«Sex Offender Laws Failed Policies, New Directions ■ Richard G. Wright, PhD NEW YORK Contents Contributors..................... ...»
Nagayama Hall and Proctor (1987) observed sexual reoffense specialization among sex offenders with adult and child victims. In their study of 342 male sex offenders, they found that sex offenders against adults tend to reoffend sexually against adults, whereas sex offenders against children tend to reoffend against children. The authors proposed that the nonsexual criminal activity by adult rapists is symptomatic of a more generalized pattern of antisocial behavior, whereas those who assault child victims are more specialized in their reoffending. They concluded that “the best single-predictor of re-arrests is arrests for sexual offending against adults, which they say explains 12 percent of the variance in rearrests for sexual offending against adults and 15 percent of the variance in nonsexual violent re-arrests” (p. 112).
42 Chapter 2Langan and Levin’s (2002) study revealed that of the 3,138 rapists who were rearrested within the 3-year period following their release, 46% were rearrested for a new crime, 18.6% for a new violent offense, 2.5% for another rape, 8.2% for a new nonsexual assault, and 11.2% for a drug offense.
These figures indicate that the majority of rearrested rapists, though arrested for something other than rape, do reflect a degree of specialization for certain offenders. Another way this can be shown is by looking at ratios. The authors explain that the odds of a released rapist being arrested for a new rape are 3.2 times greater than a nonrapist’s odds.
Using national data on about 10,000 sex offenders released from prison in 1994, Miethe, Olson, and Mitchell (2006) sought to determine the specialization and persistence levels of this group. They reported that only 2% of the released rapists in this study were rearrested for a rape. A closer examination of those offenders with at least one prior sex crime arrest determined that only 5% were exclusively sex offenders (i.e., all of their arrests were for sex crimes).
The authors said that “arrest cycles exhibit a predominant pattern of offense versatility and limited evidence of specialization” (p. 224). In other words, when they are rearrested, it is for a variety of offenses that are not sex crimes.
Lussier (2005) presented a developmental criminology paradigm to explore the issue of specialization versus generality. He argued that the specialization hypothesis is based on one parameter of criminal activity—recidivism—whereas the generality hypothesis focuses on participation and variety of criminal activity and does not consider the dynamic nature of criminal activity over time. Examining recidivism studies from a developmental criminology perspective—one that looks at criminal activity as it develops over time—the author suggested that specialization and generality can co-occur over the course of a criminal career. Additionally, the author stated that generalization and specialization “are two distinct processes characterizing the development of offending over time” (p. 284).
This view holds that versatility in offending behavior over time tempers sex offenders’ tendency to specialize in sexual crime.
Juvenile Sex Offender Recidivism Juvenile sex offenders present unique issues when the discussion focuses on recidivism. Adolescent sex offenders are The Problem of Sexual Assault heterogeneous in that they have diverse experiences that include childhood victimization, various degrees of emotional and behavioral control issues, with varying levels of sexual interest in prepubescent children. Sexual reoffending among adolescents is also heterogeneous because the factors associated with offending for this group are so variable (Worling & Långström, 2006). Because not all juveniles who sexually offend are “high risk” for reoffending, we need to be careful how authorities and treatment providers intervene with this population. The mislabeling of juvenile sex offenders as high- or low-risk can have serious consequences on their future well-being.
Using subsequent arrests as a measure of recidivism, an early study by Doshay (1943) of 256 juvenile sex offenders who had undergone counseling for sexual offending behavior revealed that 106 juveniles with no nonsexual criminal history had only two reoffenses during the 6-year follow-up period. This was in contrast to the 24 of 148 sex offenders with other criminal histories who reoffended during that same period. Overall, 40% reoffended criminally and 7% reoffended sexually.
