. Throughout the trial, judges instruct jurors that their job is to find the facts. As one typical instruction provides: "Your purpose as jurors is to find and determine the facts. Under our system of justice you are the sole judge of the facts." 1A Federal Jury Practice and Instructions: Criminal § 10.01, at 5 (5th ed. 2000). Judges also instruct jurors that they are to take the law as the judge gives it to them: "I instruct you that the law as given by the Court in these and other instructions constitute the only law for your guidance. It is your duty to accept and to follow the law as I give it to you even though you may disagree with the law." Id. Judges emphasize that the jury’s role is quite different from the judge’s role, see id. at 8 ("Your job is to decide all of the factual questions in this case . . . . I will decide all of the legal questions in this case . . . ."), and that the jury’s central role is to find facts. See, e.g., id. at 20 ("You, and you alone, are the judges of the facts.") (quoting Fifth Circuit instruction); id. at 46 ("By your verdict(s) you will decide the disputed issues of fact.") (quoting Eleventh Circuit instruction); id. at 38 ("It will be your duty to decide from the evidence what the facts are.") (quoting Ninth Circuit instruction).
 . See CBS Reports: Enter the Jury Room (CBS television broadcast, Apr. 16, 1997) (Transcript of broadcast at 49). Judge Ryan presided over the trial of Modesta Solano, who had been charged with possession and transportation of drugs. The first trial resulted in a hung jury. Judge Ryan had the rare opportunity to observe that jury’s deliberations because it had been filmed by prior arrangement. See id.
 . Judge Ryan said of the hold-out who had chosen to vote consistent with his conscience: "He just didn’t feel comfortable with what went on here, and so he’s interjecting his personal and moral beliefs into it, his—his conscience. And maybe . . . it was legally and rationally the wrong thing to do, but morally he is probably right." Id. Judge Ryan acknowledged that there might possibly be a benefit to the nullifying juror who votes consistent with his conscience.
 . See, e.g., Taylor v. Louisiana, 419 U.S. 522, 530 (1975) ("The purpose of a jury is to guard against the exercise of arbitrary power—to make available the commonsense judgment of the community as a hedge against the overzealous or mistaken prosecutor and in preference to the professional or perhaps over-conditioned or biased response of a judge."); Duncan v. Louisiana, 391 U.S. 145, 156 (1968) ("Providing an accused with the right to be tried by a jury of his peers gave him an inestimable safeguard against the corrupt or overzealous prosecutor and against the compliant, biased, or eccentric judge.").
 . See Nancy S. Marder, The Myth of the Nullifying Jury, 93 Nw. U. L. Rev. 877, 907–26 (1999) (describing a "process view" of the jury that focuses on the broad roles played by the modern jury).
 . See id. at 908–20 (describing the jury’s interpretive role).
 . See Stephen C. Yeazell, The New Jury and the Ancient Jury Conflict, 1990 U. Chi. Legal F. 87, 113–14 (describing the ways in which the jury helped to shape the law in the areas of negligence and products liability).
 . See Marder, supra note , at 921–25.
 . Alexis de Tocqueville, Democracy in America 275 (J.P. Mayer ed., George Lawrence trans., HarperPerennial 1988) (1850).
 . See Marder, supra note , at 888–89 nn.42–43.
 . 419 U.S. 522 (1975).
 . 476 U.S. 79 (1986).
 . 380 U.S. 202 (1965).
 . Batson, 476 U.S. at 92–93 (describing the burden required of defendants after Swain as "crippling," thus leaving prosecutors’ peremptory challenges "largely immune from constitutional scrutiny").
 . See Georgia v. McCollum, 505 U.S. 42, 59 (1992) (prohibiting the defendant’s use of peremptories based on race); Batson, 476 U.S. at 86 (prohibiting the prosecutor’s use of peremptories based on race). See also Powers v. Ohio, 499 U.S. 400, 402–04 (1991) (establishing that white defendants can challenge a prosecutor’s use of a race-based peremptory challenge).
 . See Hernandez v. New York, 500 U.S. 352, 358 (1991) (recognizing that peremptories cannot be exercised on the basis of ethnicity, though in this case the defendant failed to establish a prima facie case).
 . See J.E.B. v. Alabama ex rel. T.B., 511 U.S. 127, 129 (1994) (extending Batson to gender-based peremptory challenges).
 . See Edmonson v. Leesville Concrete Co., 500 U.S. 614, 616 (1991) (holding that private litigants in a civil suit cannot use peremptory challenges to exclude jurors on account of their race).
 . See sources cited supra notes –.
 . The Court’s efforts notwithstanding, discriminatory peremptories persist. See, e.g., Minetos v. City Univ., 925 F. Supp. 177, 185 (S.D.N.Y. 1996) ("It is time to put an end to this charade. We have now had enough judicial experience with the Batson test to know that it does not truly unmask racial discrimination."); Developments—The Civil Jury, 110 Harv. L. Rev. 1408, 1462 & nn.177–78 (1997).
 . See Nancy S. Marder, The Interplay of Race and False Claims of Jury Nullification, 32 U. Mich. J.L. Reform. 285 (1999) (describing two cases in which the juries were largely homogeneous and the verdicts were criticized by various communities and subjected to charges of jury nullification).
 I use the terms "diversity" and "cultural diversity" broadly and interchangeably and assume that they include many different characteristics, such as gender, age, and race, to name just a few.
 . 419 U.S. 522, 538 (1975).
 . See id. at 537 ("Accepting as we do, however, the view that the Sixth Amendment affords the defendant in a criminal trial the opportunity to have the jury drawn from venires representative of the community, we think it is no longer tenable to hold that women as a class may be excluded or given automatic exemptions based solely on sex if the consequence is that criminal jury venires are almost totally male.").
 . See id. at 542 (Rehnquist, J., dissenting) ("[P]resumably doctors, lawyers, and other groups, whose frequent exemption from jury service is endorsed by the majority, also offer qualities as distinct and important as those at issue here.").
 . See id. at 541 ("[T]he criminal defendant involved makes no claims of prejudice or bias.").
 . See id. at 543 ("Absent any suggestion that appellant’s trial was unfairly conducted, or that its result was unreliable, I would not require Louisiana to retry him . . . .").
 . Id. at 542 (quoting Ballard v. United States, 329 U.S. 187, 194 (1946)).
 . Id.
 . But see Catharine Pierce Wells, Tort Law As Corrective Justice: A Pragmatic Justification for Jury Adjudication, 88 Mich. L. Rev. 2348, 2402 (1990) (explaining that there are no omniscient observers who judge from "an ideal post of observation" but merely observers who judge, at best, from "partial perspective[s]"); Catharine A. MacKinnon, Feminism, Marxism, Method, and the State: Toward Feminist Jurisprudence, 8 Signs: J. Women Culture & Soc’y 635 (1983) (dismissing any claim to an objective, universal point of view as the male perspective).
 . See, e.g., United States v. Raszkiewicz, 169 F.3d 459, 466 (7th Cir. 1999) ("As noted, the main purpose of the fair cross-section requirement is that the defendant get the benefit of an impartial jury."); United States v. Ashley, 54 F.3d 311, 313 (7th Cir. 1995) ("While juries must be taken from a source that is representative of the community . . . , the Constitution does not require this to ensure representative juries, but rather impartial juries.") (citation omitted); Silagy v. Peters, 905 F.2d 986, 1011 (7th Cir. 1990) ("The ultimate concern of the fair-cross-section requirement is to ensure that each criminal defendant be afforded his sixth amendment right to an ‘impartial jury.’").
 . See Taylor, 419 U.S. at 530–33.
 . See id. at 530–31 ("‘[T]he broad representative character of the jury should be maintained, partly as assurance of a diffused impartiality and partly because sharing in the administration of justice is a phase of civic responsibility.’") (quoting Thiel v. S. Pac. Co., 328 U.S. 217, 227 (1946) (Frankfurter, J., dissenting)).
