«1. INTRODUCTION Throughout its history, the jury has attracted both stiff criticism and unqualified praise, viewed by turns as an incompetent, ...»
16. Empirical analysis of juries in tort cases
Shari Seidman Diamond and Jessica M. Salerno
Throughout its history, the jury has attracted both stiff criticism and unqualified praise,
viewed by turns as an incompetent, naïve, and biased decision maker and as an astute
repository of folk wisdom and common sense. Here, we examine how the American jury
actually behaves in the tort cases that produce the majority of civil jury trials. The evidence
shows that juries usually use reasonable strategies to evaluate the conflicting evidence they are given. They are active problem-solvers who typically work to produce defensible verdicts. Nonetheless, juries are occasionally misled by the strategies they rely on to reach decisions. Their decision making processes are also at times undermined by limitations, many of them unnecessary, imposed by the legal system.
In this chapter, we discuss the empirical literature relating to jury decisions in tort cases. For additional discussion on related topics in this volume, see the chapters on the empirical analysis of tort reform (Eisenberg, Chapter 20), tort damages (Viscusi, Chapter 18), and state court tort decisions (Heise, Chapter 2). Our empirical analysis of juries proceeds as follows. Empirical analysis involves a choice of both an empirical method and a theoretical framework for developing hypotheses. In Part 2 we (a) describe the various methods that are used to study jury behavior, and (b) outline the difference between the external perspective of economic theory and the internal perspective of jurors. In Part 3 we examine the behavior of the jury in tort cases, beginning with an overall view of the approaches jurors generally take in reaching decisions. We then assess how jurors deal with the key topics associated with charges of jury incompetence and bias: decisions on liability; expert testimony; corporate defendants; separating decisions on liability and damages; assessment of compensatory damages; decisions on punitive damages; and comprehension and application of the law. Finally in Part 4 we consider which aspects of less than optimal jury performance constitute insurmountable obstacles and which limitations can be overcome.
2. METHODS AND THEORYA. Empirical Methods Used to Study Juries The primary methods used for empirical studies of the jury are: (1) archival research on jury verdicts; (2) post-trial interviews with jurors; (3) experiments with real and simulated juries; (4) observations of real jury deliberations; and (5) surveys of other trial participants (judges, attorneys).
Empirical analysis of juries in tort cases
1. Archival research In archival studies of jury behavior, researchers gather data on case characteristics and jury verdicts from completed trials to analyze verdict patterns and the ability of the measured case characteristics to predict plaintiff win rates and damage patterns. Archival jury data are collected from two primary sources: courts (e.g., Cohen, 2009) and commercial jury verdict reporters (e.g., Daniels & Martin, 1995). Researchers have monitored changes in verdict patterns over time (e.g., Seabury, Pace & Reville, 2004) and used those patterns to assess the impact of various tort reform efforts on the jury (Eisenberg, 2012). Some widely used archival sources, however, are limited by selection biases that can produce misleading results. Most commercial jury verdict reporters depend on the reports of the litigating attorneys for information on their cases, which may result in incomplete data because attorneys (1) fail to provide information on some cases, and (2) do not provide accurate information on reported cases. Studies based on jury verdict reporters reveal a systematic overrepresentation of cases resulting in pro-plaintiff verdicts. Merritt and Barry (1999) found an underrepresentation in defense verdicts and lower damage awards (see also Sloan & Hsieh, 1990). More recently, Lee and Waters (2011) compared a group of verdict reporters with the actual court data collected in the 2005 Civil Justice Survey of State Courts and found that although the commercial verdict reporters did not display a significant bias in liability proportions, they, like media coverage (e.g., Bailis & MacCoun, 1996), were significantly biased in favor of the publication of cases with large awards.
In addition, a general limitation of most archival jury studies is that the case information available from even the best archival source typically includes only a limited number of variables (e.g., Goerdt, Ostrom, & Rothman, 1995). As a result, for example, studies using archival data to explain variations in jury verdicts may underestimate the variation that can be explained by characteristics of the evidence (e.g., the nature of the plaintiff’s injury) presented at trial.
2. Post-trial reports of jurors Scholars also use post-trial reports to gauge juror impressions of the evidence and to obtain a view of the deliberation process through the eyes of the participating jurors (e.g., Vidmar, 1995; Ivkovich & Hans, 1994; Sanders, 1993; Selvin & Picus, 1987). A limitation of this method is that juror post-trial reports, however sincere, are likely to be imperfect reconstructions of what occurred earlier (e.g., post-trial reports of initial impressions that changed in the course of deliberations). Some reports may be more accurate than others.
For example, in interviews conducted immediately after trial jurors are likely to have better recall and are less likely to have had post-trial influences affect their impressions. Similarly, if a substantial majority of the jurors from the same jury report, for example, that their jury discussed insurance, that report is more likely to be trustworthy than if a single juror reports that insurance was discussed. Jurors are likely to be less accurate in assessing and recounting the consequences of their discussion, (i.e., in reporting whether that discussion affected their verdict (Guinther, 1988)), particularly if admitting that it affected them would be socially undesirable because they were admonished not to consider it.
3. Experiments with jurors and juries Experimental analysis can take the form of laboratory experiments or field experiments.
Experiments provide the opportunity to isolate the effect of particular treatment (e.g., an 416 Research handbook on the economics of torts attorney’s ad damnum) on outcomes. In a well-designed experiment, the only difference between the set of participants receiving the experimental treatment and the set in the control group is the experimental treatment itself (e.g., the attorney’s ad damnum). Thus, any differences in outcomes can be directly attributed to the experimental treatment.
