Effectiveness of Residence Restrictions in Preventing Sex Offense Recidivism

http://cad.sagepub.com/Crime & Delinquency

http://cad.sagepub.com/content/58/4/491The online version of this article can be found at:

DOI: 10.1177/0011128712449230

2012 58: 491Crime & DelinquencyMatt R. Nobles, Jill S. Levenson and Tasha J. Youstin

RecidivismEffectiveness of Residence Restrictions in Preventing Sex Offense

Published by:

http://www.sagepublications.com

can be found at:Crime & DelinquencyAdditional services and information for

http://cad.sagepub.com/cgi/alertsEmail Alerts:

http://cad.sagepub.com/subscriptionsSubscriptions:

http://www.sagepub.com/journalsReprints.navReprints:

http://www.sagepub.com/journalsPermissions.navPermissions:

http://cad.sagepub.com/content/58/4/491.refs.htmlCitations:

What is This?

- Jul 2, 2012Version of Record >>

at ATSA on August 18, 2012cad.sagepub.comDownloaded from

http://cad.sagepub.com/

http://cad.sagepub.com/content/58/4/491

http://www.sagepublications.com

http://cad.sagepub.com/cgi/alerts

http://cad.sagepub.com/subscriptions

http://www.sagepub.com/journalsReprints.nav

http://www.sagepub.com/journalsPermissions.nav

http://cad.sagepub.com/content/58/4/491.refs.html

http://cad.sagepub.com/content/58/4/491.full.pdf

http://online.sagepub.com/site/sphelp/vorhelp.xhtml


Crime & Delinquency58(4) 491 –513

© The Author(s) 2012Reprints and permission:

sagepub.com/journalsPermissions.navDOI: 10.1177/0011128712449230

http://cad.sagepub.com

449230 CAD58410.1177/0011128712449230Nobles et al.Crime & Delinquency© 2011 SAGE Publications

1Sam Houston State University, Huntsville, TX, USA2Lynn University, Boca Raton, FL, USA3Florida Atlantic University, Boca Raton, FL

Corresponding Author:Matt R. Nobles, College of Criminal Justice, Sam Houston State University, P.O. Box 2296, Huntsville, TX 77341-2296, USAEmail: [email protected]

Effectiveness of Residence Restrictions in Preventing Sex Offense Recidivism

Matt R. Nobles1, Jill S. Levenson2, and Tasha J. Youstin3

Abstract

Many municipalities have recently extended residence restrictions for sex offenders beyond the provisions of state law, although the efficacy of these measures in reducing recidivism has not been empirically established. This study used arrest histories in Jacksonville, Florida, to assess the effects of a recently expanded municipal 2,500-foot residence restriction ordinance on sex crimes and sex offense recidivism. Using a quasiexperimental design, pre- and posttest measures of recidivism were compared, and no significant differences in citywide sex crimes or recidivist sex crimes were found. In addition, time-series analysis revealed no significant differences in sex crime trends over time when compared with nonsex crimes from the same of-fender sample. After controlling for several demographic factors, individual-level multivariate results indicate that the timing of the residence restriction policy was not associated with a meaningful change in sex crime arrests or sex offender recidivism after the policy implementation date, suggesting that the residence restriction did not achieve its intended goal of reducing recidivism.

Keywords

sex offense, sex offender, residence restriction, residency, recidivism

Article



492 Crime & Delinquency 58(4)

Introduction

In 1994, the Jacob Wetterling Act was passed by the U.S. Congress, requir-ing states to develop procedures to register and track sex offenders. This policy was amended in 1996 by “Megan’s Law,” which allowed states to publicize the whereabouts of registered sex offenders (RSOs). Community notification has brought national attention to the residential locations of sex criminals, resulting in the enactment of laws restricting where convicted sex offenders can live. These policies are designed to diminish the risk posed to children by repeat sexual predators by decreasing access and opportunity.

At least 30 states and thousands of municipalities across the United States have passed laws preventing RSOs from residing within a specific distance of schools, parks, daycare centers, school bus stops, or other places commonly frequented by children (Meloy, Miller, & Curtis, 2008). These laws some-times apply only to abusers of minor children, but in many localities residen-tial restrictions are in effect for all RSOs. Buffer zones typically range from 500 to 2,500 feet between the offender’s home and designated locations, encircling the concentric area around protected places. The present study fea-tures a natural quasi experiment in which an existing state law specifying a 1,000-foot residential buffer for sexual predators was enhanced by a city ordinance to a more punitive 2,500-foot restriction.1 These laws have become increasingly popular, with municipalities progressively passing restrictive measures for sex offenders that exceed those already established under state law. Unfortunately, research has not kept pace with their implementation. Few published studies have investigated the efficacy of residential restriction laws in preventing child sexual abuse or deterring recidivism. The purpose of this study is to examine the effectiveness of more severe residence restric-tions in reducing sex crime recidivism.

Literature ReviewTheoretical Models for Residence Restrictions

The relationship between where sex offenders reside and whether they reof-fend remains unclear. However, there are theoretical reasons supporting a hypothetical relationship between where offenders are located and where crimes occur. For example, the routine activities theory (Cohen & Felson, 1979) sets forth three conditions that coexist in time and space to create situ-ations in which crimes are most likely to occur: the presence of a motivated offender, access to a potential victim, and the absence of capable guardians to



Nobles et al. 493

prevent criminal activity. Consistent with this view, Rengert, Piquero, and Jones (1999) asserted that proximity to an offender increases the likelihood of victimization, an idea known as distance decay. Other researchers suggest that there is a direct link between geographic patterns of offending and crime-related decision making, perhaps involving a rational calculus process that considers factors such as target attractiveness and spatial opportunity (Brantingham & Brantingham, 1984; Capone & Nichols, 1976). Some empir-ical evidence supports this view; for instance, rapists who target adult victims have been found to often commit sex crimes within close proximity to their homes (Minnesota Department of Corrections, 2007; Warren et al., 1998).

These theories, most involving some aspect of rational choice, imply that it is when pedophiles and sexual predators have unrestricted or unmonitored access to potential victims that sexual assault is most likely to take place. Indeed, lawmakers have argued that when RSOs reside near a school or other place where children commonly congregate, opportunities for contact with children are plentiful. Furthermore, it is believed that if sex offenders are able to easily view children playing from their homes, their sexually deviant pref-erences may be reinforced, thereby encouraging or facilitating child sexual abuse. These assertions receive some empirical support in the dynamic risk factor literature suggesting that access to children and potential victims con-tributes to risk for sexual recidivism (Hanson & Harris, 1998, 2001). No empirical data exist to date, however, to validate the supposition that the location of an offender’s home necessarily increases access to children.

