alcohol in such a large proportion of some crimes had led some commentators to suggest that it is unlikely that such a large proportion of people in general would be drinking at a particular time. However, Evans notes that in the West of Scotland, the proportion of offenders who were intoxicated at the time of their offense corresponds with the proportion of men who would be expected to be intoxicated at similar times (Evans). Moreover, among convicted offenders, drinking at the time of the offense is no more than would be expected given typical drinking patterns (Kalish; Ladouceur and Temple). For example, data from the U.S. Bureau of Justice Statistics show that one-third of state prison inmates reported drinking heavily just before they committed the offense, but 20 percent of all inmates reported that they drank very heavily every day the entire year before entering prison (Greenfeld). Drinking before an offense may reflect a typical pattern rather than be specifically related to the commission of a crime.