 . 439 U.S. 357 (1979). To establish a prima facie violation of the fair-cross-section requirement, Duren requires a defendant to show:
(1) that the group alleged to be excluded is a ‘distinctive’ group in the community; (2) that the representation of this group in venires from which juries are selected is not fair and reasonable in relation to the number of such persons in the community; and (3) that this underrepresentation is due to systematic exclusion of the group in the jury-selection process.
Id. at 364.
 . See Taylor, 419 U.S. at 531–33 (citing Ballard v. United States, 329 U.S. 187, 193–94 (1946); Peters v. Kiff, 407 U.S. 493, 502–04 (1972)).
 . Another way in which the Seventh Circuit has made it more difficult for defendants to establish distinctiveness is by establishing criteria for distinctiveness, which it borrowed from the Eleventh Circuit. Whereas the Supreme Court in Duren had left the concept of distinctiveness vague and open ("that the group alleged to be excluded is a ‘distinctive’ group in the community," Duren, 439 U.S. at 364), the Seventh Circuit has said that the elements of distinctiveness are: "(1) the existence of qualities that define a group, (2) similarity of attitudes, beliefs, or experiences, and (3) a community of interest among group members." Raszkiewicz, 169 F.3d at 463. One paradox of this test is that the larger the excluded group, and therefore, the more apparent the fair-cross-section violation, the more difficult it is to show distinctiveness because of the unlikelihood that a large group would share a "similarity of attitudes, beliefs, or experiences" and "a community of interest."
 . See Raszkiewicz, 169 F.3d at 466 ("A defendant need not, in addition to the three requirements imposed by Duren, 439 U.S. at 364 . . . , also show that he was prejudiced by the alleged constitutional violation.").
 . See Peters, 407 U.S. at 493.
 . See Taylor, 419 U.S. at 531.
 . See id. at 538 ("It should also be emphasized that in holding that petit juries must be drawn from a source fairly representative of the community we impose no requirement that petit juries actually chosen mirror the community and reflect the various distinctive groups in the population.").
 . See id.
 . 407 U.S. at 503.
 . Id.
 . Justices Douglas and Stewart joined Justice Marshall in holding that the exclusion of African Americans from the venire violated a white defendant’s right to due process. Id. at 493.
 . 329 U.S. 187 (1946).
 . Id. at 193–94.
 . See id. at 193.
 . 419 U.S. at 531–32 & n.12.
 . Id. at 531.
 . See, e.g., United States v. Raszkiewicz, 169 F.3d 459, 465–67 (7th Cir. 1999) (holding that reservation Native Americans are not distinct from urban Native Americans for fair-cross-section purposes, and therefore, their exclusion from the venire did not violate defendant’s Sixth Amendment fair-cross-section right); United States v. Barry, 71 F.3d 1269, 1273–74 (7th Cir. 1995) (concluding that defendants failed to establish that accused, but not yet convicted, felons are a distinct group for fair-cross-section purposes).
 . Taylor, 419 U.S. at 531–33.
 . In Taylor, the Court also thought the systematic exclusion of a distinct group would impede the jury’s capacity to perform its other functions, including serving as a buffer between the defendant and the government, providing for "[c]ommunity participation in the administration of the criminal law," and maintaining "‘the broad representative character of the jury . . . partly as assurance of a diffused impartiality and partly because sharing in the administration of justice is a phase of civic responsibility.’" Id. at 530–31 (quoting Thiel v. S. Pac. Co., 328 U.S. 217, 227 (1946) (Frankfurter, J., dissenting)).
 . 407 U.S. at 503–04.
 . See Taylor, 419 U.S. at 538 ("It should also be emphasized that in holding that petit juries must be drawn from a source fairly representative of the community we impose no requirement that petit juries actually chosen must mirror the community and reflect the various distinctive groups in the population.").
 . 476 U.S. 79 (1986).
 . Id. at 91.
 . 380 U.S. 202 (1965).
 . See id. at 223–24.
 . Some prosecutors did this not just by happenstance but because it was part of their training. When Batson was before the Court, Justice Marshall circulated newspaper articles among the justices that reported on Texas prosecutors’ training, which included a manual instructing them to exercise their peremptories against African-American prospective jurors. See Batson, 476 U.S. at 104 & n.3 (Marshall, J., concurring) (noting that a prosecutor’s office manual used in Dallas County, Texas, instructed prosecutors to use peremptories to eliminate members of minority groups from juries); Memorandum from Justice Marshall, to the Conference (Mar. 24, 1986) (located in the Justice Marshall Papers at the Library of Congress) (copy on file with author) (sharing with the Court articles from Dallas newspapers that revealed that prosecutors used their peremptories to exclude African Americans and other minorities because of a stereotypical view that they would be more sympathetic to the defense).
 . See Marder, supra note , at 302–03 & n.86 (noting that stereotypes about African Americans’ competence had a long history in this country and shaped some of the press’ treatment of the verdict in the O.J. Simpson state criminal trial).
 . See Batson, 476 U.S. at 138 (Rehnquist, J., dissenting) ("[A]s long as [peremptories] are applied across-the-board to jurors of all races and nationalities, I do not see . . . how their use violates the Equal Protection Clause.").
 . Even Justice Rehnquist agreed: "Such use of peremptories is at best based upon seat-of-the-pants instincts, which are undoubtedly crudely stereotypical and may in many cases be hopelessly mistaken." Id.
 . See supra Part II.B.
 . See 476 U.S. at 87 ("The harm from discriminatory jury selection extends beyond that inflicted on the defendant and the excluded juror to touch the entire community.").
 . See id. at 97.
 . See id. at 98.
 . Id. at 92–93.
 . But see Purkett v. Elem, 514 U.S. 765 (1995). Although this opinion is a per curiam rather than a signed opinion based on plenary review, it could be seen as a half-step back from Batson, 476 U.S. at 79, and its progeny. In Elem, the Court held that a party’s reasons for the exercise of a peremptory need not be related to the facts of the case as long as the reasons are nondiscriminatory. This allows a party to give any reason at all, no matter how fanciful, which is at odds with the Court’s requirement in Batson that the reason must be "a neutral explanation related to the particular case to be tried." Batson, 476 U.S. at 98.
 . 499 U.S. 400 (1991).
 . 500 U.S. 614 (1991).
 . 505 U.S. 42 (1992).
 . 511 U.S. 127 (1994).
 . See Batson, 476 U.S. at 108 (Marshall, J., concurring) ("[O]nly by banning peremptories entirely can such discrimination be ended.").
 . Id. at 107.
 . See infra note .
 . See infra note .
 . The federal rules of civil and criminal procedure give federal courts discretion about whether they allow attorneys to conduct the entire voir dire or whether the court conducts the entire voir dire and merely allows attorneys to supplement questions asked by the court. See Fed. R. Civ. P. 47(a); Fed. R. Crim. P. 24(a). In practice, however, federal courts typically conduct the voir dire. According to one survey, based on 420 completed questionnaires, approximately three-fourths of federal district judges conduct the voir dire without oral participation of the attorneys. Gordon Bermant, Conduct of the Voir Dire Examination: Practices and Opinions of Federal District Judges 6 (1977). According to another study, 53.4% of federal courts allow only the judge to conduct voir dire in criminal cases; 31.1% allow the attorneys to submit supplemental questions; 13.2% allow the attorneys to ask all of the questions; and 2.3% allow the attorneys or clerks to ask questions outside the presence of the court. Works of the Comm. on the Operation of the Jury Sys. of the Judicial Conference of the United States, The Jury System in the Federal Courts 174 (1973). The trend is toward judge-conducted voir dire. See, e.g., Jury Trial Innovations 53 (G. Thomas Munsterman et al. eds., 1997) ("[T]he majority of federal judges and a substantial portion of state judges have, over time, assumed a dominant role in examining potential jurors in voir dire."); John B. Ashby, Juror Selection and the Sixth Amendment Right to an Impartial Jury, 11 Creighton L. Rev. 1137, 1158 (1978). A recent example of a state that has moved in the direction of curbing attorney-conducted voir dire and increasing judicial supervision of the process is New York. See James Barron, Top New York State Judge Issues Rules to Ease Life on Civil Juries, N.Y. Times, Nov. 2, 1995, at A1 (describing Chief Judge Judith Kaye’s efforts to reform the New York State jury system, including "streamlin[ing] a lawyer-dominated process that she called ‘an annoyance and a frustration’ to jurors who sometimes spend their two weeks of jury duty without getting selected for a case or seeing a judge").