Research Investigating Residence RestrictionsThe theories discussed above suggest that sex offenders might intentionally place themselves in close proximity to potential victims. Mapping technolo-gies provide opportunities for researchers to study key questions about the residential locations of sex offenders, and so far, results are mixed. In one Arkansas county, an analysis of the addresses of RSOs revealed that child abusers were more likely than rapists to live within 1,000 feet of schools, daycare centers, and parks (Walker, Golden, & VanHouten, 2001). A recent study of sex offender residences in New Jersey revealed that in Newark, however, child molesters lived significantly farther from schools than sex offenders who assaulted adult victims (Chajewski & Mercado, 2008). In Camden, New Jersey, 88% of sex offenders lived within 2,500 feet of schools, parks, daycare centers, or churches, compared with 80% of com-munity members, and sex offenders with adult victims lived significantly closer to schools and parks than sex offenders with child victims (Zgoba, Levenson, & McKee, 2009).




Although we might speculate that some child predators do choose to live near places frequented by children, an alternative explanation also exists. Some researchers contend that economic conditions, rather than deviant sex-ual preferences, are the variables most likely to affect housing decisions (Red Bird, 2009; Tewksbury & Mustaine, 2006; Tewksbury & Mustaine, 2009). Convicted felons often experience underemployment, and sex offend-ers face unique obstacles when attempting to secure and maintain jobs (Levenson & Cotter, 2005a; Levenson, D’Amora, & Hern, 2007; Tewksbury, 2005). RSOs, therefore, are often economically disadvantaged and are com-monly found in neighborhoods with lower incomes (Red Bird, 2009; Tewksbury, Mustaine, & Stengel, 2008). Less-affluent neighborhoods tend to be densely populated, and residential dwellings are therefore naturally in close proximity to schools and daycare centers (Zgoba et al., 2009). Although sex offenses occur more frequently in census tracts with larger proportions of young children, the number of schools in a neighborhood is not associated with a greater frequency of sex offenses (Tewksbury et al., 2008). Tewksbury et al. (2008) also found that a greater concentration of RSOs in a neighbor-hood was not significantly correlated with the incidence of sex crimes.

There is no established correlation between proximity to schools or child care facilities and sex offense recidivism. Zandbergen, Levenson, and Hart (2010) compared recidivists and nonrecidivists (n = 330) who lived within common buffer zones in Florida. Those who lived within 1,000, 1,500, or 2,500 feet of schools or daycare centers were no more likely to reoffend sexually than those who lived farther away. Furthermore, a meaningful cor-relation was not evident between sexual reoffending and proximity to schools. The sample was matched on relevant risk factors (prior arrests, age, marital status, and predator status), and when the distances to schools and day cares were entered along with risk factors into a logistic regression model, proximity measures were not significant predictors of recidivism (Zandbergen et al., 2010).

In Colorado, sex offense recidivists were found to reside randomly in the geographical area and did not appear to live closer to schools and daycare centers than nonrecidivists (Colorado Department of Public Safety, 2004). In Minnesota, the grounds of schools and parks were found to be uncommon venues of sexual assault, and in the few cases in which sex crimes did occur in those settings, the offenders lived far from the crime scene and drove a vehicle to the offense location (Minnesota Department of Corrections, 2003). A close analysis of 224 recidivistic sex offenses in Minnesota revealed that residence restriction laws would not have prevented any repeat sex crimes (Duwe, Donnay, & Tewksbury, 2008). Sex offenses against children were



Nobles et al. 495

most frequently perpetrated not by strangers lurking in schoolyards but by offenders who were well known to their victims, such as parents, caretakers, paramours of the mother, babysitters, or friends of the family. In less than 4% of the cases was the sex offender a neighbor of the victim. Initial contact with victims was usually established more than 1 mile from the offender’s home, and predatory assaults that occurred within a mile of the offender’s residence typically involved adult victims. Although relationships with minor victims were sometimes cultivated within 2,500 feet of the offender’s home, none took place in or near a school, daycare center, or park. The authors concluded that an offender’s social relationship with a child victim is much more likely to facilitate sexual abuse than residential proximity (Duwe et al., 2008).

The Iowa Department of Criminal and Juvenile Justice Planning studied the effect of Iowa’s 2,000-foot residence restrictions policy that went into effect in August 2005 (Blood, Watson, & Stageberg, 2008). In an effort to determine the impact the law had on sex crime rates, researchers compared the number of charges for sexual assaults of minor victims in the 12 months preceding implementation of the law with the number of charges filed within 24 months after the law went into effect. A decrease in the number of charges was not observed following the passage of the law. In fact, sex crime arrests increased steadily each year; there were 913 charges filed during the year prior to implementation, 928 filed the subsequent year, and 1,095 the follow-ing year. It is not clear whether this difference is statistically significant, but it is clear that a downward trend did not emerge. The authors concluded that Iowa’s residence law “does not seem to have led to fewer charges or convic-tions, indicating that there probably have not been fewer child victims” (Blood et al., 2008, p. 10).

Unintended Consequences of Residence RestrictionsData collected from surveys of sex offenders suggest that residence restric-tions create obstacles when attempting to reintegrate into communities. In particular, homelessness and transience are becoming more daunting prob-lems for RSOs. More than half of sex offenders surveyed in Oklahoma and Kansas said that they were required to relocate because of a housing restriction law (Tewksbury & Mustaine, 2009). In Indiana, more than one quarter of sex offenders surveyed said that they were unable to return to their homes after release from prison, 37% could not live with family members, and nearly one third reported that a landlord refused to rent to them or to renew a lease (Levenson & Hern, 2007). Sex offenders in Florida and Indiana have reported a scarcity of affordable housing and find themselves living farther




away from employment, public transportation, social services, and mental health clinics (Levenson, 2008; Levenson & Cotter, 2005b; Levenson & Hern, 2007). New Jersey sex offenders also indicated that residence restric-tions require them to live longer distances from employment, mental health services, and family support (Mercado, Alvarez, & Levenson, 2008). Larger buffer zones have been correlated with increased transience and homeless-ness and diminished employment opportunities (Levenson, 2008; Levenson & Hern, 2007).

