Okay, my title is purposefully provocative, but it is not an exaggeration. Based on a recently released, first-of-its-kind, comprehensive study on the effectiveness of voir dire following common practices in civil trials (Campbell et al., 2020), the verdict is that voir dire is often of no value when it comes to discovering or curing bias. “Information provided by generic, minimal voir dire questions,” the research team concludes, “did not prove useful in predicting jurors’ judgments, and thus is of no value in informing decisions about whether to exclude such individuals from a jury.” The study, a legal research paper produced by the University of Denver Sturm College of Law (available in a free download at present) confirms what most litigation consultants have long believed: that you don’t fix bias by asking jurors if they are biased, and by having them promise to be fair.
Even though all jury trials rest on the premise of an impartial jury, courts vary wildly in the procedures they follow to protect that impartiality. While some courts give attorneys great latitude to ask questions on individual attitudes and experiences, and are appropriately skeptical of jurors’ promises to “set aside” an expressed bias, other courtrooms (including most federal courts and many states) rely on limited questioning by the judge (“Do you know the parties? Is there anything in the case description that would cause you to be unfair?,” etc.) in a way that prioritizes speed and reserves cause challenges for only the admitted or patently obvious biases. These courts tend to assume that potential jurors are generally honest, and when the potential for bias is brought up, they will either control it, or they will know themselves well enough to admit to an inability to control it. The new research study undercuts these practices and assumptions; however, finding that when voir dire is limited — when it is generic rather than specific to case issues, and when it principally relies on the potential jurors themselves, rather than probing questioning to identify and correct biases — it is simply not useful. As the researchers conclude, “the questions used by many courts do nothing to predict bias and do nothing to cure it.” In this post, I will take a look at the research and its implications for civil jury selection.
The Study: What Works in Voir Dire?
The research team includes Denver University law professor John Campbell, Cornell law professor Valerie Hans, and five other scholars. Given the importance of identifying and addressing juror bias, it is surprising that there has been relatively little research on the effectiveness of voir dire, particularly in a civil trial context. The researchers observe, “despite the soaring language applied to the good of juries, we know little in the civil setting about how jury selection (voir dire) impacts the aspirations of our founders and the guarantees of our courts. Indeed, there is a spectrum of views on how much voir dire is necessary or whether it is even necessary at all.”
To test the effectiveness of voir dire in its various forms, they recruited 2,041 mock jurors who were exposed to background facts, opening statements, testimony, and closing arguments from one of three real cases (wrongful birth, medical failure to diagnose, and insurance bad faith). The study manipulated the extent of voir dire, with jurors receiving no voir dire, minimal and generic voir dire, or extensive case-relevant attitude and experience-related voir dire. They also varied whether jurors heard a message on rehabilitation (the importance of setting aside a bias) before being asked whether they could be unbiased.
I’d encourage those who study, consult on, and practice voir dire, to review the full conclusions of the study. But the short version is this: Minimal voir dire was not useful in discovering bias, and rehabilitation is not useful in curing it.
There are a few implications to this.
The Implications:
Don’t Trust Truncated Voir Dire
When voir dire is not able to include an uninhibited focus on jurors’ individual attitudes and experiences, it is not useful. Minimal voir dire unearthed factors that explained less than 1 percent (.6) of the variation of verdicts, and only 2 percent of the variance in damage awards. Extensive voir dire was able to identify factors that explained 20 percent of variance in verdicts and 19 percent of the variance in damage awards. Given that we would never expect a preexisting bias to explain all of the variation (because jurors do understand and react to case facts), 20 percent is a dramatically-high amount, and the greater than twenty-fold advantage of extensive over minimal voir dire is massive.
The researchers also found that, in the truncated modes that simply ask jurors if they are biased, very few will admit to a bias, and the few who do are no more biased than the rest of the group.
The irony is that, when voir dire is so limited that it does not provide attorneys with a solid basis for a peremptory strike, attorneys will have no choice but to rely on the panel’s demographics — factors like race, gender, and age that are not just improper, but unhelpful as well. The study found, “virtually none of the demographic factors we had included in our pre-screening questionnaire predicted verdicts or damage awards to a statistically significant degree.” The only exceptions were highly idiosyncratic: Widowers gave higher damages, and parents were more likely to find bad faith. Broadly speaking, when the panel’s demographics are all you’ve got to go on, you have no solid foundation for exercising strikes in any case.
Don’t Trust Rehabilitation
“Once a juror identifies a bias,” the team bluntly concludes, “they should be excluded. Rehabilitation does not work.” While they refer to some research showing that awareness of a bias can help in some situations, they also note that in other situations, that awareness simply increases the salience and accessibility of the bias. Telling them to ignore it, or put it out of their heads, is ineffective, just as telling you “Don’t think about a white bear” is a sure way to put a white bear into your head. In their own analysis, the researchers found that the rehabilitation message and question did not alleviate bias.
Fight for Expanded and Effective Voir Dire
As noted, extended voir dire, fulfills its functions by discovering much more of what determines bias on liability and damage awards. In addition to their own findings, the researchers also review literature, including an aggregation of 83 studies showing that attitudes predict behavior 69 percent of the time, as well as 272 studies showing the influence of authoritarianism on decision-making. Being able to explore these attitudes with potential jurors and not just ask if they feel like they have a bias, is critical. Also critical is returning the cause challenge to its proper role of being used to excuse citizens when there is bias, not just an awareness and an admission, but a reasonable basis to believe that they would not be fair to both parties. That needs to be grounded in a realistic and research-based view of juror behavior, and not based on an idealized and now-refuted view of jurors’ ability to perceive, admit to, and control their own biases.
The right to a jury trial is only as strong as our ability to work toward an unbiased jury. So, where it is a matter of judge discretion, ask for expanded attorney-conducted voir dire. If you are making a written motion, then this new research should be cited as part of the process of educating judges. They note, “The data raise serious concerns of the risks associated with seating juries that fail to meet constitutional, statutory, and procedural guarantees of impartiality when juries are formed without meaningful voir dire to eliminate biases.”
____________________
Campbell, J., Salerno, J. M., Phalen, H., Bean, S., Hans, V. P., Ross, L., & Spivack, D. (2020). An Empirical Examination of Civil Voir Dire: Implications for Meeting Constitutional Guarantees and Suggested Best Practices. U Denver Legal Studies Research Paper, (20-11).
Image credit: 123rf.com, used under license