 . Barbara Babcock observed this trend twenty-five years ago and lamented it then. See Barbara Allen Babcock, Voir Dire: Preserving "Its Wonderful Power," 27 Stan. L. Rev. 545 (1975).
 . See 28 U.S.C. § 1865(b) (1994) (providing qualifications for jury service, which include: residence in the judicial district for at least one year; attaining the age of eighteen years or older; citizenship; an ability to read, write, speak, and understand English; no physical or mental impairment that would interfere with jury service; and no charges pending or conviction for a crime punishable by imprisonment for more than one year).
 . See Hopt v. Utah, 120 U.S. 430, 433 (1887) (providing some guidance as to when a for-cause challenge should be granted).
 . See supra note .
 . Some judges have recognized the need to ask follow-up questions of prospective jurors individually and in the more private setting of the robing room. See, e.g., Babcock, supra note , at 547 ("[A]nother method of restricting the information-gathering function of voir dire is to address all questions to the jury panel at once, rather than as individuals."); Kimba M. Wood, The 1995 Justice Lester W. Roth Lecture: Reexamining the Access Doctrine, 69 S. Cal. L. Rev. 1105, 1119 (1996) ("I began asking each juror sensitive questions in the robing room . . . without having required them to request a private session. The difference in the quantity and the quality of the information the jurors revealed was striking . . . ."); Gregory E. Mize, On Better Jury Selection: Spotting UFO Jurors Before They Enter the Jury Room, Ct. Rev., Spring 1999, at 10 (recommending that all prospective jurors be individually questioned, even those who had remained silent during the general questions).
 . For a sample voir dire in federal court, see Transcript of Voir Dire, United States v. Torres, No. 77 Cr. 680 (S.D.N.Y. May 19, 1980) [hereinafter Torres Transcript]. This voir dire, conducted by Judge Knapp, was probably more searching than most conducted by a judge in federal court because the defendant, who represented herself, refused to be present. Thus, Judge Knapp took special care to make sure that any prospective jurors who felt they could not serve because of hardship or feelings about the case, no matter how vague, were immediately dismissed.
 . See, e.g., Torres Transcript, supra note , at 70–71, 74, 77, 79, 83, 89, 91, 96, 98 (providing exchanges during the voir dire when the judge asked prospective jurors if they thought they could be impartial). The judge, however, often has some discretion as to how far to push this inquiry. For example, in Ham v. South Carolina, 409 U.S. 524 (1973), in which the defendant was an African-American man with a beard, the Court held that the trial judge was obligated by due process to ask jurors whether they were prejudiced against African Americans, but not whether they were prejudiced against defendants with facial hair. Id. at 527–28.
 . See, e.g., Wood, supra note , at 1120 ("My experience left no doubt in my mind that most prospective jurors will choose to say nothing rather than reveal to everyone in the courtroom that they, for example, . . . have certain ethnic or racial biases . . . .").
 . The number of peremptory challenges varies based upon whether the case is civil or criminal or in federal or state court. For example, in a federal civil case, each side has three peremptory challenges. See 28 U.S.C. § 1870 (1994). In a federal criminal case, the number varies based upon the potential penalty. See Fed. R. Crim. P. 24(b). In a capital case, both the government and the defendant each have twenty peremptory challenges, whereas if the offense is punishable by more than one year’s imprisonment, then the government has six peremptory challenges and the defendant has ten. Id.
 . The lawyers, however, are subject to certain restraints in the exercise of their peremptory challenges. They cannot exercise peremptory challenges based upon race, gender, or ethnicity. See supra text accompanying notes –.
 . One of the lengthiest questionnaires given to jurors was in the state criminal trial of O.J. Simpson. See People v. Simpson, No. BA097211 (Cal. Super. Ct. L.A. County 1995). The questionnaire in that case contained 435 questions.
 . See, e.g., Jay Schulman, Phillip Shaver, Robert Colman, Barbara Emrich & Richard Christie, Recipe for a Jury, Psychol. Today, May 1973, at 37 (describing the role of social scientists in helping the Catonsville Nine defendants to develop a profile of the jury-eligible population in Harrisburg, Pennsylvania, where the trial was held, and to rate the prospective jurors in order to decide how defendants should exercise their peremptory challenges).
 . For example, in Ham v. South Carolina, 409 U.S. 524 (1973), the trial judge precluded questions about race and facial hair, even though the defendant, an African-American civil rights worker with a beard, was being tried for a drug violation in South Carolina. Id. at 526. The Supreme Court held that the Fourteenth Amendment required that the defendant be permitted to ask prospective jurors whether they could be impartial in spite of defendant’s race, id. at 527, but held that the Constitution did not require that defendant be permitted to ask about facial hair and its effects on juror impartiality. Id. at 528.
 . See, e.g., Babcock, supra note , at 549, 557, 558–63 (arguing that an extensive voir dire is necessary to the exercise of the peremptory challenge and is constitutionally required by due process and equal protection).
 . See Nancy S. Marder, Beyond Gender: Peremptory Challenges and the Roles of the Jury, 73 Tex. L. Rev. 1041, 1088 & n.188 (1995).
 . See Schulman et al., supra note , at 37.
 . See, e.g., Saul M. Kassin & Lawrence S. Wrightsman, The American Jury on Trial: Psychological Perspectives 52 (1988) ("[Lawyers] argue that only they are familiar enough with their cases, skilled enough in the art of asking questions, and motivated enough by the demands of their role as advocates, to conduct an effective examination.").
 . See, e.g., Judge-Conducted Voir Dire, 274 Prac. L. Inst./Litig. 541, 543 (1984) (noting that in seven states, voir dire is conducted largely by attorneys; in the remaining states, it is conducted largely by judges).
 . See Barron, supra note (describing New York’s move from attorney-conducted to attorney- and judge-conducted voir dire to reduce attorneys’ opportunities to abuse voir dire).
 . Other legal academics share this intuition as well. See, e.g., Valerie P. Hans & Neil Vidmar, Judging the Jury 50 (1986) ("[A] jury composed of individuals with a wide range of experiences, backgrounds, and knowledge is more likely to perceive the facts from different perspectives and thus engage in a vigorous and thorough debate."); Jeffrey Abramson, Two Ideals of Jury Deliberation, 1998 U. Chi. Legal F. 125, 153 (describing juries that are not representative as juries that "may weaken the information base upon which good deliberation depends, leaving jurors from one section of town uninformed about facts of life on the other side of the tracks"); Deborah Ramirez, Affirmative Jury Selection: A Proposal to Advance Both the Deliberative Ideal and Jury Diversity, 1998 U. Chi. Legal F. 161, 165. Ramirez observed:
[R]acially diverse juries bring to their deliberations a broader range of life experiences that allow them to use their common sense more effectively when they evaluate the facts presented at trial. . . . To the extent that a racially mixed jury facilitates the sharing of diverse perspectives, information, and experiences, that sharing may lead to a more thoughtful and informed verdict.
 . See Marder, supra note , at 1080 & n.157.
 . The only correlations were between gender and cases that involved rape or the death penalty. According to some studies, women were more likely than men to convict in rape cases, see Marsha B. Jacobson, Effects of Victim’s and Defendant’s Physical Attractiveness on Subjects’ Judgments in a Rape Case, 7 Sex Roles 247, 252–53 (1981), or to assign the defendant a longer sentence, see Anne Renkin Mahoney, Sexism in Voir Dire: The Use of Sex Stereotypes in Jury Selection, in Women in the Courts 126 (Winifred L. Hepperle & Laura Crites eds., 1978).
