Abstract
Sex crime policies have been associated with unintended effects. One involves harassment of registered sex offenders. Despite this finding, and the prominent role of the public in reentry efforts, few studies have examined public opinion about this collateral consequence. By extension, virtually no scholarship has investigated predictors of concern. From a utilitarian perspective, it follows that among those who rely on the registry or who feel sex offenders are driven to reoffend, concern about the registry’s unintended effects will be reduced. In testing this hypothesis the study draws on responses from a national public opinion poll conducted in 2005—a critical time when state registries and online registry websites became increasingly available. Findings support an instrumental or utilitarian argument. Use of the registry and pessimism about offender rehabilitation significantly decreased public concern about registries’ negative outcomes. Implications for research and offender reentry debates are discussed.
Introduction
Recently, an unprecedented number of “get tough” sex crime laws have been enacted by the federal government and states. Notably, the emergence of these reforms occurred during a time period when sex offenses had declined nationally (Finkelhor & Jones, 2004; Velázquez, 2008). This trend suggests that other factors—unrelated to objective indicators of crime—influenced the proliferation of sex crime laws. A potential contributor, per scholars, has been increased public concern about sexual victimization (Leon, 2011; Mears, Mancini, Gertz, & Bratton, 2008; Mustaine, Tewksbury, & Stengel, 2006; Wright, 2009), particularly for those offenses involving repeat offenders and child victims (Mancini & Mears, 2010; Sample & Kadleck, 2008).
Public opinion is also thought to be linked to the success of sex offender reentry (see, for example, Payne, Tewksbury, & Mustaine, 2010). This observation highlights a critical issue that Garland et al. (2013, p. 2) recently recognized in a larger context: “Reentry services, given their primary beneficiaries (i.e., criminals) and proximity to the community, have the potential to elicit strong opinions among the public, and these opinions have the capacity to affect the longevity and efficacy of these programs in varying ways.”
Yet, research examining sex offender reentry from a public opinion perspective remains an underdeveloped literature. Although some prior work has provided insight about public attitudes toward sex offenders returning to communities, efforts to systematically examine these views, particularly for how they might affect perceptions of the unintended or negative effects of these prominent sex crime laws, are noticeably absent. Juxtaposed against this research void is empirical evidence that some of these reforms—most notably sex offender registries—are associated with heightened levels of community fear (Beck & Travis, 2004), resident anger (Phillips, 1998), stigma and unemployment (for a review see Lasher & McGrath, 2012), and offender harm and harassment (Mercado, Alvarez, & Levenson, 2008; Tewksbury, 2005). Such research findings suggest that registry laws produce collateral consequences that might negatively impact sex offender reentry efforts. To illustrate, a larger recidivism literature has linked these deficits (such as social isolation and unemployment) with increased reoffending among sex offenders (Hanson & Morton-Bourgon, 2007). It is remarkable, then, that studies to date have not evaluated how the public—presumed to be a critical stakeholder in reentry efforts—views these unintended effects. This is all the more so given that public opinion has had a prominent role in shaping sex crime laws (Craun, Kernsmith, & Butler, 2011; Mancini & Mears, 2010) and figures prominently in the success of registered sex offender (hereafter, RSO) reentry (Payne et al., 2010). Thus an important question so far unaddressed is as follows: Are Americans concerned about a significant collateral consequence of sex offender registries?
A separate, but connected issue involves potential variability in public views. Although anecdotally it would appear that Americans unilaterally express similar opinions about sex crime and offenders, emerging empirical work has established substantial variation in public opinion (Church, Sun, & Li, 2011; Katz-Schiavone, Levenson, & Ackerman, 2008). More generally, this finding comports with a larger public opinion literature which has found that individual-level characteristics and beliefs are influential predictors in shaping perceptions about crime and justice (Applegate, Cullen, & Fisher, 2002; Payne, Gainey, Triplett, & Danner, 2004). In the context of predicting concern about collateral consequences of registries, instrumental factors stand out as potential predictors. For instance, sex offender registries are premised on the idea that greater capable guardianship—in the form of increased public awareness and precautionary efforts to prevent victimization—ultimately leads to sex crime reduction. Under a utilitarian framework, it follows that members of the public who rely on registries to feel safer or to protect their families, might be significantly less concerned about RSO harassment. In a similar vein, the endorsement of certain distorted perceptions about sex offenders—that they have unusually high rates of recidivism, or are significantly less amenable to treatment than other offenders—might also shape concern about registrant harm. This argument suggests that instrumental or rational concerns might drive public opinion about sex crime laws. Put differently, individuals who rely on the registry to protect themselves or their families and those who feel sex offenders are driven to offend, will express reduced concern about offender harassment because such an unintended effect would be perceived as a secondary concern to public safety.
Is this possibility problematic? If one takes the premise that the public is not influential in offender reentry efforts, the answer is “no.” However, to the extent that Garland et al.’s (2013) recent observation is correct—that public views figure prominently in efforts to reintegrate offenders—the answer is “yes.” Put simply, to the extent that utilitarian concerns and, more generally, public insecurity about sex crime drive apathy about the unintended effects of registry laws, efforts to reintegrate sex offenders may be difficult to achieve as there exists the potential for these collateral consequences to go undetected or unaddressed. Indeed, one could go one step further and make the argument that the public should be concerned about registries’ collateral consequences such as social isolation and harassment because they likely exacerbate the behavior registries are designed to prevent. 1
The current study would be among one of the first to examine concern about a prominent and unintended effect of sex offender registries—offender harassment—using national public opinion data. Applying a utilitarian framework, the specific focus is on the extent to which prior registry use and belief about offender amenability to reform impact public concern about the unintended effects of registries. In so doing, the study aims to contribute to a greater literature centered on public opinion about sex offenders and sex crime laws, and at the same time, highlight how findings apply to discussions and debates about offender reentry.
