Abstract
It is presumed that public opposition to sex offender treatment derives from the “nothing works” philosophy. However, few studies have systematically unpacked perceptions about sex offender treatment. Drawing on national poll data, this study uses multinomial logistic regression to identify predictors of treatment views. Results indicate that misperceptions about sex crimes, particularly the belief in “stranger danger” and offense amplification, reduce support for treatment. Parents with children under age 17 were less likely to approve of treatment efforts, because they did not believe the treatment research. Separately, belief in high sex offender recidivism was associated with a greater likelihood of not sure responses about treatment. In contrast, desiring more information about how to prevent sexual offending increased support for interventions. Implications are discussed.
Introduction
Historically, public opinion about crime and justice issues has fallen under the “nothing works” and “get tough” mantra that has pervaded the United States since the fade of the rehabilitation era, especially for those who have committed sexually based crimes (Wright, 2014). But, as Cullen, Fisher, and Applegate (2000) observed, the perception of a punitive public has been driven largely by disproportionate empirical attention given toward understanding “get tough” attitudes as opposed to those aligned with rehabilitation. That assessment is apt 15 years later, particularly concerning attitudes and the language used toward sexual offenders as “monstrous others” and “sexual predators” (Lynch, 2002; Pickett, Mancini, & Mears, 2013). Sex offender public opinion research has been dominated by queries on punishment and community social control, but relatively little research has focused on unpacking public attitudes toward sex offender rehabilitation among the general public (see, for example, Brown, 1999; Katz Schiavone & Jeglic, 2009; Mears, Mancini, Gertz, & Bratton, 2008; Payne, Tewksbury, & Mustaine, 2010). This emphasis is troubling in light of research that shows specific types of treatment programs have the potential to reduce long- and short-term recidivism for medium and even high-risk sexual offenders (Duwe & Goldman, 2009; Oliver, Wong, & Nicholaichuk, 2008). To shift the focus from gauging punitive attitudes toward sexual offenders, this research evaluates community members’ attitudes toward sex offender treatment, an assessment of rehabilitative attitudes toward the sex offender population.
Given the disparate attention toward understanding punitive attitudes toward sexual offenders, studies that seek to identify predictors of treatment views are important for several reasons. First, research focusing on views of rehabilitation has the potential to provide a more balanced assessment of public opinion (Cullen et al., 2000). Second, according to scholars, public opinion has been a driving impetus in policy creation and additional sex offender restrictions (Meloy, 2014; Sample & Kadleck, 2008; Zgoba, 2004). Yet, as the current state of scholarship is centered on punitive attitudes, there is little to be gleaned—one way or another—about public perceptions regarding the amenability for sex offenders to be reformed through the use of treatment.
The current study contributes to discussions and debates concerning perceptions of sex offenders by addressing a key research gap. Specifically, the present investigation draws on national poll data to analyze public opinion on sexual offenders’ rehabilitation amenability. We first examine whether support for treatment is influenced by prominent myths about sex offenders—“stranger danger,” high sexual recidivism, perceptions of offense amplification, and belief in the continuity of sex offending. Second, we evaluate individuals who are “not sure” of sex offender treatment. Virtually no studies attend to those individuals who are uncertain about sex offender policy or rehabilitative developments. Public views are nuanced and identifying predictors of uncertain responses could reveal important insights regarding those who are undecided concerning the treatment amenability of sex offenders.
