A False Sense of Security ?

July 22, 2009

Criminal Justice Policy Review : The Adam Walsh Act: A False Sense of Security or an Effective Public Policy Initiative?
Naomi J. Freeman, Ph.D.(New York State Office of Mental Health) and Jeffrey C. Sandler, M.A.(University at Albany)
-subscription required to view document.


With the enactment of the Adam Walsh Child Protection and Safety Act (AWA), states are required to standardize their registration and community notification practices by categorizing sex offenders into three-tier levels in the interest of increasing public safety. No empirical research, however, has investigated whether implementation of the AWA is likely to increase public safety. Using a sample of registered sex offenders in New York State, the current study examined the effectiveness of the Adam Walsh-tier system to classify offenders by likelihood of recidivism. Results indicated that the AWA falls short of increasing public safety. In fact, registered sex offenders classified by AWA as Tier 1 (lowest risk) were rearrested for both nonsexual and sexual offenses more than sex offenders in Tier 2 (moderate risk) or Tier 3 (highest risk).


The current study tested the ability of the tier system, as stipulated in SORNA, to predict sexual recidivism among a group of registered sex offenders in New York State. The results cast doubts on the ability of the SORNA provisions of the AWA to increase public safety. More specifically, results showed that those offenders classified as Tier 1 (lowest risk) were rearrested for both sexual and nonsexual offenses more quickly than both Tier 2 (moderate risk) and Tier 3 (highest risk) offenders and were rearrested for sexual offenses at a higher rate than Tiers 2 and 3 offenders. Moreover, as shown in Table 3, the results indicated that many other risk factors supported by empirical research would be better predictors of future sexual offending than the SORNA tier level. Given the results of the current study, the enactment of SORNA may give community members a false sense of security. That is, community members may believe they are safe if no Tier 3 offenders are residing in their neighborhood when, in fact, Tier 3 offenders are not at increased risk to reoffend. As such, SORNA appears unable to accurately identify high-risk offenders and, therefore, increase public safety.

This lack of any observed association between crime of conviction and sexual recidivism may be due to the fact that crime of conviction may not be an accurate indication of the type of offense that was committed. Because convictions in sexual offenses are often elusive—whether as a result of lack of evidence, victim’s hesitation to testify, credibility of the victim, or characteristic s of the defendant—prosecutors may be more likely to offer a plea bargain in sexual offense cases to secure a conviction. As such, it is possible that crime of conviction does not accurately reflect the offense that was committed and, therefore, may be a poor predictor of future risk of reoffending. Thus, it is unsurprising that, as the current study found, other easily obtainable risk factors would be better predictors of recidivism and offer a more accurate risk criterion for the classification of sex offenders.

States had until July 2009 to fully implement the regulations outlined in the AWA.7 Yet, a recent analysis conducted by the Justice Policy Institute (2008) noted that, in all 50 states, the costs of implementing SORNA far outweigh the costs of losing 10% of Byrne funding. In fact, the Justice Policy Institute estimates that in 2009 alone, New York State would spend US$31,300,125 for the implementation of SORNA, whereas forfeiting 10% of its Byrne funding would only result in a loss of US$1,127,984.

Given the large fiscal implications of implementing SORNA, as well as results of the current study which indicate that the tier system does little to accurately predict which offenders will reoffend and which will not, perhaps states should reconsider the implementation of SORNA. Rather, if states are wedded to registration and community notification practices despite the empirical research that indicates the ineffectivenes s of such laws to impact rates of sexual offending (e.g., Petrosino & Petrosino, 1999; Sandler et al., 2008; Walker et al., 2005; Zevitz, 2006; Zgoba et al., 2008), then perhaps the public would be better served if the scarce resources surrounding sex offender management were limited to the offenders who pose the greatest risk to the public’s safety (Harris & Hanson, 2004). Given the results of the current study, which indicate the lack of ability for the tiered system under SORNA to accurately identify offenders at high risk of sexual recidivism, the AWA would, in fact, target the strongest sanctions against those least likely to recidivate, while giving lesser sanctions to those most likely to recidivate (i.e., Tier 1 offenders).