Self-reported data clearly have limitations, but the experiences shared by sex offenders have been corroborated by independent empirical research. Specifically, there is a growing body of evidence that residence restrictions severely reduce housing options for sex offenders. In Florida, the locations of more than 137,000 residential dwellings in the Orlando area were examined, and 99% were found to be located within 2,500 feet of schools, parks, daycare centers, or school bus stops (Zandbergen & Hart, 2006). Likewise, 93% of residential territory in Newark, New Jersey, is located within 2,500 feet of a school (Chajewski & Mercado, 2009). In four major metropolitan areas of South Carolina, 91% of the 540,613 properties zoned for residential use were unavailable for rent, and of the few that were unoccupied, 45% were within 1,000 feet of schools or daycare centers (Barnes, Dukes, Tewksbury, & DeTroye, 2009). In Omaha, Nebraska, it was found that 79% of all residen-tial parcels were located within 2,000 feet of a school or daycare (Bruell, Swatt, & Sample, 2008).

Residence restrictions have been found to decrease access to affordable housing, employment, and social services. Of nearly 1 million residential parcels studied in Miami-Dade County, Florida, researchers found that under the state and local residence restrictions currently in effect for probationers and RSOs, about 4% of units were compliant with the restrictions, and only 1% had a monthly housing cost of US$1,250 or less (Zandbergen & Hart, 2009). In Nebraska, the average home value (US$77,028) within a buffer zone of 2,000 feet was significantly less than those outside the buffer zone (US$112,989), and the average distance to the city center was significantly greater when living outside the buffer (2.1 miles compared with 1.6 miles within the restricted zone; Bruell et al., 2008). In Ohio, compliant addresses were more likely to be located in higher income census tracts (Red Bird, 2009). In a study of more than 1,000 sex offenders in Kentucky, many lived in rural areas and traveled an average of 25 min by car to the nearest mental health treatment provider, with 10% having to travel more than an hour each way (Sloas, 2008).



Nobles et al. 497

Aside from the negative effects of residence restrictions for offenders, research has also revealed problems faced by law enforcement officers chal-lenged with implementing the restrictions. In a study in Seminole County, Florida, Tewksbury and Mustaine (2006) found that only one half of the offenders in their sample were in compliance with local residence restric-tions. Youstin and Nobles (2009) similarly found evidence of implementa-tion problems in Alachua County, Florida. Results revealed that at least 40% of the sex offender population who relocated after the imposition of a manda-tory residence restriction did so in a restricted area (Youstin & Nobles, 2009). Youstin and Nobles also found evidence that offenders may become more geographically dispersed over time, which may present hardships for police and probation officers tasked with monitoring offenders. Reports on this topic describe growing problems with sex offender tracking because offend-ers who cannot locate compliant addresses might become transient or abscond from registration (Human Rights Watch, 2007). One Iowa sheriff reported that because of difficulties in finding accessible and adequate housing, the Sheriff’s Office went from knowing where about 90% of offenders resided to knowing where only 50% to 55% of offenders are located (Stachura, 2006). In California, after a 2,000-foot state law was implemented, the number of sex offenders registering as transient rose by 60% within a year (California Sex Offender Management Board, 2008). Research has shown that transience and housing instability are consistently and strongly correlated with recidi-vism, parole violations, and absconding in samples of criminal offenders, generally, and sex offenders, specifically (Meredith, Speir, & Johnson, 2007; Schulenberg, 2007; Williams, McShane, & Dolny, 2000; Willis & Grace, 2008). Thus, these factors may create not only administrative problems for law enforcement but also an environment of greater risk for potential victims.

In practice, the goal of residence restrictions in deterring repetitive sexual predation must be balanced with the potential to exacerbate factors known to complicate effective monitoring and increase risk for criminal persistence. Residence restrictions have become a popular mechanism for attempting to hinder sex offenders’ access to children, but their underlying premise—that keeping sex offenders far from schools and other child-friendly locations will prevent sexual abuse—has yet to be empirically established. Safeguarding children from sexual predators is an important goal. Consensus is lacking, however, regarding the strategies most likely to accomplish this objective. Examining the role that residence laws play in the desistance of criminal sexual activity will assist policy makers to determine which policies work best to improve community safety.




Present Study

The aim of this study was to investigate the efficacy of sex offender resi-dence restrictions in preventing sexual recidivism. Specifically, we sought to examine the trends of sex crimes and sexual rearrests in Jacksonville, Florida, before and after the implementation of the city’s 2,500-foot residen-tial restriction ordinance, which represented an expansion of the statewide 1,000-foot boundary. We hypothesized that effectiveness of the policy would be demonstrated by a significant decline in sex crime rates and sexual recidivism rates during the postimplementation period. This research was considered to be important to clarify the role that residential restrictions might play in deterring sexual recidivism. Sex crime policies informed by scientific data are apt to be more successful in protecting communities.

Florida was considered to be an ideal state in which to conduct this research. Residence restrictions have been enforced in Florida since 1995 in the form of a statewide law requiring sex offenders on probation to reside more than 1,000 feet from a school, park, playground, daycare center, or other place where children congregate (Special Conditions of Sex Offender Probation, 1997). In 2004, school bus stops were added for offenders on con-ditional release from prison. In 2005, Miami Beach pioneered the enactment of municipal ordinances, modeled after zoning laws prevented adult enter-tainment establishments from operating near schools. According to the state’s department of corrections, more than 150 local ordinances have been adopted in Florida (typically 2,500-foot zones passed by county and city commis-sions) in addition to the state’s 1,000-foot law, which now applies to all RSOs. Thus, the research question at issue here is whether the imposition of an extended residence restriction beyond the state-mandated 1,000-foot law resulted in any observable difference in sex offense recidivism.

The current study used data from Jacksonville, Florida. Jacksonville’s city ordinance was enacted on July 1, 2005, and prohibited sexual predators from residing within 2,500 feet of any school, public library, daycare center, park, playground or other place where children regularly congregate. In Florida, predator is defined as a repeat offender convicted of more than one sex crime or an offender convicted of one particularly egregious sex crime, such as abuse of a minor by someone who is not the child’s parent or guardian.2 The ordinance did contain a clause that allowed predators to remain in a residence established before the effective date of the law. This type of “grandfather clause” is common in municipal ordinances in Florida. Jacksonville’s law was deemed unconstitutional on October 11, 2007, on the grounds that it was preempted by the state’s 1,000-foot law and that it interfered with “the com-prehensive strategy adopted by the Legislature to address the risk which



Nobles et al. 499

sexual offenders and predators pose toward all of our communities” (State of Florida v. Schmidt et al., 2007).

