Assignment: Apply Victimization Theory

Section II Extent, Theories, and Factors of Victimization

It was not exactly a typical night for Polly. Instead of studying at the library as she normally did during the week, she decided to meet two of her friends at a local bar. They spent the evening catching up and drinking a few beers before they decided to head home. Because Polly lived within walking distance of the bar, she bid her friends goodnight and started on her journey home. It was dark out, but because she had never confronted trouble in the neighborhood before—even though it was in a fairly crime-ridden part of a large city—she felt relatively safe.

As Polly walked by an alley, two young men whom she had never seen before stepped out, and one of them grabbed her arm and demanded that she give them her school bag, in which she had her wallet, computer, keys, and phone. Because Polly refused, the other man shoved her, causing her to hit her head, while the first man grabbed her bag. Despite holding on as tightly as she could, the men were able to take her bag before running off into the night. Slightly stunned, Polly stood there trying to calm down. Without her bag, which held her phone and keys, she felt there was little she could do other than continue to walk home and hope her roommates were there to let her in. As she walked home, she wondered why she had such bad luck. Why was she targeted? Was she simply in “the wrong place at the wrong time,” or did she do something to place herself in harm’s way? Although it is hard to know why Polly was victimized, we can compare her to other victims to see how similar she is to them. To this end, a description of the “typical” crime victim is presented in this section. But what about why she was targeted? Fortunately, we can use the theories presented in this section to understand why Polly fell victim on that particular night.

Photo 2.1 Polly, on her way home from the bar.

© iStockphoto.com/theprint

Measuring Victimization Before we can begin to understand why some people are the victims of crime and others are not, we must first know how often victimization occurs. Also important is knowing who the typical crime victim is. Luckily, these characteristics of victimization can be readily gleaned from

existing data sources.

Uniform Crime Reports Begun in 1929, the Uniform Crime Report (UCR) shows the amount of crime known to the police in a year. Police departments around the country submit to the Federal Bureau of Investigation (FBI) monthly law enforcement reports on crimes that are reported to them or that they otherwise know about. The FBI then compiles these data and each year publishes a report called Crime in the United States, which details the crime that occurred in the United States for the year. This report includes information on eight offenses, known as the Part I index offenses: murder and nonnegligent manslaughter, rape, robbery, aggravated assault, burglary, larceny-theft, motor vehicle theft, and arson. Arrest data are also listed in the report on Part II offenses, which include an additional 21 crime categories.

Advantages and Disadvantages

The UCR is a valuable data source for learning about crime and victimization. Because more than 97% of the population is represented by agencies participating in the UCR program, it provides an approximation of the total amount of crime experienced by almost all Americans (Federal Bureau of Investigation [FBI], 2014a). It presents the number of crimes for regions, states, cities, towns, areas under tribal law enforcement, and colleges and universities. It does so annually so that crime trends can be determined for the country and for these geographical units. Another benefit of the UCR is that crime characteristics are also reported. It includes demographic information (age, sex, and race) on people who are arrested and some information on the crimes, such as location and time of occurrence.

Despite these advantages, it does not provide detailed information on crime victims. Also important to consider, the UCR includes information only on crimes that are reported to the police or of which the police are aware. In this way, all crimes that occur are not represented, especially because, as discussed shortly, crime victims often do not report their victimization to the police. Another limitation of the UCR as a crime data source is that the Part I index offenses do not cover the wide range of crimes that occur, such as simple assault and sexual assaults other than rape, and federal crimes are not counted. Furthermore, the UCR uses the hierarchy rule. If more than one Part I offense occurs within the same incident report, the law enforcement agency counts only the highest offense in the reporting process (FBI, 2009). These exclusions also contribute to the UCR’s underestimation of the extent of crime. Accuracy of the UCR data is also affected by law enforcement’s willingness to participate in the program and to do so by reporting to the FBI all offenses of which they are aware.

Crime as Measured by the UCR

Nonetheless, the UCR can be used to paint a picture of crime in the United States. In 2015, the police became aware of 1,197,704 violent crimes and 7,993,631 property crimes. According to the UCR data shown in Figure 2.1, the most common offense is larceny-theft. Aggravated assaults are the most common violent crime, although they are outnumbered by larceny-thefts. The typical criminal who is arrested is a young (less than 30 years old) White male (although young Black males have highest offending rates) (FBI, 2015a).

Figure 2.1 Number of Crimes Occurring in 2015, Comparison for UCR and NCVS

Source: Created by author with U.S. Department of Justice data.

Note: The UCR includes only forcible rape, whereas the NCVS includes both rape and sexual assault. The UCR measures only aggravated assault, whereas the NCVS includes both aggravated and simple assault.

National Incident-Based Reporting System

As noted, the UCR includes little information about the characteristics of criminal incidents. To overcome this deficiency, the FBI began the National Incident-Based Reporting System (NIBRS), an expanded data collection effort that includes detailed information about crimes. Agencies participating in the NIBRS collect information on each crime incident and arrest in 23 offense categories (Group A offenses) that encompass 49 specific crimes. Arrest data are reported for an additional 11 offenses (Group B offenses). Information about the offender, the victim, injury, location, property loss, and weapons is included (FBI, 2015a). Also of importance, NIBRS does not use the hierarchy rule when classifying or counting crimes (FBI, n.d.-a).

Although the NIBRS represents an advancement of the UCR program, not all law enforcement agencies participate in the system. As such, crime trends similar to those based on national data produced by the UCR are not yet available. As more agencies come online, the NIBRS data will likely be an even more valuable tool for understanding patterns and trends of crime victimization.

With consideration of these limitations, in 2015, the 6,648 law enforcement agencies (36.1% of all law enforcement agencies) participating in NIBRS reported 4,902,177 incidents that involved 6,668,103 offenses, 5,979,330 victims, and 4,607,928 known offenders. Of the offenses, 62.9% were property crimes, 23.2% were crimes against persons, and 14.0 % were crimes against society (also referred to as victimless crimes). There were 3,081,609 arrests for offenses tracked in NIBRS in 2015 (FBI, 2015a).

NIBRS is also a source of information on crime victims and incidents. Slightly less than one-quarter of victims were between 21 and 30 years of age and 51% of victims were females. Almost three-fourths of victims were White (72%), 20.8% were Black or African American, 1.4% were Asian, 0.6% were American Indian or Alaska Native, and less than 0.1% were Native Hawaiian or Pacific Islander (FBI, 2015b). In a slight majority of crimes against persons (52.3%), the victim knew his or her offender but was not related to the offender, and in 10.2% of the crimes against persons, the perpetrator was a stranger (FBI, 2015c). Most crimes against a person occur at the victim’s home (62.8%), whereas slightly more than 4 in 10 property crimes occur at the victim’s home (although this was the most common location of property crime category) (FBI, 2015d).

National Crime Victimization Survey As noted, the UCR and NIBRS have some limitations as crime data sources, particularly when information on victimization is of interest. To provide a picture of the extent to which individuals experience a range of crime victimizations, the Bureau of Justice

Statistics (BJS) began, in 1973, a national survey of U.S. households. Originally called the National Crime Survey, it provides a picture of crime incidents and victims. In 1993, the BJS redesigned the survey, making extensive methodological changes, and renamed it the National Crime Victimization Survey (NCVS).

The NCVS is administered by the U.S. Census Bureau to a nationally representative sample of about 95,000 households. Each member of participating households who is 12 years old or older completes the survey, resulting in about 163,000 persons being interviewed (Truman & Morgan, 2016). Persons who live in military barracks and in institutional settings (e.g., prisons and hospitals) and who are homeless are excluded from the NCVS. Each household selected remains in the study for 3 years and completes seven interviews 6 months apart. Each interview serves a bounding purpose by giving respondents a concrete event to reference (i.e., since the last interview) when answering questions in the next interview. Bounding is used to improve recall. In general, the first interview is conducted in person, with subsequent interviews taking place either in person or over the phone (Truman & Morgan, 2016).

The NCVS is conducted in two stages. In the first stage, individuals are asked if they experienced any of seven types of victimization during the previous 6 months. The victimizations that respondents are asked about are rape and sexual assault, robbery, aggravated and simple assault, personal theft, household burglary, motor vehicle theft, and theft. The initial questions asked in the first stage are known as screen questions, which are used to cue respondents or jog their memories as to whether they experienced any of these criminal victimizations in the previous 6 months. An example of a screen question is shown in Table 2.1. In the second stage, if the respondent answers affirmatively to any of the screen questions, the respondent then completes an incident report for each victimization experienced. In this way, if an individual stated that he or she had experienced one theft and one aggravated assault, he or she would fill out two incident reports—one for the theft and a separate one for the aggravated assault. In the incident report, detailed questions are asked about the incident, such as where it happened, whether it was reported to the police and why the victim did or did not report it, who the offender was, and whether the victim did anything to protect himself or herself during the incident. Table 2.2 shows an example of a question from the incident report. As you can see, responses to the questions from the incident report can help reveal the context of victimization.

Table 2.1 Example of Screen Question From NCVS

Other than any incidents already mentioned, has anyone attacked or threatened you in any of these ways (exclude telephone threats)?

(a) With any weapon, for instance, a gun or knife

(b) With anything like a baseball bat, frying pan, scissors, or stick

(c) By something thrown, such as a rock or bottle

(d) Include any grabbing, punching, or choking

(e) Any rape, attempted rape, or other type of sexual attack

(f) Any face-to-face threats

OR

(g) Any attack or threat or use of force by anyone at all? Please mention it even if you are not certain it was a crime.

Source: Bureau of Justice Statistics (2015a).

Another advantage of this two-stage procedure is that the incident report is used to determine what, if any, incident occurred. The incident report, as discussed, includes detailed questions about what happened, including questions used to classify an incident into its appropriate crime victimization type. For example, in order for a rape to be counted as such, the questions in the incident report that concern the elements of rape, which are discussed in Section VII (force, penetration), must be answered affirmatively for the incident to be counted as rape in the NCVS. This process is fairly conservative in that all elements of the criminal victimization must have occurred for it to be included in the estimates of that type of crime victimization.

Table 2.2 Example of Question From Incident Report in NCVS

Did the offender have a weapon such as a gun or knife, or something to use as a weapon, such as a bottle or wrench?

Source: Bureau of Justice Statistics (2015b).

The NCVS has several advantages as a measure of crime victimization. First, it includes in its estimates of victimization several offenses that are not included in Part I of the UCR; for example, simple assault and sexual assault are both included in NCVS estimates of victimization. Second, the NCVS does not measure only crimes reported to the police as does the UCR. Third, the NCVS asks individuals to recall incidents that occurred only during the previous 6 months, which is a relatively short recall period. In addition, its two-stage measurement process allows for a more conservative way of estimating the amount of victimization that occurs each year in that incidents are counted only if they meet the criteria for inclusion.

Despite these advantages, the NCVS is not without its limitations. Estimates of crime victimization depend on the ability of respondents to accurately recall what occurred to them during the previous 6 months. Even though the NCVS attempts to aid in recall by spanning a short period (6 months) and by providing bounding via the previous survey administration, it is still possible that individuals will not be completely accurate in recounting the particulars of an incident. Bounding and using a short recall period also do not combat against someone intentionally being misleading or lying or answering in a way meant to please the interviewer. Another possible limitation of the NCVS is its treatment of high-frequency repeat victimizations. Called series victimizations, these incidents are those in which a person experiences the same type of victimization during the 6-month recall period at such a high rate that he or she cannot recall specific details about each incident or even recall each incident. When this occurs, an incident report is only completed for the most recent incident, and incident counts are only included for

up to 10 incidents (Truman, Langton, & Planty, 2013). As such, estimates of victimization may be lower than the actual amount because the cap for counting series victimizations is 10. On the other hand, even without recalling specific detail, these incidents are included in estimates of victimization. Including series victimizations in this way reveals little effects on the trends in violence estimates (Truman & Morgan, 2016). In addition, murder and “victimless” crimes such as prostitution and drug use are not included in NCVS estimates of crime victimization. Another limitation is that crime that occurs to commercial establishments is not included. Beyond recall issues, the NCVS sample is selected from U.S. households. This sample may not be truly representative, for it excludes individuals who are institutionalized, such as persons in prison, and does not include homeless people. Remember, too, that only those persons ages 12 and over are included. As a result, estimates about victimization of children cannot be determined.

Extent of Crime Victimization

Each year, the BJS publishes Criminal Victimization in the United States, a report about crime victimization as measured by the NCVS. From this report, we can see what the most typical victimizations are and who is most likely to be victimized. In 2015, more than 19,600,000 victimizations were experienced among the nation’s households (Truman & Morgan, 2016). Property crimes were much more likely to be experienced compared with violent crimes; 5 million violent crime victimizations were experienced compared with 14.6 million property crime victimizations. The most common type of property crime reported was theft, whereas simple assault was the most commonly occurring violent crime (see Figure 2.1).

Typical Victimization and Victim

The typical crime victim can also be identified from the NCVS. For violent victimization except for rape and sexual assault, males and females were equally likely to be victimized. Persons who are Black and those under the age of 24 have higher victimization rates than others. Characteristics of victimization incidents are also evident. Less than half of all victimizations (47%) experienced by individuals in the NCVS are reported to the police. Property crimes are less likely to be reported than are violent crimes, with some crimes being much more likely to come to the attention of police than others. For example, rape and sexual assault are the least likely of all violent crimes to be reported, whereas robbery is the most likely to be reported. Almost 70% of motor vehicle thefts are reported to the police, but only about 29% of all thefts are (Truman & Morgan, 2016). This disjuncture in reporting is likely tied to features of the victimization and motivations for reporting. For example, the lack of reporting may be

related in part to the fact that most victims of violent crime know their offender; most often, victims identified their attacker as a friend or acquaintance. Strangers perpetrated only about one-third of violent victimizations in the NCVS (Truman et al., 2013). Reporting, on the other hand, may be tied to wanting to secure property back, especially a car. In addition, when a person has his or her car stolen, a police report is necessary for insurance purposes, so a person may be particularly motivated to report this type of victimization to the police. Returning now to incident characteristics, females are more likely than males to be victimized by an intimate partner. In about 58% of incidents the offender had a weapon, and about 55% of violent crimes resulted in the victim being physically injured (Truman et al., 2013). Now that you know the characteristics of the typical victimization and the typical crime victim, how do Polly and her victimization compare?

International Crime Victims Survey As you may imagine, there are many other self-report victimization surveys that are used to understand more specific forms of victimization, such as sexual victimization and those that occur outside the United States. Many of these are discussed in later sections. One oft-cited survey of international victimization is the International Crime Victims Survey (ICVS), which was created to provide a standardized survey to compare crime victims’ experiences across countries (van Dijk, van Kesteren, & Smit, 2008). The first round of the survey

was conducted in 1989 and was repeated in 1992, 1996, 2000, and 2004/2005. Collectively, more than 340,000 persons have been surveyed in more than 78 countries as part of the ICVS program (van Dijk et al., 2008). Respondents are asked about 10 types of victimization that they could have experienced: car theft, theft from or out of a car, motorcycle theft, bicycle theft, attempted or completed burglary, sexual victimization (rapes and sexual assault), threats, assaults, robbery, and theft of personal property (van Dijk et al., 2008). If a person has experienced any of these offenses, he or she then answers follow-up questions about the incident. This survey has provided estimates of the extent of crime victimization in many countries and regions of the world. In addition, characteristics of crime victims and incidents have been produced from these surveys.

Focus on International Issues

The Crime Survey for England and Wales (CSEW) is a victimization survey of persons age 16 and over living in England and Wales. Beginning in 1982, the CSEW was conducted every 2 years until 2001, when it was changed to reflect victimizations during the previous 12 months. Beginning April 1, 2012, the CSEW changed its name to the Crime Survey for England and Wales (formerly the British Crime Survey). Using computer-assisted personal interviewing to aid in interviewing, it is a nationally representative survey of about 35,000 adults and 3,000 children in the 10- to 15-year-old supplement. Persons are asked about victimizations that their households and they experienced. To get the sample, about 1,000 interviews are conducted in each police force area. If individuals answer yes to any screen question about victimization, they complete a victim module that includes detailed questions about the event. Findings from the CSEW for year ending 2015 indicate that there were 6.4 million crimes against households and those 16 and older, with 1.3 million violent incidents (Office for National Statistics, 2015).

Source: Home Office (2011).

Theories and Explanations of Victimization Now that you have an idea about who the typical crime victim is, you are probably wondering why some people are more likely than others to find themselves victims of crime. Is it because those people provoke the victimization, as von Hentig and his contemporaries thought? Is it because crime victims are perceived by offenders to be more vulnerable than others? Is there some personality trait that influences victimization risk? All these factors may play at least some role in why victimization occurs to particular people. The following sections address these possibilities.

Link Between Victimization and Offending One facet about victimization that cannot be ignored is the link between offending and victimization and offenders and victims. As mentioned in Section I, the first forays into the study of victims included a close look at how victims contribute to their own victimization. In this way, victims were not always assumed to be innocents; rather, some victims were seen as being at least partly responsible for bringing on their victimization—for instance, by being an offender who is victimized when the victim fights back. Although the field of victimology has moved from trying to place blame on victims, the recognition that offenders and victims are often linked—and often the same person—has aided in the understanding of why people are victimized.

Victim and Offender Characteristics

The typical victim and the typical offender have many commonalities. As mentioned before in our discussion of the NCVS, those with the highest rates of violent victimization are young Black persons. The UCR also provides information on offenders. Those with the highest rates of violent offending are also young Black males. The typical victim and the typical offender, then, share common demographics. In addition, both victims and offenders are likely to live in urban areas. Thus, individuals who spend time with people who have the characteristics of offenders are more likely to be victimized than others.

Photo 2.2 This area may be a hot spot due to lots of people milling about at night.

© iStockphoto.com/RyanJLane

Explaining the Link Between Victimization and Offending

Some even argue that victims and offenders are often one and the same, with offenders being more likely to be victimized and vice versa. It is not hard to understand why this may be the case. Offending can be viewed as part of a risky lifestyle. Individuals who engage in offending are exposed more frequently to people and contexts in which victimization is likely to occur (Lauritsen, Laub, & Sampson, 1992).