Sex Offender Registries
Logic and Efficacy
States began to develop and implement sex offender registries in the mid-1990s (Welchans, 2005). At a minimum, these policies require that law enforcement agencies provide information about the addresses of convicted sex offenders to the public. Depending on the specific state statute, the registry might also include pictures of offenders, their employment addresses, and other identifying information (Mancini, Barnes, & Mears, 2013). Recent estimates indicate that the registries affect a significant proportion of offenders. Close to 750,000 offenders nationwide are required to register per federal and state laws (National Center for Missing and Exploited Children, 2012).
Since their initial emergence, policymakers have focused on developing new ways to ensure sufficient public access to the registries. For example, with the advent of the Internet, beginning in the early and mid-2000s, most states began posting this information on publicly accessible websites (Anderson & Sample, 2008). Notably in 2005, the federal government created a national registry, the Dru Sjodin National Sex Offender Public Website (NSOPW), which provided the public with access to a federal database of sex offenders as well as additional links to individual state registry websites (U.S. Department of Justice, 2012). According to the NSOPW, this innovation represents “an unprecedented public safety resource” designed to assist parents, employers, and other concerned citizens with finding “location information on sex offenders residing, working, and attending school . . . in their own neighborhoods . . . [and in] nearby states and communities as well” (U.S. Department of Justice, 2012).
Despite their national implementation, few studies have detected significant crime-reducing effects of these policies. One of the largest studies to assess the effect of Megan’s Law examined states’ rape rates after passage of the legislation. In particular, it investigated whether the implementation of sex offender registries influenced monthly rape rates in 10 states. It found no uniform effect of the law (Vásquez, Maddan, & Walker, 2008). That is, 6 of the 10 states’ rape rates remained virtually unchanged, three states had significantly lower rape rates after the law was enacted, and one state experienced a significant increase in its rape rate. A more recent examination found no significant effect of Megan’s Law (and several other types of reforms) on reports of forcible rape from 1970 to 2003 (Ackerman, Sacks, & Greenberg, in press). Other studies have produced similar, nonsignificant results (Schram & Milloy, 1995; Zgoba, Veysey, & Dalessandro, 2010). To illustrate an exception, Letourneau, Levenson, Bandyopadhyay, Armstrong, and Sinha (2010) reported an 11% reduction in sex crime arrests (first-time offenses only) in South Carolina following the passage of a registration and notification policy. Notwithstanding this one positive finding of registry effectiveness, some scholars have identified possible reasons that might explain these largely null effects. One concern revolves around public use of the registries and subsequent protective action. Some evidence indicates that majorities of the public do not report accessing the registries or making significant behavioral changes upon learning about sex offenders living in the community (Anderson & Sample, 2008). Another prominent view relates to the potential for these laws to produce criminogenic effects—which might negate any potentially positive crime-reducing benefit, as recognized by Levenson, D’Amora, and Hern (2007, p. 598):
The social stigma and shame of sex offender registration can preclude or discourage participation in prosocial roles and activities, including employment, education, parenting, and property . . . self-concept, civic participation, and one’s perceived identity as a conforming and engaged citizen are related to criminal offenders’ desistance from crime.
Under this logic and as discussed further in the next section, certain features of the registry law might potentially function to contribute to recidivism among RSOs.
Collateral Consequences
Over the last decade, research has begun to identify the unintended effects or collateral consequences of registries. The laws have been associated with several negative outcomes such as housing instability and reduced employment opportunities (see, for example, Edwards & Hensley, 2001; Levenson & Cotter, 2005; Levenson, D’Amora et al., 2007; Tewksbury & Lees, 2006); two factors that a larger reentry literature has linked to an increased likelihood of recidivism (Hanson & Morton-Bourgon, 2007; Travis, 2005). Other unintended effects appear to stem directly from public anxiety and outcry concerning the RSO population. For example, prior studies have demonstrated that registries increase levels of citizen fear and anger (Phillips, 1998). Findings from offender surveys also reveal that harassment and vigilantism are additional collateral costs associated with registry laws (Brannon, Levenson, Fortney, & Baker, 2007; Levenson & Cotter, 2005; Levenson, D’Amora et al., 2007; Mercado et al., 2008; Tewksbury, 2005; Tewksbury & Lees, 2006, 2007). It is this latter outcome that appears especially disconcerting for reintegration efforts.
For example, Tewksbury (2005) found that substantial numbers of Kentucky registrants (n = 121) reported being harassed in person (47%), having received harassing and threatening phone calls (28.2%), and mail (24.8%), and to a lesser, but still striking extent, some reported being physically assaulted (16.2%). Remarkably similar findings were revealed by Mercado et al. (2008) in their later study of sex offender registrants in New Jersey (n = 138). In that investigation, approximately half of the RSOs reported having been physically threatened or harassed (48%) and 11% admitted having actually been physically assaulted or injured. In addition, the Mercado et al. (2008) study inquired about the impact of these events on offenders’ well-being. A majority of the RSOs reported that they were “afraid for their own safety,” “worried about the safety and well-being of family members,” and that “shame and embarrassment ke[pt] them from engaging in social activities” (p. 195).