Beyond this, a well-designed experiment captures the core features of the environment the experimenter seeks to study and represents the participant characteristics that may affect participant responses.1 Thus, for example, the external validity of mock jury experiments may be weakened when the trial materials omit cross-examination or use college student participants. In a field experiment, actual trials are assigned to varying conditions (e.g., jurors in some trials are permitted to ask questions, while jurors in other trials are not permitted to ask questions). When these experiments involve jury trials, they require substantial cooperation from the courts and close monitoring to ensure that random assignment to experimental conditions is being successfully implemented (e.g., Heuer & Penrod, 1994;
Diamond et al., 2003). Not surprisingly, field experiments on jury trials are relatively rare.
In mock jury experiments, by far the most widely used method of studying the jury, jurors or juries are also randomly assigned to experimental conditions, but the trials are simulated (although they may be drawn from real trials). A key strength of mock jury experiments is that they combine control with the opportunity to examine the process of decision making. Simulated trials vary in the degree to which they approximate the conditions of a real trial. The closest approximations involve videotaped trials that include witnesses, arguments and instructions, jurors who participate during their jury service, and deliberations. The extent to which behaviors observed in the less elaborate simulations, or indeed in the more elaborate simulations, mirror the behavior of real juries is unknown.
Some evidence, however, shows substantial correspondence between verdict results from simulations that vary in realism, suggesting that not all differences will affect the results (Bornstein, 1999). Nonetheless, if experiments use minimalist stimulus materials that omit elements likely to affect responses, generalization from those experiments can be misleading.
4. Observing jury deliberations An obvious way to study the behavior of any decision making body is to watch it as it reaches its decisions. An early effort to directly monitor several juries by audio taping their deliberations (Kalven & Zeisel, 1966) was short lived. It ended in an uproar when one of the tapes was played at a judicial conference (Katz, 1972). The result was legislation prohibiting the recording of jury deliberations in federal courts and in most state courts. Aside from the analysis of a few deliberations videotaped for television (e.g., Maynard & Manzo, 1993; Manzo, 1996), which might not generalize to jurors who did not agree in advance to have their deliberations televised (American Judicature Society, 2003), researchers have not been permitted to observe or record jury deliberations. In the Arizona Jury Project—the single exception to this pattern—a team of researchers, with permission from the Pima County Superior Court in Arizona, jurors, attorneys, and litigants, was allowed to videotape 50 actual civil jury trials and deliberations as part of a For additional discussion of jury simulation experimental methodology, see Diamond (1997);
for a more general treatment of experimental methodology, see Shadish, Cook, & Campbell (2001).
Empirical analysis of juries in tort cases field experiment examining the effect of recent jury reforms on jury behavior (Diamond et al., 2003). Conclusions about jury decision making processes drawn from this dataset are discussed in Parts 3A and H and Part 4 below.
5. Comparing judges and juries Another empirical approach bridges these different data sources (e.g., archival, experimental) and is based on a comparison between jury and judges’ decisions, rather than focusing exclusively on factors affecting jury decisions. Scholars have compared the trial verdicts of judges and juries in an attempt to assess whether they have different verdict patterns (e.g., Clermont & Eisenberg, 1992; Eisenberg, Rachlinski & Wells, 2002; Hersch & Viscusi, 2004; Heuer & Penrod, 1994). This comparison is important given that inherent in any praise or criticism of the jury system is an implicit comparison to the only viable alternative: a judge. As Clermont and Eisenberg concluded, the selection processes that lead to a bench versus a jury trial always leave doubt about the comparability of the sets of cases decided by each of them. Because litigant choices determine whether a jury or judge is the decision maker, there are likely to be differences between the types of cases that juries (versus judges) hear which may explain verdict patterns. As a result of this selection bias, conclusions about both similarities and differences may be misleading.
Kalven and Zeisel (1966) developed a creative method for comparing jury and judge decisions that largely overcomes this selection problem. They asked trial judges to report how they would have decided each jury trial they presided over if it had been a bench trial. To the extent that judges generally filled out the questionnaire before knowing the jury’s verdict, this method provides an independent judicial decision in precisely the same case that the jury decided. Kalven and Zeisel asked the judges to indicate their preferred verdict before the jury returned, but did not ask whether they actually did. A few researchers (Diamond et al., 2003; Hannaford, Hans, & Munsterman, 2000) have used the same method, but have also asked if the judge answered the question before the jury returned its verdict.
A third form of judge-jury comparison uses lay and judicial participants in parallel versions of the same survey or experiment. This method enables the researcher to compare judge and jury responses to the same trial stimulus materials. For example, Landsman and Rakos (1994) tested the ability of judges and jurors to ignore inadmissible evidence. In an experiment, they produced three versions of a vignette involving a products liability tort case that differed only in whether a piece of potentially biasing evidence had been admitted, whether the decision maker learned about it but it had been ruled inadmissible, or whether they did not learn about it. Exposure to the biasing information, whether inadmissible or not, increased liability verdicts of both the judges and jurors.
6. Combining methods When we combine the results from archival, experimental, post-trial interviews, and observational studies of the jury, the picture that emerges does not resemble the extreme image of the runaway jury. Nor does it reflect an economic ideal decision maker primed to reach decisions that produce optimal care.
418 Research handbook on the economics of torts
B. Role of Theory
Good empirical analysis is informed by theory. We focus here on two theoretical perspectives that lead to contrasting predictions about jury decision making: (a) traditional economic theory, predicting that juror decision making will conform to rational choice theory, versus (b) psychological theories, predicting ways in which jurors’ decision making should systematically deviate from rational choice theory.
1. An economic perspective Traditional economic theory assumes that individuals behave in ways that maximize utility. The implications of this theory depend on whether individuals gain utility from performing a public service and promoting justice. If people care only about their own direct welfare, citizens would avoid jury duty and, when unable to escape service, minimize the time they spend as jurors with hasty deliberations aimed at swift verdicts.