Currently, the provisions outlined in SORNA do not discriminate between those sex offenders who can be rehabilitated and those who may continue to sexually offend. Yet, in recent years much has been learned about risk factors related to sexual recidivism, and a growing number of actuarial risk assessment instruments have been developed to identify those high risk sex offenders who pose the greatest threat to public safety. The two most well-known risk assessment instruments used for the prediction of sexual recidivism among male sex offenders are the Static-99 (Hanson & Thornton, 1999) and the MnSOST-R (Epperson et al., 1998), both of which have been shown to have moderate predictive accuracy in numerous international samples of sex offenders (Knight & Thornton, 2007). Although these risk assessment instruments do not account for all factors that could be associated with recidivism, they provide a moderate prediction of recidivism and allow for a means to distinguish sex offenders based on risk (usually into categories of low, medium, and high risk). In fact, results of the current study suggest that individual items found on these instruments are significantly associated with recidivism for a group of sex offenders in New York State. Specifically, the presence of prior sexual offenses, the number of previous sentencing dates, having male victims, and being younger (all items on the Static-99) were all related to an increase in the likelihood of sexual recidivism. Although some sex offenders are extremely dangerous and pose a threat to public safety, others present a low risk and can be effectively managed in the community with appropriate levels of supervision and treatment. Thus, the registration and community notification provisions of the AWA may be more effective if actuarial risk assessment instruments that measure both static and dynamic factors are used as a way to identify those most at risk to reoffend (see Levenson & D’Amora, 2007) instead of the currently proposed three-tier system based solely on crime of conviction. Not only would this approach prevent low-risk offenders from receiving the same sanctions as high-risk offenders, it would also conserve resources and allow registration and community notification practices to be directed at those most at risk to reoffend. Targeting intervention programs and legislative initiatives to specific types of sex offenders will more effectively reduce the likelihood of recidivism, ultimately increasing public safety, and will conserve the limited resources aimed at sex offender management strategies.

The idea behind the enactment of the AWA, to standardize registration and notification procedures nationwide, appeared to address limitations of the current system. In reality, however, the three-tiered system, as outlined in SORNA, fails to increase the effectiveness of current registration and community notification practices. In fact, as indicated by the results of the current study, the system proposed in SORNA actually decreases the ability of states to predict which sex offenders will sexually reoffend and which ones will not. More specifically, the use of almost any empirically based risk factor would yield more accurate predictions than the SORNA tier level, which is based solely on crime of conviction. Although no risk prediction system can accurately predict sexual recidivism 100% of the time, the results of the current study indicate that SORNA is almost completely ineffective at categorizing sex offenders based on risk of sexual recidivism. As such, it appears enactment of the AWA (and, therefore, SORNA) would not only cost states more money than they would lose if they were not to enact it, but also that such enactment would unlikely increase public safety.

There is, however, a broader question surrounding the ability of any sex offender registration and notification law to increase public safety. Specifically, several recent studies (e.g., Petrosino & Petrosino, 1999; Sandler et al., 2008; Walker et al., 2005; Zevitz, 2006; Zgoba et al., 2008) have found registration and notification laws to be ineffective methods of reducing sexual victimizations . Furthermore, there is some evidence to suggest that these types of laws are increasing recidivism, as the unintended consequences of these laws may aggravate stressors known to be associated with sexual reoffending (Freeman, in press). Winick (1998) argued that by denying them [sex offenders] a variety of employment, social, and educational opportunities, the sex offender label may prevent these individuals from starting a new life and making new acquaintances, with the result that it may be extremely difficult for them to discard their criminal patterns. (p. 556)

Given that the SORNA provisions increase the reporting requirements as well as the public distribution of housing and employment information, it is possible that the enactment of the tier system, as outlined in SORNA, may actually increase reoffending rates of convicted sex offenders. As such, perhaps it is time to replace these wellintended, yet ineffective, public policy initiatives (e.g., registration, community notification) with ones that are scientifically supported.

%d bloggers like this: