- Welcome to the National Center for Victims of Crime.
- What Are the Challenges in Measuring Defensive Gun Use?.
- International Journal of Cyber Criminology-bosslerholtjan.
Women are more likely to experience repeat sexual victimization than repeat violence incidents. Repeat victimiza- tion tends to happen in the same month of the initial victimization, and the most likely next type of victimization is by far the same type of victimiza- tion.
Risk and Protective Factors for Perpetration
Comparing incident-level characteristics of repeat incidents to single incidents, there are few differences, with the exception that, in a larger pro- portion of single incidents, women took self-protective action. Implications for prevention and educational programs are discussed. Davis, Combs-Lane, and Jackson suggested as well that to understand the prevention of victim- ization, researchers should broaden their investigations to assess multiple incidents of sexual and physical assault that may occur within developmental periods.
First, to address the existing gap in the extant literature, we assess the extent to which college women have experienced different types of repeat violent and sexual victimization during an academic year. Second, we provide descriptive information on the time course for repeat sexual and violent victimization incidents.
Studies of repeat property vic- timization have revealed that following an initial incident, subsequent vic- timization tends to recur quickly; there appears to be a delimited period of heightened risk for repeat victimization, which then decreases and eventu- ally levels off Farrell, Furthermore, researchers have explored whether victims are prone to experience victimizations of the same type; however, these analyses have not been conducted on sexual victimization incidents for college women.
We examine a crime-switch matrix to depict the sequential pattern of sexual victimization incidents. Third, we explore the preincident, situational, and postincident charac- teristics of repeat incidents and compare them to the characteristics of sin- gle incidents. Prior research has suggested factors of potential relevance. Studies have also identified high-risk situations for sexual assault such as being in isolated locations e.
The use of self- protective action is another situational factor that merits consideration; researchers have consistently reported that self-protective action can effec- tively thwart an attack see Ullman, We also address this issue. In the first stage, institu- tions were stratified into 12 strata based on their location and total student enrollment; schools were randomly selected. In the second stage, students were chosen using a probability proportionate to the size of the female enrollment.
The surveys were administered during spring Female interview- ers, who were trained in general principles of survey design and skills used to obtain sensitive information from respondents, administered the surveys using a computer-assisted telephone interviewing system. The response rate was Measurement Process Each study used a two-stage measurement process. The number of different incidents is important because for each different inci- dent, the interviewer completed a separate incident report. Starting with the most recent incident, respondents were asked detailed questions about the specific incident.
Violence includes a robbery, b simple assault, and c aggravated assault. Measures Repeat victimization. Repeat victimization is defined as having experi- enced more than one of the same type of victimization e. At the individual level, a woman who experienced two simple assaults within the time frame of since school began in the fall and when she was interviewed in the spring would be classified as a repeat simple assault victim.
At the incident level, each type of incident was coded as whether it was a single specific type of victimization incident or whether it was an incident within a repeat victimization episode. A single incident is one in which the victim experienced only one incident of that specific type of victimization e.
Incident-level characteristics. The first preincident measure, the victim— offender relationship, includes a current or former intimate partner, b someone else known e. The situational characteristics included location of the incident and use of self-protective action by the victim during the incident. First, location measured whether the incident occurred in living quarters or did not occur in living quarters e. Second, each respondent was asked if she had done anything with the idea of protecting herself or stopping the behavior while the incident was going on.
The response categories were not mutually exclusive, so the respondent could have used more than one type of self-protective action e. Self-protective action and the type of protective action used are measured as dichotomies. Two postincident characteristics were whether a the respondent reported the incident to any campus official, including campus police, and b the respondent had told someone else about the incident e.
Sample Characteristics The characteristics of the college women in the two samples were very similar. The mean age was Almost Counts and percentages describe the distribution of the repeat victims and incidents. A chi-square test of independence was used to determine if there was a relationship between the type of victimization in the preceding inci- dent and the type of victimization in the following incident.
Two-sample tests of proportion were performed to determine a if the proportion of women who experienced different types of repeat victimization were sig- nificantly different and b if single-incident characteristics significantly differed from characteristics of repeat incidents. Results The Extent of Repeat Victimization Table 1 shows the proportion of college women who were victimized, how many times they were victimized, and the proportion of incidents that happened to these women.
