|
Abstract
|
VioLit summary:
OBJECTIVE:
The purpose of this study by Lauritsen et al. was to provide a detailed, longitudinal analysis which explored both victimization among juveniles and connections between offending and victimization. Using the first five waves of the National Youth Survey (NYS), this study examined the effects of delinquent lifestyles on the criminal victimization of juveniles.
METHODOLOGY:
This study examined the victimization of young adults and the connections between these youth's victimization and their offending both of which have traditionally been areas ignored by victimization research. Demographic factors, involvement in delinquent lifestyles, and physical proximity to crime and social disorder are the three areas used to test a lifestyle theoretical perspective of victimization. The NYS, a large longitudinal study consisting of five waves of personal interviews beginning in 1977 with youth ages 11 to 17 years who were randomly selected across the U.S., was utilized to analyze the connections between the three lifestyle factors and victimization. Age, sex (1=male, 0=female), race (1=black, 0=white), family structure (1=two parent home, 0=all other home structures), and family income (1 to 10 scale of lower to higher income) were the demographic factors measured. A delinquent lifestyle was measured in two ways: the number of times the youth self-reported engaging in predator activities in the last year and the amount of time the youth spent with their peers per week multiplied by the extent of their peers' involvement in delinquency. Physical proximity to crime and social disorder was measured using the results of a question only in wave 1 asking the parents of the youth about the extent of vandalism, winos and junkies, abandoned housing, burglary and theft in their neighborhoods. Each of these factors was weighed to produce a reliable index called the neighborhood disorder index. However, because this information was only available in wave 1, it should be interpreted with caution. The dependent variable of victimization was measured by four types of victimization. Assault victimization measured the number of times in the past year the youth reported being a victim either of attempted or completed beatings (other than by parents) or attacks with a weapon. Robbery victimization was measured by the number of times in the last year the youth had something taken from them by force (attempted or completed). Larceny victimization included the number of times the youth reported having a bicycle or motorbike stolen, something stolen from their bicycle or motorbike, or something stolen from a public place. Vandalism victimization was measured by the number of times the youth reported having personal belongings purposely destroyed or damaged. The demographic factors did not change over the five waves; some variables were not available for all waves; both delinquency and victimization decreased with age. Prior victimization and delinquency experienced by the subjects were statistically controlled. The methodological concerns about self-reporting were addressed, as well as issues of reliability and validity. This study used both bivariate and multivariate (i.e. ordinary least squares regression) statistical tests, as well as percentages. Using a full-information maximum-likelihood (FIML) simultaneous model, the methodological concerns of lagged dependent variables and the possibility of victimization leading to delinquency were addressed.
FINDINGS/DISCUSSION:
In the bivariate analysis, the association between the four types of victimization and the subject's own delinquent behavior were examined. Delinquents appeared to be about four times more likely to be assaulted than nondelinquents. They were also, across all five waves, significantly more likely to be the victim of robbery, larceny, and vandalism. Using multivariate analyses, a more complete examination of the relationships between individual characteristics, lifestyles and victimization within the sample was possible. First, when interpreting the results of assault victimization, it was found that males reported being assaulted more than women; wealthier youth were less likely to report being assaulted; youth in neighborhoods with high crime levels reported experiencing more assaults; and whites reported experiencing more assaults than blacks (which may be result of under-reporting by blacks or over-reporting by whites). Also, increases in delinquent lifestyle were positively related to increased risk of assault (p<.01). The direct effect of other factors on victimization were greatly reduced when controlling for delinquent lifestyle. In other words, the majority of factors which were found to be statistically significant appeared to operate indirectly through delinquent lifestyle. Similar results, which report that the dominant direct effect on victimization was delinquent lifestyle, were found for robbery, larceny and vandalism (p<.01). Engagement in prosocial activities (i.e. time with family, community activities) appeared to only be important in that they reduced the likelihood of delinquent behavior. From the FIML simultaneous model, it was assumed that the effects of prior or lagged delinquency on victimization operated only through their effect on current delinquency; and that the effects of prior victimization on delinquency were only relevant through their effect on current victimization. The FIML models supported the reciprocal relationship between victimization and delinquency. This study concluded, using the NYS data, that delinquency does increase adolescents' risk of victimization.
AUTHORS' RECOMMENDATIONS:
The authors warned against interpreting their results in a manner which would blame the victims of crimes for their victimization. They also stressed the difficulty in and problems with measuring and assessing victimization risk among youth. They suggested the need for more research combining large longitudinal studies, like NYS, with rich, detailed ethnographies and life histories of juveniles which would provide details of the situational dynamics between victimization and delinquency. (CSPV Abstract - Copyright © 1992-2007 by the Center for the Study and Prevention of Violence, Institute of Behavioral Science, Regents of the University of Colorado) |