Several studies have found that women are more likely than men to oppose the death penalty, and consequently, to be excluded from "death-qualified" juries. See, e.g., Claudia L. Cowan, William C. Thompson & Phoebe C. Ellsworth, The Effects of Death Qualification on Jurors’ Predisposition To Convict and on the Quality of Deliberation, 8 Law & Hum. Behav. 53, 67 (1984) (finding that potential jurors excluded from death-qualified juries are more likely to be women); Robert Fitzgerald & Phoebe C. Ellsworth, Due Process vs. Crime Control: Death Qualification and Jury Attitudes, 8 Law & Hum. Behav. 31, 46 (1984) (finding that death-qualified juries are more likely to exclude women and African-American men). A "death-qualified" jury is one in which those who say that they could not vote for the death penalty in any case are excluded from the jury. See Witherspoon v. Illinois, 391 U.S. 510, 520 (1968) (holding that the State’s exclusion of prospective jurors who were not just unalterably opposed to the death penalty, but who simply "expressed conscientious or religious scruples against capital punishment and all who opposed it in principle" violated the requirements of an impartial jury as provided by the Sixth and Fourteenth Amendments); Lockhart v. McCree, 476 U.S. 162, 167 (1986) (holding that "‘death-qualification,’ or the removal for cause of the so-called ‘Witherspoon-excludable’ prospective jurors" from the guilt phase does not violate the fair-cross-section requirement of the Sixth Amendment).
 . Although my focus is on the deliberation rather than the verdict, I think that the deliberation will, in turn, affect the verdict.
 . See Lisa Hope Pelled, Demographic Diversity, Conflict, and Work Group Outcomes: An Intervening Process Theory, 7 Org. Sci. 615, 626 (1996) ("A limitation of most diversity studies is that they consider only one or two types of diversity simultaneously rather than a larger set; they do not deal with the combined effects of diversity across multiple dimensions.").
 . See infra Appendix B, question 15 (asking jurors to rate the diversity of their jury on a seven-point scale).
 . See, e.g., Priscilla M. Elsass & Laura M. Graves, Demographic Diversity in Decision-Making Groups: The Experiences of Women and People of Color, 22 Acad. Mgmt. Rev. 946 (1997) ("In diverse decision-making groups, members have different experiences, values, attitudes, and cognitive approaches; consequently, they bring divergent perspectives to the group’s problem.").
 . Even if jurors have different demographic characteristics, this does not necessarily mean they will have different views. Cf. Brooke Harrington, Cohesion, Conflict and Group Demography: A Multi-Method Integration 6 (1999) (unpublished paper on file with author) ("As this study will show, both kinds of diversity [demographic diversity and diversity of opinion] have an important influence on [investment club] performance, but diversity of opinion cannot be inferred directly from demography."). Even if jurors do have different views, they may not always feel free to express those different views during the deliberations. In both jury and nonjury contexts, studies have found that women and minority members do not participate at the same rate as white men and do not occupy the same positions of influence as white men. See, e.g., Elsass & Graves, supra note , at 954 ("Evidence confirms that women and people of color contribute to group tasks at a lower level, make fewer influence attempts, are less frequently chosen as leaders, and are less committed to group outcomes than White males."); Nancy S. Marder, Note, Gender Dynamics and Jury Deliberations, 96 Yale L.J. 593, 594–98 (1987) (citing studies comparing men’s and women’s participation rates and influence during jury deliberations). Another study observed that in groups in which members had some shared and some unshared information, there was a tendency to discuss the shared information and to seek consensus on that, rather than for members to offer the information that was unique to them. See Garold Stasser & William Titus, Pooling of Unshared Information in Group Decision Making: Biased Information Sampling During Discussion, 48 J. Personality & Soc. Psychol. 1467, 1470 (1985) (describing a situation in which a group would make a better judgment if it pooled the information of its members, but instead, its discussion focused on shared information and the initial preferences of the group members, which were formed without knowledge of the individual members’ unshared information). If group members do express different views, then "[e]xploration of opposing positions can help them gather new data, delve into issues more deeply, and develop a more complete understanding of problems and alternative solutions." Pelled, supra note , at 624.
 . See infra Appendix B (Juror Questionnaire).
 . See infra Appendix C (Lawyer’s Questionnaire).
 . The research assistants consisted of three women and two men. The two men, both of whom are white, were law students in different area law schools in Los Angeles. The three women included one white woman, one Asian-American woman, and one South Asian-American woman. Two of the women were law students in area law schools in Los Angeles, and one was an undergraduate in Los Angeles.
 . It turned out that there were no jury trials in Santa Monica and West Los Angeles during the data collection period.
 . See infra Appendix B.
 . Case complexity was not defined for the attorneys. Although a case can be complex for a variety of reasons, from the sheer volume of information, to the number of interconnections needed to decide what happened, to the inferences that jurors would have to draw, see Geoffrey P. Kramer & Norbert L. Kerr, Laboratory Simulation and Bias in the Study of Juror Behavior: A Methodological Note, 13 Law & Hum. Behav. 89 (1989), attorneys were asked to rate the complexity of the case without being given a definition of the term. See infra Appendix C, question 7 ("Circle a number on the [seven-point] scale below that you think best indicates how complicated this case was.").
 . Background information included the case name, docket number, number of defendants, nature of the charge, attorneys’ names and addresses, verdict, and date on which the verdict was rendered.
 . In criminal cases in California, judges use "confidential juries," in which "names and identifying information are withheld even after a verdict, but the names are shown to the lawyers during jury selection." Jerry Markon, Judges Pushing for More Privacy of Jurors’ Names, Wall St. J., June 27, 2001, at B1. This practice, which has been upheld on appeal, was motivated by concerns for juror safety and privacy. Id.
 . Distributing the questionnaires by mail was not an option because, as noted earlier, juror names and addresses were unavailable. See supra note and accompanying text. In addition, mailed questionnaires tend to have a low response rate. See, e.g., Don A. Dillman, Mail and Telephone Surveys: The Total Design Method (1978); Maria Elena Sanchez, Effects of Questionnaire Design on the Quality of Survey Data, 56 Pub. Opinion Q. 206–17 (1992).
 . There still would have remained the problem of obtaining jurors’ telephone numbers or addresses. See supra note and accompanying text.
 . Data could only be collected during the summer months when research assistants could spend all day waiting for a trial to end. Trials could end at any time, and bailiffs, though generally quite helpful, could only give their best guess as to when deliberations might end, the verdict would be announced, and the jury would be dismissed.
 . Sociologists would say that the Los Angeles juries cannot be seen as representative of juries throughout the rest of the country. Rather, if I had wanted a representative sampling, I would have had to distribute questionnaires throughout the country. Using Los Angeles to stand for the rest of the country would be an example of "convenience sampling," which is not a reliable method for studying juries on a national level. See generally Divided Opportunities: Minorities, Poverty, and Social Policy (Gary D. Sandefur & Marta Tienda eds., 1988).
 . Of the twenty-six juries, sixteen were located in Los Angeles, five were in Santa Ana, three were in Pasadena, and two were in Van Nuys.
 . Of the 138 respondents, three did not answer this question.
 . Of the 138 respondents, one did not answer this question. Percentages exceed 100% due to rounding.
 . This is one skewing effect of studies involving questionnaires: those who respond are more likely to be educated than those who do not respond. This was also found to be the case when jurors are interviewed. See, e.g., Diane L. Bridgeman & David Marlowe, Jury Decision Making: An Empirical Study Based on Actual Felony Trials, 64 J. Applied Psych. 91, 93 (1979) ("The 65 jurors who volunteered to be interviewed . . . tended to be well educated: Only 24 (37%) reported attending only grade or high school, whereas 41 jurors (63%) attended college, and this number included 6 who held graduate degrees.").
 . See B. Beckham & H. Aronson, Selection of Jury Foremen As a Measure of the Social Status of Women, 43 Psychol. Rep. 475, 476–77 (1978) (finding that in 155 juries, women were elected as foreperson one-fifth as often as their numbers would indicate); Fred L. Strodtbeck, Rita M. James & Charles Hawkins, Social Status in Jury Deliberations, 22 Am. Soc. Rev. 713, 715 (1957) ("[O]nly one-fifth as many women were made foreman as would be expected by chance."); Charles Hawkins, Interaction and Coalition Realignments in Consensus-Seeking Groups: A Study of Experimental Jury Deliberations 24 (Aug. 17, 1960) (unpublished doctoral dissertation, University of Chicago) (finding that women constituted 36% of jurors, but only 3% of forepersons). See also Reid Hastie, Steven B. Penrod & Nancy Pennington, Inside the Jury 28 (1983) ("[m]ales, higher classes, and end seating are overrepresented" in the role of foreperson); Rita James Simon, The Jury and the Defense of Insanity 114 (1967) (noting that businessmen had four times better chance of being selected as foreperson than male laborers; housewives were never selected).
