Richard Tewksbury, Ph.D.
Department of Justice Administration, University of Louisville
RECENTLY CHANGES and innovations in public policies and sentencing structures have extended criminal sanctions beyond the immediate needs of offenders, victims, and society in general. One clear example of reaching beyond immediate needs and extending the form, length and consequences of sentencing is the use of community notification and registration of sex offenders. The present research identifies how such practices have created unintended and potentially serious collateral consequences for convicted sex offenders, with a special focus on female sex offenders.
Research on sex offenders has historically focused, almost exclusively, on male offenders. Studies of female sex offenders are relatively rare, at least in part because most known sex offenders are male. Females comprise only 1.2 percent of arrests for rape and 8.0 percent of arrests for all other sex offenses (U.S. Department of Justice, 2002). Women who are sex offenders are most often convicted of offenses against children (Faller, 1987; Lewis and Stanley, 2000; Rosencrans, 1997; Vandiver and Walker, 2002) or low level felonies (other than rape); because of the latter, female sex offenders are often considered "less serious" sex offenders (Hetherton, 1999).
In the criminological literature, attention to female sex offenders is rare, with the first articles appearing in the 1980s. Although recognized, female sex offending is not only often considered less serious than that of males (Hetherton, 1999), but is also acknowledged as possibly less likely to be detected or reported (Berliner and Barbieri, 1984; Johnson and Shrier, 1987).
Recently, much attention has been focused on community notification and registration of (male and female) sex offenders. To date, there are few studies of sex offender registries. The existing assessments are of four varieties: overviews and "profiles" of the population of registered sex offenders, evaluations of recidivism rates for registered sex offenders, examinations of the accuracy of information in the registries, and assessments of the experience of registration from the point of view of offenders. However, only one study to date has focused on female registered sex offenders.
The Bureau of Justice Statistics (Adams, 2002) reports that a total of 386,000 convicted sex offenders were registered in 49 states1 and the District of Columbia in 2001. The use of sex offender registries has grown rapidly; the 2001 total represents a 46.2 percent increase over the registered offender population in 1998. However, statistics on the sex of registered sex offenders is not available. Individual studies have reported that females comprise 0.8 percent of registered sex offenders in Hawaii (Szymkowiak and Fraser, 2002), 3 percent of sex offenders in Iowa, 2.4 percent of registered sex offenders in Arkansas, and in the present research, 2.7 percent and 2 percent of registered sex offenders in Kentucky and Indiana respectively.
The most comprehensive assessment of registered sex offenders to date is the overview of 1,458 offenders on the Hawaii registry (Szymkowiak and Fraser, 2002). The demographic assessment shows that the "average" registered sex offender in Hawaii is between the ages of 40 and 49, lives in the greater Honolulu metropolitan area, has a criminal record of between one and five (typically non-violent) felonies (and a similar number of misdemeanor convictions), and has only one sex offense conviction. In Iowa, the "typical" registered sex offender was a white male with a median age of 31.1 at time of conviction; a majority (57.9 percent) have a previous criminal (but not necessarily sexual) conviction (Adkins, Huff, and Stageberg, 2000).
The only assessment to date of female registered sex offenders (Vandiver and Walker, 2002) focused on identifying a typology of offending patterns, including victim and offender characteristics. This review of official records revealed that these offenders were almost all white, with a mean age of 31 at the time of their first sex offense. Females comprised a slight majority (55 percent) of the victims of these female sex offenders. All had juvenile victims and less serious (if any) criminal records than their male counterparts. Vandiver and Walker (2002) were not able to gather complete data on how many of their sample of female sex offenders were related to their victims; from the data available, though, 94 percent of the victims were related to the offender.
The Iowa study also assessed recidivism of registered sex offenders over a 4.3 year period and showed "mixed effects on recidivism rates" when comparing sex offenders that were and were not (due to a different time period in question) required to register (Adkins, et al., 2000: 19). Registered sex offenders had a sex offense recidivism rate of 3.0 percent; the comparison group had a recidivism rate of 3.5 percent; total recidivism (for all offenses) was 24.5 percent for the registered offenders and 33.3 percent for the comparison group.
All of the examples of research focused on identifying the characteristics and recidivism of offenders listed on sex offender registries is superficial and macro in nature.
