Full opinion text
Argued Nov. 19, 1991. Before: MANSMANN, COWEN and ROTH, Circuit Judges. Reargued Sept. 10, 1992. Before, SLOYITER, Chief Judge, and STAPLETON, MANSMANN, GREENBERG, HUTCHINSON, SCIRICA, COWEN, NYGAARD, ALITO, and ROTH, Circuit Judges. OPINION OF THE COURT ROTH, Circuit Judge. During the selection of the grand jury which indicted appellant, Thomas Ramseur, the assignment judge, through statements and actions, treated certain African-American members of the venire differently because of their race. There is, however, nothing in the record to indicate any actual exclusion from appellant’s grand jury of African-American jurors on account of their race. This appeal requires us to address the difficult question of whether these events comprise a constitutional violation. We must also determine whether the grand and petit juries that tried and convicted the appellant were drawn from lists that unconstitutionally underrepresented African-Americans and whether the procedures used in Essex County, New Jersey, to select grand jury forepersons violated the Equal Protection Clause of the Fourteenth Amendment or the Sixth Amendment’s guarantee of a trial by a jury drawn from a cross-section of the community. Finally, we must determine whether misconduct by the prosecutor in this case denied appellant his constitutional right to a fair trial. Appellant, Thomas Ramseur, has advanced these grounds in his petition for a writ of habeas corpus. The district court denied his petition. For the reasons that follow, we will affirm the denial of the petition but will do so for the reasons we state below. I. On May 12, 1983, a jury found Thomas Ramseur guilty of the murder of Asaline Stokes, his former girlfriend. He was convicted on all counts charged in the indictment against him: (1) murder (N.J.S.A. 2C:11 — 3); (2) unlawful possession of a knife under circumstances not manifestly appropriate for lawful use (N.J.S.A. 2C:39-5d); and (3) unlawful possession of a knife with the purpose of using it against another (N.J.S.A. 2C:39-4d). Following the sentencing phase of the bifurcated trial, the jury rendered a sentence of death that was imposed by the trial court on June 17,1983. On March 5, 1987, the New Jersey Supreme Court affirmed Ramseur’s convictions on all counts but reversed his death sentence. See State v. Ramseur, 106 N.J. 123, 524 A.2d 188 (1987). Ramseur then filed a petition for a writ of habeas corpus under 28 U.S.C. § 2254 in the U.S. District Court for the District of New Jersey. On March 14, 1990, the district court denied Ramseur’s petition for a writ of habeas corpus. This appeal followed. We have jurisdiction over this appeal pursuant to 28 U.S.C. §§ 1291, 2253. Our scope of review of a district court’s conclusions of law with regard to a state prisoner’s petition for a writ of habeas corpus is plenary. See Humanik v. Beyer, 871 F.2d 432, 435 (3d Cir.1989), cert. denied, 493 U.S. 812, 110 S.Ct. 57, 107 L.Ed.2d 25 (1989). II. The crux of Ramseur’s complaint is that the procedure used to empanel grand jurors in Essex County violated his right to equal protection of the laws under the Fourteenth Amendment. Resolving this issue requires a close examination of the procedure used to empanel grand juries in the county prior to and at the time of Ramseur’s indictment. We will describe in Part III how the juror source lists were created in Essex County, panels of qualified jurors selected from those lists, and summonses sent to persons randomly designated for the grand jury panels. Many potential jurors, upon receipt of the summons to jury service and the realization that grand jury service would last for a six week period, would submit a written request for an excuse. The clerk’s office would screen these excuses and meritorious requests would be granted before the jurors actually appeared for service. Prior to the assembly of the venire, the assignment judges for Essex County would review the letters and questionnaires received. At the actual selection, the assignment judges would question each juror and reconsider the excuses previously requested but denied. The assignment judge who empaneled Ramseur’s grand jury used the following procedure once the venire was assembled: First, the judge briefly interviewed each juror. Then he did one of three things: excused the juror for cause, asked the juror to take a seat in the body of the courtroom for the time being, or asked the juror to take a seat on the panel. Those he asked to sit in the body of the courtroom fell into one of two groups. The first group consisted of those persons who had asked to be excused but whose excuses had been denied. They were asked to sit in the body of the courtroom with the understanding that they might be called upon to serve later, after all of the other prospective jurors had been questioned. The second group consisted of persons who proffered no excuse, stated they were willing to serve, but were nonetheless asked to sit aside. In the course of empaneling Ramseur’s grand jury, the judge announced that he was attempting to “pick a cross section of the community” to serve on the grand jury. App. at 2429. Later, he asked Esther Cata-gen, app. at 2447, and George Smith, app. at 2438, to sit in the body of the courtroom although both had stated they were willing to serve. When the assignment judge reached Betty Patrick, the forty-third prospective juror in the selection of Ramseur’s grand jury, Ms. Patrick indicated that she was willing to serve. However, the judge asked her to take a seat in the body of the courtroom and stated: I don’t mind telling you, ladies and gentlemen of the jury or the panel of the grand jury, I am trying to get a cross-section; and as you’ve probably noticed, I have asked two of the blacks who have indicated a willingness to serve to sit in the body of the courtroom. I am deliberately trying to get an even mix of people from background and races, and things like that. And if any of you think that I am in any way being sneaky about it, please understand that I am not. I am telling you like it is, and that is the reason I have done what I have done. App. at 2449-2450. Following this announcement, the judge asked two other panel members who expressed a willingness to serve, Francena Hardwick, app. at 2453, and Orro Ikena, app. at 2454, to sit in the body of the courtroom. After panel members one through twenty-two had been selected, but prior to voir dire, the assignment judge asked Ms. Cata-gen to come up from the body of the courtroom and take seat number twenty-three. The judge then embarked on voir dire of the jurors assembled. One of the jurors explained that she was “prejudiced against certain people, certain races.” The judge excused that juror, explaining, “I appreciate your honesty, and we don’t want people like you to serve on the Grand Jury.” App. at 2464. That juror was replaced by Orro Ikena. At that point Mr. Ikena proffered an excuse which the