 . The percentages add up to more than 100% due to rounding.
 . One study of ten felony cases in Santa Cruz, California found a conviction rate of 82% and an overall Santa Cruz County conviction rate in felony trials of about 90%. See Bridgeman & Marlowe, supra note , at 94.
 . The percentage of male (85%) and female (15%) defendants in this study matches the percentage of male (84%) and female (16%) defendants arrested in the seventy-five largest counties in the U.S. in 1996. See Bureau of Justice Statistics, Dep’t of Justice, Sourcebook of Criminal Justice Statistics—2000, at 460 tbl.5.48 (Kathleen Maguire & Ann L. Pastore eds., 2001). Race is a little harder to compare because the categories used by the Bureau of Justice are "White," "Black," and "Other." Id. It is unclear whether Latinos/as were included under "White" (40%), "Black" (58%), or "Other" (2%) because the U.S. Census Bureau does not consider Latinos/as to be a racial category. Id. ("Without consideration of Hispanic origin, U.S. Census Bureau data for 1996 indicate that the racial distribution . . . .").
 . There was only one measure of deliberation length: jurors’ reports of how long their deliberations lasted. Thus, there was no finding on this item.
 . See supra Part IV.A.
 . See infra Appendix B, questions 23–24.
 . See infra Appendix B, questions 32, 34–36.
 . See infra Appendix B, questions 25, 26(a)–(c).
 . See infra Appendix B, questions 30(a) & (b).
 . See infra Appendix A (providing tables and details of the analyses).
 . See infra Appendix A (describing adjustments made in response to the factor analysis).
 . In retrospect, I should have asked jurors to assess how much "conflict" their deliberations produced, rather than the tone of their deliberations because while there may have been disagreement or conflict, there may not have been hostility. My prediction was that diverse juries would have more conflict, and indeed they might have, but the conflict might not have risen to the level of hostility, which may explain why my finding was contrary to my hypothesis. For a study in which diverse groups produced greater conflict, but the group members saw the conflict as beneficial to the tasks they were to perform when they were members of groups that valued collective efforts, see Jennifer A. Chatman, Jeffrey T. Polzer, Sigal G. Barsade & Margaret A. Neale, Being Different Yet Feeling Similar: The Influence of Demographic Composition and Organizational Culture on Work Processes and Outcomes, 43 Admin. Sci. Q. 749 (1998).
 . See Elizabeth Aries, Interaction Patterns and Themes of Male, Female, and Mixed Groups, 7 Small Group Behav. 7, 13–14 (1976) (noting that discussions in all-male groups were marked by competition and aggression and discussions in all-female groups centered on feelings, affiliation, home, and family); B. G. Reed, Gender Issues in Training Group Leaders, J. Specialists Group Work, Aug. 1981, at 161, 162 (finding that all-male groups were characterized by hierarchy and competition, and all-female groups shared feelings and discussed subjects in greater depth). See also Elizabeth Aries, Male-Female Interpersonal Styles in All Male, All Female and Mixed Groups, in Beyond Sex Roles 292, 294 (Alice G. Sargent ed., 1977) [hereinafter Male-Female Interpersonal Styles] (describing supportive behavior in all-female groups in which a member who missed a session was encouraged to participate, in contrast to all-male groups in which a member who missed a session was excluded from discussion); John E. Baird, Sex Differences in Group Communication: A Review of Relevant Research, 62 Q.J. Speech 179, 189–90 (1976) (observing that in all-male groups, the weakest member is excluded; in all-female groups, any member in danger of neglect is given encouragement).
 . See Peter W. Hahn & Susan D. Clayton, The Effects of Attorney Presentation Style, Attorney Gender, and Juror Gender on Juror Decisions, 20 Law & Hum. Behav. 533, 537 (1996) (referring to other studies "showing that men dominate in mixed-gender conversations"); Male-Female Interpersonal Styles, supra note , at 297 (finding that in mixed groups, women spoke less, initiating only 34% of total interactions); Reed, supra note , at 163 (observing that in mixed groups, women spoke less, spoke primarily to men, shared less personal information, and were less involved in topics and tasks associated with masculinity).
 . See Deborah Tannen, You Just Don’t Understand Me: Women and Men in Conversation (1990) (describing men’s tendencies to offer information, to interrupt, and to compete verbally and women’s tendencies to listen, to encourage, and to include other speakers in the conversation); Hahn & Clayton, supra note , at 537 ("Other studies have found general distinguishing characteristics between males’ and females’ speech. Compared with men, women are less likely to interrupt, and less successful when they do interrupt, while they are more likely to hedge, ask tag questions, and use disclaimers and intensifiers.") (citations omitted); Nancy M. Henley, Power, Sex, and Nonverbal Communication, in Language and Sex 184 (Barrie Thorne & Nancy M. Henley eds., 1975) (describing how nonverbal acts, such as gesture, movement, touch, and gaze, can contribute to the speaker’s power); Candace West & Don H. Zimmerman, Small Insults: A Study of Interruptions in Cross-Sex Conversations Between Unacquainted Persons, in Language, Gender and Society 107 (Barrie Thorne et al. eds., 1983) (finding that in cross-sex conversations, men are three times more likely than women to interrupt the other speaker).
 . See, e.g., Joyce Cohen, He-Mails, She-Mails: Where Sender Meets Gender, N.Y. Times, May 17, 2001, at D1 (describing research in which "‘men tend to make strong assertions,’ disagree with others and use profanity, insults and sarcasm [in e-mail exchanges]. By contrast, women tend to use mitigated assertions along with questions, offers, suggestions and polite expressions . . . . They are supportive and agreeable [in e-mail exchanges] . . . .") (quoting Susan C. Herring, Associate Professor of Information Science and Linguistics at Indiana University at Bloomington).
 . See Carol Gilligan, In a Different Voice 100 (1982) (contrasting the way in which men understand the "moral imperative . . . as an injunction to respect the rights of others and thus to protect from interference the rights to life and self-fulfillment" with the way in which women view the moral imperative as "an injunction to care, a responsibility to discern and alleviate the ‘real and recognizable trouble’ of this world").
 . See, e.g., Female Jurors Complain of Sexism Tainting Menendez Deliberations, N.Y. Times, Jan. 31, 1994, at A7; Mary B.W. Tabor, Stereotyping Men, Women and Juries by Trial and Error, N.Y. Times, Feb. 6, 1994, at 3.
 . See, e.g., Rosabeth Moss Kanter, Women and the Structure of Organizations: Explorations in Theory and Behavior, in Another Voice 34, 56–60 (Marcia Millman & Rosabeth Moss Kanter eds., 1975).
 . See Rosabeth Moss Kanter, Men and Women of the Corporation 208 (1977) (suggesting that women in work settings may assume traditional roles because they are outnumbered by men and that male-female interactions will change when both sexes are equally represented in the workplace). See also Elsass & Graves, supra note , at 949 ("[I]f the representation of any gender or racioethnic group is small relative to the total group size . . . , members’ categorical identities are likely to be highly salient.") (citations omitted).
 . See Kanter, supra note , at 208. See also Elsass & Graves, supra note , at 948 ("Although research on group composition effects provides conflicting information about how the exact sizes of the numerical minority and majority affect individuals’ experiences in diverse groups, it does suggest that women and people of color especially are likely to encounter negative dynamics when they do not make up the majority of a group.") (citations omitted).
 . It would be interesting to know at what point the dynamics change. If juries of six men and six women produce harmonious deliberations, what do the numbers of men and women have to be before the deliberations become hostile? Is there a "tipping point?"
