Dr. McGrew provided two thought provoking postings on the use of partial IQ scores in the Atkins context. But the problem with inappropriate use of partial IQ scores does not lie solely with expert witnesses. The J. Williams (2007, Texas) Atkins case involves a judge’s use of partial IQ scores, without any suggestion by the experts that this was an appropriate way to deal with the Atkins claimant’s IQ scores. In Jeffrey Demond Williams’ case, the federal magistrate judge court concluded that the partial IQ scores should control over the full scale IQ scores, with no explanation of why (and under what authority) these partial IQ scores should be used as the measure of general intellectual functioning. Williams had taken three IQ tests, with full scale IQ scores of 70, 70, and 71, all consistent with a diagnosis of mental retardation. To deal with these scores, the magistrate judge stated, “While those three scores appear superficially to show a consistent subaverage intellectual functioning over time, a closer examination reveals that Williams was capable of scoring a Verbal IQ of 79 and a Performance IQ of 77, well outside the mentally retarded range.”  The court concluded its decision by stating that, “Considering Williams’ borderline IQ scores in light of his academic performance . . . ,” signifying that it had, in fact, used the partial scores which were in the borderline range (77 and 79) instead of the full scale IQ scores.
Interestingly, even though the magistrate judge sided with the prosecution expert in the end, according to the decision, the prosecution’s expert did not suggest that the judge should use the partial IQ scores of 77 and 79. The prosecution expert testified that “Williams’ three IQ scores at or near 70 are the result of poor effort on all three tests.”  Moreover, this expert’s report filed with the court made no such recommendation. Instead, the state’s expert suggested that a score of 83 on a non-verbal IQ test (TONI-3) given to Williams by the Texas Department of Corrections was the best indicator of Williams intellectual functioning. 
So what we have is a magistrate judge playing psychologist and determining on his own that the partial IQ scores were the best indication of Williams intellectual functioning, without any analysis or citation to authority, legal or scientific, to support his conclusion in this regard. The magistrate’s opinion was adopted by the district court judge, and her findings were affirmed on appeal. The appeals court compounded the error by approving this approach.
 Memorandum and Recommendation, Williams v. Quarterman, Civil Case No. H-04-2945 (S.D. Tex. filed Jan. 8, 2007), at page 70-71. Even if the court ultimately made the correct decision – that Williams was not mentally retarded - that does not excuse this inappropriate manner of dealing with the three IQ scores. This especially so in a case like this one, where the magistrate judge admitted that, “reasonable jurists can disagree about whether the evidence supports Williams’ claim that he is mentally retarded.” Id. at 77. In close cases, the courts need to especially vigilant to ensure that their analysis is both legally and scientifically sound.
 Id. at 68.
 Psychological Report of Thomas G. Allen, Ph.D., in the matter of Williams v. Dretke, Civil Case No. H-04-2945, (Feb. 8, 2006)
Technorati Tags: psychology, forensice psychology, criminal psychology, ABA, American Bar Association, school psychology, educational psychology, neuropsychology, clinical psychology, intelligence, IQ, IQ tests, IQ scores, part v full scale IQ, Atkins case, MR, mental retardation, intellectual disability, developmental disability, death penalty, capital punishment