court accepted. Ms. Patrick was then asked to take the seat to which Mr. Ikena had been assigned. Ultimately, George Smith, the fifth juror who had expressed a willingness to serve but was initially asked to sit in the body of the courtroom, also found his way onto the panel although the transcript does not reflect how this came about. Ms. Hardwick, who had expressed a willingness to serve but was nonetheless asked to sit in the body of the courtroom, was never chosen as a grand juror. The record does not clearly establish the race of Mr. Ikena, Mr. Smith, or Ms. Hardwick. N.J.S.A. 2A:73-1 provides: If from any grand jury panel, more persons remain available for service, after excuses have been allowed, than are necessary to constitute the grand jury, the persons whose names are first drawn and not excused, not to exceed twenty-three in number, shall constitute the grand jury. The random selection procedure set forth in the New Jersey statute was routinely ignored by the assignment judges of Essex County at the time Ramseur's grand jury was chosen. At a hearing regarding Ramseur’s allegations in 1983, one assignment judge stated that rather than selecting jurors randomly “it may be on the basis of my observation of them that I feel that they should not be selected.” App. at 3883-3884. Asked if his choices were made as a result of his “discretionary judgment about each person,” the judge responded, “There is no question about that.” Id. The record in this case contains many similar statements that show that Essex County assignment judges used subjective criteria to select grand jurors and often considered race, “a racial balance” or a “cross section” of black and white jurors when assembling grand juries. Based upon the protections of the Equal Protection Clause of the Fourteenth Amendment, the United States Supreme Court “gradually has abolished race as a consideration for jury service.” Georgia v. McCollum, — U.S. -, -, 112 S.Ct. 2348, 2352, 120 L.Ed.2d 33 (1992). “[R]a-cial discrimination in the qualification or selection of jurors offends the dignity of persons and the integrity of the courts.” Powers v. Ohio, — U.S. -, -, 111 S.Ct. 1364, 1366, 113 L.Ed.2d 411 (1991). Discrimination on the basis of race in the selection of grand jurors is unacceptable and “ ‘strikes at the fundamental values of our judicial system and our society as a whole.’ ” Vasquez v. Hillery, 474 U.S. 254, 262, 106 S.Ct. 617, 622, 88 L.Ed.2d 598 (1986) (quoting Rose v. Mitchell, 443 U.S. 545, 556, 99 S.Ct. 2993, 3000, 61 L.Ed.2d 739 (1979)). Discrimination in the jury selection process harms the defendant, prospective and actual jurors, and the community as a whole. The defendant has an “interest in neutral jury selection procedures ... because racial discrimination in the selection of jurors casts doubt on the integrity of the judicial process, and places the fairness of a criminal proceeding in doubt.” Powers, — U.S. at-, 111 S.Ct. at 1371. Jurors have the right to be unmarred by public discrimination in the justice system. In Edmonson v. Leesville Concrete Co., — U.S. -, -, 111 S.Ct. 2077, 2087, 114 L.Ed.2d 660 (1991), the Court explained that the harm of discriminatory peremptory challenges includes the danger that “persons could be required by summons to be put at risk of open and public discrimination as a condition of their participation in the justice system.” Discrimination in the jury selection process undermines the justice system, and, thereby, the whole of our society. [T]he injury caused by the discrimination [in the jury selection process] is made more severe because the government permits it to occur within the courthouse itself. Few places are a more real expression of the constitutional authority of the government than a courtroom, where the law itself unfolds. Within the courtroom, the government invokes its laws to determine the rights of those who stand before it. In full view of the public, litigants press their cases, witnesses give testimony, juries render verdicts, and judges act with the utmost care to ensure that justice is done. Race discrimination within the courtroom raises serious questions as to the fairness of the proceedings conducted there. Racial bias mars the integrity of the judicial system and prevents the idea of democratic government from becoming a reality. Edmonson, — U.S. at-, 111 S.Ct. at 2077. The Court has ruled that a determination of racial discrimination in the selection of grand jurors will support the quashing of a resulting indictment and reversal of the conviction. Intentional discrimination in the selection of grand jurors is a grave constitutional trespass, possible only under color of state authority, and wholly within the power of the state to prevent. Thus, the remedy we have embraced for over a century — the only effective remedy for this violation — is not disproportionate to the evil that it seeks to deter. If grand jury discrimination becomes a thing of the past, no conviction will ever again be lost on account of it. Vasquez, 474 U.S. at 262, 106 S.Ct. at 623. In order to establish an equal protection violation, a party must show that there has been some actual “purposeful discrimination” in the jury selection process. See Batson v. Kentucky, 476 U.S. 79, 96, 106 S.Ct. 1712, 1722, 90 L.Ed.2d 69 (1986). The crucial question at issue in the present case is whether such purposeful discrimination has been demonstrated. In Batson, which involved the discriminatory use of peremptory challenges by the prosecution in choosing a petit jury, the Court held that “a single invidiously discriminatory governmental act” is sufficient to constitute a constitutional violation. See id. at 95-96, 106 S.Ct. at 1722. In Batson this meant that, rather than having to show a systematic race-based exclusion of petit jurors through the use of peremptory challenges, a criminal defendant could show that in his case alone the prosecutor exercised peremptory challenges to exclude members of the defendant’s race from the jury. See id. “The basic principles prohibiting exclusion of persons from participation in jury service on account of their race ‘are essentially the same for grand juries and for petit juries.’ ” Id. at 84 n. 3, 106 S.Ct. at 1716 n. 3 (quoting Alexander v. Louisiana, 405 U.S. 625, 626 n. 3, 92 S.Ct. 1221, 1223 n. 3, 31 L.Ed.2d 536 (1972)). Therefore, the analog to Batson in the context of grand jury selection is that, to establish a Fourteenth Amendment violation, a defendant must demonstrate purposeful discrimination in the selection of the grand jury panel that indicted him. The present case, however, presents a rather unusual factual situation. Here, it cannot be said that any prospective grand jurors were actually excluded from grand jury service on the basis of their race. The record indicates that two African-American prospective grand jurors were asked to sit in the body of the courtroom, for possible later selection, because they were African-American. These two prospective jurors were eventually empaneled. While an additional prospective grand juror was asked to sit in the body of the courtroom and was not subsequently empaneled, it is impossible to discern from the record either her race or whether she was initially passed over because of her race of because of some other factor such as her “background.” Therefore, this case presents the difficult question of whether purposeful discrimination may be shown absent proof of the actual exclusion from jury service of someone on the basis of her race. Under the rationale articulated in Batson and its progeny, it is necessary to establish three elements to raise an inference of discrimination in the context of grand jury selection. First, the prospective juror allegedly discriminated against must be a member of a cognizable racial group. See Batson, 476 U.S. at 96, 106 S.Ct. at 1722 (prohibiting use of peremptory challenges to strike members of defendant’s racial group); Powers, — U.S. at-, 111 S.Ct. at 1368 (extending Batson to prohibit use of peremptory challenges to strike jurors on the basis of their race regardless of whether their race is the same as the defendant’s race). Second, there must be “a jury selection practice that permits 'those to discriminate who are of a mind to discriminate.' ” Batson, 476 U.S. at 96, 106 S.Ct. at 1723 (quoting Avery v. Georgia, 345 U.S. 559, 562, 73 S.Ct. 891, 891, 97 L.Ed. 1244 (1953)). Finally, the defendant must show that the “opportunity for discrimination” was utilized. See Batson, 476 U.S. at 96-97, 106 S.Ct. at 1723 (in peremptory challenge context, the “defendant must show that these facts and any other relevant circumstances raise an inference that the prosecutor used that practice to exclude the veniremen from the petit jury on account of their race”); Whitus v. Georgia, 385 U.S. 545, 552, 87 S.Ct. 643, 647, 17 L.Ed.2d 599 (1967) (discrimination in the selection of a jury venire found when names of prospective black and white jurors were separated on the list used to select jurors and there existed a substantial disparity between the percentage of blacks in the population and that in the jury veni-re). In the present case, the first criterion is clearly met. The assignment judge’s statement, after three jurors were passed over for service, that he had asked two African-Americans to sit in the body of the courtroom in his effort to create a cross section of the community on the grand jury panel demonstrates that two prospective jurors, treated in this manner, were members of a cognizable racial group. Moreover, the second criterion is met. The practices employed by the assignment judges in the Essex County grand jury selection processes provided an opportunity for discrimination. As documented by the assignment judge’s statements during the grand jury selection at issue and by his later testimony, the practice permitted the judge subjectively to include or exclude jurors and provided the opportunity for discrimination based upon race. Cf. Vasquez, 474 U.S. at 256, 106 S.Ct. at 619 (subjective judicial selection of grand jurors provided an opportunity for discrimination); Cassell v. Texas, 339 U.S. 282, 287, 70 S.Ct. 629, 631, 94 L.Ed. 839 (1950) (jury commissioners’ subjective selection of jury venire provided an opportunity for discrimination); Smith v. Yeager, 465 F.2d 272, 280 (3d Cir.1972) (same); see also United States v. Calabrese, 942 F.2d 218, 227 (3d Cir.1991) (procedures which allowed courts subjectively to add categories for juror exclusion to the Jury Selection-and Service Act of 1988 provided an opportunity for discrimination). The real problem here is whether that opportunity for discrimination was utilized absent the actual exclusion of a juror from the panel on the basis of race. The previous case -law dealing with discrimination in the context of jury selection involves exclusion on account of race. See, e.g., McCollum, — U.S. at -, 112 S.Ct. at 2352 (defendant’s use of peremptory challenges to exclude African-Americans); Batson, 476 U.S. at 83, 106 S.Ct. at 1715 (prosecutor’s use of peremptory challenges to exclude blacks); Vasquez, 474 U.S. at 256, 106 S.Ct. at 619 (grand jury assignment judge’s exclusion of blacks from grand jury service); Castaneda v. Partida, 430 U.S. 482, 493-94, 97 S.Ct. 1272, 1279-80, 51 L.Ed.2d 498 (1977) (jury list’s substantial underrepresentation of Mexican-Americans); Cassell, 339 U.S. at 286-87, 70 S.Ct. at 631-32 (jury commissioners’ proportional limitation of blacks on grand jury). Moreover, an analytical focus of discriminatory jury selection claims is upon the opportunity of our citizens to deliberate as jurors. As the Supreme Court has recently stated in Powers: The opportunity for ordinary citizens to participate in the administration of justice has long been recognized as one of the principal justifications for retaining the jury system. Jury service preserves the democratic element of the law, as it guards the rights of the parties and insures continued acceptance of the laws by all of the people.... Indeed, with the exception of voting, for most citizens the honor and privilege of jury duty is their most significant opportunity to participate in the democratic process. Powers, — U.S. at -, 111 S.Ct. at 1368-69. The Court has also recently emphasized that, in general, discrimination in the grand jury selection process could “im-permissibly infect” court proceedings to the detriment of the defendant, prospective and actual jurors, and the community. See Vasquez, 474 U.S. at 263, 106 S.Ct. at 623 (infection of proceedings could deny the defendant a fair trial); see also Edmonson, — U.S. at-, 111 S.Ct. at 2087 (noting the “risk of open and public discrimination as a condition of [jurors’] participation in the justice system”); Powers, — U.S. at -, 111 S.Ct. at 1373 (“race neutrality in jury selection [is] a visible, and inevitable, measure of the judicial system’s own commitment to the commands of the Constitution”). Viewing the facts of the present case in the light of precedent, we conclude that the statements and actions of the assignment judge in the present case did not impermissibly infect the proceedings at issue and do not comprise an equal protection violation. First, there was no actual exclusion of a prospective juror on account of her race. The two jurors who were initially passed over based upon their race were eventually empaneled. Moreover, we are not willing to accept appellant’s invitation to speculate, absent supporting evidence in the record, that Francena Hard-wick, who was passed over and not seated, was both African-American and passed over on account of her race. Therefore, no prospective juror’s opportunity to deliberate was impermissibly denied for a tainted reason. Second, we do not believe that the assignment judge’s statements and actions short of actual exclusion comprised an equal protection violation. The