There also may be a link between victimization and offending that is part of a broader cultural belief in the acceptability and sometimes necessity of violence, known as the subculture of violence theory. This theory proposes that for certain subgroups of the population and in certain areas, violence is part of a value system that supports the use of violence, in response to disrespect in particular (Wolfgang & Ferracuti, 1967). In this way, when a subculture that supports violence exists, victims will be likely to respond by retaliating. Offenders may initiate violence that leads to their victimization by, for example, getting into a physical fight to resolve a dispute. Recent research shows that the victim–offender overlap does indeed vary across neighborhoods and that this variation is related to the neighborhood’s strength of attachment to the “code of the streets” and degree of structural deprivation (Berg & Loeber, 2012; Berg, Stewart, Schreck, & Simons, 2012).

Being victimized may be related to offending in ways that are not directly tied to retaliation. In fact, being victimized at one point in life may increase the likelihood that a person will engage in delinquency and crime later in life. This link has been found especially in individuals who are abused during childhood. As discussed in Section IX on victimization at the beginning and end of life, those who are victimized as children are significantly more likely than those who do not experience child abuse to be arrested in adulthood (Widom, 2000) or to engage in violence and property offending (Menard, 2002).

The reasons why victimization may lead to participation in crime are not fully understood, but it may be that being victimized carries psychological consequences, such as depression, anxiety, or post-traumatic stress disorder, that can lead to coping through the use of alcohol or drugs. Victimization may also carry physical consequences, such as brain damage, that can further impede

success later in life. Cognitive ability may also be tempered by maltreatment, particularly in childhood, which can hinder school performance. Behavior may also change as a result of being victimized. People may experience problems in their interpersonal relationships or become violent or aggressive. Whatever the reason, it is evident that victimization and offending are intimately intertwined.

Insomuch as victimization and offending are linked, it makes sense then, as you will see in the following sections, that the same influences on offending may also affect victimization and hence may explain the link between victimization and offending. This is not to say that the only explanations of victimization should be tied to or be an extension of explanations of offending—just remember that when you read about the research that has used criminological theories to explain victimization, it is largely because of the connection between victimization and offending.

Routine Activities and Lifestyles Theories In the 1970s, two theoretical perspectives—routine activities and lifestyles theories—were put forth that both linked crime victimization risk to the fact that victims had to come into contact with a potential offender. Before discussing these theories in detail, first, it is important to understand what a victimization theory is. A victimization theory is generally a set of testable propositions designed to explain why a person is victimized. Both routine activities and lifestyles theories propose that a person’s victimization risk can best be understood by the extent to which the victim’s routine activities or lifestyle creates opportunities for a motivated offender to commit crime.

In developing routine activities theory, Lawrence Cohen and Marcus Felson (1979) proposed that a person’s routine activities, or daily routine patterns, impact risk of being a crime victim. Insomuch as a person’s routine activities bring him or her into contact with motivated offenders, crime victimization risk abounds. Cohen and Felson thought that motivated offenders were plentiful and that their motivation to offend did not need to be explained. Rather, their selection of particular victims was more interesting. Cohen and Felson noted that there must be something about particular targets, both individuals and places, that encouraged selection by these motivated offenders. In fact, those individuals deemed to be suitable targets based on their attractiveness would be chosen by offenders. Attractiveness relates to qualities about the target, such as ease of transport, which is why a burglar may break into a home and leave with an iPod or laptop computer rather than a couch. Attractiveness is further evident when the target does not have capable guardianship. Capable guardianship is conceived as a means by which a person or target can be effectively guarded to prevent a victimization from occurring. Guardianship is typically considered to be social, when the presence of another person makes someone less attractive as a target. Guardianship can also be provided through physical means, such as a home with a burglar alarm or a person who carries a weapon for self-protection. A home with a burglar alarm and a person who carries a weapon are certainly less attractive crime targets! When these three elements—motivated offenders, suitable targets, and lack of capable guardianship—coalesce in time and space, victimization is likely to occur.

When Cohen and Felson (1979) originally developed their theory, they focused on predatory crimes—those that involve a target and offender making contact. They originally were interested in explaining changes in rates of these types of crime over time. In doing so, they argued that people’s routines had shifted since World War II, taking them away from home and making their homes attractive targets. People began spending more time outside the home in leisure activities and going to and from work and school. As people spent more time interacting with others, they were more likely to come into contact with motivated offenders. Capable guardianship was unlikely to be present; thus, the risk of criminal victimization increased. Cohen and Felson also linked the increase in crime to the production of durable goods. Electronics began to be produced in portable sizes, making them easier to steal. Similarly, cars and other expensive items that could be stolen, reused, and resold became targets. As Cohen and Felson saw it, prosperity of society could produce an increase in criminal victimization rather than a decline. Also important, they linked victimization to everyday activities rather than to social ills, such as poverty.

Michael Hindelang, Michael Gottfredson, and James Garofalo’s (1978) lifestyles theory is a close relative of routine activities theory. Hindelang and colleagues posited that certain lifestyles or behaviors place people in situations in which victimization is likely to occur. Your lifestyle, such as going to bars or working late at night in relative seclusion, places you at more risk of being a crime victim than others. Although the authors of lifestyles theory did not specify how opportunity structures risk as clearly as did the authors of routine activities theory, at its heart, lifestyles theory closely resembles routine activities theory and its propositions. As a person comes into contact —via lifestyle and behavior—with potential offenders, he or she is likely creating opportunities for crime victimization to occur. The lifestyle factors identified by Hindelang and his colleagues that create opportunities for victimization are the people with whom one associates, working outside the home, and engaging in leisure activities. In this way, a person who associates with criminals, works outside the home, and participates in activities—particularly at night, away from home, and with nonfamily members—is a more likely target for personal victimization than others. Hindelang and colleagues noted that a person’s lifestyle is structured by social constraints and role expectations. That is, because of a person’s demographic characteristics, he or she may be afforded less opportunity to engage in particular activities. Consider the fact that females are socialized differently from males. Females may be expected to be the caretaker of the home and, when younger, may be supervised more closely than males. Accordingly, females may spend more time at home and spend more time under the supervision of their parents or other guardians. Given these social constraints and role expectations, females may be less likely to engage in activities outside the home that

would place them at risk for victimization, hence explaining why females are at lower risk for victimization than males.

Hindelang et al. (1978) further delineated why victimization risk is higher for some people than others using the principle of homogamy. According to this principle, the more frequently a person comes into contact with persons in demographic groups with likely offenders, the more likely it is the person will be victimized. This frequency may be a function of demographics or lifestyle. For example, males are more likely to be criminal offenders than females. Males, then, are at greater risk for victimization because they are more likely to spend time with other males. Now that you know about routine activities theory, do you think Polly’s routines or lifestyle placed her at risk for being victimized? Today, researchers largely treat routine activities theory and lifestyles theory interchangeably and often refer to them as the routine activities and lifestyles theory perspectives.

One of the reasons that routine activities and lifestyles theories have been the prevailing theories of victimization for more than 30 years is the wide empirical support researchers have found when testing them. It has been shown that a person’s routine activities and lifestyle impact risk of being sexually victimized (Cass, 2007; Fisher, Daigle, & Cullen, 2010a, 2010b; Mustaine & Tewksbury, 1999, 2007; Schwartz & Pitts, 1995). This perspective also has been used to explain auto theft (Rice & Smith, 2002), stalking (Mustaine & Tewksbury, 1999), cybercrime victimization (Holt & Bossler, 2009), adolescent violent victimization (Lauritsen et al., 1992), theft (Mustaine & Tewksbury, 1998), victimization at work (Lynch, 1997), and street robbery (Groff, 2007).

Recent research on routines suggests that people may also alter them after being victimized. You may expect that a person who is victimized may engage in more protective behaviors such as installing a burglar alarm following having his or her house broken into or would avoid walking alone at night after being mugged at night. Researchers have investigated whether such changes in behaviors occur. Some of the first works in this area showed that victims had greater use of defensive behaviors (things like avoiding certain areas or people) and that property crime victims engaged in higher use of household protective efforts (such as installing lights and timers) (Skogan, 1987). Victimization has also been linked to moving, which would certainly alter your routines! (Dugan, 1999; Xie & McDowall, 2008). For example, using data from the NCVS, Bunch, Clay-Warner, and McMahon-Howard (2014) found that whereas victims did change some of their behaviors after being victimized compared with nonvictims (such as going out at night more often!), these differences were not due to the victimization event but could be attributed to preexisting differences between victims

and nonvictims that influence victimization risk.

Ripped From the Headlines

At a pre–MTV Video Music Awards party hosted by Chris Brown, multiple gunshots rang out inside a packed club in West Hollywood. Several people were hit, including Suge Knight, the founder of Death Row Records. This was not the first time that Knight has been shot—he was shot at another pre-awards party that was hosted by Kanye West, and he was driving the vehicle when Tupac Shakur was shot and killed (and was shot himself). Although the exact motives behind Knight’s shooting are unknown, the preliminary investigation has linked the shooting to gang ties. Knight has been connected to the Compton’s Bloods gang, and several known members were at the party. Knight also has a criminal history, having spent time in prison for violating probation regarding an assault case and for violating parole when he assaulted a parking lot attendant. Can you apply lifestyles theory to the shooting that occurred at the party?

Source: Branson-Potts (2014).

Structural and Social Process Factors In addition to routine activities and lifestyles theories, other factors also increase a person’s risk of being victimized. Key components of life—such as neighborhood context, family, friends, and personal interaction—also play a role in victimization. To read about how immigration may be linked to victimization read Box 2.1 at the end of this discussion.

Neighborhood Context

We have already discussed how certain individuals are more at risk of becoming victims of crime than others. So far, we have tied this risk to factors related to the person’s lifestyle. Where that person lives and spends time, however, may also place him or her at risk of victimization. Indeed, you are probably not surprised to learn that certain areas have higher rates of victimization than others. Some areas are so crime-prone that they are considered to be hot spots for crime. Highlighted by Lawrence Sherman (Sherman, Gartin, & Buerger, 1989), hot spots are areas that have a concentrated amount of crime. He found through examining police call data in Minneapolis that only 3% of all locations made up most calls to the police. A person living in or frequenting a hot spot will be putting himself or herself in danger. The features of these hot spots and other high-risk areas may create opportunities for victimization that, independent of a person’s lifestyle or demographic characteristics, enhance chances of being victimized.

What is it about certain areas that relates them to victimization? A body of research has

identified many features, particularly of neighborhoods (notice we are not discussing hot spots specifically). One factor related to victimization is family structure. Robert Sampson (1985), in his seminal piece on neighborhoods and crime, found that neighborhoods that have a large percentage of female-headed households have higher rates of theft and violent victimization. He also found that structural density, as measured by the percentage of units in structures of five or more units, is positively related to victimization. Residential mobility, or the percentage of persons 5 years and older living in a different house from 5 years before, also predicted victimization.

Beyond finding that the structure of a neighborhood influences victimization rates for that area, it also has been shown that neighborhood features influence personal risk. In this way, living in a neighborhood that is disadvantaged places individuals at risk of being victimized, even if they do not have risky lifestyles or other characteristics related to victimization (Browning & Erickson, 2009). For example, neighborhood disadvantage and neighborhood residential instability are related to experiencing violent victimization at the hands of an intimate partner (Benson, Fox, DeMaris, & Van Wyk, 2003). Using the notions of collective efficacy, it makes sense that neighborhoods that are disadvantaged are less able to mobilize effective sources of informal social control (Sampson, Raudenbush, & Earls, 1997). Informal social controls are often used as mechanisms to maintain order, stability, and safety in neighborhoods. When communities do not have strong informal mechanisms in place, violence and other deviancy are likely to abound. Such communities are less safe; hence, their residents are more likely to be victimized than residents of more socially organized areas.

Exposure to Delinquent Peers

The neighborhood context is but one factor related to risk of victimization. Social process factors, such as peers and family, are also important in understanding crime victimization. Generally, one of the strongest influences on youth is their peers. Peer pressure can lead people, especially juveniles, to act in ways they normally would not and to engage in behavior they otherwise would not. Having delinquent peers places youth not only at risk of engaging in delinquent behavior—juvenile delinquency does, after all, often take place in groups—but also of being victimized (Lauritsen, Sampson, & Laub, 1991; Schreck & Fisher, 2004). Spending time with delinquent peers places people at risk of being victimized because, as lifestyles and routine activities theory suggests, spending time in the presence of motivated offenders increases risk. Never mind that these would-be offenders are your friends! Another reason having delinquent peers may be related to victimization is that a person may find himself or herself in risky situations (such as being present for a fight) in which being harmed is not unlikely. In this situation, it may not be your friends per se who harm you, but others involved in the fight may attack you, or you may feel the need to come to the aid of your friends. Taylor, Peterson, Esbensen, and Feng (2007) note that being a member of a gang increases a young person’s risk of experiencing violence.

Box 2.1 Immigration and Victimization: Are They Related?

If you have been paying attention to the news, you may have heard people blame the crime problem in the United States on immigrants—legal or otherwise—who have come across our

borders. This argument has been made all the more salient in the wake of mass shootings on our soil like the one that occurred in Orlando, Florida, on June 12, 2016, at Pulse nightclub that resulted in 49 deaths and 53 people injured. Even though the shooter was a U.S. citizen (and was even born here!), some people reinforced their calls for tighter security and reduced ability for people to enter the country. This concern is related to crime that may be committed by people coming to our country, but there is also concern that immigration is related to victimization. There are many reasons to be concerned about the victimization experiences of immigrants. In a study of criminal justice personnel throughout the nation, it was found that these individuals believe that recent immigrants are less likely than others to report their victimization experiences to the police because of language barriers, fear of retaliation, and lack of knowledge about the criminal justice system (Davis & Erez, 1998). Other research has found that increases in immigration are not linked to increases in crime victimization (this study examined immigration in western Europe) (Nunziata, 2015). At the individual level, some research has documented that immigrant youth are at particular risk for being bullied in schools, whereas other research has not found an elevated risk. Still other research has suggested that assimilation is the driving factor behind increased risks for victimization and that lifestyles and routines can help understand this relationship (Peguero, 2013). Immigration and being an immigrant need to be more fully studied to understand if they play a role in victimization risk.

Family

Especially during adolescence, the family also plays an important role in individual experiences. Having strong attachments to family members, particularly parents, is likely to insulate a person from many negative events, including being victimized. Not surprisingly, research has found that weak emotional attachment between family members is a strong predictor of victimization (Esbensen, Huizinga, & Menard, 1999; Lauritsen et al., 1992). This may be due to parents being unable and unwilling to exert control over the behavior of their children, such that they are more likely to end up in risky situations. Family units may also spend more time together when there is strong attachment, thus reducing exposure to motivated offenders. Youth may also be less likely to place themselves in risky situations because they do not want to disappoint their parents, for they place high value on the relationships they have with them. In these ways, emotional attachment to family members serves to reduce risky behavior. At this point, you may be noting that familial attachment may be related to lifestyles and routine activities theories—and you would be right! This relationship is explored, along with how delinquent peers increase victimization risk, by Christopher J. Schreck and Bonnie S. Fisher (2004) in Reading 1 included in this section.

Social Learning Theory According to social learning theory (Akers, 1973), criminal behavior is learned behavior. Specifically, it is learned through differential association (spending time with

delinquent or criminal others) whereby imitation or modeling of behavior occurs. A person learns behavior as well as the definitions about behavior, such as whether it is acceptable to engage in crime. The likelihood that a behavior will persist depends on the degree of reward or punishment. In this way, behaviors are differentially reinforced, and people continue to engage in behaviors that are rewarded and cease to engage in behaviors that are punished. When a behavior is rewarded, the definitions favorable toward

that act will eventually outweigh the definitions against that act. Although this social learning process was originally posited to explain delinquency, it has also been used to explain victimization, especially intimate partner violence in the sense that children who are exposed to violence between parents in the home are more likely to be victims of intimate partner violence than others later in life (see Section VIII for a more detailed discussion). Other research has linked social learning theory to stalking victimization (Fox, Nobles, & Akers, 2011).

Control-Balance Theory A general theory of deviancy, control-balance theory, may also apply to victimization. Developed by Charles Tittle (1995, 1997), this theory proposes that the amount of control that people possess over others and the amount of control to which one is subject factor into their risk of engaging in deviancy. When considered together, a control ratio can be determined for individuals. Control-balance theory posits that when the control a person has exceeds the amount of control he or she is subject to, that person has a control surplus. When the amount of control a person exercises is outweighed by the control he or she is subject to, that person has a control deficit. When a person has a control surplus or deficit, he or she is likely to be predisposed toward deviant behavior. The type of deviant behavior to which a person will be predisposed depends on the control ratio. A control surplus is linked to autonomous forms of deviance such as exploitation of others. Control deficits, on the other hand, are linked to repressive forms of deviance such as defiance.

Although not expressly a theory of victimization, control-balance theory is used by Piquero and Hickman (2003) to explain victimization. They proposed that having a control surplus or control deficit would increase victimization risk as compared with having a control balance. Individuals with a control surplus are used to having their needs and desires met and have a desire to extend their control. In short, they engage in risky behaviors (in terms of victimization) because there is little to restrain their actions. They may treat others who have control deficits with disrespect in such a way that those individuals act out and victimize them. Those with control deficits are at risk for victimization for different reasons. So used to having little control at their disposal, they lack the confidence or belief that they can protect themselves and are, thus, vulnerable targets. They may also try to overcome their control deficits by lashing out or victimizing those who exercise control over them. Piquero and Hickman tested control- balance’s ability to predict victimization and found that both control deficits and control surpluses predicted general and theft victimization.

Social Interactionist Perspective Marcus Felson (1992) posited that distress may be related to victimization. When experiencing stress, peoples’ behavior and demeanor are impacted. People are more likely to break rules and to be generally irritating to others. Distressed individuals, thus, may entice a certain measure of aggression from others given their poor attitudes and rule-breaking behavior. Consider a student who goes to class having just learned that he failed a test in his previous class that effectively ruined his chances of passing that class. This student is likely experiencing a level of stress that will negatively impact his behavior in class. While in class, then, he may explode after a fellow student makes a comment that he finds unreasonable. The student who is the “victim” of the outburst may find the other student’s behavior unacceptable and offensive. The attacked student then may, as a result, respond aggressively, effectively starting an aggressive exchange. This distress-and-reaction sequence is at the heart of the social interactionist perspective.

