Abstract
Much attention has been paid to the examination of community sentiment regarding convicted sex offenders and the policy that governs these offenders’ behavior. This literature, however, has largely been absent of international comparisons of sex offender community sentiment. The current study seeks to fill this gap by drawing from the results of surveys (n = 333) conducted in both the United States (US) and the United Kingdom (UK). Results indicate that sex offender policy is generally supported in both the US and the UK. Contrary to our expectations, we found that participants from the UK were less tolerant of sex offenders residing in their neighborhoods than participants from the US. Additionally, there is support for the notion that sex offender policy holds a symbolic value for both study locations. Theoretical and practical implications of these findings are discussed.
Introduction
Currently there exists a wealth of literature which assesses the community sentiment and public opinion of individuals convicted of sexual crimes and relevant sexual offense (SO) policies (Levenson & Cotter, 2005; Levenson & Tewksbury, 2009; Tewksbury & Lees, 2006). More specifically, much of the literature is dedicated to examining the predictors of these community sentiments and determining why people feel as they do (Levenson et al., 2007; Olver & Barlow, 2010). However, despite its importance in understanding geographical and policy development differences, no current research studies exist that describe or compare public sentiment across countries that have implemented similar sexual offense legislation.
As sexual offense legislation continues to grow across the globe, the United States (US) has been the most proactive with the advent of Megan’s Law, the Adam Walsh Act, the International Megan’s Law and the Angel Watch Center, while the United Kingdom (UK) has been progressive with Sarah’s Law and international travel notifications for sexual offenders. Despite these countries taking similar action toward implementing sexual offense policies, a small number of studies have noted interesting differences in the community response to such legislation. For example, McCartan (2013) reported that participants of a focus group in the UK would not be supportive of practices embraced in the US that disclosed personal information, such as home and work addresses, of those convicted of sex offenses. Parliament and the Home Office have reacted in a similar manner and have consistently denied the public access to private information regarding individuals convicted of sexual crimes (Griffin & Blacker, 2010). Conversely, however, The Telegraph recently reported outrage over the manner by which the UK’s sexual offense policies were being enforced, as individuals who receive information on sexual offenders residing in their proximity were mandated to sign nondisclosure agreements (NDAs) ensuring they would not disclose the information to others (Swerling, 2020). The Telegraph’s position was that this information may be valuable to others who come into contact with the previously convicted sexual offender and that enforcing NDAs through legal and criminal means defeats the disclosure purpose (Swerling, 2020). Additionally, Britain’s News of the World reported that they would take matters into their own hands by publicly outing sexual offenders through a “Name and Shame” campaign (Griffin & Blacker, 2010). Within the US, however, broad notification of sexual registry information has consistently been supported among the community and lawmakers. Research demonstrates that the public widely supports knowledge of private information regarding convicted sex offenders (Brannon et al., 2007), even if databases that house this information are rarely accessed (Harris & Cudmore, 2018). Unlike the mixed response in the UK, the majority of those in the US have voiced unwavering support for increasing the public’s knowledge of sex offenders’ whereabouts and decreasing their right to privacy.
Thus, it would appear there may be differences related to the perceptions of sex offenders and resulting policies in these two geographic regions. As the disclosure and notification processes of the sexual offense legislation varies greatly between the countries, it is noted that the differences in community perceptions may have impacted the drafting and implementation of the broad notification in the US and the limited disclosure in the UK. These distinctions in allowing the public access to private information may be rooted in the variable governmental beliefs between the countries and citizens’ beliefs in privacy. As noted by Griffin and Blacker (2010), the UK places a higher value on private reputational interests compared to the US and highlights these distinctions by noting differences in libel law, defamation and camera use in court. In each of these scenarios, the UK is significantly more concerned about privacy and reputation, while the US places higher value on an individual’s right to free speech (Griffin & Blacker, 2010). For example, UK news outlets that reveal a suspect’s identity or adversely affect an individual’s reputation will face harsh sanctions, and libel suits are more often victorious. Conversely, in the US, if the defendant is not proven to have disregarded the truth, the case defaults on the side of the First Amendment and the defendant holds no responsibility if privacy or reputation are placed at risk (Griffin & Blacker, 2010). In sum, these distinctions may aid in the explanation as to why the United States allows for such broad publication of their sexual offender registry, while the United Kingdom maintains its privacy. The current study seeks to be the first to delve deeper into these possible disparities in opinion and examine public perception of sexual offense laws and individuals convicted of sexual offenses in the US and UK.