Data and MethodData used in this analysis were provided by the Jacksonville Sheriff’s Office (JSO) at the request of the authors. Specifically, JSO provided arrest data for all individuals arrested between March 1, 2003, and December 31, 2007, for sex-related offenses (rape, sexual battery, lewd or lascivious behavior, or sexual assault). This time frame was chosen to examine offenses before and after the implementation of Jacksonville’s residence restriction ordinance on July 1, 2005. This date is defined as the “intervention.” Arrests that occurred between the noted dates were considered the index offenses; thus, only indi-viduals arrested for sex offenses between those dates were included in our sample. But, full criminal histories were provided that included prior and subsequent (recidivistic) sexual and nonsexual arrests for the offenders who matched our initial query; thus, the cases span more than 30 years, from March 1976 to March 2008.

The original data were truncated by date to provide equivalent time before and after the residence restriction policy was implemented; to make maxi-mum use of the available data, we selected all offenses ranging from the intervention to the end of the data set, and then selected a symmetrical range from the preintervention data—specifically −978 days to +978 days sur-rounding the Jacksonville law’s implementation date, July 1, 2005. This resulted in a final working sample size of 8,597 unique cases accountable to 2,630 individual offenders. Of these cases, n = 3,567 (41.5%) were classified as felonies under state law. However, a minority of unique cases in the data, n = 2,557 (29.8%), featured one or more sex offenses as previously defined. Even fewer cases, n = 670 (7.8%), met the criteria for a recidivist sex offense. Descriptive statistics are provided in Table 1.

In response to concerns about potential confounds, the authors of this study also truncated the data a second time and reestimated each set of results to ensure consistency across conditions. The nature of this potential confound dealt with a court ruling3 stating that Jacksonville’s residence restriction ordi-nance was unconstitutional; as a result, some arrests in our original data occurred after the law was no longer in effect, potentially affecting the valid-ity of our measurements. Therefore, the court ruling, issued October 11, 2007, required the range of data used in the analysis to be shortened from ±978 days surrounding the intervention date to ±832 days. Truncating in this way resulted in a reduction of 761 cases. However, although the data were altered slightly according to the time frame for observations, the substantive




conclusions from each of our analyses were upheld. As it is plausible that the hypothetical effects of Jacksonville’s policy would persist beyond the court ruling because offender relocation is costly and time-consuming, we report parameter estimates from the original data set only.

Sex offender arrest histories were analyzed in several ways to thoroughly evaluate available evidence. First, bivariate analyses compared sex offense counts in the preintervention period with counts in the postintervention period. Significance testing was accomplished using Pearson’s χ2. Next, the effects of the Jacksonville ordinance were hypothesized to be observable in a time-series sequence plot, which displays monthly counts of sex offenses in the preintervention and postintervention periods, where a significant effect would be represented by a change in the level (a “step” function) or the rate (an increase or decrease in the series trend line) of the dependent variable. The time-series data were analyzed to illustrate any observed changes through autoregressive-integrated moving average (ARIMA) regression modeling. ARIMA is designed to correct confounding factors, such as serial autocorre-lation, that violate traditional regression assumptions, producing biased parameters and thus impaired interpretations (Cook & Campbell, 1979). In addition, the ARIMA models featured here were estimated with robust standard errors. This step ensures further resistance to potential data problems, such as outliers in the series, that could result in inaccurate conclusions.

Two separate binary logistic regression models were estimated. The first model in the analysis used sex offense (0 = no, 1 = yes) as the dependent vari-able.4 Independent variables included race (0 = White, 1 = non-White), sex (0 = female, 1 = male), age5 (continuous), felony charge (0 = nonfelony, 1 = felony, where the “felony” category included one or more coincident charges at the felony level under Florida law; the “nonfelony” category included all other offenses), and whether the arrest occurred after the policy change (July 1, 2005; 0 = no, 1 = yes). The objective here was to determine whether the more

Table 1. Descriptive Statistics of All Arrests

n M SD Min Max

Sex arrest (0 = no, 1 = yes) 8,597 0.292 0.455 0 1Recidivist sex crime arrest (0 = no, 1 = yes) 8,592 0.078 0.268 0 1Race (0 = White, 1 = minority) 8,569 0.593 0.491 0 1Sex (0 = female, 1 = male) 8,597 0.924 0.266 0 1Age 8,597 34.146 12.056 12 84Felony (0 = no, 1 = yes) 8,597 0.415 0.493 0 1After policy date (0 = no, 1 = yes) 8,597 0.487 0.500 0 1



Nobles et al. 501

restrictive residence policy was associated with a reduction in the overall number of arrests for sex offenses while holding offender demographic and crime-type factors constant. The second model used sexual recidivism (0 = no, 1 = yes; where “yes” indicated that the arrest was for a subsequent sex offense by the same offender) as the dependent variable, with the same inde-pendent/control variables included. The objective was to determine whether the more restrictive residence restriction policy was related to a decline in the overall number of rearrests for sex offenses, a direct measure of recidivism, while holding demographic and crime-type factors constant; this model is the most appropriate for a direct test of the efficacy of residence restrictions on reducing sexual recidivism, presumably the primary goal of Jacksonville’s policy. In both regression models, because of the large number of cases in the analysis (N = 8,597), a conservative threshold for statistical significance (p < .001) was used, and standardized coefficients are reported to aid in the inter-pretation of substantive meaning (Allison, 1999).

ResultsTable 2 displays bivariate analyses that compared offense counts in the pre-intervention period with those in the postintervention period. Results indicated that there were significant differences found in sex crime arrests before and after the passage of the Jacksonville ordinance (χ2 = 24.306, p < .001) but not in the expected direction. Sex crime arrests increased in the postpolicy period. In addition, recidivistic sex crimes increased in frequency after the ordinance went into effect (χ2 = 17.752, p < .001).

Analysis of the sequence plot for monthly sex crime charges and corre-sponding ARIMA regression model indicated that, although there appears to have been a substantial spike in the number of sex charges prior to the inter-vention (see Figure 1), the trends for sex crimes and nonsex crimes appear relatively flat pre- and post intervention. A correctly specified ARIMA(0,0,2) model showed a nonsignificant relationship (b = −23.504, p < .439) between the timing of the intervention and the monthly counts of all sex charges in

Table 2. Group Comparisons (N = 8,592)

% of total prepolicy arrests

% of total postpolicy arrests

Pearson χ2 Significance

Sex crime arrests 13.76 15.45 24.306 .000Sexual arrest

recidivism3.39 4.41 17.752 .000




Jacksonville (see Table 3). Postestimation diagnostics indicated that the series did not require correction for either residual serial autocorrelation or stationarity. With respect to autocorrelation, the Durbin–Watson test (d = 1.866, p > .05) indicated no serial autocorrelation. With respect to stationar-ity, the augmented Dickey–Fuller test (d = −7.631, p < .00) supported the rejection of the null hypothesis of a unit root, thus providing evidence for series stationarity. Correction of the moving average component was estab-lished by the evaluation of residual ACF (autocorrelation function) and PACF (partial autocorrelation function) plots, which indicated a cut point at Lag 2. Postestimation diagnostics established model fit. After correction with the MA (moving average) term at Lag 2, the ACF and PACF plots showed that no lags exceeded critical values and that p values for Q-statistics at all lags were nonsignificant.