Stated more formally, M. Felson (1992) argues that aggressive encounters occur when distressed individuals break social rules and those who are aggrieved by the breaking of rules respond aggressively. The distressed individual is then placed in a situation in which he or she has to respond to aggression. If this person does so unsatisfactorily, the original aggrieved person is likely to implement punishment—in other words, victimization. The distressed individual then may retaliate, thus continuing the cycle of aggression. In this way, distress is a cause of victimization.

Life-Course Perspective Emerging in the 1990s in the field of criminology, the life-course perspective considers the development of offending over time. In doing so, it uses elements from biology, sociology, and psychology to explain why persons initiate into, continue with, and desist or move out of a life of crime. Contributing to the growth of this field, in large part due to the overlap between victims and offenders discussed shortly, victimologists have recently begun applying and testing the principles of life-course criminology to victimization. A summary of these theories is presented in Table 2.3.

Table 2.3 Life-Course Criminological Theories Relevant for Victimization

Theory Author(s) Key Factor Related to Outcome

General theory of Gottfredson and

crime Hirschi (1990) Low self-control

Age-graded theory of adult social bonds

Sampson and Laub (1993)

Social bonds: marriage and employment; marriage related to desistance

General Theory of Crime

In 1990, Michael Gottfredson and Travis Hirschi published A General Theory of Crime. In this seminal work, they presented their general theory of crime, proposing that criminal behavior is caused by a single factor—namely, low self-control. They argued that a person with low self-control, when presented with opportunity, will engage in criminal and other analogous behaviors, such as excessive drinking. When examining the characteristics of persons with low self-control, the reasons this trait might lead to criminal behavior are clear. A person with low self-control will exhibit six elements, the first being inability to delay gratification; a person with low self-control will be impulsive and unable or unwilling to delay gratification. Second, the person will be a risk taker who engages in thrill-seeking behavior without thought of consequence. Third, an individual with low self-control will be shortsighted, without any clear long- term goals. Fourth, low self-control is indicated by a preference for physical as compared with mental activity. This preference may lead an individual to respond to disrespect with violence rather than having a discussion about the finer points of being respectful. Fifth, low self-control is evidenced by low frustration tolerance, which results in a person being quick to anger. Sixth, insensitivity and self-centeredness are hallmarks of low self-control. A person with low self-control will be unlikely to exhibit empathy toward others.

Gottfredson and Hirschi (1990) argue that low self-control is fairly immutable once developed, which occurs during early childhood. They believe that, although self- control is an individual-level characteristic, it is not inherent; rather, it is developed through parental socialization. Once the level is set (around age 8), people will be hard- pressed to develop greater abilities to moderate their behavior. Without self-control, a person will act on impulses and seek personal gratification—often engaging in crime. Importantly, as noted, low self-control will lead individuals to engage in other behaviors that are similar to crime.

In 1999, Schreck applied the general theory of crime to victimization. He was one of the first researchers to apply to victimization what had been conceived as a theory of crime. This innovative approach was rooted in his recognition that persons who engage in

crime are also likely to be victimized, a point we return to later. He also noted that because crime and victimization may be closely related, often with the same people engaging in both, the same factors that explain crime participation may also explain crime victimization. He tested his theory and found that low self-control increased the likelihood that a person would experience both personal and property victimization, even when controlling for participation in criminal behavior. This finding suggests that solely being involved in crime does not increase risk of victimization but that low self- control has significant, independent effects on victimization. In a recent meta-analysis, which is a type of study that examines all the research that has been conducted on a subject—in this

case, the link between low self-control and victimization—collectively and produces an effect size for the magnitude of this relationship, self-control was found to have a modest effect on victimization. The overall mean effect size for low self-control on victimization is .154, which means that a one standard deviation increase in low self- control corresponds to a .154 standard deviation increase in victimization. The relationship was strongest for victimizations that were noncontact in nature, such as online victimization (Pratt, Turanovic, Fox, & Wright, 2014). Recent research has linked self-control with neighborhood disadvantage to victimization risk. Chris Gibson’s (2012) research using data from the Project on Human Development in Chicago Neighborhoods is one example. Read this work in Reading 2.

Age-Graded Theory of Adult Social Bonds

Not all criminologists agree that there is a single cause of crime (or victimization) called low self-control. Others noted that people do indeed move in and out of criminal activity, a phenomenon that is difficult to explain with a persistent trait, low self-control. Robert Sampson and John Laub (1993) instead believed that a person’s social bonds could serve to insulate him or her from criminal activity. In their age-graded theory of adult social bonds, Sampson and Laub identified two key social bonds—marriage and employment—that can aid people in moving out of a life of delinquency and crime as they emerge into young adulthood. If a person enters into marriage and has gainful employment, he or she is developing valuable social capital. In other words, a person who has these two social bonds will have much to lose by engaging in crime, which will promote crime desistance if he or she was previously involved in crime. If that person was not involved in crime, social capital would enable him or her to continue living a crime-free life.

Although this obviously is not a victimization theory, because of the link between victimization and offending, researchers have attempted to connect the attainment of adult social bonds with victimization in that individuals who are married and working will be less likely to be crime victims than those with little to lose. Daigle, Beaver, and Hartman (2008) found that entering into marriage did in fact predict desistance from victimization as individuals moved into early adulthood. They found that employment was not similarly protective; instead, employment reduced the chances that a person would desist from victimization. Looking at routine activities and lifestyles theories, however, this finding is none too surprising. The more time a person spends outside the home, at work, or in other activities, the greater the chances of being victimized.

Genes and Victimization

The life-course perspective in criminology has also centered on individual factors, such as genetics, that promote offending. This body of research has found a link between different genetic polymorphisms and behaviors relevant to criminology, such as criminal involvement and alcohol and drug use. A genetic polymorphism is a variant on a gene. Research has shown that sometimes these variations impact the likelihood of engaging in certain behaviors, such as violence, aggression, and delinquency. The genes that have been identified as linked to criminality are those that code for neurotransmitters, such as monoamine oxidase, serotonin, and dopamine. Neurotransmitters are chemical messengers responsible for information transmission. In terms of criminal behavior, relevant neurotransmitters are those linked to behavioral inhibition, mood, reward, and attention deficits. One important aspect of the link between genetics and crime is that possessing a variant for a gene, or having a certain polymorphism for a particular neurotransmitter, appears to “matter” only in certain environments. This is known as a gene x environment interaction. Genes tend to be important not for everyone in every circumstance but for particular individuals in particular contexts. For example, a person who is genetically predisposed toward alcoholism will express these alcoholic tendencies only if first exposed to alcohol.

As noted with other life-course perspective approaches, the applicability of genetic factors to the study of victimization has been explored. In fact, a gene 3 environment interaction for one gene in particular, dopamine, has been found to increase victimization risk. Dopamine is a neurotransmitter linked to the reward and punishment systems of the brain. Dopamine is released when we engage in pleasurable activities, thus reinforcing such behavior. Too much dopamine, however, can be a bad thing. High levels of dopamine are linked to enhanced problem solving and attentiveness, but overproduction of dopamine can be problematic. In fact, it has been linked to violence and aggression. One gene that codes for dopamine is the DRD2 gene, a dopamine receptor gene. Research has found evidence for a gene 3 environment interaction between DRD2 and having delinquent peers. White males with low levels of delinquent peers and who have a certain genetic polymorphism for DRD2 are more likely than others to be violently victimized (Beaver, Wright, DeLisi, Daigle, et al., 2007). Genes have also been implicated in the victim–offender overlap. Research has found that genetic factors account for from 54% to 98% of the covariation between delinquency and victimization (Barnes & Beaver, 2012). The link between genes and victimization is an emerging area of research, and therefore, additional research is certainly needed to understand fully how genes impact victimization.

Role of Alcohol in Victimization One of the common elements present in victimization is alcohol. According to data from

the NCVS in 2008, 36% of victims perceived their offender to be under the influence of alcohol at the time of the incident (Rand, Sabol, Sinclair, & Snyder, 2010). Alcohol use is commonplace among crime offenders, but many crime victims also report that just prior to their victimizations, they had consumed alcohol. Patricia Tjaden and Nancy Thoennes (2006) found in their National Violence Against Women study that 20% of women and 38% of men who experienced rape in adulthood had consumed alcohol or drugs prior to being victimized. Alcohol use is associated with other forms of victimization as well, such as physical assault. This fact should not be too surprising given the effects of alcohol on individuals. Generally, alcohol is linked to victimization because it reduces inhibition and also impedes people’s ability to recognize or respond effectively to dangerous situations. Offenders may also see intoxicated persons as particularly vulnerable targets for these reasons. Where a person consumes alcohol is also important. A person who drinks at home alone or with family is less likely to be victimized than a person who drinks in a bar at night. The latter person is likely interacting with motivated offenders without capable guardianship and may be perceived as a suitable target.

Alcohol use may place a person at risk of being victimized but also may impact how the victim responds to the incident. Research by Ruback, Me´nard, Outlaw, and Shaffer (1999) shows how alcohol use may be relevant to understanding why victims often do not report their experiences to police. In their study, college students evaluated various hypothetical scenarios that depicted victimization. Study participants were asked whether they would advise a victimized friend to report to the police based on a given scenario. When the friend in the scenario had been drinking, college students were less likely to advise that the police be contacted, and this relationship was particularly strong for victims depicted as being underage and drinking.

As you can see, the explanations of victimization are many. The hallmark victimization theory is routine activities and lifestyles theories, based on the notion that a person’s routines and lifestyle, not social conditions, place him or her at risk. As you have read, however, explanations of victimization have expanded beyond this to include social process and structural factors. The explanations you are drawn to may be tied to the data you are examining, which you now know are impacted by methodology. To understand the causes of victimization, you must first know who the “typical” victim is and what characterizes the “typical” victimization. In some of the following sections, specific types of victims are examined. Think about what theories can be used to explain their victimizations.

Summary

The National Crime Victimization Survey (NCVS) uses a nationally representative sample of U.S. households. Individuals ages 12 and over in selected households are asked questions about victimization experiences they faced during the previous 6 months. According to the NCVS, the typical victim is a young White male—although Blacks have higher victimization rates than other racial or ethnic groups and females experience higher rates of rape and sexual assault than males. The typical victimization incident is perpetrated by someone known to the victim, is not reported to the police, and does not involve a weapon. There is a clear link between victimization and offending, as well as between victims and offenders. Persons who live risky lifestyles are more likely to engage in criminal or delinquent activity and to be victims of crime. Victims and offenders also share similar demographic profiles. Routine activities theory suggests that crime victimization is likely to occur when motivated offenders, lack of capable guardianship, and suitable targets coalesce in time and space. Lifestyles theory is closely linked with routine activities theory in proposing that a person who leads a risky lifestyle is at risk of being victimized. Neighborhoods are not equally safe. The risk of being victimized, then, differs across geographical areas, and even when controlling for individual-level factors such as risky lifestyle, neighborhood disadvantage predicts victimization. Spending time with friends who participate in delinquent activities places a person at risk of being victimized. These “friends” may victimize their nondelinquent peers and encourage them to participate in risky behaviors that may lead to victimization. Strong attachments to family may serve to protect individuals from victimization, whereas weak attachments may increase victimization risk. Victimization may also be a learned process, whereby victims have learned the motives, definitions, and behaviors of victimization and had them reinforced. According to control-balance theory, individuals with an unequal control-balance ratio—either having a control deficit or a control surplus—are more prone to victimization than those with a balanced ratio. Those with control deficits may be seen as easy targets. They also may get tired of being targeted and lash out, thus increasing their involvement in situations associated with violent victimization. Those with control surpluses may engage in risky behavior with impunity, which could set them up for being victimized or retaliated against. Research on the general theory of crime suggests that those individuals who have low self-control are more likely to be victimized than those with higher levels of self-control. Adult social bonds may explain why people who were once victimized are not

The Uniform Crime Report (UCR) is an official measure of the amount of crime known to the police. According to this report, which is published annually by the Federal Bureau of Investigation (FBI), the most common crime type is larceny- theft. The most common type of violent crime is aggravated assault. Criminal offending rates are highest for young Black males.

victimized again as they age into young adulthood. Marriage appears to protect individuals from victimization. Genetic factors may also play a role in victimization. One specific genetic polymorphism of the DRD2 gene has been found to increase risk for White males who have delinquent peers. A genetic effect that occurs only under certain environmental conditions is known as a gene × environment interaction. Alcohol and victimization appear to go hand in hand. Alcohol impacts cognitive ability, and persons who are drinking are less likely to assess and recognize situations as being risky even when they are. In addition, alcohol is linked to behavioral inhibition, such that people may act in ways they otherwise would not, which may incite aggression in others. Alcohol is also linked to victimization when offenders purposefully select intoxicated victims because they are seen as easy targets.

Discussion Questions 1. Compare and contrast the UCR and the NCVS. What are the advantages and

disadvantages of each? Which is the best measure of victimization? 2. Apply the concepts of routine activities and lifestyles theories to evaluate your

own risk of being victimized. What could you change to reduce your risk? 3. What are the individual-level factors that place people at risk of being crime

victims? What are the structural factors and social process factors that place individuals at risk of being crime victims?

4. How should the criminal justice system handle victimized individuals who may also have been involved in a crime? Do you believe in the idea of truly innocent victims? If so, should they be treated differently?

5. Given what you have read about the theories and factors that influence crime victimization, how can victimization be prevented? Be sure to tie your prevention ideas to what is thought to cause victimization.

Key terms age-graded theory of adult social bonds 28 bounding 17 capable guardianship 21 “code of the streets” 21 control-balance theory 26 control deficit 26 control ratio 26

control surplus 26 delinquent peers 24 family structure 24 general theory of crime 27 gene x environment interaction 28 hierarchy rule 15 hot spots 23 incident report 17 life-course perspective 27 motivated offenders 21 National Crime Victimization Survey (NCVS) 17 neighborhood context 23 principle of homogamy 22 residential mobility 24 routine activities and lifestyles theories 21 screen questions 17 series victimizations 18 social interactionist perspective 26 structural density 24 suitable targets 21 Uniform Crime Report (UCR) 15 victimization theory 21

Internet Resources “Alcohol and Crime”: http://bjs.ojp.usdoj.gov/content/pub/pdf/ac.pdf

This report by the Bureau of Justice Statistics, in connection with the U.S. Department of Justice, looks at the link between alcohol and crime. It includes several graphs and figures that show the link between crime, specifically violent crime, and alcohol. These statistics also show that alcohol-related crime is generally decreasing.

Bureau of Justice Statistics: Victim Characteristics: http://bjs.ojp.usdoj.gov/index.cfm?ty=tp&tid=92

The NCVS provides information on characteristics of victims, including age, race, ethnicity, gender, marital status, and household income. For violent crimes (rape, sexual assault, assault, and robbery) the characteristics are based on the victim who experienced the crime. For property crimes (household burglary, motor vehicle theft, and property theft) the characteristics are based on the household of the respondent who provided information about these crimes. Property crimes are defined as affecting the entire household.

Crime in the United States: The Nation’s Two Crime Measures: http://www.fbi.gov/about- us/cjis/ucr/crime-in-the-u.s/2010/crime-in-the-u.s.-2010/the-nations-two-crime-measures

This website is part of the FBI’s research on various crimes. This one specifically examines the differences, advantages, and disadvantages of the UCR and the NCVS. Both forms of research are important to the study of crime.

“Opportunity Makes the Thief: Practical Theory for Crime Prevention”: http://webarchive.nationalarchives.gov.uk/20110218135832/rds.homeoffice.gov.uk/rds/prgpdfs/fprs98.pdf

This article combines several theories that focus on the “opportunity” of crimes. This includes the routine activities approach, the rational choice perspective, and crime pattern theory. This publication argues that the root cause of crime is opportunity. This allows for prevention techniques to focus on how to lessen the opportunity for crime to occur.

Project on Human Development in Chicago Neighborhoods: http://www.icpsr.umich.edu/PHDCN/about.html

The Project on Human Development in Chicago Neighborhoods is an interdisciplinary study of how families, schools, and neighborhoods affect child and adolescent development. It was designed to advance understanding of the developmental pathways of both positive and negative human social behaviors. In particular, the project examined the pathways to juvenile delinquency, adult crime, substance abuse, and violence. At the same time, the project also provided a detailed look at the environments in which these social behaviors take place by collecting substantial amounts of data about urban Chicago, including its people, institutions, and resources.

How to Read a Research Article Let’s use one of the articles in Section II, “Specifying the Influence of Family and Peers on Violent Victimization: Extending Routine Activities and Lifestyles Theories” by Christopher J. Schreck and Bonnie S. Fisher, as an example for the following steps.

1. What is the thesis or main idea of this article? The thesis or main idea is found in the introduction of this article. Although it is not directly stated, in the final paragraph of the introduction, Schreck and Fisher note what they are testing. That is, they are examining whether the family and associating with delinquent peers impacts juveniles’ level of violent victimization.

2. What is the hypothesis? Schreck and Fisher identify two hypotheses, although these hypotheses are not specifically

stated. You can identify the general hypotheses by reading the introduction. The first hypothesis is located in the second paragraph. The authors suggest that a strong family will reduce the likelihood that an adolescent will experience violent victimization. The second hypothesis is discussed in the following paragraph. The authors hypothesize that membership in delinquent peer groups will lead to increased risk of violent victimization.

3. Is there any prior literature related to the hypothesis? Literature linking victimization to offending is cited throughout the introduction. In addition, Schreck and Fisher use routine activities/lifestyles theory to justify why the family and peer group should be considered when trying to understand violent victimization risk among adolescents. They link the elements of routine activities theory to family and peers and explain how these activities impact violent victimization risk.

4. What methods are used to test the hypothesis? Schreck and Fisher use survey research to test their hypotheses. They used data from the National Longitudinal Study of Adolescent Health, which had been collected by other researchers. This is known as secondary data analysis. They used a type of regression analysis called overdispersed Poisson regression as their analytic tool to examine factors related to higher levels of violent victimization.

5. Is this a qualitative or a quantitative study? Qualitative studies are those that use nonnumerical data or methods, while quantitative studies use numbers to test hypotheses. Schreck and Fisher use statistics to test their hypotheses; therefore, their study is quantitative.

