Letter to the Editor

Editor:

Morris, Haroun, and Naimark (J Am Acad Psychiatry Law 32:231–45, 2004) surveyed forensic psychologists and psychiatrists and failed to find consensus in judgments of competency to stand trial based on a presented vignette. Essentially equal numbers of subjects found the same defendant to be competent as found the defendant to be incompetent. The authors characterized this finding as “not a mere fluke of the sample” and offered some probability calculations to suggest that the likelihood of this result is as low as 1 in 10 quadrillion (a result that requires certain assumptions about an expected higher base rate of examiner agreement). Apparently regarding these results as an embarrassment to forensic expertise, the authors concluded that “the defendant’s fate depends only on who performs the evaluation.”

Before zealous attorneys seize upon this research report as a cross-examination tool, it should be noted that a far simpler and arguably more compelling hypothesis was available. The authors never demonstrated that the brief vignettes employed as research stimuli were psychometrically adequate to the task of eliciting and accurately measuring the subjects’ forensic decision-making. A psychometrically inadequate vignette could effectively reduce the subjects to mere guessing, and random guessing would also produce the observed pattern of equally divided judgments.