Assessment

Overview

The term “sex offender” implies that the individuals who engage in sexually abusive behavior are all alike and can be managed in the same way. However, because adult and juvenile sex offenders are such diverse populations, “one size fits all” approaches are neither appropriate nor effective. Determining what to do with which offenders, how and when to do it, and why it should be done demands careful consideration to the varied levels of risk, needs, development, and functioning of these individuals. This requires having access to—and making good use of—comprehensive assessment information. Put simply, well executed assessments are the key to informed decisionmaking with adult and juvenile sex offenders.

The following are among the many decisions that can be guided by assessments throughout the criminal and juvenile justice process:

Because adult and juvenile sex offenders are such diverse populations, “one size fits all” approaches are neither appropriate nor effective.

Each of these decisions has significant implications for public safety and, as such, the stakeholders involved in sex offender management will benefit from having reliable and comprehensive assessment information upon which to base their decisions.

Some assessments with sexually abusive individuals occur at a specified point in time within the criminal or juvenile justice process (e.g., pre–sentencing, pre–release) and are intended to assist decisionmakers with a specific inquiry. Other assessments occur at routine intervals (e.g., every six months, every contact) and are designed to provide practitioners with an ongoing snapshot of an individual’s stability, adjustment, compliance, progress, and overall circumstances, such that case management strategies can be maintained or adjusted accordingly.

As jurisdictions begin to explore the ways in which assessments are approached and used within their current systems of adult and juvenile sex offender management, it is helpful to consider them within the context of four broad categories:

None of these categories is mutually exclusive; important intersections and interactions are expected. Regardless of the category into which an assessment is placed, the quality and utility of most assessments often hinge on the synthesis of information from multiple parties.

By virtue of their distinct roles in the overall management process, different stakeholders tend to be limited to the kinds of information that they independently collect or observe about a given individual (e.g., from psychological testing, field contacts, treatment encounters, reports from victims), which provides a narrow and incomplete understanding of the adult or youthful offender. When key information is gathered and shared across disciplines and agencies, a more complete “picture” emerges. As a general rule, therefore, relying on a single data point or sole data source for critical decisionmaking is inadvisable.

Using multiple data sources as part of the assessment process enhances the reliability, validity, and ultimate usefulness of these assessments. Areas of convergence increase the confidence of professionals’ decisionmaking, whereas areas of divergence should lead to further exploration. When assessments of adult and juvenile sex offenders are comprehensive and reliable, stakeholders are better able to develop corresponding interventions and responses that reduce the likelihood of future victimization, increase public safety, and maximize limited resources.

A Useful Framework to Guide Assessments

Over the past several years, as policymakers and practitioners within the criminal and juvenile justice fields have become more invested in assessment–driven and evidence–based interventions, many have come to rely on three core principles of effective correctional intervention: risk, needs, and responsivity (see Andrews & Bonta, 2007; Cullen & Gendreau, 2000). When used as a framework for assessment with adult and juvenile sex offenders, these principles address the following questions:

The quality and utility of most assessments often hinges on the synthesis of information from multiple parties.

Risk Principle: Which Sex Offenders Will Benefit Most from Treatment and Supervision Interventions?

A wide range of potential interventions and strategies is available for managing sexually abusive individuals. However, attempting to employ all of these strategies (many of which are time and resource intensive) for all offenders is not a prudent approach, nor will it have the greatest potential to increase public safety. This is because researchers have found that better outcomes are achieved when the intensity of interventions is matched based on assessed level of risk (see Andrews & Bonta, 2007). Specifically, prioritizing higher risk adult and juvenile offenders for higher intensity services will have a greater impact on reducing recidivism than providing that same level of intervention to their lower risk counterparts; delivering intensive interventions to low risk offenders has limited to no impact and, in some cases, can actually result in increased rates of recidivism (see, e.g., Andrews & Bonta, 2007; Cullen & Gendreau, 2000; Gendreau, Goggin, Cullen, & Andrews, 2001).