In an effort to confirm the stability of this finding, alternative ARIMA models were estimated with and without MA correction at Lag 1, and sub-stantive conclusions with respect to the intervention effect were identical in both cases. In addition, further examination of the preintervention sex crime

Figure 1. Time-series sequence plots of criminal charges for sex offenders in Jacksonville by month (November 2002 to February 2008)



Nobles et al. 503

trend revealed a monthly sum of 537 charges in May 2005, due to a single offender who was charged with 260 individual counts of possessing child pornography. In addition to the use of robust standard error estimates, this outlier was manually corrected in the series, and the models were reesti-mated, revealing no difference in the substantive conclusions with respect to the timing of the intervention when the outlier was removed (b = −2.441, p < .901). Finally, in an effort to provide a comparison with the time series for sex crime charges, a separate ARIMA(1,0,1) model was estimated for the nonsex crime charges observed within the same offender sample. Results indicated no significant change in the monthly counts for nonsex crime charges associated with the intervention (b = −13.442, p < .758).

Results from the logistic regression model using sex crime arrests as the dependent variable are shown in Table 4. The model explained about 22% of the variance in all sex crime arrests. Race was inversely correlated with sex crime arrest, indicating that non-Whites had slightly lower odds of being arrested for sex crimes. Males and older offenders in the sample had higher odds of being arrested for a sex offense. Felony offenders had greater than 11 times the odds of being arrested for a sex offense. The postpolicy period was associated with a slight increase in the odds of sex crime arrest, but the effect was nonsignificant and very small compared with other factors, particularly the sex and age of the offender, and the presence of one or more felony charges, as evidenced by a comparison of standardized coefficients and odds ratios for these factors.

Table 3. ARIMA Regression Models for Monthly Sex Crime and Nonsex Crime Charges

B Semirobust SE Z Significance

Sex crimes—ARIMA(0,0,2) Intervention date Lag 0 −23.504 30.381 −0.770 .439 MA Lag 2 0.504* 0.256 1.97 .049 Constant 106.873** 24.098 4.43 .000 Log pseudolikelihood = −370.5725Other crimes—ARIMA(1,0,1) Intervention date Lag 0 13.442 43.626 0.31 .758 AR Lag 1 0.971** 0.037 26.59 .000 MA Lag 1 −0.581** 0.113 −5.15 .000 Constant 178.076** 51.771 3.44 .001 Log pseudolikelihood = −344.8263

Note: ARIMA = autoregressive-integrated moving average.*p < .05. **p < .01.




Results from the logistic regression analysis using sex crime recidivism as the dependent variable are reported in Table 5. The results indicate that the model explained only about 17% of the variance in sex crime recidivism. Being male and older increased the odds of being arrested for sexual reoff-ending. Felony offenders had nearly 13.5 times the odds of being arrested for a recidivistic sex offense. The postpolicy period was associated with a slight increase in the odds of a recidivism arrest; however, as before, the effect was very small compared with other factors, particularly the coincidence of a felony charge, as evidenced by a comparison of standardized coefficients and odds ratios for these factors. Notably, the postpolicy period was not associ-ated with a statistically significant change in the odds of sexual recidivism.

DiscussionThis study used multiple types of analyses to determine whether Jacksonville’s residential restriction law was successful in achieving its goal of community

Table 4. Logistic Regression Models for All Sex Crime Arrests

b SE B Exp(B)

Race −0.786*** 0.058 −0.213 0.456Sex 0.519*** 0.120 0.076 1.680Age 0.021*** 0.002 0.141 1.022Felony 2.432*** 0.061 0.394 11.385After policy 0.143 0.055 0.022 1.154Constant −3.004*** 0.154 — —

Note: log likelihood = −4,058.0093; pseudo-R2 = .2156.***p < .001.

Table 5. Logistic Regression Models for Recidivist Sex Crime Arrests

b SE B Exp(B)

Race −0.002 0.089 −0.001 0.998Sex 0.917*** 0.271 0.134 2.501Age 0.029*** 0.003 0.191 1.029Felony 2.602*** 0.124 0.707 13.490After policy 0.235 0.085 0.065 1.264Constant −6.195*** 0.322 — —

Note: log likelihood = −1,952.7182; pseudo-R2 = .1665.***p < .001.



Nobles et al. 505

protection from sexual abuse. We hypothesized that if such a law were effec-tive, the data would demonstrate a significant decline in sex crime arrests and sexual recidivism after the implementation of the policy. None of the analy-ses demonstrated a significant effect, even when controlling for several individual-level factors and for aggregate-level variation in the sex crime arrest time series. Sex crimes in general, as well as sexual reoffenses, increased in Jacksonville rather than decreased in the postpolicy time period. It is possible that the slight increase in arrests can be attributed to enhanced attention to sex offenders after the policy, leading to increased detection and reporting of sex crimes. In other words, the increase might represent changes in detection and reporting trends rather than changes in offending behavior.

When examining effect sizes for the postpolicy time period, the results indicated that any potential protective benefit from the more restrictive 2,500-foot ordinance is extremely small. One limitation of inferential analyses is that when a significant effect is detected, it is sometimes difficult to ascertain causality. At the same time, explanations for the absence of an effect can be equally difficult to determine or defend. Nonetheless, there are some possible explanations for failing to detect a significant relationship between Jacksonville’s implementation of the residential restriction policy and sexual recidivism. It might be the case that these types of laws are simply based on faulty logic. As there is no extant evidence that sex offenders’ residential proximity to schools, parks, or daycare centers plays a role in recidivism (Duwe et al., 2008; Zandbergen et al., 2010), our results are perhaps unsur-prising. Another possible explanation for the absence of an effect is that the ordinance targeted only sexual predators rather than all sex offenders; in other words, the ordinance restricted only a small proportion of sexual perpe-trators. However, by targeting the “worst of the worst,” the ordinance was apparently designed to deter those most likely to reoffend—a smaller group of predatory offenders who would presumably have the greatest number of victims. Alternatively, the lack of a significant effect might be attributed to the short time the ordinance was in place—only 2 years 3 months. However, using monthly offense counts before and after implementation, the time period provides a large enough number of observations to detect an effect using time-series analysis (Cook & Campbell, 1979). Finally, it is possible that low base rates of sex offense recidivism make it unlikely that policy changes will appreciably change recidivism outcomes. Our results found a sexual recidivism rate of just below 8% overall in this sample, which is con-sistent with low rates in other published reports, such as 5.3% over 3 years (Bureau of Justice Statistics, 2003) and 13% to 14% over 4 to 6 years (Hanson & Bussiere, 1998; Hanson & Morton-Bourgon, 2005).




