We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Mathematical Probabilities Inadmissable As Evidence.
- Authors
Ginsburg, Allen J.
- Abstract
The article discusses the U.S. Supreme Court decision in the case People v. Collins, filed in 1968. During the defendants' trial for robbery, the state was experiencing difficulty in establishing the identities of the perpetrators of the crime, because the victim was unable to identify them, and descriptions by the witnesses were incomplete. Witnesses to the robbery established that it was committed by a Caucasian woman with a blond ponytail who left the scene accompanied by a Negro man with a beard and mustache, who was driving a yellow Cadillac. The defendants answered this description. To bolster the identification, the prosecutor called a math instructor as a witness to determine the probability that the crime was committed by any couple answering such distinctive characteristics. The witness had arrived at an answer based upon the "product rule" which states that the probability of the joint occurrence of a number of mutually independent events is equal to the product of the individual probabilities that each of the events will occur. He concluded that the probability was but one chance in 12 million that any couple possessed the distinctive characteristics of the defendants. The court held that the use of mathematical probabilities was error, for (1) the testimony itself lacked an adequate foundation both in evidence and in statistical theory, (2) the testimony was used to distract the jury from its proper function of weighing the evidence of guilt. The witness had failed to produce any statistical evidence in support of the probabilities of the factors selected, and had offered no proof that the characteristics were mutually independent, though this was essential to the product rule. As the court pointed out, error in the probabilities of the factors, or the extent that the factors were not mutually independent, would yield an erroneous result. Mathematical odds are not admissible as evidence to identify a defendant.
- Subjects
LEGAL judgments; TRIALS (Robbery); MATHEMATICS; EYEWITNESS identification; PROBABILITY theory
- Publication
Journal of Criminal Law, Criminology & Police Science, 1968, Vol 59, Issue 3, p408
- ISSN
0022-0205
- Publication type
Article