 . U.S. Dep’t of Labor, Employment & Earnings, July 2001, at 15 (showing that in the period June–Dec. 2000, men constituted 54% of those employed and women constituted 46% of those employed); Current Labor Statistics, Monthly Lab. Rev., June 2001, at 62 (providing numbers that indicate that for the year 2000, 54% of those employed were men and 46% of those employed were women).
 . Almost forty-five years ago, Fred Strodtbeck and Richard Mann observed that in mock juries consisting of men and women, women tended to be concerned with the socio-emotional dynamics of the group, similar to their role in the family. Fred L. Strodtbeck & Richard D. Mann, Sex Role Differentiation in Jury Deliberations, 19 Sociometry 3, 5 (1956). In contrast, men played a more active role on the jury and focused on accomplishing the task set out for them, consistent with their role in the family. See id. Of course, Strodtbeck and Mann were doing their empirical work at a time when men were breadwinners and women were homemakers. Although the divisions between men’s and women’s roles in the workplace and home are no longer so clear-cut, they have not been outgrown altogether. For a more recent study of mixed groups in which men engaged in task behavior and women engaged in social behavior, see Wendy Wood & Stephen J. Karten, Sex Differences in Interaction Style As a Product of Perceived Sex Differences in Competence, 50 J. Personality & Soc. Psychol. 341 (1986). For an illustration of such behavior in the jury room, see Frontline: Inside the Jury Room (WGBH television broadcast, Apr. 8, 1986) (Transcript of broadcast at 24–25). After the jury had spent two hours in deliberation, and had succeeded in convincing the lone holdout to capitulate and to vote to acquit the defendant, the response of several male jurors was to call for a vote, whereas the response of several female jurors was to ask the holdout if he felt comfortable with his change of vote.
 . See, e.g., Barry Ruback, The Sexually Integrated Prison: A Legal and Policy Evaluation, in Coed Prison 33, 44 (John Ortiz Smykla ed., 1980) ("Based on their personal observations, residents at FCI Fort Worth [a sexually integrated prison] opined that physical violence between the men was less common than in other prisons. From the women’s standpoint, the favorable sexual ratio . . . and the resulting attention, minimizes hostilities among women."); Rosemary Herbert, Note, Women’s Prisons: An Equal Protection Evaluation, 94 Yale L.J. 1182, 1184-85 (1985) ("The limited data available suggest that these experiments [in "‘co-correctional’" prisons] have been successful in the terms traditionally used to evaluate correctional programs . . . .") (footnote omitted); Coed Incarceration, Time, Sept. 16, 1974, at 84 ("Though it is too early to judge conclusively, penal experts believe that coed incarceration is a success—at least in improving behavior."); Aric Press, Ron LaBrecque & Peggy Clausen, When Prisons Go Coed, Newsweek, Jan. 11, 1982, at 66 ("‘There is a gentling effect which seems to occur’" when men and women are in prison together.) (quoting Charles Campbell, Alaska’s Commissioner of Corrections); id. ("There tends to be much less violence and more desire to participate in work-and[-]education programs . . . .").
 . Hahn & Clayton, supra note , at 550.
 . See Fred L. Strodtbeck & Richard M. Lipinski, Women Jurors, Then and Now 2 (1988) (unpublished paper on file with author). I thank Reid Hastie for bringing this study to my attention.
 . See id. at 22–25.
 . See id. at 4–7.
 . See id. at 3–4.
 . Juries were categorized as "majority-male" or "majority-female" juries if at least eight of the twelve jurors were male or female respectively. If there were not at least eight jurors of one gender, the jury was categorized as having no gender majority ("no majority"). In this study, there were no majority-male juries, six majority-female juries, and twenty no-majority juries.
 . Of the twenty-six juries, sixteen were racially mixed ("no majority"), four had at least eight white jurors ("largely white"), and six had at least eight minority jurors ("largely minority"). Although the small number of largely-white and largely-minority juries made it difficult to detect any significant effects, it does mean that most cases in this study were heard by racially diverse juries.
 . Strodtbeck and Lipinski, who used data from jury experiments done in 1956 in Chicago and 1978 in Boston, to see if there were changes in male and female jurors’ styles, such as seating choice, participation rates, and persuasiveness, found that women in the 1956 Chicago study "reported significantly greater satisfaction with their jury service than men," Strodtbeck & Lipinski, supra note at 26, even though women’s participation rate was only two-thirds that of the men. See id. at 18, 26.
 . Of course, there is always the problem with self-reporting of how accurate and candid jurors are able and willing to be. However, this is a problem inherent in juror questionnaires, which necessarily depend on self-reporting rather than on an outside evaluator’s observations or assessment.
 . For an earlier study finding a connection between participation and satisfaction, see Strodtbeck et al., supra note , at 716 (finding that the level of an individual’s satisfaction with his or her jury duty was positively correlated with the level of his or her participation).
 . One study of employees in a state agency and two Fortune 100 companies found that men’s and women’s satisfaction with their work situation was affected by the gender composition of their workplace, with men feeling less satisfaction as their workplace became more gender diverse (and therefore less homogeneous), whereas women did not experience less satisfaction when their workplace became more gender diverse. See Anne S. Tsui, Terri D. Egan & Charles A. O’Reilly III, Being Different: Relational Demography and Organizational Attachment, 37 Admin. Sci. Q. 549, 569 (1992). This same study also found a similar effect for race, namely that whites became less satisfied with their workplace as the workplace became more racially diverse. Id. The study suggested that there was support for these findings in earlier work, see Amy S. Wharton & James N. Baron, So Happy Together? The Impact of Gender Segregation on Men at Work, 52 Am. Soc. Rev. 574 (1987) (finding that men in mixed-gender work settings reported significantly lower job satisfaction and self-esteem and more job-related depression than men in either largely male or female settings), but that there were also contrary hypotheses. For example, Rosabeth Moss Kanter observed:
Blocked opportunity, powerlessness, and tokenism tend to generate employees who, among other things, have low aspirations, lack commitment to the organization, become hostile to leaders, behave ineffectively in leadership roles themselves, take few risks, or become socially isolated and personally stressed. Aside from the cost to such individuals—often women, but also men—organizations are wasting a large measure of their human talent.
Kanter, supra note , at 266. She proposed "number-balancing" as one antidote to the deleterious effects of tokenism because "[o]rganizations with a better balance of people would be more tolerant of the differences among them." Id. at 283.
 . Throughout the discussion of thoroughness, I will be referring to jurors’ beliefs about their jury’s thoroughness. Jurors rated how thorough they thought their deliberations were; thus, there are only jurors’ perceptions of their thoroughness rather than any external measure.
 . See Harrington, supra note , at 1 ("The findings suggest that demographic diversity is most likely to enhance group performance when organizations encourage members to cohere around their shared commitment to a task rather than social attraction."). See also Donna Bozzo, Taking Stock of Investment Clubs: Teamwork May Bring Many Happy Returns, Chi. Trib., Dec. 8, 1999, at 6 ("[A]nother recent study at Brown University indicated that mixed-gender clubs fare better—earning about 2 percent more than same-sex clubs."); Richard Teitelbaum, Mars and Venus Do Better Together, N.Y. Times, Sept. 5, 1999, at 8 ("A new study suggests that when it comes to investment clubs, those that include members of both sexes outperform those of the single-gender variety.").
 . Harrington uses the phrase "mixed-gender" whereas I use the phrase "gender diverse." In discussing her study, I will use her terminology.
 . See Harrington, supra note , at 12, 17–18.
 . See id. at 17–18.
 . Id. at 21.
 . Id. at 22.
 . Harrington described task orientation as "the dimension of group cohesion in which groups prioritize shared commitment to the task" and social orientation as "the dimension of group cohesion in which groups prioritize similarity and consensus." Id. at 3. These terms were originated by Robert Bales in 1953. See Robert F. Bales, The Equilibrium Problem in Small Groups, in Working Papers in the Theory of Action 111 (Talcott Parsons et al. eds., 1953).
 . Harrington, supra note , at 28.