assignment judge mentioned that he employed race as a factor in his effort to “pick a cross section of the community” and “get an even mix of people from background and races, and things like that.” App. at 2429, 2449. He also temporarily asked two African-American prospective jurors to sit in the body of the courtroom until they were belatedly empaneled. While we find objectionable this subjective sorting of the jury members and the judge’s statements about balancing the jury according to race, we cannot conclude that these activities violated the Equal Protection Clause. While subjectively rigging the jury to represent his vision of the appropriate representation of Essex County’s population was ill-conceived, it apparently was not motivated by a desire to discriminate purposefully against African-Americans, nor was it apparently an attempt expressly to limit the number of African-Americans who could serve on an Essex County grand jury. These factors distinguish the present case from Cassell, 339 U.S. at 288-89, 70 S.Ct. at 632-33. Cassell involved jury commissioners’ limitation of African-Americans on grand jury panels to one African-American per panel. The jury commissioners expressed both their belief in the legitimacy of proportionally limiting the number of African-Americans on grand jury panels to their representation in the general population and their opinions that they did not know any African-Americans who were qualified to serve as grand jurors. See id. The Court inferred, based upon such statements, that the jury commissioners had purposefully discriminated against African-Americans and engaged in “proportional limitation” of African-Americans on grand jury panels. See id. at 287, 70 S.Ct. at 631. Here, the assignment judges’ statements demonstrate no such desire proportionally to limit the number of blacks to some cutoff figure, nor do they indicate the presence of purposeful manifestations of invidious discrimination. The judge in the present case apparently wished the non-invidious objective of a representative jury. We are reluctant to infer an invidious discriminatory purpose when the record does not support such an inference. See, e.g., United States v. Bedonie, 913 F.2d 782, 795 (10th Cir.1990), cert. denied, — U.S. -, 111 S.Ct. 2895, 115 L.Ed.2d 1059 (1991) (refusing to infer that use of peremptory strike of Native American prospective juror was racially motivated). Moreover, the judge who empaneled Ramseur’s grand jury explicitly rebuked the legitimacy of racial discrimination in grand jury proceedings. When a juror had expressed that she was prejudiced against “certain people, certain races,” the judge, in excusing her, stated, “I appreciate your honesty, we don’t want people like you to serve on the Grand Jury and I will be just as honest with you.” See App. at 2464. To the extent that prospective jurors and the public might have interpreted the assignment judge’s jury sorting as purposeful discrimination, this rebuke of juror prejudice would countermand such an interpretation. Unlike Batson and its peremptory challenge progeny, the present case involves no instances of the imprimatur of the state being given to discriminatory actions, such as exclusions. Rather, the present case involves statements and conduct, which although objectionable, simply do not warrant an inference of “purposeful” discrimination. Indubitably, we seek to eradicate discrimination from the grand jury selection process. However, the record here does not justify our finding either an invidious purpose to discriminate or a communication to the public by the assignment judge that the state countenanced racial prejudice. III. We will next address Ramseur’s challenge to the juror source lists used in Essex County. Ramseur contends that, because it underrepresents the African-American community in Essex County, the composition of the juror source list and resulting qualified pool of jurors violates his Fourteenth Amendment right to Equal Protection of the laws and his Sixth Amendment right to a grand and petit jury drawn from a cross-section of his community. We will first set forth the factual basis for his claim and then explore the legal framework within which these facts must be examined. Since 1979, grand and petit juries in Essex County have been chosen from a source list consisting of the names found on the Department of Motor Vehicles licensed driver list and the voter registration list. From this source list, which is arranged by municipality, there is derived a “master” list consisting of the names of all people to whom qualifying questionnaires may be sent. Jury managers decide the number of questionnaires to be sent based on the anticipated need for jurors and past experience regarding the rate of return expected from the questionnaire mailing. The appropriate number of questionnaires are then sent to people selected randomly from the source list. This method is designed to ensure that prospective jurors are selected from each street in each municipality without choosing more than one person from any one household. Twenty to twenty-eight percent of the questionnaires sent are completed and returned. Those questionnaires are then screened for eligibility. Those who have served on £ jury within the past seven years or who have received a questionnaire within the past four years are not eligible. If the response to the questionnaire indicates extreme hardship, the potential juror is excused. The remaining names are placed on the “qualified” list. Once the qualified list is completed, jury managers make a random selection of grand jurors from it. Those who are not chosen as grand jurors are designated as petit jurors. The resulting lists are divided into panels and summonses are sent to the jurors on the panels, ordering them to report for jury duty. The 1980 census figures indicated that African-American adults in Essex County comprised 35.9 percent of the population of those between the ages of eighteen and seventy-four. Defense experts in this case conducted three separate surveys between 1981 and 1982 in order to determine the percentage of African-Americans on the jury lists. Two were telephone surveys— one conducted in May 1981, the other in May 1982. The third was a geographic inference survey, a study in which the race of each juror is inferred from census information regarding the racial makeup of the area of Essex County in which the juror lives. See Ramseur, 106 N.J. at 214 n. 41; 524 A.2d at 233 n. 41. By averaging the results of the telephone surveys and the geographic inference study, defense experts concluded that only 21.2% of the persons on the source list and 21.8% of those on the qualified list were African-American. The Equal Protection Clause of the Fourteenth Amendment requires the eradication of “racial discrimination in the procedures used to select the venire from which individual jurors are drawn.” Batson v. Kentucky, 476 U.S. 79, 86, 106 