The results show that a small proportion of women experienced a large percentage of all types of violent and sexual incidents during the academic year. More than 7. Noteworthy is that the most sexually victimized, those 3. Table 2 reports the rates of repeat victimization broken down by types of violent and sexual victimization. As shown, a much larger proportion of women experienced more than one sexual incident compared to those who experienced more than one violent incident. Close to half of the women, A significantly smaller yet still substantial proportion of women, Repeat violent victimization.
None of the robbery or aggravated assault victims were repeat victims. Repeat sexual victimization. Three noteworthy results are evident in Table 2 regarding repeat sexual victimization. First, repeat sexual victimization was common. From The percentage of repeat incidents was also striking. Even more of the threats, Second, within each type of sexual victimization, a disproportionately small percentage of the victims experienced a large proportion of the inci- dents.
A quarter of the repeat threat victims experienced Third, women were more likely to repeatedly experience any type of sexual victimization compared to any type of violence. A significantly larger percentage of women experienced repeated rape The Nature of Repeat Victimization The time course of repeat victimization.
Figure 1 presents the time course of repeated simple assault and each type of sexual victimization.
For example, if the most recent rape occurred in January and the rape before this one happened in November, then the number of months between the two incidents was 2 months. The number of months between paired repeat incidents that happened in the same month was 0. As can be seen in Figure 1, there is an elevated risk of repeat violence or any type of repeat sexual victimization in a short time. In particular, this elevated risk is greatest within the same month. The risk of unwanted sexual contact without force happening a second time was highest a month after the first incident.
As can also be seen in Figure 1, the risk of repeat sexual and violent vic- timization, with the exception of rape,4 steadily declined over the passage of time when looking at the proportion of repeat incidents having had occurred 1, 2, 3, 4, 5, or 6 months apart. This pattern suggests that the risk of a second violent or sexual victimization decreased after the passage of 1 month. Personal experience with violent crime in general is a strong determinant for change routine daily activities rather than conducting the costly precautionary measures.
Unlike the case if staying in the region with low crime levels. Comparison of the effects propcrate and violcrate in columns 2 and 3 of table 4 with columns 5 and 6 of table 5, also showed that the high level of the property crime makes household prefer costly precautionary measures to prevent crime. Conclution The study results indicated that level of both type fo crime whether property and violent is a consistent predictor of precautionary behavior, frequency and intensity, suggesting that future research should continue examining the victimization level-precautionary behavior nexus.
In general we found that our two dependent variables have similar determinants. Nevertheless we also found some significant disparities, indicating that there is a conceptual difference between these two measures of crime cost. In concrete we found that men are more likely to incur a monetary cost than making changes to their routine daily activities to avoid crime.
Also they are less likely to adopt any precautionary behavior than women. Personal experience with violent crime is generally a stronger determinant for changing routine activities than for incurring a monetary cost. The opposite holds for own experience with property crime.
In particular, property crime has the highest impact of all objective explanatory variables considered. Also a household living in a state with high levels of property crime has higher probability of incurring a costly measure to avoid crime; than a household that lives in a state that has high level of violent crime. We found that social status does matter in the way people is affected by crime. In general education and income have a positive effect on both the probability of incurring a cost and changing own behavior. Although rich people have more to loose from crime than the poor economically speaking our results show that members of rich families are less likely to change their routine daily activities because of crime or perception of crime.
The effect for people with high education is similar. Note that these effects do not hold for incurring a monetary cost of crime. Onether interesting finding, it was reporting the incidence of criminality by individual members of the household to the police, have an impact on the precautionary behavior pattern. However, the relation between these two variables was also mediated by both criminality level, whether property or violent, in the area where the household resides. Finally, the study of precautionary behavior is relevant for policy makers.
- Japan in Crisis: What Will It Take for Japan to Rise Again??
- Time of the Fourth Horseman?
- The Challenges of Defining and Measuring Defensive Gun Use | RAND.
- Metal Cutting.
- Pro Apache Tomcat 6?
- Nature and distribution of known offenses.
- Testing and Performance of Geosynthetics in Subsurface Drainage (ASTM Special Technical Publication, 1390)?
Crime is "expensive" for society. It affects both from an economic point of view and in people's everyday life, as shown in this study by changing individuals' daily activities and incurring costly measures. Despite this, policy implications are not straightforward. Although adopting any type of precautionary behavior implies a cost for the individuals; according to the literature it is desirable that these measures are taken as they serve to decrease crime.