As a third focus, Tewksbury (2002) examined a sample of 537 sex offender listings on the Kentucky Sex Offender Registry in 2001, examining whether offenders' listed information was complete and accurate. Results showed that while most offenders' information was provided, the registry showed a significant degree of missing data. One in twelve (8.2 percent) registrants had "unknown" addresses listed. The problem of accuracy was most acute for sex offenders listed as residing in an urban county: 10.5 percent had "unknown" addresses, 10.5 percent listed addresses that turned out to be commercial locations and 5.4 percent had addresses that did not exist.
Finally, a fourth focus of research on sex offender registries has examined the experiences of registered sex offenders, examining the collateral consequences of registration. Focusing on registered sex offenders in Kentucky, Tewksbury (in press) found that serious social consequences were reported by more than one in four registrants. Specifically, at least one-quarter of registrants reported having received harassing/threatening mail and telephone calls, losing a job, being denied a promotion at work, losing (or being unable to obtain) a place to live, being treated rudely in public, being harassed/threatened in person, and losing at least one friend. These experiences were more common for registrants from non-metropolitan communities, and (surprisingly) less common for offenders with child victims. Tewksbury (in press) further suggested that child-victimizing sex offenders were able to more closely control information about their status as a sex offender, and consequently limit the collateral consequences experienced. One shortcoming of this research, however, is that females comprised only 7.5 percent of the sample, rendering an assessment of female registered sex offenders impossible.
This shortcoming provides the impetus for the current research. With an exclusive focus on female registered sex offenders, the present study examines if and to what degree female registered sex offenders perceive they are known in their community as sex offenders; what consequences are experienced as a result of being listed on the publicly accessible sex offender registry; and registrants' attitudes regarding the registration process.
Method
Data for this study were collected through a mailed, anonymous questionnaire sent to all female offenders listed on the Kentucky Sex Offender Registry (http://kspsor.state.ky.us) and Indiana Sex and Violent Offender Registry (http://www.indianasheriffs.org/default.asp). Once identified, sample members' addresses were recorded from their individual registry pages. All sample members were mailed a cover letter, informed consent explanation, survey, and postage-paid return envelope. The Human Studies Protection Program office at the author's university reviewed all materials. Data collection was conducted in May, 2004.
Sample
A review of all entries on the Indiana and Kentucky registries reveals a total of 227 females. The Kentucky registry had 97 females listed among the total of 3,586 individuals. Females accounted for 130 of 6,407 registrations on the Indiana registry. This means that 2.7 percent of the registrants in Kentucky and 2.0 percent in Indiana are female.
A total of 40 completed and usable surveys were obtained,2 for a response rate of 20.5 percent. While this is not a very high response rate,3 this needs to be understood as a difficult to access population. Previous research looking at registrants has relied on small samples (2.4 percent, Vandiver and Walker, 2002; 14.3 percent, Tewksbury, in press) or has used only officially recorded data, avoiding collection of data directly from registrants (Adkins, et al., 2000; Szymkowiak and Fraser, 2002; Tewksbury, 2002). And, as Vandiver and Walker (2002:286) state, "the number of subjects in female sex offender research has consistently remained low,… The number of subjects in female sex offender literature has been as low as 2 (Peluso and Putnam, 1996) and as high as 93 (Rosencrans, 1997)." Additionally, studies of sex offenders in general have almost always collected data either from offenders who are incarcerated or in treatment, or researchers have collected data from professionals working with sex offenders (treatment providers, probation officers, etc.). Only two studies have gathered data directly from sex offenders in the community (Sack and Mason, 1980; Tewksbury, in press), and both have samples of 112.
Table 1 presents the demographic and registration information for the respondents.
Instrument
The data collection instrument was designed specifically for this study. The instrument is a four-page questionnaire containing 35 closed-ended items. The items assess demographics, offenses characteristics, questions about whether, by whom, and how often the offender is recognized as a registered sex offender, and attitudes regarding registries in general and the registration experience specifically.
The dependent variables for this analysis are self-reports by registered sex offenders regarding ten different negative consequences they may have experienced (loss of a job, denial of promotion, loss/denial of a place to live, being treated "rudely" in a public place, being asked to leave a business, loss of a friend, harassment or assault and receipt of harassing/threatening telephone calls or mail). Also used as dependent variables are items assessing registrants' perceptions of shame, being unfairly punished by registration, understanding the purpose/goal for the registry and perceiving social stigmatization, all as a result of registration.
Results
Analysis focused on identifying the distribution of negative consequences reported by registrants as arising from registration, as well as perceptions and attitudes of registrants toward registries and the activities of officials charged with maintaining the registries.