6. What are the results, and how does the author present them?

1. The results of the study are presented in the “Results” section. The authors present their results in two forms—in a table (Table 1) and in the write-up of the results section. If the statistical technique is one with which you are unfamiliar, the table can be examined by finding those variables that have asterisks by their coefficients or values. The asterisks indicate that the variable is significant in the model. Those variables’ coefficients or values that do not have asterisks beside them are not significant. The write-up also will tell you what variables are significant and how to interpret the values. Schreck and Fisher found that “teenagers whose friends engaged in higher levels of delinquency and risky activity tended to experience more victimization.” Spending time with those friends also increased risk of experiencing violence. Two family context variables were related to violent victimization. A warm, accepting family climate and parents’ positive feelings toward the adolescent reduced risk of violent victimization.

2. Do you believe that the authors provided a persuasive argument? Why or why not?

The answer to this question is subjective; however, the authors do make a strong case for examining both the peer and family context in risk for violent victimization. A close reading of the front end of the paper provides the reader the basis for the argument. The analytic test shows support for their hypotheses—variables measuring both peer and family context do “matter” in predicting risk of violent victimization. It should be noted, though, that variables operationalizing the routine activities/lifestyles perspective were also shown to be relevant (they were significant). This finding suggests that both the routine activities/lifestyles perspective and the peer and family context should be further examined.

3. Who is the intended audience of this article? As you read the article, you should consider to whom the authors are speaking. Most academic articles have professors as their audience, of course, but they are also often written for others. Schreck and Fisher are writing for students, practitioners, and policymakers as well as for professors. When determining who the audience is, you should ask yourself who would benefit from reading the article. Also, consider for whom the findings would be beneficial.

4. What does the article add to your knowledge of the subject? This question is obviously specific to the reader. One way to answer this question may be as follows: This article helps the reader understand how the peer and family context influence risk for violent victimization. It also helps the reader understand how the impact of peer and family variables on

the risk of violent victimization can be explained through a routine activities/lifestyles theory approach.

5. What are the implications for criminal justice policy that can be derived from this article?

Implications for criminal justice policy are likely to be found in the discussion or conclusion sections of the article. In the conclusion, Schreck and Fisher discuss the implications of their findings mainly in terms of how future research and theory may be impacted. They do, however, question policies intended to reduce victimization by educating potential victims about ways to reduce their risk, since it is possible that victimization risk is tied to family attachment, which they link to socialization. Socialization through the family via its effect on the development of self-control has been previously linked to victimization risk. Schreck and Fisher suggest that victimization may be linked to something that cannot be easily changed (i.e., low self-control) rather than something that can be easily manipulated (i.e., a person’s routine activities). In this way, policy implications are discussed. You will notice that some articles have more criminal justice policy implications.

Now that we have examined the elements of a research article, it is your turn to answer these questions as you read through the various articles in the text. Some of the articles may be easier for you to read than others, and not all the articles will have all the elements you just read about: introduction, literature review, methods, results, and discussion. You should refer to this introduction on how to read a research article for guidance if you have any problems.

Reading /// 1

In this article, Christopher Schreck and Bonnie Fisher (2004) examined how both the family and peers impact a person’s likelihood of experiencing a violent victimization. Four types of violent victimization were examined: being threatened with a knife or gun, being shot, being stabbed, or being “jumped.” They also examined variables designed to measure risky lifestyles. In doing so, they use data from the National Longitudinal Study of Adolescent Health on more than 3,500 respondents. The authors found that the lifestyle factors of sneaking out of the house, driving around in a car, and exercising increased the chances of a person being victimized. They also found that general feelings the respondent had for his or her family were negatively related to victimization—the better a person felt about his or her family, the less likely he or she was to be victimized. Finally, they found that peer delinquency and spending time with peers both increased victimization risk. Taken together, they found support for three of the theories or risk factors identified in this chapter: routine activities/lifestyle, family, and delinquent peers.

Specifying the Influence of Family and Peers on Violent VictimizationSpecifying the Influence of Family and Peers on Violent Victimization

Extending Routine Activities and Lifestyles Theories

Christopher J. Schreck and Bonnie S. Fisher

Criminologists have long been interested in family conditions as an antecedent of youthful delinquent behavior (Farrington, 1987; Glueck & Glueck, 1950; Gottfredson & Hirschi, 1990; Loeber & Stouthamer-Loeber, 1986; Rebellon, 2002; Sampson & Laub, 1993). As important as family context is in the delinquency research, the family may have broader significance in view of the finding that delinquents tend to experience high levels of violent victimization (Lauritsen, Sampson, & Laub, 1991). This connection between involvement in crime and the experience of victimization has led some criminologists to propose that established correlates of crime—which might include family characteristics—are also relevant for understanding victimization (Piquero & Hickman, 2003; Schreck, 1999; Schreck, Wright, & Miller, 2002). In addition, it appears reasonable to think that families have an interest in preventing their children from being threatened, beaten up, or robbed. Consequently, family context may be particularly important among teenagers, who bear a disproportionately high level of victimization relative to every other age group (Bureau of Justice Statistics, 2003).

In contrast to the longstanding tradition of family and delinquency research, there have been few attempts by researchers to explore a connection between family context and the experience of general violent victimization. The few studies to consider the relationship between family context and general victimization found that level of emotional attachment between family members is perhaps the strongest family-relevant predictor of victimization (Esbensen, Huizinga, & Menard, 1999; Lauritsen, Laub, & Sampson, 1992). Consistent with these studies, we argue that a strong family—one that

possesses durable ties of emotional warmth and support—might reduce the chances that adolescents will experience violent victimization.

The primary competitor to the influence of the family is most likely the adolescent peer group, which criminologists and victimologists find to be a risk factor for criminal behavior as well as victimization (Akers, 1985; Lauritsen et al., 1992; Schreck et al., 2002; Warr, 2002). The peer group represents one of the primary social contexts in the lives of adolescents, aside from the family (Coleman, 1980; Corsaro & Eder, 1990). At the same time, the presence of criminal associates in a teenager’s life is traditionally one of the strongest and most consistent correlates of individual delinquency (Akers, 1994; Hindelang, Hirschi, & Weis, 1981; Warr & Stafford, 1991; c.f. Haynie, 2001). Membership in peer groups, however, also corresponds with a higher risk of victimization. Researchers have theorized that membership in a delinquent group can invite retaliation from the group’s victims or else group norms might favor the use of violence as an accepted means of settling within-group disputes (Baron, Forde, & Kennedy, 2001; Baron, Kennedy, & Forde, 2001; Decker & Van Winkle, 1996; Singer, 1981). In addition, proximity to other participants in crime leads members of the peer group to use each other as convenient targets (Jensen & Brownfield, 1986; Lauritsen et al., 1991; Schreck et al., 2002). Even research looking exclusively at criminal youth gang members found that membership is a source of risk rather than protection (Miller & Decker, 2001; Sanders, 1994). Taken together, attempts at understanding the sources of adolescent victimization cannot afford to ignore risk factors originating in the peer context. Nevertheless, the literature has not given much attention to the effectiveness of family context as a predictor of victimization after controlling for delinquent peer group characteristics (see Esbensen et al., 1999; Lauritsen et al., 1992).

In view of the importance of the family and peer group for delinquency causation, both appear to have significance beyond simply being precursors of juvenile delinquency. This article explores family and peer contexts and examines how these contexts link to routine activities/lifestyles theories of victimization.

Using data from the National Longitudinal Study of Adolescent Health (Add Health), we then empirically tested whether the combined influence of family and delinquent peer-group associations significantly affects adolescents’ level of violent victimization. One important methodological advantage of the Add Health is its national coverage of adolescents. The Add Health also offers a significant improvement over other major data sets (such as the National Youth Survey or the Monitoring the Future study) in that the staff collected data from not only the respondent and his or her parents (which would be relevant for measuring family attachment) but also from the respondent’s peers. This data set thus provides a unique opportunity to measure the influence of peer context in

conjunction with family characteristics.

Source: Schreck, C. J., & Fisher, B. S. (2004). Specifying the influence of family and peers on violent victimization: Extending routine activities and lifestyles theories. Journal of Interpersonal Violence, 19, 1021–1041.

Theoretical Framework Our general framework for explaining how family and peer contexts might influence victimization draws from the routine activities/lifestyles theories (Cohen & Felson, 1979; Hindelang, Gottfredson, & Garofalo, 1978). Together, these theories stress how situations or contexts carry their own level of risk for victimization. The routine activities theory maintains that the convergence in time and space of motivated offenders, attractive targets, and ineffective guardianship determines the risk of victimization. People who spend a lot of time in settings where they are exposed to offenders and cannot adequately protect themselves will tend to have a higher risk of victimization. Social structure and demographic characteristics generally affect daily routines and, thus, exposure and vulnerability to likely offenders (Hindelang et al., 1978; Miethe, Stafford, & Long, 1987; Sampson & Wooldredge, 1987). Family and peer contexts could also be important mechanisms for the meeting of motivated offenders and worthwhile, vulnerable targets; that is, social bonds with family and peers might structure or determine routine daily activities that have relevance for victimization risk (Felson, 1986; Horney, Osgood, & Marshall, 1995; Schreck et al., 2002).

Attachment to the Family and Violent Victimization

The bond of attachment attends to the degree of sensitivity and emotional closeness that people have for others (see Hirschi, 1969). Although the level of emotional closeness with others might directly influence risk, the connection between bonds, similar to attachment, and routine

activity is better documented (Felson, 1986; Schreck et al., 2002); that is, attachment could serve to constrain the types of activities we routinely engage in and bring individuals into closer proximity to those with an interest or social obligation to act as protectors. The following discussion thus shows how attachment to the family has relevance for each of the three necessary elements for crime: effective guardianship, exposure to motivated offenders, and target attractiveness.

Family context and exposure to motivated offenders

Bonds to family members should remove would-be victims from the proximity of likely offenders. Strong attachment for parents should tend to keep children closer to home and away from the company of strangers. Felson (1998) noted that although some people have family members who are unquestionably dangerous to be with, it is typically much more dangerous to spend time around strangers and outside the home (see also, Bureau of Justice Statistics, 2003; Hindelang et al., 1978). In addition, strong feelings of attachment might compel parents to regulate their children’s friendship groups (which includes getting to know the friends, including them in family activities, and getting to know their families). As noted earlier, the delinquent peer group may well represent a pool of unusually motivated offenders. In short, attachment keeps children closer to home, makes parents interested in the friends of their child, and thus keeps children away from motivated offenders.

Family context and guardianship

Strong bonds represent a catalyst for the meeting, in time and space, of potential victims with effective guardians. Gottfredson and Hirschi (1990) maintained that strong bonds of parental attachment to the child are a prerequisite for effective supervision and, presumably, protection; that is, one might expect that parental presence would make criminal activity against their offspring inconvenient. One may also assume that children who are attached to their parents will choose to spend more time with them enjoying their protection than children who dislike or distrust their parents. Attachment can also lead to improved guardianship in ways other than direct supervision. Given the scattered nature of informal control in modern American society (see Felson, 1998), with parents who work and students who go to school and who may work themselves, parents can ensure guardianship by arranging for surrogate parents, such as child care providers or coaches. Thus, although strong attachment cannot guarantee that parents will always be able to immediately supervise their children, attachment appears more likely to facilitate the direct presence of alternative guardians.

Family context and target attractiveness

Bonds might create proximate so-called handlers, or people who are socially obligated to prevent likely victims from making themselves into attractive targets (for more on handlers, see Felson, 1986); that is, effective handlers will strive to prevent those they feel responsible for from behaving provocatively and risking retaliation, which could happen if the child is behaving belligerently or insolently to others or is victimizing them. Handlers may also be in a position to require precautions intended specifically to promote safety (e.g., “Come straight home after school” or “Make sure you take the bus; don’t walk home by yourself”).

Social bonds aside, we would expect that deviant parents will be less capable of protecting their children from crime. Gottfredson and Hirschi (1990) suggested that such parents are little inclined to exert themselves monitoring their children. Beyond the effort and time that effective supervision entails, parents who impair their judgment by consuming illegal drugs or drinking alcohol to excess will tend to be less effective as protectors. Deviant parents may also engage in illegal behavior against their own children and represent proximate likely offenders in their own right. The presence of two married adults in the household may also be important, as researchers have speculated that family disruption makes supervision more difficult as well as interferes with the maintenance of bonds with children (Hirschi, 1991; Rutter & Giller, 1983).

Peer Context and Violent Victimization

Although adolescent life typically revolves around the family, the peer group represents a second major context. We have already noted the fact that teenagers spend considerable amounts of time engaged in social activity with their friends. As is the case with the family, the routine activities/lifestyles approaches can offer insights about why contact with the peer group, most notably the delinquent peer group, also suggests exposure to motivated offenders, weakened guardianship, and increased attractiveness as a target.

Delinquent Peer Association and Exposure to Motivated Offenders

Although routine activities theory does not attempt to explain why people are motivated to commit crime, research reveals that there are individual differences in how easy it is to tempt someone into crime (Felson, 1998; Gottfredson & Hirschi, 1990). To illustrate, exposure to persons with a record of delinquency—particularly when one is vulnerable —would therefore be risky. The principle of homogamy in lifestyle theory (Hindelang et al., 1978, pp. 256–257) addresses this issue, positing that individuals who share daily

living conditions and social activities with high-risk offenders (because they possess similar demographic backgrounds) will have a disproportionately high risk of victimization (see also, Miethe & Meier, 1994). Ties of friendship to a delinquent would therefore make it more likely that someone will routinely be in proximity to a motivated offender and vulnerable. If most crimes do not require significant departures from the normal routines of the offender, the closeness of a delinquent’s friends would make them accessible targets (see Felson, 1998).

Delinquent peer association and guardianship

In light of the amount of time teens spend with their peer group, other members of that group are potentially well situated to act as guardians and protect their peers. This role for the peer group presupposes that one’s associates are interested in serving as guardians. Guardianship, however, potentially necessitates time, effort, and risk. Felson (1986, 1998) suggested that strong social bonds create a stake in protecting others. These bonds influence the amount of time people spend in risky activity away from those who can protect them from the criminal inclinations of others; that is, caring about others may make one spend more time with them, which promotes personal enjoyment and safety from victimization.

The delinquent peer group, on the other hand, undermines guardianship in multiple ways. First, Gottfredson and Hirschi (1990) characterized delinquents as lacking diligence and being impulsive and selfish, thus suggesting that they are unlikely to exert themselves to protect their closest associates when their associates are vulnerable. Second, as Sparks (1982) indicated, someone in a delinquent group would be unlikely to report victimization to the police and can therefore count on having less police protection. Part of this may reflect the delinquent’s understandable lack of comfort in dealing with legal authorities. For instance, delinquents might assume that the police will be quick to dismiss their claims because of their connection to other troublemakers, or else might run a higher risk of implicating themselves merely by talking to the police.

Delinquent peer association and target attractiveness

As noted in criminological research, children who participate in youth gangs and who claim the friendship of other delinquent youth are much more likely to participate in crime themselves. In fact, juvenile delinquency frequently occurs as a group activity (Erickson & Jensen, 1977; Warr, 1996). A number of researchers have suggested that participation in crime—or even of merely identifying with a delinquent group—can make a youth a worthwhile target for others. Singer (1981) proposed that retaliation is part of the normative characteristics of a violent subculture (see also, Baron, Forde, et

al., 2001). Someone thus committing a crime or who is a member of a group responsible for committing a crime might thus alternate from offender to victim when the former victim attempts to retaliate. Youth groups that encourage aggressive behavior may likewise make themselves attractive targets by precipitating violence through their confrontational behavior.

Research Questions There is good reason to think that families with conventional parents and where strong bonds are present between family members will be more effective at promoting the safety of offspring. At the same time, the prominence of the peer group in the lives of adolescents represents a plausible source of risk. More specifically, do stronger bonds of attachment between family members—especially between parents and children— promote greater safety from violent victimization? Does spending time with friends, and having delinquent friends, increase risk?

Most existing data sources survey the victimization experiences and traits of only the adolescent respondent, and not their parents or members of their peer network. The National Longitudinal Study of Adolescent Health (Add Health), however, integrated information from each of these three sources—the adolescent, parents, and

friends—which provides more valid measures than would be the case if one relied on data from only the teenagers. This analysis below, which uses these data, considers the relative influences of family context and peer group factors on the level of violent victimization.

Data and Methods The current research used the first wave of the public-use version of the Add Health study, which was conducted between September 1994 and December 1995. This wave consisted of in-school and in-home administrations. The goal of the Add Health study was to create a sample that is nationally representative of students attending U.S. schools in Grades 7 through 12. Our analyses excluded cases with any missing data, leaving slightly more than 3,500 respondents who provided sufficient data for analysis.

Dependent Variable—Violent Victimization

To measure the occurrences of violent victimization for each individual, we created an additive index. Respondents were asked about having been the victim of four types of serious violence. These items each measure serious forms of violence: (a) threat with a knife or gun, (b) shooting, (c) stabbing, and (d) being “jumped.” The reference period was the 12 months preceding the survey. The Add Health survey recorded responses using ordinal categories (never = 0, once = 1, more than once = 2). As expected, the distribution of violent victimization is skewed; nearly 80% of the respondents were not victims of any of the violent crimes noted above during the reference period, approximately 10% reported one instance of victimization, and the remainder accounted for two or more incidences of victimization.

Independent Measures

Family context

There were six family-related measures included in the analysis. The first measure, welfare receipt (coded 1 = yes, 0 = no), measures the economic condition of the family. Less than 10% of the sample reported receiving any public assistance. The second measure, parental drinking, measures how often parents consumed five or more drinks in the previous month (coded 1 = never through 6 = five or more times). We used this variable as a proxy for deviant parents. The median parent reported never consuming five or more drinks. The third item, positive parental feelings toward child, is a

composite index of four items measuring the parent’s opinions about the child, with a higher positive score indicating warmer feelings toward the child. Because the component items in this index had different metrics, we converted the raw item score indicating more positive feelings toward the child. The fourth measure is a two-item index tapping the child’s attachment toward the mother, with a higher score indicating stronger attachment (coded 1 = not at all through 5 = very much). The median score was 5, indicating that the typical child is strongly attached to his or her mother. The fifth item is also a two-item index made up of similar questions but measuring the child’s attachment toward the father (median = 4.5). The final item, family climate, is a four- item index measuring the general feelings of the respondent for the family (as opposed to the parents specifically). Descriptive statistics indicate that the average respondent lived in a family characterized by closeness between the members (median = 4).

