Probabilistic Scientific Evidence - Jurors' Inferences

• NCJ Number: 105015
• Author(s): J Goodman
• Date Published: 1986
• Length: 270 pages
• Grant Number(s): 86-IJ-CX-0004
• Agencies: NIJ-Sponsored
• Publication Type: Thesis/Dissertation

Annotation

Two jury simulation studies examined inferences drawn by mock jurors from probabilistic frequency evidence and found no support for the position of some judges that probabilistic information has undue influence on jurors and should therefore be excluded from jury trials.

Abstract

The first study varied the frequency probabilities of a suspect's blood type. The 233 subjects for the study were psychology undergraduates who returned verdicts individually in a pencil-and-paper exercise. The subjects underused the probabilistic evidence and had high error rates on questions about it. The second study used three variables: burden of proof, the linguistic form of the probabilistic testimony, and the presence or absence of a visual aid depicting results of chromatographic tests of gasoline samples in a hypothetical arson case. The 223 mock jurors had high error rates on questions about probabilistic evidence. None of the three variables significantly affected the verdicts. Jurors gave significantly more weight to the expert witness who presented scientific evidence than to the expert witness who did not. Neither study supported the hypothesis that jurors are susceptible to the prosecutor's fallacy; thus, subjects did not confuse the burden of proof and the probabilistic evidence. Data tables, appended study instruments, and 161 references. (Author abstract modified)