Background of US and UK Policy
Both the US and the UK have a variety of laws that dictate post-conviction requirements for sexual offenders, such as community notification, residence restrictions, and mandatory registration. Community notification laws require that individuals and neighbors living within the vicinity of an individual convicted of a sexual offense be notified of their status, while residence restrictions refer to prohibitions upon sex offender residential locations (such as near schools, playgrounds, or daycare centers). Registration policies mandate that law enforcement agencies retain repositories of personal information for those convicted of sexual crimes. Within the US, community notification and registration laws are generally referred to as Megan’s Law, while in the UK, they are informally referred to as Sarah’s Law (McCartan, 2013; Zgoba, 2017; Zgoba & Cowan, 2019). Both of the names of these laws stem from individual cases in which a violent, sexually based offense occurred against a minor, resulting in wide media publicity, community outrage and the advocacy of stricter laws against individuals convicted of sexually based offenses.
In the US, sexual offense registration and notification (SORN) practices resulted from the adoption of Megan’s Law and its successor, the Adam Walsh Child Protection and Safety Act of 2006. These acts mandated that states adopt practices to register sex offenders of several different classifications in an effort to store offender information into a national database. This database would subsequently be used to track the whereabouts of known offenders in an effort to reduce recidivism (Adam Walsh Act, 2006). Conversely, in the UK, registries were established through the Sex Offenders Act of 1997, and far less personal information was made publicly available (Sexual Offences Act, 1997, 2003). This act was subsequently updated under the Sexual Offences Act of 2003. Unlike the US, which allows for nearly unfiltered access to the databases, the UK requires individuals to file requests for information on a case-by-case basis and if information is provided, it cannot be shared (Blacker & Griffin, 2012; Zgoba, 2017; Zgoba & Cowan, 2019). For example, parents who are seeking a caregiver for their child can file a petition with authorities to see if the potential caregiver has previous convictions for sexually based offenses. In addition to this limited system of disclosure within the UK, authorities may also disclose this information to the public if an individual is deemed to be a significant risk to the community (Sexual Offences Act, 2003).
Public Perceptions of Law and Offenders in the United States and United Kingdom
Research on the perceptions of laws and offenders can be broken down into two main categories that are based on the respondents within the sample—offending and non-offending populations. Studies that have utilized sexual offenders as their survey subjects have found that they experience a variety of negative consequences from residence restrictions, registration, and community notification laws, including loss of employment, harassment and stigmatization from the public, property damage, and assault (Brannon et al., 2007; Levenson & Cotter, 2005; Levenson et al., 2007; Tewksbury & Lees, 2006). Due to these extra-legal consequences and the obvious restrictions, many sex offenders have an overall negative view of SORN laws and residence constraints (Brannon et al., 2007; Tewksbury & Lees, 2006). Moreover, Tewksbury and Lees (2006) argued that the label of “sex offender” often carried more weight and social harm than that of the term “felon.” Meaning, the label of “sex offender,” in the eyes of society, has a more negative stigma associated with it than other types of criminal activity. This enhanced negative stigma can thus result in more social consequences, such as an inability to find gainful employment (Tewksbury & Lees, 2006). Consistent with labeling theory, the labeling process and subsequent stigmatization gives credence to further examine how the criminal justice system can lessen the impact resulting from punishments for these offenders.
In terms of the perceived effectiveness and impact of SORN laws by both offenders and non-offenders, Brannon et al. (2007) reported that individuals convicted of sex offenses perceived community notification laws to be ineffective at reducing recidivism rates. While the overall sentiment of SORN laws is negative with offenders, Levenson and Cotter (2005) reported some positive indirect outcomes of these mandates. Drawn from in-depth interviews, registered sex offenders reported that SORN laws provided them accountability in their daily activities and helped them maintain honesty with their friends and family (Levenson & Cotter, 2005). Apart from these examinations of perceived effectiveness and impact, scholars have also examined adherence to sex offender myths. These myths include notions that the majority of sexually based offenses involve a perpetrator who is a stranger to the victim or that the majority of offenders are victims of sexual offenses themselves. To test these adherences, Fortney et al. (2007) conducted a survey and found that offenders, much like their non-offending counterparts, overestimate the rates of offenses involving a stranger and victim, as well as the number of offenders who are previous victims themselves. It is argued that the adherence to these myths may play a part in the policy that was subsequently enacted (Fortney et al., 2007).
When conducting surveys with individuals not convicted of sexual offenses, Fortney et al. (2007) found that the public overestimates recidivism rates for sex offenders and also underestimates the efficacy of post-conviction treatment. With the exception of family members of sexual offenders, research from non-offenders also indicates that these individuals are largely not aware of the vigilantism that offenders face (Brannon et al., 2007; Levenson & Tewksbury, 2009). In an online survey of families of registered sex offenders, Levenson and Tewksbury (2009) reported that some families themselves experienced the same harassment and stigmatization that offenders report experiencing. These unintended consequences of SORN laws often created more challenges to reintegration of sex offenders by causing further instability in their lives and by not being able to find support from their families who might, at times, distance themselves from the offender in order to avoid subsequent harassment to themselves (Levenson & Tewksbury, 2009).