This scholarship highlights several notable points. First, registry laws produce unintended effects. Second, one particular type—offender harassment—appears prevalent. A substantial number of the RSO population—about half, or extrapolating to the general RSO population, nearly 375,000 offenders—have reported harassment or vigilante victimization from community members (Mercado et al., 2008; Tewksbury, 2005). Third, as Mercado et al.’s (2008) findings indicate RSO harassment appears to reflect a disintegrative rather than reintegrative shaming philosophy (Braithwaite, 1999). That is, it is unlikely that public provocation functions to produce incentives for offenders to desist from crime. Taken together then, to the extent that RSOs experience harassment by community members and to the degree that it leads to disenfranchisement, the goal of successful reintegration will be more difficult to achieve. While illuminating, this literature leaves open the question of how the general public—stakeholders in reentry efforts (Garland et al., 2013; see also, generally Levenson & Tewksbury, 2009, p. 56)—perceive this collateral consequence. Given the goal of the current study—to address this research gap—we turn now to reviewing studies centered on public opinion about sex crime to provide context for the subsequent analysis.
Registry Laws, Sexual Offending, and Public Opinion
Beyond examining the impact of sex crime laws, a relatively new area of sex offender scholarship has focused on public opinion. Here the emphasis has largely been on exploring public support for these crime initiatives (see, for example, Kernsmith, Craun, & Foster, 2009; Phillips, 1998; Redlich, 2001; Zevitz & Farkas, 2000). One striking finding from this research is that the public strongly endorses support for these policies. For instance, in a study examining Melbourne, Florida residents’ perceptions about a range of sex offender sanctions, Levenson, Brannon, Fortney, and Baker (2007) reported that the greatest majority support was found for registry and notification laws (83%). Similar findings regarding the popularity of these laws have been replicated in different studies examining policy support among other populations (see, for example, Anderson, Evans, & Sample, 2009).
Notwithstanding this focus, a smaller number of studies have begun to explore the extent of registry awareness and use. To illustrate, Craun (2010) used hierarchical linear modeling (HLM) and assessed the extent of community awareness of convicted sex offenders’ residences in a county in the southeastern United States. Approximately 3 of 10 residents who lived within one tenth of a mile of an RSO were aware of a sex offender’s presence. In one of the only studies to systematically examine public awareness, use of sex offender registries, and subsequent protective action, Anderson and Sample (2008) found approximately 35% of Nebraskans reported accessing the state’s sex offender registry. Of those respondents who admitted accessing the registry, most endorsed positive views about its potential efficacy. For example, overwhelming majorities reported “feel[ing] family is safer” (87.6%) and “personally feel[ing] safer” (88%) after using the registries. A smaller, but still significant proportion of users (38%)—reported taking some type of preventive action after accessing the registry (e.g., installing an alarm system, monitoring children more closely; p. 383). Collectively, these findings are notable for underscoring that substantial minorities of the public appear aware of the registries, report having accessed them in the past, and convey feeling safer after using them.
Another related strand of this literature has examined perceptions about sex offender reform. For example, in a study of Virginia residents, Payne et al. (2010) analyzed predictors of rehabilitation views. In general, their findings indicate that the public holds largely pessimistic views about sex offender reform. Most respondents (52%) reported that sex offender rehabilitation was not likely; however, a nontrivial amount (36%) of the public appeared convinced that sex offender reform was indeed possible (the remaining percentage of respondents was unsure). At odds with these findings, as the authors noted, is research indicating that sex offenders have low odds of recidivating and appear as amenable to reform as other serious offenders (p. 585).
Notably, the Payne et al. (2010) study focused on identifying correlates of rehabilitation views and not on measuring the impact of such views on perceptions about sex offender policy or sex offenders. Given the seemingly widespread endorsement of sex crime myths, particularly about offender reform, a natural question is how they impact attitudes toward sex offenders. While few investigations have centered directly on how stereotypical beliefs affect public perceptions about sex offender harm or of sex crime policy generally (see, for example, Willis, Levenson, & Ward, 2010 for a review) some extant findings provide insight for the current study.
For example, Sanghara and Wilson (2006) using an experimental design examined an “experienced” group of sex offender practitioners (n = 60) and an “inexperienced” group (a sample of teachers who had no previous contact with sex offenders; n = 74). The researchers found that the inexperienced group was significantly more likely to hold inaccurate views about sex crime and offending (e.g., that sex offenders are typically strangers, are compelled to offend because of underlying mental illness). In turn, these beliefs triggered negative attitudes toward sex offenders (measured using a scale that included items such as “child sexual offenders never change;” “trying to rehabilitate a sex offender is a waste of time”). Although this research is not directly tied to measuring the perceptions about the unintended effects of sex crime laws, it more broadly highlights that the endorsement of sex crime–related stereotypes—particularly those about sex offender rehabilitation—shapes negative attitudes about sex offenders.
In short, this strand of sex crime research has greatly illuminated our understanding of public support for registry laws, other aspects of them (e.g., awareness and use), and general perceptions of sex offenders. Yet, substantially less is known about how the public views the unintended or collateral consequences of sex crime reforms. To a broader extent, this gap has been recently recognized by scholars as Willis et al.’s (2010, p. 546) observation suggests: “In view of the crucial role of social acceptance in the reintegration process it is surprising that only a few studies have addressed public attitudes towards sex offenders.” To illustrate an exception, Manza, Brooks, and Uggen (2004) using a national sample of Americans, examined public views toward felon disenfranchisement. They reported that the public was least willing to restore voting rights to convicted sex offenders, compared to white collar or violent, nonsex offenders. However, beyond this disenfranchisement outcome, the study revealed little about public attitudes toward collateral consequences of prominent sex crime laws. One other notable exception, a study conducted by Levenson and Tewksbury (2009) while not focused directly on general public views, examined RSOs’ family members’ perceptions about aspects of the registry law. It revealed that most family members expressed substantial concern about the registry, particularly concerning its efficacy and fairness. A last exception, a study conducted by Katz-Schiavone and Jeglic (2009) used an online message board and solicited information from respondents about their perceptions of Megan’s Law (n = 115). A finding relevant to reentry views and registration was that a majority of respondents (n = 95) agreed that Megan’s Law results in sex offenders being fearful for their safety. The study however used only a descriptive method and so left open the question of what factors predict these views, and more precisely—whether respondents indicated concern about the unintended effects of registries.
