Ask and What Shall Ye Receive? A Guide for Using & Interpreting What Jurors Tell Us

University of Pennsylvania Journal of Law and Social Change

 14 U. Pa. J.L. & Soc. Change 201

ASK AND WHAT SHALL YE RECEIVE? A GUIDE FOR USING AND INTERPRETING WHAT JURORS TELL US

By Barbara O'Brien*, Samuel R. Sommers** and Phoebe C. Ellsworth*** 

BIO: * Associate Professor of Law, Michigan State University College of Law. Ph.D., 2007, University of Michigan; J.D., 1996, University of Colorado School of Law; A.B., 1993, Bowdoin College. Many thanks to the MSU College of Law's outstanding librarians for their assistance and support. During preparation of this Article, this author was supported by a summer research grant from the Michigan State University College of Law.

** Associate Professor of Psychology, Tufts University. Ph.D., 2002, University of Michigan. B.A., 1997, Williams College. During preparation of this Article, this author was supported by a Tufts University Faculty Research Award.

*** Frank Murphy Distinguished University Professor of Psychology and Law, Radcliffe; Ph.D. 1970, Stanford University; University of Michigan. A.B. 1966. During preparation of this Article, this author was a Visiting Scholar at the Russell Sage Foundation.

[*202]

INTRODUCTION

Juries' decisions have profound consequences, yet the decision-making process is largely opaque. Researchers have therefore sought to understand as much as they can about how jurors perceive their task, interpret evidence, and render decisions. n1 One obvious approach to these questions is the direct one: simply ask jurors what they thought and why they reached the decision that they did. This straightforward tactic is certainly appealing, but it carries risks that can render the information it produces suspect. At the same time, however, some research questions are most accessible through this method. Rigorous, multi-jurisdictional studies have greatly advanced our understanding of jury decision-making, particularly when they combine self-report and archival data. n2 Although many researchers who use interviews are aware of the limitations and advantages of the method, n3 no thorough and systematic evaluation of the general juror interview method has been published to date. Accordingly, we envision this Article as a guide for researchers considering the use of juror interviews in their own work, as well as for legal practitioners, scholars, and policymakers who seek to draw conclusions based on studies using this method. As judges and legal policymakers increasingly recognize the value of empirical research to guide trial strategy and reform, it becomes even more critical to conduct these types of studies properly and to draw the appropriate inferences from them. Without adequate attention to methodological limitations, overconfidence can easily distort the conclusions we draw from what jurors tell us.

In a recent conversation with a conscientious judge, one of the authors mentioned the [*203] consistent finding that jurors have far more difficulty understanding and applying legal instructions than they do understanding the facts of a case. n4 The judge responded that although jurors in general may have trouble understanding the law, in his courtroom they do not. When asked how he knew, he explained that he regularly includes the following question in his post-trial questionnaire: "Did you have any difficulty with the Court's instructions on the law? YES___ NO____," and that an overwhelming majority of jurors check "NO." Indeed, in spite of the clear behavioral evidence that jurors find it difficult to apply legal instructions, few judges with whom we have spoken see juror comprehension as a problem in their own courtrooms.

Discrepancies such as this one - between what jurors tell us about their decision-making and what archival and observational data demonstrate - raise questions about the reliability of information obtained from juror interviews. n5 Nevertheless, the use of juror interviews is widespread. Trial consultants advise litigators to interview jurors after a trial to "shed light on arguments the jurors found most persuasive, witnesses who were most convincing, demonstrative evidence that was most understandable ... as well as provide a clearer illumination of the process of juror decision making and juror reaction to the key features of the trial." n6

On the other side of the coin, another of the authors was recently asked to serve as an expert for a pre-trial defense motion in a capital case. The prosecution took issue with defense assertions founded on studies showing that capital jurors often make up their minds about sentencing before guilt has been determined and that the race of the parties can affect capital juries' decision-making. The prosecution argued that these studies were fatally flawed because they were based in part on jurors' self-reports. However, the limitations of interview studies must be considered in light of the content of the jurors' answers and the purpose of the research. For instance, self-presentational concerns can bias jurors' answers to questions; people are generally motivated to present themselves in a positive light. But when the interviews suggest that a substantial number of jurors go against the judge's instructions by deciding the sentence before the trial is over, such evidence of premature judgment is quite persuasive because it occurs in spite of jurors' motivation to make themselves look good.

Researchers rely on the information jurors report in post-trial interviews for a wide range of empirical objectives. In Part I of this Article, we review the psychological research on the pitfalls associated with self-report data. Problems associated with the passage of time, people's desire to present themselves in a positive light, and lapses of memory are just a few of the factors [*204] that can threaten the validity of findings premised on jurors' self-reports. In Part II, we review the literature of published studies relying on jurors' responses to interviews and questionnaires. We synthesize these findings and assess the value of the self-report method for answering different kinds of questions in light of the various threats to validity outlined in Part I. In Part III, we provide a detailed case study of one particular question: the influence of race on jury decision-making. Specifically, we compare findings from juror interview studies with those obtained using behavioral measures, including data from a study of race and mock jury deliberations. We end by evaluating the types of questions that are most (and least) appropriate for assessment via juror interviews, and offer suggestions for their more effective use. Our primary aim is neither to encourage nor to discourage the use of juror interviews, but rather to provide a review and critique of this methodology that may serve as a reference for those who conduct and evaluate such studies in the future.

I. POTENTIAL THREATS TO THE VALIDITY OF SELF-REPORT DATA

The appeal of the juror interview methodology is clear. Many empirical questions cannot be addressed by archival analyses of trial outcomes, and talking with jurors about real cases avoids problems raised by examinations of mock jurors' responses to simulated trials, such as whether or not the results are generally applicable. n7 As Professor Vidmar argues, even carefully designed jury simulation studies tend to lack the rich complexity of even the simplest actual trial. n8 Moreover, the likelihood that mock jurors are at least somewhat less invested in their task than real jurors raises questions about the generalizability of simulation studies.

But it is well documented that self-report measures - despite their popularity across disciplines - have inherent limitations. n9 It is important for jury researchers - as well as policymakers and litigators who rely on their findings - to be aware of the potential threats to validity posed by reliance on self-report measures and to recognize that their value depends very much on the nature of the specific research question. Specifically, psychologists have identified several factors that threaten the validity and reliability of self-report data. n10 To set the stage for our review of the studies relying on jurors' self-reports, we first identify several factors that threaten the validity and reliability of self-report data and then briefly consider their potential relevance to the jury context.

[*205]

A. Passage of Time

Interviews often require research participants to reconstruct events or describe what they were thinking at an earlier point in time. Participants' responses are only as accurate as the memories underlying them, and memory is less reliable the longer the delay between event and interview. n11 Moreover, information learned after an event can influence people's memory of it. n12 In a legal context, having joined the jury's unanimous verdict, a juror may reinterpret earlier events as being more consistent with that outcome. n13 Put differently, what jurors think about a case after the deliberation process is often not what they thought about it at the time. Given that jurors are generally unaware of these subtle shifts, n14 the utility of retrospective self-report is more suspect when a researcher seeks to capture a juror's attitudes at an earlier time. Complicating matters further is the great variability in the lapse of time between jury service and interview, across and even within studies. For example, one published study was based on interviews conducted between three months and seven years after trial, n15 and another study was based on juror interviews conducted immediately after trial. n16

B. Inaccessible Information

Some interview questions are problematic because they ask for information that respondents cannot realistically be expected to provide. In a seminal article, psychologists Nisbett and Wilson argued that the value of introspection for determining the reasons behind thoughts and behaviors is limited because people are often unaware of the true influences on their preferences and actions. n17 Even more problematic, this ignorance does not prevent people from offering plausible explanations for why they do what they do. n18 In one experiment, participants who memorized "ocean-moon" as part of an ostensibly unrelated earlier task were twice as likely as other participants to say "Tide" when asked to name a laundry detergent. n19 When questioned [*206] about this response, participants did not mention the earlier, semantically relevant word pair, but rather offered plausible, more personal explanations such as "my mother uses Tide." n20 These findings raise concerns about questions that require jurors to report on higher order cognitive processes (e.g., "What evidence had the greatest impact on you?") or to identify factors that influenced decision-making (e.g., "To what extent did the jury's gender composition affect deliberations?").

C. Self-presentational Concerns

Even when a respondent's memory is strong and an interviewer's questions focus on accessible information, the validity of self-report measures is threatened by the possibility that people may wish to conceal information. The potential problems posed by respondents' motivation to "look good" - or, at least, to avoid "looking bad" - are illustrated by studies in which participants who believe they are hooked up to a lie detector give more accurate and less socially acceptable answers to questions than do naturally responding participants. n21 In a legal context, self-presentational concerns could prevent jurors from admitting, for example, that they did not understand complex scientific evidence or that they were influenced by inadmissible considerations. Concern with self-presentation is likely exacerbated when a court official or authority figure conducts the interview, as in the judicial example at the beginning of this article. n22

D. Response Rate

Another important consideration in evaluating self-report data is participant response rate. When judges or attorneys poll a jury at the end of a trial, they have the advantage of speaking to jurors before they leave the courthouse. Researchers conducting post-trial interviews rarely have that luxury; generally, they must track down as many jurors as possible and persuade them to talk about their experience, and they hardly ever manage to talk to all of the jurors. n23 In addition to practical difficulties, researchers are also faced with the problem that the jurors they successfully recruit may differ from those they cannot locate or convince to cooperate. If [*207] systematic differences exist between those who respond and those who do not, interview data will not represent the experiences of all jurors. n24 For example, a researcher interested in juror satisfaction might interview 80% of jurors from a sample of trials, but if the 20% who refuse to participate do so, at least in part, because they hated their experience, the researcher's conclusions will paint an overly optimistic picture of juror satisfaction.

E. Question Wording

Even something as seemingly trivial as question wording can affect self-reports. n25 For example, Loftus and Palmer showed participants films of a car accident and asked them to estimate the speed of the cars on impact. n26 Those asked how fast the cars were going when they "smashed" into each other gave significantly higher estimates than did participants asked about when the cars "hit," "bumped," "collided," or "contacted" each other. n27 Moreover, these effects were still evident one week later: participants who had been asked the "smashed" question were most likely to agree that they had seen broken glass. n28 Along similar lines, a question about deliberations might produce different responses if phrased "how easy was it for the jury to reach a verdict?" versus "how difficult was it to reach a verdict?" Different wordings could also influence jurors' responses to subsequent questions and later memories of deliberation.

II. REVIEW OF RESEARCH USING JUROR SELF-REPORT

To understand the advantages and limitations of juror interviews, it is important to consider the objectives of the researchers. That is, what exactly do interviewers seek to find out from jurors? Are jurors asked to recall past events, report current perceptions, reconstruct thought processes, or something else entirely? Even more important, what sorts of conclusions do researchers seek to draw based on these self-reports? In this section we review the different objectives of researchers who have used juror interviews or questionnaires. In examining what researchers seek to achieve with this method, we identify particular issues of concern in light of psychological findings regarding the limitations of self-report data, and offer suggestions for future use and interpretation of the juror interview method. n29

[*208]

A. What are Jurors' Subjective Experiences?

Our review of the literature indicates that a principal use of interviews is to examine jurors' emotional or subjective experiences. Interviews have focused on the stress jurors experience when serving on a disturbing case, n30 their feelings about the fairness of various courtroom procedures, n31 their satisfaction with the deliberation process, n32 and their perceptions of the difficulty of the task in general. n33 Sometimes researchers have attempted to discover the factors that predict jurors' experiences. Professors Eisenberg and Wells, for example, found that [*209] compared to jurors from juries that imposed the death penalty, jurors from juries that imposed life sentences reported more impatience with deliberations and a stronger sense that the jury failed to follow the judge's instructions. n34

Interviews have also been used to learn how jurors' subjective experiences vary with their demographics or jury composition. Professor Marder analyzed questionnaire responses of jurors who had served in criminal trials to examine how gender, age, and racial diversity affected deliberations. n35 She found that gender diversity was associated with reports of reduced hostility, greater support among jurors, and the perception that deliberations were thorough; racial diversity, however, was not related to these outcomes. n36 Researchers from the Capital Jury Project used interview data to conclude that Black jurors tended to perceive their White counterparts as predisposed toward the death penalty and indifferent to mitigating evidence, n37 whereas members of all-White juries reported more unity and consensus. n38

How reliable are assessments of subjective experience? On the one hand, these questions are well within the limits of jurors' cognitive ability in that they do not require reconstruction of complex past events or analysis of higher-order mental processes. On the other hand, threats to their validity clearly exist. Jurors may be motivated to portray themselves in a certain light - unwilling, for example, to admit that they were confused by the judge's instructions. The passage of time may also color perceptions; upon post-trial reflection, jurors may grow increasingly confident or doubtful about their jury's verdict. Although self-report is the most straightforward way to assess subjective reactions to jury service, there are no guarantees that reports of subjective experiences are accurate. At the very least, researchers should think carefully about the cognitive and motivational factors that might distort responses and look for ways to counteract - or at least to assess - them. Likewise, policymakers interested in reforming aspects of jury service to improve the experience for jurors should consider such issues in attempting to draw conclusions from this research.

B. Which Juror Characteristics Predict Trial Perceptions and Votes?

Another frequent objective of juror interview studies is to find out how different kinds of people decide different kinds of cases by asking jurors how they voted during deliberations. Usually jurors are asked to recall not just their final vote, but also preliminary votes cast during the course of deliberations. n39 Other studies have examined the relation between jurors' general [*210] legal attitudes and their voting tendencies. n40 For example, Capital Jury Project ("CJP") jurors' belief that death is generally the right punishment for murder predicted both their first and final votes for death. n41 Other studies have asked jurors their opinions on topics such as defendants' rights, n42 the civil tort system, n43 and whether they consider themselves "tough on crime." n44

Of course, the data obtained in these studies are only as reliable as the voting history recollections on which they are based. As alluded to above, one concern is that the longer the time between deliberation and interview, the more likely it is that random memory error and systematic bias (e.g., recollections of initial votes becoming skewed in the direction of the final verdict) will emerge. This is particularly the case when jurors are asked to recall several votes over the course of a long deliberation. Clearly, passage of time is a factor that should be considered when evaluating responses.

