Journal of Research in Crime and Delinquency 49(4)

Journal of Research in Crime and Delinquency, November 2012: Volume 49, Issue 4

Carjacking and Copresence
Bruce A. Jacobs

Drawing from in-depth qualitative interviews, this research note explores how carjacking offenders establish copresence with victims. Two tactics are explored: normalcy illusions and blitzes. The paper links these tactics conceptually to the delivery of coercive actions. Discussion focuses on the offenders' manipulation of celerity and publicity in the approach process and its implications for compliance generation in an alternative coercive climate.

Structural Covariates of Gang Homicide in Large U.S. Cities
David C. Pyrooz

Objectives: This study examined the structural covariates of gang homicide in large U.S. cities and whether the structural conditions associated with gang homicide differed from non-gang homicide. Methods: Several national data sources were used to gather information on the structural conditions of the 88 largest U.S. cities, including the U.S. Census Bureau, Law Enforcement Management and Administrative Statistics, Uniform Crime Report, and National Gang Center. Negative binomial regression was used to model the relationship between the structural conditions of cities and homicide rates. Results: Socioeconomic deprivation, official rates of gang membership, and population density explained between-city variability in gang homicide rates. In addition, quadratic associations were observed for socioeconomic deprivation and population density. Equality of coefficients tests revealed that the structural covariates of gang homicide differed in magnitude from non-gang homicide. Conclusions: Prior to this study, the etiology of gang homicide was found to differ from other homicide types in terms of event characteristics and sub-city correlates. This macro-level study extended this line of research to cities, providing evidence that the structural correlates of violence operated differently for gang homicide.

White Perceptions of Whether African Americans and Hispanics are Prone to Violence and Support for the Death Penalty
James D. Unnever and Francis T. Cullen

Objective: To explore whether the impact of racial and ethnic pejorative stereotypes and prejudice on White support for the death penalty changes over time. Methods: The data were drawn from the 1990 and 2000 General Social Surveys. This trend analysis included a four-item racial–ethnic prejudice scale and two stereotype or “typification” measures that assessed the extent to which the respondents believed that African Americans and Hispanics were more prone to violence than Whites. Controls were introduced for standard demographic measures and for known covariates of punitiveness, such as conservative political ideology and religious beliefs and involvement. The dependent variable was whether the respondents favored the death penalty for persons convicted of murder. Results: The analyses revealed that racial and ethnic prejudice increases Whites' embrace of capital punishment in both periods of time. By contrast, the negative stereotypes that typify African Americans and Hispanics as prone to violence exerted a significant impact on support for the death penalty in 1990 but not in 2000. Conclusions: These findings suggest that Whites' views on minority group members' proneness to violence potentially shift over time and exert significant impacts on death penalty support depending on the broader social context that prevails at a given historical juncture. Racial–ethnic feelings of animus appear to be a stable, if not intractable source of punitiveness. Future research should continue to unpack the complex nature of Whites' racial and ethnic views and to explore how, in given sociopolitical contexts, they potentially serve to justify punitive policy agendas.

Partners in Crime? Criminal Offending, Marriage Formation, and Partner Selection
Marieke van Schellen, Anne-Rigt Poortman, and Paul Nieuwbeerta

Objectives. This study aims to examine the impact of criminal offending on marriage formation and partner selection. Although it is well established that marriage has the potential to foster desistance from crime, far less attention has been paid to offenders' marital chances. The few studies that did investigate the effects of crime on marriage formation have not taken into account spouses' criminal behavior. However, desistance from crime may strongly depend on the criminal history of the spouse. Method. We employ data from a unique longitudinal study: The Criminal Career and Life-course Study (CCLS). The CCLS contains data on the officially registered criminal careers of 4,615 Dutch offenders and their spouses. To analyze the relationship between criminal offending and outcomes in the marriage market, we use event history models. Results. The results show that not only the seriousness of a criminal history but also the timing of criminal convictions affects offenders' outcomes in the marriage market. The more criminal offenses individuals have committed, the lower the chances of marrying and, given marriage, the higher the chances of marrying a criminal partner. The impact of a criminal record on marriage likelihood becomes weaker when offenders have been convicted a longer time ago. Conclusions. On the basis of our findings, we have to nuance the prominent idea that marriage reduces criminal behavior. Because of their lower marital chances and their tendency to marry criminal partners, offenders are less likely to experience protective effects of marriage.

The Influence of Travel Distance on Treatment Noncompletion for Juvenile Offenders
Brian Lockwood

Objectives: This study seeks to identify the effects of travel distance on treatment noncompletion for juveniles attending community-based offender programs. Methods: A population of more than 6,000 juvenile offenders adjudicated in Philadelphia’s Family Court is analyzed using hierarchical linear models. Distance to treatment is operationalized with Euclidean distance. Treatment noncompletion is disaggregated by type to distinguish between noncompletion due to dropout and expulsion. Results: Results indicate that distance to treatment influences noncompletion due to dropout, but not due to expulsion. In the cross-classified models, an increase of approximately 3 miles to treatment is shown to increase the odds of treatment noncompletion due to dropout by nearly 100 percent. Conclusions: Implications of this research suggest that Euclidean distance is an appropriate measure with which to estimate the travel of young offenders in urban space and that juvenile justice policymakers should consider distance to treatment when matching young offenders to treatment facilities.

On the Operational Validity of Perceptual Peer Delinquency: Exploring Projection and Elements Contained in Perceptions
John H. Boman, IV, John M. Stogner, Bryan Lee Miller, O. Hayden Griffin, III, and Marvin D. Krohn

Objectives. The authors examine perceptions of a peer’s substance use to determine whether and to what degree individuals project their own behavior onto their perceptions of peer’s delinquency, and to determine whether the constructs of self-control and peer attachment are related to perceptions. Methods. Using a sample of 2,154 young adult respondents within friendship pairs in which each respondent reported their own substance use and their perception of the friend’s use, the authors estimate a series of regression models with perceptions of a peer’s alcohol, marijuana, Salvia divinorum, and hard drug use as dependent variables. Results. Perceptions of a peer’s substance use are approximately equally related to a peer’s and a respondent’s use of each substance. Projection occurs to a greater extent when perceiving low-frequency behaviors. Low self-control is sporadically associated with higher perceived substance use. Conclusions. Peer self-reported delinquency and perceptions of peer delinquency are distinct constructs. Because projection appears to be worse for infrequent behaviors, researchers should use caution when using low-frequency behaviors to measure perceptual peer delinquency. Although the data used are cross sectional, the perceptual measure is confounded by too many variables other than a peer’s actual delinquency to be considered a valid measure of the sole construct of peer delinquency.