Voir Dire

By Bruce Barrett in the November 2007 issue of the American Statistician (version is gated):

The process of jury selection typically requires opposing counsel to reduce a pool of prospective jurors to the prescribed jury size by alternately excusing or striking individuals from service. These decisions, called peremptory strikes, are executed without the need of revealing any underlying rationale. However, recent U.S. Supreme Court rulings have held that attorneys may not exercise their peremptory strikes to systematically exclude prospective jurors on the basis of race or gender. The first step in establishing a charge of such improper bias requires the challenging party to show evidence that his or her opponent’s strikes are inconsistent with random consideration of these protected characteristics. Since court procedure dictates that there is some alternating between Prosecution and Defense in the striking process, choices for each side impact those of the other, and a simple comparison of the jury pool with the peremptory strikes is insufficient for establishing any inference of bias. For these situations, we present a methodology for assessing the neutrality of juror strikes, based on the Poisson binomial distribution.

This is a neat idea since it’s really a probability-based approach to understanding how two sets of decisions impact one another in a sequential decision-making process. Of course, there’s a long history of building detection-type statistical techniques in this journal (years ago, I remember one on detecting cheating in jai alai), but this is one of the first I’ve seen on this particular problem.

I teach the hypergeometric model in my masters-level classes on statistical decision-making and am always looking for nice applications. Leads to any others are most appreciated!