Answers to questions about general legal attitudes may also be influenced by concerns about self-presentation. Jurors may avoid admitting beliefs that they consider unpopular, especially if interviewed in public or by an authority figure such as a judge. A less obvious threat to validity is the assumption that general legal attitudes are stable and unaffected by recent trial experiences. n45 When a juror states that she opposes the three-strikes law, for example, a researcher may assume that she had this attitude before she was called for jury service. However, attitudes reported after jury service may have been influenced by the trial and deliberation. This is particularly likely when the topic is unfamiliar: someone who has never thought much about punitive damages may develop strong opinions on this subject during a tort trial. Thus, jurors' post-trial reports of attitudes may represent pre-existing opinions that influenced deliberations, or they may reflect ideas that developed during the trial itself. Typically, the interviewer has no way to distinguish between these possibilities. Researchers who see this as a potential problem for the attitudes they want to measure should look for ways to assess these attitudes before the trial.

[*211]

C. What Do Jurors Cite as Important to Their Decisions?

Researchers have often sought to identify the factors that influenced jurors' decisions. n46 One way to do this is to ask jurors for unconstrained, open-ended narrative accounts of their experience. n47 This method can reveal information that the researcher would never have thought to seek. For example, in the CJP interviews, a White juror reported that the jury feared that the Black defendant's family would retaliate if they voted for a death sentence, prompting jurors to ask for a court escort to their cars after the verdict. Another White juror mentioned that two Black men stood at the back of the courtroom during trial, which she interpreted as an attempt to communicate to the Black jurors that they should stand by the Black defendant because of their shared race. n48

Another way that researchers have sought to identify influential factors has been to ask jurors about specific aspects of the case. n49 For example, National Center for State Courts ("NCSC") interviews revealed that post-trial doubts about police officers' credibility were associated with votes for acquittal. n50 Many researchers have been particularly interested in jurors' responses to expert witnesses. n51 As one example, Professor Sundby analyzed CJP interviews and concluded that experts are most persuasive when their testimony helps provide a context for the evidence presented by other witnesses. n52

[*212] Other researchers have attempted to find out how the topics discussed during jury deliberations influenced verdicts. n53 For instance, one study found that eighty-five percent of jurors said that they discussed attorney fees when calculating damages, even though they had been instructed not to. n54 In the criminal context, CJP interviews have revealed a number of topics common to death penalty deliberations. The defendant's remorse n55 and future dangerousness n56 were among the most frequently discussed issues during penalty deliberations, as was the victim's character and role in crime. n57 On the other hand, jurors reported that they tended to neglect mitigating factors during penalty phase discussions, as they saw such evidence not as providing a context in which to evaluate the crime, but rather as an attempt to excuse it. n58

Finally, in an effort to understand how jurors reached their decisions, some researchers have simply asked them directly. n59 For example, Professors Ivkovich and Hans asked jurors [*213] about the impact of expert witness testimony. Jurors claimed to have evaluated experts' credentials and motives, as well as the content and presentation style of their testimony. n60 In a criminal setting, researchers asked jurors whether they considered a defendant's criminal record in assessing guilt. Some jurors responded that although other members of their jury mentioned the defendant's known or suspected criminal history, it did not ultimately affect the jury's decision. n61 Professors Geimer and Amsterdam interviewed jurors in ten Florida capital cases about the reasons for their decisions. Jurors who voted for a life sentence cited lingering doubt about guilt as the most important factor in their decision. n62 Several jurors also said that their knowledge that the trial judge could override their recommendation made them more willing to vote for death. n63

How useful are juror interviews for identifying the factors that influence jurors' decisions? Open-ended narrative accounts like those used in the CJP study have revealed unexpected factors that the researchers had not thought of, and that are, thus, enormously valuable. The limitations of responses to more specific questions are similar to those for assessment of subjective emotional experiences. Jurors' evaluations of particular aspects of the trial may be colored by the experience of deliberation and the passage of time. For instance, it may be difficult to distinguish between jurors' perceptions of a witness's credibility at the time of the trial and their current, post-trial assessment. Jurors may be motivated to communicate that their decisions were based on the evidence, as opposed to the emotions or illegitimate considerations that actually influenced them. Thus, when jurors do admit to being swayed by fear, influenced by impermissible information, or to making their minds up before the end of the case, their responses are more trustworthy than when they say they decided based on "the strength of the evidence."

Most problematic are questions that require jurors to estimate the extent to which particular elements of the case affected their judgment. The utility of such questions depends on respondents' honesty and ability to remember not just a single fact (such as their vote) or a series of events (such as topics of deliberation), but also to analyze their decision process and infer the factors that influenced it. Given the dubious accessibility of such information, these questions [*214] regarding higher-order cognitive processes are problematic. n64 Even if obstacles such as memory decay, biasing post-event information, and self-presentational concerns are overcome, jurors may simply be unable to provide accurate insight into why they made the decision they did. n65

D. How Does the Deliberation Process Work?

Interviews have been used not only to investigate individual jurors' decision-making, but also to examine the process of jury deliberation. Researchers sometimes explicitly ask jurors to evaluate their jury's deliberations n66 or to report on their interactions with fellow jurors. n67 For example, one study found that most civil jurors formed at least a tentative opinion about the verdict before receiving jury instructions, with almost half admitting that they had made up their minds by that point. n68 CJP researchers observed similar evidence of early judgment, as many jurors admitted making up their minds about the penalty before the end of the guilt phase of the trial and before hearing any evidence regarding aggravation and mitigation. n69

At first blush, questions about the course of the deliberations do not seem particularly problematic. After all, like recollection of personal voting history, reporting the topics of deliberation discussions involves memory for specific facts. But unlike votes, discussions are not discrete events and, therefore, are more susceptible to inaccuracies in memory. Most interview questions about deliberations not only ask which topics were raised, but also how extensively they were discussed. Such questions require jurors to reconstruct the deliberation process, a more challenging task than remembering a handful of discrete events. Add to this the likelihood that jurors are influenced by self-presentational concerns - such as reluctance to admit that they failed to consider a significant issue or were biased by race - and the utility of such questions becomes more questionable, particularly when the research objective is to determine whether jurors did something they were not supposed to do.

E. How Good is Juror Comprehension?

Some interviews have focused on juror comprehension, often regarding technical or scientific evidence, and often in complex civil cases. n70 As an example, Professor Sanders [*215] interviewed jurors in a Bendectin case and found that many of them did not understand the complex evidence about the cause of plaintiffs' birth defects. n71

Jurors' responses to interview questions have also been interpreted as indicating misunderstandings of the law. For instance, researchers have asked jurors whether they made up their minds before the court's instructions, n72 how they perceived extralegal factors, n73 and whether they understood that criminal defendants are presumed innocent. n74 In another study, psychologists surveyed jurors who served on civil and criminal trials to test their understanding of both procedural and substantive legal issues. They found that jurors understood fewer than half of the judge's instructions and those who asked for the judge's help during deliberations showed better comprehension than those who did not. n75 Studies such as these identify particular shortcomings of the jury system and suggest possibilities for improving jurors' understanding of the law. n76

[*216] More generally, interviews have shed light on jurors' beliefs about their role in the trial process, some of which are consistent with legal requirements, some of which are not. For example, in a study of civil juries, many jurors expressed an ethic of individual responsibility, seeing themselves as guardians against frivolous lawsuits. n77 However, jurors may also talk about their experiences in a way that clearly demonstrates that they did not fully understand their role. For instance, CJP researchers found that many capital jurors erroneously believed that they were merely making a recommendation to the judge, who was the ultimate decision maker. n78

The value of interviews for assessing jurors' comprehension of the case, the law, and their own role varies depending on the strategy adopted by researchers. Questions that require jurors to evaluate their own comprehension are unlikely to yield informative responses: jurors may be unwilling to admit that the testimony or the law confused them or may not realize that they failed to grasp important issues. However, interviews can be extremely useful for understanding juror comprehension when jurors are asked directly about the law or complex evidence, allowing for later assessment of accuracy. If jurors describe complex testimony, their recollections can be compared to the trial transcript; jurors' responses to questions about the law can be compared to the judge's instructions. In this manner, questioning jurors about what the law or the evidence says is an excellent way to measure comprehension. In short, testing actual understanding is far more accurate and informative than simply asking jurors how well they understood.

F. How Well is the Legal Process Functioning?

Juror interviews have also been used to evaluate other aspects of the trial process. For example, in interviews conducted by the NCSC, jurors who served on deadlocked juries described concerns both about the fairness of the law and legal process generally, and about the adequacy of the evidence presented in their specific cases. n79 Interview studies have also assessed the effectiveness of voir dire for identifying biased jurors who nevertheless survived the jury [*217] selection process. n80 One study found that many jurors lied during voir dire - failing to disclose, for instance, that they or a family member had been the victim of a crime. n81

Interviews have also been used to study procedural changes. One such study tested the effects of an Arizona rule allowing jurors to discuss evidence before deliberations. This change did not affect the level of agreement between judges and juries, but juror interviews revealed that pre-deliberation discussion increased the perceived conflict on the jury, as well jurors' certainty about their pre-deliberation verdict preference. n82 However, the same cautions regarding the examination of juror comprehension apply to the assessment of procedural changes: it is more informative to combine interview data with other evidence about the effectiveness of new procedures than it is to rely on jurors' own evaluations of these changes.

For instance, psychologists Heuer and Penrod reported that although jurors who received written judicial instructions reported that they were helpful, their actual understanding of the law did not improve. n83 Researchers Young, Cameron, and Tinsley did not directly test jurors' comprehension, but their post-trial interviews suggested that jurors believed that they understood the law better than they actually did. n84 Although more than eighty-five percent of jurors reported finding the judge's instructions clear, the researchers found evidence of fundamental misunderstandings of the relevant law in thirty-five of the forty-eight trials. n85 Similarly, jurors allowed to take notes during trial reported greater satisfaction, n86 but their understanding of the law was no better than that of their counterparts, who had to rely on their memories alone. n87

[*218]

G. Summary

Juror interviews have been used to address a range of empirical questions about civil, criminal, and capital jury decision-making. For some research objectives, it is difficult to conceive of methods that do not involve self-report measures; other questions are more clearly amenable to archival or observational analysis. Many research objectives are compromised to some degree by the limitations of self-report data, such as the malleability of human memory, systematic bias in social judgment, the inherent difficulty of analyzing one's own complex cognitive processes, and concerns with self-presentation. However, the extent and precise nature of these threats to validity depend on the empirical questions under investigation and the kinds of answers respondents are asked to provide. To illustrate this point and to provide a more concrete synthesis of the conclusions above, we now examine more closely one particular research area: the influence of race on juror decision-making.

III. CASE STUDY: RACE AND JUROR DECISION-MAKING

Studying the role of race in legal decision-making poses a number of challenges, in large part because the effects of race on judgment are often subtle and even unconscious. n88 In addition, jurors' concerns about self-presentation are likely to be especially acute when they are asked about race. n89 Because of the variety of problems with self-report data in this domain, legal research on the influence of race serves as an illustrative example of the kind of situation where the juror interview method is especially risky. We begin this section by comparing the findings of interview studies concerning race with the results of studies that have relied on behavioral data. We then compare these two types of data within the framework of a single study, describing self-report and behavioral findings from a recent experimental investigation of the influence of the jury's racial composition on deliberation processes. n90

A. Juror Interview Studies Examining the Influence of Race

Some studies that have relied on interviews with actual jurors have examined racial differences in self-reported satisfaction with the trial process. Professors Antonio and Hans examined NCSC data and concluded that White jurors were more satisfied with their jury experiences than non-White jurors. n91 CJP researcher Bowers and colleagues reported that Black [*219] jurors were less confident than their White counterparts about the capital sentencing process, and more concerned that their jury had made a mistake. n92 In capital cases with Black defendants and White victims, Black jurors also said that they felt that the White jurors failed to understand Black defendants' circumstances and relied on racial stereotypes in interpreting the evidence. n93

Only a few interview studies have tried to address the causal influence of race on jury decision-making, although many interview, archival, and observational studies have examined the correlation between race and legal perceptions or outcomes. Professor Broeder interviewed jurors from four criminal trials with Black defendants, three of which ended in conviction, and found that several jurors who favored conviction expressed racial stereotypes during deliberations, and a few even suggested that the defendant's race was grounds for conviction. n94 In discussing the case with interviewers, some of the jurors made unsolicited prejudicial remarks about Blacks. n95

It seems unlikely that interviews conducted forty-five years later, in today's more politically correct era, would produce evidence of racial prejudice anywhere near as blatant as that found by Broeder. n96 Indeed, we know of no contemporary interview study that finds overt juror racial bias in response to direct questions. More recent efforts to assess the influence of race using interview data have typically compared the responses of jurors who served in trials with White victims or defendants to the responses of jurors who served in trials with Black victims or defendants. Sociologist LaFree found race effects in jurors' perceptions of the alleged victim in sexual assault cases, with White jurors viewing Blacks as less credible than Whites; n97 psychologist Edelman reported similar findings for victim race in CJP cases, but found no effect for race of defendant. n98 CJP researchers reported that White jurors in capital cases were more likely than Black jurors to say that Black defendants were dangerous, remorseless, and emotionally disturbed. n99 Beyond this handful of studies, there are scant published interview data that provide evidence - direct or indirect - of the ways in which the race of trial principals influences jurors.

