Showing papers in "Journal of Quantitative Criminology in 2002"

Friendship Networks and Delinquency: The Relative Nature of Peer Delinquency

Abstract: Although acknowledging the importance of adolescent friendships in the etiology of delinquency, prior studies have yet to provide a detailed examination of the role of actual friendship networks in delinquency. Using data from the National Longitudinal Study of Adolescent Health (1995–1996), this study’s incorporation of friendship networks allows for a more rigorous conceptualization and measurement of peer delinquency based on carefully defined networks of adolescent friendships. Findings illustrate that friendship networks are very heterogenous in terms of members’ participation in delinquent behavior with the majority of adolescents belonging to networks containing both delinquent and non-delinquent friends. In support of differential association’s premise that delinquent behavior is influenced by the ratio of definitions favorable to those unfavorable to law violation (Sutherland, 1947), the proportion of delinquent friends in a respondent’s network is most strongly associated with respondents’ subsequent delinquency. This relative measure of peer delinquency is preferable to a measure of the absolute level of delinquency occurring by friends, the average delinquency committed by friends, or the absolute number of delinquent friends. Enmeshment in a friendship network where consensus about the appropriateness of delinquency is maximized (i.e., all friends are delinquent or non-delinquent) most effectively constrains the behaviors of network members to resemble the groups’ behavior.

Analyzing multiple-item measures of crime and deviance I: Item response theory scaling

Abstract: Multiple-item measures of self-reported offending typically provide the principal outcome measures for individual level research on the causes of crime and deviance. This article directs attention to the substantial problems presented by the task of forming composite scores for these measures, and it presents a possible solution to those problems. We consider scaling by means of the graded response model from item response theory (IRT) as a potential means of overcoming the shortcomings of traditional summative scaling and of obtaining valuable information about the strengths and weaknesses of our measures. We illustrate this strategy through a scale analysis of a fourteen-item, self-report measure of delinquency, using three years of data from the Monitoring the Future study, an annual national survey of high school seniors. The graded response model proves to be consistent with the data, and it provides results that address important substantive questions about self-report measures. The findings are informative about the strengths and weaknesses of alternative strategies for developing self-report instruments, indicating that there is little to be gained by making fine distinctions in the frequency of individual delinquent acts.

A Note on the Use of County-Level UCR Data

Abstract: County-level crime data have major gaps, and the imputation schemes for filling in the gaps are inadequate and inconsistent. Such data were used in a recent study of guns and crime without considering the errors resulting from imputation. This note describes the errors and how they may have affected this study. Until improved methods of imputing county-level crime data are developed, tested, and implemented, they should not be used, especially in policy studies.

Aoristic Signatures and the Spatio-Temporal Analysis of High Volume Crime Patterns

Abstract: The spatial analysis of crime and the current focus on hotspots has pushed the area of crime mapping to the fore, especially in regard to high volume offenses such as vehicle theft and burglary. Hotspots also have a temporal component, yet police recorded crime databases rarely record the actual time of offense as this is seldom known. Police crime data tends, more often than not, to reflect the routine activities of the victims rather than the offense patterns of the offenders. This paper demonstrates a technique that uses police START and END crime times to generate a crime occurrence probability at any given time that can be mapped or visualized graphically. A study in the eastern suburbs of Sydney, Australia, demonstrates that crime hotspots with a geographical proximity can have distinctly different temporal patterns.

Difference, Disparity, and Race/Ethnic Bias in Federal Sentencing

Abstract: Federal sentencing guidelines were enacted to reduce unwarranted disparities in sentencing. In this paper we examine the degree to which disparity in sentencing on the basis of race and ethnicity occurred in federal sentencing after the guidelines were implemented. We consider how much of the disparity is explained by offense-related factors as specified in the guidelines. We find that African Americans, Hispanics, and Native Americans receive relatively harsher sentences than whites and that these differentials are only partly explained by offense-related characteristics. We interpret our findings in light of attribution, uncertainty avoidance, and conflict theories.

Analyzing multiple-item measures of crime and deviance II: Tobit regression analysis of transformed scores.

Abstract: The purpose of this article is to inform criminological researchers about tobit regression, an alternative regression model that deserves more attention in this field. Tobit regression is intended for continuous data that are censored, or bounded at a limiting value. The tobit model may be a particularly good match to measures of self-reported offending, provided they have been transformed to reduce skewness. We present empirical analyses that evaluate the match of self-report measures to the assumptions of ordinary least square (OLS) and tobit regression models and that assess the consequences of any violations of assumptions. The analyses use a fourteen-item, self-report measure of delinquency from the Monitoring the Future study, a national survey of high school seniors. These analyses provide clear evidence that (1) transformations to reduce skewness improve the match of OLS to the data but still leave considerable discrepancies, and (2) the tobit model is well suited to the transformed measure. We conclude by assessing the purposes for which tobit offers greater and smaller advantages over OLS regression.

