|
Article Excerpt
INTRODUCED BY COHEN AND FELSON (1979), routine activities theory outlines how victimization is associated with everyday patterns of social interaction. Routine activities are defined as "any recurrent or prevalent activities which Provide for basic population and individual needs" (Cohen and Felson, 1979: 593). These activities range from formal work to leisure pursuits to the methods that people utilize to acquire food, shelter, education and other basic needs. The perspective works from the premise that routine activities influence criminal opportunity by bringing together offenders and victims. The focus is on "direct-contact predatory violations." These are violations where there is "direct physical contact between at least one offender and at least one person or object which that offender attempts to take or damage" (Cohen and Felson, 1979: 589). While other criminological perspectives focus on the motivated offender, Cohen and Felson take this as a given and instead focus on suitable targets (things worth stealing, a person to assault, or a home to break into), and the lack of guardianship (the absence of anyone or anything that might prevent the offense from occurring). They argue that most criminal acts require the convergence of motivated offenders and suitable targets, in the absence of capable guardians, and this convergence is influenced by people's routine activities. Routine activities can influence the location of personal and property targets at particular times. Due to routine activities, suitable targets are said to be found more often in certain settings than others. For example, those who are employed may present opportunities for burglary by presenting a home (a suitable target) that lacks guardianship. Leisure activities may also leave homes unguarded and bring together people of different backgrounds at various times of night or day in locations that leave them vulnerable to violent victimization. Cohen and Felson add that the convergence of suitable targets and the absence of capable guardians "may even lead to large increases in crime rates without necessarily requiring any increase in structural conditions that motivate individuals to engage in crime." Furthermore, they write that "if the proportion of motivated offenders or even suitable targets were to remain stable in a community, changes in routine activities could nonetheless alter the likelihood of their convergence in space and time" (1979: 589). This suggests that, if one can account for variations in routine activities, then one can account for potential and actual victimization experiences. It also suggests that people can change their risk for victimization by altering their routine activities and lifestyles.
Most research in this area has taken the motivated offender as a given and has focussed on targets and guardians. Studies have found that being a drinker, having limited income, and frequently going out in the evening increase the probability of victimization. Demographic characteristics, such as being male, young, unmarried, unemployed and an urban dweller also increase victimization (Gottfredson, 1981; Miethe, Stafford and Long, 1987; Lasley and Rosenbaum, 1988, Kennedy and Forde, 1990; Sacco and Johnson, 1990, Sacco, Johnson and Arnold, 1993; Keane and Arnold, 1996). Studies examining both individual and contextual effects have found, in general, that community factors such as social cohesion, socio-economic status, crime rates, ethnic heterogeneity, and residential instability influence victimization risk (Sampson and Wooldredge, 1987; Sampson and Lauritsen, 1990; Miethe and McDowall, 1993; Rountree, Land and Miethe, 1994; Sampson, Raudenbusch and Earls, 1997; Lauritsen, 2001).
Other studies have found that being the victim of a crime increases the likelihood that one will be victimized in the future (Miethe, Stafford and Sloan, 1990; Farrell, 1995; Farrell, Phillips and Pease, 1995; Lauritsen and Davis Quinet, 1995). Wittebrood and Nieuwbeerta (2000), however, found that changes in routine activities produced significant changes in the risk of victimization (112). Changing from high-risk to low-risk routine activities/lifestyles sharply reduced the risk of victimization.
Given that changes in routine activities affect individual victimization, we ask how aggregate victimization rates would change if many individuals were to modify their behaviour. Put another way, how would overall victimization rates change if individuals engaging in high-risk behaviour behaved more like those at low risk? Would this affect different types of victimizations in the same way? These questions lead us to epidemiology, which is greatly concerned with how changes to causal factors impact the risk of illness or mortality. Epidemiologists wish to estimate how much of an effect a treatment may have, and the number of cases that can be averted if a cause is removed. Adopting their approach allows us to ask not only who is at risk for victimization, but also how many crimes might be prevented if characteristics of victims or their settings were changed. Further, it allows us to go beyond the past research, which focusses on how much better, or worse, a condition is in one group in comparison to another, to being able to determine which factors account for the greatest amount of risk or the magnitude of the problem. In this paper we bring epidemiological concepts of risk into victimization research.
Concepts of Risk
The notion of risk is the probability of some usually undesirable outcome. While in epidemiology risk is typically disease, or mortality, it can be expanded to include criminal victimization. For example, we can say that the mean risk of violent victimization for those individuals aged 15 and over in 2003 was .054. This means that the probability of violent victimization, averaged over all those 15 and over, was .054. This number corresponds to the proportion actually victimized. We can also examine risk at different levels. If we say that the minimum risk was .002, this means that for those in the category of lowest risk the probability of violent victimization was .002. Again, this number corresponds to the proportion in this category actually victimized.
Forms of Risk
From the basic concept of risk we can examine several variants (Streiner, 1998a; 1998b). Relative risk is the ratio of the risk experienced by someone who has a given characteristic, a "risk factor," and someone who has not. For example, if someone living in a rural setting has a risk of violent crime of .050, and someone in a city has a risk of .055, the relative risk for the urbanite is 1.10 (.055/.05). This means an urban dweller has a 10% greater risk of victimization relative to a rural dweller. In medical research, outcomes such as disease and mortality are most often predicted through logistic regression. The odds ratios obtained from this procedure are usually employed as indicators of how much the danger of illness or mortality is multiplied by a risk factor. As long as the probability of an outcome is relatively small, odds ratios differ only slightly from risk ratios, and...
|