Background
Sex Offender Treatment
Generally, research shows that treated offenders are significantly less likely to commit additional crime (Duwe & Goldman, 2009; Oliver et al., 2008). A comprehensive meta-analysis of more than 20 studies evaluated as “high quality” demonstrated that sex offenders who received a treatment intervention sexually recidivated less, at approximately 11% recidivism, versus untreated offenders, at approximately 19% recidivism (Hanson, Bourgon, Helmus, & Hodgson, 2009). A similar pattern emerged for general recidivism; approximately 32% of treated sex offenders committed another non-sex offense compared with approximately 48% of offenders who received no intervention. Comparable findings have been reported in other meta-analyses. For example, Lösel and Schmucker (2005) analyzed 69 studies involving more than 20,000 offenders and found that compared with controls, treated offenders showed 37% less sexual recidivism and 31% less general recidivism. Their study is distinct from prior investigations as it also evaluated the impact of the type of treatment on sexual recidivism. Cognitive behavioral therapy and organic interventions, such as surgical and chemical castration, were shown to be the most effective interventions.
Public Perceptions on Sex Offender Treatment
How does the public perceive treatment efforts for sex offenders? Providing an answer is difficult given that a majority of public opinion studies have centered on examining approval for various sex offender laws—such as registries and community notification (Levenson, Brannon, Fortney, & Baker, 2007), residence restrictions (Mancini, Shields, Mears, & Beaver, 2010), chemical castration (Comartin, Kernsmith, & Kernsmith, 2009), child pornography legislation (Mears et al., 2008), and civil commitment (Pickett, Mears, Stewart, & Gertz, 2013). Far fewer studies have centered on unpacking public perceptions 1 of treatment and rehabilitation for sex offenders with two exceptions.
Payne et al. (2010) sampled 746 residents from Norfolk and Virginia Beach, Virginia, to investigate whether demographic, victimization, and community-level factors were predictive of public views about sex offender rehabilitation. Although few significant predictors emerged, being a victim of corporal punishment as a child and using force against a partner were associated with the perception that rehabilitation was not possible for sex offenders. In addition, minority respondents compared with Whites were significantly more pessimistic regarding rehabilitation for those convicted of sex crimes.
The study also highlighted divides in public support for rehabilitation. Fifty-two percent of the sample strongly agreed or agreed that it was “impossible to rehabilitate or reform a sex offender,” whereas 36% strongly disagreed to disagreed with that statement. A sizable minority, 12%, were unsure about rehabilitation efforts. The study was not originally designed to capture views about sex offenders beyond rehabilitation and so could not speak to the relationship between sex offender myth endorsement, such as “stranger danger,” and its influence on rehabilitation perceptions. As a result, Payne et al. (2010) called for future work to focus on capturing “broader factors” (p. 584) possibly related to sex crime rehabilitation views—such as these misperceptions about sexual offenders.
In a more recent study, Pickett, Mancini et al. (2013) looked at stereotype endorsement of sex offenders and its influence on assessments of sex offender treatment. They found that two perceptions—sex offenders are “unreformable” and have an “immoral character”—were associated with reduced treatment support. The study however did not evaluate predictors of “unsure” respondents and did not specifically examine other myths such as continuity in offending over time. In addition, it did not control for media influence on perceptions that may serve as an important control due to the media’s attention on atypical sex crimes, such as sexual assaults that end in murder (Federal Bureau of Investigation, 2013; Fox, 2013; Lancaster, 2011). The authors recommend that future work test a range of competing theories on diverse samples to continue to assess perceptions of sex offender treatment.
In short, although numerous studies examining public attitudes toward individuals convicted of sex crimes have been published, only a small handful have investigated treatment perceptions. The current study extends the focus of this underdeveloped scholarship on public perceptions of sex offender treatment amenability. Drawing on national public opinion data, we evaluate whether myth endorsement—stranger danger, high recidivism, amplification of deviance, and continuity in offending—affects treatment perceptions. In contrast to prior scholarship, we also assess “unsure” responses. This line of inquiry can inform efforts to educate the public regarding treatment interventions for sex offenders with the goal to breakdown misperceptions that sex offenders are “unreformable.”