Collectively, this small handful of studies provides a springboard for the present analysis. It bears emphasizing, however, that this prior research has not examined public views about the negative consequences of registries. It is also noteworthy that extant investigations have typically not focused on national views (for an exception see Manza et al., 2004); relying instead on state or county-level data. The current study addresses both of these gaps. Specifically it analyzes national survey data to investigate the determinants of public concern about the collateral consequences of sex offender registry laws. The study’s theoretical focus is outlined below.
Utilitarian Concerns and Views About Sex Crime Policies
Theoretical explanations for public views are currently underdeveloped (Church et al., 2011; Payne et al., 2010). However, one set of factors that may influence attitudes toward sex offender laws include crime-related or instrumental concerns. To illustrate, those who use the registries as they are intended (as a safety tool) and those who feel sex offenders constitute a particularly dangerous offender population (i.e., one that cannot be reformed) can be expected to express substantially less concern about RSO harassment. This theoretical focus derives from a larger literature focused on public views toward offending and “get tough” laws. That is, research has found that utilitarian or instrumental factors—concern about crime, recidivism, and violent offending—tend to shape largely pessimistic views about offenders, and increase policy support for punitive initiatives among the public (King & Wheelock, 2007; Zimring & Hawkins, 1995).
To be clear and as Warr (1994, p. 48) has cautioned, while it is tempting to theorize about public opinion in simplistic terms (e.g., as punitive, indifferent, instrumental), “the complexity of legal punishment means that there are numerous facets of public opinion that require attention.” This observation indicates that public perceptions are nuanced; thus, instrumental concerns may be just one set of theoretically relevant predictors of public views about crime and justice policy. Having said that, sex crime laws are premised on an instrumental logic—that is, these laws were designed and implemented to control what has been seen by policymakers as an especially dangerous and threatening population of offenders (see, for example, Sample & Kadleck, 2008). Thus a natural extension of prior research involves investigating whether utilitarian factors also shape public perceptions about aspects of sex crime policy.
In the context of measuring views related to sex crime laws, it follows that individuals who use the registries out of belief that they effectively reduce crime (see, generally, Anderson & Sample, 2008) might not view offender harassment as a significant negative because the positives of the law—crime reduction—significantly outweigh the drawbacks—offender harm.
A similar argument applies regarding the public’s assessment of successful offender rehabilitation. According to some scholars, the proliferation of various “get tough” reforms targeting sex offenders stems largely from the belief that they cannot be rehabilitated (Payne et al., 2010; Wright, 2009). Indeed, judging from the special restrictions placed exclusively on released sex offenders (registration, residence boundaries, chemical castration), a predominant and related perception is that sex offenders are significantly less amenable to reform compared to other offenders (see, for example, Center for Sex Offender Management, 2000; Sample & Bray, 2003). To the extent that the public believes rehabilitation is not possible for sex offenders, concern about RSO harm will be significantly reduced. Put differently, pessimistic views about sex offender rehabilitation might be associated with reduced concern because the perceived benefit of the registry law—to control a threatening population of offenders—significantly outweighs its unintended effects (i.e., offender harassment).
Current Study
Research Questions
To summarize, research demonstrates that sex crime policies have negative collateral consequences, but scant empirical attention has examined public perceptions of such consequences. In addition, theoretical work on such views is remarkably underdeveloped. The current study seeks to address these voids, and more generally, “incorporate an understanding of how attitudes toward sex offenders overlap or contrast with broader justice-related attitudes” (see Church et al., 2011, p. 93). In particular, it examines the extent to which utilitarian factors affect concern about unintended effects of registry laws among the public by focusing on the following research questions: First, is prior use of the sex offender registry associated with concern about its collateral consequences—specifically, registrant harassment? Second, are rehabilitation beliefs associated with reduced concern about offender harm?
Hypotheses
From these questions, and applying findings from the previously reviewed research, the following two hypotheses are developed:
Hypothesis 1 (H1): Prior use of the registry will drive public apathy about the unintended effects of registries.
Hypothesis 2 (H2): Negative assessments of sex offender reform will reduce concern about the registry’s collateral consequences.
Data and Method
Data
The current study draws on responses from a national Gallup/Cable News Network/USA Today public opinion telephone poll housed in the Roper Public Opinion Network (“Poll # 2005-21: Social Security/Iraq/Filibuster Procedures/Child Sexual Molestation Crimes”). The survey (n = 1,006) was conducted in 2005 (April 29 to May 1) and asked Americans about their views concerning several controversial political and crime-related topics. 2 We focus on perceptions about child molestation offenses and the nationally implemented registry law.
The emphasis on 2005 data is strategic because this time period marks an important technological advance in how registry information was shared with the public. To illustrate, and as discussed earlier, in 2005, the year the study’s data were collected, the federal government created a national registry which provided Americans with immediate access to information about sex offenders across various jurisdictions (U.S. Department of Justice, 2012). In addition to the implementation of a national registry in 2005, most states during the early to mid-2000s began to develop online registry websites. These technological innovations—the establishment of a national sex offender registry and the development of online state registries—provided for greater opportunities for residents to become aware of and access these resources. Accordingly, reliance on this dataset ensures that we are tapping into an especially important era in understanding views and perceptions about sex crime and the registry law.