Another study that used subsequent referrals to the juvenile court as a measure of recidivism found that 14% of the youth were referred for sexual crimes and 35% for nonsexual crimes for an overall total 49% recidivism rate (Smith & Monastersky, 1986). Worling and Curwen (2000) evaluated the success of a specialized community-based treatment program for treating adolescent sexual offending. Working with a group of 58 sex offenders who took part in a 12-month treatment program and a comparison group of 90 adolescents, they reported that “recidivism rates for sexual, violent nonsexual, and nonviolent offenses for treated adolescents were 5.17%, 18.9%, and 20.7%, respectively. The comparison group had significantly higher rates of sexual (17.8%), violent nonsexual (32.2%), and nonviolent (50%) recidivism. Sexual recidivism was predicted by sexual interest in children. Nonsexual recidivism was related to factors commonly predictive of general delinquency such as history of previous offenses, low self-esteem, and antisocial personality” (p. 965).
A meta-analysis conducted by Reitzel and Carbonell (2006) found that the average recidivism rates for sexual, nonsexual violent, nonsexual nonviolent and unspecified nonsexual crimes were 12.53%, 24.73%, 28.51%, and 20.40%,
44 Chapter 2respectively (n = 2,986, 2,604 known males) based on an average of a 59-month follow-up period across studies. Caldwell (2007) found that in a cohort of 249 juvenile sex offenders and 1,780 nonsexual juvenile offenders released from custody, the recidivism rate was 6.8% and 5.7%, respectively, during a 5-year follow-up period. It was 10 times more likely that a juvenile sex offender would be charged with a nonsexual offense as compared to a sexual offense. The non–sex offending cohort accounted for 85% of the new sexual offenses.
Worling and Långström (2006) investigated juvenile sex offender recidivism rates by examining 22 published follow-up investigations (studies of previously institutionalized juveniles) of unique samples of juveniles. (See Table 2.1 for a list of risk factors associated with recidivism.) Recidivism rates ranged from 0% to almost 40% across studies with the follow-up periods ranging from 6 months to 9 years. Sexual assault recidivism rates using criminal charges as an estimate of reoffending were 15% (127 of 846).
Those studies using more conservative estimates such as convictions, court records, self-report, or adult-only charges reported a recidivism rate of 14% (226 of 1,593). These findings echo the research on adult treatment in that most sex offenders do not inevitably reoffend, and even less do they do so perpetually, raising questions about sex offender–specific criminal sanctions.
Is Sex Offender Treatment Effective?
Questions about sex offender recidivism are invariably linked to treatment efficacy as the ultimate goal of treatment is to preclude the offender from reoffending. Treatment efficacy is often judged on whether or not there was a new sexual offense committed after treatment had been administered.
Sex offenders are not a homogeneous group, as several studies have noted (Sample & Bray, 2006). Neither are they totally specialists, as some argue, though they do exhibit lower recidivism rates and have less extensive offense histories overall (Langan, Schmitt, & Durose, 2003; Sample & Bray, 2003; Soothill et al., 2000; Speir, Meredith, Johnson, Bird, & Bedell, 2001).
The Problem of Sexual Assault
Information for this table was adapted from Worling and Långström (2006) The Problem of Sexual Assault An important question to ask is does sex offender treatment work? Or, more specifically, does sex offender treatment prevent sex offenders from recidivating? In an early study from 1989, Furby, Weinrott, and Blackshaw’s comprehensive review of 42 studies concluded that there is no evidence to support the efficacy of treatment in reducing sex offender recidivism rates. Part of their conclusion was based on the fact that many of the studies evidenced multiple and serious methodological problems. Marshall, Jones, Ward, Johnston, and Barbaree (1991) asked a similar question: Can sex offenders be effectively treated so as to reduce subsequent recidivism? In their review of treatment outcomes studies, they found that comprehensive cognitive-behavioral programs and combined psychological and hormonal treatments are effective with child molesters and exhibitionists but not with rapists.
This conclusion is disputed however by Quinsey, Harris, Rice and Lalumière, (1993) in particular, who argue that narrative reviews of studies have not demonstrated that sex offender treatment is effective. Their argument centers around several factors, including methodological issues that exclude the use of comparison groups, the overestimation of treatment effectiveness by not including those who refuse or drop out of treatment as compared to outcomes of those who complete treatment with outcomes of untreated offenders, and the lack of sampling groups from the same jurisdiction and cohort. Quinsey and colleagues advocate for meta-analyses that can provide estimates of treatment effects, effect size, relationships between effect size and type of control group, and finally variability in outcome studies and mediator variables.