Current Study
Public opinion toward crime is complex (Cullen, Pealer, Fisher, Applegate, & Santana, 2011; Mancini et al., 2010; Payne et al., 2010; Pickett, Mancini et al., 2013). One of these nuances involves the realization that although Americans strongly favor punitive measures to address crime, they do not simultaneously discount efforts to rehabilitate and treat offenders (Cullen et al., 2011). At the same time, perceptions related to the nature and extent of crime and offending likely shape and contour public views toward rehabilitation and treatment (Cullen et al., 2000). This point bears emphasis when the focus is on sex offending given the numerous misperceptions surrounding sex crimes, such as the myth of “stranger danger,” the perception of abnormally high sexual recidivism, relative to other serious offenders, offense amplification, and continuity in offending (e.g., Levenson et al., 2007; Logan, 2009).
Despite the image of the predatory stranger offender so often propagated by the media (Dowler, 2006), most sex offenders—nearly 75%—are known to their victims (Greenfeld, 1997). Separately, although heterogeneity exists in offending patterns, the vast majority of individuals convicted of a sex crime will not reoffend; according to experts, the rate of sexual recidivism among those with a prior sex crime conviction is relatively low (Durose, Langan, & Schmitt, 2003; Hanson & Bussière, 1998; Sample & Bray, 2003). Even so, it is a perception consistently linked with sex offenders (Quinn, Forsyth, & Mullen-Quinn, 2004). The related impression, that sex offenders inevitably progress to more serious and predatory sex offenses, also appears prominently communicated in news accounts and serves as the underpinning for “get tough” sex offender laws (Jenkins, 1998). Still though, it is one without solid empirical backing (Sample, 2006).
Finally, there is the myth of “continuity” among sex offenders. Although a generally understudied aspect of sex offending, at least three published studies indicate little support for the assumption that juvenile sex offenses are predictive of later adult sex crime. Zimring, Jennings, Piquero, and Hays (2009) analyzed the Second Philadelphia Cohort Data (N = 13,160 males and N = 14,000 females) and reported that 92% of adult sex offenders had no prior history of sex offending as juveniles (see earlier, Zimring, Piquero, & Jennings, 2007). In addition, in a separate investigation, Piquero, Farrington, Jennings, Diamond, and Craig (2012) analyzed a cohort of South London males participating in the Cambridge Study in Delinquent Development (CSDD; N = 411). Their results demonstrated strikingly similar findings to Zimring et al. (2009, 2007)—specifically, the researchers concluded that “there was no continuity in sex offending from the juvenile to adult periods and very few recidivist sex offenders” (p. 412, emphasis added).
Ascribing to such myths is likely consequential in shaping public attitudes. Scholars have argued that they activate certain stereotypes among the public (Harris & Socia, 2014); that is, their endorsement makes it difficult for the public to extend any positive attributes to sex offenders, including, as we argue, their potential to desist as a result of treatment interventions (Mancini & Pickett, 2014). Under that theoretical backdrop, we ask the following research questions:
In particular, we test four hypotheses concerning treatment perceptions.
Recall that another avenue of inquiry of our study is to explore predictors of “unsure” responses. Most prior research has excluded this category. As a result, we have little theoretical basis to anticipate what factors may be associated with uncertain views. For this reason, we do not present specific hypotheses regarding this group but rather simply test our second research question concerning uncertain responses.
Data and Methods
Sample
The study draws on responses from a national poll (N = 1,005) commissioned by the federally funded Center for Sex Offender Management (CSOM). The CSOM disseminates information about sex crime and victimization to the public and policymakers. The survey was specifically designed to capture public opinion concerning knowledge of sex crimes, beliefs about sex offenders, and policy perceptions. In particular, it includes measures that permit us to tap into misperceptions of sex crimes and policy preferences, including treatment support. Our analysis is one of the first to capitalize on this data source. This approach advances scholarship given that some prior related investigations have relied on surveys that were not originally intended to capture perceptions of sex crimes and offenders (e.g., Mancini et al., 2010; Payne et al., 2010). Individuals with expertise in sex offending and public opinion research assisted with the creation of the survey items and pre-testing. An internationally recognized polling firm, Zogby International, administered the poll to a randomly selected national sample, relying on techniques that ensured the selection probabilities were proportional to population size within area codes and exchanges. The margin of error of the poll was calculated to be plus or minus 3.5 percentage points.