Peer delinquency

The friends located in the respondent’s peer network reported participation in five minor types of delinquency: (a) cigarette smoking, (b) drinking alcohol, (c) getting drunk, (d) skipping school, and (e) doing risky things on a dare. All items had the following response categories: 0 = not at all, 1 = once or twice, 2 = 3 or 4 times, and 3 = 5 or more times. The Add Health survey designers then computed the average level of delinquency among those peers nominated by the respondent, or who nominated the respondent, and recorded them on the respondent’s record. We summed the averages for the five items to create a peer delinquency index. The average respondent had a peer network that engaged in these delinquent or risky activities approximately 4 times during the reference period. Most members of the sample had friends who reported relatively low levels of delinquency, averaging four different acts of delinquency committed once or twice in the previous 12 months.

Peer context

Two variables measure additional dimensions of peer context. The first variable, friends care, measures the feeling that friends have about the respondent.

Scores range from 1 (not at all) to 5 (very much). The second variable, activities with friends, reflects the amount of time the respondent spends simply hanging out with friends (coding ranges from 0 = not at all to 3 = 5 or more times per week). The average respondent felt that friends generally cared (mean = 4.29) and reported going out with friends 3 or 4 times per week.

Lifestyles

We also employed several activities that bivariate analysis indicated as being correlated with victimization: sneaking out of the house at night, driving a car, and exercising. Each of these items represents activity outside the home, which should elevate victimization risk. One might reasonably interpret each of these activities as events that are likely to occur away from parental guardians and home, and therefore likely to incur additional risk.

Controls

The multivariate data analyses include several demographic control variables. Gender represents the effect of being a male respondent (with female respondents being the reference category). Males constituted 47% of the sample. Racial identity is divided between Black and non-Black (reference), with 25% of the sample being Black. The variable for respondent age measures age in discrete years, ranging from 11 to 20. Each of these variables is a correlate of violent victimization (see, e.g., Bureau of Justice Statistics, 2003; Cohen, Kluegel, & Land, 1981; Hindelang et al., 1978).

Results We were interested in learning whether family and peer-related variables predict violent victimization and significantly improve prediction beyond baseline demographic variables. In view of the highly skewed distribution of the dependent variable, we use an overdispersed Poisson regression model to estimate the effect coefficients of the independent variables. The first column of coefficient in Table 1 (i.e., Model 1) represents the baseline model. As expected, minorities tend to have significantly higher levels of victimization, as do male and younger members of the sample. Model 1 also includes the three indicators of unsupervised and unstructured activities. Consistent with the finding of Schreck and colleagues (2002), the data show that teenagers who engage in leisure activities away from home (and, in many cases, away from supervision) also tend to have higher levels of violent victimization.

Model 2 incorporates the family-related variables into the analysis. The strongest family-related predictors of violent victimization are family climate and parental feelings for their children. Adolescents living in households where there is a warmer, more accepting climate experience less violence, while those living in families where parents dislike them have a greater risk for becoming victims. This supports our explanation for children as well as reduced exposure to motivated offenders. Some types of family closeness do not appear to matter, however. The attachment of child to mother or father did not significantly affect victimization risk. Because attachment from the parent to the child is significant, this finding suggests that parents have the more central role when it comes to protecting teenagers from violence. More interesting, drinking activity by parents failed to significantly influence victimization risk. In addition, incorporating the family-relevant variables did not substantially alter the baseline variable coefficients, thus indicating that demographic and lifestyle effects are independent of the family context variables used in the analysis.

Model 3 adds the peer context variable. Peer delinquency and activities with friends were the significant peer-context predictors. Teenagers whose friends engaged in higher levels of delinquency and risky activity tended to experience more victimization. Readers should be aware, however, that the peer delinquency measures tap minor forms of delinquency, and that a measure for more serious forms could yield a larger influence. This finding supports our explanation that those who associate with delinquent peers risk enhanced exposure to motivated offenders, ineffective guardianship, and suitability as a target for violence. In addition, spending time with friends—whether or not they are delinquent—also increases the risk of experiencing violence. Having friends who care about the respondent, on the other hand, did not significantly influence level of violent victimization. Peer context, however, predicted victimization independently of demographic, lifestyle, or family context variables—the effect coefficients of these variables, in fact, did not seem much affected by the inclusion of the peer context variables. That is, the effects of peer context do not seem to detract from the influence of family variables; each appears to predict violent victimization independently. In addition, the continued strength of demographic predictors even in the full model indicates that researchers looking to explain demographic variation in violent victimization would have to look at risk factors other than the ones considered in the analysis.

Table 1 Poisson Regression Estimates for Violent Victimization

Model 1 Model 2 Model3

b β b β b β

Intercept −1.72 — −.47 — −.40 —

Demographic

Male .90*** .25 .98*** .27 .95*** .26

Black .42*** .10 .35*** .08 .41*** .10

Age −.05 −.41 −.06* −.49 −.08* −.65

Lifestyles

Driving around .35* .08 .32* .08 .30* .07

Sneaks out 1.03*** .18 .78*** .13 .74*** .13

Exercises .15*** .09 .15*** .09 .15*** .09

Family context

Welfare receipt — — .32 .05 .31 .04

Climate — — −.26** −.10 −.25** −.09

Parent drinks — — .04 .02 .03 .01

Parents’ feelings for child — — −.06** −.09 −.05* −.07

Attached to mother — — −.02 −.01 −.02 −.01

Attached to father — — −.03 −.03 −.02 −.02

Peers

Peer delinquency — — — — .03** .06

Friends care — — — — −.04 −.02

Activities w/ friends — — — — .10* .05

–2 Log Likelihood 2097.76 — 2091.52 — 2083.46 —

Generalized R2 .08 — .10 — .11 —

*p < .05.

**p < .01.

***p < .001.

Conclusion What relevance do family and peer variables have for general victimization risk among adolescents? Building from the similarities between the correlates of crime and victimization, the importance of the family and peer contexts in the etiology of delinquency indicates that victimization research might benefit from exploring these contexts further. Moreover, the centrality of family and peers in the lives of teenagers and the likely influence these contexts have on the convergence of motivated offenders, attractive targets, and guardianship further recommend research on family and peer environments.

In the current study, we linked family and peer context to routine activity/lifestyles theories. In the case of the family, we hypothesized that strong bonds of attachment decrease the likelihood that meetings between motivated offenders and attractive and vulnerable targets will occur. To test, we measured attachment in several ways: Attachment included (a) general feelings of closeness with the family, (b) feelings of warmth toward parents, and (c) parental feelings toward the child. Our results indicated that teenagers in households where the family context is characterized by closeness and understanding tend to be safe from violent crime. Likewise, children whose parents are emotionally alienated from them tend to experience a higher risk of victimization. Given the preliminary nature of research linking family and peer contexts to victimization, readers should be aware that although the results are consistent with the expectations of routine activity/lifestyle theories (that family climate and positive parental feelings for

offspring appear to promote better guardianship, increase distance from motivated offenders, and prevent children from becoming attractive targets), the interpretation we offer might not be the only one. For instance, strong bonds of attachment might also create unwillingness to take risks that might lead to victimization (e.g., by behaving belligerently or insultingly toward others) because victimization jeopardizes the happiness of loved ones; that is, attachment might show how much more one has to lose through victimization and thus deter provocative behavior. Further research about the nature of the attachment-victimization link might clarify this issue.

The presence of delinquent peers appears to be a risk factor, in addition to family-related correlates. The results support the interpretation that spending time around delinquent peers is a risky venture. First, group delinquency, or identification with delinquents, can invite retaliation (Baron, Forde, et al., 2001; Singer, 1981). Affiliation with delinquent peers also suggests exposure to motivated offenders (Lauritsen et al., 1991; Schreck et al., 2002). Second, delinquents may exacerbate risk because they are not the most effective of guardians for their friends. We should note that the magnitude coefficients in our estimated model appear somewhat weaker than those coefficients reported in Lauritsen et al. (1992). This may be a reflection of the fact that the National Youth Survey asked respondents to report on the behavior of their peers, whereas the Add Health asked the peers themselves. Haynie (2001) found a similar reduction in effect magnitude, although in the case of delinquent behavior. As is the case with the family, the relation between peer group affiliation and general risk of victimization deserves further investigation.

The findings also revealed that demographic variables remain important predictors net of the routine activities/lifestyles, family, and peer variables. Lifestyles theory was originally developed to make sense of demographic differences in the risk of victimization; however, measures of lifestyles have had very limited success mediating the effects of demographic variables. Our research shows further that demographic differences appear to come from sources other than family and peer factors as well. Clearly, there is further room for improvement in our attempts to explain gender and racial differences in violent victimization.

The research on family and peer correlates of delinquency is too numerous to easily cite here, yet parallel research linking family and peer characteristics to general violent victimization has been comparatively scant. Our research indicates that these two contexts of adolescent life might be useful for understanding variation in risk of violent victimization. This importance suggests several new areas of inquiry for theory and research. The first new direction follows from the findings about family attachment. If the family determines vulnerability, then could family characteristics socialize children to be either victims

or nonvictims in the long term? The recent extension of self-control theory to victimization suggests that this might be the case; however, because Gottfredson and Hirschi (1990) posited that differences in self-control stabilize in late childhood, the Add Health sample is not young enough to address this question about the long-term effects of family characteristics. Thus, the failure of the parental drinking measure to significantly influence risk might not say anything about the empirical status of self- control theory. Nevertheless, the possibility that vulnerability to crime is a time-stable personal trait clearly casts into question policies intended to reduce victimization by educating potential victims about what they can do to protect themselves from crime. This possibility also makes sense of the fact that crime prevention educational programs have been ineffective at making students safer from victimization (Finkelhor, Asdigian, & Dziuba-Leatherman, 1995). The family may therefore have broader relevance than the immediate-situation importance that we suggest here.

The second new direction relates to our results regarding the peer effect finding. Some of the major criminological theories have built themselves around explaining how peer associations cause crime (Akers, 1985). We explicitly adopted a situational interpretation for the peer effect on victimization; however, the extension of criminological theories that stress the socializing role of peer associations might offer new insights about how peer group membership matters when it comes to victimization. Although we do not necessarily endorse this approach, further extensions of criminological theory to victimization can add richness to an area that has long suffered from a lack of theoretical innovation and diversity. The vast literature on peers and delinquency, and the substantive debates about the meaning of the peer effects, may well apply to victimization. Even though our analyses did not exhaust the full range of risk factors for violent victimization, the results suggest avenues for further enriching theories of victimization.

Discussion Questions 1. What role do you think family and peers have on victimization of young adults?

What about older adults? Are these constructs relevant for understanding the victimization of persons who are not teens?

2. What are the policy implications of the findings that peers and family both matter in the occurrence of victimization?

3. How might the construct of low self-control impact the relationship among family, peers, and victimization? Is it possible that persons with low self-control may seek out delinquent peers? Given what you know about the development of self- control, how might the family be related to both self-control and victimization?

4. What other factors not identified in this article or in this section might be related to victimization? Why do you think the factors you identified would be important in understanding why some people are victimized?

References Akers, R. L. (1985). Deviant behavior: A social learning approach. Belmont, CA:

Wadsworth.

Akers, R. L. (1994). Criminological theories: Introduction and evaluation. Los Angeles: Roxbury.

Baron, S. W., Forde, D. R., & Kennedy, L. W. (2001). Rough justice: Street youth and violence. Journal of Interpersonal Violence, 16, 662–678.

Baron, S. W., Kennedy, L. W., & Forde, D. R. (2001). Male street youths’ conflict: The role of background, subcultural, and situational factors. Justice Quarterly, 18, 759–790.

Bureau of Justice Statistics. (2003). Criminal victimization in the United States: 2001 statistical tables. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.

Cohen, L. E., & Felson, M. (1979). Social change and crime rate trends: A routine activity approach. American Sociological Review, 52, 170–183.

Cohen, L. E., Kluegel, J. R., & Land, K. (1981). Social inequality and predatory criminal victimization: An exposition and test of a formal theory. American Sociological Review, 46, 505–524.

Coleman, J. S. (1980). Friendship and the peer group in adolescence. In J. Adelson (Ed.), Handbook of adolescent psychology (pp. 408–431). New York: John Wiley.

Corsaro, W., & Eder, D. (1990). Children’s peer cultures. Annual Review of Sociology, 16, 197–220.

Decker, S. H., & Van Winkle, B. (1996). Life in the gang. Cambridge, UK: Cambridge University Press.

Erickson, M. L., & Jensen, G. F. (1977). Delinquency is still group behavior!: Toward revitalizing the group premise in the sociology of deviance. Journal of Criminal Law and Criminology, 68, 262–273.

Esbensen, F., Huizinga, D., & Menard, S. (1999). Family context and criminal victimization in adolescence. Youth & Society, 31, 168–198.

Farrington, D. P. (1987). Early precursors of frequent offending. In J. Q. Wilson & G. C. Loury (Eds.), From children to citizens, Volume III: Families, schools, and delinquency prevention (pp. 27–50). New York: Springer-Verlag.

Felson, M. (1986). Linking criminal choices, routine activities, informal control, and criminal outcomes. In D. B. Cornish & R. V. Clarke (Eds.), The reasoning criminal: Rational choice perspectives on offending (pp. 119–129). New York: Springer-Verlag.

Felson, M. (1998). Crime and everyday life. Thousand Oaks, CA: Pine Forge Press.

Finkelhor, D., Asdigian, N., & Dziuba-Leatherman, J. (1995). The effectiveness of victimization prevention instruction: An evaluation of children’s responses to actual threats and assaults. Child Abuse and Neglect, 19, 141–153.

Glueck, S., & Glueck, E. (1950). Unraveling juvenile delinquency. New York: Commonwealth Fund.

Gottfredson, M. R., & Hirschi, T. (1990). A general theory of crime. Palo Alto, CA: Stanford University Press.

Haynie, D. L. (2001). Delinquent peers revisited: Does network structure matter? American Journal of Sociology, 106, 1013–1057.

Hindelang, M. J., Gottfredson, M. R., & Garofalo, J. (1978). Victims of personal crime. Cambridge, MA: Ballinger.

Hindelang, M. J., Hirschi, T., & Weis, J. G. (1981). Measuring delinquency. Beverly Hills, CA: Sage.

Hirschi, T. (1969). Cause of delinquency. Berkeley: University of California Press.

Hirschi, T. (1991). Family structure and crime. In B. J. Christensen (Ed.), When families fail . . . the social costs (pp. 43–65). Lanham, MD: University Press of America.

Horney, J., Osgood, D. W., & Marshall, I. H. (1995). Criminal careers in the short-term: Intra-individual variability in crime and its relation to local life circumstances. American Sociological Review, 60, 655–673.

Jensen, G. F., & Brownfield, D. (1986). Gender, lifestyles, and victimization: Beyond routine activity theory. Violence and Victims, 1, 85–99.

Lauritsen, J. L., Laub, J. H., & Sampson, R. J. (1992). Conventional and delinquent activities: Implications for the prevention of victimization among adolescents. Violence and Victims, 7, 91–108.

Lauritsen, J. L., Sampson, R. J., & Laub, J. H. (1991). Addressing the link between

offending and victimization among adolescents. Criminology, 29, 265–291.

Loeber, R., & Stouthamer-Loeber, M. (1986). Family factors as correlates and predictors of juvenile conduct problems and delinquency. In M. Tonry & N. Morris (Eds.), Crime and Justice: A Review of Research (Vol. 7, pp. 29–149). Chicago: University of Chicago.

Miethe, T. D., & Meier, R. F. (1994). Crime and social context: Toward an integrated theory of offenders, victims, and situations. Albany: State University of New York Press.

Miethe, T. D., Stafford, M. C., & Long, J. S. (1987). Social differentiation in criminal victimization: A test of routine activities/lifestyles theories. American Sociological Review, 52, 184–194.

Miller, J., & Decker, S. H. (2001). Young women and gang violence: Gender, street offending and violent victimization in gangs. Justice Quarterly, 18, 115–140.

Piquero, A. R., & Hickman, M. (2003). Extending Tittle’s control balance theory to account for victimization. Criminal Justice and Behavior, 30, 282–301.

Rebellon, C. J. (2002). Reconsidering the broken homes/delinquency relationship and exploring its mediating mechanisms. Criminology, 40, 103–135.

Rutter, M., & Giller, H. (1983). Juvenile delinquency: Trends and perspectives. New York: Guilford Press.

Sampson, R. J., & Laub, J. H. (1993). Crime in the making: Pathways and turning points through life. Cambridge, MA: Harvard University Press.

Sampson, R. J., & Wooldredge, J. (1987). Linking the micro- and macro-dimensions of lifestyle-routine activity and opportunity models of predatory victimization. Journal of Quantitative Criminology, 3, 371–393.

Sanders, W. (1994). Gangbangs and drive-bys: Grounded culture and juvenile gang violence. New York: Aldine de Gruyter.

Schreck, C. J. (1999). Criminal victimization and low self-control: An extension and test of a general theory of crime. Justice Quarterly, 16, 633–654.

Schreck, C. J., Wright, R. A., & Miller, J. M. (2002). A study of individual and situational antecedents of violent victimization. Justice Quarterly, 19, 159–180.

Singer, S. I. (1981). Homogeneous victim-offender populations: A review and some research implications. Journal of Criminal Law and Criminology, 72, 779–788.

Sparks, R. F. (1982). Research on victims of crime. Washington, DC: U.S. Government Printing Office.

Warr, M. (1996). Organization and instigation in delinquent groups. Criminology, 31, 17–40.

Warr, M. (2002). Companions in crime: The social aspects of criminal conduct. Cambridge, UK: Cambridge University Press.

Warr, M., & Stafford, M. C. (1991). The influence of delinquent peers: What they think or what they do? Criminology, 29, 851–866.

Reading /// 2

Neighborhoods have long been studied for their crime proneness but have been less studied for how individual-level characteristics of people interact with neighborhood conditions to enhance or reduce risks for victimization. In his research using data from the Longitudinal Cohort in the Project on Human Development in Chicago Neighborhoods, Chris L. Gibson (2012) investigates how levels of self-control relate to neighborhood disadvantage to influence violent victimization risk. Of particular note, he finds evidence for the “social push” perspective in that low self- control mattered most in the least disadvantaged neighborhoods.