Some researchers have examined the relation between the racial make-up of a jury and its decision-making. For example, using CJP data Bowers and colleagues found that a predominance of White male jurors was associated with a sentence of death, and that the presence of Black males on the jury was strongly associated with the imposition of a life sentence. n100 With [*220] regard to how a jury's composition affects its deliberation process, a traditional argument in favor of jury representativeness is that diverse juries will bring up a broader range of information and knowledge during deliberations than will homogeneous juries. n101 Yet juror interview studies provide little to no empirical support for this proposition. On the one hand, Professor Marder interviewed criminal jurors from Los Angeles and found no relation between jury racial composition and jurors' ratings of the nature and scope of their deliberations. n102 On the other hand, Bowers and colleagues found that all-White capital juries reported greater consensus during deliberations than did diverse juries. n103 In sum, juror interview studies provide few and inconsistent empirical findings concerning traditional assumptions about the influence of jury racial composition on deliberations.

B. Behavioral Studies Examining the Influence of Race

In contrast to the findings of interview studies, archival and observational studies using behavioral measures generally find that race does influence jury decision-making. n104 A recent meta-analysis of thirty-four mock jury studies involving over 7,000 participants revealed a statistically significant association between defendants' race and verdicts, with mock jurors less likely to vote to convict a same-race defendant than a defendant of a different race. n105 Another meta-analysis of fourteen studies with over 2,800 participants produced similar conclusions regarding the sentencing decisions of White mock jurors. n106 The race of a defendant has been found to influence not only mock verdicts, but also participants' explanations for a defendant's behavior, n107 ratings of a defendant's personality, n108 susceptibility to bias from inadmissible evidence n109 and pretrial publicity, n110 and how deeply participants process information about a [*221] case. n111

Fewer observational experiments have focused on the racial composition of juries, but these studies provide consistent evidence of the influence of race, too. Several simulation studies have concluded that the higher the percentage of Whites on a jury, the more likely the jury is to convict a Black or Latino defendant. n112 In fact, one study indicated that even the expectation of deliberating as part of a mixed-race jury can influence mock jurors' judgments. n113 These findings are consistent with analyses of actual jury outcomes, which also demonstrate a relation between jury demographics and verdicts. n114

In a behavioral study that explicitly addressed the relation between mock jury racial composition and deliberation content, Sommers compared the deliberations of all White to racially diverse mock juries. n115 Most of these jurors were recruited at a local courthouse during their actual jury duty. n116 Participants were randomly assigned to six-person mock juries, half of which were all White and half of which were racially diverse (four White and two Black jurors). n117 All juries watched the same video summary of a real sexual assault case involving a [*222] Black defendant and a White victim and deliberated as a group. n118 These deliberations were video-recorded, allowing for a subsequent comparison of jurors' actual behavior with their responses on post-deliberation questionnaires. n119 Sommers reported that the deliberations of diverse juries were longer, covered more of the evidence, and were more factually accurate than those of all White juries, consistent with traditional assumptions about jury racial composition. n120 Interestingly, these effects were not wholly attributable to the contributions of the Black jurors; White jurors were more thorough and accurate in their discussion of the trial evidence (and more willing to discuss controversial race-related issues such as profiling) when they deliberated on diverse juries. n121 Below, we offer previously unpublished analyses of the findings of this study to illustrate the types of research questions amenable to self-report, and those that are less so.

C. Comparing Interview and Behavioral Studies

Our literature review suggests that behavioral studies are more likely than interview studies to provide evidence of the influence of race on jury decision-making. There are several possible explanations for this disparity. Many behavioral studies examine mock jurors, not actual jurors. Of course, mock jurors render judgments that have no real consequences, unlike the decisions of actual jurors; perhaps race is more likely to be influential in a simulation because the stakes are lower than those in an actual trial. However, archival data regarding race and actual juror judgments parallel the findings of the mock jury experiments, suggesting that race influences real juries as well. n122

Another possible explanation is that juror interview studies are less able than behavioral experiments to capture the influence of race on legal decision-making. The best way to evaluate this possibility is to conduct research that includes both interview and behavioral measures. Sommers and Ellsworth presented groups of mock jurors with a summary of an assault case in which only the defendant's race varied. n123 When the incident in question was not racially charged, behavioral measures indicated that White mock jurors were influenced by the defendant's race, voting to convict a Black more often than a White defendant. n124 In unpublished pre-testing for this study, the researchers presented White participants with the same trial summary broken into five smaller passages. n125 After each passage, participants were asked to describe the events in question and their current reactions to the case. n126 Of the thirty-three Whites who took part in this pre-testing, only five (15%) mentioned race in their written responses. n127 Breaking this result down even further, of the 165 written comments (five responses [*223] from each of thirty-three participants), only seven (4.2%) mentioned race. In other words, even though behavioral data using the same trial summary indicated that Whites were significantly more likely to vote to convict the Black defendant than the White defendant, Whites almost never mentioned race in response to open-ended questions about the case.

The Sommers deliberation study described above provides an even clearer comparison of discrepancies between interview and behavioral measures. n128 The outcomes reported in the published Sommers paper were primarily behavioral, such as the number of case facts raised during deliberations and the number of inaccurate statements. n129 However, analysis of unpublished data from the post-deliberation questionnaire allows for a comparison between participants' self-reported perceptions of the deliberation process and the behaviors they actually exhibited while deliberating. n130

Which aspects of their experiences were mock jurors able to talk about most accurately? Perceptions of their own relative contributions to the deliberation process were reasonably accurate. Mock jurors' ratings of how active they were during deliberations were significantly correlated with the amount of time they actually spoke. n131 That is, the more active jurors said they were during deliberations, the more active they actually were. We also examined deliberation content by considering the forty-seven discrete evidentiary facts conveyed in the video trial. Jurors' ratings of how active they were positively (though non-significantly) predicted the number of different trial facts they raised during deliberations. n132

In addition to self-assessment, jurors were reasonably accurate in their evaluations of other jurors' participation. One question asked them to rate the extent to which "one or two jurors dominated discussion." In analyzing the videotaped discussions, we identified domination where one juror talked for thirty-three percent or more of the total deliberation time or two jurors combined for fifty percent or more of the speaking time. Analysis of self-reports indicated that jurors on juries that fit our dominance criteria perceived a greater amount of domination n133 than did jurors on other juries. n134

Self-reported data proved less useful with regard to perceptions of personal influence over fellow jurors during deliberations. Mock jurors were asked to rate the extent to which they influenced the other members of the jury, as well as the extent to which the other jurors influenced them. Individual jurors' ratings of their own influence on others were not reliably correlated with fellow jurors' perceptions. n135 It is not surprising that this comparison has less predictive utility, as these questions required participants to assess their own performance as well as to take the perspective of fellow jurors, thus inviting self-serving bias. Jurors were also asked [*224] to rate the extent to which they felt that they could speak their minds freely during deliberations, as well as the extent to which they believed their fellow jurors felt this way. Participants' ratings of their fellow jurors' perceptions were actually negatively (though non-significantly) related to the ratings provided by those other jurors. n136 Interestingly, jurors' self-ratings for this measure were strong, positive predictors of their perceptions of other jurors' feelings. n137 In other words, jurors were unable to gauge how free fellow jurors felt to express their opinions, relying instead on the assumption that everyone else felt the way that they did. Such overestimation of the extent to which others share one's own viewpoint has been demonstrated for a wide range of social perceptions. n138

Jurors were quite poor at evaluating the extent to which their jury considered all possible perspectives during deliberations. n139 We correlated these estimates with the amount of information actually discussed by each jury. Information was categorized into four types: facts related to the assault itself, the victims' identification of the defendant, forensic analyses conducted by police labs, and the judge's instructions on the law. Jurors' ratings of the thoroughness of their jury's deliberations were not correlated with three of these categories, n140 and were only moderately (but non-significantly) correlated with the number of facts their jury considered regarding the victims' eyewitness identification. n141

Another measure of the thoroughness of the deliberations was the extent to which juries discussed "missing" information, or evidence that they wished had been introduced at trial but was not (e.g., testimony from a child who witnessed part of one assault or fingerprint evidence). Jurors' reports of deliberation thoroughness emerged as significant negative predictors of the tendency to discuss missing evidence. n142 That is, the more thorough the jurors believed that the deliberations were, the less likely their jury was to have talked about the evidence they thought was missing from the prosecution's case. Moreover, self-reported deliberation breadth was a marginally significant negative predictor of the number of race-related conversations occurring on the jury. n143 In other words, jurors' assessments of the breadth of their deliberation discussions were typically no more accurate than chance, and on some dimensions, these ratings were actually inaccurate.

Finally, jurors were asked two direct questions about the influence of race on their decision processes: they were asked to rate the extent to which the jury's racial composition affected deliberations, and the extent to which "racial issues" influenced their decision-making in general. Not surprisingly, jurors' responses to these questions were at odds with the behavioral data. Although there were substantial differences between the deliberations of all White and racially diverse juries, n144 jurors rated their jury's racial composition as largely non-influential, n145 [*225] and responses to this question did not differ by jury composition. n146 Jurors also disagreed with the proposition that racial issues influenced their decision-making. n147 These responses did not differ by jury composition, nor by whether jurors had answered race-relevant or race-neutral pre-trial voir dire questions, n148 a manipulation that had a significant impact on individual jurors' pre-deliberation verdict preferences.

The interview and behavioral data from the Sommers study support the conclusion that juror interviews have limited value for identifying impact of race on legal decision-making. n149 Among other explanations, these limitations presumably derive from jurors' lack of awareness of the effects of race, as well as self-presentational concerns about talking about race. n150 Analysis of the Sommers study also illustrates some of the more general strengths and weaknesses of relying on jurors' self-reports. n151 Interview questions about factual recollections of juror participation during deliberations provided fairly accurate information. n152 However, self-reports were less reliable for estimates of the influence jurors exerted on others and the scope of the jury's deliberations. n153 Higher-order questions about the influence of the jury's composition on its deliberations were even less informative, as they produced data that contradicted significant behavioral findings. n154

Put another way, the principal findings of Sommers were that racially diverse mock juries had longer, more complete, and more factually accurate deliberations than homogeneous juries, and that race-relevant voir dire questions led jurors to render more lenient judgments than did race-neutral questions. n155 Had the study relied exclusively on self-reports, however, its conclusions would have drastically underestimated the impact of race. The data would have indicated that neither the racial composition of the jury nor the racial content of voir dire influenced jurors' verdict preferences or deliberations. Furthermore, with the exception of the assessment of jurors' participation rates, conclusions regarding different aspects of the deliberations would have been inconclusive and contradictory.

It is quite possible that the influence of race is a topic particularly ill-suited to research using self-report measures. Race may affect decision-making and perceptions in ways that are less accessible to conscious awareness than the effects of other factors. Moreover, self-presentational concerns may be especially acute when race is involved. But self-report measures concerning race are not unique in that their reliability varies considerably depending on the precise nature of the research question. For example, researchers interested in examining [*226] sensitive questions related to prejudice of any sort - whether based on race, gender, class, or even physical attractiveness - should consider the possibility that self-report measures are particularly problematic. Moreover, the mock jurors in the Sommers study were also wrong about matters completely unrelated to race, such as the thoroughness of their deliberations and fellow jurors' feelings about their experience. n156 Thus, although using self-report measures to research the effects of race on jury decision-making may be cause for special concern, the present case study provides no reason to think that the perils of self-report measures are limited to questions involving sensitive topics.

The present case study compares behavioral outcomes and interview data obtained from mock jurors in a simulation, not from actual jurors. n157 Comparisons within the same studies conducted with actual jurors would be extremely informative, but with few exceptions (e.g., the AJP research in which jury deliberations were videotaped), the only behavioral outcomes that can be measured with real jurors are crude ones such as length of deliberation and final verdict. Experimental simulations therefore provide the best opportunity to compare interview and behavioral data about the same issues.

IV. IMPLICATIONS FOR THE USE OF JUROR SELF-REPORTS

Asking jurors about their attitudes, their reactions to the trial, their impressions of the deliberation process, or the factors that influenced them is one of the simplest and most straightforward ways to learn about jury decision-making. Jurors answer our questions, often quite confidently, but for a variety of reasons we must sometimes be cautious about taking their answers at face value. Before turning to when and under what circumstances relying on self-reports is most problematic, consider the major concerns we have discussed. n158 First, memory fades over time and it can be distorted by information learned after the original event. That confident recollections of eyewitnesses often turn out to be mistaken is well known to empirical legal researchers, and we should be alert to the danger of inaccurate memories in jurors as well. Second, jurors, like everyone, want to look competent and unbiased, and so tend to be unwilling to admit that they were influenced by impermissible factors; that their decisions were more emotional than rational; or that the legal instructions were so turgid and boring that they stopped listening. Finally, few human beings in any context can consistently and accurately identify the true causes of their behavior - although they are almost always willing to try - and researchers should be skeptical about answers to questions about higher-order cognitive processes.

A. When Are Jurors' Self Reports Most and Least Useful?

If a researcher's goal is to generate hypotheses about what ordinary people notice and care about when they serve as jurors, asking for free narrative descriptions about their thoughts during the trial and their discussions during deliberations can produce a wealth of information, revealing issues that mattered to the jurors but might not have occurred to the researchers. For instance, CJP interviews found that the defendant's remorse and future dangerousness were central to many juries' deliberations, as were the character and behavior of the victim, even when [*227] these considerations were not emphasized by the attorneys. n159 Open-ended interviews tend to be the best method for discovering issues that were important to individual jurors in a case, even better than observing videotaped deliberations - after all, many jurors vote but say very little, and so analyses of deliberations do not reveal much about their thoughts. n160

Interviews are also the best way to discover jurors' feelings about their experience after the case has been decided. Are they satisfied? Ambivalent? Anxious? Regretful? Do they now think of arguments they wish they had made during deliberations? Reflecting back, are there things that might have made their task easier? What did they like most about the experience? Least? These are retrospective questions about jurors' subjective impressions - not about what actually influenced them, but rather about how they feel about their experiences after the fact. These feelings might affect their performance as jurors if they serve again, as well as their general attitudes towards the jury system.