In order to understand the negative consequences that may come from a sex offender being placed on the registry, it is important to assess the degree to which others in a registrant's social milieu know of the registrant's status and offenses. When asked to indicate what portion of their "family, friends, coworkers, and other people you consider a part of your life know about your sexual offense conviction(s)," responses indicated that for almost all offenders, at least a sizable minority if not all or nearly all of these persons know of the offender's offenses. Only 5.0 percent of registrants report that fewer than 10 percent of others in their lives know about their offenses. However, fully 45.0 percent report that 90 percent or more of others know of their offenses, with 25.0 percent saying everyone they know has knowledge of their offenses. Whether this knowledge is attributable to the registration process and site is not known; however, the important point is that for most registered sex offenders, others know their status as sexual offenders.
Perceived Collateral Consequences of Registration
As shown in Table 2, a number of negative experiences stemming from sex offender registration are commonly reported by registrants. More than 30 percent of registrants report having lost a job, losing or being denied a place to live, being treated rudely in public, losing friends, and being personally harassed as a result of public knowledge of one's offenses.
Table 3 presents the distribution of negative experiences resulting from registration for registrants based on length of time on the sex offender registry. It is apparent that for all ten collateral consequences, a greater percentage of women who have been on the registry for longer than the sample median of 32 months report having had such a negative experience.
Attitudes Toward Registration
In addition, to assess registered female sex offenders' perceptions and reports of negative consequences arising from their listing on the registry, analysis also examined registrants' responses to five attitudinal items. Women were asked to report their level of agreement with each of 5 statements (1= strongly disagree, 10 = strongly agree), as shown in Table 4 . Registrants report a high level of shame about their registration and largely believe that registration is an unfair form of punishment; yet typically say they understand why society desires a sex offender registry.
There are no statistically significant mean differences in responses to the five attitudinal items across registered female sex offenders based on the length of time they have been on the sex offender registry.
All registrants were also asked whether they believed that "because my name and personal information is listed on the Sex Offender Registry I am less likely to commit another sexual offense in the future." The mean response to this item is 7.42. Nearly two-thirds (61.1 percent) of registrants report complete agreement with the statement, although the actual effect of registration on recidivism cannot be determined. No statistically significant differences are seen for registrants based on length of time on the registry.
Discussion
As one of the more recent responses to sexual offending, the use of sex offender registries clearly has far-reaching implications for society as well as for individuals listed on registries. However, little previous research has examined these implications. The present research is one of the first attempts to examine both the consequences of sex offender registration for offenders and one of the few assessments of female sex offenders outside of clinical settings.
The results of this research make clear that registered female sex offenders frequently experience collateral consequences that may have serious deleterious effects on their social, economic, and physical well-being. While the goal of shaming sex offenders seems to be achieved through registration, and registered female sex offenders report an understanding of why society would want to have such registries, there are also obvious indications of registration having lasting negative consequences for individual offenders. Approximately one in three (or more) registered female sex offenders report that as a result of their listing on a sex offender registry they have lost a job, lost or been denied a place to live, lost friends, and been personally harassed. Such experiences are directly contradictory with the goals and resources known to be critical to successful community reentry and the reduction of recidivism. And, as women remain on a sex offender registry for longer periods of time, these (and other) collateral consequences become more common.
In light of these findings, the importance of sex offender registration as a tool for promoting public safety needs to be questioned. While the present research does not definitively conclude that sex offender registration leads to recidivism or poor community adjustment following conviction, it does suggest that the very resources identified as centrally important for successful reentry are diminished and weakened by registration. As such, it is important to continue to assess the consequences of sex offender registration on recidivism, and on accompanying costs (both financial and social) experienced by offenders and communities. If registration cannot be shown to be associated with significantly lower rates of recidivism (see Adkins, et al., 2000), the costs may well outweigh the benefits of registration. And, if registration in fact is associated with lower rates of sexual offending recidivism, it may be useful to examine whether the current method for registering (and publicizing information about registrants) can be modified in a way that maintains the positive outcomes while reducing the costly collateral consequences.
The present study is a first step toward evaluating these costs and benefits. Future research needs to look more closely at the costs of sex offender registration and the benefits. At present it appears that registration of sex offenders—or at least the female sex offenders questioned in this study—may generate more societal costs and negative consequences for individuals than intended, necessary, and appropriate.