New study: Do popular actuarials work?

4-21-2010 National:

From In The News by Karen Franklin:

Newer instruments outperform Static-99 and RRASOR

If you evaluate sex offenders, you may be overwhelmed by the dizzying array of actuarial risk assessment instruments vying for your vote: Static-99, RM 2000, RRASOR, Static-2002, SORAG, MnSOST-R. The list of weird-looking acronyms grows ever longer, and the validity studies more complex.

For help figuring out which instruments do what, and how well, I recommend a study by two Canadians, Jan Looman and Jeffrey Abracen, just published in the Journal of Interpersonal Violence. The study compared the ability of four leading instruments to predict recidivism among a sample of 419 high-risk Canadian sex offenders. Overall, about 15% of the released offenders were convicted of a new sex offense when the follow-up time period was held at a constant of seven years.

Among the interesting findings:

•When rapists and child molesters were examined separately, none of the instruments were able to predict sexual recidivism (defined as a new conviction) for child molesters.

•The RRASOR (Rapid Risk Assessment for Sex Offense Recidivism) bombed completely as a predictive tool. The Static-99, a descendant of the RRASOR and the most widely used actuarial for assessing sex offender risk, did not do much better. As the authors note, since their sample was similar to U.S. offenders subject to civil commitment as Sexually Violent Predators (SVPs), the lack of predictive validity of these instruments raises concerns about their use in SVP proceedings.

•Two newer instruments -- the Static-2002 and the Risk Matrix 2000, developed in England -- did somewhat better. Especially interesting is how the Static-2002 performed. Overall, only two of the instrument's five subscales predicted sexual recidivism. Not surprisingly, these were Age at Release and Persistence of Sexual Offending. This is rather common sensical, in that sex offending declines dramatically with age, and a high score on the Persistence scale indicates an entrenched pattern of sex offending, with an early onset, a high rate, and multiple convictions. When rapists and child molesters were separated out, instead of being lumped together as generic “sex offenders” as they often are, only the Deviant Sexual Interests scale predicted sexual recidivism for the child molesters.

•Finally, age matters. It really does. The few who continue to deny this obvious fact are just being stubborn. Similar to other studies, only about 2% of offenders over 50 reoffended (one child molester and one rapist, in this case), compared with 20% of those ages 25 to 35.

Of course, this is just one study. Other studies comparing the accuracy of instruments for assessing sex offender risk have come to various and contradictory conclusions. For example, contrast the present findings with those of the Austrian study that I blogged about last month, testing German versions of the RRASOR, Static-99, SORAG, and SVR-20. The publisher of the International Journal of Offender Therapy and Comparative Criminology has just made that study available for free, by the way. Keeping up with this burgeoning literature is a full-time job, unfortunately necessary for practice in this area.

Read More of Article...

Article Summary only...

Evaluating and Improving Risk Assessment Schemes for Sexual Recidivism: A Long-Term Follow-Up of Convicted Sexual Offenders

March 2007 217618

EXECUTIVE SUMMARY
Risk assessment plays a central role in the management of sexual offenders in the criminal justice system, impacting every level of decision-making. The specification of which offenders should be included in state registration or community notification, the distribution of treatment opportunities in prison, the determination of parole eligibility, the allocation of resources in monitoring and supervising offenders in the community, and the ultimate decision to remove the most serious offenders from the community through civil commitment all involve critical choices involving risk. The serious consequences to potential victims, to those offenders who no longer pose a threat to the community, but are denied their civil liberties, and to critical community funds that may be squandered as the result of inaccurate decisions all demand the guidance of the most accurate actuarial algorithms available. Because decisions must be made—even doing nothing constitutes a decision—to choose not to employ the most accurate decision-making algorithms is to choose to pay the costly price of a suboptimal decision.

The aim of the present study was to evaluate and to improve the decision-making algorithms that have been generated to assess risk in sexual offenders. More specifically, it was the task of this research project to evaluate the extant actuarials in a sample of sexual offenders on whom long-term follow up were available. We assessed the comparative accuracy of the major risk instruments over time and over subsamples, explored their underlying factor structure, examined the accuracy of a new assessment protocol (SRA Need Assessment; Thornton, 2002), and explored the potential for generating improved predictive instruments.

In a prior study we (Knight, 1999; Prentky, Knight, & Lee, 1997) had followed 599 offenders who had been referred to the Massachusetts Treatment Center (MTC) for evaluation between 1959 and 1984. Of these 266 (Bridgewater Treatment [BT] sample) had been committed to MTC as “sexually dangerous” and subsequently released, and 333 (Bridgewater Observation [BO] sample) had been determined not to be sexually dangerous and returned to finish their sentences. Of the 333 BO offenders 200 constituted a matched sample (on age at evaluation, marital status, and number of prior crimes) and 200 were randomly sampled from the entire sample BO population evaluated. There was an overlap of 67 offenders selected by both the random and matched process. For all these offenders we had accessed and integrated four outcome record sources, including the Massachusetts Board of Probation records, the Massachusetts Parole Board records, the Massachusetts Treatment Center Authorized Absence Program records, and the Federal Bureau of Investigation (FBI) records.

