Psychologists have grappled for decades with a basic, sobering reality of our profession: absent certain specialized circumstances, we can’t predict the future. It makes us feel only a little better to reflect to ourselves, "well, who can?" Incidentally, when we mention this aloud, it doesn’t make judges feel any better at all.

It’s easy to see why clinicians welcome claims that newly developed instruments will enable us to perform reliable and valid "risk assessments" regarding persons convicted of sexual offenses. It’s easier still to understand how judges are willing to accept that the administration of these psychological tests will lead to accurate determinations of potential dangerousness – particularly when these evaluative procedures are mandated by statutory law. Perhaps easiest to understand, however, is the mounting frustration of all participants in this process who come to believe that these measures are not "administered," nor "psychological," nor even "tests" in the sense we have employed such terms in the past.

In a recent issue of The Advocate, I commented on the emerging doctrine of "Jurisprudent Therapy" and provided the following definition:

This point of view is forcefully reflected in such cases as Daubert v. Merrell Dow Pharmaceuticals, Inc. [2] and Kumho Tire Co., Ltd. v. Carmichael. [3] Trial attorneys are thus encouraged – even compelled -- to ask: "are psychological theories, their clinical and policy-making applications, and the people who develop and provide them making a fair, just, and legally supportable contribution to the lives of the people they are intended to serve?" [4]

Kentucky’s current scheme for the assessment of "sex offenders" (see KRS 17.500 et seq.) provides an excellent example of how this perspective can be brought to life. The law mandates the use of certain actuarial measures in order to determine the degree of "risk" associated with the background of a given "offender." The two core instruments may be characterized as follows:

Any errors or omissions in an institutional record are likely to detract from a given instrument’s accuracy in a particular case. Attorneys should give serious consideration to the correctional sources from which this information is obtained, and the training and background of the personnel providing this material to the designated evaluators.

If actuarial approaches are truly superior to clinical judgment (as the research consistently suggests they are), then why are "psychological" experts employed to conduct them? It is here that a Jurisprudent Therapy analysis attaches: these actuarial assessments are cloaked in the guise of "clinical" practice (to a litigant’s advantage or detriment, depending on a particular judge’s regard for clinicians), obscuring their true nature, and thus complicating the fact-finder’s ability to gauge their import and value as scientific evidence. [6]

These observations should not be construed as gratuitous criticism of the psychologists chosen to perform these evaluations in the Commonwealth of Kentucky. These valued colleagues have availed themselves of skilled consultation from within and without the state, often possess considerable experience from providing services in other forensic contexts, and work under extreme time and workload pressures. The reliability and validity of their contributions will be limited, however, like those of any professional, by any deficiencies in mandated measures, as well as difficulties in interpreting the statistical and/or psychometric properties of instruments employed.

A few examples from Kentucky’s recent mandatory 32-hour Sex Offender Risk Assessment Advisory Board (SORAAB) training serve to illustrate this point. In the first, a clinician performing evaluations to gauge the likelihood of adolescent recidivism admitted directly to conference attendees that all currently available measures designed for that population had only "face" validity, concluding that "we’re back to just going by our judgment."

Another presenter, asked by a fellow psychologist to explain why materials touting the efficacy of the RRASOR claimed an ability to "capture .27 of variance" while also describing a "predictive accuracy [of] r = .27," admitted that he was unable to explain this assertion.

Still another presenter, when a trainee noted that in one instance a higher MnSOST-R score was actually less predictive of re-offending than a lower one, dismissed this phenomenon as a minor statistical anomaly, and intimated that researchers were avoiding making such data readily accessible to courts because it might lead to allegedly groundless criticism of the instrument in forensic applications.

Again, attorneys should note that much important, useful, and clinically and forensically valid information was imparted at the above-referenced training conference. No one should fail to recognize the effort necessary to keep up with the immense caseloads faced by SORAAB evaluators. This having been acknowledged, however, both prosecutors and defense counsel should be in a position to undertake a measured, stepwise analysis of the sources, nature, and generalizability of the data employed in these evaluations. [7]

One source of guidance in this regard is the codes and guidelines from which psychologists derive ethical standards for professional conduct. They include specific reference to ways in which testing must be conducted and interpreted. Foremost in influence among these resources are the Specialty Guidelines for Forensic Psychologists [8] and the Ethical Principles of Psychologists and Code of Conduct [9]. In addition, recently promulgated regulations concerning psychological practice in the Commonwealth of Kentucky may be found at 201 KAR 26:115 et seq.

