May 20, 2013

Never Rely on Self-Diagnosis of Bias

By Dr. Ken Broda-Bahm: 

PulseThink of the situations where self-diagnosis wouldn't work very well: A police officer asking, "Do you think you were speeding?" or a doctor inquiring, "Do you believe your cancer is in remission?" Yet we still rely on self-diagnosis when trying to discover and eliminate bias in civil and criminal cases by essentially asking prospective jurors, "Are you biased?" A new study (Robertson, Yokum & Palmer, 2013) takes a look at whether we can rely on jurors to identify their own attitudes and know the sources of their own judgments well enough to say whether they would be biased or not. The result confirms our intuition: They can't. 

Of course, those of us with a background in social science have always known that a juror's self-appraisal isn't a perfect indicator of actual beliefs or behavior. But what this study shows is that the self-appraisal is not just imperfect, it's not even helpful. Joining a number of other studies pointing in the same direction, the three researchers from the University of Arizona have systematically demonstrated the basic unreliability of this central premise of voir dire. But rather than showing that the search for bias doesn't matter, the research tells litigators, judges, and trial consultants to become more savvy and more strategic when asking about bias. And they need to stop taking jurors at their word when it comes to the critical "Can you be fair" questions. This post will take a look at the study and share some thoughts on how legal practitioners should adapt. 

The Research: Don't Trust An "I Can Be Fair"

The authors are Christopher Robertson, David Yokum, and Matt Palmer, all with the James E. Rogers College of Law at the University of Arizona. Their 2013 paper can be downloaded here, and I should also point out that two out of three authors will be presenting a Juror Bias Symposium at the American Society of Trial Consultants' Annual Conference next month (June, 2013) in Las Vegas.

The study's conclusion should be striking to anyone who relies on self-report during voir dire, which is to say, every judge and most attorneys. That conclusion is that "juror self-assessments were not related to actual bias." The reason is that people are usually unaware of their own internal processes in rendering judgment. Given the choice between trusting what they say and trusting what they do, the preference in social science is to trust the latter. The law is another matter. "The Supreme Court has instructed courts," the authors note, "to use a simple method to determine whether jurors are biased: Ask them." 

To test whether this works, the authors conducted an experiment in several phases. First, they randomized the groups and exposed them to either irrelevant material or to prejudicial publicity in a medical malpractice civil trial context. In phase two, they asked participants the voir dire questions endorsed in the most recent applicable Supreme Court case (Skillings v. U.S.). In the third phase, all jurors watched and decided the same truncated medical malpractice case. Then finally, in phase four, the researchers looked at what the effect would have been of screening out jurors based on their answers in phase two. 

By presenting the prejudicial materials immediately prior to trial, the researchers purposefully designed the study to make it as easy as possible for jurors to self-diagnose their own state of mind. "If they are unable to accurately assess the impact of an article they read just moments before in a laboratory setting," they note, "it is even less likely that they could accurately assess the impact of an exposure occurring weeks prior." 

Predictably, the authors found that the pretrial reading created a bias. Even after being instructed to base their verdict on the trial alone, exposure to the publicity effectively doubled the odds of a verdict against the defendant and more than tripled the median damage awards. Despite this profound effect, however, a clear majority of jurors (87 percent of those in the exposed condition) denied bias and instead expressed a certainty that they would be able to decide the case impartially, considering only the evidence presented in the truncated trial. Comparing the verdict rates before and after screening the jurors, the rates were nearly identical. Without screening, 52 percent of those exposed to the prejudicial material found against the defendant. After weeding out those who said they couldn't be fair, 53 percent found against the defendant. Overall, the University of Arizona team found no statistical relationship between a likelihood of bias and a likelihood of admitting to that bias. 

The bottom line according to the authors: "Courts' reliance on such unreliable self-diagnoses to seat biased jurors may be a cause of wrongful convictions and wrongful impositions of civil liability."  