Measures
Dependent Variable
Reduced concern about registries
As discussed earlier, research has identified several unintended effects of the nationally implemented registries (Levenson & Tewksbury, 2009; Mercado et al., 2008; Tewksbury, 2005); however, efforts to assess public perceptions about these collateral consequences are noticeably absent. Thus, a specific measure of a prominent unintended effect is included in the analysis. In particular, the survey inquired about whether respondents were concerned about any unintended effects of registry laws, such as harassment of sex offenders. The specific measure was “How concerned are you that those [registry] laws will lead to harassment of the people whose names are on the lists?” The coding here was “1 = not at all/not too concerned” and “0 = very concerned/somewhat concerned.”
Independent Variables
Registry use
As theorized, those individuals who rely on the registry to learn about sex offenders living in the community, or use them as a crime-prevention method to protect themselves or their families, might not fully appreciate the unintended effects of registries, and so may be significantly more likely to express reduced concern about offender harassment. To determine the extent of the public’s use of the registry, the survey asked a subsample of respondents (those who reported being aware of the registry), “Have you, personally, ever checked the list, or not?” Options were “1 = yes” and “0 = no.” Of this subsample (n = 381), 57% of respondents reported accessing the registry (alternatively and as shown in Table 1, this means 22% of the entire sample reported prior registry use).
Descriptive Statistics.
Sex offender amenability to rehabilitation
Prominent sex crime–related stereotypes underscore the belief that rehabilitation efforts for sex offenders are unlikely to be successful. These views, in turn, may affect public judgments about the dangerousness and risk of convicted sex offenders. Thus, in the survey, respondents were asked to rate offenders’ amenability to be rehabilitated. In particular, they were asked two separate questions. The first measure gauged opinions specifically about threats to child victims: “Just your best guess, do you think people who commit the crime of child sexual molestation can be successfully rehabilitated to the point where they are no longer a threat to children, or not?” Responses were coded as “1 = not possible,” and “0 = possible.” Since additional postrelease restrictions are placed on sex offenders and not on other violent offenders, the second measure was designed to tap into views about sex offenders’ amenability to be rehabilitated compared to other nonsex offenders: “Just your best guess, do you think people who commit the crime of child sexual molestation are more likely or less likely to be successfully rehabilitated than people who commit other serious crimes?” The coding for this variable was “1 = less likely” and “0 = more/as likely.”
Controls
Although not the main focus of the study, prior research has identified several control variables shown to affect attitudes about offenders and crime policy (see generally, Borg, 1998; Hirschfield & Piquero, 2010; Moon, Wright, Cullen, & Pealer, 2000; Payne et al., 2010). Below, the theoretical justification for including these correlates is discussed.
Age
Some previous studies find that age is significantly linked to views about offenders and crime laws (Mancini, Barrick, Diponio, & Gertz, 2010; Moon et al., 2000; Wanner & Caputo, 1987). For this reason, age is included in the models as a control. Respondent age was measured in years.
Sex
Prior studies have uncovered a “gender effect” in explaining differences between men and women in their opinions about offending and criminal justice initiatives (Applegate et al., 2002; Sandys & McGarrell, 1995). That is, women appear more optimistic about offender rehabilitation, and this view in turn might explain their lower levels of support for punitive, “get tough” measures such as capital punishment, compared to men. At the same time, prior work suggests that women, in contrast to men, express greater concern about disadvantaged groups and vulnerable populations; in short, they endorse an “ethic of care” (Steffensmeier & Allan, 1996, p. 476). Given these findings the current study controls for sex. In the analyses, gender was dummy coded (“1 = male” and “0 = female”).
Race
Prior studies have identified race as “one of the strongest known correlates of attitudes toward . . . punitive forms of social control” (Messner, Baumer, & Rosenfeld, 2006, p. 566). As Messner et al. note, this scholarship has generally revealed that Whites endorse greater support for “get tough” punishments such as the death penalty compared to minorities. At the same time, non-Whites tend to express greater skepticism concerning the fairness of the criminal justice system compared to Whites (Henderson, Cullen, Cao, Browning, & Kopache, 1997). Given these race effects, analysis includes race as a control. The poll asked the following question: “What is your race? Are you White, African-American, Asian, or some other race?” Eighty-six percent of the sample reported being White, almost 8% reported they were African American, the remaining respondents reported “other.” Given this distribution and to have a measure that more closely corresponds to those used in prior research (e.g., Borg, 1998; Moon et al., 2000), race was coded as “1 = White” and “0 = non-White.”
Education
Studies have consistently found that individuals with more education are generally less punitive toward offenders as compared to individuals with less education (Fox, Radelet, & Bonsteel, 1991; Grasmick & McGill, 1994; Payne et al., 2004). For that reason, education is included as a control in the models. Education was measured in categories based on responses to two questions. First, respondents were asked the open-ended question, “What is the last grade or class you completed in school?” Second the survey inquired, “Do you have any post-graduate education beyond a 4-year Bachelor’s degree?” The responses to these two questions were combined to create an education measure coded as: “1 = less than high school degree,” “2 = high school/vocational degree,” “3 = some college,” “4 = college degree,” and “5 = graduate degree.”
Political conservative
Most studies find political conservatism to be linked with punitive attitudes toward offenders (Applegate, Cullen, Fisher, & Vander Ven, 2000; Borg, 1998; Unnever & Cullen, 2005) and less optimism toward offender reform (Nagin, Piquero, Scott, & Steinberg, 2006). Concomitantly, politically conservative individuals are more supportive of law enforcement efforts to control crime, even if negative effects on offenders are observed (Unnever & Cullen, 2005). Thus, a measure of it is included in the current analysis. In the poll, respondents were asked, “How would you describe your political views?” They could choose “very conservative,” “conservative,” “moderate,” “liberal,” or “very liberal.” In keeping with previous coding schemes (e.g., Chiricos, Welch, & Gertz, 2004), respondents were coded as “1 = conservative” if they chose “very conservative” or “conservative.” (All others were coded as “0 = not conservative.”)
