Mock juror sampling issues in jury simulation research: A meta-analysis

The advantages and disadvantages of jury simulation research have often been debated in the literature. Critics chiefly argue that jury simulations lack verisimilitude, particularly through their use of student mock jurors, and that this limits the generalizabilty of the findings. In the present article, the question of sample differences (student v. nonstudent) in jury research was meta-analyzed for 6 dependent variables: 3 criminal (guilty verdicts, culpability, and sentencing) and 3 civil (liability verdicts, continuous liability, and damages). In total, 53 studies (N = 17,716) were included in the analysis (40 criminal and 13 civil). The results revealed that guilty verdicts, culpability ratings, and damage awards did not vary with sample. Furthermore, the variables that revealed significant or marginally significant differences, sentencing and liability judgments, had small or contradictory effect sizes (e.g., effects on dichotomous and continuous liability judgments were in opposite directions). In addition, with the exception of trial presentation medium, moderator effects were small and inconsistent. These results may help to alleviate concerns regarding the use of student samples in jury simulation research.