An Empirical Assessment of Presentation Formats for Trace Evidence with a Relatively Large and Quantifiable Random Match Probability

Abstract

Since the early 1970s, academics have debated the wisdom of an explicit use of statistical information concerning the probability of a coincidental match in the presentation of forensic match evidence at trial. Critics pointed to the dangers of such numbers, arguing that jurors would misunderstand the numbers and exaggerate their importance in the case, being seduced into ignoring the other unquantified evidence. Data from a pool of people called for jury service in Kane County, Illinois, support the conclusion reached in most previous empirical research: Jurors tend to undervalue the scientific evidence when measured against a Bayesian norm. More importantly, the present results indicate that a careful use of Bayesian methods in the courtroom can assist the jury in reaching more accurate verdicts, a conclusion with less support in previous studies. This study also suggests that it may be possible in some contexts to achieve comparable improvements in accuracy by giving juries less information than is conventionally given to them. The significance of this last result, however, is more ambiguous.