In large part, non-offending populations are supportive of community notification, residence restrictions, and registration (Comartin et al., 2009; Schiavone & Jeglic, 2009; Zgoba & Cowan, 2019), although some would question the effectiveness of these practices at reducing recidivism (Schiavone & Jeglic, 2009). Thus, it would appear that perceived effectiveness is not the sole driving force behind support of these practices. A number of research studies suggest that the support for SORN laws could emanate not from effectiveness but may function through a fear of victimization and sex offenders in general (Comartin et al., 2009; Kernsmith et al., 2009). In a US-based telephone survey of non-offenders, Kernsmith et al. (2009) found that crimes against children were the most feared out of sexually based offenses. Additionally, this fear of offenders was correlated with agreement of registration practices for convicted offenders. These heightened emotional reactions to fears of child-based offenses could be an additional catalyst toward favor of SORN policies (Kernsmith et al., 2009).
Predictive Factors of the Perceptions of Sex Offenders and Associated Policy
Literature that seeks to examine why people feel the way they do, meaning the predictive factors of the perceptions of sex offenders and associated policy, has been vastly documented over the last two decades. The studies have assessed several types of predictors, including participant demographics (Brown et al., 2008; Harper et al., 2017; King & Roberts, 2017; Levenson et al., 2007; Olver & Barlow, 2010; Payne et al., 2010, 2016; Rogers et al., 2011; Willis et al., 2013; Zgoba & Cowan, 2019), familiarity with victims and /or offenders (Harper et al., 2017; Mustaine et al., 2015; Tewksbury & Mustaine, 2013; Willis et al., 2013), victim characteristics (Rogers et al., 2011), and belief in sex offender myths (King, 2019). Additionally, scholars have also examined a variety of criminal justice and programming attitudes, including, general attitudes toward offenders (Harper et al., 2017; Levenson et al., 2007; Rogers et al., 2011; Tewksbury & Mustaine, 2013; Willis et al., 2013), ideal punishment for offenders (Brown et al., 2008; King & Roberts, 2017), support for punishment (King, 2019; Levenson et al., 2007; Payne et al., 2016), and treatability of offenders (Payne et al., 2010; Olver & Barlow, 2010).
Research examining demographic factors as predictors of perceptions of sex offenders and relevant policy have found little significance (Harper et al., 2017; Olver & Barlow, 2010; Rogers et al., 2011), although there were some marked differences between several demographics (Brown et al., 2008; King & Roberts, 2017; Payne et al., 2016; Willis et al., 2013). In a UK based study of the perceptions of sex offenders, Brown et al. (2008) found that females estimate higher recidivism rates than their male counterparts. Additionally, and of particular interest to the current study, younger individuals and those with higher educational attainment were also more likely to report that the media exaggerates the risk that sex offenders pose to society (Brown et al., 2008). This finding could perhaps point toward these individuals being less likely to subscribe to myths regarding sex offenders and the subsequent likelihood of re-offending following a first offense. With reference to punishment preferences, King and Roberts (2017), in an assessment of attitudes toward punishment for sex offenders within the US, found that females, those with less educational attainment, and those with children under 18 held more punitive views toward offenders. More specifically, females and those with less educational attainment preferred longer prison sentences and registration requirements of sex offenders. In alignment with these findings, Willis et al. (2013), found that females and individuals with less educational attainment held more negative views of sex offenders in New Zealand.
The occupation of survey participants has been also associated with the perceptions that individuals hold toward sex offenders (Connor & Tewksbury, 2017; Mustaine et al., 2015; Tewksbury & Mustaine, 2013). Specifically, Tewksbury and Mustaine (2013) found that law enforcement officers and their managers generally have negative views of offenders, while Mustaine et al. (2015) found that judges also hold overall negative views of sex offenders. Furthermore, Fortney et al. (2009) noted that individuals who worked with victims perceived higher recidivism rates among sex offenders than those who worked with offenders. Interestingly, Connor and Tewksbury (2017) reported that individuals who assist in the treatment of sex offenders hold overall positive views of their progress and recidivism risk. Reasons for this contrast could be related to the nature of the contact with offenders. More specifically, law enforcement officers are tasked with apprehending individuals who have or are suspected of committing offenses and often do not have experiences with individuals outside of a legal/criminal situation. On the other hand, individuals who are involved with the treatment of sex offenders will come in contact with individuals who are not in prison and are participating in efforts to refrain from subsequent offending (Connor & Tewksbury, 2017). According to Lea et al. (1999), these variations in perceptions could be related to the differences in training individuals receive regarding offenders. Law enforcement professions tend to gravitate toward perceptions of security and justice, while therapists often view the world through a therapeutic lens.