In the present study we accessed the archival clinical files for these offenders and coded these records both on modern empirically-derived, mechanical actuarials that have been developed since 1997 for predicting sexual recidivism, including the RRASOR (Hanson, 1997), the Static-99 (Hanson & Thornton, 2000), the Static-2002 (Hanson & Thornton, 2003), the SORAG (Quinsey, Harris, Rice, & Cormier, 1998), the MnSOST-R (Epperson, Kaul, Huot, Hesselton, Alexander, & Goldman, 1998), and the Risk Matrix 2000 (Thornton et al., 2003), on two structured clinical guidelines, the SVR-20 (Boer, Hart, Kropp, & Webster, 1997) and the A-SOAP-II (Prentky & Righthand, 2003), and on a new experimental measure, the SRA Needs Assessment (Thornton, 2002). Offenders were also categorized in the Massachusetts Treatment Center Rapist (MTC:R3) and Child Molester (MTC:CM3) typologies and diagnosed according the DSM-IV Antisocial Personality Disorder and Conduct Disorder criteria. All codings and judgments were carried out without any access to or knowledge about the follow-up status of the offenders.

The final report addressed six basic questions. We summarize the findings for each using this framework: ..more.. by Raymond A. Knight, Ph.D. ; David Thornton, Ph.D.

Read More of Article...

Article Summary only...

DOES THE PRISON TREATMENT PROGRAM REDUCE RECIDIVISM?

I find the results of this recidivism study statistically invalid due to the selection process used which only allows high risk offenders into treatment. The state, in 2006, concluded that therapy does not reduce recidivism.

Explanation:
Initially the researchers chose everyone who was willing to participate in therapy. Then they divided that into two groups: A) Those who DID NOT participate in therapy; and, b) Those who DID particpate in therapy. No matter which group the person fell into each had to have a prior sex offense conviction.

Now the catch, in order to particpate in therapy a prescreening process was applied. i.e. risk assessment tools MnSORT-R, RRASOR, and Static 99 were used to chose who could go into therapy. ONLY those who "had a high likelihood to reoffend" were permitted in therapy.

Anyone who DID NOT fit that requirement was pushed into the group of those who DID NOT participate in therapy group. In other words the DID NOT participate group was a mixed group of offenders. It contained folks who were willing to take therapy but did not, and those who fell out of the prescreening process.

Now a review of Exhibit 2 (pg-3) shows the study was not interested in just new sex offenses, instead recidivism means any kind of offense. i.e., the definition of recidivism meant more than just sex offenses. It appears ridiculous to assume that sex offender treatment program (SOTP) is designed to reduce recidivism of all kinds of offenses.

Exhibit 2 shows of those who went through therapy, 12 committed a new sex offense, and only 6 of those in the opposing group committed a new sex offense. Based upon that outcome researchers concluded, that therapy did not reduce recidivism.

What? Remember, the deck is stacked, only high risk of reoffense folks went to therapy, and low or medium risk did not. The outcome was predictible based on the prescreening process.

This is why I feel the study is statistically invalid. Additionally I see a few other things but what I mentioned above was more than enough to go no further.

eAdvocate

With that said, the study:

SEX OFFENDER SENTENCING IN WASHINGTON STATE: DOES THE PRISON TREATMENT PROGRAM REDUCE RECIDIVISM?

The 2004 Legislature directed the Washington State Institute for Public Policy (Institute) to conduct a comprehensive evaluation of the impact and effectiveness of current sex offender sentencing policies.1 Because this is an extensive topic, we are publishing a series of reports.

The Washington State Department of Corrections (DOC) has operated a prison-based Sex Offender Treatment Program (SOTP) at the Twin Rivers Corrections Center since 1988. The program has undergone a series of changes since its inception. Since 1996, the program has used a combination of treatment techniques including group therapy, psycho-educational classes, behavioral treatment, and family involvement. The length of treatment has decreased from two years in 1996 to approximately one year currently. Since 2000, sex offenders assessed as having a high likelihood to reoffend, based on their criminal history, are prioritized for program entry.2 (FN2: The SOTP uses three risk for sexual reoffense
assessments: MnSOST-R, RRASOR, and Static 99.)

Offenders selected for the treatment program must meet the following five requirements:
• Sex offense conviction
• Voluntary participation
• Admission of guilt
• One year minimum remaining in prison
• Medium or lower custody classification
This report estimates whether SOTP reduces recidivism by comparing the recidivism rates of sex offenders who were willing but did not participate in SOTP with those who did participate in the program. ..more.. by Washington State Institute for Public Policy

Read More of Article...

Article Summary only...

Sex Offender Risk Assessment

June 30, 2006

Prepared for: Missouri Sentencing Advisory Commission

Executive Summary
Compared to clinical methods, actuarial risk instruments are a preferred method to discern sex offenders risk for sexual as well as violent recidivism because, unlike clinical practices, they are considered inexpensive, objective and modestly accurate. Scientists argue that risk instruments that employ only static, or historic measures of offender characteristics, rather than dynamic, are certainly sufficient for the purposes of gauging individuals’ likelihood of recidivism. In fact, Harris and Rice (2003:207) contend that dynamic constructs are “unnecessary for anticipating who will recidivate in a given time period”; furthermore they state that “very accurate statements about the likelihood of another…offense can be based upon knowledge of an individual’s lifetime conduct.” In their view, offender risk scales that incorporate only static information are essentially capturing factors that refl ect a person’s underlying antisocial propensity.

Although there are a considerable number of risk instruments available for corrections officials to utilize, far fewer have been rigorously evaluated. Of those that have, Harris and Rice (2003) recommend that the MnSOST-R and the Static-99 are two of the most “promising” scales for predicting sexual recidivism. An emerging body of work also suggests that the SORAG is quite effectual in terms of its predictive accuracy. Additional empirical research is likely to surface which will provide further evidence of the statistical accuracy of sex offender risk instruments.

For the remainder of this paper: by Dr. Beth Huebner (Prepared by: Institute of Public Policy, Truman School of Public Affairs, University of Missouri – Columbia)

Read More of Article...

Article Summary only...