Thus it is possible that policies aiming to increase precautionary behavior would be unpopular as they might increase discomfort in the society, and thus undesirable for any government. We therefore would need a deeper analysis on the costs and benefits that such policies would imply for the society to make a better judgment. References Ayres, Ian and Levitt, Steven Quarterly Journal of Economics, Baker, M.
Law and Society Review, 17I, Becker, Gary Crime and Punishment: An Economic Approach. Journal of Political Economy, Borooah, Vani K. Crime and Fear : Evidence from Australia. British Journal of Criminology, Bushway, Shawn and Reuter, Peter Cohen, Mark A. Cook, Philip The Demand and Supply of Criminal Opportunities.
Gun Violence: The Real Costs. NY: Oxford University. Christens, B. Predicting violent crime using urban and suburban densities. Behavior and Social Issues, Dills, Angela K. Dugan, Laura. Criminology, Crime and Victimization: an Economic Perspective. Economfa Ferraro, K. Fear of crime: Interpreting victimization risk. Social Forces, 75, The measurement of fear of crime. Sociological Inquiry, 57, Are older people most afraid of crime? Fisher, C. Toward a Subcultural Theory of Urbanism. American Journal of Sociology, Gaviria, Alejandro and Pages-Serra, Carmen, Patterns of Crime Victimization in Latin America.
Gaviria, Alejandro and Velez, Carlos Eduardo Who Bears the Bur- den of Crime in Colombia. Paper No. Gomme, Ian M. Fear of crime among Canadians: A multivariate analysis. Lagrange, Randy L. Journal of Research in Crime and Delinquency, Lane, J. Social disorganization perceptions, fear of gang crime, and behavioral precautions among Whites, Latinos, and Vietnamese. Journal of Criminal Justice, 32, Ethnicity, information sources, and fear of crime.
Deviant Behavior, 24, Levitt, Steven, The changing relationship between income and crime victimization. Levitt, Steven, and Miles, Thomas Mitchell Polinsky, and Steven Shavell. Levitt, Steven. Juvenile Crime and Punishment. Journal of Political Economy. Liska, Allen E. Functions of Crime: A Paradoxical Process. Reed Social Forces, 66, Lochner, Lance, and Moretti, Enrico American Economic Review, Madriz, E. Images of criminals and victims: A study on women's fear and social control. Gender and Society, 11 3 , Nothing bad happens to good girls: Fear of crime in women's lives. Powdthavee, Nattavudh Unhappiness and Crime: Evidence from South Africa.
Economica, Reid, L. The gender gap in fear: Assessing the interactive effects of gender and perceived risk on fear of crime. Sociological Spectrum, 24, Rountree, P. A reexamination of the crime-fear linkage. Journal of Research in Crime and Delinquency, 35, Burglary victimization, perceptions of crime risk, and routine activities: A multilevel analysis across Seattle neighborhoods and census tracts. Journal of Research in Crime and Delinquency, 33, Perceived risk versus fear of crime: Empirical evidence of conceptually distinct reactions in survey data.
Social Forces, 74, Journal of Quantitative Criminology, 3, Samuel L. Correlations showed that victimization and offending were affected by many of the same shared environmental factors. This result implied that shared environmental factors were important for explaining the victim-offender overlap. Given the result that both genes and environment were important for explaining victimization and offending, we next tested the extent to which the genetic and environmental factors influencing each outcome were similar by transforming the results of the Cholesky decomposition into the mathematically identical correlated factors solution Loehlin The correlation r between the factors was estimated using the following formula;.
However, the shared environmental factors that accounted for the overlap between victimization and offending were correlated at 0. The unique environmental factors between victimization and offending were correlated at 0. In sum, our results show that in this cohort, victimization and offending were under genetic and environmental influence. A large part of the overlap in victimization and offending could be attributed to many of the same unmeasured environmental factors. We analyzed sex heterogeneity in the victim-offender overlap and found that, while there were some minor sex differences, the results between males and females were substantively similar.
With about pairs of boys and pairs of girls, our cohort was not large enough to yield decisive tests of sex differences [ ]. Sex-specific correlations followed the same pattern and can be found in Table S. The second part of our analysis used a variable-centered approach to test the hypotheses that early childhood personal risk factors and the accumulation of adverse childhood experiences positively contribute to the victim-offender overlap.