Apart from demographic factors, studies that examine whether respondents are a victim or know a victim of sexual assault have found little support for predictive power regarding the perceptions of sex offenders (Harper et al., 2017). Additionally, research has found that family members of offenders typically have positive views of the offenders themselves, as well as negative views of the punishments associated with the committed offenses (Connor & Tewksbury, 2017; Levenson & Cotter, 2005). Research has also examined the impact of belief in misconceptions, or myths, on perceptions of sex offenders (King, 2019). Misconceptions refer to beliefs that are non-factual, such as the notion that sex offenders are not amendable to treatment and that sex offender recidivism rates are high or increasing. These misconceptions were tested for their relation to punitiveness toward sex offenders and results indicated that support for these misconceptions was a significant predictor of punitiveness. Quinn et al. (2004) posit that these misconceptions and the beliefs that sex offenders are irredeemable are largely perpetuated by over-sensationalization in the media and are transmitted among individuals. In total, these studies highlight the importance in understanding the nuances among the predictive factors that guide the public’s perceptions of sex offenders.
Theoretical Frameworks for Explaining Variation in Community Sentiment
To explain the public’s perception of sex offenders, researchers have put forth two prominent theoretical foundations, including implicit theory (Harper & Bartels, 2017) and moral foundations theory (Harper & Harris, 2017). These two theories have specific implications from which predictors of community sentiment can be inferred. For the current study, each of these proposed frameworks will be utilized to explain community sentiment of sex offenders and relevant public policy. Implicit theory posits that there are two primary ways of thinking that guide perceptions—entity- and incrementally-based rationales. Entity-based thinking refers to the notion that behavior, such as sexual offending, is fixed and individuals are not amendable to change (Harper & Bartels, 2017). Conversely, incrementally-based thinking adheres to the notion that individuals are malleable in their behavior. Using an implicit theories framework, perceptions are guided by stereotypes of human behavior. Thus, for sex offenders, the common assumption is that they are male, predatory, and have a child victim (Harper & Bartels, 2017). To test this theory, Harper and Bartels (2017) conducted a survey and found that individuals who aligned with entity-based thinking regarding sex offenders typically held more negative views, while the opposite was true for incrementalists. With reference to the current study, this theoretical framework would be identified through belief in sex offender myths. Belief in these myths, such as sex offenders not being amendable to treatment, would align with an entity-based rationale for offending patterns. Thus, support for this theoretical framework would be found in the current study, if participants’ belief in myth is predictive of the proposed dependent variables.
Apart from implicit theory, the basis for moral foundations theory in the context of sexual offending is that individuals have varying moral compositions which ultimately lead to varying degrees of support for policy (Harper & Harris, 2017). Much like implicit theories discussed, individuals often base their sense of morality off preconceived notions, or stereotypes. To expand on this, Harper and Harris (2017) discuss these preconceived notions on a “feel first, rationalize later” mentality (p. 115). This initial reaction to an event or idea is guided by stereotypes. With reference to sexually based offenses, moral foundations theory would put forth that the common individual initial reaction would be that of disgust or moral condemnation. These reactions would be different on an individual basis depending on the type of offense being discussed. As mentioned, research has shown that offenses against children are seen as especially heinous (Kernsmith et al., 2009), thus, moral foundations theory has utility in explaining the individual variation that is found within the perceptions of sex offenders and relevant policy (Harper & Harris, 2017). As discussed, within this theoretical framework, one can anticipate overall negative reactions to sexually based offenses. For the current study, then, one might expect to find overall negative perceptions relating to sex offenders and more support for SORN policies that govern these convicted offenders. Additionally, when comparing different nations, it is possible that differing moral compositions of the population will lead to variations in the community sentiment of these offenders. Thus, the US and the UK may have differing perceptions of policy and offenders.
Methods
Purpose of Study
The purpose of this analysis was to assess the public opinion of UK residents regarding the implementation of Sarah’s Law, in comparison to the perceptions of American residents after the application of Megan’s Law. While the subjects had been surveyed years apart, both samples were questioned after the law went into effect and within a short duration of high-profile media reports of child sex abuse cases. While news media’s opinions on the topic of child protection laws through SORN are widely publicized, the public’s sentiment and opinions are much lesser known. In particular, as the US and the UK are the countries that have most widely adopted sexual offense laws, no existing studies have examined the similarities or differences in residents’ attitudes toward sexual offense laws, sexual offenders and the treatment of convicted offenders. Since previous research has demonstrated a potential ‘contagion effect’ of sexual crime panic and coverage spreading across continents, it is important to capture public opinion firsthand (Neuilly & Zgoba, 2005, 2006). As Sarah’s Law operates under a different disclosure and privacy mechanism from Megan’s Law, we posit that residents will be largely in favor of the law in each of the countries, however, given the limited disclosure of private information within the UK, we believe residents will expect a balance between the release of ex-offender private information and the public’s right to know. Accordingly, we believe that US citizens will be less tolerant toward sexual offenders and treatment. Consistent with these hypotheses, this study tests whether geographic location (US and UK) predicts the likelihood of perceiving sex offender laws to be fair, belief in sex offense treatment, and tolerance toward those convicted of sexual offenses. This study fills an existing void in transnational public perception research on the important topic of sexual offenders and the laws that strive to govern their post-release behavior.