Growing Up Poor: Examining the Link between Persistent Childhood Poverty and Delinquency

Abstract: Findings from aggregate-level and ethnographic research suggest that poverty and delinquency are related. The inability of individual-level quantitative research to demonstrate consistent evidence of this relationship, however, has been used to call into question whether poverty is indeed related to an increased propensity for delinquent involvement. This may be due to the difficulty individual-level analyses have in identifying the group most important in uncovering the relationship of poverty to delinquency—those individuals that experience persistent childhood poverty. This paper provides an assessment of the effects of both the level of exposure to poverty and its timing on delinquent involvement using fourteen years of longitudinal data for a national sample of younger adolescents. Findings indicate that exposure to poverty and the timing of such exposure are indeed related to an increased likelihood of involvement in delinquency.

Sex Differences in Age Patterns of Delinquent/Criminal Careers: Results from Poisson Latent Class Analyses of the Philadelphia Cohort Study

Abstract: Applications of latent class analyses to the study of criminal careers have yielded results with implications for criminological theory. Distinct latent classes of individuals within various samples have been identified based upon the similarity of individuals with respect to their rate of offending across the teen and adult years, net of the effects of other regressors. In previous research on samples of males taken from the cities of London and Philadelphia, four and five such categories have been identified respectively, ranging from a group of nonoffenders to a group of chronic offenders. However, the question of whether similar findings hold for females has not been adequately addressed, in part due to the scarcity of longitudinal samples with sizable female populations. Data from the Second Philadelphia Cohort are used to address this and related questions. First, are there latent classes of female offenders? Second, if such categories do exist, how do they compare? Third, how do classes of male and female offenders compare on key measures of criminal careers? Analyses of the samples yield differing numbers of classes for males and females. Gender invariances as well as differences in patterns of offending are also found and are discussed.

Reassessing the Cross-National Relationship Between Income Inequality and Homicide Rates: Implications of Data Quality Control in the Measurement of Income Distribution

Abstract: A significant positive relationship between income inequality and homicide rates has been found in a large number of cross-sectional studies and a few longitudinal analyses; a theoretically interesting interaction effect between income inequality and social welfare has also been found For the most part researchers have been willing to use generous quality criteria in collecting income-distribution data, to maximize sample size and thereby enhance statistical power and representativeness In the present paper we use a recently developed data set with explicit quality ratings for the income-distribution data, which permits systematic assessment of the consequences of relying on income-distribution measures of varying quality in examining the inequality-homicide relationship Our analyses reveal that the income inequality-homicide relationship is remarkably robust in cross-sectional analyses Regardless of the quality of income-distribution data, we observe significant positive effects of the Gini coefficient on homicide rates in cross-sectional multivariate models Consistent results are also observed when an interaction term for income inequality and decommodification is examined The results of longitudinal analyses differ; we observe a significant positive effect of changes in inequality on changes in homicide rates only when income-distribution measures of low quality are used

Routine Activities and Deviant Behaviors: American, Dutch, Hungarian, and Swiss Youth

Abstract: The current investigation examined cross-national similarities and differences in routine activities, measures of deviance, and their relationship in representative samples of 7,000 adolescents aged 1519 years (mean age: 175 years) from Hungary, the Netherlands, Switzerland, and the United States For the majority of youth, most of their time was spent in solitary activities, followed by peer activities, community/sports activities, and family activities; Hungarian youth reported spending a much greater amount of time with the family than adolescents from other countries, while Dutch youth spent far more time in solitary activities than their peers Rates of total deviance were remarkably similar for American, Dutch, and Swiss youth; Hungarian youth reported substantially lower rates than all other adolescents Finally, findings indicated that routine activities accounted for 18% for males and 16% for females of the variance explained in total deviance Furthermore, with the exceptions of alcohol and drug use, country had very little or no explanatory power in deviance The current study suggests that the utility and the explanatory power of the routine activities framework replicates across national boundaries

Reconsidering Domestic Violence Recidivism: Conditioned Effects of Legal Controls by Individual and Aggregate Levels of Stake in Conformity

Abstract: Social scientists examining whether stake in conformity conditions the deterrent effect of arrest for domestic violence recidivism have applied criminological theory to an important criminal justice issue. We extend this research with a discussion and multi-level analysis of the possible interplay between court dispositions and (a) an offender's stake in conformity, and (b) the proportion of “higher stake” residents in an offender's census tract of residence. The prevalence of re-arrest for intimate assault (misdemeanor and felony) is examined for 3110 suspects of misdemeanor intimate assault in Cincinnati. Findings reveal a significant main effect involving higher re-arrest likelihoods for arrested suspects with no formal charges filed against them. Results for the conditioned effects of court dispositions reveal significantly lower re-arrest likelihoods for higher stake offenders undergoing a counseling program (a predicted relationship), and significantly lower re-arrest likelihoods for lower stake offenders serving probation and/or jail (opposite to the predicted relationship). At the neighborhood-level, sentences of probation and/or jail correspond with significantly lower re-arrest likelihoods for offenders living in neighborhoods with more residentially stable populations (as predicted). We discuss the implications of our study for future research.