Although the initial research supporting the risk principle involved “general” offenders, a growing body of evidence suggests that it is also applicable to individuals who have committed sex offenses (Friendship, Mann, & Beech, 2003; Gordon & Nicholaichuk, 1996; Hanson, 2006; Marques, Wiederanders, Day, Nelson, & van Ommeren, 2005; Mailloux et al., 2003). This has significant implications for the ways in which sex offender management strategies are implemented.

Applying the risk principle does not mean that adult and juvenile sex offenders who pose a relatively low risk of recidivism should go untreated or unsupervised. Rather, it indicates that—when faced with the challenges of growing numbers of sex offenders within prisons and residential facilities, increasing supervision caseloads in the community, and often limited resources—agencies and their staff will be best served by reserving the more intensive strategies (e.g., residential or prison–based treatment, intensive supervision, electronic monitoring) for those who pose a greater risk to reoffend. Simply put, the risk principle ensures that the agencies responsible for sex offender management will “get the most bang for their buck.” Therefore, a critical goal of the assessment process is to identify the individual levels of risk posed by adult and juvenile sex offenders so that interventions and strategies can be prioritized accordingly.

Need Principle: What Are the Targets of Treatment and Supervision that Will have the Greatest Impact on Reducing Recidivism Potential among Sex Offenders?

Working with sexually abusive individuals carries with it the considerable responsibility of ensuring community safety, which requires that practitioners focus their time and energy on the “right” problem areas. The need principle provides important guidance in this respect. According to this principle, the greatest impact occurs when programs and services target the changeable factors that are directly linked to recidivism among adult and juvenile offenders (Andrews & Bonta, 2007; Cullen & Gendreau, 2000; Gendreau, 1996). These crime–producing factors, also known as criminogenic needs, are comprised of two types: stable dynamic and acute dynamic risk factors. Stable dynamic risk factors are relatively enduring but nonetheless changeable, whereas acute dynamic factors can fluctuate rapidly.

Among the stable dynamic risk factors specific to adult sex offenders are intimacy deficits, pro–offending attitudes, pervasive anger, and deviant sexual interests; examples of acute dynamic risk factors are substance abuse, sexual preoccupations, access to victims, and non–compliance with supervision (Hanson & Harris, 2000b, 2001; Hanson & Morton–Bourgon, 2005). And for youthful sex offenders, similar dynamic factors (e.g., deviant interests, antisocial values, pro–offending attitudes, impulsivity) are associated with reoffending (see Prescott, 2006; Worling & Langstrom, 2006). Because of their relationship with recidivism both in the short and long term, identifying these criminogenic needs must be a key focus of assessment efforts. This will ensure that the efforts of practitioners are more efficient and effective. Applying the need principle assists supervision and treatment professionals with determining “what” to target and “when” to intervene (Hanson & Harris, 2000b, 2001; Krisberg, 2005; Lipsey & Wilson, 1998).

Responsivity Principle: How Should Treatment and Supervision Interventions for Sex Offenders be Delivered?

Assessments with adult and juvenile sex offenders should also be geared toward identifying specific client characteristics that may impact their response to interventions. Learning style, motivation to change, denial and level of functioning are key examples of these kinds of characteristics, which are known as responsivity factors. The responsivity principle indicates that when programs and services specifically take into account these factors, better outcomes are achieved (Andrews & Bonta, 2007; Cullen & Gendreau, 2000). Ways in which responsivity factors can be addressed include matching clients to specific services based on the content, format, modality, or “teaching approach” used, and by matching clients to specific providers or officers based on skill sets, personality attributes, or style. Responsivity factors are, therefore, an important consideration in the assessment process with sex offenders.

Taken together, the principles of risk, need, and responsivity provide a useful underlying framework for assessments with adult and juvenile sex offenders, as they can guide management approaches in a manner that will preserve limited resources, maximize outcomes, and reduce recidivism.