Measures
Dependent variable
The dependent variable of interest involves treatment perceptions. In particular, the survey has a unique measure of treatment support. The specific item used in the survey asked, “Some research demonstrates that treatment designed specifically to prevent sexual reoffending can be effective. Given this, which of the following statements do you agree with most?” Response categories were designed to tap nuances in support views: “1 = I support treatment for sex offenders, if the research shows it can be effective in preventing repeat sex offenses”; “2 = I do not support treatment for sex offenders, even if the research shows that it can be effective in preventing repeat sex offenses”; “3 = I do not believe the research about treatment for sex offenders”; and “4 = not sure/other.” Given that few studies have evaluated “unsure” responses, we chose to retain the nominal measurement of these categories.
The use of this measure improves on prior research and follows the trend in capital punishment public opinion literature (McGarrell & Sandys, 1996; Radelet & Borg, 2000), where additional context is given regarding the criminal sanction (e.g., its costs, potential unintended effects, deterrence impact). The end result is that respondents can better evaluate their levels of support in light of factual information regarding the specific initiative. In addition, the response categories of the treatment measure allow for investigation into the precise reasons for non-support (e.g., skepticism of research findings). Finally, by including “unsure” individuals, it is possible to study factors associated with individuals who have not yet formed an opinion about treatment.
Independent variables
Given prior accounts identifying prominent misperceptions about sex crimes, the study examined four independent variables of interest: “stranger danger,” the perception of substantial sexual recidivism, offense amplification, and continuity in offending. In measuring “stranger danger” endorsement, the survey asked, “Which of the following statements about sex offenders who victimize children do you think is most true?” Categories included, “1 = Most sex offenders victimize children who are well known or related to them,” “2 = Most sex offenders victimize children who they do not know,” or “3 = Most sex offenders victimize children who they have met, but do not know well.” The first and third categories (measuring familiar/acquaintance perceptions) were combined and coded as “0.” The second choice, emphasizing unknown perpetrators, was categorized as “1.” Unsure or don’t know responses were dropped.
Assessments of sexual recidivism were gauged continuously using this item: “What percentage of sex offenders do you think repeat their crimes after they are convicted?” Response options were “1 = less than 25%,” “2 = between 25% and 50%,” “3 = between 50% and 75%,” or “4 = 75% or more.” Uncertain responses were excluded.
Offense amplification, in contrast, was operationalized in the following manner: “Which of the following statements do you think is most true about sex offenders who reoffend?” The response choices were as follows: “1 = If they reoffend, sex offenders are most likely to commit a non-violent, non-sexual offense”; “2 = If they reoffend, sex offenders are most likely to commit another, similar sex offense”; or “3 = If they reoffend, sex offenders are most likely to commit an offense that is more serious and violent.” The three choices were dichotomized with the latter category, indicative of stronger amplification views, coded as “1” and other responses as “0.” Unsure responses were not included.
Finally, perceptions of continuity in sex offending were measured continuously by asking the following: “Which of the following statements about teenagers who commit sex offenses do you think is most true?” Respondents could indicate, “1 = less than 25% of these teenagers will go on to commit sex offenses as adults,” “2 = between 25% and 50% of these teenagers will go on to commit sex offenses as adults,” “3 = between 50% and 75% of these teenagers will go on to commit sex offenses as adults,” and “4 = more than 75% of these teenagers will go on to commit sex offenses as adults.” Respondents who admitted being “unsure” were dropped from analysis.
