Crime and time: The temporal patterning of causal variables

Criminologists tend to assume that most of the variables that cause crime are stable over long periods, with some variables having a brief, episodic existence in the lives of individuals. This article challenges that assumption and instead argues that variables are best described in terms of three temporal levels: baseline levels or long-term averages; short-term deviations around these baseline levels, with such deviations lasting from hours to days; and situational deviations, lasting from seconds to minutes. Criminologists can more accurately describe the standing of individuals on the causes of crime using these levels, thereby improving the ability to explain crime.