In this approach, we examined partial correlations between victimization and offending while controlling for each risk factor separately. A reduced correlation would imply that the risk factor helps explain the victim-offender overlap. Association between risk factors, victimization v and offending o in a birth cohort. The table shows correlations between each risk factor with victimization and offending, respectively, and the partial correlation between victimization and offending, after partialling out effect of each risk factor.
As anticipated, when taken one-by-one, the single adversities contained in the ACEs scale produced weak reductions in the correlation between victimization and offending. We tested whether victim-offenders differed on risk predictors from offenders-only, victims-only, or adolescents who were neither. The interesting question was whether victim-offenders differed from victims-only and offenders-only. In terms of personal risk factors measured in childhood, victim-offenders, compared to all other groups, had lower self-control, were significantly more likely to be diagnosed with conduct disorder, self-reported delinquency and had experimented with substance use.
In terms of adverse childhood experiences, victim-offenders, compared to both victims-only and offenders-only, had scored significantly higher on cumulative ACEs. Victim-offenders tended to be worse-off using the individual items from the ACEs scales, but this difference failed to reach statistical significance for many individual ACEs.
Are victim-offenders unique? Comparing victim-offenders to three groups individuals who are neither victims nor offenders, victims-only, and offenders-only. These controls were important because it is possible that any observed differences between victim-offenders and victims-only and offenders-only could arise simply because victim-offenders are more frequently victimized or are more frequent, high-volume offenders. For each of our eight key risk factors, we used logistic regression to test the likelihood of being a victim-offender versus being a a victim-only and b an offender-only, controlling for victimization frequency and offending frequency.
After controlling for adolescent victimization frequency, the odds that a Study member was identified as a victim-offender compared to a victim-only were still significantly increased by lower self-control 1. After controlling for adolescent offending frequency, the odds that a Study member was identified as a victim-offender compared to an offender-only were still significantly increased by lower self-control 1. Association between personal risk factors and ACEs and being a victim-offender versus a a victim-only and b an offender-only. Models of model 1 type include the risk factor and male.
Models of model 2 type include the risk factor, male, and victimization or offending variety. In summary, these models showed that in this cohort the victim-offender overlap could be partly explained through both personal risk factors and cumulative ACEs. We additionally conducted a sensitivity analysis using offending as recorded in the UK Police National Computer. However, when using official records as the indicator of offending, offenders-only and victim-offenders did not appear significantly different from each other, probably because police-recorded offenders—whether victims or not—are in the upper end of the spectrum of risk.
Criminal justice policies have historically tended to contrast victims with offenders, but they are often the same people [ 66 ]. Our findings provide new evidence that during the peak age-period of crime victimization and offending, many victims are offenders and many offenders are victims. Moreover, victim-offenders seem to be characterized by early-onset antisocial behavior and are exposed to multiple adverse childhood experiences. Our findings make three novel contributions. First, our quantitative twin models showed that the victim-offender overlap is influenced by the environment and that victimization and offending have many environmental risks in common.
Recall, prior studies had shown that most of the variation in the victim-offender overlap was due to genetic factors. Second, in line with past developmental research comparing single risks to an accumulation of risks for other, non-crime, outcomes, our findings showed that childhood risk for the victim-offender overlap is cumulative in nature.
Non-offending or not being victimized was associated with few or no risks. Being both a victim and an offender was associated with the largest accumulation of risks. Being either a pure victim or a pure offender fell in-between and was associated with a modest number of risks. Prior studies of the victim-offender overlap often reported that risk factors had little or no effect on the overlap, but this may have arisen from testing one risk factor at a time.
Third, we studied risk factors prospectively assessed during childhood, and our results illustrate that a developmental approach is useful for predicting the victim-offender overlap. Victim-offender overlap research can be further advanced through tests of other life course theories e. This focus does not detract from the value and importance of research on the situational determinants of the victim-offender overlap. Many of the factors that we analyzed could easily predispose victims and offenders to switch roles during a single incident, or within a few days or weeks.
This research aims to highlight the relevance of a longitudinal view of the overlap and highlight its developmental etiology. Current theories of crime over the life course should be augmented to consider victimization as an outcome predicted by similar risks as offending and an outcome likely to occur simultaneously with offending in the presence of an abundance of risk. In line with meta-analyses on the heritability of antisocial behavior [ ], our twin models supported our first hypothesis that both genes and environmental factors would be important for explaining the victim-offender overlap.