An Investigation of Neighborhood Disadvantage, Low Self-Control, and Violent Victimization Among Youth

Chris L. Gibson

Vulnerability to crime and violence is influenced by the context in which individuals reside, the activities they choose to engage in, and the individual characteristics that make them different from each other (Gibson, Morris, & Beaver, 2009; Leventhal & Brooks-Gunn, 2003; Maimon & Browning, 2010; Mayer & Jencks, 1989; Sampson, Morenoff, & Raudenbush, 2005; Sampson, Raudenbush, & Earls, 1997; Schreck, Wright, & Miller, 2002; Sharkey, 2006; Rountree, Land, & Miethe, 1994). Historically, situational and contextual theories have had the largest influence on understanding variation in personal victimization (Cohen & Felson, 1979; Hindelang, Gottfredson, & Garofalo, 1978). Ranging from lifestyle choices to the imposed structural conditions of neighborhoods, a primary focus in personal victimization research has been on how environments encountered by people can put them at risk of victimization, and more specifically expose them to situations where offenders are more proximate. On the other hand, studies show that individual traits increase individual risk of being victimized. In particular, low self-control—defined as the likelihood that individuals will consider the short- and long-term consequences of their actions in pursuit of self-gratifying behavior —is a trait that has been shown to influence victimization risk (Schreck, 1999). These antecedents of personal victimization have largely been independently explored and research has often neglected how the two may be integrated to better understand victimization risk.

Neighborhood Context and Personal Victimization It is established that crime and violence tend to be concentrated in disadvantaged urban areas. That is, compared to other neighborhoods those with elevated rates of poverty, residential turnover, and ethnic heterogeneity often experience more crime, and those living in them are at an increased vulnerability for becoming victims of violence

(Rountree et al., 1994; Sampson et al., 1997; Sampson & Wooldredge, 1987; Smith & Jarjoura, 1989).

Several scholars have empirically extended Shaw and McKay’s [1942] research by measuring intervening mechanisms that offer explanations for why residents in disorganized neighborhoods are less capable of controlling crime. Factors such as weak interpersonal networks, inadequate informal social control, subcultural and competition models that condone and encourage violence, and mistrust among neighbors all have been studied as mechanisms that are important for explaining the connection between structural disadvantage and neighborhood crime (e.g., Anderson, 1999; Kornhauser, 1978; Sampson & Groves, 1989; Smith & Jarjoura, 1989). For instance, stable residency in a neighborhood can promote friendships with other residents as well as the discovery of common interests and values.

A recent extension of social disorganization identified the relationships that may promote order and consequentially lead to less vulnerability to crime and violence, or what Sampson et al. (1997) coined as “collective efficacy.” Collective efficacy is the willingness of residents within neighborhoods to exercise informal social control in situations of disorderly conduct (e.g., kids loitering outside while truant, painting graffiti) and to intervene for the common good of their neighborhood, coupled with trust and cohesion among neighbors (Sampson, 2006; Sampson et al., 1997).

Sampson et al. (1997) were the first to show the influence of neighborhood collective efficacy on violence reduction. Using data from the PHDCN, they found that collective efficacy mediated neighborhood structural influences, such as concentrated disadvantage and residential stability, on neighborhood violence. Children living in areas of high collective efficacy are less likely to engage in violence (Berg & Loeber, 2011; Maimon & Browning, 2010; Sampson et al., 2005). In sum, neighborhoods have higher concentrations of crime, violence, and adverse outcomes for children and adolescents when residents do not trust one another, are less willing to work together for the common good of their neighborhood, and when subcultural values are dominated by unconventional values where competition for respect among individuals often results in crime and violence.

Being that violent offending and victimization are strongly correlated (Berg & Loeber, 2011; Hindelang et al., 1978; Lauritsen, Sampson, & Laub, 1991), it comes as no surprise that neighborhood structural and social factors are also related to violent victimization risk (e.g., Berg & Loeber, 2011; Lauritsen, 2001; Sampson, 1985; Sampson & Wooldredge, 1987). Research shows that—independent of demographic characteristics—neighborhood poverty, residential mobility, and family structure increase the risk of personal victimization (Lauritsen, 2001; Miethe & McDowall, 1993;

Sampson, 1983, 1985). With a desire to explain community variation in victimization, researchers have integrated the routine activity and lifestyles perspectives with social disorganization and collective efficacy (e.g., Lee, 2000; Wilcox, Land, & Hunt, 2003). Clearly, community social control and collective efficacy intersect with the routine activity concept of guardianship. Marcus Felson (1998) noticed that the most effective guardianship is informal, where the presence of others deters victimization (see also Felson, 1986). This is especially true since formal agents of social control (e.g., the police) only infrequently appear in citizens’ lives; routine guardianship must therefore result from informal control. As Lee (2000, p. 687) stated, guardianship exists within “the constellation of networks and social interactions with other individuals and groups external to the individual.”

Using data from the PHDCN, the British Crime Survey (BCS), and the Seattle phone survey, researchers have scrutinized the joint contribution of individual and neighborhood-level risk factors for victimization. For instance, Sampson and Wooldredge (1987) used data from the BCS and found that while lifestyles of adults (e.g., living alone or frequent out-of-home activity) influenced violent victimization risk, aggregated neighborhood factors independently influenced victimization risk. Specifically, contextual factors including family disruption, social cohesion, and the amount of activity on neighborhood streets all had significant associations with risk of being robbed. More recent work in a variety of contexts shows similar results (e.g., Fisher, Sloan, Cullen, & Lu, 1998; Kennedy & Forde, 1990; Miethe & McDowall, 1993; Thompson & Fisher, 1996). For instance, Miethe and McDowall (1993) used data from the Seattle phone survey and found that people living in poorer and busier neighborhoods were at risk of being victims of stranger violence. In a reanalysis of the Seattle data, Rountree et al. (1994) used a multilevel model that accounted for the fact that community residents are not randomly assigned to neighborhoods and found similar findings; that is, risk of victimization varied significantly across neighborhoods and victimization was influenced by contextual variables including density, disorder, and ethnic heterogeneity.

Indeed, neighborhood variables appeared to determine how effective individual protective behaviors were. Sampson et al. (1997) found that risk of violence among residents was reduced in higher collective efficacy neighborhoods in Chicago. Taken together, there is impressive consistency in the results that link community characteristics to personal victimization, especially when one considers the variety of contexts examined.

Studies conducted to date on the influence of neighborhood conditions on personal victimization share some noteworthy limitations. First, they have largely used adult samples, leaving less knowledge of how neighborhoods influence personal victimization among adolescents. This is an important area of research because individuals at most risk of becoming victims of violence are adolescents. Second, although relationships were found between neighborhood conditions and personal victimization independent of individual characteristics, most studies linking neighborhoods and personal victimization rarely go beyond incorporating individual-level demographic and lifestyle characteristics that distinguish between individuals (Berg & Loeber, 2011). This makes it unclear what role other theoretically derived individual-level risk factors play in (e.g., low self-control) understanding neighborhood contextual influences on victimization experiences. Finally, given the limited attention to neighborhoods and personal victimization among adolescents, this initial study focuses on the intersection of neighborhood structural characteristics—commonly used in the social disorganization literature—and individual differences between adolescents to better understand their personal victimization.

Source: Gibson, C. L. (2012). An investigation of neighborhood disadvantage, low self- control, and violent victimization among youth. Youth Violence and Juvenile Justice, 10(1), 41–63.

Self-Control, Lifestyles, and Personal Victimization Gottfredson and Hirschi’s general theory (aka, self-control theory) is one of the most tested theories of crime, violence, and deviant behavior (Pratt & Cullen, 2000), [and] Schreck (1999) was one of the first to recognize that Gottfredson and Hirschi’s self- control theory is also applicable to personal victimization. He observed that self-control theory is compatible with situational theories of victimization by theoretically describing how those with lower self-control make decisions about their lifestyles and have personal characteristics that make them highly vulnerable to crime (Schreck et al., 2002). For instance, it is likely that low self-control individuals will have more grievances against them because they are more aggressive and have aversive personal

styles. Felson (1984) found that most physical assaults took place when aggrieved parties attacked the person who had injured them, and this pattern appears often in the violence literature (e.g., Anderson, 1999; Jacobs & Wright, 2006). With respect to property victimization, the tendency to lack foresight works against efforts at self- protection as well, since self-protection entails at least some inconvenience (Schreck, 1999). Further, characteristics such as impulsivity and risk seeking can lead individuals into situations that put them in closer proximity to offenders and situations that more generally come with a higher risk of being victimized.

Employing various samples, methods, and data collection techniques, recent research by Schreck and others (e.g., Higgins et al., 2010; Schreck et al., 2002, 2006; Stewart, Elifson, & Sterk, 2004) has consistently confirmed that those who have lower self- control are at increased risk of victimization. Also, this research has shown that self- control influences victimization independently of lifestyles and other risk factors such as peers, indicating that the presence of self-control can independently make a situation safer while low self-control can worsen personal safety and one’s possessions in otherwise harmless situations (Schreck et al., 2002; Stewart et al., 2004). In addition to research showing that low self-control has influences on trajectories of violent victimization, at least one study indicates that low self-control individuals who are victims are unlikely to change their routines that may protect them from future victimization (Schreck et al., 2006).

The Current Study It has been established that neighborhood structural conditions and individual differences in low self-control can independently influence adolescents’ vulnerability to crime. At present, the role of low self-control as a facilitator of personal victimization vis-à-vis imposed neighborhood structural factors is unknown. That is, does the risk of personal victimization depend on both or do they each have independent influences? The importance of the current study is then to understand how neighborhood structure and individual differences in low self-control jointly shape violent victimization risk.

Although not specifically focusing on victimization, several perspectives have been articulated to predict how context and traits will interact to shape behavior. First, intended for understanding biosocial phenomena, Raine (1988, 2002) proposed a “social push” perspective to discuss the interplay between social environments and traits for understanding how biologically related variables influence antisocial outcomes. His notion of “social push” is consistent with the idea that environments can be “weak” or “strong.” Particularly important for the current study is the idea of a “strong” environment, which is a context that purportedly minimizes the influence of individual

differences because of the strong pressures to behave in certain ways (Lynam et al., 2000; Mischel, 1977). Disadvantaged neighborhoods are one context identified by researchers as a strong environment; they are contexts where normative cultures are more likely to strongly influence the behavior of people, especially antisocial and violent behaviors. Anderson’s (1999) qualitative work on the code of the streets in urban, impoverished neighborhoods exemplifies the notion of strong social environments in that even kids who come from what he coined as decent families must adopt the street code to gain respect among their neighborhood peers to avoid future vulnerabilities to criminal violence. In such strong neighborhood contexts, it has been argued that individual differences among adolescents may not matter much at all because of the social pressures that are present. On the other hand, weak environments are those where the influence of individual differences in traits will have more pronounced impacts on behavior because of the indefinite nature of the social contexts, which may lack pressures to engage in antisocial and criminal activities. In these environments, adolescents’ behaviors are less dictated by environmental social pressures and more likely to be influenced by individual differences. From a neighborhood perspective, and for the purpose of the current study, such a context is the least disadvantaged neighborhoods that, in comparison to others, have less social pressures for adolescents to engage in crime, consequently leading to a reduced threat of victimization regardless of individual differences between adolescents.

Second, Gottfredson and Hirschi would likely argue that neighborhood factors and influences are to a substantial degree determined by self-selection into neighborhoods. For instance, those with low self-control would be unable to hold work that paid well and therefore would, absent other sources of support, drift toward disadvantaged neighborhoods, thus increasing the potential number of prospective offenders in the area. Those with low self-control might have friends, but their ties with others are likely of low quality (Gottfredson & Hirschi, 1990). To them, neighborhood disorganization and vulnerability to crime in the area would be largely a by-product of an excess of residents with low self-control. It might follow that any connection between neighborhood structure and victimization should be spurious due to an excess of residents with low self-control living in structurally disadvantaged neighborhoods.

Third, neighborhood contextual variables may moderate the relationship between low self-control and victimization. As discussed earlier, many studies have linked neighborhood structural conditions to victimization risk. Disadvantaged neighborhoods may increase vulnerability to crime among individuals due to a lack of collective efficacy and/or informal social control that may increase the concentration of crime in a neighborhood and, consequently, their proximity to offenders (Sampson et al., 1997). Further, subcultural norms in such neighborhoods may condone violence as an

acceptable means for handling grievances and/or confrontations with others (Anderson, 1999). These conditions can lead individuals with low self-control to be especially vulnerable to crime because they are likely to find themselves in situations with lower social control and socializing with delinquent peers. Further, such individuals have temperaments that make them more susceptible to grievances and this type of personal style combined with living in neighborhoods where violence is often used as a means for dealing with confrontation and gaining respect can make the risk of violence against them amplified. On the other hand, individuals possessing more self-control may have a lowered risk of becoming a victim in disadvantaged neighborhoods because they may have a heightened sensitivity to neighborhood dangers and try to make sure they avoid situations and people that would otherwise put them at elevated risk of personal victimization. They will likely consider the potential consequences of their actions; in this case, it would be the real threat of becoming a victim of violence. Further, in view of the fact that those with

high self-control will act in ways that are less likely to provoke others, as well as be cognizant of the necessity of protective behaviors, then it would seem to follow that high self-control youth living in high-risk environments may have reduced risk of becoming victims because they are more sensitive to the environmental risks.

Hypotheses Several competing hypotheses can be discerned from the literature on victimization and the different perspectives discussed above.

Hypothesis 1: Regardless of neighborhood context, lifestyles, and prior involvement in violence, those with low self-control will have an increased risk of being victims of violence. Hypothesis 2: Regardless of self-control and other individual characteristics that increase the risk of violent victimization, individuals residing in structurally disadvantaged neighborhoods will have a greater risk of being victims of violence. Hypothesis 3: As argued in the “social push” perspective, low self-control will not predict violent victimization among those residing in disadvantaged neighborhoods, but it will for those residing in least disadvantaged neighborhoods that lack social pressures for crime. Hypothesis 4: Individuals with lower self-control will be at a heightened risk of violent victimization in disadvantaged neighborhoods, whereas those possessing more self-control are less likely to become victims in disadvantaged neighborhoods.

Method Data from the 9-, 12-, and 15-year-old cohorts from the PHDCN-LCS—an interdisciplinary study on how the contexts in which children and adolescents reside contribute to their behavior and psychological development—were analyzed to test hypotheses in the current study. Violent victimization is predicted using individual-level predictors from Waves 1 and 2 of the LCS. Additionally, 1990 U.S. Census data were used to measure neighborhood-level structure of the 80 neighborhood clusters (NCs), that is, aggregations of 1–3 census tracts. Through combining data from the LCS and the 1990 U.S. Census, the PHDCN is appropriate for assessing the influence and intersection between neighborhood and individual-level factors on violent victimization experiences among children and adolescents. A wealth of neighborhood structural, organizational, and social process measures exist that have been validated over time, making the PHDCN a unique study on neighborhoods (Raudenbush & Sampson, 1999;

Sampson et al., 1997).

Data and Sampling The PHDCN was implemented in 1995 to collect quality data on the neighborhood contexts in which children and adolescents develop. The sampling design of the PHDCN contains two components: selection of NCs and selection of children and their primary caregivers for the longitudinal study. A sample of 80 NCs was selected from within the strata of 343 NCs for the PHDCN-LCS. With few exceptions, the 80 sampled NCs were representative of the racial, ethnic, and socioeconomically diverse Chicago neighborhoods (Sampson et al., 1997; Sampson, Morenoff, & Earls, 1999).

Within the 80 NCs, approximately 6,500 children, adolescents, and their primary caregivers were randomly selected. Starting with a random selection of block groups within NCs, a list of dwellings was compiled and household members were enumerated. In total, approximately 40,000 dwellings were screened. Study recruitment consisted of infants, children, adolescents (including 18-year-olds) and their primary caregivers. For selection, subjects had to be within approximately 6 months of the following age categories: 0, 3, 6, 9, 12, 15, and 18 years. Extensive interviews were conducted with these children and their primary caregivers over three waves of data collection spaced approximately 2.5 years apart. The longitudinal study began in 1995; the second wave of data collection occurred between 1997 and 1999; and the third wave of data was collected between 2000 and 2001. This effort resulted in seven cohorts that span a period of development between infancy and early to mid adulthood.1

Analysis Sample The analysis sample consists of 1,889 children and adolescents from the 9- (n = 669), 12- (n = 674), 15-year-old cohorts (n = 546), and their primary caregivers from Waves 1 and 2.2 Table 1 reports several descriptive statistics for the analysis sample. The sample is 15% White, 34% Black, 47% Hispanic, and 3.5% Other. Males make up 50% of the sample and the average age at Wave 1 is approximately 12 years old. Approximately 23% of the children were from single-parent households. These children and their families resided within 79 of the NCs.

Table 1 Descriptive Statistics (n = 1,889)

Analysis Sample Victim Nonvictim

Dependent Variable

Violent victimization .294 (.456) — —

Demographic

Sex .502 (.500) .574 (.021) .473 (.014)

Age 11.955 (2.422) 12.555(.102) 11.705 (.065)*

Single parent .229 (.420) .295 (.019) .202 (.011)*

Race/Ethnicity

White .150 (.357) .133 (.014) .157 (.010)

Black .343 (.475) .403 (.021) .318 (.466)*

Hispanic .472 (.500) .433 (.021) .488 (.014)*

Other .035 (.185) .031 (.007) .038 (.005)

Socioeconomic status −.102 (1.422) −.067(.062) −.116 (.039)

Family and parenting

Family attachment .016 (.964) −.100 (.045) .065 (.025)*

Warmth .027 (.951) −.046(.041) .056 (.026)*

Supervision .026 (.960) −.023(.043) .046 (.026)

Lack of hostility −.001 (.990) −.021(.043) .007 (.027)

Antisocial propensity

Low self-control −.005 (.978) .217 (.043) −.097(.026)*

Behavioral and lifestyle

Delinquent peers −.022 (.969) .250 (.046) −.136(.024)*

Violent offending .330 (.470) .478 (.021) .268 (.012)*

Unstructured socializing −.001 (2.661) .743 (.114)

−.312 (.071)*

Note: *p < .05.

Measures

Dependent Variable

Violent victimization

A violent victimization measure was created using responses to seven questions from the Exposure to Violence (ETV) interview administered to subjects during Wave 2 interviews. Subjects were asked whether in the past 12 months they have been hit, slapped or beaten up, attacked with a weapon, shot, shot at, sexually assaulted, and if someone has threatened to seriously hurt them. Given that approximately 70% (n = 1,333) of the analysis sample reported not being a victim of any of the seven types of violent victimizations and that very few experienced two or more of these,3 a decision was made to use a prevalence measure indicating whether youths reported being violently victimized in the past 12 months. To this end, 29.4% (n = 556) of the children and adolescents in the analysis sample reported being a victim of one or more of the seven types of violent victimizations during the 12 months prior to Wave 2 interviews.

