The Persuasion Effect: A Traditional Two-Stage Jury Model
The Persuasion Effect: A Traditional Two-Stage Jury Model
An influential model of jury behavior in American criminal cases suggests that in an initial stage jurors vote independently on the guilt or innocence of the defendant and that, in a second "persuasion stage," the outcome of that vote will affect the probable final distribution of votes to acquit or convict. Thus the following factors yield various statistics about the behavior of the jury system: (1) the incidence of guilt among those tried, (2) the probability that the jury will convict a "truly guilty" defendant, (3) the probability that the jury will acquit a "truly innocent" defendant, (4) a mapping between the distribution of initial vote and final vote (the "persuasion effect") and (5) an "outcome mapping" (legal rules determining the types of final votes that result in the conviction of the defendant, the acquittal of the defendant, or a so-called "hung jury").
This Demonstration simulates this traditional two-stage model of jury behavior. You may set a prosecutorial accuracy parameter (the probability that the person tried is "truly guilty" of the crime), two jury accuracy parameters (the probability they will convict the guilty and the probability they will acquit the innocent), rules on jury size and outcome mappings, and two parameters that characterize the persuasion effect. The Demonstration produces a plot showing the distribution of votes to convict at the first stage and a plot showing the distribution of votes at the second stage, and it makes extensive use of Bayes' theorem to compute various statistics on the justice system.