S.Ct. 1712, 1717, 90 L.Ed.2d 69 (1986). The Sixth Amendment requires that jurors be drawn from pools that represent a “fair cross-section” of the community. “[Jjury wheels, pools of names, panels, or venires from which juries are drawn must not systematically exclude distinctive groups in the community and thereby fail to be reasonably representative thereof.” Duren v. Missouri, 439 U.S. 357, 363-64, 99 S.Ct. 664, 668, 58 L.Ed.2d 579 (1979) (quoting Taylor v. Louisiana, 419 U.S. 522, 538, 95 S.Ct. 692, 701, 42 L.Ed.2d 690 (1975)). The requirements a party must meet when challenging the jury selection process as being racially discriminatory are comparable under equal protection and fair cross-section analysis. To prove either claim, a defendant must first identify a constitutionally cognizable group, that is, a group capable of being singled out for discriminatory treatment. See Castaneda v. Partida, 430 U.S. 482, 494, 97 S.Ct. 1272, 1280, 51 L.Ed.2d 498 (1977); Duren, 439 U.S. at 364, 99 S.Ct. at 668. Ramseur has clearly met the first prong of both the equal protection and fair cross-section tests because African-Americans are unquestionably a constitutionally cognizable group. See, e.g., Batson, 476 U.S. at 84, 106 S.Ct. at 1716; Rose v. Mitchell, 443 U.S. 545, 551, 99 S.Ct. 2993, 2997, 61 L.Ed.2d 739 (1979). Second, to prove an equal protection violation, the defendant must show that the cognizable group was subject to “substantial underrepresentation” over a significant period of time. See Castaneda, 430 U.S. at 494, 97 S.Ct. at 1280. The second requirement in a Sixth Amendment challenge may differ somewhat in application: the defendant must show that the representation of the cognizable group was not “fair and reasonable in relation to the number of such persons in the community.” Duren, 439 U.S. at 364, 99 S.Ct. at 668. Finally, to prove an equal protection violation, the defendant must show that the procedure which is being used to select the jurors is “susceptible of abuse or is not racially neutral.” Castaneda, 430 U.S. at 494, 97 S.Ct. at 1280. To succeed in a Sixth Amendment challenge, the defendant need show only that the underrepresentation was a result of systematic exclusion of the group in the jury selection process. See Duren, 439 U.S. at 364, 99 S.Ct. at 668. The imbalance necessary to establish an equal protection or Sixth Amendment violation in the composition of a jury venire is not determined by a bright line test. The Supreme Court “has never announced mathematical standards” that would apply to all such challenges. The Court has, however, recognized that it may be possible to infer that unconstitutional exclusion of cognizable groups exists when there is a disparity between a group’s population figures and its representation in the jury venire sufficiently large that it is extremely unlikely that the disparity results from random chance. See Castaneda, 430 U.S. at 496 & n. 17, 97 S.Ct. at 1281 & n. 17; Alexander v. Louisiana, 405 U.S. 625, 630 & n. 9, 92 S.Ct. 1221, 1225 & n. 9, 31 L.Ed.2d 536 (1972). Such an inference may be bolstered or weakened by a direct examination of the reasons for any seemingly non-random disparity that exists. See Alexander, 405 U.S. at 630, 92 S.Ct. at 1225. In the present case, we find the New Jersey Supreme Court’s analysis of Ramseur’s challenges to the composition of his jury venire carefully reasoned and persuasively explained. See Ramseur, 106 N.J. at 212-28, 524 A.2d at 232-40. It is both logical and consistent with the United States Supreme Court’s requirements that these challenges be evaluated, first, by determining whether there is a non-random disparity between the percentage of African-Americans who live in Essex County and the percentage of African-Americans on Essex County jury lists, and second, if there is an imbalance, by addressing the reasons for it. In proceeding to determine whether a non-random disparity existed, we examine evidence of absolute disparity, comparative disparity, and deviation from expected random selection. The results of Ramseur’s studies reflected an absolute disparity of 14.1% for the qualified list and an absolute disparity of 14.6% for the source list. Absolute disparity in the jury selection context is defined as the difference between the percentage of a certain population group eligible for jury duty and the percentage of that group who actually appear in the venire. The results reflected a comparative disparity of 39.3% for the qualified list and 40.1% for the source list. Comparative disparity is calculated by dividing the absolute disparity by the population figure for a population group. It measures the diminished likelihood that members of an underrepresented group, when compared to the population as a whole, will be called for jury service. Finally, according to the defendant’s unchallenged statistical analysis, a standard deviation analysis revealed that there was an observed deviation of 28.9 “standard deviations” from the expected deviation. This analysis explains the probability that the disparity between the percentages of African-Americans in the population of Essex County and African-Americans in the qualified pool and source list is a result of random chance. A deviation of 28.9 standard deviations means that the probability that the disparity at issue is the result of random chance is less than 1 in 10140. See Castaneda v. Partida, 430 U.S. 482, 496 & n. 17, 97 S.Ct. 1272, 1281 & n. 17, 51 L.Ed.2d 498 (1977). If we compare' the 14.1% absolute disparity in the Essex County juror source list with the results in other cases, we find the 14.1% to be of borderline significance. Courts addressing the question of whether a given absolute disparity constitutes “substantial underrepresentation” have held that absolute disparities between 2.0% and 11.5% do not constitute substantial under-representation. However, some courts have found disparities of between 10% and 16% sufficient to establish “substantial underrepresentation.” See Jones v. Georgia, 389 U.S. 24, 25, 88 S.Ct. 4, 5, 19 L.Ed.2d 25 (1967) (15.7%); Hernandez v. Texas, 347 U.S. 475, 480-81, 74 S.Ct. 667, 671-72, 98 L.Ed. 866 (1954) (14.0%); Stephens v. Cox, 449 F.2d 657, 659-60 (4th Cir.1971) (15.0%). Ramseur’s evidence of a comparative disparity of about 40% is also borderline. It is a bit lower than the 45.4% condemned in Preston v. Mandville, 428 F.2d 1392 (5th Cir.1970) and close to the 42% comparative disparity found permissible in Swain v. Alabama, 380 U.S. 202, 85 S.Ct. 824, 13 L.Ed.2d 759 (1965). Thus, both the absolute and comparative disparity analyses present results at the margin of the range found acceptable by the courts. Perhaps more importantly, however, the standard deviation analysis revealed 28.9 standard deviations, a departure from the expected value which would occur by random chance in less than 1 in 10140 occasions. This figure is virtually identical to the 29 standard