Risk Assessment of Sex Offenders

• Author: Daniel Montaldi
• Pages: 79-94

79

Chapter

Risk Assessment of

Sex Offenders

Daniel Montaldi

6

You represent a person charged with a sex offense, and you are confronted

with a risk assessment that contains all sorts of terms like “rate” and “risk”

and “dangerousness.” The report compares actuarial rates and lifetime

rates. The report cites tests known as the Static-99 and the Static-99R.

The prosecutor’s expert has an alphabet soup of initials following his or her

name. And to top it all off, you know that as a lawyer representing some-

one charged with a sexually oriented offense, you are representing someone

almost everyone believes is Public Enemy Number One. What are you to do?

This chapter by Dr. Daniel Montaldi shows how to read a report and

interpret its terms. He details weaknesses and assumptions of risk assess-

ment tools. And he provides several lines of questioning to pursue during

cross-examination of the expert.

• • •

This discussion is devoted to some basic concepts and common mis-

conceptions about sex offenders and risk assessment. These issues rarely

receive specific treatment in the literature. Familiarity with these topics

will help attorneys make decisions about how to examine experts testify-

ing about risk.

More generally, the issues and misconceptions discussed in this chap-

ter relate to how sex offender risk assessments often exaggerate risk. This

commonly occurs when the evaluator mistakenly treats risk assessment

80 • Representing People with Mental Disabilities

instruments as underestimating risk. Risk exaggeration results in increas-

ingly scarce resources being devoted to the management and treatment

of individuals who are actually low risk, leaving too few resources for

managing the truly dangerous. Inaccurate risk assessments also harm the

person assessed by adding stigma and further blocking access to hous-

ing, employment, and social support. Finally, excessive resources devoted

to the management of low-risk sex offenders are resources not available

for helping people with serious mental illness and nonsexual issues. The

needs of this latter population are overwhelming the legal and correc-

tional systems of many states. Let’s begin with some basic risk assessment

tools and concepts.

Static-99 and 99R

References are made throughout this discussion to a widely used actuarial

risk assessment instrument for sex offenders. The Static-99 line of instru-

ments is the most researched set of actuarial risk assessment tools for sex

offenders. The Static-99 and (now) the revised version (Static-99R) are

the primary actuarial devices used in Sexually Violent Predator (SVP) civil

commitment evaluations. The tool consists of a list of ten “risk factors” or

recidivism-related characteristics that are typically unchanging in nature

(static) and act as scoring criteria for the instrument. One such factor is

previous sex offense convictions. The factor is unchanging because previ-

ous convictions do not go away.

The original tool came out in 1999, thus the name. The tool was

normed on offenders released in the 1950s through the 1980s. Later

research using samples of offenders released more recently found dif-

ferent (and lower) rates, and also different rates for different samples of

offenders with the same score. Researchers also discovered that increasing

age has a much stronger impact on recidivism rates (lowering them) than

previously thought. The Static was revised (Static-99R) in light of these

findings to include more age cutoffs (for decreased score) and multiple

reference groups instead of just one. Each reference group has its own set

of (new sex-related charge or conviction) rates per score. The 99R went

into widespread use in 2009.

The Static-99 has one actuarial or “experience” table. The table lists

sex offense reconviction rates for samples of offenders with a given score

on the instrument. These are observed rates in that researchers followed

each sample after release to find out what proportion of the sample’s