Income
Some studies have found that those with higher incomes are more supportive of “get tough” crime reforms (Keil & Vito, 1991). Thus, an indicator of it is included as a control in the multivariate models. Toward the end of the survey, respondents were asked to report their “total annual household income before taxes.” This variable was coded categorically as, “1 = less than US$20,000,” “2 = US$20,000-US$29,000,” “3 = US$30,000-US$39,000,” “4 = US$40,000-US$49,000,” “5 = US$50,000-US$74,000,” “6 = US$75,000-US$99,000,” “7 = US$100,000-US$150,000,” or “8 = over US$150,000.”
Parental status
Surprisingly, scholars have just recently begun examining parental status as a predictor of crime-related views. Emerging research shows that individuals with children are significantly less supportive of punitive crime control and express greater support for offender reform—albeit, this effect appears limited to youthful offenders (Mears, Hay, Gertz, & Mancini, 2007; Welch, 2011). Moreover, some studies find that parents are significantly more supportive of sex crime laws compared to individuals without children (Mancini, Shields et al., 2010). Given these findings, multivariate analysis controls for the effect of parental status. It was measured by asking, “Do you personally have any children under age 18?” Respondents could report either “1 = yes,” or “0 = no.”
Residence
Similar to recent investigations of parental status, scholars have just begun to uncover residence effects on views about crime and punishment. Some prior studies indicate that urban residence “softens” punitive views toward offenders. Hirschfield and Piquero (2010) recently observed that urbanites were significantly more optimistic toward offender reintegration; to be precise, they were more willing to accept former offenders into their communities compared to rural or suburban Americans. Thus, the survey included a measure of residence (calculated by the respondent’s zip code). This measure was coded as: “1 = urban” and “0 = nonurban.”
Analytic Plan
Logistic regression was used to test the study’s hypotheses. Analysis proceeded as follows. First, the extent of reduced public concern was regressed on prior registry use in Table 2. The second set of models presented in Table 2 examined the effect of believing offenders are unlikely to be reformed to the extent that they no longer threaten children on concern. A third set of models tested whether the belief that sex offenders are not as rehabilitatable as other offenders predicted reduced concern about unintended effects. A fourth model included all of these instrumental factors on examining public concern. A final step involved estimating a series of multiplicative analyses designed to measure whether interaction effects existed between use of the registry and rehabilitation views.
Logistic Regression of Registry Reliance and Rehabilitation Assessments on Concern about Unintended Effects.
Note. Unstandardized coefficients (with standard errors in parentheses) and ORs are presented.
Findings
We turn first to descriptive statistics of the study presented in Table 1. As inspection of the table indicates, approximately two thirds of Americans reported reduced concerned about offender harassment. Alternatively, a nontrivial minority (almost 40%) indicate being somewhat or very concerned about offender harassment. This finding merits emphasis because it goes against anecdotal judgments about public opinion. That is, the argument that Americans invariably express similar opinions about sex offenders and policy does not appear to be empirically supported in this instance. Turning to registry use, in line with previous findings (Anderson & Sample, 2008), a minority of Americans (about one quarter of the total sample) reported having accessed them. As anecdotal evidence would indicate, Americans’ perceptions of RSO rehabilitation appeared less than positive; however, views were dependent on the specific indicator of rehabilitation amenability. A majority (73%) felt that rehabilitation of sex offenders to the extent that they no longer threatened children was not possible; however, 27%—a substantial minority—was more favorable toward RSO rehabilitation. A greater number of respondents (85%) were convinced that sex offenders were less rehabilitatable compared to other serious offenders; conversely, this indicates 15% of Americans believed rehabilitation was as likely to occur in the sex offender population as in other criminal populations (e.g., burglars, nonsexually violent offenders).
We now turn to results of the logistic regression analysis presented in Table 2 and our first research question. 3 Does prior registry use predict concern about the law’s unintended effect? Results from the first model indicate “yes”; prior use of sex offender registries reduced public concern about their collateral consequences. This effect remains even with the addition of theoretically relevant controls. On average, respondents who relied on the registries in the past had increased odds (odds ratio [OR] = 3.02) of expressing reduced concern about RSO harassment. These results provide partial support for the utilitarian or instrumental H1. Although not the focus of the study, one note about the control variables bears emphasis. That is, in model 1 the only statistically significant controls in the model were income and urban residence. In particular, in line with prior studies (Keil & Vito, 1990) income served to amplify negative attitudes toward offenders; that it, it was associated with a reduction in concern about the registries’ collateral consequences. In contrast, urban residence was linked with increased concern about offender harm among Americans. This finding directly accords with Hirschfield and Piquero’s (2010) research indicating that urban residence tends to “soften” stigmatizing attitudes toward ex-offenders.
The finding that few of the controls emerged as statistically significant accords with some prior scholarship that has documented little social and demographic variation in public opinion about sex offenders. For example, Payne et al. (2010, p. 584) reported in their study of Virginia residents that only one demographic variable significantly predicted views about sex crime policy. What this suggests is that views directly related to sexual offending (and not simply social and demographic characteristics)—may serve as more robust predictors of public perceptions of sex crime policy (see also, Mancini & Mears, 2010).