deviations condemned in Castaneda, 430 U.S. at 496 & n. 17, 97 S.Ct. at 1281 & n. 17. We, like the New Jersey Supreme Court, find that this standard deviation analysis reveals that the un-derrepresentation of blacks in the Essex County jury pools is not the result of random selection. However, the presence of non-random underrepresentation does not necessarily mean that purposeful discrimination has occurred; rather, any implication of discrimination that exists must be evaluated in light of the circumstances surrounding the present case. Particularly significant to our analysis are the time period covered by Mr. Ramseur’s evidence of underrepresentation and the nature of the source lists and the procedures by which they are compiled. The evidence which is relied upon to document the underrepresentation of a minority group on jury lists must be carefully scrutinized. See Castaneda, 430 U.S. at 494, 97 S.Ct. at 1280 (requiring evidence that demonstrates underrepresentation “over a significant period of time”). As one court has stated: A disparity of fifteen percentage points is much greater in a case where the class or group represents only twenty percent of the general population, than where the class or group represents seventy percent of the population. Similarly, a disparity of fifteen percentage points is much more significant if it has continued for ten years, than if it has occurred in only one isolated year. The magnitude of a disparity may also depend on whether the statistics are based on one grand jury venire of thirty people, or on dozens of grand jury venires representing thousands of people. Bryant, 686 F.2d at 1377 (emphasis added). In the present case, the defense conducted two telephone surveys covering a two year period and contacted a total of 739 persons on the juror source list. In addition, the geographically inferred study covered the 7,149 persons on the May 1982 qualified list. We believe that the brief duration and limited sample size of the surveys undermine an inference that substantial underrepresentation of blacks took place over a significant period of time. Those studies which have been found to satisfy Castaneda's requirement of a “significant period of time” have covered periods substantially longer than the two years covered by this study. See Hobby v. United States, 468 U.S. 339, 341, 104 S.Ct. 3093, 3094, 82 L.Ed.2d 260 (1984) (7 years); Castaneda, 430 U.S. at 487, 97 S.Ct. at 1276 (11 years); Hernandez, 347 U.S. at 481, 74 S.Ct. at 671 (25 years). Therefore, while the figures presented by Ramseur indicate that the underrepresentation was not random, we find that the figures are insufficient to meet Castaneda’s requirement that underrepresentation occur over a significant period of time to comprise an equal protection violation. Moreover, an examination of the source lists reveals that the mechanism used to create the source lists was facially neutral with respect to race. Essex County, New Jersey utilized voter registration and-Department of Motor Vehicle lists to create its jury venire. These lists are constituted using facially neutral criteria and allow no opportunity for subjective or racially motivated judgments. See Alexander, 405 U.S. at 630, 92 S.Ct. at 1225 (finding discrimination when jury commissioners had “a clear and easy opportunity for racial discrimination”); Yeager, 465 F.2d at 280 (striking down system that permitted “tainted subjective decision[s]”). When we combine the presence of multiple, facially neutral selection lists with the presence of studies indicating non-random underrepresentation of blacks that cover only two years duration, we find that the “substantial underrepresentation” requirement of Castaneda is unfulfilled. Any implication of discrimination provided by Mr. Ramseur’s data is undercut both by the limited usefulness of the data itself and the context of the Essex County jury system. Therefore, we conclude that Mr. Ramseur’s data is insufficient to support the presumption of discrimination necessary under a Castaneda equal protection claim in order to create a prima facia case. With regard to Ramseur’s Sixth Amendment challenge to the source lists, we conclude that the studies conducted here, which do not reflect substantial un-derrepresentation over a significant period of time, also do not satisfy Duren’s fair cross-section analysis. Duren’s second prong requires a cognizable group’s representation to be “fair and reasonable in relation to the number of such persons in the community.” Duren, 439 U.S. at 364, 99 S.Ct. at 668. In making our Sixth Amendment analysis, we use standards that are somewhat different than those under the “substantial underrepresentation” requirement of the equal protection analysis. A significant reason for this is that the focus of Sixth Amendment protections, more than Fourteenth Amendment protections, is upon the concept of the jury as a system rather than upon individual rights. As Justice White stated in Taylor v. Louisiana: 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 overconditioned or biased response of a judge.... This prophylactic vehicle is not provided if the jury pool is made up of only special segments of the populace or if large, distinctive groups are excluded from the pool. Community participation in the administration of the criminal law, moreover, is not only consistent with our democratic heritage but is also critical to public confidence in the fairness of the criminal justice system. Restricting jury service to only special groups or excluding identifiable segments playing major roles in the community cannot be squared with the constitutional concept of jury trial. “Trial by jury presupposes a jury drawn from a pool broadly representative of the community as well as impartial in a specific case.... [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.” 419 U.S. at 530-31, 95 S.Ct. at 697-98 (quoting Thiel v. Southern Pacific Co., 328 U.S. 217, 227, 66 S.Ct. 984, 989, 90 L.Ed. 1181 (1946) (Frankfurter, J., dissenting); other citations omitted). Therefore, a Sixth Amendment analysis will examine whether a group’s representation on jury lists is “unfair” such that the proper functioning of the jury system is threatened. We believe that factors such as the nature of,the process by which jury lists are composed, the length of time of underrepresentation, and the strength of the evidence that purports to establish an “unfair and unreasonable” representation should be examined under Duren. See, e.g., Ford v. Seabold, 841 F.2d 677, 685 (6th Cir.1988), cert. denied, 488 U.S. 928, 109 S.Ct. 315, 102 L.Ed.2d 334 (1988) (examining length of time of under-representation and nature of process by which jury lists are composed); Timmel v. Phillips, 799 F.2d 1083, 1086 (5th Cir.1986) (same); United States v. LaChance, 788 F.2d 856, 866-69 (2d Cir.1986) (examining length of time of underrepresentation and strength of evidence that purports to establish