Controls
The analysis also accounted for the effects of two specific controls that may confound the myth–treatment support relationship. The survey instrument included an item tapping the extent to which Americans rely on the media to form opinions about sex crime and offenders. Some have argued that news media coverage (Dowler, 2006) has a tendency to emphasize bizarre and atypical sex crime cases. The survey asked, “From what source have you received most of your knowledge about sex offenders?” Response categories were “news media,” “your state’s sex offender registry,” “family members,” “community members,” “professionals who work in the areas of sexual assault, criminal justice, and related fields,” or “your own search for information on the Internet.” For the media reliance variable, those who reported “news media” were coded as “1” and all others as “0.” In addition, respondents were also asked “To what extent do you agree or disagree with the following statement . . . I want more information than I have now about the ways that sex offending can be prevented in my own community.” It follows that individuals who desire greater knowledge about sex crimes are distinct from those who feel they have a solid understanding of the issue. The variable, want more information, was coded as follows: 1 = strongly agree, 2 = somewhat agree, 3 = somewhat disagree, and 4 = strongly disagree.
Beyond these two specific confounders, we included several other controls shown to be important in prior sex crime research (Payne et al., 2010; Pickett, Mancini et al., 2013). Age was measured continuously (range = 19-91). In the survey, education was recorded as follows: 1 = less than high school, 2 = high school graduate, 3 = some college, and 4 = college degree or graduate degree. Income was estimated by asking participants to report their annual income coded as 1 = less than US$25,000, 2 = US$25,000 to $35,000, 3 = US$35,001 to US$50,000, 4 = US$50,001 to US$75,000, 5 = US$75,001 to $100,000, and 6 = more than US$100,000. Conservatism was measured continuously from 1 = extremely liberal to 9 = extremely conservative. The survey asked respondents if they lived in a city, suburb, or rural area. Dichotomized variables were created and “suburb” was used as the omitted comparison category. Marital status was measured dichotomously, “0 = not married” and “1 = married.” Three separate dummy variables were used to capture race: “White,” “Black,” and “Other.” “Other” was used as the omitted comparison category. Parental status was also coded as two separate categories, “0 = not a parent of child younger than 17” and “1 = parent of child 17 or younger.” Finally, sex was recorded by the interviewer, “1 = female” and “0 = male.”
Analytic Plan
To answer our research questions and corresponding hypotheses, we used multinomial logistic regression. Multinomial logistic regression was the preferred method of analysis because the outcome variable, or dependent variable, is unordered. With this type of regression model, the probability of membership in each of the other response categories was compared with the probability of membership in the reference, or base, category (Long & Freese, 2006). For this analysis, “I support treatment for sex offenders, if the research shows it can be effective in preventing repeat sex offenses” is the reference, or base, category. Because the multinomial logistic regression model produces log odds that have no substantive meaning, we transformed the log odds into odds ratios (ORs) for ease of interpretation. In addition, because of the complex output of the model, we discuss significance between different outcome categories. These results 2 are not included in the tables, as the tables only report results for the above base category.
For clarity, ORs are only presented for variables that are statistically significant in comparison with the reference category.
Results
Going against the assumption that the public is overwhelmingly punitive, survey results shown in Table 1 indicate that nearly three out of four people support treatment for sex offenders. Even so, more than 25% do not extend their support to treatment efforts. Of this proportion, 8% oppose treatment even if research demonstrates that it can be effective in preventing sex crime. Nearly 18% report not believing research related to effectiveness and so, for this reason, do not support treatment efforts. Two percent of the public are unsure about treatment.
Descriptive Statistics (N = 658).
Note. SD = standard deviation.
Moving beyond descriptive differences, in line with our hypotheses, inspection of the multivariate analysis results in Table 2 indicates that myth endorsement was associated with pessimism toward sex offender treatment efforts. Hypothesis 1, that endorsement of stranger danger will be associated with reduced treatment support, was supported. Believing that sex offenders victimize strangers increased the odds of not supporting treatment by 2.31 times relative to supporting treatment, holding all other variables constant (p ≤ .05). This pattern also extended to those who do not believe the research on treatment. Believing that sex offenders victimize strangers increased the odds of not supporting treatment by 3.48 times relative to not supporting treatment due to non-belief in the treatment research, holding all other variables constant (p ≤ .05).