Support for an instrumental logic—and the study’s H2—is also evident in the second set of analyses that addressed our second research question—whether rehabilitation perceptions affected concern about the law’s unintended effect. As inspection of Table 2 (models 2 and 3) indicates pessimistic views about sex offender rehabilitation served to significantly reduce concern about collateral consequences. In particular, as shown in model 2, Americans who believed sex offender rehabilitation was not possible were over four times more likely to be unconcerned about offender harassment. In this model, three control variables were statistically significant. Income, again, emerged as a significant predictor of reduced concern. In contrast, race was associated with increased concern. Specifically, Whites were significantly more likely to be concerned about collateral consequences. As discussed earlier, Whites generally express greater punitivity toward offenders. How then can this race effect be explained? Payne et al. (2010)—who also demonstrated that minorities in their sample held negative views about sex offenders—provide an astute explanation (p. 581): “Different factors potentially influence attitudes about criminal justice issues for Whites and Blacks. In other words, Blacks [are] believed to be punitive for one set of reasons, while Whites are believed to be punitive for another set of reasons.” Thus, it may be that certain issues—in this case, sexual offending—differentially activate punitive attitudes among Whites and Blacks. 4 Along a similar direction, urban residence was linked with increased concern among Americans, in line with extant research (Hirschfield & Piquero, 2010).
Endorsing the related, but distinct view—that sex offenders were less rehabilitatable compared to non-sex offenders—was also associated with reduced odds of concern (model 3; OR = 2.41). Here, once again, respondents with greater income were more likely to report reduced concern. In a different direction, higher education and urban residence were predictors associated with increased concern about offender harm.
A final and competing model (model 4) included all utilitarian predictors to assess which exerted the greatest influence on concern about RSO harassment. These results indicate that registry use was still associated with reduced concern. In addition, the first rehabilitation view—doubting that sex offenders could be reformed to the extent that they “no longer represented a threat to children” was also a significant predictor. In this model, the effect of believing that sex offenders could not be rehabilitated as successfully as other, serious offenders was no longer statistically significant. As few studies have examined public views toward sex offender rehabilitation (see, for example, Payne et al., 2010) it is not entirely evident why this pattern emerged. It may be that the former measure—which emphasized threats to child victims—invokes a greater response among the public, and therefore serves as a more direct predictor of rehabilitation attitudes. Notably, these utilitarian predictors fully mediate any statistically significant effect of the control variables. This is noteworthy because it indicates, as some prior research has demonstrated (Payne et al., 2010), that in the context of measuring public views about sex offenders, specific sex crime–related indicators should be examined.
Collectively, these findings provide support for the utilitarian hypothesis. Per the theory, individuals who are safety oriented (a driving motivator for registry use per Anderson & Sample, 2008) and who are unconvinced about the possibility of sex offender rehabilitation might see the unintended and negative aspects of registry laws as less problematic than the potential for offenders to recidivate. It bears emphasis as well that these factors related to utilitarian concerns—registry use and rehabilitation beliefs—exert unique and direct effects on public opinion. Put differently, no statistically significant interaction between them (“registry use × rehabilitation assessments”) surfaced in ancillary models (results not shown). 5
Conclusion and Implications
To address calls for further attention to public views about sex offenders and sex crime laws (Church et al., 2011; Rogers & Ferguson, 2011; Willis et al., 2010), this study provided the first examination of the correlates of public concern about the unintended effects of registries using national survey data. Two findings are notable. First, registry use served to reduce concern about a prominent unintended effect—offender harassment. Second, pessimism about offender rehabilitation was associated with significantly less concern about registry’s collateral consequences. In line with prior research (Mancini & Mears, 2010; Payne et al., 2010) few social and demographic controls consistently predicted concern across the models. This finding highlights the need to consider specific sex crime–related indicators of views (e.g., perceptions of offender reform) when examining public opinion about sex crime.
Taken together, these findings support the utilitarian or instrumental argument regarding crime control. In short, individuals who rely on the registry as a means of crime prevention might not sufficiently acknowledge the unintentional aspects of the law—as safety concerns might supersede concern about offender harassment. A similar logic applies to rehabilitation beliefs. Among those who remain doubtful about offender rehabilitation there exists a greater potential to discount potential drawbacks of the registry law, as the motivation to control a presumably dangerous population driven to recidivate might weigh as a greater priority among the public than ensuring that the law does not produce unintended, collateral effects. Research and implications for offender reentry debates flow from these findings.
A broad research implication involves greater attention toward measuring public views about offender reentry and predictors of these views. The current study tapped into views concerning an unintended effect of sex offender registries—the extent of public concern about registrant harassment. This perception constitutes only one type of view related to offender reintegration. Other measures of reentry views might include questions that tap into direct, real-world experiences. To illustrate, future studies should consider measuring the extent to which Americans would be likely to hire convicted sex offenders, to socialize with them, to rent a home to them, or to allow their children to befriend another child whose parent is a sex offender.
In addition, studies along this line should frame these questions in a theoretical context to more fully understand how the public views sex offender reentry. For example, as Payne et al. (2010, p. 587) explained, “it is clear that sex offenders are stigmatized and this stigma has hindered appropriate responses to sex offenders. What is not clear is how this stigma arises . . .” Although utilitarian factors—use of the registry law and pessimistic assessments about offender rehabilitation—served to drive decidedly negative views in the current study, it may be that experiential factors, such as actual contact, or interpersonal relationships with convicted sex offenders in the workplace, neighborhood, or community amplifies more positive assessments regarding their reform and reentry prospects.