Questions that require jurors to recall particular, discrete events, such as their votes during deliberation, also produce fairly reliable data. In the Sommers mock jury experiment, participants accurately estimated how often they and other jurors spoke during deliberations, a task that seems even more difficult than recalling voting history. n161 Of course, this study required them to report on an hour-long deliberation immediately after it ended, whereas interviews with actual jurors often involve much longer deliberations that occurred much earlier. n162 Jurors' answers to questions about what they felt before the trial are less trustworthy than questions about specific events. People often believe that what they feel now is what they felt earlier, and they may reinterpret the past to match the present. n163 However, jurors may also fear that their responses will cast them in a negative light, and therefore admissions of inappropriate behavior, such as deciding on a verdict before hearing all the evidence, are particularly persuasive.

More problematic are questions that move beyond memory of the case and subjective experience, assessing instead jurors' evaluations of the trial or deliberation process. Interviewers often ask jurors for their opinions about the evidence, witnesses, attorneys, and other trial principals - questions more susceptible to reconstructive, post-trial biases than simple factual recollections about votes. Initial perceptions of a witness might change when subsequent events cast doubt on her testimony, and retrospective reports might not reflect the juror's initial impressions.

Moreover, unless a researcher's primary goal is to learn what enhances juror satisfaction, she will likely learn little from jurors' assessments of the effectiveness of various aspects of the trial. Jurors may report that a particular feature of the trial helped them do their job, but objective measures often suggest otherwise. For instance, jurors reported that getting written instructions [*228] helped them to better understand the law, but they showed no better understanding than did jurors who got them only orally. n164 Jurors' satisfaction with the clarity of the judge's instructions also reveals little about their effectiveness: Young, Cameron, and Tinsley found fundamental misunderstandings of the law even among jurors who reported finding the judge's instructions helpful and clear. n165

Similar problems arise with questions that do not focus on the particular trial, but rather assess more general attitudes about the legal system. The assumption that people have stable attitudes about legal issues is questionable, particularly when the attitudes involve subjects the person knew little about before serving on the jury. Some legal attitudes, such as attitudes towards capital punishment, are familiar and fairly stable over time. n166 Others, such as attitudes towards punitive damages or the duty to retreat before using force in self defense, may not have existed before trial, and jurors' answers to questions about them may not reflect long-standing pre-existing attitudes, but rather ideas that originated during their jury service. n167 If researchers want to learn about jurors' attitudes after serving on the jury, there is no problem; if they want to study the effect of stable, pre-existing attitudes on jury decision-making, there may be.

Our analyses suggest that self-report data are almost hopelessly problematic when jurors are asked to report on aspects of the trial or deliberation - such as fairness, comprehensiveness, or overall quality - that implicitly but necessarily require a basis for comparison. Answers to these questions can tell us how satisfied the juror is, but not how fair the trial or how thorough the deliberation actually was. Comparison of self-report and behavioral data also confirms the unreliability of jurors' self-reported perceptions of higher-order deliberation processes. The Sommers study provides an illustrative example, as mock jurors could not accurately assess aspects of deliberations such as the comprehensiveness of their jury's discussions or their own influence on fellow jurors. n168 Jurors were also poor reporters of the subjective experiences of other jurors, tending to assume that all members of the jury reacted the same way they did. n169

Likewise, people are often unable to provide accurate explanations for their own or their jury's decision. Asking jurors to describe how they made up their mind and arrived at a verdict is highly unlikely to provide valid information, whether the questions are pitched at the individual level - asking a particular juror to explain the factors behind her decision - or at the group level - asking respondents to recall how their jury reached unanimity. Mock jurors in the Sommers study, for example, denied that they were affected by race during deliberations or that their jury's racial composition affected their decision-making process. n170 Similarly, Professors Hans and Doob found that mock jurors were more likely to find a defendant guilty when they learned of his criminal history, despite their insistence that this evidence played no part in their decision. n171 [*229] Though they may have offered this information in good faith, behavioral data demonstrate that they were mistaken. n172 Actual jury deliberations are secret and unrecorded, and behavioral data from actual juries are few and far between. Therefore, higher-order questions regarding causal influences or requiring relative judgments - questions typically beyond the cognitive capabilities of respondents - are better addressed by simulation studies that include behavioral measures.

B. Alternatives to Self-Reports

There is no single best alternative for the researcher interested in examining the processes by which jurors and juries make decisions. Different methods are suitable for different questions. Archival analyses can identify some factors that predict jury outcomes, but they shed little light on the causal processes underlying this influence. n173 For that purpose, a better alternative is the use of jury simulation studies to determine the ways in which individual judgments are formed and translated into group-level verdicts. Such studies are time-and labor-intensive, but as our examination of race and jury decision-making studies indicates, total reliance on interview data can prevent researchers from discovering important processes or influences, and can even lead to inaccurate conclusions about the direction of relations among variables. n174 Of course, mock juror studies that rely exclusively on post-deliberation self-reports are susceptible to the same criticisms as interviews with actual jurors, and face additional limitations of realism and generalizability.

Alternative methods are available to researchers who seek to relate jurors' perceptions of specific case factors to their verdict preferences. Experiments can isolate the influence of a specific factor on juror decisions. For example, mock jury research has demonstrated the extent to which an expert witness influences juror decision-making, as well as the characteristics that render such testimony more or less persuasive. n175 Another strategy is to get independent ratings of the comprehensibility and persuasiveness of expert testimony, and then to determine whether these ratings predict verdicts in a sample of cases.

Some research objectives are easier than others to achieve using alternative measures. In research on juror comprehension, for example, jurors' assessments of their own competence can be supplemented or replaced by direct tests of their understanding of evidence or judicial instructions. But we propose that with a bit of creativity, there are few jury-related research [*230] questions accessible only through self-report data. Consider jurors' subjective experiences, such as emotional responses to evidence. It is true that the most straightforward approach would be to simply ask jurors how different aspects of the case made them feel. n176 Indeed, even mock jury experiments that measure juror emotion often use self-report measures. n177 But it also would be possible to code mock deliberations for jurors' emotional responses. Face-to-face interviews with former jurors could be video recorded and analyzed to determine whether self-reported emotional reactions to the trial coincide with behavioral evidence of emotion. No research methodology is without its limitations, and any empirical findings are most persuasive when supported by multiple methods.

C. Techniques for Improving the Validity of Self-Report Data

Knowing the hazards of interview research enables a researcher to take measures to counteract them, although these measures may be more or less effective depending upon the research question. Failures of memory grow worse with time. Therefore, the sooner the jurors can be interviewed, the better. It would be best to interview jurors before they leave the courthouse, before they have a chance to describe and justify their behavior to family and friends. Occasionally this is possible, though not often. Lawyers and jury consultants sometimes interview jurors right after a trial, and may be willing to include researchers in the process. When immediate interviews are impossible, researchers should be sure to record the amount of time between the interview and the events in question. In order to ensure standardization across jurors and cases, a researcher should seek to maintain a consistent, ideally short duration throughout a study. But no matter what, keeping track of duration is essential because it makes possible statistical analysis that includes passage of time as a factor. For instance, such analyses would permit a researcher to determine whether varying levels of comprehension observed during interviews truly indicate that some jurors did not understand judicial instructions at trial, or rather are attributable to differences among respondents in the gap between the trial and interview. It might also be the case that the relation between two variables - such as perceptions of a particular aspect of the case and final verdict - differs in strength among respondents depending on how much time has elapsed since trial.

If a researcher is interested in the effects of pre-existing legal attitudes, she would be well-advised to measure them before trial. There is good reason to believe that a juror's experiences during the trial and deliberations can influence more general attitudes towards various aspects of the law and the legal system. n178 Furthermore, it is likely that such processes are subtle, so that jurors are unaware that their attitudes about the system have been affected by their recent experience. When researchers plan to interview actual jurors, they should consider seeking [*231] permission to administer measures of pre-existing attitudes to the jury pool before petit juries are empanelled. Another strategy would be to add questions to the attorneys' pretrial questionnaire or to later obtain access to voir dire transcripts or jury selection questionnaires to discover pre-trial legal attitudes.

For some questions, there is a gold standard to which jurors' memory can be compared. For example, if the researcher is interested in the jurors' memories for lay or expert testimony, responses can be compared to the trial transcript. If the researcher is interested in the jurors' understanding of the law, simply asking them how well they understood the judge's instructions is unlikely to provide useful information. A better strategy is to ask specific questions about the law and compare the answers to the instructions. n179 Discrepancies between jurors' responses and the transcript or judicial instructions reveal not only general levels of accuracy, but also where mistakes occur.

On the other hand, there is no gold standard for a juror's memory of what happened during deliberations, except in those exceedingly rare cases when deliberations are recorded. However, the researcher may consider using agreement among jurors as a check on the accuracy of self-report information. n180 To the extent that jurors can provide accurate information about the content and scope of deliberations, there should be relative agreement among responses of jurors on the same juries, or among jurors in the majority and jurors in the minority. Professor Vidmar also offers the useful recommendation of interviewing jurors in groups to allow them to refresh each others' memories. n181 Of course, the more subjective the assessment, the less consistency among respondents that can be expected. But for self-reports regarding factual information, such a reliability check could serve as a useful way to confirm that a particular question generates accurate and meaningful data, as well as to identify individual jurors and juries whose reliability appears to be more problematic.

Biases related to jurors' desire to present themselves in a favorable light are harder to counteract. The usual guarantees of anonymity are important, but probably not completely effective. If the researcher is interested in whether inadmissible evidence or other extralegal considerations surfaced during deliberations, a good technique is to ask whether anyone raised these topics, and how other jurors responded, so that the respondent is not made to feel personally accountable. For instance, Young, Cameron, and Tinsley asked jurors if anyone raised concerns based on sympathy or prejudice, and how the other jurors reacted. n182 Of course, as alluded to above, any prejudicial or otherwise socially undesirable answers that jurors do give are particularly worth noting. As for finding out the real reasons for jurors' perceptions or decisions, no amount of determination will extract true answers from self-reports. People are generally unable to identify the factors that affected their decisions, although they are perfectly able to generate spurious answers to an interviewer's questions. n183

[*232]

V. CONCLUSION

Every research method has strengths and weaknesses, and almost every research question can be answered more cogently by using more than one method. Using multiple methods allows the researcher to use the strengths of one method to compensate for the weaknesses of another. n184 Indeed, several of the more impressive and persuasive interview studies we have described did not rely exclusively on interview data, but also include archival or behavioral analysis. For example, the CJP n185 and the research conducted by the NCSC n186 relied in large part on interviews with jurors and judges, but also included analysis of actual jury outcomes and compositions. So, too, did the Arizona Jury Project study, which was actually a field experiment comparing data - self-report and observation of actual deliberations - from juries randomly assigned to one of two conditions: permitted or not permitted to engage in pre-deliberation discussions. n187 In this manner, the investigators were able to combine the enhanced generalizability of juror interviews and archival analysis with the experimental control of a mock jury study. n188

Our goal in this Article has not been to discourage juror interviews or the use of self-reports, or to suggest that findings from studies using these methods should be disregarded. To the contrary, juror interviews remain an essential tool in the empirical investigation of juries. We hope that our literature review and detailed case study of race and legal judgment prompt researchers to think carefully about the differences among types of interview questions, the importance of multiple methods, and the ways in which interviews can be utilized to their fullest potential. We anticipate that others will add their own suggestions to our list of recommendations and cautionary tales, generating a more complete assessment of the methods used by the burgeoning field of empirical legal research. And we hope that future researchers, while still making use of interview methodologies, will also heed the call to continue to develop new and creative ways of investigating jury decision-making that are not wholly reliant upon self-report.

