Although the Constitution does not specify jury size, over the years the number 12 has become widely accepted as the proper jury size. This research identifies variation in judicial output that results from competing operating structures of jury decisionmaking. A quasi-experimental design is used to address important problems of simulation, such as structural and functional and functional verismilitude. A total of 110 juries composed of nearly 1,000 jurors were studied. Participants were drawn from the jury rolls in Fayette County, Kentucky. Jurors were assigned to either 6-member or 12-member juries. Findings showed that viable minorities are more likely to cause juries to hang than are nonviable minorities. Viable minorities in six-member juries are also more successful in converting the majority than are viable minorities in a larger group. Larger juries hang more often than smaller ones do. The degree to which this avoids the conviction of an innocent person or release of a guilty person remains unanswered. Nevertheless, the Court was wrong in assuming that there are no differences in the behavior of 12-member and 6-member juries. Three pages of references are included, and a list of cited cases is provided. (Author abstract modified)