Methodology
In an effort to compare UK and US perceptions on SORN laws, the present study uses the archived public opinion data collected in Florida (FL) from Levenson et al. (2007), while also collecting new data on perceptions from the UK public after the implementation of Sarah’s Law. In addition, it replicates the methodology and utilizes the same survey tool from the Levenson et al. (2007) study when asking the public their opinions of SORN policies in the UK. The researchers conducted surveys in Spring 2015 in London and Bristol, England. These two cities provided an ideal backdrop to the discussion on Sarah’s Law; the law had been implemented within the past few years, and the law’s namesake, Sarah Payne, was abducted and murdered in Kingston Gorse, England (Zgoba & Cowan, 2019). Both locations however, allowed for variability in opinions and beliefs—Bristol is very close to Wales and has a strong university influence, while London provides a more urban setting. This is a similar methodology employed by Levenson et al. (2007), as Florida had implemented Megan’s Law years prior and in 2005, they strengthened the standards for Megan’s Law by passing the Jessica Lunsford Act. Florida was considered an ideal state in the United States for the first round of data collection because it maintained one of the most comprehensive SORN policies in the nation, employing broad notification of all sex offenders through aggressive community-based strategies and via the state’s public internet registry (Levenson et al., 2007). As such, this study relied on the combination of survey answers from these two samples, yielding a total sample size of (N = 333). Table 1 displays the demographics from the selected samples.
Demographics of Sample.
Sample 1—Melbourne, Florida
The data were collected in August 2005 from 193 random participants in Driver Motor Vehicle (DMV) agencies in Melbourne, Florida. Melbourne is located in Brevard County, on the central east coast of Florida (see Levenson et al., 2007 for a detailed description of the representativeness of the geographic area). The data were collected in two DMV offices within four consecutive business days. The researchers were cognizant to not bias the sample and therefore approached a variety of people. The researchers followed a standardized format- they introduced themselves, explained the purpose of the study, sought participation, answered questions, provided a clip board, along with a pencil, the survey, and the informed consent. The researchers explained that survey completion implied their consent to participate. Once the surveys were complete, the researchers placed them in an envelope. No identifying information was written on the survey or the envelope. Lastly, as noted in Levenson et al. (2007), it is necessary to mention that the sample’s social awareness was likely heightened during this time, as two highly publicized cases of abduction and murder of young girls by previously convicted sex offenders took place in central Florida. Thus, the data were collected during a time of extensive media coverage regarding violent sex crimes against children.
Sample 2—Bristol and London, United Kingdom
The sample of survey participants is comprised of 140 random individuals who were walking near the Driver and Vehicle Licensing Agency in Bristol (England’s version of the US DMV Agency), students gathered or walking near the student center at the University of the West of England (UWE) in Bristol and individuals near the London Waterloo train station (Zgoba & Cowan, 2019). At each of the locations, the researchers solicited participation from the street. Based on the choice of the three locations, an effort was made to include a representative sample of survey respondents on income, age, gender and race (see Zgoba & Cowan, 2019 for a detailed description of the representativeness of the geographic area). Similar to Levenson et al. (2007), the surveys were completed over the course of multiple days, one day at each location, and a variety of individuals were approached to participate in the study. Survey takers were explained that their participation was completely voluntary, there was no incentivized gift or cash provided for participation and the survey was expected to take between 10 and 15 min of their time. The researchers stressed that it was important to know the publics’ opinions on the implementation of Sarah’s Law in the UK and issues related to sexual offenders. Those who agreed to complete the survey were read a consent statement and full consent was realized through their completion of the survey tool. Researchers were present to aid in any unclear questions or to provide a more detailed explanation. Surveys were conducted anonymously, as participants provided no identifying information; they did, however, fill out a portion of the survey on their demographics. After completion, the survey respondents returned the completed surveys into an unmarked envelope. Data collection for this sample of the study was completed with funding from the United States/United Kingdom Fulbright Commission.
