Jury Simulation

• Activity Number: 448
• Type: Contributed
• Date/Time: Tuesday, August 11, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Section for Statistical Programmers and Analysts
• Author(s): Catherine Durso*
• Companies: University of Denver
• Keywords: jury ; simulation ; bootstrap

Abstract

The effects on juries of varying strategies on the part of the plaintiff and the defendant have economic implications and implications for legitimacy of jury damage awards. An experiment modeling juror decisions and awards collected verdicts and awards from participants exposed to simulated trials with randomized plaintiff and defense strategies. These juror data were assembled into jury simulations using several models for jury decisions given the individual jurors' verdict and award data. One such model for a twelve person jury uses an award of zero if six or fewer jurors favor the plaintiff, and an award equal to the median if at least seven jurors favor the plaintiff. Given juries drawn as weighted samples from the participants and a jury decision and award model,the distribution of awards may be calculated, along with summary properties of the distributions. The effects of strategies may be compared for the respondent pool. Bootstrap methods allow analysis of the sensitivity of the comparison to the specific sample.