Past research, in contrast, found that shared environmental factors had no significant effect on the victim-offender overlap [ 6 , ]. Sample differences in age could not explain the contrasting results, as the age range in our study was not noticeably different from ages covered by past research. In contrast, our measure of victimization captured a wider variety of types of victimization, a few of which were family-based. The prevalence and frequency of victimization in our study are congruent with nationwide estimates of victimization derived from the NSPCC survey [ 45 ].
Regardless of particular data-related issues, a past meta-analysis of antisocial behavior [ ] implies that many more behavioral genetic studies will be needed before drawing firm conclusions about genetic and environmental effects on the victim-offender overlap. Speaking to the broader controversy of using a twin design [ 8 , 18 , 77 ], our results showed that twin designs in criminology are valuable for showing environmental effects. Our person- and variable-centered models partially supported our second hypothesis that personal risk factors would be important for explaining the victim-offender overlap.
Similarly, in the person-centered approach, Study members who had been diagnosed with conduct disorder, and those with high amounts of low self-control, childhood delinquency, and childhood substance use were significantly more likely to become a victim-offender versus a victim-only or an offender-only during adolescence. Our results thus supported the prediction from theories of individual heterogeneity that low self-control partly explains the victim-offender overlap.
The two other personal risk factors, low cognitive ability and early-onset puberty, were not supported as risks for the victim-offender overlap. To the best of our knowledge, this is the first analysis to test the contribution of these two risk factors to the victim-offender overlap and replication of this novel null finding is needed. Both our person- and variable-centered models supported our third hypothesis that cumulative adverse childhood experiences ACEs would be important for explaining the victim-offender overlap.
The results of the variable-centered and person-centered approaches showed that individual ACEs, considered one-by-one, were unhelpful in explaining the victim-offender overlap. In contrast, cumulative ACEs produced relatively stronger reductions in the correlation between victimization and offending, differed significantly between victim-offenders and both victims-only and offenders-only, and significantly increased the odds of being a victim-offender versus an offender-only.
Cumulative risk appeared especially important for neighborhood measures which, when measured individually, had little association with the victim-offender overlap. There are limitations to our study. First, Study members were asked about their victimization over a six-year period age 13—18 , while they were asked about their offending over a one-year period age 17— For our study, this meant that the overlap may have been based on events a number of years apart. A measure using the same 1-year exposure window may have led to fewer Study members reporting victimization and possibly fewer victim-offenders.
In this regard, for our analysis, a longer window of offending may have been more appropriate than restricting the window of victimization. Recall that we were able to capture a multi-year window of offending from age 10 onwards with the police-reported offense data Table S.
Second, we are limited in our ability to make causal inferences. We could not assess the temporal order of victimization and offending, only the overlap. We were, thus, unable to evaluate whether victimization caused offending or vice-versa. Additionally, our research can only support personal risk factors and cumulative adversities as indicators of risk for being a victim-offender and not necessarily as indicators of causation.
Yet, randomized controlled experiments of parenting programs have connected lower childhood antisocial behavior and lower childhood maltreatment to lower adolescent delinquency and victimization [ 34 , 84 , 89 , ]. Thus, it would seem that reducing childhood antisocial behavior and ACEs could also lower the risk of children becoming adolescent victim-offenders. Moreover, given the nature of ACEs and the behavioral genetic results showing the influence of the shared environment, the family appears to be a good target for intervention. Third, given the complexity of the current study, we have not delved further into how childhood risks may predict different types of victimization and offending.
However, offense specialization in adolescence is uncommon [ 90 ] and many of our Study members were victims of more than one type of crime see Table S. Future research should nonetheless consider the possibility of different childhood risks predicting different types of victimization and offending.
Fourth, the sample was composed of twins, so the results may not generalize to singletons. However, past studies of singletons show results similar to our prevalence of adolescent victimization and offending and childhood conventional ACEs. Additionally, past research has demonstrated the similarity between twins and singletons in antisocial behavior and risk factors for antisocial behavior [ 7 ]. Our results thus support the developmental axiom that past behavior and experience are predictive of future behavior and experience, and we extend this observation to show that it applies to the victim-offender overlap.
Established delinquency-prevention programs have the opportunity to warn their clients about the risks of victimization and, perhaps, educate them on self-protective measures. Our study also showed that existing theories of the victim-offender overlap should be adapted to take a developmental perspective of risk beginning in childhood. Efforts to prevent adolescent victimization and offending may begin with a situational focus, such as programs to prevent retaliatory violence, but could improve their efficacy by recognizing that victimization and offending have their roots in childhood.
