LEVELS AND PREDICTORS OF RECIDIVISM: THE MALTA EXPERIENCE*

Research on recidivism has been confined primarily to a few highly industrialized Western nations (e.g., United States, Great Britain, Canada, and Australia) in which the data and resources needed for such research are readily available. The restriction of recidivism research to such a small number of nations begs the question: Do these results reflect patterns of offending and products of the criminal justice systems unique to these nations, or do they describe patterns of recidivism across a much wider range of social and cultural contexts? In this study I extend the scope of recidivism research by examining levels and patterns of recidivism in the Republic of Malta, a small Mediterranean island that differs considerably from the typical context in which recidivism is studied. Specifically, I examine the likelihood of recidivism among persons released from Malta's only prison between 1976 and 1994. In addition, I examine factors shown in previous research to be strong predictors of recidivism to assess their value as predictors of recidivism among Maltese prisoners. Proportional hazards regression models reveal that levels and predictors of recidivism in Malta approximate those observed in societies that are socially and culturally quite different. The findings suggest that the role of social institutions in reintegrating offenders into society may be more complex than commonly believed.     

THE IMPACT OF MEGAN'S LAW ON SEX OFFENDER RECIDIVISM: THE MINNESOTA EXPERIENCE*

Although Megan's Law was passed more than 10 years ago, very little is known as to whether it reduces sex offender recidivism significantly. Using a retrospective quasi‐experimental design, we examine whether community notification has a deterrent effect by comparing the recidivism rates of 155 level 3 (“high public risk”) sex offenders released from Minnesota prisons between 1997 and 2002 who were subject to broad notification with two separate control groups who were not. The first control group (referred to as the prenotification group) contained 125 sex offenders released between 1990 and 1996 (the 7 years preceding the implementation of the Community Notification Act) who likely would have been subject to broad community notification had the law been in effect at the time of their release. The second control group (referred to as the non‐notification group) was composed of 155 offenders (37 level 1 and 118 level 2) released between 1997 and 2002 who were not subject to broad community notification. The results from the Cox proportional hazards models reveal that broad community notification significantly reduced the risk of time to a sex reoffense (rearrest, reconviction, and reincarceration) compared with both control groups. The findings were mixed, however, for both non‐sex and general reoffending. Whereas broad community notification significantly reduced the risk of timing to both non‐sex and general recidivism compared with the prenotification group, no such effects were found in the non‐notification group analyses. We discuss the implications of these results and attempt to explain why Megan's Law seems to reduce sex offense recidivism in Minnesota.     

STABLE PREDICTORS OF RECIDIVISM:

Seventy-one studies are reviewed which present data on the relationship between biographical predictors and recidivism in 177 independent samples of offenders. A summary table lists the number of samples in which a particular predictor was found to be related to recidivism and the number of samples in which that predictor was found to be unrelated to recidivisim. Detailed summaries of the relation between “type of instant offense” and recidivism and “age at first arrest” and recidivism are also included.     

A META-ANALYSIS OF THE GENE-CRIME RELATIONSHIP*

This investigation used the statistical technique of meta-analysis to probe the putative association between heredity and crime. The data for this study were 54 effect sizes obtained from 38 family, twin, and adoption studies on crime. In addition to the overall gene-crime relationship, the potential moderating effects of gender, sample nationality, date of publication, and quality of the research design were also investigated. It was predicted that heredity and crime would not coincide, although subsequent analyses disclosed a low-moderate correlation between these two variables (mean unweighted phi coefficient = .25; mean weighted phi coefficient = .09). Further analysis of these data revealed that better designed and more recently published studies provided less support for the gene-crime hypothesis than more poorly designed and earlier published investigations. The individual strengths and weaknesses of the meta-analytic technique relative to this effort to achieve insight into the gene-crime relationship are discussed.     

PREDICTING THE VIOLENT OFFENDER:THE DISCRIMINANT VALIDITY OF THE SUBCULTURE OF VIOLENCE*

This study tests the extent to which an adherence to the subculture of violence uniquely predicts a tendency to favor violence or instead predicts a more generalized offending repertoire, of which violence is part. Specifically, we use a unique analytic technique that provides the opportunity to distinguish empirically between the “violent offender” and/or the “frequent offender.” The results suggest that holding values favorable toward violence consistently predicts general offending but do not identify youth who systematically favor violence over nonviolence. This discussion considers the impact of these findings for the continued utility of the subculture of violence perspective.     

THE IMPACT OF FLORIDA'S HABITUAL OFFENDER LAW ON CRIME*

Research examining the impact of habitual offender laws on crime has suffered from numerous methodological problems, including failure to consider incapacitation effects that may be responsible for most of the laws' impact but that may not appear until years after the law is passed; not addressing simultaneity issues; and omitting needed control variables to avoid spurious or suppressed results. The present study specifies procedures that can be used to mitigate these problems. These procedures are used in the current study to estimate the impact of Florida's habitual offender law on crime using a multiple time series design, with data for 58 counties from 1980–1998.     