Another organizing tool for attorneys exploring the reliability and validity of any forensic measure is Professor Kirk Heilbrun’s seminal 1992 article on "The Role of Psychological Testing in Forensic Assessment," [10] a core workshop and board preparation training reference for the American Academy of Forensic Psychology. The key points of this resource may be summarized as follows:
 

1. The test is commercially available and adequately documented in two sources. First, it is accompanied by a manual describing its development, psychometric properties, and procedure for administration. Second, it is listed and reviewed in Mental Measurements Yearbook or some other readily available source.

2. Reliability should be considered. The use of tests with a reliability coefficient of less than .80 is not advisable. The use of less reliable tests would require an explicit justification by the psychologist.

3. The test should be relevant to the underlying legal issue, or to a psychological construct underlying the legal issue. Whenever possible, this relevance should be supported by the availability of validation research published in refereed journals.

4. Standard administration should be used, with testing conditions as close as possible to the quiet, distraction-free ideal.

5. Applicability to this population and for this purpose should guide both test selection and interpretation. The results of a test (distinct from behavior observed during testing) should not be applied toward a purpose for which the test was not developed (e.g., inferring psychopathology from the results of an intelligence test).

6. Objective tests and actuarial data are preferable when there are appropriate outcome data and a "formula" exists.

7. Response style should be explicitly assessed using approaches sensitive to distortion, and the results of psychological testing interpreted within the context of the individual’s response style. When response style appears to be malingering, defensive, or irrelevant rather than honest/reliable, the results of psychological testing need to be discounted or even ignored and other data sources emphasized to a greater degree.

Return to the Table of Contents

NOTES

[1] Eric Drogin, "WAIS Not, Want Not?": A Jurisprudent Therapy Approach to Innovations in Forensic Assessment of Intellectual Functioning, Advocate 4-5 (September, 1999). See also Eric Drogin, "Evidence and Expert Mental Health Witnesses: A Jurisprudent Therapy Perspective," in New Developments in Personal Injury Litigation 295-333 (E. Pierson, ed., 2000); and Eric Drogin, "The Behavioral Science Committee: A Foray into Jurisprudent Therapy?" Bull. L. Sci. Tech. 9-10 (October, 1999).

[2] 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed. 469 (1993).

[3] 526 U.S. 137, 119 S.Ct. 1167, 143 L.Ed. 238 (1999).

[4] Eric Drogin, "From Therapeutic Jurisprudence … To Jurisprudent Therapy," 18 Behav. Sci. L. --- (2000, in press).

[5] Eric Drogin, "Sexual Offender Risk Assessment and Scientific Evidence: A Jurisprudent Therapy Perspective," Bull. L. Sci. Tech. 3-4 (December, 1999).

[6] See Donald Bersoff, "Judicial Deference to Nonlegal Decisionmakers: Imposing Simplistic Solutions on Problems of Cognitive Complexity in Mental Disability Law," 46 SMU L. Rev. 329 (1992).

[7] See Eric Drogin & Curtis Barrett, "‘But Doctor, Isn’t that Just Your Opinion?’: Contributing to the Decision-Making Process of the Forensic Psychologist as Expert Witness," Advocate 14-20 (May, 1996); Eric Drogin & Curtis Barrett, "Daubert v. Merrell Dow Pharmaceuticals and Scientific Evidence: A Note of Caution," 7 Ky. Prac. Advances 3-4 (1996); and Eric Drogin & Curtis Barrett, "Forensic Mental Health Assessment: Moving from Examination to Evaluation," Advocate 129-33 (January, 1996).

[8] Committee on Ethical Guidelines for Forensic Psychologists, "Specialty Guidelines for Forensic Psychologists," 15 L. Hum. Behav. 655-65 (1991).

[9] American Psychological Association, "Ethical Principles of Psychologists and Code of Conduct," 47 Am. Psychol. 1597-1611 (1992).

[10] Kirk Heilbrun, "The Role of Psychological Testing in Forensic Assessment," 16 L. Hum. Behav. 257-72 (1992).

Eric Drogin, J.D., Ph.D., ABPP is an attorney and board-certified forensic psychologist on the clinical faculty of the University of Louisville School of Medicine. Serving on the Commission on Mental and Physical Disability Law of the American Bar Association, Dr. Drogin also chairs the ABA Behavioral Sciences Committee.
 

Return to the Table of Contents