We thank David L.
The Challenges of Defining and Measuring Defensive Gun Use | RAND
Corcoran and Joseph A. Amber L. The unstandardized variance components, standardized variance components, and the formulae for calculating the correlations between the factors can be found in Table S. This, however, is not the case with antisocial behavior. Instead, the influence of both genetic and shared environmental factors tends to decline with age while the influence of unique environmental effects increases [ ].
National Center for Biotechnology Information , U. Journal of Developmental and Life-Course Criminology. J Dev Life Course Criminol. Published online Oct 9. Barnes , 6 Helen L. Odgers , 8 Jasmin Wertz , 1 and Terrie E. Moffitt 1, 3, 4, 5. Helen L. Candice L. Terrie E. Author information Article notes Copyright and License information Disclaimer. Beckley, Email: ude. Corresponding author. This article has been cited by other articles in PMC. Abstract Purpose It is well-established that victims and offenders are often the same people, a phenomenon known as the victim-offender overlap, but the developmental nature of this overlap remains uncertain.
Results In contrast to past twin studies, we found that environment as well as genes contributed to the victim-offender overlap. Conclusions This study showed that the victim-offender overlap is, at least partially, developmental in nature and predictable from personal childhood characteristics and an accumulation of many adverse childhood experiences. Electronic supplementary material The online version of this article Keywords: Victim-offender overlap, Developmental criminology, Adverse childhood experiences, Low selfcontrol.
Past Victim-Offender Overlap Research from a Developmental Perspective Three general streams of research form the basis of knowledge on the victim-offender overlap. The Present Study In this study, we analyzed data on year-old twin pairs to test explanations of the overlap between victimization and offending behavior.
Using variable-centered and person-centered approaches, we tested the hypotheses that H1: Both genes and environment contribute to the victim-offender overlap. Results Our preliminary analytic step was to verify that the victim-offender overlap existed in our data. Open in a separate window. Table 2 Association between risk factors, victimization v and offending o in a birth cohort. Pairwise correlations for full sample a Percent difference between zero-order correlation 0. Table 3 Are victim-offenders unique? Table 4 Association between personal risk factors and ACEs and being a victim-offender versus a a victim-only and b an offender-only.
Models of model 2 type include the risk factor, male, and victimization or offending variety OR odds ratio, CI confidence interval. Discussion Criminal justice policies have historically tended to contrast victims with offenders, but they are often the same people [ 66 ]. Limitations There are limitations to our study.
Footnotes 1 The effect of cumulative adversities that we discuss and measure differs from the process of cumulative disadvantage described in life-course theory [ ], which describes how, through the process of labeling, young offenders experience subsequent life failures, leading to adult offending.
References 1. Aaltonen, M. To whom do prior offenders pose a risk? Victim—offender similarity in police-reported violent crime. Accessed 19 July. The enduring effects of abuse and related adverse experiences in childhood: a convergence of evidence from neurobiology and epidemiology. European Archives of Psychiatry and Clinical Neuroscience. Averdijk M. Reciprocal effects of victimization and routine activities. Journal of Quantitative Criminology. Genetic and environmental influences on victims, bullies and bully-victims in childhood. Journal of Child Psychology and Psychiatry.
Extending research on the victim—offender overlap evidence from a genetically informative analysis. Journal of Interpersonal Violence. A demonstration of the generalizability of twin-based research on antisocial behavior. Behavior Genetics. Demonstrating the validity of twin research in criminology. Street youth and criminal violence. Journal of Research in Crime and Delinquency.
The association between intelligence and personal victimization in adolescence and adulthood.
Welcome to the National Center for Victims of Crime
Personality and Individual Differences. Etiological features of borderline personality related characteristics in a birth cohort of year-old children. Development and Psychopathology. Berg MT, Loeber R. Examining the neighborhood context of the violent offending-victimization relationship: a prospective investigation.
The victim—offender overlap in context: examining the role of neighborhood street culture. The influence of early attachments on other relationships. New York: Guilford Press; Human agency, capable guardians, and structural constraints: a lifestyle approach to the study of violent victimization.
Journal of Youth and Adolescence. Boker, S. Open Mx 2. Brezina T.