Instead, Escape from Self-Diagnosis

This research, along with the large number of complementary studies reviewed in the article, should lead to a serious reconsideration of whether we can continue to leave jurors as the judges of their own bias. We can't snap our fingers and change the legal standard, but we can push for change and we can alter our own practices as a result. 

Educate Your Judges

Appeals based on juror bias tend to center on the factual question of whether a juror was actually influenced or not. Despite this, it is surprising how little attention is given in the appeals to the social science that would answer that question. For example, though multiple courts have ruled on these issues of juror bias, the authors describe a very rigorous study looking at the lack of a connection between jurors' likelihood of being influenced by pretrial publicity and likelihood of admitting to that influence (Kerr et al., 1990). That study is now 23 years old, but it has never been cited by the Supreme Court or a Court of Appeals. Instead, the Court most recently upheld (Skillings v. U.S., 561 U.S. 2010) the adequacy of questions like, "Would any opinion you have prevent your impartial consideration of the evidence at trial?" and "Could you base a verdict only on the evidence at trial?" This is despite a lack of substantive social science evidence indicating that these questions are actually effective. "The Supreme Court has simply ignored the foregoing scientific literature," the researchers note, "and instead proceeded blithely, relying on jurors' self-diagnoses to affirm convictions and the imposition of civil liabilities, without requiring evidence about whether those self-diagnoses are reliable." 

At a minimum, this current study (Robertson, Yokum & Palmer, 2013) should be cited in any motion or appeal bearing on the adequacy of a potential juror's own disavowal of bias. 

Show More Creativity on Cause

Of course, attorneys seeking to remove a juror for cause need to, at some point, elicit whatever magic words your judge is applying in practice. But that doesn't mean that attorneys should reduce the standard to just that. Instead of beginning with the removal question, ask some lead-ins:  

Tell me more about [this biased belief of yours] 

How strongly or weakly are you committed to that view? 

How easy or how hard would it be for you to simply set that aside? 

Those questions are useful even if they don't get you to the goal by themselves. Once jurors have expanded on their views or committed themselves to a position, they're more likely to stick with that position when you or the judge asks them if they can be fair. 

In addition, the study authors suggest a few additional ideas for more creatively asking cause questions. "Juror self-diagnosis might also be improved if they focused on the causal question, and were asked to imagine themselves in the counterfactual position of having not been so exposed." In other words, "If you were to imagine yourself not having had that experience, do you think you would react to a case like this any differently?" Another practical idea the authors suggest is asking potential jurors to answer the hypothetical question of whether the given belief or experience would be likely to bias another person. I've also witnessed attorneys somewhat sneaking past the Golden Rule objection by asking the venire members, "If you were in court in circumstances similar to my client, would you want someone with that [experience/belief] on your jury?"  

Don't Let a Juror Promise Their Way Out of a Strike

Perhaps the most important piece of advice is that, even if the judge adopts a flawed standard for cause, don't replicate that same flawed standard when you are thinking about your own strikes. For example, here is a scenario I've been in several times in jury selection:

Me: You've really got to strike number 7. She said she hates companies like your's…

Client: But…she also said that she would set that aside and follow the evidence.

Me: And you believe her?

Based on this research, there is no indication that a promise of neutrality would make that juror any less biased. So when it comes to thinking of your strikes, think about their experiences and their attitudes, not their promises or their own self-assessments of how those experiences and attitudes would influence them.  

The researchers add one other interesting point: Judicial disqualification doesn't depend on self-assessed bias, but instead on the standard of whether "impartiality might reasonably be questioned." That would be a far better standard to apply to cause challenges. If it can't be applied in law, then at least apply it in practice when dealing with your peremptories.


Other Posts on Juror Bias: 

____________________ Robertson, C., Yokum, D., & Palmer, M. (2013). The Inability of Jurors to Self-Diagnose Bias 7th Annual Conference on Empirical Legal Studies Paper, 12-35


Image Credit:, used under license


Related Posts Plugin for WordPress, Blogger...