Multinomial Regression of Perceptions of Support for Sex Offender Treatment, Reference Category: Support (N = 658).
Note. SE = standard error; OR = odds ratio.
p < .05. **p < .01. ***p < .001; statistical tests use “supports” as the reference category.
Hypothesis 2, that perceiving sex offenders to have high sexual recidivism will decrease treatments, was not supported. Unexpectedly, buying into the high recidivism myth was associated with a significantly higher probability of being “unsure” regarding the benefits of treatment. Believing that sex offenders have high recidivism rates increased the odds of being unsure by 2.82 times relative to supporting treatment, holding all other variables constant (p ≤ .05).
Hypothesis 3, that individuals who believe in offense amplification will have reduced odds of supporting sex offender treatment, was supported. Believing that sex offenders progress in their offending over time increased the odds of not supporting treatment due to non-belief in the treatment research by 2.17 and 5.18 relative to supporting treatment or being unsure, holding all other variables constant (p ≤ .001, p ≤ .05). Said differently, members of the public who believed in sex offense amplification had increased odds of not believing the research on treatment efficacy and therefore not supporting treatment relative to those who support treatment or are unsure.
The fourth hypothesis proposed that those who endorsed the “continuity in offending” perception, that juvenile sex offenders will go on to sexually offend in adulthood, would be significantly less likely to favor treatment. It was not supported. There were no significant results pertaining to this myth.
Two control variables were significant predictors of whether there was treatment support: parental status and the desire for more information on preventing sexual offenses. Parents with children below the age of 17 had a higher likelihood of not supporting treatment, because they did not believe the treatment efficacy research. Specifically, the odds of not believing research demonstrating treatment efficacy and therefore not supporting treatment relative to supporting treatment were 1.84 times higher for individuals with children below the age of 17 (p ≤ .05). If individuals desired more information about how sex offending can be prevented in their community, the likelihood of supporting treatment increased. The odds of supporting treatment relative to not supporting treatment increased by 2.19 if more information was desired about how to prevent sex offending, holding all other variables constant (p ≤ .05).
Discussion
Nearly four decades ago, Martinson’s “Nothing Works” doctrine began to shape what scholars have designated the “punitive era” in the United States. (Phelps, 2011, p. 33; also Cullen et al., 2000; Lynch, 2000). As part of this shift, “get tough” reforms—such as mass incarceration, penal austerity, and “invisible” post-incarceration sanctions—were favored over efforts to treat and rehabilitate offenders (Logan, 2009). Although research indicates that treatment can be effective in reforming offenders, the perception that the public is unreceptive to treatment interventions pervades crime policy discussions (e.g., Cullen et al., 2000). We unpacked this latter assumption about treatment by focusing on a criminal population deemed to be the “worst of the worst” in the eyes of the public (e.g., Pickett, Mancini et al., 2013)—sex offenders. Overall, results support this view, but also offer evidence to the contrary.
With respect to the latter, nearly three out of four Americans indicated their approval for treatment efforts for those convicted of sex crimes. Notably, the percentage for support is higher than those reported in previous work (e.g., Payne et al., 2010). The disparity is likely driven by our unique operationalization of treatment support. Our version specified the extent of approval if research demonstrates the efficacy of treatment. We took this approach for two reasons. First, building off capital punishment support literature (McGarrell & Sandys, 1996; Radelet & Borg, 2000), prefacing with this information provides context for respondents to better evaluate their levels of support. As mentioned earlier, prior scholarship has indeed demonstrated that treated sex offenders are significantly less likely to reoffend sexually and generally (i.e., commit non-sex crimes) than those offenders who receive no treatment (Duwe & Goldman, 2009; Hanson et al., 2009; Lösel & Schmucker, 2005; Oliver et al., 2008). Second, it would seem prudent to understand nuances in non-support and unsure responses. Most prior research has not explored the specific rationales underlying non-support for rehabilitation and treatment, and none that we could identify tapped into “unsure” responses. For this reason, our dependent measure specified the conditions under which the public would not favor treatment (i.e., due to disbelief of research indicating efficacy or simply not endorsing support, even if the research is credible) and also retained the “unsure” category.