an unfair and unreasonable representation). In the present case, the evidence does not convincingly demonstrate that the representation of African-Americans on jury pools was unfair or unreasonable. The length of time documented by Ramseur’s studies was only two years. Moreover, the selection process was facially neutral and included names from both the voter registration and DMV lists. Additionally, the New Jersey Supreme Court found it significant that these lists were employed as part of an on-going effort in New Jersey to increase the representativeness of the State’s jury lists: [W]e look to the State’s efforts at reform. We are not dealing here with a system in which there has been longstanding abuse with no attempts at reform. New Jersey has been conscious of its obligation to achieve greater neutrality and representativeness in its jury selection system. The addition of the DMV lists in 1979 — at a time when very few jurisdictions, state or federal, required the use of multiple lists in addition to voter lists — was obviously intended to broaden the representativeness of the pool. In addition, a 1981 Task Force chaired by Justice Clifford to study the current jury system has made numerous recommendations that may serve to increase the representativeness of juries. We are certain that those currently working on improvements in jury procedures will continue to seek to improve the yield of jurors from the source lists. Ramseur, 106 N.J. at 226, 524 A.2d at 239. Such efforts at reform to increase the representativeness of jury lists have some relevance to the question of whether a group’s representation on those lists is “fair and reasonable.” If a system appears ex ante likely to create representative jury lists there should be some presumption of its legitimacy, even though evidence ex post may demonstrate that the lists are not representative enough. We find the evidence ex post presented in this case insufficient to establish that Essex County’s jury selection system was unfair and unreasonable enough to constitute a Sixth Amendment violation. We hold, therefore, that Ramseur has not made a prima facie showing that the use of these juror source lists denied him his rights under the Equal Protection Clause of the Fourteenth Amendment, nor has he established a denial of his right to trial by a fair cross-section of the community guaranteed by the Sixth Amendment. IV. Next, Ramseur contends that the manner in which grand jury forepersons are selected by Essex County assignment judges violates the Equal Protection Clause of the Fourteenth Amendment as well as the Sixth Amendment’s guarantee of a trial by a jury drawn from a cross-section of the community. We look first to the methods employed in the selection of forepersons in Essex County. Once a grand jury is chosen in Essex County, the assignment judges use their discretion in choosing a foreperson and deputy foreperson. The judges review all the questionnaires and proffered excuses prior to empaneling the jury, and once the jury is empaneled, the judges confer with the grand jury manager to determine which jurors would make the best forepersons and deputies. The assignment judges, who testified in Ramseur’s New Jersey state court proceedings about the procedures employed in Essex County to select grand jury forepersons, related that race was a factor in their choice of a foreperson or deputy foreperson. One judge stated, “When their names were called to serve as jurors I would observe them, I would determine whether they were black or white because, obviously, that does not appear in the questionnaire and it does not appear on the jury list and so I had no way of knowing whether or not all of these people were Caucasian or black or Chinese or what they were.” App. at 3791. Another judge explained: My ultimate choice may be based upon my desire to get a cross-section even in the selection of the foreperson or deputy foreperson; although as I indicated before, I do not think that is a requirement. I think that there should be a certain number of men and I think there should be a certain number of women and I think there should be a certain number of blacks and Hispanics and whites and laboring groups and executive groups and housewife groups who should not only be on the Grand Jury body but even sharing as foreperson or deputy foreperson. App. at 3880. To support his allegation that grand jury forepersons were chosen in a manner viola-tive of the Constitution, Ramseur contacted thirty-three of the sixty-six persons who served as grand jury forepersons between the start of the 1979 Term and September of 1982, the month in which Ramseur was indicted. Ramseur’s survey determined that only two, or 6.1%, of the thirty-three former forepersons contacted were African-American, although 35.9% of the voting population was African-American. Turning first to Ramseur’s Sixth Amendment challenge, the constitutional guarantee of a speedy trial by an “impartial jury” demands that jury members be drawn from a fair cross-section of the community. Taylor v. Louisiana, 419 U.S. 522, 527, 95 S.Ct. 692, 696, 42 L.Ed.2d 690 (1975). At the outset it should be noted that this guarantee differs substantially from the protections that flow from the Equal Protection Clause. See United States v. Musto, 540 F.Supp. 346, 362 (D.N.J.1982), aff'd sub nom., United States v. Aimone, 715 F.2d 822 (3d Cir.1982). In equal protection analysis, the focus is on purposeful discrimination against individuals, whereas in a fair cross-section challenge “the focus is not on discriminatory conduct but instead is on whether the jury selection system is impartial and will yield a microcosm of the community which can fairly represent the views of all persons within the society.” Id. (citing Taylor, 419 U.S. at 522, 95 S.Ct. at 692). Unlike the equal protection analysis under the Fourteenth Amendment or the analogous equal protection analysis under the Fifth Amendment which are based primarily upon directly protecting individuals from government abuse, the “fair cross-section” analysis under the Sixth Amendment, as we note in Section III above, is grounded primarily to provide “jury review” as a buffer for the people from the abuses of government power. The emphasis is on the system and its proper functioning rather than on the individual citizen. Because of these differing emphases of the Sixth and Fourteenth Amendments, decisions of other courts holding that forepersons occupy constitutionally significant positions for equal protection purposes do not compel the conclusion that, as a matter of law, the role of foreperson is constitutionally significant in the context of a Sixth Amendment challenge. See Musto, 540 F.Supp. at 360-61. Instead, when a defendant alleges that fair cross-section values are tainted by the underrepresentation of a particular group in the office of grand jury foreperson, the defendant must show that the office of foreperson