Careful comparison of the standardized coefficients and odds ratios in our models suggested that other factors besides the residential restriction law, in particular the presence of coexisting charges for one or more felonies, tended to account for a large proportion of explained variance in the dependent vari-ables. The results from both models are consistent with established research that prior criminal behavior, especially violent behavior, is a predictor of sexual recidivism (Hanson & Bussiere, 1998). Unfortunately, a dearth of theoretically meaningful risk measures available for this offender sample precludes additional tests to determine which hypothesized processes (e.g., rational choice, routine activities, perceptual deterrence, and so on), if any, were observed to influence offender behavior. Whatever the theoretical ratio-nale, it appears that the more restrictive residence ordinance was not func-tional in systematically changing outcomes at the individual or aggregate levels.

This study was conceived as a generalized empirical test of the current trend toward harsher sex offender residence restrictions throughout the United States at the state and local levels. An important justification often used for these restrictions is that they will prevent sex crimes. Taken broadly, the results presented herein suggest that lawmakers should view residence restrictions neither as an effective method of preventing sexual victimization perpetrated by known sex offenders nor as a general deterrent for sex crimes. These analyses convincingly and consistently fail to demonstrate a meaning-ful effect of enhanced residence restrictions on recidivistic sex crimes, which should be the single most important consideration from an evidence-based policy standpoint. The results should also be considered in conjunction with the known unintended consequences of residence restrictions for sexual offenders, such as housing instability, increased environmental and psychosocial stressors, and relocation farther from employment, family, and treatment facilities (Levenson & Cotter, 2005b; Mercado et al., 2008; Tewksbury & Levenson, 2007). Given that these are precisely the types of factors found to contribute to increased risk of recidivism for criminal and sexual offenders (Meredith et al., 2007; Schulenberg, 2007; Willis & Grace, 2008), the net effect of residence restrictions might be less than helpful. In the absence of empirical evidence supporting the premise that these laws protect communi-ties, there appears to be no compelling rationale for continuing to perpetuate a policy that, at best, promotes a false sense of security and, at worst, may increase risk.

As always, there are limitations that must be acknowledged in the present study. Although a large number of cases was used for the analyses, the data included only reported crimes resulting in an arrest. Arrests always underes-timate actual crime rates, as many offenses go undetected or unreported.



Nobles et al. 507

However, because underreporting is generally stable across time, the analy-ses should serve as a credible measure of any relative changes associated with the pre- and postpolicy periods. The types of offenses targeted by resi-dence restrictions (predatory offenses against children by unrelated perpetra-tors) are among the most likely sex crimes to be reported to authorities. In addition, these analyses were limited to offenders from only one municipal-ity, so caution must be applied when generalizing the results to other jurisdictions.

The use of a single site presents a potential issue in generalizing. However, the burden of proof to establish validity is considerably less in light of the null finding. In other words, had we discovered a significant intervention effect, there would be a necessity to determine whether that effect was repre-sentative of a larger, perhaps statewide trend, or whether it was accountable solely to the Jacksonville ordinance. As we found no significant effect from the intervention, we would not expect effects elsewhere. It is possible that Jacksonville experienced no effect in terms of the rate of sex crimes while the rest of the state or nation increased, thus the observed null finding would indicate relative success compared to other levels of aggregation. However, that conclusion is inconsistent with the analyses we present and with results from several other published studies, which also report no effect from resi-dential restrictions when using data from Florida (see Zandbergen et al, 2010). Finally, since we lack appropriate data to test that hypothesis at the state level, we consider this issue to be one best left for future research. Nevertheless, we attempted to capture some type of comparison trend via the monthly charges for nonsex crimes among this same offender sample. Recall that the majority of crimes in our criminal histories data set, more than 70%, involved nonsex crimes. Thus, if crimes were decreasing in general during the period under study for whatever reason, we might expect a significant decline in the nonsex crime charges, suggesting that any observed effect in the sex crime series would be accountable to a general decline rather than a specific intervention. Replication of this type of research in other jurisdic-tions will increase confidence in the findings. Despite these limitations, there are no peculiarities in the available data to suggest that offenders in the cur-rent study area were qualitatively different from offenders in other areas or that Jacksonville was unique in some other way. As such, it is expected that these results can provide a reasonable basis for estimation of the effective-ness, or lack thereof, of residence restrictions in other jurisdictions.

ConclusionAlthough this study fills a gap in the current state of knowledge on residence restrictions, there are still many pertinent questions left unanswered.




Longitudinal research should attempt to evaluate changes in rates of recidi-vistic sex crimes since the inception of residence restrictions; this type of “natural experiment” would be advantageous because a design incorporating random assignment to the residence restriction condition seems unlikely. Moreover, as studies have shown that implementation problems may plague residence restrictions, researchers should try to determine the extent to which implementation variation plays a role in overall effectiveness. For example, researchers might investigate aggregate or individual-level recidivism risk as a function of resources dedicated to monitoring and maintenance. Finally, one of the core assumptions of residence restrictions remains unanswered to date: Do these policies actively and directly deter motivated offenders? Future research should continue to explore whether first-time sex offenders discriminately target victims in close proximity to their own residences and whether recidivist sex offenders modify their target selection behavior because of residence restrictions.

From a review of these findings as well as extant knowledge, it seems as though residence restrictions were implemented throughout the United States with haste, greatly outpacing empirical inquiry and scientific data. Despite growing evidence questioning the utility of these laws, they remain popular and are unlikely to fall out of favor with lawmakers and with the general public in the near future. But, through the continued efforts of researchers concerned with identifying empirical relationships between sex offenders, risk factors, and victimization patterns, it may eventually be possible to pro-mote policies that will prevent future sexual victimization in a consistent and reliable way.

Acknowledgments

The authors gratefully acknowledge feedback from Matt White, the Jacksonville Sheriff’s Office, and comments from anonymous reviewers.