A Test of Generalizability of the Social Development Model Across Gender and Income Groups with Longitudinal Data from the Elementary School Developmental Period

Abstract: The social development model (SDM) is a theory of behavior that has proven useful in explaining the etiology of delinquency, violence, and substance use among adolescents as well as early antisocial behavior among pre-adolescents. A further test of the model is its generalizability across population groups. A section of the SDM representing prosocial influences in the etiology of problem behavior was compared for boys and girls and for children from low- and non low-income families using three waves of child, parent and teacher survey data on a sample of 851 elementary school students. Multiple group structural equation modeling was used to assess differences across groups in both measurement of model constructs and hypothesized structural paths between constructs. The results indicate overall similarity in the reliability of measurement models and validity of structural models.

Disentangling the effects of self-protective behaviors on the risk of injury in assaults against women

Abstract: Using data from the National Crime Victimization Survey, this paper attempts to disentangle the effects of self-protective behaviors on the risk of injury in assaults against women. Unlike previous research, in this study we address simultaneously three important conceptual and methodological issues: (1) type of self-protective behavior, (2) temporal sequencing of self-protective behavior in relation to injury, and (3) the victim/offender relationship. Results indicate that even after controlling for other contextual characteristics of an assault, the probability of a woman being injured was lowest when she employed non-physical resistance strategies such as arguing or reasoning with the offender. This was true for all types of offenders. However, for assaults involving intimates, the probability of injury was increased for women who physically resisted their attackers.

Assessing the Effect of Batterer Program Completion on Reassault: An Instrumental Variables Analysis

Abstract: The main objective of this paper is to investigate the utility of using the instrumental variables (IV) method to estimate batterer program efficacy, i.e., the program effect among batterers who complete batterer programs. This method takes account of possible confounding due to unmeasured traits of compliers and non-compliers. A structural model is used to estimate the impact of program completion on reassault using instrumental variables (IV) regression. Data on 640 batterers enrolled at three batterer programs are used. Results obtained from IV regression are compared with those obtained from a more traditional regression analysis. The results indicate that usual regression methods yield estimates of program effect that may be biased due to confounding by unmeasured batterer characteristics. Unfortunately, IV estimates may be unreliable due to failure of some of the assumptions on which they are based. If equations are adequately identified by the non-linear functional form used to estimate them, then IV results indicate that among a very select group of batterers, program completion significantly reduces the probability of reassault. The implications of confounding due to program non-compliance, program non-enrollment and attrition for future evaluations of batterer programs are discussed.

The Impact of the 1994 Federal Assault Weapons Ban on Gun Markets: An Assessment of Short-Term Primary and Secondary Market Effects

Abstract: The reactions of the gun market, including those of producers, wholesalers, retailers, and consumers, play an important role in shaping the potential impact of gun control policies on gun crime. As a case in point, this paper examines the federal Violent Crime Control and Law Enforcement Act of 1994, which bans a group of military-style semiautomatic firearms (i.e., assault weapons). Using a variety of national and local data sources, we assess the short-term (1994–1996) impact of the assault weapons ban on gun markets, examining trends in prices and production of the banned weapons in legal markets and assessing the availability of the banned weapons in illicit markets as measured by criminal use. Prices of assault weapons rose substantially around the time of the ban's enactment, reducing the availability of assault weapons to criminal users in the very short run. However, a surge in assault weapon production just before the ban caused prices to fall in the months following the ban. Implications of the findings for assessing this and other gun control policies are discussed.

Estimates of Repeat Victimization Using the National Crime Victimization Survey

Abstract: Many crime victims experience multiple victimizations over time Estimating the rate of repeat victimization from a longitudinal survey, however, is difficult because individuals often have missing data for some of the interviews We use data from the 1996–98 US National Crime Victimization Survey to explore potential effects of missing data on estimated rates of repeat violent victimizations in individuals We introduce two algorithms for estimating repeat victimization rates, using logistic models to impute values for individuals who have partial data These models are applied to estimate rates of repeat victimization for all violent crimes, and separately for crimes of domestic violence Estimates show substantial sensitivity to the form of the model used

Identifying Unit-Dependency and Time-Specificity in Longitudinal Analysis: A Graphical Methodology

Abstract: Longitudinal analysis in criminology and other social sciences has become an important research tool because it allows us to draw conclusions from observing how multiple units change over time. Unfortunately, its results are more vulnerable to potential influences of unusual observational units or periods of time. Current leverage diagnostics are designed for cross-sectional analysis and are fallible when applied to longitudinal models. This article introduces a graphical diagnostic methodology to systematically examine the sensitivity of longitudinal results to extreme observational units and periods of time—unit-dependency and time-specificity. Further the article illustrates its use with an example testing policy effects on black and white female victimization of intimate partner homicide. Results are displayed in an easily understood graph that provides a snapshot of the results' time-specific patterns and robustness to unit-dependency. Currently, comparable tests for panel analysis are tedious and cumbersome. With this new illuminating methodology, researchers and policy-makers can easily decide whether a time-specific or unit-dependent pattern is consequential.