Legal Topics:

For related research and practice materials, see the following legal topics:

Criminal Law & ProcedureJury InstructionsParticular InstructionsPresumption of InnocenceCriminal Law & ProcedureJury InstructionsParticular InstructionsUse of Particular EvidenceCriminal Law & ProcedureVerdictsUnanimity

FOOTNOTES:

n1. See generally Neil Vidmar & Valerie P. Hans, American Juries: The Verdict (2007) (examining how jurors make decisions and whether jurors are competent to perform the tasks expected of them); Nancy S. Marder, The Jury Process 50-104 (2005).

n2. See, e.g., Shari Seidman Diamond, Neil Vidmar, Mary Rose, Leslie Ellis, & Beth Murphy, Juror Discussions During Civil Trials: Studying an Arizona Innovation, 45 Ariz. L. Rev. 1, 48 (2003) (discussing the Arizona Jury Project in which researchers examined the implications of procedural changes regarding jury deliberations); William J. Bowers, The Capital Jury Project: Rationale, Design, and Preview of Early Findings, 70 Ind. L. J. 1043 (1995) (discussing the Capital Jury Project in which researchers interviewed people who had served on capital juries).

n3. See Stephen P. Garvey, The Emotional Economy of Capital Sentencing, 75 N.Y.U. L. Rev. 26, 29 (2000) (describing the limitations of juror interviews, which include lack of candor, the desire to please the interviewer, the social desirability of certain responses, deterioration of memory, and hindsight bias).

n4. See generally Phoebe C. Ellsworth & Alan Reifman, Juror Comprehension and Public Policy: Perceived Problems and Proposed Solutions, 6 Psychol. Pub. Pol'y & L. 788 (2000) (summarizing studies demonstrating that jurors often have trouble understanding the law that they are asked to apply to decide a case); Joel D. Lieberman & Bruce D. Sales, What Social Science Teaches Us About the Jury Instruction Process, 3 Psychol. Pub. Pol'y & L. 589 (1997) (showing that jurors' comprehension of legal instructions is poor due to unnecessary procedural obstacles in the trial process).

n5. In this Article, we use "jury interviews" and "juror interviews" to refer not only to face-to-face interviews between researchers and jurors, but also the administration of post-trial questionnaires. Both methods essentially rely on jurors' self-reported data.

n6. Marjorie Fargo, Postverdict Juror Interviews, in ALI-ABA, A Handbook of Jury Research 25-2 (Walter F. Abbott & John Batt, eds., 1999). See also R. Robert Samples, All Things Jury: Looking Forward with Post-Trial Juror Interviews (May 21, 2008), http://www.wvrecord.com/news/212651-all-things-jury-looking-forward-with-post-trial-juror-interviews ("One of the most effective research methods in a trial consultant's toolbox is the post-trial interview with members of the actual jury panel."); Jay M. Finkelman, Juror Perception of Employment Litigation, 8 The Psychologist-Manager J. 45, 46 (2005) (employing post-trial interviews with jurors to evaluate how jurors process information and determine verdicts in employment litigation).

n7. See, e.g., Neil Vidmar, Civil Juries in Ecological Context: Methodological Implications for Research, in Civil Juries and Civil Justice: Psychological and Legal Perspectives 35, 35-65 (Brian H. Bornstein et al., eds., 2008) [hereinafter Vidmar, Civil Juries]; Shari Seidman Diamond, Beyond Fantasy and Nightmare: A Portrait of the Jury, 54 Buff. L. Rev. 717, 726-27 (2006) [hereinafter Diamond, Beyond Fantasy and Nightmare]; Samuel R. Sommers & Phoebe C. Ellsworth, How Much Do We Really Know About Race and Juries? A Review of Social Science Theory and Research, 78 Chi.-Kent L. Rev. 997, 1000-02 (2003) [hereinafter Sommers & Ellsworth, How Much Do We Really Know].

n8. See generally Vidmar, Civil Juries, supra note 7.

n9. See, e.g., Elliot Aronson, Phoebe C. Ellsworth, J. Merrill Carlsmith & Marti Hope Gonzales, Methods of Research in Social Psychology 60-63 (2d ed. 1990) (reviewing research demonstrating that people are often unable or unwilling to report their cognitive and affective processes accurately); Delroy L. Paulhus & Oliver P. John, Egoistic and Moralistic Biases in Self-Perception: The Interplay of Self-Deceptive Styles With Basic Traits and Motives, 66 J. Personality 1025, 1040-45 (1998); Norbert Schwarz, Robert M. Groves & Howard Schuman, Survey Methods, in 1 Handbook of Social Psychology 143, 143-44 (Daniel T. Gilbert et al. eds., 4th ed. 1998).

n10. See infra Part I.A-E.

n11. Hermann Ebbinghaus, Memory: A Contribution to Experimental Psychology 62-69 (Henry A. Ruger & Clara E. Bussenius trans., 1998) (1913).

n12. Elizabeth F. Loftus & Hunter G. Hoffman, Misinformation and Memory: The Creation of New Memories, 118 J. Experimental Psychol. 100 (1989); Harry T. Reis & Shelly L. Gable, Event Sampling and Other Methods for Studying Everyday Experience, in Handbook of Research Methods in Social and Personality Psychology 190-222 (Harry T. Reis & Charles M. Judd eds., 2000).

n13. See Daniel L. Schacter, The Seven Sins of Memory: Insights From Psychology and Cognitive Neuroscience, 54 Am. Psychol. 182, 193 (1999) [hereinafter Schacter, The Seven Sins of Memory] (reviewing research on consistency bias, which leads people to overestimate consistency between their past and current attitudes).

n14. See generally Keith J. Holyoak & Dan Simon, Bidirectional Reasoning in Decision Making by Constraint Satisfaction, 128 J. Experimental Psychol. Gen. 3 (1999); Derek J. Koehler, Explanation, Imagination, and Confidence in Judgment, 110 Psychol. Bull. 499 (1991).

n15. Garvey, supra note 3, at 29 n.16.

n16. Paula L. Hannaford, Valerie P. Hans, Nicole L. Mott & G. Thomas Munsterman, The Timing of Opinion Formation by Jurors in Civil Cases: An Empirical Examination, 67 Tenn. L. Rev. 627, 634 (2000).

n17. Richard E. Nisbett & Timothy DeCamp Wilson, Telling More Than We Can Know: Verbal Reports on Mental Processes, 84 Psychol. Rev. 231, 255 (1977).

n18. Id. at 255. See generally Michael I. Norton, Joseph A. Vandello & John M. Darley, Casuistry and Social Category Bias, 87 J. Personality & Soc. Psychol, 817 (2004); Eldar Shafir, Itamar Simonson & Amos Tversky, Reason-based Choice, 49 Cognition 11 (1993).

n19. Nisbett & Wilson, supra note 17, at 243.

n20. Id.

n21. See Smith v. Phillips, 455 U.S. 209, 230 (1982) (Marshall, J., dissenting) ("Given the human propensity for self-justification, it is very difficult "to learn from a juror's own testimony after the verdict whether he was in fact "impartial." Certainly, a juror is unlikely to admit that he had consciously plotted against the defendant during the course of the trial." (quoting Phillips v. Smith, 632 F.2d 1019, 1022 (2d Cir. 1980) (internal citation omitted)), rev'd 455 U.S. 209 (1982)). See also Edward E. Jones & Harold Sigall, The Bogus Pipeline: A New Paradigm for Measuring Affect and Attitude, 76 Psychol. Bull. 349, 358-62 (1971) (reviewing research showing that people tend to give less socially acceptable responses about sensitive issues when they believe that experimenters can detect lies). But see E. Ashby Plant, Patricia G. Devine & Paige C. Brazy, The Bogus Pipeline and Motivations to Respond Without Prejudice: Revising the Fading and Faking of Racial Prejudice, 6 Group Processes & Intergroup Rel. 187, 188-91 (2003); Neal J. Roese & David W. Jamieson, Twenty Years of Bogus Pipeline Research: A Critical Review and Meta-Analysis, 114 Psychol. Bull. 363, 369 (1993) (presenting results from a meta-analysis showing that use of the bogus pipeline technique makes people provide more honest self-reports).

n22. Barry R. Schlenker & Mark L. Leary, Social Anxiety and Self-Presentation: A Conceptualization and Model, 92 Psych. Bull. 641, 647 (1982) (reviewing research on the factors that heighten self-presentational concerns).

n23. See, e.g., Bowers, supra note 2, n.205 (explaining random selection of four jurors per capital case in the CJP studies); Mary R. Rose, A Dutiful Voice: Justice in the Distribution of Jury Service, 39 Law & Soc'y Rev. 601, 614 (reporting a 67% response rate by former jurors to requests for interviews).

n24. Schwarz et al., supra note 9, at 143-79.

n25. See generally Norbert Schwarz, Self-Reports: How the Questions Shape the Answers, 54 Am. Psychologist 93 (1999).

n26. Elizabeth F. Loftus & John C. Palmer, Reconstruction of Automobile Destruction: An Example of the Interaction Between Language and Memory, 13 J. Verbal Learning &Verbal Behav. 585, 585 (1974).

n27. Id. at 586.

n28. Id. at 585.

n29. Researchers use juror self-reports in both single study and larger scale research projects. The most extensive of these larger projects is the Capital Jury Project (CJP), an ongoing fourteen-state field study of jury decision-making in capital trials. See generally William J. Bowers, The Capital Jury Project: Rationale, Design, and Preview of Early Findings, 70 Ind. L. J. 1043 (1995). CJP data have been collected in 20 to 30 capital trials per state, divided equally between cases that ended in death and alternative sentences. Four randomly selected jurors from each trial are interviewed at length about the trial and deliberations, touching on topics including the reasons for their decisions, the course of their decision-making, their assessments of the evidence, and their impressions of the witnesses. William J. Bowers, Benjamin D. Steiner & Marla Sandys, Death Sentencing in Black and White: An Empirical Analysis of the Role of Jurors' Race and Jury Racial Composition, 3 U. Pa. J. Const. L. 171, 189, 197 (2001) [hereinafter Bowers, Steiner & Sandys, Death Sentencing]. In addition, the National Center for State Courts (NCSC) conducted a large-scale study of juries in which judges, attorneys, and jurors in 372 felony trials were surveyed to examine interactions among case characteristics, group dynamics, and juror demographics and attitudes in order to uncover factors associated with hung juries. Paula L. Hannaford-Agor & Valerie P. Hans, Nullification at Work? A Glimpse from the National Center for State Courts Study of Hung Juries, 78 Chi.-Kent L. Rev. 1249, 1263 (2003) [hereinafter Hannaford-Agor & Hans, Nullification at Work]. See also Paula L. Hannaford, Valerie P. Hans & G. Thomas Munsterman, Permitting Jury Discussions During Trial: Impact of the Arizona Reform, 24 Law & Hum. Behav. 359 (2000) [hereinafter Hannaford, Hans & Munsterman, Permitting Jury Discussions]; Paula L. Hannaford-Agor, Valerie P. Hans, Nicole L. Mott & G. Thomas Munsterman, Nat'l Inst. of Justice, Are Hung Juries a Problem? NCJ 201096 (2002).

n30. See generally Leigh B. Bienen, Helping Jurors Out: Post-Verdict Debriefing for Jurors in Emotionally Disturbing Trials, 68 Ind. L. J. 1333 (1993); Theodore Eisenberg, Stephen P. Garvey & Martin T. Wells, Jury Responsibility in Capital Sentencing: An Empirical Study, 44 Buff. L. Rev. 339 (1996) [Eisenberg, Garvey & Wells, Jury Responsibility in Capital Sentencing]; Thomas L. Hafemeister, Juror Stress, 8 Violence & Victims 177 (1993); Joseph L. Hoffman, Where's the Buck? - Juror Misperception of Sentencing Responsibility in Death Penalty Cases, 70 Ind. L. J. 1137 (1995); Honorable James E. Kelley, Addressing Juror Stress: A Trial Judge's Perspective, 43 Drake L. Rev. 97 (1994); Scott E. Sundby, A Life and Death Decision: A Jury Weighs the Death Penalty 56-57, 97-100, 130, 159, 169-173 (2005) (examining the emotional impact of a capital case on individual jurors); Michael E. Antonio, I Didn't Know It'd Be So Hard: Jurors' Emotional Reactions to Serving on a Capital Trial, 89 Judicature 282 (2006).

n31. Carroll Seron et al., Report of the Working Committees to the Second Circuit Task Force on Gender, Racial and Ethnic Fairness in the Courts, 1997 Ann. Surv. Am. L., 117 (1997); William J. Bowers, Marla Sandys & Thomas W. Brewer, Crossing Racial Boundaries: A Closer Look at the Roots of Racial Bias in Capital Sentencing When the Defendant is Black and the Victim is White, 53 DePaul L. Rev. 1497 (2004); Roger Matthews, Lynn Hancock & Daniel Briggs, Home Office, Jurors' Perceptions, Understanding, Confidence and Satisfaction in the Jury System: A Study in Six Courts (2004); Mary R. Rose, A Dutiful Voice: Justice in the Distribution of Jury Service, 39 Law & Soc'y Rev. 601 (2005); Mary R. Rose, A Voir Dire of Voir Dire: Listening to Jurors' Views Regarding the Peremptory Challenge, 78 Chi.-Kent L. Rev. 1061 (2003); Franklin Delano Strier, Through the Jurors' Eyes, 74 A.B.A. J. 79 (1988).

n32. See generally Nancy S. Marder, Juries, Justice & Multiculturalism, 75 S. Cal. L. Rev. 659 (2002); Bowers, Sandys & Brewer, supra note 31; Shari Seidman Diamond, Mary R. Rose & Beth Murphy, Revisiting the Unanimity Requirement: The Behavior of the Non-Unanimous Civil Jury, 100 Nw. U. L. Rev. 201 (2006); Valerie P. Hans, Paula L. Hannaford & G. Thomas Munsterman, The Arizona Jury Reform Permitting Civil Jury Trial Discussions: The Views of Trial Participants, Judges, and Jurors, 32 U. Mich. J.L. Reform 349 (1999).

n33. See Joe S. Cecil, E. Allan Lind & Gordon Bermant, Jury service in lengthy civil trials 27, 33 (Federal Judicial Center 1987) (surveying jurors about how difficult it was to understand information presented at trial and instructions from the judge); Larry Heuer & Steven Penrod, Increasing Jurors' Participation in Trials: A Field Experiment with Jury Notetaking and Question Asking, 12 Law & Hum. Behav. 231, 243 (1988) [hereinafter Heuer & Penrod, Increasing Jurors' Participation in Trials] (using juror's ratings of the difficulty of reaching unanimity and the difficulty of personal voting as factors to analyze jury deliberations). See generally Warren Young, Neil Cameron & Yvette Tinsley, Juries in Criminal Trials, Pt. 2, in Preliminary Papers 1999 (New Zealand Law Comm'n, NZLC PP37, 1999); Warren Young, Yvette Tinsley & Neil Cameron, The Effectiveness and Efficiency of Jury Decision-Making, 24 Crim. L.J. 89 (2000); Michael Zander & Paul Henderson, The Royal Commission on Criminal Justice: Crown Court Study (HMSO 1993).

n34. Theodore Eisenberg & Martin T. Wells, Deadly Confusion: Juror Instructions in Capital Cases, 79 Cornell L. Rev. 1, 14 (1993).

n35. Marder, supra note 32, at 681, 682.

n36. Id. at 692-93, 697-98.

n37. Bowers, Steiner & Sandys, Death Sentencing, supra note 29, at 244.