Authors’ Note

A previous version of this article was presented at the American Society of Criminology annual meeting in St. Louis, Missouri, in November 2008.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publica-tion of this article.



Nobles et al. 509

Notes

1. Although technically this study attempts to capture effects from supplementing an existing law rather than from creating a new law, the principles behind the poli-cies may operate similarly. For example, general and specific deterrence, rational choice, and routine activities could all be theoretically applicable considering the nature of the sex crimes and the offender population to which these laws pertain. Moreover, the literature in criminal justice has broadly addressed efforts to toughen laws in accordance with crime trends, political interest, and other factors. One com-mon example is the zero tolerance approach (see Bowling, 1999; Greene, 1999).

2. Florida’s sexual predator statutes have received numerous modifications in the past decade. Several additional provisions now qualify offenders for predator sta-tus, including a history of civil commitment and relocation to Florida after being on another state’s sexual predator registry. Although the present study was conceived as a generalization rather than a test of the policy efficacy with a specific subpopulation, an effort was nonetheless made to address severity of sexual crimes by incorporating an additional felony variable into the individual-level multivariate analyses. In the current offender sample, the felony variable was strongly associated (χ2 = 2290.4, p < .000) with sexual predator status corresponding to one or more qualifying charges listed in Florida statutes as of 2009. Data on prior civil commitments or registration on sexual predator registries outside Florida were unavailable for this sample.

3. The ruling, issued by the Fourth Judicial Circuit in Duval County, held that the 2,500-foot ordinance was preempted by state legislation and was unconstitutional on the basis of violating substantive due process. Our understanding is that the decision did not affect municipalities outside of Duval County. In early 2010, Florida Senate Bill 1284 introduced legislation that would have, among other actions, preempted all local residence restrictions and permitted municipalities to extend existing state restrictions up to 2,500 feet. The bill was withdrawn by its sponsor before voting occurred.

4. Arguably, the Jacksonville ordinance would not necessarily expect to achieve a reduction in overall sex offenses because it applies specifically to the subpopula-tion of sexual predators as defined under Florida statutes. Thus, the policy could be conceptualized strictly as an antirecidivism measure. However, to the extent that the policy affects offender decision making and behavior either directly (e.g., through some type of deterrence) or indirectly (e.g., by increasing community awareness, reporting, and so on), certain effects may be observable for nonre-cidivistic sex crimes. In addition, this is a key selling point for the policies—The public believes that they make the community safer overall.

5. A small number of cases involved offenders who were minors (n = 19, 0.22%). In response to a reviewer comment, we removed these cases from the analysis and reestimated the multivariate regression models, finding no substantive differences. Thus, we report parameter estimates from the full sample.




References

Allison, P. D. (1999). Fixed Effects Regression Models. Thousand Oaks: SAGE.Barnes, J. C., Dukes, T., Tewksbury, R., & DeTroye, T. (2009). Predicting the impact

of a statewide residence restriction law on South Carolina sex offenders. Criminal Justice Policy Review, 20, 21-43.

Blood, P., Watson, L., & Stageberg, P. (2008). State legislation monitoring report. Des Moines, IA: Criminal and Juvenile Justice Planning.

Bowling, B. (1999). The rise and fall of New York murder: Zero tolerance or crack’s decline? British Journal of Criminology, 39, 531-554.

Brantingham, P. L., & Brantingham, P. J. (1984). Patterns in crime. New York, NY: Macmillan.

Bruell, C., Swatt, M., & Sample, L. (2008). Potential consequences of sex offender residency restriction laws: Housing availability and offender displacement. Paper presented at the American Society of Criminology, St. Louis, MO.

Bureau of Justice Statistics. (2003). Recidivism of sex offenders released from prison in 1994 (Rep. No. NCJ 198281). Washington, DC: U.S. Department of Justice.

California Sex Offender Management Board. (2008). Homelessness among registered sex offender in California: The numbers, the risks, and the response. Sacramento, CA: Author.

Capone, D. L., & Nichols, W. W. (1976). Urban structure and criminal mobility. American Behavioral Scientist, 20, 1999-1213.

Chajewski, M., & Mercado, C. C. (2008). An analysis of sex offender residency restrictions in Newark, New Jersey. Sex Offender Law Report, 9, 1-6.

Chajewski, M., & Mercado, C. C. (2009). An evaluation of sex offender residency restriction functioning in town, county, and city-wide jurisdictions. Criminal Jus-tice Policy Review, 20, 44-61.

Cohen, L. E., & Felson, M. (1979). Social change and crime rate trends: A routine activity approach. American Sociological Review, 44, 588-608.

Colorado Department of Public Safety. (2004). Report on safety issues raised by liv-ing arrangements for and location of sex offenders in the community. Denver, CO: Sex Offender Management Board.

Cook, T. D., & Campbell, D. T. (1979). Quasi-experimentation: Design & analysis issues for field settings. Boston, MA: Houghton Mifflin.

Duwe, G., Donnay, W., & Tewksbury, R. (2008). Does residential proximity matter? A geographic analysis of sex offense recidivism. Criminal Justice and Behavior, 35, 484-504.

Greene, J. A. (1999). Zero tolerance: A case study of police policies and practices in New York city. Crime & Delinquency, 45, 171-187.

Hanson, R. K., & Bussiere, M. T. (1998). Predicting relapse: A meta-analysis of sex-ual offender recidivism studies. Journal of Consulting and Clinical Psychology, 66, 348-362.



Nobles et al. 511

Hanson, R. K., & Harris, A. J. R. (1998). Dynamic predictors of sexual recidivism (Report No. 1998-01). Ottawa, Ontario, Canada: Department of the Solicitor Gen-eral of Canada.

Hanson, R. K., & Harris, A. J. R. (2001). A structured approach to evaluating change among sexual offenders. Sexual Abuse: A Journal of Research and Treatment, 13, 105-122.

Hanson, R. K., & Morton-Bourgon, K. (2005). The characteristics of persistent sexual offenders: A meta-analysis of recidivism studies. Journal of Consulting and Clin-ical Psychology, 73, 1154-1163.

Human Rights Watch. (2007). No easy answers: Sex offender laws in the United States. Retrieved from http://www.hrw.org/reports/2007/us0907/

Levenson, J. S. (2008). Collateral consequences of sex offender residence restrictions. Criminal Justice Studies, 21, 153-166.

Levenson, J. S., & Cotter, L. P. (2005a). The effect of Megan’s Law on sex offender reintegration. Journal of Contemporary Criminal Justice, 21, 49-66.