Indeed, a non-trivial percentage of Americans are unsupportive or unsure. In particular, 8% do not endorse their support, even if research demonstrates treatment efficacy. Almost 17% express not believing the research, and so, on this basis, do not extend their support. About 2% of the public are uncertain about treatment efforts—remaining unsure of sex offenders’ ability to be rehabilitated through treatment. In short, a much smaller proportion of the American public, approximately one quarter, are dubious of treatment efforts, whereas, a substantial majority indicate support for treatment, given prior research demonstrating its efficacy in reforming sex offenders.
Still though, there is evidence that a punitive public exists if punitive attitudes are embedded in myths about sexual offending. In line with prior research examining punitive attitudes (Pickett, Mancini et al., 2013), myth endorsement—particularly believing in stranger danger and also offense amplification—was associated with a reduced likelihood of supporting treatment interventions for sex offenders. Concerning the continuity in offending myth, the lack of significant findings may derive from the measurement of continuity. Few prior studies existed to guide our conceptualization of continuity and so perhaps the measure is incomplete. Alternatively, perhaps this myth affects other perceptions of sex offenders but not treatment support.
Pertaining to stranger danger and offense amplification, a possible explanation of these myths influencing perceptions of treatment effectiveness in light of research evidence derives from the application of subconscious processes or heuristics (e.g., Harris & Socia, 2014). Ascribing to distorted perceptions of the nature of sex offending obscures affording any positive traits to those who have committed sex offenses in the past—including the potential for reform and desistance. In particular, the stranger danger myth was associated with a greater likelihood of indicating no support for treatment even if the research is credible. Separately, endorsing the offense progression perception (i.e., believing sex offenders amplify sexual deviance over time) was linked with not believing prior research and so not extending support to treatment efforts. In contrast, evaluating sex offenders to be at high risk of recommitting sex crime was the most salient predictor of unsure beliefs. Collectively, these findings indicate the nuanced effects of myth endorsement—certain myths appear to shape particular beliefs concerning treatment. It may be that views about sex offenders, as we found, are more likely to be shaped by what Payne and colleagues (2010) have characterized as “irrational cognitive processes” (p. 584), or a complementary perspective supported with this research, a long-standing and publicly engrained endorsement of sex offender myths.
Only one social and demographic factor, having children, was associated with decreased support, specifically, a disbelief of research indicating efficacy of treatment. Notably, other research examining sex crime opinion have revealed a similar pattern, namely observing few social and demographic divides of opinion related to sex crime policy (Mancini & Mears, 2010; Payne et al., 2010). Parental status divides may derive from a number of factors. One relevant explanation is that parents, given their obligation to ensure their children’s well-being and safety, may be especially unwilling to believe research suggesting that sex offender rehabilitation is effective. In short, they may have a felt need to discount such evidence as the concern for their children tempers any recognition that sex offenders can be effectively treated to the extent they no longer pose threats.
Limitations
Although this study reveals new insights concerning public perceptions of sex offender treatment, it is not without its limitations. One involves generalizability concerns. The study, conducted by the federally funded CSOM adhered to best practices in the administration of the survey; the instrument was also pre-tested by experts in the field and efforts were made to garner a nationally representative sample. Even so, as is typical with survey sampling, women and those who are older were overrepresented. Therefore, generalizations from these findings may be somewhat limited in scope. In addition, we did not have the opportunity to follow-up with respondents to probe why they supported treatment, why they did not support treatment, why they were skeptical of the research demonstrating treatment efficacy, or why they were unsure. Of particular interest is gaining more insight to those who do not support treatment because they do not believe scientific literature that treatment has the potential to rehabilitate, or said differently, to decrease the risk of recidivism, for sex offenders. Future research should investigate these nuanced aspects of public perceptions pertaining to treatment, especially if a segment of the public is resistant to “buying into” sex offender treatment that is supported empirically to decrease recidivism.