Under this logic, Hirschfield and Piquero (2010) recently tested the contact hypothesis and found that exposure to former offenders reduced stigmatizing attitudes about other ex-offenders among the public; contact in fact was the strongest predictor of such views about this population. Notably, however, the study did not evaluate the effects of different types of contact on attitudes toward ex-offenders, nor did it assess views about specific types of offenders. It is possible that different types of contact with offenders may have varying effects on views about the former—indeed, the authors (p. 45) emphasize that this line of inquiry “merit[s] future research.” More generally, this line of investigation would go far in addressing Warr’s (1994) concern that future studies consider the complexity in public opinion and work toward identifying nuances in public views toward offending and crime control.
We turn now to a central theme of the study, implications for discussions and debates about offender reentry. As underscored by study findings, although a nontrivial percentage of Americans is at least somewhat concerned about offender harassment, the majority of the public remains indifferent over accounts that RSOs are harassed or hassled by others. Put differently, most Americans, while certainly not condoning RSO harassment, do not perceive it as a significant problem or negative aspect of the registries. In particular, utilitarian or instrumental factors—reliance on the registries and negative assessments about successful offender rehabilitation—served to amplify these views.
How might these perceptions affect reentry efforts? Given that the public is presumed to be a stakeholder in efforts to reduce crime and victimization, these findings indicate possible cause for alarm. Although public opinion may not directly drive public policy, it is thought to matter in a democracy (Burstein, 2009). As sex crime experts have emphasized, increased public concern has been a key catalyst in the creation of new sex offender reforms across the country (Wright, 2009). More generally scholars have recognized that public views can contribute to successful offender reentry (Garland et al., 2013). To the extent that the public, a critical stakeholder in efforts to prevent sexual victimization, expresses little concern about the negative and possibly crime-inducing effects of sex crime laws, there exists the potential for these collateral consequences to go undetected or unaddressed, and ultimately lead to future reoffending among sex offenders.
Judging from the substantial percentages of Americans who endorse stereotypical views about RSOs, particularly concerning their propensity to recidivate, it may be that the public is largely misinformed about the reality of sex offending. At the same time, utilitarian motives (e.g., to ensure safety of children, vulnerable populations) appear to drive public use of the registries (Anderson & Sample, 2008). It may be that the registries—which are presumably accessed by individuals concerned about sex crime—thus could be used as an outreach tool to provide factual information about sex crime and sexual offenses. This approach would involve standardizing the context of the over 100 registries that currently exist nationally (50 state sites and numerous others serving tribal communities; U.S. Department of Justice, 2012).
To be clear, a number of critiques have been leveled against efforts to standardize state registries (see generally, Freeman & Sandler, 2010; Harris, Levenson, & Ackerman, 2014). To illustrate, Title 1 of the Adam Walsh Act Child Protection and Safety Act (2006; Sex Offender Registration and Notification Act, SORNA) is designed to promote greater uniformity across registries by requiring states to implement a tier system which indicates offender risk of recidivism—Tier I (low), Tier II (medium), and Tier III (high). Per the Act, offenders’ risk levels are based on the crime of conviction. Notably, a majority of states are not in “substantial compliance” of the law (Brewster, DeLong, & Moloney, 2013). Critics contend the new system is overly broad, restrictive, and costly (as reported by Ragusa-Salerno & Zgoba, 2012). Thus, recent debates of standardization have primarily centered on the scope and content of the law (i.e., which offenders are required to registry). Having said that, substantially less attention has been directed toward efforts to standardize registry context—that is, the introductory or supplemental information provided on registry sites which assist the public with appropriate use of the registry (for a recent content analysis of state registries, see Brewster et al., 2013). 6
It is this dimension that merits further policy consideration. Put differently, given the goal of developing registries that are more accessible and informative it would be appropriate to consider a few modifications. To illustrate, these sites might emphasize that a wide variety of sex offenders appear on it. This is an important distinction because scholars contend that the media have fueled distortions about sex offenders, typically reporting on only the most violent, predatory, and homicidal offenders (Brown, 1999; Sample & Kadleck, 2008). On the contrary, wide heterogeneity exists in offender “types”—from those who have victimized children and strangers, were physically violent during their offense, and those who have committed multiple sex crimes—to those offenders who have committed lower level sex crimes such as statutory rape, indecent exposure, or non-contact sex offenses (e.g., accessing child pornography; Mears et al., 2008). In addition, it should be mentioned upon accessing the registry that research indicates wide variation in recidivism rates among these offender types; but, most sex offenders, however, are unlikely to recidivate (Sample & Bray, 2003). This approach would paint a more accurate portrait of the diverse population of offenders who appear on registry sites. Not least, the registry sites could contain user-friendly applications (e.g., glossaries for technical legal terms, “web quizzes,” tutorials, learning modules), which furnish information from empirical studies and provide the public with best practices to prevent victimization. Ideally, these features would be designed to allow the average layperson to easily understand the intent of the registry law, and offender “types” and offense heterogeneity. It may be that once Americans are exposed to the reality of sexual offending, they will be in a better position to fully acknowledge the strengths and limitations of the registry law, and other reforms, in preventing sex crime. All of these considerations are for naught if in fact citizens do not access the registries. Thus, a concomitant implication would focus on investigation of how best to introduce these approaches and to improve usability and navigation of the registry sites. This might include using social media to introduce concerned citizens to the websites, sending flyers or e-mail notifications to residents, ensuring registry information is also provided in other languages (see, for example, Brewster et al., 2013), and similar approaches.
These modifications may not radically change public opinion, but they may lead to greater dialogue about how best to present factual information to a concerned public in an effort to enhance sex offender reentry. It is clear that the public constitutes an important stakeholder in sex crime reform and policy. Accordingly, their views—to the extent that they accord with empirical realities or not—should be carefully measured and understood.
Footnotes
Acknowledgements
The author thanks two anonymous reviewers and the Editor for helpful comments on an earlier draft.
Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author received no financial support for the research, authorship, and/or publication of this article.