Levenson, J. S., & Cotter, L. P. (2005b). The impact of sex offender residence restric-tions: 1,000 feet from danger or one step from absurd? International Journal of Offender Therapy and Comparative Criminology, 49, 168-178.

Levenson, J. S., D’Amora, D. A., & Hern, A. (2007). Megan’s Law and its impact on community re-entry for sex offenders. Behavioral Sciences & the Law, 25, 587-602.

Levenson, J. S., & Hern, A. (2007). Sex offender residence restrictions: Unintended consequences and community re-entry. Justice Research and Policy, 9(1), 59-73.

Meloy, M. L., Miller, S. L., & Curtis, K. M. (2008). Making sense out of nonsense: The deconstruction of state-level sex offender residence restrictions. American Journal of Criminal Justice, 33, 209-222.

Mercado, C. C., Alvarez, S., & Levenson, J. S. (2008). The impact of specialized sex offender legislation on community re-entry. Sexual Abuse: A Journal of Research and Treatment, 20, 188-205.

Meredith, T., Speir, J., & Johnson, S. (2007). Developing and implementing auto-mated risk assessments in parole. Justice Research and Policy, 9(1), 1-21.

Minnesota Department of Corrections. (2003). Level three sex offenders residential placement issues. St. Paul, MN: Author.

Minnesota Department of Corrections. (2007). Residential proximity and sex offense recidivism in Minnesota. St. Paul, MN: Author.

Red Bird, B. (2009). Assessing housing availability under Ohio’s sex offender resi-dency restrictions. Columbus: Ohio State University.

Rengert, G. F., Piquero, A. R., & Jones, P. R. (1999). Distance decay reexamined. Criminology, 37, 427-446.




Schulenberg, J. L. (2007). Predicting noncompliant behavior: Disparities in the social locations of male and female probationers. Justice Research and Policy, 9(1), 25-57.

Sloas, L. (2008). Assessing treatment accessibility for adult sex offenders on proba-tion and parole in Kentucky. Paper presented at the American Society of Crimi-nology, St. Louis, MO.

Special Conditions of Sex Offender Probation, Florida Statute 948.03 (1997).Stachura, S. (2006). The consequences of zoning sex offenders [Electronic ver-

sion]. Retrieved from http://minnesota.publicradio.org/display/web/2006/04/21/sexoffconsequences/

State of Florida v. Schmidt et al. Fourth Judicial Circuit, Duval County Florida (2007).Tewksbury, R. (2005). Collateral consequences of sex offender registration. Journal

of Contemporary Criminal Justice, 21, 67-82.Tewksbury, R., & Levenson, J. S. (2007). When evidence is ignored: Residential

restrictions for sex offenders. Corrections Today, 69(6), 54-57.Tewksbury, R., & Mustaine, E. E. (2006). Where to find sex offenders: An examina-

tion of residential locations and neighborhood conditions. Criminal Justice Stud-ies, 19, 61-75.

Tewksbury, R., & Mustaine, E. E. (2009). Stress and collateral consequences for reg-istered sex offenders. Journal of Public Management & Social Policy, 15, 215-239.

Tewksbury, R., Mustaine, E. E., & Stengel, K. M. (2008). Examining rates of sexual offenses from a routine activities perspective. Victims & Offenders, 3, 75-85.

Walker, J. T., Golden, J. W., & VanHouten, A. C. (2001). The geographic link between sex offenders and potential victims: A routine activities approach. Jus-tice Research and Policy, 3(2), 15-33.

Warren, J., Reboussin, R., Hazelwood, R., Cummings, A., Gibbs, N., & Trumbetta, S. (1998). Crime scene and distance correlates of serial rape. Journal of Quantitative Criminology, 14, 35-59.

Williams, F. P., McShane, M. D., & Dolny, M. H. (2000). Predicting parole abscond-ers. The Prison Journal, 80, 24-38.

Willis, G. M., & Grace, R. C. (2008). The quality of community reintegration plan-ning for child molesters: Effects on sexual recidivism. Sexual Abuse: A Journal of Research and Treatment, 20, 218-240.

Youstin, T. J., & Nobles, M. R. (2009). Residency restrictions: A geospatial analysis of sex offender movement over time. Crime Mapping: A Journal of Research and Practice, 1(1), 55-76.

Zandbergen, P. A., & Hart, T. (2009). Availability and spatial distribution of afford-able housing in Miami-Dade County and implications of residency restriction zones for registered sex offenders. Retrieved from http://www.aclufl.org/pdfs/SORRStudy.pdf



Nobles et al. 513

Zandbergen, P. A., & Hart, T. C. (2006). Reducing housing options for convicted sex offenders: Investigating the impact of residency restriction laws using GIS. Justice Research and Policy, 8(2), 1-24.

Zandbergen, P. A., Levenson, J. S., & Hart, T. (2010). Residential proximity to schools and daycares: An empirical analysis of sex offense recidivism. Criminal Justice and Behavior, 37, 482-502.

Zgoba, K., Levenson, J. S., & McKee, T. (2009). Examining the impact of sex offender residence restrictions on housing availability. Criminal Justice Policy Review, 20, 91-110.

Bios

Matt R. Nobles is an Assistant Professor in the College of Criminal Justice at Sam Houston State University. His research interests include violent and interpersonal crimes, gun policy, GIS and spatial econometrics, and quantitative methods. His recent work has appeared in Justice Quarterly, Crime & Delinquency, Aggression and Violent Behavior, Journal of Quantitative Criminology, Journal of Criminal Justice, Journal of Interpersonal Violence, Criminal Justice and Behavior, and is forthcoming in the American Journal of Public Health.

Jill S. Levenson is an Associate Professor of Human Services at Lynn University in Boca Raton, FL., as well as a licensed clinical social worker with over 20 years of experience counseling sexual abuse victims, survivors, perpetrators, and non-offending parents. Dr. Levenson’s numerous publications include studies investigat-ing the impact and effectiveness of social policies designed to prevent sexual violence. She is a co-investigator on two grants funded by the National Institute of Justice investigating the effectiveness of sex offender registration and notification in reducing recidivism. She has published over fifty articles and book chapters and has co-authored three books on the treatment of sex offenders and their families.

Tasha J. Youstin received her Ph.D. from John Jay College/Graduate Center-City University of New York, and is currently an Assistant Professor at Florida Atlantic University. Her general research interests include theory testing, environmental criminology, and sexual offending and offender policies. Her research has been pub-lished in Criminal Justice and Behavior, Justice Quarterly, and Crime & Delinquency.



Effectiveness of Residence Restrictions in Preventing Sex Offense Recidivism

Documents