Individual-Level Independent Variable

Low self-control

Following past research (Gibson et al., 2009; Gibson, Sullivan, Jones, & Piquero, 2010), a 17-item measure of behavioral indicators of low self-control was constructed using the Emotionality, Activity, Sociability, and Impulsivity Temperament Survey (EASI)- temperament instrument. Administered during Wave 1 interviews, primary caregivers were asked to report on their children’s inhibitory control, decision making, risk and sensation seeking, and diligence or persistence in completing tasks (see also Buss & Plomin, 1975), which are consistent with Gottfredson and Hirschi’s (1990) definition of self-control as well as past empirical research (see Grasmick, Tittle, Bursik, & Arneklev, 1993).

Reflecting what Gottfredson and Hirschi call impulsivity, inhibitory control reflects the inability to delay gratification and control frustrations, such as trouble controlling impulses and having problems with resisting temptations. Items measuring decision time measure the (in)ability to delay decision making until alternatives can be seriously considered including saying the first thing that comes into one’s head, trouble making up my mind, and acting on the spur of the moment. Taken together, inhibitory control and decision time reflect what Gottfredson and Hirschi (1990) refer to as impulsivity, a main component of low self-control. Sensation seeking, or what Gottfredson and Hirschi (1990) refer to as risk seeking, reflects a preference for novel stimuli, such as seeking new and exciting experiences, doing “crazy” things, and willingness to try anything. Persistence, or diligence as Gottfredson and Hirschi (1990) call it, is the likelihood that a child will complete or follow through with tasks they start. Children with low self- control are often the first to initiate a task, but also the first to abort them, especially

when it does not involve immediate gratification.

For each subject, item responses were summed and standardized α = .68. More positive scores indicate lower self-control, whereas more negative scores indicate higher self- control. Other studies using the PHDCN also find this measure to be reliable (also see Gibson et al., 2009, 2010).

Individual-Level Control Variables

Family attachment and support

Family attachment and support has been repeatedly shown to be a predictor of violence and victimization and it has also been related to children and adolescents’ self-control development (Gottfredson & Hirschi, 1990). Following Maimon and Browning (2010), a family attachment and support measure was constructed using 6 items from the Provision of Social Relations instrument administered during the Wave 1 interviews (α = .60). This measure gauges the level of perceived emotional and social support provided to the respondents from their families. On a scale ranging from not true (1) to very true (3), respondents were asked to respond to the following statements regarding their family: “I know my family will always be there for me,” “my family tells me they think I am valuable,” “my family has confidence in me,” “my family helps me find solutions to problems,” “I know my family will always stand by me,” and “sometime I am not sure I can rely on family” (reverse coded). The measure was standardized with higher scores indicating more family attachment and support.

Parenting measures

Parenting has been implicated in the development of self-control, as a social control mechanism for decreasing the likelihood of misbehavior, and has been associated with adolescent victimization. Parenting measures were taken from the Home Observation for Measurement of the Environment (HOME) Inventory (Caldwell & Bradley, 1984; Leventhal, Selner-O’Hagan, Brooks-Gunns, Bingenheimer, & Earls, 2004). This inventory is generally designed to measure the quantity and quality of stimulation in a child’s home environment from primary caregivers.

Parental sensitivity and responsiveness is one domain of parenting measured by the HOME, which consists of parental warmth and parental lack of hostility (and punitiveness). Derived from observations of primary caregivers and children during the in-home interviews at Wave 1, the parental warmth measure consists of 9 items. Likewise, parental lack of hostility consists of several observations that interviewers

recorded during the in-home interviews. Reflecting a primary caregiver’s knowledge of their child’s whereabouts, familiarity with child’s friends, and rules surrounding curfews and activities with friends, the parental supervision measure is a 13-item scale of how primary caregivers directly and indirectly monitor and supervise their children. Each measure was standardized (Gibson et al., 2010 reported these measures to have reliabilities ranging between .50 and .72; also see Leventhal et al., 2004). More positive scores indicate more supervision, more warmth, and less hostility.

Lifestyle, situational, and behavioral variables

Similar to prior research (Maimon & Browning, 2010; Osgood & Haynie, 2005) peer delinquency is measured using 11 items from the Deviance of Peers instrument administered at Wave 1 to children and adolescents to obtain information regarding the quantity of their peers engaging in a variety of delinquent behaviors. For example, respondents were asked “during the past year, how many of the people you spend time with” have “stolen something worth more than $5 but less than $500,” “gone into or tried to go into a building to steal something,” “hit someone with the idea of hurting them,” and “attacked someone with a weapon with the idea of seriously hurting them.” Responses to questions ranged from 1 (none of them) to 3 (all of them). The measure was standardized (α = .81) and higher scores indicate having a larger quantity of delinquent peers.

Prior violent offending behavior is included because [of] the robust relationship between offenders and victims and the evidence that shows low self-control is related to both violent offending and victimization. During the first wave of data collection, children and adolescents were asked to self-report their involvement in a variety of criminal and serious delinquent acts. Seven questions that asked about violence are used to create a dichotomous measure indicating whether a respondent ever engaged in at least one violent offense. For instance, violent offending items include items such as “have you ever:” “attacked someone with a weapon,” “used a weapon or force to get money or things from people,” “been involved in a gang fight in which someone was hurt [or] threatened with harm,” and “thrown objects, such as a rock or bottles, at people.” Those reporting at least one violent offense were coded as 1 and those reporting none of the violent offenses are coded 0. Of the analysis sample, 33% (n = 623) reported engaging in at least one of the violent offenses.

Unstructured socializing with peers is also included as a control because past research shows that it is related to delinquent behavior (Osgood, Wilson, O’Mally, Bachman, & Johnston, 1996), and such lifestyles more generally have been linked to victimization risk (Schreck et al., 2002). Similar to Osgood et al. (1996) and Maimon and Browning

(2010), a measure of unstructured socializing with peers was constructed using a 4-item scale of responses to questions that asked children and adolescents during Wave 2: “How often do you ride around in a car/motorcycle for fun,” “how often do you hang out with friends,” “how often do you go to parties,” and “how many days a week do you go out after school/at night.” Responses to question[s] range from 1 (never) to 5 (almost every day). The measure was standardized with higher scores indicating more time spent in unstructured activities with peers (α = .58).

Demographic characteristics

SES is measured using the principal component of three variables including household income, maximum education level of primary caregiver and partner, and the socioeconomic index (SEI) for primary caregivers and partner’s jobs.4 Gender of the subject is coded 0 (female) or 1 (male). Age of subject at Wave 1 is measured as a continuous variable.

Neighborhood-Level Variable

Concentrated disadvantage

Neighborhood disadvantage—or neighborhood poverty and segregation—is measured using a composite measure of several structural characteristics within each NC of the PHDCN-LCS. Originally created by Sampson et al. (1997) and shown to be a reliable construct, this measure is composed of 6 items taken from the 1990 U.S. Decennial Census and includes the following variables: percentage of neighborhood residents below the poverty line, percentage on public assistance, percentage of female-headed families, percentage unemployed, density of children by percentage younger than 18, and percentage of Blacks. Higher scores on this measure reflect more concentrated disadvantage.

Analytic Strategy

The analysis will proceed in several steps for testing hypotheses regarding neighborhood structure, low self-control, and violent victimization. First, bivariate comparisons are made for victims of violence and nonvictims to determine the demographic and theoretically derived characteristics that distinguish the two groups. Second, a series of multivariate logistic regression models are estimated to determine the direct influence of neighborhood structure and low self-control on violent victimization. Third, additional multivariate logistic regression models are estimated to determine whether the influence of low self-control on violent victimization varies as a function of the level of neighborhood disadvantage in which adolescents reside. Finally, conditional predicted probabilities are estimated to allow a better visualization of the association between low self-control and violent victimization in the least and most disadvantaged neighborhoods.

Results

Differences Between Victims and Nonvictims of Violence

Table 1 shows comparisons between subjects who self-reported at least one violent victimization to those not reporting a violent victimization 12 months prior to Wave 2 on demographic characteristics, family/parenting variables, behavioral and lifestyle variables. First, a significantly larger percentage of males reported being victimized compared to females; of those who were victims, 57% were males and of those who were nonvictims, 47% were males. On average, victims of violence were significantly older than nonvictims; victims had a mean age of 12.55 at Wave 1, whereas nonvictims’ average age was 11.705. Respondents who were victims of violence were also more likely to be from single-parent households where the primary caregiver is typically the

biological mother, 29.5% of victims compared to 20.2% of nonvictims. Black and Hispanic youth were significantly more likely to be victims of violence, but Whites were no more likely to be a victim than others. Second, child and adolescent victims of violence, on average, had significantly less family support and warmth from their primary caregiver compared to nonvictims, yet no significant differences were observed for average levels of supervision and hostility from their primary caregivers. Third, several significant differences were found between victims and nonvictims for lifestyle, situational, and behavioral characteristics. Victims of violence were significantly more likely to report engaging in violent offending at Wave 1; 33% reported engaging in at least one of the violent offenses in their lifetime. Victims of violence also reported having significantly more delinquent peers and spending more time in unstructured activities with peers. Fourth, victims of violence, on average, have significantly lower self-control compared to nonvictims; victims, on average, have .217 standard deviation units greater than the mean, whereas nonvictims, on average, have .097 standard deviation units below the mean. Finally, victims were not significantly more likely to live in disadvantaged neighborhoods compared to nonvictims.

Multivariate Analysis Predicting Violent Victimization

To first address the hypothesis that violent victimization is more likely among those living in more structurally disadvantaged neighborhoods, a hierarchical generalized linear model (HGLM) was estimated. Although not reported, results showed that while a positive relationship was observed between concentrated disadvantage and violent victimization, this relationship was not statistically significant. This analysis confirms the bivariate comparison that neighborhood structural disadvantage did not

have a direct, statistically significant influence on violent victimization at Wave 2 among adolescents.

Table 2 shows [a] logistic regression model predicting Wave 2 violent victimization, [with robust standard errors to account for the fact that youth are nested in neighborhood].

Table 2 shows a full model that includes demographic characteristics, family and parenting variables, low self-control, and behavioral and lifestyle characteristics. Delinquent peers, self-reported violent offending, and unstructured time spent with peers each have independent significant associations with violent victimization. Youth who have a larger portion of their friends that engage in delinquent activities are more likely to report being violently victimized (OR = 1.196; p < .05). Those reporting that they have committed at least one violent offense in their lifetime at Wave 1 were also more likely to have reported a violent victimization compared to their nonviolent counterparts. Unstructured time spent with peers also has a statistically significant association with violent victimization (OR = 1.117; p < .05); youth spending more time in unstructured activities with peers are more likely to report being victims of violence. And, after statistically controlling for behavioral and lifestyle measures commonly known to have moderate-to-strong influences on violent victimization, youth with lower self-control were still significantly more likely to report being a victim of violence compared to those who possess more self- control (OR = 1.287; p < .05). While sex, age, and single-parent family structure retained significant associations, the difference in violent victimization between Blacks and Whites is accounted for by behavioral and lifestyle characteristics. Blacks are no longer more likely to be victimized than Whites after statistically controlling for peers, violent offending, and unstructured time spent in activities.

Table 2 Multivariate Logistic Regression Predicting Violent Victimization (n =

1,889)

OR (RSE)

Demographics

Sex 1.325* (.154)

Age 1.090* (.026)

Single parent 1.466* (.209)

Race (Ref = White)

Black 1.313 (.205)

Hispanic 1.402 (.242)

Other 1.097 (.336)

Socioeconomic status 1.047 (.050)

Family and parenting

Family attachment .964 (.064)

Warmth .988 (.059)

Supervision 1.013 (.062)

Antisocial propensity

Low self-control 1.287* (.073)

Behavioral and lifestyle

Delinquent peers 1.196* (.063)

Violent offending 1.464* (.172)

Unstructured time 1.117* (.025)

χ2 172.51*

Log likelihood −1050.96

Pseudo R2 .08 Note: 79 neighborhood clusters. OR = odds ratio; RSE = Robust standard errors.

*p < .05.

Neighborhood Disadvantage and Violent Victimization

Table 3 explores the association between low self-control and violent victimization for youth residing in neighborhoods that are characterized by different levels of concentrated disadvantage. Neighborhoods were classified as low, medium, or high concentrated disadvantage by defining those at the lowest 25th percentile and lower as low disadvantaged; highest 25th percentile and higher as the most disadvantaged; and those in between as medium disadvantaged. The pattern of findings that emerged regarding low self-control and violent victimization across the range of disadvantaged neighborhoods is telling.

Table 3 Logistic Regression Analysis Predicting Violent Victimization by Neighborhood Concentrated Disadvantage (n = 1,889)

Low Disadvantage

(n = 501)

OR (RSE)

Medium Disadvantage

(n = 919)

OR (RSE)

High Disadvantage

(n = 469)

OR (RSE)

Demographics

Sex 1.502* (.311) 1.041 (.168) 1.756 (.408)

Age 1.034 (.052) 1.104* (.033) 1.104 (.067)

Single parent 1.300 (.370) 1.794* (.365) 1.163 (.323)

Race (Ref = White)

Black 1.783* (.514) 1.306 (.352) 3.835 (3.234)

Hispanic 1.401 (.354) 1.710 (.508) 2.980 (2.565)

Other .630 (.317) 1.615 (.786) 3.803 (4.236)

Socioeconomic status .994 (.082) 1.108 (.069) .935 (.105)

Family and parenting

Family attachment .866 (.132) .938 (.089) 1.123 (.148)

Warmth 1.071 (.117) .959 (.074) .959 (.146)

Supervision 1.027 (.116) 1.018 (.104) 1.045 (.149)

Lack of hostility 1.162 (.195) 1.052 (.080) .927 (.076)

Antisocial propensity

Low self-control 1.390* (.177) 1.335* (.106) 1.125 (.134)

Behavioral and lifestyle

Delinquent peers 1.382* (.219) 1.151* (.077) 1.180 (.122)

Violent offending 1.345* (.322) 1.468* (.259) 1.496 (.349)

Unstructured socializing 1.122* (.050) 1.138* (.037) 1.108* (.049)

χ2 601.47 139.97 133.08

Log likelihood −272.868 −495.676 −270.539

Pseudo R2 .100 .090 .084 Note: OR = odds ratio; RSE = robust standard errors.

*p < .05.

First, a logistic regression model for youth living in the lowest disadvantaged neighborhoods reveals that the association between low self- control and victimization was statistically significant and positive. Consistent with the model reported above, youth with lower self-control were more likely to report being victims of violence. Youth in the lowest disadvantaged neighborhoods with more delinquent peers; those reporting past involvement in violent offending; and those spending more time in unstructured activities with peers were significantly more likely to report being victims of violence. In addition, males and Blacks were more likely to report being victims of violence in the lowest disadvantaged neighborhoods. Unlike in the analysis in Table 2, significant differences in violent victimization between Blacks and Whites persisted after controlling for violent offending, delinquent peers, and unstructured time spent with peers.

Second, Table 3 shows results predicting violent victimization for those youth residing in medium or moderate disadvantaged neighborhoods. Results reported in this model are largely consistent with those from the low neighborhood disadvantage analysis. That is, low self-control continued to have a significant and positive association with violent victimization for youth residing in moderately disadvantaged neighborhoods. Moreover, adolescents in moderately disadvantaged neighborhoods who socialize with more delinquent peers, engage in violent offending, and spend more time in unstructured activities with peers continue to have significantly heightened risk of being victimized by violence.

Finally, Table 3 shows strikingly different results for predictors of violent victimization among those adolescents living in the most disadvantaged neighborhoods. Unlike low and moderately disadvantaged neighborhoods, self-control did not have a statistically significant association with violent victimization in the most disadvantaged neighborhoods. That is, adolescents residing in high disadvantaged neighborhoods with lower self-control were no more likely than those with more self-control to report being a victim of violence during Wave 2 interviews. In addition, delinquent peers and violent offending—factors that past research has consistently found to be correlates of victimization—were not significantly associated with violent victimization. Unstructured time spent with peers is the only lifestyle characteristic that significantly increased the likelihood of adolescents’ violent victimization in highly disadvantaged neighborhoods.5

Conditional Probabilities of Violent Victimization: The Differing Influences of Self-Control

This section reports the conditional predicted probabilities for violent victimization across levels of low self-control for youth living in the lowest and highest disadvantaged neighborhoods while holding constant all demographic, family, and lifestyle variables in previous models. Figure 1 shows a range of standardized self-control scores on the x-axis such that 0 is the average self-control score and negative and positive values represent standard deviations above and below the mean; those values above the mean indicate lower self-control and those below represent more self-control. Probabilities of violent victimization are represented on the y-axis.

Figure 1 Conditional Probabilities of Violent Victimization

Figure 1 shows low self-control’s influence on violent victimization across neighborhood contexts. First, the probability of violent victimization increases as a function of low self-control scores, with individuals having lower self-control (higher scores) also having a greater chance of being victims of violence, although in the most disadvantaged neighborhoods the association between self-control and victimization was not statistically significant. In the lowest disadvantaged neighborhoods, the probabilities of violent victimization increase exponentially from a low to high likelihood of victimization, whereas the chance of being victimized across low self-control scores is not as nearly pronounced in the high disadvantaged neighborhoods and did not reach statistical significance. Second, even at the highest levels of self-control, it appears that the difference in violent victimization for those living in disadvantaged neighborhoods is nearly double the risk compared to those who live in the lowest disadvantaged neighborhoods. For instance, youth scoring a –2.5 on the low self-control measure, that is, possessing the most self-control, have almost a 15% chance of being victimized but those in the highest disadvantaged neighborhoods who have a similar score have almost a 25% chance. In contrast, youth living in the least disadvantaged neighborhoods and who have the lowest self-control are at most risk of being violently victimized compared to those who have the lowest self-control and reside in the most

disadvantaged neighborhoods. In fact, youth scoring a 2.5 on the low self-control measure, that is, possessing the lowest amount of self-control, in the lowest disadvantaged neighborhoods have approximately a 45% chance of being a victim of violence compared to approximately 35% for those who live in the most disadvantaged neighborhoods and have a score of 2.5 on the low self-control measure.