Risk Assessment

Estimating recidivism risk is perhaps the most common assessment issue raised during the sex offender management process. Indeed, risk estimates can be useful for informing many key decisions with adult and juvenile sex offenders, such as disposition or sentencing, the type of placement or required level of care, release from facilities, and the application of registration and community notification policies. In addition, as highlighted above, assessing risk is particularly helpful for guiding decisions about which individuals will benefit most from interventions and strategies that are both time and resource intensive (e.g., prison–based or residential sex offender treatment, intensive supervision, ancillary accountability measures such as electronic monitoring).

In order to assess risk, practitioners generally employ one of the following approaches:

Unstructured Clinical Judgment

With the unstructured clinical approach, evaluators rely on their “instincts” or intuition about the individual who is the subject of the assessment. Although subjective judgments of some professionals may have some utility, research demonstrates that an unstructured method of assessing risk is not particularly reliable (Andrews, Bonta, & Wormith, 2006; Grove & Meehl, 1996; Grove, Zald, Lebow, Snitz, & Nelson, 2000). The purely subjective, and therefore inconsistent, nature of this assessment strategy means that different assessors may reach very different conclusions about a given offender. Because the potential implications of inaccurate assessments and the associated management decisions with sex offenders are significant (e.g., additional victims in the community, restricted liberties of an offender), all attempts should be made to increase the reliability of risk assessments within the context of sex offender management. Therefore, the exclusive use of unstructured clinical judgment is largely inadvisable.

Empirically–Guided

An alternative is the empirically–guided approach, in which the evaluator uses a structured scale or checklist to rate the presence or absence of specific risk factors associated with recidivism, and then makes an informed determination about the presumed level of risk. The Risk for Sexual Violence Protocol (RSVP), a modification of the Sexual Violence Risk–20 checklist (SVR–20) is one example of an empirically–guided approach to risk assessment with adult sex offenders (see Boer, Hart, Kropp, & Webster, 1997; Hart, Kropp, & Laws, 2004). The Estimate of Risk of Adolescent Sexual Offense Recidivism (ERASOR; Worling & Curwen, 2001) and the Juvenile Sex Offender Assessment Protocol–II (J–SOAP–II; Prentky & Righthand, 2003) are the primary examples of this approach with youthful sex offenders.

Although the empirically–guided risk assessment strategy tends to be more reliable than unstructured clinical judgment with sex offenders (Hanson & Morton–Bourgon, 2007), inconsistency between assessors remains a noteworthy concern. This is in part because no specific direction is generally provided regarding how much “weight” should be given to each of the risk factors that is being considered within the assessment.

Actuarial

Yet another risk assessment strategy commonly employed in the sex offender management field is the actuarial approach, in which an assessor uses an empirically–validated instrument with a fixed and relatively small number of research–supported items. Each item is assigned a specific weight, and the items are summed to yield a total score that is associated with a broad risk category (e.g., low, moderate, high). Risk categories are linked to the known recidivism rates of groups of sex offenders who were followed at routine intervals (e.g., 5, 10, and 15 years).

Actuarial tools are grounded in extensive research to ensure that the tools predict what they are designed to predict (i.e., sexual or violent recidivism) and that different assessors will reach the same conclusion about the same offender when using the tools. As a result, actuarial tools provide more accurate estimates of risk than both the unstructured and empirically–guided approaches with sex offenders (see Hanson & Morton–Bourgon, 2007; Quinsey, Harris, Rice, & Cormier, 2006).

Multiple actuarial risk assessment tools specific to adult sex offenders have been developed over the past decade. Sex offender–specific instruments are necessary because although there is some overlap between risk factors for sex offenders and non–sex offenders, several factors are uniquely associated with sexual recidivism (Hanson & Bussiere, 1998). As such, simply using a tool designed to estimate recidivism risk with “general” offenders will provide only part of the picture, whereas actuarial tools designed for sex offenders specifically can offer more accurate estimates. The following are key examples of actuarial tools developed for use with adult sex offenders:

Sex offender–specific instruments are necessary because although there is some overlap between risk factors for sex offenders and non–sex offenders, several factors are uniquely associated with sexual recidivism.