Survey tool
This study utilized the same survey instrument as that employed by Levenson et al. (2007) in Florida, however, where necessary for the UK sample, minor amendments were made to the wording (see Appendix). Specifically, questions that asked US respondents about far reaching public notification strategies were amended to gauge whether UK residents would approve this level of public notification should it become available for use (Zgoba & Cowan, 2019). The utilized survey incorporated questions from previous research that also gauged public perceptions about sex offenders (Levenson & Cotter, 2005; Levenson et al., 2007; Zevitz & Farkas, 2000). The survey was comprised of five areas for assessment of beliefs including public perceptions of sexual offenders, public perception and familiarity with the laws, perception of the law’s effectiveness, support for various notification practices and the manner by which the public was apprised of the laws.
Analysis
For the current study, a two-part analysis was undertaken. The first step involved assessing bivariate relationships between survey participants’ demographics and perceptions. For this chi-square analysis, a total of three dependent variables were used, along with eight independent variables. The dependent variables were the perceived fairness of community notification laws, tolerance of a convicted sex offender living the subject’s neighborhood, and the belief that treatment of sex offenders will reduce subsequent offending. Each of the dependent variables were operationalized as a dichotomous measure (Fairness [n = 329], 0 = Not Fair, 1 = Fair; Tolerance [n = 268], 0 = Not Tolerant, 1 = Tolerant; Belief in Tx [333], 0 = Not Effective, 1 = Effective).
The Fairness measure was taken from a survey item which asked respondents to “Rate the law on fairness,” based on a Likert scale from 0 to 4 [0 = Unfair; 1 = Somewhat Fair; 2 = Mostly Fair; 3 = Fair; 4 = Completely Fair]. This item was subsequently recoded into a dichotomous measure for use within a logistic regression analysis with 0 = Not Fair and 1 = Fair [0,1 = 0; 2–4 = 1]. The Tolerance measure was based on a survey item that asked respondents, “How tolerant of sex offenders residing in your neighborhood are you?” [0 = Not tolerant at all; 1 = A little tolerant; 2 = Neutral; 3 = Tolerant most of the time; 4 = Very tolerant]. This item was also recoded into a dichotomous measure with 0 = Not Tolerant and 1 = Tolerant [0,1 = 0; 3–4 = 1]. The Belief in Treatment measure is based on a survey item that asked, “Would requiring sex offender treatment in prison prior to release be helpful in reducing child sexual abuse?”. This item was already dichotomized with response being either Yes [1] or No [0].
The independent variables are also all operationalized as dichotomous measures, including whether the respondent took the survey in the US or UK, gender (M/F), race (White/Non-White), educational attainment (Below HS/HS or Above), whether they have children (Yes/No), if they had a sexual experience under the age of 15 with an individual who was at least 5 years older (Yes/No), if the respondent knows someone who has experienced negative repercussions from notification laws (Yes/No), and whether the respondent believes in sex offender myths (Yes/No). For the last independent variable, the myth referred to was whether the respondent believes that sex offenders have a higher recidivism rate than the general offending population. Each of the above variables were also utilized in the logistic regression, with the addition of one independent variable—age. This variable was operationalized as a continuous measure of respondent age at the time of the survey.
Results
Two primary analyses were run to assess the research questions presented earlier. First, a bivariate analysis was conducted to assess the relationship between the three dependent variables and the participant’s geographic location (US/UK) and their descriptive variables. Secondly, a series of logistic regression tests were calculated predicting participants’ opinions of the SORN law’s fairness, their tolerance toward those convicted of sexual offenses, and belief in treatment. As can be seen in Table 2, individuals from the UK were more likely to indicate that they were tolerant of an individual convicted of a sex offense residing in their neighborhood than a respondent from the US (χ2 [1, N = 268] = 22.418, p < .001). Additionally, individuals who have children were less likely to indicate that they were tolerant of an individual convicted of a sexual offense residing in their neighborhood (χ2 [1, N = 261] = 9.024, p < .01). Respondents who believe in sex offender myths were also less likely to indicate they were tolerant of individuals who were convicted of a sexually based offense residing in their neighborhood (χ2 [1, N = 267] = 20.793, p < .05). Respondents with children were also more likely to report that the laws governing convicted sex offenders were fair, when compared with individuals that did not have children (χ2 [1, N = 320] = 8.174, p < .05). Individuals from the US were more likely to believe that mandatory treatment in prison would reduce subsequent child sexual abuse than a respondent from the UK (χ2 [1, N = 333] = 5.211, p < .05). When compared with non-White respondents, White participants were more likely to believe mandatory treatment in prison reduces subsequent child sexual abuse (χ2 [1, N = 320] = 7.078, p < .05).
Descriptive Statistics by Dependent Variable.
p < .05. **p < .01. ***p < .001.