Discussion and Conclusion The current study contributed to [the] literature by weaving together two strains of criminological research—social disorganization and self-control—to better understand youths’ vulnerability to violence. Results from a variety of different multivariate models support the idea that these two perspectives should not be considered separately for understanding victimization. Three findings are particularly noteworthy. First, this study’s findings confirm work by Schreck and others (Higgins et al., 2010; Schreck, 1999; Schreck et al., 2002) showing that adolescents with less self-control are independently at more risk of becoming victims of crime and violence. More impressive is the fact that this relationship remained in light of situational, behavioral, and lifestyle characteristics that are commonly chosen by low self-control individuals and that also influence their vulnerability to violence. Second, and running counter to research on neighborhoods and victimization among adults, adolescents residing in the most disadvantaged neighborhoods were not significantly more likely to report being victimized compared to those residing in less disadvantaged neighborhoods, although the substantive differences were in the expected direction. Finally, and most importantly, it was found that self-control’s influence on victimization is conditioned by neighborhood type. Low self-control had a statistically important influence on victimization for those who live in the least disadvantaged neighborhoods where levels of self-control did not influence victimization among those living in the most disadvantaged neighborhoods. I interpret this finding as largely consistent with the “social push” perspective (Raine, 2002).

To recap, the social push perspective implies that in disadvantaged and criminogenic environments more social pressure exists for individuals to engage in crime, violence, and other illegal activities, especially for adolescents and those emerging into adulthood. As such, individual differences between people will have less impact on their behavior because the environmental context overrides individual risk. While Raine (1988, 2002) intended for this explanation to be specific to the influence of biological-based traits on antisocial outcomes, it is highly consistent with the measures and findings in the present study for two reasons. First, although Gottfredson and Hirschi (1990) theoretically argue that self-control is largely influenced by parental disciplining practices, recent empirical

findings show that variation in self-control is partially explained by biological factors such as genetics (Beaver et al., 2008; Wright & Beaver, 2005), making self-control a trait that fits well under the social push framework. Second, given the strong overlap between offending and victimization (Lauritsen et al., 1991), and the studies that find similar characteristics that predict both, it is easier to make the leap from antisocial outcomes that Raine discusses (e.g., violence, conduct problems, etc.) to violent victimization risk.

Questions still remain as to why variation in violent victimization among adolescents from disadvantaged neighborhoods is not significantly influenced by low self-control, and why those living in the least disadvantaged neighborhoods are. The amount of crime, proximity to offenders, and social and cultural processes occurring in disadvantaged neighborhoods may make all adolescents equally vulnerable to violence, and consequently overpowering the effects of individual differences. For instance, it is plausible to speculate that victimization risk is similar for those living in neighborhoods where residents are less likely to look after youth, especially in high crime environments. Given the lack of community informal social controls and high crime rates, it is no surprise that these children are supervised more often compared to those in less disadvantaged neighborhoods, yet supervision in this study does not seem to reduce risk of victimization either in disadvantaged neighborhoods. Further, subcultural explanations are also plausible. Risk may be similar, regardless of individual differences among adolescents, because low and high self-control adolescents face similar obstacles for attaining status and economic success, which in turn may lead them to adopt unconventional ideals to enhance their status. This can contribute to the emergence of values that encourage and condone violence or an “I will get them before they get me” mentality. Such subcultural values encourage hostile and aggressive sanctions for those who are perceived to be disrespectful, but environments such as these can make all youth vulnerable to crime and more aware of the need to use violence and exhibit aggression in attempts to avoid attacks from their peers (Anderson, 1999). As Anderson states (1999, p. 33), street cultures present in such neighborhoods provide a “rationale allowing those who are inclined to aggression to precipitate encounters in an approved way.” Unfortunately, these adaptations can lead to being on the receiving end of violence too (Stewart, Schreck, & Simons, 2006). Future research should explore these reasons for why individual difference variables such as self-control do not influence victimization risk in disadvantage[d] neighborhoods.

This study does not provide a conclusive explanation as to why low self-control influences violent victimization for adolescents residing in the least and moderately disadvantaged neighborhoods. As argued by Raine (2002), this could be because the social pressures found in the most disadvantaged contexts are less likely, nonexistent, or

attenuated. It could also be that different social processes are operating in these neighborhoods that make low self-control’s influence on victimization more pronounced. Whatever the case, it will be left for future research to explore this finding.

Implications, Limitations, and Future Research

Findings from this study have several potential implications for reducing victimization risk among adolescents. Because low self-control has such far reaching influences on a spectrum of social and health-related outcomes across developmental stages (Moffitt et al., 2011), it is important to identify early childhood precursors to low self-control to prevent the accumulation of negative outcomes in children’s futures, one of which is vulnerability to violence. Although early prevention is important, promising research suggests that self-control is plastic and is susceptible to change in later stages of one’s life course (Roberts, Walton, & Viechtbauer, 2006). Thus, unique and universal prevention and intervention efforts are warranted for targeting developmentally sensitive periods to train and encourage individuals in making informed decisions and to consider the consequences of their own behavior. Nonetheless, as Moffitt et al. (2011, p. 2967) stated, “Early childhood intervention that enhances self-control is likely to bring a greater return on investment than harm reduction programs targeting adolescence alone.” According to the present study’s findings, such efforts may be particularly important for those living in the least disadvantaged neighborhood environments.

This is not to say that we should neglect opportunities to change self-control among those children and youth in more disadvantaged environments. In fact, in an analysis not reported, youth in the most disadvantaged neighborhoods had, on average, the lowest levels of self-control. Given the broader negative consequences that can stem from this trait, it is equally important to improve self-control among youth from impoverished neighborhoods, especially since families residing in such neighborhoods may not have the resources to do so. It will also be important to identify and address the environmental forces that override the influence of individual differences on crime vulnerability in disadvantaged neighborhoods. Finally, unstructured time spent with

peers was the only factor that exhibited a statistically significant relationship with victimization risk of adolescents across each category of neighborhood disadvantage. Parents may have success at reducing their children’s risk of victimization by limiting the unstructured activities they engage in with peers.

The challenging ideas proposed set a modest research agenda that will allow criminologists to further understand how the intersection between neighborhoods and individual traits may jointly and independently help minimize youths’ vulnerability to crime and more specifically reduce their risk of victimization.

Discussion Questions 1. What evidence is found for the role of lifestyles/routine activities and

victimization? 2. Why do you think that no significant relationship was found between

neighborhood structural disadvantage and violent victimization? 3. What is the “social push” perspective? What is the implication of the findings

from this perspective for policy? 4. This paper investigates multiple theoretical perspectives. Given its findings, what

theory do you think is able to best explain victimization and why?

Notes The author disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This project was supported by Grant No. 2009-IJ-CX-0041 awarded by the National Institute of Justice, Office of Justice Programs, U.S. Department of Justice.

1. At Wave 1 of data collection, there were 1,269 subjects in Cohort 0; 1,003 subjects in Cohort 3; 980 subjects in Cohort 6; 828 subjects in Cohort 9; 820 subjects in Cohort 12; 696 subjects in Cohort 15; and 632 subjects in Cohort 18.

2. Due to attrition, the original sample of 2,345 subjects in Cohorts 9, 12, and 15 was reduced to 1,889. T-test comparisons of demographic and individual characteristics for those in the attrition group to those remaining in the sample at Wave 2 revealed no statistically significant difference for age, gender, SES, and low self-control; however, statistically significant difference emerged for delinquent peers, single-parent family structure, and supervision, with those not in the sample at Wave 2 having, on average, more delinquent peers, from single-parent households, and less supervision. To account

for additional missing data, regression imputation was used to arrive at the analysis sample.

3. In the 12 months prior to Wave 2 interviews, 18.5% (n = 351) reported experiencing one type of victimizations, 6.99% (n = 132) reported experiencing two types of victimizations, 2.38% (n = 45) reported experiencing three types of victimizations, 0.95% (n = 18) experienced four types of victimizations, and 0.53% experienced five types of victimization.

4. See http://www.icpsr.umich.edu/PHDCN/imputations.html http://www.icpsr.umich.edu/PHDCN/imputations.html for a detailed explanation of the computation and imputation of this index.

5. Each of the models was assessed for multicollinearity. With variance inflation factors (VIFs) below 10, I ruled out the possibility of severe multicollinearity in each of the logistic regression models. Those variables showing the largest VIFs were the race variables in Model 3 of Table 3 when estimating the influence of low self-control on violent victimization in the most disadvantaged neighborhoods. When race was excluded from the model, it did not affect the results; low self-control still did not reach a level of statistical significance.

References Anderson, E. (1999). Code of the street: Decency, violence, and the moral life of the

inner city. New York: W. W. Norton.

Beaver, K. M., Wright, J. P., DeLisi, M., & Vaughn, M. G. (2008). Genetic influence on the stability of low self-control: Results from a longitudinal sample of twins. Journal of Criminal Justice, 36, 478–485.

Berg, M., & Loeber, R. (2011). Examining the neighborhood context of the violent offending-victimization relationship: A prospective victimization. Online first. doi:10.1007/s10940–011–9129–7

Buss, A. H., & Plomin, R. (1975). A temperament theory of personality development. New York: John Wiley.

Caldwell, B., & Bradley, R. (1984). Home observation for measurement of the environment. Little Rock: University of Arkansas.

Cohen, L., & Felson, M. (1979). Social change and crime rate trends: A routine activities approach. American Sociological Review, 44, 588–608.

Felson, M. (1986). Linking criminal choices, routine activities, informal control, and criminal outcomes. In D. B. Cornish & R. V. Clarke (Eds.), The reasoning criminal: Rational choice perspectives on offending (pp. 119–128). New York: Springer-Verlag.

Felson, M. (1998). Crime and everyday life. Thousand Oaks, CA: Pine Forge Press.

Felson, R. B. (1984). Patterns of aggressive interaction. In A. Mummendey (Ed.), Social psychology of aggression: From individual behavior to social interaction (pp. 107–126). Berlin: Springer-Verlag.

Fisher, B. S., Sloan, J. J., Cullen, F. T., & Lu, C. (1998). Crime in the ivory tower: The level and sources of student victimization. Criminology, 36, 671–710.

Gibson, C., Morris, S., & Beaver, K. (2009). Secondary exposure to violence during adolescence: Does neighborhood context matter? Justice Quarterly, 26, 30–57.

Gibson, C., Sullivan, C., Jones, S., & Piquero, A. (2010). Does it take a village? Assessing neighborhood effects on children’s self-control. Journal of Research in Crime and Delinquency, 47, 31–62.

Gottfredson, M. R., & Hirschi, T. (1990). A general theory of crime. Stanford, CA: Stanford University Press.

Grasmick, H. G., Tittle, C. R., Bursik, R. J., & Arneklev, B. J. (1993). Testing the core implications of Gottfredson and Hirschi’s general theory of crime. Journal of Research in Crime and Delinquency, 30, 5–29.

Higgins, G., Jennings, W., Tewksbury, R., & Gibson, C. (2010). Exploring the link between low self-control and violent victimization trajectories. Criminal Justice and Behavior, 36, 1070–1084.

Hindelang, M. J., Gottfredson, M. R., & Garofalo, J. (1978). Victims of personal crime: An empirical foundation for a theory of personal victimization. Cambridge, MA: Ballinger.

Jacobs, B. A., & Wright, R. (2006). Street justice: Retaliation in the criminal underworld. New York: Cambridge.

Kennedy, L. W., & Forde, D. R. (1990). Routine activity and crime: An analysis of victimization in Canada. Criminology, 28, 137–152.

Kornhauser, R. (1978). Social sources of delinquency. Chicago: University of Chicago Press.

Lauritsen, J. L. (2001). The social ecology of violent victimization: Individual and contextual effects in the NCVS. Journal of Quantitative Criminology, 17, 3–32.

Lauritsen, J. L., Sampson, R. J., & Laub, J. H. (1991). The link between offending and victimization among adolescents. Criminology, 29, 265–291.

Lee, M. R. (2000). Community cohesion and violent predatory victimization: A theoretical extension and cross-national test of opportunity theory. Social Forces, 79, 683–688.

Leventhal, T., & Brooks-Gunn, J. (2003). Children and youth in neighborhood contexts. Current Directions in Psychological Science, 12, 27–31.

Leventhal, T., Selner-O’Hagan, M., Brooks-Gunns, J., Bingenheimer, J. B., & Earls, F. (2004). The homelife interview from the project on human development in Chicago neighborhoods: Assessment of parenting and home environment for 3- to 15-year- olds. Parenting: Science and Practice, 4, 211–241.

Lynam, P., Caspi, A., Moffitt, T., Wikstrom, P., Loeber, R., & Novak, K. (2000). The interaction between impulsivity and neighborhood context on offending: The effects of impulsivity are stronger in poorer neighborhoods. Journal of Abnormal Psychology, 109, 563–574.

Maimon, D., & Browning, C. (2010). Unstructured socializing, collective efficacy, and violent behavior among urban youth. Criminology, 48, 443–474.

Mayer, S., & Jencks, C. (1989). Growing up in poor neighborhoods: How much does it matter? Science, 243, 1441–1445.

Miethe, T. D., & McDowall, D. (1993). Contextual effects in models of criminal victimization. Social Forces, 71, 741–759.

Mischel, W. (1977). The interaction of person and situation. In D. Magnusson & N. S. Endler (Eds.), Personality at the crossroads: Current issues in interactional psychology (pp. 33–352). Hillsdale, NJ: Erlbaum.

Moffitt, T. E., Arseneault, L., Belsky, P., Dickson, N., Hancox, R. J., Harrington, H., … Caspi, A. (2011). A gradient of childhood self-control predicts health, wealth, and public safety. Proceedings of the National Academy of Sciences, 108, 2693–2698.

Osgood, D. W., & Haynie, D. L. (2005). Reconsidering peers and delinquency: How do peers matter? Social Forces, 84, 1109–1130.

Osgood, D. W., Wilson, J., O’Mally, P., Bachman, J., & Johnston, L. (1996). Routine activities and individual deviant behavior. American Sociological Review, 61, 635–655.

Pratt, T. C., & Cullen, F. T. (2000). The empirical status of Gottfredson and Hirschi’s general theory of crime: A meta-analysis. Criminology, 38, 931–964.

Raine, A. (1988). Antisocial behavior and psychophysiology. In H. L. Wagner (Ed.), Social psychophysiology and emotion: Theory and clinical application (pp. 231–250). New York: John Wiley.

Raine, A. (2002). Biosocial studies of antisocial and violent behavior in children and adults. Journal of Abnormal Child Psychology, 30, 311–326.

Raudenbush, S. W., & Sampson, R. J. (1999). “Ecometrics”: Toward a science of assessing ecological settings, with application to the systematic social observation of neighborhoods. Sociological Methodology, 29, 1–41.

Roberts, B. W., Walton, K. E., & Viechtbauer, W. (2006). Patterns of mean-level change in personality traits across the life course: A meta-analysis of longitudinal studies. Psychological Bulletin, 132, 1–25.

Rountree, P. W., Land, K. C., & Miethe, T. D. (1994). Macro-micro integration in the study of victimization: A hierarchical logistic model analysis across Seattle neighborhoods. Criminology, 32, 387–414.

Sampson, R. J. (1983).

Structural density and criminal victimization. Criminology, 21, 276–293.

Sampson, R. J. (1985). Neighborhood and crime: The structural determinants of personal victimization. Journal of Research in Crime and Delinquency, 22, 7–40.

Sampson, R. J. (2006). How does community context matter? Social mechanisms and the explanation of crime rates. In P. H. Wikstrom & R. Sampson (Eds.), The explanation of crime: Context, mechanisms, and development (pp. 31–60). Cambridge, MA: Cambridge University Press.

Sampson, R. J., & Groves, W. B. (1989). Community structure and crime: Testing social disorganization theory. American Journal of Sociology, 94, 774–802.

Sampson, R. J., Morenoff, J., & Earls, F. (1999). Beyond social capital: Spatial dynamics of collective efficacy for children. American Sociological Review, 64, 633–660.

Sampson, R. J., Morenoff, J. D., & Raudenbush, S. (2005). Social anatomy of racial and ethnic disparities in violence. American Journal of Public Health, 95, 224–232.

Sampson, R. J., Raudenbush, S., & Earls, F. (1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277, 918–924.

Sampson, R. J., & Wooldredge, J. (1987). Linking the micro and macro-level dimensions of lifestyle-routine activity and opportunity models of predatory victimization. Journal of Quantitative Criminology, 3, 371–393.

Schreck, C. J. (1999). Criminal victimization and low self-control: An extension and test of a general theory of crime. Justice Quarterly, 16, 633–654.

Schreck, C. J., Stewart, E. A., & Fisher, B. S. (2006). Self-control, victimization, and their influence on risky activities and delinquent friends: A longitudinal analysis using panel data. Journal of Quantitative Criminology, 22, 319–340.

Schreck, C. J., Wright, R. A., & Miller, J. M. (2002). A study of individual and situational antecedents of violent victimization. Justice Quarterly, 19, 159–180.

Sharkey, P. (2006). Navigating dangerous streets: The sources and consequences of street efficacy. American Sociological Review, 31, 826–846.

Shaw, C., & McKay, H. (1942). Juvenile delinquency in urban areas. Chicago: University of Chicago Press.

Stewart, E. A., Elifson, K., & Sterk, C. (2004). Integrating the general theory of crime into an explanation of violent victimization among female offenders. Justice Quarterly, 21, 159–182.

Stewart, E. A., Schreck, C. J., & Simons, R. L. (2006). I ain’t gonna let no one disrespect me: Does the code of the streets increase or reduce violent victimization? Journal of Research in Crime and Delinquency, 43, 427–458.

Thompson, C. Y., & Fisher, B. S. (1996). Predicting household victimization utilizing a multi-level routine activity approach. Journal of Crime and Justice, 19, 49–66.

Wilcox, P., Land, K. C., & Hunt, S. (2003). Criminal circumstance: A dynamic, multicontextual criminal opportunity theory. New York: Aldyne de Gruyter.

Wright, J., & Beaver, K. (2005). Do parents matter in creating self-control in their children? A genetically informed test of Gottfredson and Hirschi’s theory of low self- control. Criminology, 43, 1169–1202.