 . In another study, business students were assigned to small groups to perform various tasks typical of a business organization and were told that their organization valued either individualism or collective effort. In groups that were diverse by demographic characteristics (nationality, race, and gender) and were encouraged to work collectively, members described their solutions as more creative than members in groups that were homogeneous and encouraged to work individually. See Chatman et al., supra note , at 749. According to the authors of this study, diversity led to a greater array of ideas for the group to consider and the emphasis on collective work encouraged members to share their different ideas with each other. See id. at 777 ("[C]reativity was enhanced more in heterogeneous organizations when the organization emphasized collectivism . . . .").
 . Irving L. Janis, Groupthink 7, 262, 270–71 (2d ed. 1982) (labelling as "groupthink" the situation in which group members conform to the prevalent view and effectively limit the range of ideas expressed and considered by the group).
 . Geoffrey Colvin, The 50 Best Companies for Asians, Blacks, and Hispanics, Fortune, July 19, 1999, at 52.
 . Id. ("More impressively, these companies as a group have performed terrifically, about matching the S&P 500 over the past year and beating it over the past three and five years.").
 . Id. at 54.
 . Id. (quoting Ivan Seidenberg).
 . Although the observations of CEOs in minority-friendly companies were anecdotal, the increases in their companies’ stock prices as a group were not merely anecdotal.
 . Another study found this effect with respect to technology-based companies whose founding members were diverse according to industry experience. See Kathleen M. Eisenhardt & Claudia Bird Schoonhoven, Organizational Growth: Linking Founding Team, Strategy, Environment, and Growth Among U.S. Semiconductor Ventures, 1978–1988, 35 Admin. Sci. Q. 504 (1990). This study hypothesized:
Teams with individuals who have entered the industry at different times are likely to have different points of view about technology, competitive tactics, and so forth. People with long experience in the industry bring a knowledge of how the industry operates. Those with less experience bring freshness in perspective. These different points of view encourage conflict, which, in turn, counteracts the danger that the team reaches premature closure or has an insufficient airing of alternatives. Combining conflicting views may yield innovative and yet viable ways to compete, giving competitive advantage to the young firm.
Id. at 510. After evaluating semiconductor firms founded between 1978–85 in the U.S. according to certain criteria, the researchers found that "the size of the team, members’ past experience together, and members’ heterogeneity in industry experience are linked with higher growth." Id. at 524.
 . According to one study, which tested the effect of attorneys’ gender and presentation style (aggressive or passive) on jurors, "[m]ale jurors found the defendant significantly more guilty when the attorney was aggressive, but female jurors found the defendant just as guilty when the attorney was aggressive as when the attorney was passive." Hahn & Clayton, supra note , at 548. Thus, "[w]hile men are clearly influenced by the attorney’s speech style, women may consider the evidence in the trial more important than the style of the attorney . . . ." Id. Although this was a mock jury study, which did not replicate the courtroom experience in a number of ways, its finding that women may focus on the evidence while men may give more weight to the attorney’s style of speech, suggests one way in which male and female jurors might view the evidence differently.
 . See Marder, supra note , at 1070–72 & n.120 (describing studies finding that men tend to overestimate eyewitnesses’ ability to identify a suspect more than women).
 . See, e.g., Susan Glaspell, A Jury of Her Peers, reprinted in The Best Short Stories of 1917, at 256–82 (Edward J. O’Brien ed., 1918) (providing a fictional account of how men and women viewed facts differently based on the separate spheres they occupied).
 . In a study testing whether an individual’s demographic dissimilarity related to his or her perception of emotional conflict within the group, there were positive relationships with gender dissimilarity and tenure dissimilarity, but not with race dissimilarity. See Lisa Hope Pelled, Relational Demography and Perceptions of Group Conflict and Performance: A Field Investigation, 7 Int’l J. Conflict Mgmt. 230 (1996). The author explained that the gender and tenure dissimilar individual might feel different and therefore "experience the group as more conflict-ridden regardless of the actual level of intragroup conflict" or that "a demographically distinct group member makes all (or most) group members more uncomfortable . . . fostering emotional conflict." Id. at 241. She suggested that race dissimilarity might not have had the same effect in a group in which there was one Asian, one Hispanic, one African American and one Caucasian, for example, because in such a group "‘being different’" was a common condition among the group members. Id. This explanation might have application to this jury study as well.
 . One jury study, for example, found a connection between jurors’ race and jurors’ liability findings and damage awards in that African Americans tended to be more pro-plaintiff in their liability determinations and to award plaintiffs higher damages than did Caucasians and Hispanics. See Chris F. Denove & Edward J. Imwinkelried, Jury Selection: An Empirical Investigation of Demographic Bias, 19 Am. J. Trial Advoc. 285 (1995). Although the authors cautioned that litigators should not rely on one demographic characteristic in searching for favorable jurors because that characteristic could be favorable in one fact situation and not in another, they did find race to be "the single most important factor in predicting juror orientation." Id. at 293. They found this to be the case in terms of liability findings, see id., and damage awards. See id. at 313.
In an early mock jury study involving a rape case in which the race of the defendant was either white or black and the racial composition of the jury was varied (one jury was all-black; one jury was three-fourths black and one-fourth white; one jury was one-half black and one-half white; one jury was three-fourths white and one-fourth black; one jury was all-white), researchers did find a racial effect: "[W]hite jurors who found the Black defendant ‘guilty’ on their first ballot tended to hold to this decision and not be influenced by group discussion [and] Blacks as a whole were much more likely than whites to reach a ‘not guilty’ verdict, regardless of the race of the defendant." J.L. Bernard, Interaction Between the Race of the Defendant and That of Jurors in Determining Verdicts, 5 Law & Psychol. Rev. 103, 109 (1979).
 . For a study finding a statistical relationship between the racial diversity of the jury and the likelihood of reaching a verdict, with greater jury diversity leading to more hung juries when the defendant was African-American, see Kenneth S. Klein & Theodore D. Klastorin, Do Diverse Juries Aid or Impede Justice? 1999 Wis. L. Rev. 553, 564. This same study, however, found no statistical relationship between gender diverse juries and the likelihood that a jury would reach a verdict. See id.
 . See infra Appendix C, question 7.
 . See supra text accompanying notes 123-24.
 . See Powers v. Ohio, 499 U.S. 400, 402 (1991) (holding that a white defendant can challenge the prosecutor’s use of peremptory challenges to exclude prospective jurors who are African American).
 . Ballard v. United States, 329 U.S. 187, 193–94 (1946).
 . Peters v. Kiff, 407 U.S. 493 (1972).
 . Ballard, 329 U.S. at 193-94.
 . According to this view, the gender, race, or age of the jurors does not affect how they see the case or the way they deliberate or perceive the deliberations. See supra Part II.A (describing the reasonable person view).
 . One way to create venires that are drawn broadly from the community is to summon prospective jurors from multiple lists, rather than from only the voter registration list that has been relied on at least in the past. The theory is that by drawing prospective jurors from an array of lists, such as tax rolls, drivers’ licenses, utility bills, and unemployment compensation, outreach will be much greater than if only the voter registration list is used. See, e.g., Dennis Bilecki, Program Improves Minority Group Representation on Federal Juries, 77 Judicature 221 (1994) (describing a pilot program in northern California that used lists of licensed drivers and California identification card holders in addition to voter registration lists to improve minority representation on the venire); David Kairys, Joseph B. Kadane & John P. Lehoczky, Jury Representativeness: A Mandate for Multiple Source Lists, 65 Cal. L. Rev. 776 (1977) (recommending the use of multiple source lists for the venire, rather than simply relying on voter registration lists); G. Thomas Munsterman & Paula L. Hannaford, Reshaping the Bedrock of Democracy: American Jury Reform During the Last 30 Years, Judges’ J., Fall 1997, at 6 ("A widely used technique was to supplement voter registration lists with lists of licensed drivers. More recently, states have added unemployment compensation . . . recipients, welfare recipients, and state and local income tax filers as supplemental source lists.") (citation omitted).