Before discussing the individual results of the multivariate models, it should be noted that each of the three models were assessed for model fit using the Hosmer-Lemeshow test. All three of the models were not shown to have a statistically significant relationship, indicating a good fit of the data to the chosen models. As demonstrated in Table 3, the first multivariate analysis examines whether living in the US or UK predicts the likelihood of perceiving sex offender laws to be fair. The model explains 17.2% variability in the dependent variable (Nagelkerke R2 = .172). The variable that differentiated between the respondent country was not statistically significant, meaning there is no higher likelihood of an affirmative response based on residing country. Beyond this, three variables were found to be statistically significant. An individual who is tolerant of a sex offender residing in their neighborhood is approximately 4.2 times more likely to indicate that sex offender laws are fair (p < .01). In addition, individuals who do not know an offender who has experienced negative consequences are approximately 2.7 times more likely to believe the laws against convicted sex offenders are fair (p < .05). Likewise, individuals who believe in sex offender myths were also approximately 2.7 times more likely to respond in affirmation of the fairness of the current laws governing those convicted of sexual offenses.
Logistic Regression Predicting Fairness.
p < .05. **p < .01.
Table 4 examines whether living in the US or UK predicts the likelihood of respondents being tolerant toward a sex offender residing in their own neighborhood. Of the three multivariate models, this model had the most explanatory power (Nagelkerke R2 = .289). Findings reveal a UK resident is significantly less likely to be tolerant of a sex offender residing in their neighborhood than a respondent from the US (p < .05). Additionally, an individual who does not know an offender who has experienced negative consequences from current sex offender laws is approximately 4.5 times more likely to be tolerant of a sex offender residing in their neighborhood (p < .05). Moreover, a respondent who believes in sex offender myths is approximately 3.4 times more likely to be tolerant of a sex offender residing in their neighborhood (p < .05).
Logistic Regression Predicting Tolerance.
p < .05. **p < .01. ***p < .001.
Table 5 examines whether respondents believe that in-prison treatment for individuals convicted of sexual offenses will reduce subsequent child sexual abuse (Nagelkerke R2 = .111). Much like the previous model, residents in the UK are significantly less likely to believe treatment in prison will reduce child sexual abuse than respondents from the US (p < .01). There is no statistically significant relationship between respondent belief in sex offender myth and the belief in treatment of sex offenders. This aligns with expectations in that individuals who believe in the myth that sex offenders are more likely to recidivate than the general offending population also believe these offenders are less amendable to treatment within prison. Implications for these results are discussed below.
Logistic Regression Predicting Belief in Treatment.
*p < .05. **p < .01.
Discussion
The current study sought to answer several research questions, including, whether respondents in the US and UK were in favor of current sex offender policies, whether respondents desired a balance between the release of information and the privacy of those convicted, and whether geographic location influenced whether respondents were tolerant of sex offenders and their treatment. As noted, bivariate analyses indicated several statistically significant relationships between the three dependent variables for the current study. Notably, these analyses found that there were statistically significant differences between respondents’ tolerance of sex offenders residing in their neighborhood and the belief in treatment of convicted offenders with respondents from the US and UK. Additionally, respondents who had children were more likely to perceive current laws governing convicted sex offenders to be fair and were less likely to be tolerant of a convicted sex offender residing in their neighborhood. This finding, in particular, highlights the notion that sexually based offenses that target children are typically the most feared type of sexual offending (Kernsmith et al., 2009).
With regard to the favorability of current laws, respondents from both the US and UK were generally supportive of current sex offender laws in their respective countries. This finding mirrors that of previous research which asserts that the general public, in large part, supports SORN practices (Comartin et al., 2009; Zgoba & Cowan, 2019). Findings from the current study also indicate that there are no statistically significant differences between gender, race, education, whether the respondent has children, and whether the respondent had a sexual experience prior to the age of 15. Previous research also made note that, in terms of the perceived fairness of sex offender policies, there were no differences between male and female (Levenson et al., 2007). In alignment with previous research, Levenson et al. (2007) found no statistically significant differences in the perceived fairness of sex offender policy if an individual had been abused before the age of 15. While the current study did not explicitly ask whether respondents were abused before 15, the measure that asked whether the respondent had a sexual experience prior to the age of 15 was utilized as a proxy measure for this concept. Specifically, the lowest age of consent throughout the US and UK, is 16, thus lending credibility to the notion that affirmative responses to this item could indicate abuse had taken place. Levenson et al. (2007) also found no statistically significant differences between the perceived fairness of sex offender policy and whether respondents had children, which is consistent with the findings of the current study.