The actuarial approach is not without its own set of limitations. One of the most salient issues is that many of the most commonly used instrument, (e.g., RRASOR, STATIC–99) were constructed using only static or unchangeable factors (e.g., number of prior sex offenses, gender of victims). As a result, these tools do not take into account the various dynamic or changeable risk factors that are also associated with recidivism among adult sex offenders. This not only makes the assessment of risk less comprehensive, but also limits professionals’ abilities to assess important changes in risk over time. Additionally, because the actuarial approach is, by design, a structured and objective strategy for assessing risk, idiosyncratic characteristics of a given offender are generally not taken into account when arriving at the risk determination.

Another important caveat regarding the use of actuarial tools is that they cannot indicate whether a particular individual will or will not recidivate; rather, these tools are simply designed to offer “relative” risk estimates. In other words, actuarials can assist practitioners with considering whether a given individual poses a greater or lesser risk of recidivism in comparison to other offenders based on their score on the tool. The risk categorization (i.e., low, moderate, or high) may or may not ultimately prove to be an accurate reflection of the offender’s risk to recidivate. For example, it is possible that an individual categorized as high risk will not ultimately reoffend (false positive), and that an offender rated as low risk will commit a new sex offense (false negative).

In summary, actuarial risk assessment tools are an extremely important innovation in the sex offender assessment process, but practitioners must be cognizant of and carefully consider the following issues:

Assessing Risk with Juvenile Sex Offenders

Although research on risk assessment with adult sex offenders has advanced significantly in recent years, the state of risk assessment for juvenile sex offenders remains in its infancy (see, e.g., Prescott, 2006). Challenges with assessing risk among juvenile sex offenders are a function of the low base rates of juvenile sexual recidivism, a lack of controlled, empirical studies pertaining to risk estimation with this population, and limited efforts to develop risk assessment tools specifically for juveniles to date (see, e.g., Worling & Langstrom, 2006; Prescott, 2006).

These conditions have affected professionals’ abilities to make research–based risk estimates about juveniles who have committed sex offenses. Consequently, some agencies and organizations have either developed their own internal risk assessment tools for juvenile sex offenders or relied on more generic and nonsex offense specific risk assessment tools that were designed for “general” justice–involved youth. These approaches, however, are unlikely to provide accurate risk estimates, in that they have not been established as reliable or valid measures for this population and fail to take into account the specific variables associated with sexual recidivism among these youth.

It is important to recognize that because the actuarial tools listed above (e.g., RRASOR, STATIC–99) were developed to assess risk with adult sex offenders, they are not automatically generalizable for use with juveniles who have committed sex offenses. Youthful sex offenders differ from adult sex offenders in multiple ways, and the risk factors associated with recidivism for adults and youth are not identical (see, e.g., Longo & Prescott, 2006; Prescott, 2006; Worling & Langstrom, 2006). Unfortunately, the extensive research necessary to develop and validate actuarial tools has not yet been sufficiently conducted within the juvenile sex offender field. At present, the Juvenile Sexual Offense Recidivism Risk Assessment Tool–II (JSORRATII; Epperson, Ralston, Fowers, DeWitt, & Gore, 2006) is the only tool that can be considered an actuarial instrument for youthful sex offenders; however, it has not been independently validated and, as such, it is only an experimental or research tool (Epperson et al., 2006).

The next best alternative is the empirically–guided approach to risk assessment. As previously mentioned, two empirically–guided risk assessment tools (i.e., ERASOR, J–SOAP–II) are available for use with juvenile sex offenders. Although additional information is needed, the research conducted on these tools thus far is very promising (Prentky, Harris, Frizzell, & Righthand, 2000; Righthand, et al., 2005; Worling, 2004). As such, these tools are likely better than the alternatives (i.e., relying solely on a non–research supported tool or using clinical judgment alone). Over the next several years, because of the growing interest in and need for additional research in this area, advances in risk prediction with youthful sex offenders are likely to be made. Practitioners will be well served by remaining abreast of these developments.

Click here to assess your policies and practices in this area.