THE EFFECT OF IMPRISONMENT ON RECIDIVISM RATES OF FELONY OFFENDERS: A FOCUS ON DRUG OFFENDERS*

The purpose of this study is to evaluate the deterrent effect of imprisonment. Using data on offenders convicted of felonies in 1993 in Jackson County (Kansas City), Missouri, we compare recidivism rates for offenders sentenced to prison with those for offenders placed on probation. We find no evidence that imprisonment reduces the likelihood of recidivism. Instead, we find compelling evidence that offenders who are sentenced to prison have higher rates of recidivism and recidivate more quickly than do offenders placed on probation. We also find persuasive evidence that imprisonment has a more pronounced criminogenic effect on drug offenders than on other types of offenders.     

PREDICTING THE RECIDIVISM OF SERIOUS YOUTHFUL OFFENDERS USING SURVIVAL MODELS*

Increases over the past decade in the number and seriousness of crimes committed by young offenders have resulted in renewed interest in control of the serious, violent, or chronic youthful offender. Recent research has established a number of theoretically relevant variables associated with persistent offending and offending defined recidivism. Few studies, however, have examined the individual characteristics that might predict the timing of youthful recidivism. This paper seeks to fill some gaps in knowledge of the serious youthful offender by estimating a multivariate survival model for a sample of youths paroled from California Youth Authority institutions. The model's results are then used to develop risk-assessment profiles for the sample. We suggest that these profiles may provide a reasonable approach for establishing treatment protocols or supervision requirements for paroled youths.     

THE LABELING OF CONVICTED FELONS AND ITS CONSEQUENCES FOR RECIDIVISM*

Florida law allows judges to withhold adjudication of guilt for individuals who have been found guilty of a felony and are being sentenced to probation. Such individuals lose no civil rights and may lawfully assert they had not been convicted of a felony. Labeling theory would predict that the receipt of a felony label could increase the likelihood of recidivism. Reconviction data for 95,919 men and women who were either adjudicated or had adjudication withheld show that those formally labeled are significantly more likely to recidivate in 2 years than those who are not. Labeling effects are stronger for women, whites, and those who reach the age of 30 years without a prior conviction. Second‐level indicators of county characteristics (e.g., crime rates or concentrated disadvantage) have no significant effect on the adjudication/recidivism relationship.     

POLITICIZATION OF THE CRIMINAL OFFENDER

A “political model” of crime and punishment is described and the utility of this model is explored in the light of depth interviews with incarcerated offenders. It is suggested that the average offender brings to prison political socialization experiences which have resulted in alienation but not ideological estrangement from the political order. It is further suggested that politicizing processes in prison, coupled with the failure of rehabilitative efforts, have mode the political model acceptable to a large proportion of offenders. this model is not acceptable to prison administrators, however, and the result is increasingly that administrators and clients are operating from conflicting paradigms.     

PREDICTING WHO REOFFENDS: THE NEGLECTED ROLE OF NEIGHBORHOOD CONTEXT IN RECIDIVISM STUDIES*

Prior studies of recidivism have focused almost exclusively on individual‐level characteristics of offenders and their offenses to explore the correlates of reoffending. Notably absent from these studies are measures reflecting the neighborhood contexts in which individuals live. The current research addresses this shortcoming. Using data on a sample of ex‐offenders in Multnomah County, Oregon (Portland and surrounding area) in conjunction with 2000 census data, we answer two questions. First, which individual‐level factors influence rates of recidivism? Second, to what extent does neighborhood socioeconomic status account for variation in the reoffending behavior of ex‐prisoners that is not explained by their individual‐level characteristics? We find that those who return to disadvantaged neighborhoods recidivate at a greater rate while those who return to resource rich or affluent communities recidivate at a lesser rate, controlling for individual‐level factors.     

A NEW LOOK AT THE EMPLOYMENT AND RECIDIVISM RELATIONSHIP THROUGH THE LENS OF A CRIMINAL BACKGROUND CHECK*

Criminal background checks are increasingly being incorporated into hiring decisions by employers. Although originally uncompromising—almost anyone with a criminal record could be denied employment—court rulings and policy changes have forced criminal background checks to become more nuanced. One motivation for allowing more individuals with criminal records to work is to decrease recidivism and encourage desistance. In this article, we estimate the causal impact of receiving a clearance to work on subsequent arrests for individuals with criminal records who have been provisionally hired to work in certain nonlicensed health‐care jobs in New York State (N = 6,648). We employ an instrumental variable approach based on a substantive understanding of the state‐mandated criminal background check process. We examine age‐graded effects within this group of motivated individuals and differential effects by sex in the rapidly growing health‐care industry, which is typically dominated by women. Our estimated local average treatment effect indicates a 2.2‐percentage‐point decrease in the likelihood of a subsequent arrest in 1 year and a 4.2‐percentage‐point decrease over 3 years. We find meaningful variations by sex; men are 8.4 percentage points less likely to be arrested over the 3‐year period when cleared compared with a 2.4‐percentage‐point (and nonsignificant) effect for women. Older women in particular are driving the nonsignificant results for women.     