Conclusion
When evaluating these research findings, and findings on public opinion on sex offenders and sex offender policy as a whole, there is support that the public’s perceptions about these offenders and the laws used to control or rehabilitate them are built upon faulty perceptions. Therefore, myths play a powerful role in not only punitive crime policy but also rehabilitative crime policy in the sense that they may be derived from “irrational cognitive processes” (Harris & Socia, 2014). Although evidence is presented here that a sizable majority of the public wants more information about sex offenders, the question remains how this information will be processed and understood and if it will influence, what seems to be, negative stagnant views on sex offender management and crime policy that are based on distorted perceptions.
To continue to understand rehabilitative views on sex crime policy and about sex offenders themselves, carefully designed surveys and polls that are national in scope and consistently administered are sorely needed. It is rare for a survey such as this one, specifically designed to capture perceptions about sex offenders and sex offender policy, to be administered to the public. We have argued that it is superior to prior investigations given its scope and inclusion of nuanced measures of treatment as well as theoretically relevant predictors of these views. Even still, it was administered at one point in time and thus cannot capture how views might change over time or in response to celebrated cases or changes in policy. The same recommendation holds for understanding practitioner perceptions of sex offender treatment. Correctional officers, community corrections officials, and other professionals who work with and supervise sex offenders may differ from the public in their views about treatment. There may even be nuances in judgments across these groups (e.g., prison officers vs. probation officers). Some work has recently begun to tap into views among these understudied populations (generally, Mustaine, Tewksbury, Connor, & Payne, 2015; Tewksbury, Mustaine, & Payne, 2012). However, no scholarship that we are aware of has systematically compared and contrasted treatment views across these professional groups and the general public.
Ideally, future research will continue to investigate treatment effects for sex offenders. Although sex offenders who complete treatment pose less of a threat to public safety than those who receive no treatment, it is equally true that some treatment modalities are less effective than others (Hanson et al., 2009). This finding thus suggests the need for robust research designs—carefully controlled meta-analyses and experiments—that evaluate which treatment methods are most effective in reducing sexual recidivism and under which conditions (e.g., prison vs. community treatment, across offender “types”). This suggestion leads to another relevant implication. To the extent that treatment effects vary by intervention, it would seem prudent to assess which treatments the public endorses the most and least support for and why these perceptions might vary. Hormonal interventions and cognitive behavioral therapy, for example, have shown significant promise in reducing sexual recidivism (Lösel, & Schmucker, 2005). However, virtually no research that we identified has examined treatment perceptions by intervention.
Finally, concluding comments surround policy. Public awareness and educational campaigns to advance understanding of the reality of sex offending, although often cited, cannot be discounted. The public has clearly had tremendous influence on the national proliferation of “get tough” sex crime laws and restrictions (Meloy, Curtis, & Boatwright, 2013). The current study, which found nearly three out of four Americans expressing treatment support for sex offenders, suggests that it is possible for Americans to simultaneously support both punitive and rehabilitative efforts. The challenge then for policymakers is to disseminate information about sex crimes in a fair and balanced manner. Under this perspective, social media campaigns, polls conducted by agencies assessing levels of public knowledge and awareness, and evaluations of such efforts are worthwhile investments to consider and implement.
Footnotes
Authors’ Note
The opinions, findings, and conclusions expressed in this publication are those of the authors and do not necessarily reflect those of the Center for Sex Offender Management or Zogby International.
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) received no financial support for the research, authorship, and/or publication of this article.