Marshall and colleagues’ (1991) study was later followed by Alexander (1994) and still later Hall’s (1995) meta-analysis of a post-Furby et al. (1989) study that provided more optimism about the efficacy of hormonal treatments and cognitive-behavioral treatments. Though Hall’s meta-analysis also suffered from methodological issues, mainly because he was able to use only 12 of the original 90 identified studies (only three of which used randomly assigned subjects for both control and treatment groups), it still provided important information about the efficacy of sex offender treatment.
48 Chapter 2Since Quinsey et al’s. (1993) work, much has been done in the way of addressing whether or not treatment is effective for sex offenders. Many victims of sexual assault not only seek prosecution for their attackers but also want officials to do something about treating these individuals in both institutions and communities. As pointed out earlier, the majority of perpetrators of sexual assault are known to the victim therefore a criminal justice response that does not include treatment discourages many victims from participating in the prosecution process. Nevertheless, a consensus view has emerged of how sex offenders should be dealt with. This view holds that treatment can be effective for some sex offenders, voluntary treatment is more effective, treatment reduces recidivism, and treatment must include more than just counseling (Matson, 2002).vii In light of the question as to whether or not treatment works for sex offenders, there is also some debate about which treatments work best and for which offenders.
Conducting research on the effectiveness of sex offender treatment has a number of obstacles both practical and methodological. Following up on released sex offenders for long periods of time is difficult and expensive. In addition, as emphasized by Quinsey and his colleagues (1993), the use of control (untreated) groups and treated groups is vital for conducting empirical research for this population. Similarities in group makeup would also be a necessary component;
otherwise it would be specious to say that treatment affected recidivism rates between the groups because other factors could be involved.
The use of control groups itself presents some issues that could affect outcomes. It is typical that study subjects participate voluntarily, that is, they have some initial motivation to change. Those who do not want to participate, who are not selected to participate, or who drop out are generally used for the control group, if the study is using a control group. So the best possible candidates are selected for the treatment groups. Assigning subjects randomly to control or treatment groups presents moral and ethical issues since purposely denying treatment to released sex offenders could open up the researchers to criticism and certainly to lawsuits by future victims of the untreated group. What the research does show, given the extent of the studies, is that recidivism among sex offenders is lower (though more variable) than The Problem of Sexual Assault the general criminal offender, thus raising questions about the premise of sex offender laws (Alexander, 1999; Doren, 1998; Furby et al., 1989; Greenberg, 1998; Grubin & Wingate, 1996; Hall, 1995; Hanson & Bussiere, 1998; Hanson et al., 2002; Proulx, Tardif, Lamoureux, & Lussier, 2000; Quinsey, Lalumière, Rice, & Harris, 1995).
Sex Offender Treatment Types of treatment that have been shown to have success come in two categories, biological (chemical or physical interventions) and psychological (behavior modification), and in most cases these are given in tandem. Biological approaches focus on reducing the sex drive; whether that is accomplished by use of pharmacology (medication) or surgical castration, the aim is to decrease or eliminate the sex drive. Several medications have been used including antiandrogens, which are used to reduce the amount of natural androgens—naturally occurring substances in the body (i.e., testosterone); hormones—medroxyprogesterone acetate (MPA), also known as Depo-Provera, help reduce the sex drive; and other antipsychotic medications have been known to dampen the sex drive, though these medications are still being tested for effectiveness (Dwyer & Laufersweiler-Dwyer, in Reddington & Kriesel, 2005). A fuller discussion of the role of medicine (e.g., chemical castration laws) in sex offender treatment can be found in Scott and del Busto’s chapter 10.
Psychological approaches attempt to change offenders by modifying behaviors. These approaches include behavioral, cognitive, and psychodynamic interventions.