The findings from the current study are also supportive of the notion that sex offender policy holds a symbolic value for the public (Koon-Magnin, 2015; Socia & Harris, 2016). That is, believing in sex offender myth was predictive of both the perception of fairness of sex offender policy and whether respondents were tolerant of sex offenders residing in their neighborhoods. When discussing the symbolic nature of sex offender policy, this is suggestive that the general public perceives the laws to be fair, even without scientific evidence to support the effectiveness of the policy (Koon-Magnin, 2015; Levenson et al., 2007; Socia & Harris, 2016). One reason for this relationship would be the public’s general disdain toward sexually based offenses (Socia & Harris, 2016), especially, as noted, when these crimes are against children (Kernsmith et al., 2009). In essence, the public generally perceives sex offenders to be non-amendable to treatment, thus increasing the likelihood the public would be in support of SORN policy (King, 2019). This finding is also in alignment with implicit theory that was presented previously, specifically, the belief that sex offenders are not amendable to treatment follows an entity-based rationale (Harper & Bartels, 2017).
Contrary to our proposed hypothesis, and the presumption of the UK’s value on privacy, survey participants from the UK were less likely to be tolerant of individuals convicted of sexually based offenses residing in their neighborhood. Moreover, individuals from the UK were less likely to believe treatment for convicted sex offenders in prison would be helpful in reducing subsequent sexual abuse against children. While these results did not align with our proposed hypothesis, they are in alignment with the moral foundations theory discussed by Harper and Harris (2017). To elaborate, the main foundation of moral foundations theory is that individuals with differing moral compositions will have differing sentiments relating to offending (Harper & Harris, 2017). Thus, the differences noted between the US and the UK could be indicative of differing moral compositions of those sampled for the current study. Additionally, there is a possibility that UK citizens’ have been affected by the British media’s wide-reaching pedophilia coverage and belief in more punitive sanctions.
Apart from the study location, results indicate that study participants who do not know a convicted offender who has experienced negative repercussions from sex offender policy are more likely to perceive the current policy to be fair. This is in partial alignment with previous research that indicates family members of convicted offenders generally have more positive views of offenders (Connor & Tewksbury, 2017; Levenson & Cotter, 2005). Thus, the social proximity of individual to offender could be predictive of general perceptions of offenders and the policies that govern them. Another relationship identified from previous literature would be that of educational attainment and age (Brown et al., 2008). Specifically, Brown et al. (2008) found that younger individuals and those with higher educational attainment perceived convicted sex offenders to be less of a risk to the general public than older individuals and those with lesser educational attainment. These results would not be supported by the findings of the current study as both age and education were not predictive in any of the three models that were produced.
This study is not without limitations. Specifically, due to the sampling methodology, the results of the current study may not be generalizable to populations at large. Additionally, literature to date has discussed the use of the term “sex offender” as potentially problematic in that this term carries a heavy, negative stigma that could result in bias against these offenders (Socia & Harris, 2016). To elaborate, using an experimental design with matched pairs, Socia and Harris (2016) found that when these terms were used within items of a survey, respondents were more likely to support punitive policies, such as internet disclosure, residency restrictions, and social networking bans. Thus, the use of these terms within the current study survey instrumentation could have led participants to hold more negative views of the offenders in question. Additionally, there is a 10-year gap between the survey administration dates between the US and UK. This gap in administration could raise issues relating to history effects. Potentially minimizing this impact, however, would be that the administration of each respective survey was done relatively close, temporally speaking, to the implementation of these sex offender policies. Thus, the results of the current study would be situated in the context of timely reactions from citizens of each nation to these relevant policies. Future research could benefit from softening the language utilized to describe these individuals to assess community sentiment more accurately. Future research could also continue to examine differences across nations with regard to community sentiment as differing perceptions between countries may mean that enacted policies may also differ. Lastly, future research could benefit from the use of multi-item scales. The current study utilized primarily single-item measures which could impact the validity of the chosen constructs.
To lessen these overall negative perceptions, evidence-based policy could also be enacted, and public information campaigns can be implemented to inform the public of the realities behind sexually based re-offending (Levenson et al., 2007). A public information campaign that works to dispel myths of sexual offending will assist in garnering public support for evidence-based policy. To echo what previous literature has discussed, politicians will draft policy that satisfies public desire, thus public information campaigns will then allow for politicians to present policy more in line with empirical findings (Koon-Magnin, 2015; Levenson et al., 2007).
Supplemental Material
Appendix – Supplemental material for Do Views on Sex Offending Vary by Nationality? A Comparative Analysis of Community Sentiment Toward Sex Offense Legislation in the United States and United Kingdom
Supplemental material, Appendix for Do Views on Sex Offending Vary by Nationality? A Comparative Analysis of Community Sentiment Toward Sex Offense Legislation in the United States and United Kingdom by Devin Cowan, Kristen M. Zgoba, Rob T. Guerette and Jill S. Levenson in International Journal of Offender Therapy and Comparative Criminology
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported through the United States Fulbright Scholar Program.
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