 . See, e.g., David Margolick, Question for the ‘90s: Just What Is a Jury of One’s Peers?, Chi. Daily L. Bull., Feb. 18, 1992, at 2 (describing Georgia’s practice in death penalty cases of requiring venires that resemble the racial and gender composition of the counties from which the jurors are drawn). The Eastern District of Michigan also attempted this, see Nancy J. King & G. Thomas Munsterman, Stratefied Juror Selection: Cross-Section by Design, 79 Judicature 273, 275 (1996) ("The court . . . randomly strikes from the list of persons qualified the specific number of ‘white and other’ potential jurors needed to obtain a qualified list with racial demographics identical to that of the population."), but the Sixth Circuit held that it violated 28 U.S.C. § 1862 and the equal protection component of the Fifth Amendment. See United States v. Ovalle, 136 F.3d 1092, 1109 (6th Cir. 1998).
 . Such an inquiry would have been difficult in this study because most of the verdicts were convictions in whole or in part (20) with few acquittals (4) and hung juries (2). See supra text accompanying note .
 . See Marder, supra note , at 1104–07 & nn.270–80.
 . See id. at 1104–05 & nn.270–75.
 . This was Barbara Babcock’s position in an early article in which she urged that voir dire be extended so that poor litigants have access to information about prospective jurors and can exercise their peremptories based on that information. See Babcock, supra note , at 546.
 . See, e.g., David L. Hamilton, Steven J. Sherman & Catherine M. Ruvalo, Stereotype-Based Expectancies: Effects on Information Processing and Social Behavior, 46 J. Soc. Issues 35, 43–44 (1990) ("Stereotypic expectancies are influential to the extent that information-processing and judgments are based on the group membership information rather than on the individuating information. The implication, then, is that these effects can be reduced if greater importance is attached to the information pertaining specifically to the target person."). But see Galen V. Bodenhausen & Meryl Lichtenstein, Social Stereotypes and Information-Processing Strategies: The Impact of Task Complexity, 52 J. Personality & Soc. Psychol. 871, 879 (1987) ("[I]t is clearly premature to claim that social stereotypes become impotent in the presence of individuating information. Rather, it appears that the nature of the judgment task must be considered in understanding the impact that stereotypes will exert.").
 . See, e.g., Batson v. Kentucky, 476 U.S. 79, 107–08 (1986) (Marshall, J., concurring); Minetos v. City Univ., 925 F. Supp. 177, 185 (S.D.N.Y. 1996); Jeffrey Abramson, We, the Jury 137–39 (1994); Morris B. Hoffman, Peremptory Challenges Should Be Abolished: A Trial Judge’s Perspective, 64 U. Chi. L. Rev. 809, 810, 850 (1997); Marder, supra note , at 1044–47, 1052–86, 1095–99.
 . See, e.g., Jean Montoya, The Future of the Post-Batson Peremptory Challenge: Voir Dire by Questionnaire and the "Blind" Peremptory, 29 U. Mich. J.L. Reform 981, 1011 & n.144 (1996) (citing scholars and commentators who have made this proposal).
 . Such use of peremptories would be contrary to J.E.B. and in violation of the Equal Protection Clause. See J.E.B. v. Alabama ex rel. T.B., 511 U.S. 127, 129 (1994). Since J.E.B., the Court has retrenched slightly in Purkett v. Elem, 514 U.S. 765 (1995) (per curiam) (allowing reasons for the exercise of the peremptory challenge to be unrelated to the facts of the case as long as they are nondiscriminatory). Even if Elem had not given attorneys more leeway for avoiding Batson and its progeny, it was clear that many attorneys had managed to do so before Elem. See supra note (noting sources that indicated that Batson was failing, in practice, to eliminate discriminatory peremptories).
 . See United States v. Thomas, 116 F.3d 606, 608–09 (2d Cir. 1997) (holding that the trial judge, after being notified by the jury, should interview any juror urging nullification to determine if that was the juror’s intent, and if so, to remove that juror from the jury, even if the jury was already in the midst of its deliberations). In my view, this is an unfortunate development. See Marder, supra note , at 947–52 (describing the ways in which the Second Circuit erred in Thomas).
 . See, e.g., Am. Judicature Soc’y, Behind Closed Doors: A Guide for Jury Deliberations (1999) (providing general information about jury deliberations).
 . This view of jurors explains why some courts are still reluctant to allow jurors to take notes and to submit written questions, even though the benefits of both are apparent to any educator. Many interested in jury reform have recommended these changes. See, e.g., ABA/Brookings Inst., Charting a Future for the Civil Jury System 18–19 (1992) (recommending notetaking); id. at 20 (recommending the submission of written questions to the judge); Ariz. Sup. Ct. Comm. on More Effective Use of Juries, Jurors: The Power of 12 (1994) (including a list of recommendations and a proposed bill of rights); Kassin & Wrightsman, supra note , at 128–29 (considering why there is so much resistance to allowing jurors to take notes); B. Michael Dann, "Learning Lessons" and "Speaking Rights": Creating Educated and Democratic Juries, 68 Indiana L.J. 1229 (1993) (describing two models of jurors, active and passive, and arguing that current practices encourage passive jurors and proposing reforms that would foster active jurors); B. Michael Dann & George Logan III, Jury Reform: The Arizona Experience, 79 Judicature 280 (1994) (describing some of Arizona’s reforms to its jury system, including note taking and juror questions); Harold J. Bursztajn, Linda Stout Saunders & Archie Brodsky, Keeping a Jury Involved During a Long Trial, Crim. Just., Winter 1997, at 8 (recommending that jurors be permitted to ask questions because it keeps them actively engaged in the trial).
 . Another study of actual jurors, which used a questionnaire and interviews, found that jurors "expressed a strong need for closure to their jury experience" and "some of the jurors acknowledged that participating in the study brought about such closure." Bridgeman & Marlowe, supra note , at 97. Admittedly, those of us who conduct research based on jurors’ views are likely to believe that not only is this a useful method for gaining insight into the jury experience, but also that we might contribute to the jury experience in some positive way.
 . For example, Judges Schwartz and Chotiner in L.A. Municipal Court have distributed questionnaires to jurors for years.
 . For example, Judge Dann in Arizona and Judge Mize in Washington, D.C. have implemented jury reforms in their own courtrooms and have advocated for and written about jury reform on a state (or district) level. See, e.g., Dann, supra note , at 1247–79 (recommending changes in jury practice that would lead to greater juror participation); Dann & Logan, supra note , at 280 (describing Arizona’s jury reforms); Mize, supra note , at 10 (describing voir dire reforms that he implemented in his courtroom).
 . Judge Dann described the need for judges to seek feedback from jurors during the trial. See Dann, supra note , at 1243. Although he did not mention a questionnaire after the trial, it would be yet another means of enabling jurors to be "active" and of providing courts and judges with feedback as to their experiences. See id.
 . See supra Part I. See also Marder, supra note , at 907–26 (describing more fully the broad roles of the modern jury).
 . See de Tocqueville, supra note , at 275.
 . Future studies could expand upon this study in a variety of ways: they could focus on other geographical areas to study the effects of diversity on deliberations and compare them to the findings of this study, which was based on Los Angeles juries; they could take one particular finding, such as juror satisfaction, and explore it more fully, particularly in connection to participation; and they could focus on judicial resistance to the vital tool of jury questionnaires and examine why this is so and how it might be overcome.
 . See infra Appendix A, tbl.2.
 . Hostility scores were coded so that higher scores indicate more hostility and lower scores indicate less hostility (and recall that lower diversity scores indicate greater diversity).
 . Satisfaction ratings were coded so that higher scores indicated higher satisfaction and lower scores indicated lower satisfaction (and recall that lower diversity scores indicate greater diversity).
 . The index consists of the first three measures of satisfaction.
 . The significant model statistic is driven by a significant covariate, case complexity, not a significant predictor variable, race diversity.
 . As above, the significant model includes a significant covariate (case complexity), not a significant predictor (average diversity).
 . See infra Appendix A, tbl.4.
. Thoroughness scores were coded so that higher scores indicate more thoroughness and lower scores indicate less thoroughness (and recall that lower diversity scores indicate greater diversity).