n38. Bowers, Sandys & Brewer, supra note 31, at 1525.

n39. See Theodore Eisenberg, Stephen P. Garvey & Martin T. Wells, Forecasting Life and Death: Juror Race, Religion, and Attitude Toward the Death Penalty, 30 J. Legal Stud. 277, 279 (2001) [hereinafter Eisenberg, Garvey & Wells, Forecasting Life and Death] (examining how certain juror characteristics may influence a juror's first vote in capital cases); Stephen P. Garvey, et al., Juror First Votes in Criminal Trials, 1 J. Empirical Legal Stud. 371, 373-74 (2004) (controlling for the influence of the majority on the final vote by asking for jurors' first votes); Carol J. Mills & Wayne E. Bohannon, Juror Characteristics: To What Extent Are They Related to Jury Verdicts? 64 Judicature 23, 24 (1980); John P. Reed, Jury Deliberations, Voting, and Verdict Trends, 45 Southwestern Soc. Sci. Q. 361, 366-67 (1965); Marla Sandys & R. C. Dillehay, First-Ballot Votes, Predeliberation Dispositions, and Final Verdicts in Jury Trials, 19 Law & Hum. Behav. 175, 178 (1995); Christy A. Visher, Juror Decision Making: The Importance of Evidence, 11 Law & Hum. Behav. 1, 13-14 (1987).

n40. See Gary Moran & John C. Comfort, Scientific Juror Selection: Sex as a Moderator of Demographic and Personality Predictors of Impaneled Felony Juror Behavior, 43 J. Personality & Soc. Psychol. 1052, 1058 (1982); Gary Moran & John C. Comfort, Neither "Tentative" Nor "Fragmentary": Verdict Preference of Impaneled Felony Jurors as a Function of Attitude Toward Capital Punishment, 71 J. Applied Psychol. 146, 147-48 (1986) (examining multiple studies on the correlation between jurors' attitudes on capital punishment and the verdict).

n41. Eisenberg, Garvey & Wells, Forecasting Life and Death, supra note 39, at 308. See also William J. Bowers, Marla Sandys & Benjamin D. Steiner, Foreclosed Impartiality in Capital Sentencing: Jurors' Predispositions, Guilt Trial Experience, and Premature Decision Making, 83 Cornell L. Rev. 1474, 1512-14 (1998) [hereinafter Bowers, Sandys & Steiner, Foreclosed Impartiality].

n42. See Cathy Johnson & Craig Haney, Felony Voir Dire: An Exploratory Study of Its Content and Effect, 18 Law & Hum. Behav. 487, 498 (1994) (noting how jurors who had been allowed to serve still had attitudes directly against the presumption of innocence); Richard Seltzer, Mark A. Venuti & Grace M. Lopes, Juror Honesty During the Voir Dire, 19 J. Crim. Just. 451, 457-58 (1991) (revealing that jurors did not respond during voir dire to attitudinal questions such as whether the defendant bears the burden of proof).

n43. See Valerie P. Hans & Stephanie Albertson, Empirical Research and Civil Jury Reform, 78 Notre Dame L. Rev. 1497, 1507-08 (2003) (debunking claims that juries are pro-plaintiff and anti-business).

n44. Visher, supra note 39, at 9.

n45. E.g., Virginia R. Boehm, Mr. Prejudice, Miss Sympathy, and the Authoritarian Personality: An Application of Psychological Measuring Techniques to the Problem of Jury Bias, 1968 Wis. L. Rev. 734, 742; David A. Kravitz, Brian L. Cutler & Petra Brock, Reliability and Validity of the Original and Revised Legal Attitudes Questionnaire, 17 Law & Hum. Behav. 661, 663 (1993) (dealing with this issue by administering their Legal Attitudes Questionnaire before presenting students with a murder case).

n46. See, e.g., Samples, supra note 6 (stating that "Certainly, the most important questions focus on the decision criteria utilized by the panel and the group interaction that influenced the verdict," and, "Without question, the most important area of exploration with jurors is a thorough investigation of their personal, and the group's, interaction and discussion as to how they arrived at a particular verdict. Probing questions are utilized to understand the thinking and reasoning behind the various elements that lead to the verdict. These findings are critical to similar trials that litigators may face in the future.").

n47. See, e.g., Benjamin Fleury-Steiner, Narratives of the Death Sentence: Toward a Theory of Legal Narrativity, 36 Law & Soc'y Rev. 549, 553 (2002) (asking open-ended questions to see what legal or extra-legal factors affected jurors' decisions).

n48. Bowers, Sandys & Brewer, supra note 31, at 1532; see also Bryan C. Edelman, Racial Prejudice, Juror Empathy, and Sentencing in Death Penalty Cases (2006); Garvey, supra note 3.

n49. See Valerie P. Hans, Business on Trial: The Civil Jury and Corporate Responsibility (2000) (interviewing jurors about their responses to parties and evidence). See generally Michael E. Antonio & Nicole E. Arone, Damned if They Do, Damned if They Don't: Jurors' Reaction to Defendant Testimony or Silence During a Capital Trial, 89 Judicature 60 (2005) (reviewing interviews with CJP jurors about the inferences they drew from a defendant's failure to testify as well as their perceptions of defendants who did testify); Thomas W. Brewer, The Attorney-Client Relationship in Capital Cases and its Impact on Juror Receptivity to Mitigation Evidence, 22 Just. Q. 340 (2005) (using interviews with 725 capital jurors to assess how jurors' perception of the attorney-client relationship affects their willingness to consider mitigation evidence).

n50. Garvey et al., supra note 39, at 396.

n51. See generally Diane L. Bridgeman & David Marlowe, Jury Decision Making: An Empirical Study Based on Actual Felony Trials, 64 J. Applied Psychol. 91, 95 (1979) (finding that jurors placed little importance on the testimony of experts relative to the testimony of police, the defendant and other witnesses); Shari Seidman Diamond & Jonathan D. Casper, Blindfolding the Jury to Verdict Consequences: Damages, Experts, and the Civil Jury, 26 Law & Soc'y Rev. 513 (1992); Michael J. Saks & Roselle L. Wissler, Legal and Psychological Bases of Expert Testimony Surveys of the Law and of Jurors, 2 Behav. Sci. & L. 435 (1984) (interviewing individuals for their assessments of the relative credibility of different kinds of experts, such as doctors, lab experts, and accountants).

n52. Scott E. Sundby, The Jury as Critic: An Empirical Look at How Capital Juries Perceive Expert and Lay Testimony, 83 Va. L. Rev. 1109, 1176-77 (1997). See generally John H. Montgomery, J. Richard Ciccone, Stephen P. Garvey & Theodore Eisenberg, Expert Testimony in Capital Sentencing: Juror Responses, 33 J. Am. Acad. Psychiatry & L. 509 (2005) (concluding that psychiatric expert testimony was important in affecting jurors' view of mental illness as a mitigating factor but not so impactful on jurors' view of defendant's future dangerousness in capital cases).

n53. Bowers, Sandys & Brewer, supra note 31, at 1500 (examining how race of jury and defendant influences the verdict). See generally Shari Seidman Diamond, Jonathan D. Casper, Carnie L. Heiert & Anna-Maria Marshall, Criminology: Juror Reactions to Attorneys at Trial, 87 J. Crim. L. & Criminology 17 (1996) (using simulations of jury service to determine reactions to attorneys' performance during deliberations); Scott E. Sundby, The Capital Jury and Empathy: The Problem of Worthy and Unworthy Victims, Cornell L. Rev. 343, 344 (2003) (studying how victims' characteristics influence the jury's decision in capital cases); Theodore Eisenberg, Stephen P. Garvey & Martin T. Wells, Victim Characteristics and Victim Impact Evidence in South Carolina Capital Cases, 88 Cornell L. Rev. 306 (2003) (examining the relationship between victim impact evidence, perceived victim admirability and the effect of victim admirability on juries imposing death sentences).

n54. Nicole L. Mott, Valerie P. Hans & Lindsay Simpson, What's Half A Lung Worth? Civil Jurors' Accounts of Their Award Decision Making, 24 Law & Hum. Behav. 401, 410 (2000).

n55. See generally Scott E. Sundby, The Capital Jury and Absolution: The Intersection of Trial Strategy, Remorse, and the Death Penalty, 83 Cornell L. Rev. 1557 (1998) (examining how juries use defendant's remorse in deciding between the death penalty and a life sentence); Theodore Eisenberg, Stephen P. Garvey & Martin T. Wells, But Was He Sorry? The Role of Remorse in Capital Sentencing, 83 Cornell L. Rev. 1599 (1998) (studying what factors the jurors use to assess defendant's remorsefulness).

n56. See Austin Sarat, Violence, Representation, and Responsibility in Capital Trials: The View from the Jury, 70 Ind. L.J. 1103, 1131-32 (1995). Sarat analyzed what jurors told interviewers about how they perceived their role in the process and what they thought a death sentence really meant. Many jurors reported that they did not believe that the death penalty deterred, but they nevertheless supported it as a way to incapacitate offenders, especially when life without the possibility of parole was not a sentencing option. See generally John H. Blume, Stephen P. Garvey & Sheri Lynn Johnson, Future Dangerousness in Capital Cases: Always "At Issue," 86 Cornell L. Rev. 397, 397-410 (2001); Sally Costanzo & Mark Costanzo, Life or Death Decisions: An Analysis of Capital Jury Decision Making Under the Special Issues Sentencing Framework, 18 Law & Hum. Behav. 151 (1994) (determining that future dangerousness combined with the possibility of parole had a strong influence in deciding between life and death sentences); Eisenberg & Wells, supra note 34 (asserting that future dangerousness is on the minds of most jurors without the help of the prosecutor).

n57. Sundby, supra note 52.

n58. Ursula Bentele & William J. Bowers, How Jurors Decide on Death: Guilt is Overwhelming; Aggravation Requires Death; and Mitigation is No Excuse, 66 Brook. L. Rev. 1013, 1042 (2001).

n59. See e.g., Bridgeman & Marlowe, supra note 51, at 91 (interviewing sixty-five jurors who served in felony trials to examine the relationship between jury characteristics and decision-making); Eisenberg, Garvey & Wells, Jury Responsibility in Capital Sentencing, supra note 30, at 340 (interviewing 153 jurors in South Carolina capital cases to examine whether capital sentencing jurors assume responsibility for the sentences imposed); Valerie P. Hans & William S. Lofquist, Jurors' Judgments of Business Liability in Tort Cases: Implications For The Litigation Explosion Debate, 26 Law & Soc'y Rev 85, 92 (1992) (analyzing interviews regarding jurors' reactions to parties and evidence in business liability cases); Matthews et al., supra note 31; John E. B. Myers et al., Jurors' Perceptions of Hearsay in Child Sexual Abuse Cases, 5 Psychol. Pub. Pol'y & L. 388, 393 (1999) (analyzing interviews regarding jurors' reactions to evidence in child abuse cases); Barbara F. Reskin & Christy A. Visher, The Impacts of Evidence and Extralegal Factors in Jurors' Decisions, 20 Law & Soc'y Rev. 423, 424-25 (1986) (analyzing interviews regarding jurors' background, evidence, and response to verdict to discover the impact of evidence and extralegal factors). See generally Young, Tinsley & Cameron, supra note 33 (interviewing jurors and judges to research jury decision-making in criminal cases in New Zealand); Ann Burnett Pettus, The Verdict Is In: A Study of Jury Decision Making Factors, Moment of Personal Decision, and Jury Deliberations - From the Jurors' Point of View, 38 Communication Q. 83 (1990).

n60. Sanja Kutjnak Ivkovich & Valerie P. Hans, Jurors and Experts, 16 Advoc. Mag. Del. Trial L. 17, 18 (1994). See generally Anthony Champagne, Daniel W. Shuman & Elizabeth Whitaker, An Empirical Examination of the Use of Expert Witnesses in American Courts, 31 Jurimetrics J. 375, 379 (1991); Daniel W. Shuman, Anthony Champagne & Elizabeth Whitaker, Assessing the Believability of Expert Witnesses: Science in the Jury Box, 37 Jurimetrics J. 23 (1996) (arguing that jurors make expert-specific decisions based on a rational set of criteria).

n61. Young, Cameron & Tinsley, supra note 33, at 43-44.

n62. William S. Geimer & Jonathan Amsterdam, Why Jurors Vote Life or Death: Operative Factors in Ten Florida Death Penalty Cases, 15 Am. J. Crim. L. 1, 4 n.11 (1988); see also Bowers, Sandys & Steiner, Foreclosed Impartiality; supra note 41, at 1534-36.

n63. Geimer & Amsterdam, supra note 62, at 8. See generally Stephen P. Garvey, Aggravation and Mitigation in Capital Cases: What Do Jurors Think?, 98 Colum. L. Rev. 1538 (1998).

n64. See generally Nisbett & Wilson, supra note 17 (explaining that the accuracy of cognitive processes are determined by the saliency and plausibility of the stimuli).

n65. Id.

n66. See, e.g., Bowers, Sandys & Brewer, supra note 31, at 1500 (examining the impact of racial composition of a jury on the decision through juror evaluations); Marder, supra note 32, at 662-63 (considering the implications of more culturally diverse juries and how the concept connects with a jury's reasonableness standard).

n67. Young, Tinsley & Cameron, supra note 33, at 28.

n68. Hannaford et al., supra note 16, at 636-37.

n69. Bowers, Sandys & Steiner, Foreclosed Impartiality, supra note 41, at 1477; see also Marla Sandys, Cross-Overs: Capital Jurors Who Change Their Minds About Punishment: A Litmus Test for Sentencing Guidelines, 70 Ind. L. J. 1183, 1191 (1995).