THE VICTIM–OFFENDER OVERLAP IN CONTEXT: EXAMINING THE ROLE OF NEIGHBORHOOD STREET CULTURE*

Although numerous studies have found a strong relationship between offending and victimization risk, the etiology of this relationship is not well understood. Largely absent from this research is an explicit focus on neighborhood processes. However, theoretical work found in the subculture of violence literature implies that neighborhood street culture may help to account for the etiology of this phenomenon. Specifically, we should expect the magnitude of the victim–offender overlap to vary closely with neighborhood‐based violent conduct norms. This research uses waves 1 and 2 of the Family and Community Health Study (FACHS) to test the empirical validity of these notions. Our results show that the victim–offender overlap is not generalizeable across neighborhood contexts; in fact, it is especially strong in neighborhoods where the street culture predominates, whereas it is significantly weaker in areas where this culture is less prominent. These results indicate that neighborhood‐level cultural processes help to explain the victim–offender overlap, and they may cause this phenomenon to be context specific.     

PAROLEE RECIDIVISM IN CALIFORNIA: THE EFFECT OF NEIGHBORHOOD CONTEXT AND SOCIAL SERVICE AGENCY CHARACTERISTICS*

We studied a sample of reentering parolees in California in 2005–2006 to examine whether the social structural context of the census tract, as well as nearby tracts, along with the relative physical closeness of social service providers affects serious recidivism resulting in imprisonment. We found that a 1 standard deviation increase in the presence of nearby social service providers (within 2 miles) decreases the likelihood of recidivating 41 percent and that this protective effect was particularly strong for African American parolees. This protective effect was diminished by overtaxed services (as proxied by potential demand). We found that higher concentrated disadvantage and social disorder (as measured by bar and liquor store capacity) in the tract increases recidivism and that higher levels of disadvantage and disorder in nearby tracts increase recidivism. A 1 standard deviation increase in the concentrated disadvantage of the focal neighborhood and the surrounding neighborhoods increases the likelihood of recidivating by 26 percent. The findings suggest that the social context to which parolees return (both in their own neighborhood and in nearby neighborhoods), as well as the geographic accessibility of social service agencies, play important roles in their successful reintegration.     

CHILDHOOD AND ADOLESCENT PREDICTORS OF PHYSICAL ASSAULT: A PROSPECTIVE LONGITUDINAL STUDY*

Data gathered over the course of an 18‐year prospective longitudinal study of more than 1,000 New Zealand children was used to examine the contextual, lifestyle, and childhood risk factors associated with young people's exposure to physical assault in late adolescence. Twenty‐three percent of males compared with 14% of females reported an assault between the ages of 16 and 18 years. However, although the prevalence and nature of young people's physical assault experiences differed in gender specific ways, the concurrent and antecedent risk factors that placed males and females at risk of physical assault were similar. The major predictors of physical assault during late adolescence included childhood measures of behavioral disturbance and parental dysfunction, in addition to measures of adolescent participation in a delinquent lifestyle, such as violent offending, status offending, and the misuse of alcohol. These findings support previous research suggesting a strong link between juvenile delinquency and victimization risk, and they contribute to an understanding of the role of gender and childhood experiences in predicting later risk of physical assault.     

HIGH SCHOOL EXPERIENCE AND THE RISK OF ADULT INCARCERATION*

This study assesses the effects of high school educational experiences on the risk of incarceration for young men aged 19–36 using event history analysis and National Longitudinal Survey of Youth data. High school education serves as a defining moment in an individual's life course. Young men who enroll in secondary occupational course work significantly reduce their likelihood of incarceration both overall and net of differences in the adult labor market. High school student/teacher ratios and student composition also significantly affect an individual's risk of incarceration.     

RECIDIVISM AMONG PAROLED PROPERTY OFFENDERS RELEASED DURING A PERIOD OF PRISON REFORM*

Data from four successive yearly cohorts and one special early release cohort of parolees are used to explore the question of whether rapid statewide changes in the administration of criminal justice affected the patterns of recidivism among persons on parole for property offenses. Given the earlier broadly constructed research reported by Ekland–Olson et al. (1993), and their conclusion that variation in shifting policies would have different effects on different types of offenses, we decided to sharpen the focus of the research questions posed by concentrating on recidivism patterns among property offenders. Three alternative explanations—compositional effects, administrative discretion, and deterrence—are explored to interpret the differences found across cohorts While suggestive, these alternative explanations remain open to question given the limitations inherent in quasi-experimental research. Conclusions related to issues of prison construction policy suggest that more attention be paid to the “replacement factor,” whereby “vacancies” left by incarcerated offenders are rapidly filled by others. If future research supports the rapid replacement hypothesis, increased levels of incarceration will yield a larger, more experienced criminal “work force” and ironically a heightened collective potential for crime.