n70. See, e.g., Arthur D. Austin, Complex Litigation Confronts the Jury System: A Case Study 55, 63 (1984) (interviewing two sets of jurors in a complex antitrust suit and finding that jurors had a great deal of difficulty with the concepts involved and the poor language of the instructions); Harold M. Hoffman & Joseph Brodley, Jurors on Trial, 17 Mo. L. Rev. 235, 250 (1952) (revealing from juror interviews that jurors misunderstand a variety of aspects of trial). See also Paul Rosenthal, Nature of Jury Response to the Expert Witness, 28 J. Forensic Sci. 528-31 (1983) (discussing a case where jurors based decisions on expert witness' appearance when they did not comprehend voiceprint identification-technological evidence); Molly Selvin & Larry Picus, Inst. Civ. Just., The Debate Over Jury Performance (1987) (researching jury behavior in tort liability claims).

n71. See generally Joseph Sanders, The Jury Deliberation in a Complex Case: Havner v. Merrell Dow Pharmaceuticals, 16 Just. Sys. J. 45 (1993) (evaluating Havner, a case where the jury had difficulties understanding the scientific evidence presented by expert witnesses); see also Special Comm. A.B.A. Sec. Litig., Jury Comprehension in Complex Cases, 1989.

n72. Hannaford, et al., supra note 16, at 636-38; Valerie P. Hans & Krista Sweigart, Jurors' Views of Civil Lawyers: Implications for Courtroom Communication, 68 Ind. L. J. 1297, 1304 (1993) (analyzing jurors and mock jurors to determine if they remained neutral after opening statements and prior to closing arguments and arguing that in initially deciding criminal cases, jurors focus on evidence, with limited influence by personal attitude or demographics, among other factors).

n73. Reskin & Visher, supra note 59, at 423-24 (examining how evidence and extralegal variables influenced jurors in sexual assault trials).

n74. Mitchell J. Frank & Dawn Broschard, The Silent Defendant and the Presumption of Innocence: In the Hands of Real Juror, Is Either of Them Safe?, 10 Lewis & Clark L. Rev. 237, 248-52 (2006).

n75. Alan Reifman, Spencer M. Gusick & Phoebe C. Ellsworth, Real Jurors' Understanding of the Law in Real Cases, 16 Law & Hum. Behav. 539, 539 (1992); see also Shari Seidman Diamond, Mary R. Rose, Beth Murphy & Sven Smith, Juror Questions During Trial: A Window Into Juror Thinking, 59 Vand. L. Rev. 1927 (2006) (arguing that jurors used questioning to supplement and deepen understanding of evidence).

n76. See Bowers, supra note 29, at 1044 (studying how capital jurors make their life or death sentencing decisions and how the juridical system is still criticized for some arbitrariness). See generally William J. Bowers & Benjamin D. Steiner, Death By Default: An Empirical Demonstration of False and Forced Choices in Capital Sentencing, 77 Tex. L. Rev. 605 (1999) (examining whether capital jurors should know prisoner parole timelines and whether jurors can make a reasonable decision regarding punishment without such information); Shari Seidman Diamond & J. N. Levi, Improving Decisions on Death By Revising and Testing Jury Instructions, 79 Judicature 224 (1996) (arguing that the instructions that jurors receive in some capital cases do not provide adequate guidance, and deliberations cannot be relied on to cure juror miscomprehensions); Eisenberg & Wells, supra note 34, at 2 (arguing that jurors are often improperly instructed in murder trials, which leads to initial inclination to sentence to death); Geoffrey P. Kramer & Dorean M. Koenig, Do Jurors Understand Criminal Jury Instructions? Analyzing the Results of the Michigan Juror Comprehension Project, 23 U. Mich. J. L. Reform 401, 406 (1990) (surveying Michigan judges to identify possible weaknesses in criminal jury instructions leading to juror confusion). See generally Craig Haney, Lorelei Sontag & Sally Constanzo, Deciding to Take a Life: Capital Juries, Sentencing Instructions, and the Jurisprudence of Death, 50 J. Soc. Issues 149 (1994) (analyzing interviews with jurors in two states with different sentencing frameworks to determine how capital sentencing instructions, a juror's interpretation of evidence, and conceptions of the juror's task alter decision-making process); James Luginbuhl & Julie Howe, Discretion in Capital Sentencing Instructions: Guided or Misguided?, 70 Ind. L. J. 1161 (1995) (arguing that improving judicial instruction to jurors in capital punishment cases can decrease influence by erroneous perception of law); Reifman et al., supra note 75; Bradley Saxton, How Well Do Jurors Understand Jury Instructions? A Field Test Using Real Juries and Real Trials in Wyoming, 33 Land & Water L. Rev. 59 (1998) (arguing for improvement of jury instruction practices in Wyoming); Benjamin D. Steiner, William J. Bowers & Austin Sarat, Folk Knowledge as Legal Action: Death Penalty Judgments and the Tenet of Early Release in a Culture of Mistrust and Punitiveness, 33 Law & Soc'y Rev. 461 (1999) (arguing that individual perception influences jury deliberation in capital punishment cases despite court admonitions); Young, Tinsley & Cameron, supra note 33, at 97-98; Zander & Henderson, supra note 33.

n77. Hans & Albertson, supra note 43, at 1498; see also Hans & Lofquist, supra note 59, at 86 (discussing coverage of civil juries as overcompensating plaintiff's claiming negligence even when the claim is questionable); Young, Tinsley & Cameron, supra note 33.

n78. Hoffman, supra note 30 (arguing that many capital case jurors avoid personal moral responsibility for jury death penalty decisions); see also Bentele et al., supra note 58, at 1013 (discussing diminished responsibility among jurors in death sentence cases and its unconstitutionality); William J. Bowers, Wanda D. Foglia, Jean E. Giles & Michael E. Antonio, The Decision Maker Matters: An Empirical Examination of the Way The Role of the Judge and the Jury Influence Death Penalty Decision-Making, 63 Wash. & Lee L. Rev. 931, 1010 (2006); Eisenberg, Garvey & Wells, Jury Responsibility in Capital Sentencing, supra note 30, at 341 (arguing that some jurors do not believe they are causally responsibility for death sentences and do not believe that most death sentences will be carried out). See generally Wanda D. Foglia, They Know Not What They Do: Unguided and Misguided Discretion in Pennsylvania Capital Cases, 20 Just. Q. 187 (2003).

n79. Hannaford-Agor & Hans, Nullification at Work, supra note 29, at 1276-77.

n80. Johnson et al., supra note 42, at 487 (researching the use of voir dire in felony trials and addressing criticisms, including that it fails to eliminate prejudice of jurors); see also Dennis J. Devine, Kristi M. Olafson, LaRita L. Jarvis, Jennifer P. Bott, Laura D. Clayton & Jami M. T. Wolfe, Explaining Jury Verdicts: Is Leniency Bias for Real?, 34 J. of Applied Soc. Psychol. 2069 (2004) (using post-trial questionnaires to test whether initial votes predicted final verdicts).

n81. Seltzer et al., supra note 42, at 455; see also Reid Hastie, Is Attorney-Conducted Voir Dire an Effective Procedure for the Selection of Impartial Juries?, 40 Am. U. L. Rev. 703, 708 (1991); Norbert L. Kerr, Geoffrey P. Kramer, John S. Carroll & James J. Alfini, On the Effectiveness of Voir Dire in Criminal Cases with Prejudicial Pretrial Publicity: An Empirical Study, 40 Am. U. L. Rev. 665, 699 (1991) (commenting on jurors' failure to disclose pre-litigation exposure to the case in the media); Samuel R. Sommers & Michael I. Norton, Race-Based Judgments, Race-Neutral Justifications: Experimental Examination of Peremptory Use and the Batson Challenge Procedure, 31 Law & Hum. Behav. 261 (2007).

n82. See generally Hannaford, Hans & Munsterman, Permitting Jury Discussions During Trial, supra note 29. See also Shari Seidman Diamond, Neil Vidmar, Mary Rose, Leslie Ellis, & Beth Murphy, Juror Discussions During Civil Trials: Studying an Arizona Innovation, 45 Ariz. L. Rev. 1, 48 (2003); Shari Seidman Diamond & Neil Vidmar, Jury Room Ruminations on Forbidden Topics, 87 Va. L. Rev. 1857 (2001) (discussing the Arizona Jury Project that juxtaposed juries with pre-trial discussion from those without).

n83. Larry Heuer & Steven D. Penrod, Instructing Jurors: A Field Experiment with Written and Preliminary Instructions, 13 Law & Hum. Behav. 409, 424 (1989) [hereinafter Heuer & Penrod, Instructing Jurors]; cf. Cecil et al., supra note 33, at 30, 39 (finding that jurors reported certain instructions and procedures to have varying degrees of helpfulness but not conducting any research on jurors' actual understanding of the law). Contra Kramer & Koenig, supra note 76, at 428 (concluding that written jury instructions did improve juror comprehension).

n84. Young, Tinsley & Cameron, supra note 33, at 97-98.

n85. Id.; see also Zander & Henderson, supra note 33, at 216-17.

n86. Along similar lines, Sand and Reis found that seven of twelve jurors they interviewed stated that their notes helped them to remember facts and to keep track of exhibits. Leonard B. Sand & Steven A. Reiss, A Report On Seven Experiments Conducted by District Court Judges in the Second Circuit, 60 N.Y.U. L. Rev. 423, 450 (1985).

n87. Heuer & Penrod, Instructing Jurors, supra note 83, at 411; Heuer & Penrod, Increasing Jurors' Participation in Trials, supra note 33, at 245-46.

n88. For a discussion of the ways racial stereotypes can be activated without the perceiver realizing it, see Susan T. Fiske, Stereotyping, Prejudice, and Discrimination, in The Handbook of Social Psychology 357, 473 (D.T. Gilbert et al. eds., 4th ed. 1998).

n89. Samuel R. Sommers, On Racial Diversity and Group Decision-Making: Identifying Multiple Effects of Racial Composition on Jury Deliberations, 90 J. Personality & Soc. Psychol. 597, 599-600 (2006) [hereinafter Sommers, On Racial Diversity] (citing multiple studies that suggest that a focus on racial prejudice triggers jurors' concerns about appearing racially biased such that the effect of jurors' prejudice decreases); see Samuel R. Sommers & Michael I. Norton, Lay Theories About White Racists: What Constitutes Racism (And What Doesn't), 9 Group Processes & Intergroup Relations 117, 118, 125 (2006) (finding that people hold lay theories of racism which allow them to distance themselves from the "racist" label).

n90. Sommers, On Racial Diversity, supra note 89.

n91. Michael E. Antonio & Valerie P. Hans, Race and the Civil Jury: How Does a Juror's Race Shape the Jury Experience?, in Psychology in the courts: International advances in knowledge 69, 75 (Ronald Roesch et al., eds., 2001).

n92. Bowers, Sandys & Brewer, supra note 31, at 1502.

n93. Bowers, Steiner & Sandys, Death Sentencing, supra note 29, at 244.

n94. Dale W. Broeder, The Negro in Court, 1965 Duke L.J. 19, 21-24, 28 (1965).

n95. Id. at 30.

n96. See id.

n97. Gary D. LaFree, Rape and Criminal Justice: The Social Construction of Sexual Assault 219-20 (1989).

n98. Edelman, supra note 48, at 4-5, 59; see also Thomas W. Brewer, Race and Jurors' Receptivity to Mitigation in Capital Cases: The Effect of Jurors', Defendants', and Victims' Race in Combination, 28 Law & Hum. Behav. 529, 540-42 (2004) (finding that both Black and White jurors were more receptive to mitigation when the defendant was the same race as the juror and the victim was of a different race).

n99. Samuel R. Sommers, Race and the Decision-Making of Juries, 12 Legal & Criminological Psychol. 171, 177-78 (2007) (citing Bowers, Sandys & Brewer, supra note 31, at 1502-03); Bowers, Steiner & Sandys, Death Sentencing, supra note 29, at 257-58 (explaining that according to the CJP study, White jurors are less likely to mitigate because they may interpret Black defendants' lack of emotion as remorselessness or expressions of remorse as deceptive).

n100. Bowers, Sandys & Brewer, supra note 31, at 1501; see also Bowers, Steiner & Sandys, Death Sentencing, supra note 29, at 254-56 (describing juror experiences in which a Georgia jury had to "coach" a Black male juror holdout to impose the death sentence and another case in which a Black male juror refused to impose the death sentence despite the other jurors' hostility).

n101. Peters v. Kiff, 407 U.S. 493, 503-04 (1972) (The "exclusion [of a group] deprives the jury of a perspective on human events that may have unsuspected importance in any case."); Phoebe C. Ellsworth, Are Twelve Heads Better Than One?, 52 Law & Contemp. Probs. 205, 205-06 (1989); see Sommers, Race and the Decision-Making of Juries, supra note 99, at 180-81 (claiming that jury diversity is one possible explanation for the causal link between a jury's racial composition and its final decision).

n102. Marder, supra note 32, at 694, 699, 711 (finding that racial and age diversity had no effect on jurors' reported thoroughness, hostility, or unanimity of deliberations).

n103. Bowers, Sandys & Brewer, supra note 31, at 1519.

n104. Sommers, Race and the Decision-Making of Juries, supra note 99, at 174, 180, 183.

n105. Tara L. Mitchell, Ryann M. Haw, Jeffrey E. Pfeifer & Christian A. Meissner, Racial Bias in Mock Juror Decision-Making: A Meta-Analytic Review of Defendant Treatment, 29 Law & Hum. Behav. 621, 625, 627 (2005).

n106. Laura T. Sweeney & Craig Haney, The Influence of Race on Sentencing: A Meta-Analytic Review of Experimental Studies, 10 Behav. Sci. & L. 179, 190-91 (1992).

n107. Samuel R. Sommers & Phoebe C. Ellsworth, Race in the Courtroom: Perceptions of Guilt and Dispositional Attributions, 26 Personality & Soc. Psychol. Bull. 1367, 1368 (2000) [hereinafter Sommers & Ellsworth, Race in the Courtroom].

n108. See, e.g., Denis Chimaeze E. Ugwuegbu, Black Jurors' Personality Trait Attribution to a Rape Case Defendant, 4 Soc. Behav. & Personality 193, 194 (1976) (presenting simulated studies about the influence of race of the defendant on Black jurors in rape cases).

n109. See, e.g., James D. Johnson, Erik Whitestone, Lee Anderson Jackson & Leslie Gatto, Justice is Still Not Colorblind: Differential Racial Effects of Exposure to Inadmissible Evidence, 21 Personality & Soc. Psychol. Bull. 893, 893 (1995) (presenting studies on how inadmissible evidence impacts the verdict in simulated criminal trials where the defendant is Black versus where the defendant is White).

n110. See, e.g., Steven Fein, Seth J. Morgan, Michael I. Norton & Samuel R. Sommers, Hype and Suspicion: The Effects of Pretrial Publicity, Race, and Suspicion on Jurors' Verdicts, 53 J. Soc. Issues 487, 487 (1997) (presenting studies on the impact of pretrial publicity on mock jury verdicts and how verdicts differed based on similar or dissimilar race of juror and defendant).

n111. See, e.g., Michael J. Sargent & Amy L. Bradfield, Race and Information Processing in Criminal Trials: Does the Defendant's Race Affect How the Facts Are Evaluated?, 30 Personality & Soc. Psychol. Bull. 995, 995 (2004) (presenting studies concluding that a Black defendant can elicit more sensitivity to legally relevant facts than a White defendant).

n112. See J. L. Bernard, Interaction Between the Race of the Defendant and that of Jurors in Determining Verdicts, 5 Law & Psychol. Rev. 103, 109 (1979) (finding a shift from "guilty' to "not guilty' verdicts between first ballots decided alone and final ballots decided with other jurors); Jack P. Lipton, Racism in the Jury Box: The Hispanic Defendant, 5 Hisp. J. Behav. Sci. 275, 282 (1983) (finding that Anglo jurors were more likely to shift their decision towards innocence if the jury was predominantly Chicano while Chicano jurors were more likely to shift their decision towards guilt if the jury was predominantly White); Delores A. Perez, Harmon M. Hosch, Bruce Ponder & Gloria Chanez Trejo, Ethnicity of Defendants and Jurors as Influences on Jury Decisions, 23 J. Applied Soc. Psychol. 1249 (1993) (concluding that majority White juries were more lenient with White defendants than Hispanic defendants).

n113. See, e.g., Norbert L. Kerr, Robert W. Hymes, Alonzo B. Anderson & James E. Weathers, Defendant-Juror Similarity and Mock Juror Judgments, 19 Law & Hum. Behav., 545, 560 (1995) (finding, for example, that mock jurors will judge another ingroup member more harshly than an outgroup member when they anticipate that most of the jury will be composed of outgroup members).

n114. David C. Baldus, George Woodworth, David Zuckerman, Neil Alan Weiner & Barbara Broffitt, The Use of Peremptory Challenges in Capital Murder Trials: A Legal and Empirical Analysis, 3 U. Pa. J. Const. L. 3, 18 (2001) (citing studies that show a distinct relationship between the probability of conviction and the number of Black jurors); Bowers, Steiner & Sandys, Death Sentencing, supra note 29, at 193 (finding, for example, that the dominance of White male jurors is strongly associated with the imposition of a death sentence in Black defendant/White victim cases, and the presence of Black male jurors is strongly associated with the imposition of a life sentence in Black defendant/White victim cases). See generally Bowers, Sandys & Brewer, supra note 31; Howard C. Daudistel, Harmon M. Hosch, Malcolm D. Holmes & Joseph B. Graves, Effects of Defendant Ethnicity on Juries' Dispositions of Felony Cases, 29 J. Applied Soc. Psychol. 317 (1999).

n115. See generally Sommers, On Racial Diversity, supra note 89.

n116. Id. at 602.

n117. Id. at 601.

n118. Id. at 601-02.

n119. Id. at 602.

n120. Id. at 604.

n121. Sommers, On Racial Diversity, supra note 89, at 600.

n122. Nancy J. King, Postconviction Review of Jury Discrimination: Measuring the Effects of Juror Race on Jury Decisions, 92 Mich. L. Rev. 63 (1993); Sommers & Ellsworth, How Much Do We Really Know, supra note 7.

n123. Sommers & Ellsworth, Race in the Courtroom, supra note 107, at 1369.

n124. Id. at 1374.

n125. Samuel R. Sommers & Phoebe C. Ellsworth, Unpublished Pilot Study for Race in the Courtroom: Perceptions of Guilt and Dispositional Attributions. The data from the study are on file with the second author, to whom any questions can be directed.

n126. Id.

n127. Id.

n128. See generally Sommers, On Racial Diversity, supra note 89.

n129. See id.

n130. For details regarding the data collection methods and data analyses of this study, see generally id. As detailed therein, mock jurors in the study were jury-eligible citizens recruited in the midst of actual jury duty at a local courthouse. Of course, given that it was a simulated trial, mock juries' verdicts carried no real repercussions. The implications of this aspect of the design, as well as the general nature of the juror-and jury-level results observed, are discussed in detail in this previous article. The data from the study are on file with the second author, to whom any questions can be directed.

n131. Average group r (n = 28) = .47, p < .01

n132. Average r = .31, n.s.

n133. Average group M = 4.90 on a scale of 1-9.

n134. Average M = 3.74, t (26) = 2.52, p < .02

n135. Average r = .19, n.s.

n136. Average r = -.28, n.s.

n137. Average r = .56, p < .01.

n138. See Lee Ross, David Greene & Pamela House, The "False Consensus Effect": An Egocentric Bias in Social Perception and Attribution Processes, 13 J. Experimental Soc. Psychol. 279 (1977) (citing four studies demonstrating that social observers tend to perceive a "false consensus" with respect to the relative commonness of their own responses based on social inferences).

n139. See generally Sommers, On Racial Diversity, supra note 89.

n140. Jury level rs < .08.

n141. r (n = 29) = .23, n.s.

n142. r (n = 29) = -.39, p < .04.

n143. r (n = 29) -.35, p = .07.

n144. See generally Sommers, On Racial Diversity, supra note 89.

n145. Group M = 2.59 on a 1-9 scale.

n146. F (1, 25) < 1.

n147. Group M = 2.61.

n148. Fs <1.

n149. See also Sommers & Ellsworth, How Much Do We Really Know, supra note 7, at 1000-01 (reviewing the use of juror interviews in the investigation of race in deciding criminal cases). See generally Sommers, On Racial Diversity, supra note 89 (examining the effects of racial diversity on jury decision-making).

n150. Id., at 601; Sommers & Ellsworth, How Much Do We Really Know, supra note 7, at 1001.

n151. Sommers & Ellsworth, How Much Do We Really Know, supra note 7, at 1000-01.

n152. See Sommers & Ellsworth, Unpublished Data, supra note 125.

n153. Id.

n154. Id.

n155. Sommers, On Racial Diversity, supra note 89, at 608; Sommers & Ellsworth, How Much Do We Really Know, supra note 7, at 1026-28.

n156. See Sommers & Ellsworth, Unpublished Data, supra note 125.

n157. Id.

n158. See supra Part I.

n159. See generally Bowers, supra note 29.

n160. See Bowers, supra note 2, n.208, 210 (noting that asking open-ended questions in CJP interviews produced richer and more detailed information about deliberations than structured questions alone); Diamond, Beyond Fantasy and Nightmare, supra note 7, at 757-58 (discussing variability in how much individual jurors talk during deliberations).

n161. See Sommers & Ellsworth, Unpublished Data, supra note 125.

n162. See, e.g., William J. Bowers, Wanda D. Foglia, Susan Ehrhard-Dietzel, and Christopher E. Kelly, Jurors' Failure to Understand or Comport with Constitutional Standards in Capital Sentencing: Strength of the Evidence, 6 Crim. L. Bull. 1147, 1173 (2010) (calculating the median time between trial and interview of CJP jurors from 353 cases as 2.22 years).

n163. See generally Daryl J. Bem & H. Keith McConnell, Testing the Self-Perception Explanation of Dissonance Phenomena: On the Salience of Premanipulation Attitudes, 14 J. Personality & Soc. Psychol. 23 (1970).

n164. Heuer & Penrod, Instructing Jurors, supra note 83, at 420.

n165. Young, Tinsley & Cameron, supra note 33, at 97; cf. Zander & Henderson, supra note 33, at 216-17 (finding that few British jurors reported difficulty in understanding the judge's summing up of the law).

n166. Phoebe C. Ellsworth & Samuel R. Gross, Hardening of the Attitudes: Americans' Views on the Death Penalty, 50 J. Social Issues 19, 25 (1994).

n167. See Schacter, The Seven Sins of Memory, supra note 13, for a discussion of how people underestimate the extent to which their attitudes have changed over time.

n168. Sommers & Ellsworth, Unpublished Data, supra note 125.

n169. See supra Part III.B-C.

n170. Sommers, On Racial Diversity, supra note 89, at 174.

n171. Valerie P. Hans & Anthony N. Doob, Section 12 of the Canada Evidence Act and the Deliberations of Simulated Juries, 18 Crim. L. Q. 235, 242, 249 (1975-76).

n172. Id. at 242-49.

n173. See Elliot Aronson, Timothy D. Wilson, & Marilynn B. Brewer, Experimentation in Social Psychology, in The Handbook of Social Psychology 99, 100-05 (Daniel T. Gilbert, Susan T. Fiske, & Gardner Lindsey eds., 4th ed. 1998) (discussing the superiority of experiments to correlational methods for uncovering causal processes).

n174. See infra Part III.

n175. See generally Brian L. Cutler, Steven D. Penrod & Hedy R. Dexter, The Eyewitness, the Expert Psychologist, and the Jury, 13 Law & Hum. Behav. 311 (1989) (demonstrating that the presence of expert testimony improved juror sensitivity to eyewitness evidence in a study where mock jurors were shown a realistic videotaped trial centered around eyewitness evidence); Natalie J. Gabora, Nicholas P. Spanos & Amanda Joab, The Effects of Complainant Age and Expert Psychological Testimony in a Simulated Child Sexual Abuse Trial, 17 Law & Hum. Behav. 103 (1993) (demonstrating that, in a simulated sexual abuse trial, jurors voted to convict more often and rated the defendant as less credible when presented with expert psychological testimony specific to the case); Margaret Bull Kovera, April W. Gresham, Eugene Borgida, Ellen Gray & Pamela C. Regan, Does Expert Psychological Testimony Inform or Influence Juror Decision Making? A Social Cognitive Analysis, 82 J. Applied Psychol. 178 (1997) (demonstrating that in a simulated sexual abuse trial with a child witness, repetitive expert testimony bolstered the child's testimony, whereas concrete and standard expert testimony did not).

n176. See, e.g., Bienen, supra note 30 (employing interviews with jurors from homicide and capital trials in New Jersey to inform a study of the usefulness of post-verdict debriefing for jurors in emotionally disturbing trials); Kelley, supra note 30 (using the results of a questionnaire sent to jurors who deliberated in forty-four murder trials in Iowa to draw conclusions about juror stress).

n177. See, e.g., David A. Bright & Jane Goodman-Delahunty, Gruesome Evidence and Emotion: Anger, Blame, and Jury Decision-Making, 30 Law & Hum. Behav. 183 (2006) (employing self-report measures to determine the influence of gruesome evidence on the emotional states of mock jurors).

n178. Dennis Devine, Laura Clayton, Benjamin Dunford, Ramsy Seying & Jennifer Pryce, Jury Decision Making: 45 Years of Empirical Research on Deliberating Groups, 7 Psychol. Pub. Pol'y & L. 622, 677 (2001) (discussing research on the effect of jury service on people's legal attitudes and how they approach subsequent jury service).

n179. See, e.g., Heuer & Penrod, Instructing Jurors, supra note 83, at 420 (noting that despite jurors' assessments of written judicial instructions as helpful, those who received them performed no better than others on a post-trial multiple choice test about the relevant law); see also supra notes 69-74 and accompanying text.

n180. See, e.g., Young, Tinsley & Cameron, supra note 33, at 92.

n181. Vidmar, supra note 7, at 57-58.

n182. Young, Tinsley & Cameron, supra note 33, at 97.

n183. See generally Nisbett & Wilson, supra note 17.

n184. Donald T. Campbell & Donald W. Fiske, Convergent and Discriminant Validation by the Multitrait-Multimethod Matrix, 56 Psychol. Bull. 81, 104 (1959); see also Sommers & Ellsworth, How Much Do We Really Know, supra note 7, at 997-98 (listing the primary methodologies used to analyze the relationship between race and criminal jury trials); Hans, supra note 49, at 17 (combining experimental, survey, and interview methodologies in the context of civil juries in business regulation and litigation as a "research approach to triangulation").

n185. See Bowers, supra note 29, at 1077.

n186. See Hannaford-Agor & Hans, Nullification at Work, supra note 29, at 1253; Hans, Hannaford & Munsterman, supra note 32, at 349.

n187. See Diamond, Vidmar, Rose, Ellis & Murphy, supra note 82, at 5 (reporting findings from videotaped deliberations of fifty juries as well as the results of post-deliberation questionnaires). See generally Mary R. Rose & Shari Seidman Diamond, Offstage Behavior: Real Jurors' Scrutiny of Non-Testimonial Conduct, 58 DePaul L. Rev. 311 (2009) (observing videotaped jury discussion and deliberation in Arizona to evaluate the practice of allowing civil jurors to discuss evidence mid-trial).

n188. See Diamond, Vidmar, Rose, Ellis & Murphy, supra note 82, at 5.