carries with it some power to affect the deliberative process of the array which otherwise represents a cross-section of the community. As one district court has aptly observed: When a group as a whole is excluded or significantly underrepresented on a jury, the defendant is denied the attitudes, experiences, outlook, and accumulated wisdom of that group. The relevance of the similar question to the office of foreperson, however, is not so clear. Assuming a fair cross-section on the júry as a whole, the defendant enjoys the richness of the community's general make-up, even where certain groups are underrepresented as forepersons. The benefit of the fair cross-section to the defendant is destroyed only if the “impact of the grand jury foreperson is so substantial as to influence or alter the unique qualities and characteristics of the jury’s individual members.” United States v. Cabrera-Sarmiento, 533 F.Supp. 799, 808 (S.D.Fla.1982) (citations omitted). The successful defendant will show that the foreperson alters the representative character of the jury when the foreperson exerts an “overpowering influence” over the other jurors such that their views are substantially diminished during the deliberative process. See Musto, 540 F.Supp. at 362. If the grand jury foreperson plays no more significant role than any other grand juror and “jury arrays are representative of the community,” Sixth Amendment values are not disturbed even if there is underrepresentation in the foreperson position. Id. In Musto, a case involving the role of the foreperson in a federal grand jury, the court determined that no Sixth Amendment violation existed because the role of the federal grand jury foreperson was purely ministerial and such a foreperson would be unlikely to affect improperly an otherwise representative array of grand jurors. Id. Similarly, we must determine whether Essex County grand jury forepersons “have such a significant impact on the criminal justice system that discrimination in their selection amounts to a constitutional violation.” United States v. Aimone, 715 F.2d 822, 826 (3d Cir.1983). The Supreme Court of New Jersey concluded that the duties of the grand jury foreperson in Essex County were “not constitutionally significant.” Ramseur, 106 N.J. at 238, 524 A.2d at 244. Because forepersons in Essex County have no more significant powers than federal grand jury forepersons, whom we determined in Aimone do not play a constitutionally significant role in the indictment process, we agree with the district court’s conclusion that the role of Essex County foreperson is not constitutionally significant in a fair cross-section challenge of this nature. Therefore, we hold that even if African-Americans and women were underrepresented in the position of foreperson in Essex County, such underrepresentation did not violate the Sixth Amendment because their underrepresentation had no significant impact on the otherwise representative array of the panel. With regard to Ramseur's equal protection challenge, unlike a Sixth Amendment challenge, an equal protection challenge to the selection of grand jury forepersons does not require that the forepersons’ role be more than ministerial or invested with substantial influence. Because the Fourteenth Amendment focuses primarily upon direct protection for individuals, any invidious discrimination in the jury selection process gives rise to an equal protection challenge to that discrimination. See Georgia v. McCollum, — U.S. -, -, 112 S.Ct. 2348, 2352, 120 L.Ed.2d 33 (1992) (noting that “the harm from discriminatory jury selection extends beyond that inflicted on the defendant and the excluded juror to touch the entire community” and that equal protection analysis serves to “remedy the harm done to the dignity of persons and to the integrity of the courts”) (quotations omitted). The State of New Jersey contended that Hobby v. United States, 468 U.S. 339, 104 S.Ct. 3093, 82 L.Ed.2d 260 (1984), governs the outcome of Ramseur’s equal protection challenge. We find the state’s reliance on Hobby misplaced. In Hobby, the Supreme Court held that, even if discrimination was a factor in the selection of federal grand jury forepersons, such discrimination did not constitute a violation of the Due Process Clause of the Fifth Amendment justifying reversal of the petitioner’s conviction or the dismissal of the indictment. In reaching its conclusion, the Court examined the role of the foreperson in a federal grand jury and found it to be insignificant because “discrimination in the selection of one person from among the members of a properly constituted grand jury can have little, if indeed any, appreciable effect upon the defendant’s due process right to fundamental fairness.” 468 U.S. at 345, 104 S.Ct. at 3096. In Rose v. Mitchell, 443 U.S. 545, 551 n. 4, 99 S.Ct. 2993, 2998 n. 4, 61 L.Ed.2d 739 (1979), a case in which the petitioner alleged that his rights under the Equal Protection Clause had been violated, the Court assumed without deciding that discrimination in the selection of a grand jury foreperson would violate equal protection and require that a conviction be set aside. The Court’s analysis in Hobby regarding the “constitutional” role of the foreperson in the context of a Fifth Amendment due process challenge did not limit the Court’s previous analysis in Rose with respect to equal protection principles in the grand jury selection process. According to Rose, where the claim is injury to equal protection values, the injury occurs whether the discriminatory exclusion affects the selection of individual grand jurors to the panel or the selection of the foreperson from among the grand jurors on the panel. These injuries exist regardless of the extent of the foreperson’s authority which was the Court’s principle concern in Hobby. See Johnson v. Puckett, 929 F.2d 1067, 1071 (5th Cir.1991), cert. denied, — U.S. -, 112 S.Ct. 274, 116 L.Ed.2d 226 (1991). Therefore, Rose, not Hobby, governs the outcome of Ramseur’s equal protection claim. The Court in Rose cautioned that “[notwithstanding these holdings that claims of discrimination in the selection of members of the grand jury are cognizable on federal habeas corpus, and will support issuance of a writ setting aside a state conviction and ordering the indictment quashed, it remains true that to be entitled to habeas relief the [defendant is] required to prove discrimination under the standards set out in this Court’s cases.” Id. 443 U.S. at 564-65, 99 S.Ct. at 3004-05. Therefore, we must now determine if the requirements of Castaneda, set forth in Section III of this opinion, have been met with respect to the selection of forepersons in Essex County. We conclude that.Ramseur has not established a prima facie case under Castaneda. We conclude that Ramseur’s survey of only half of the sixty-six forepersons who served over a three year period did not conclusively establish that the procedure “resulted in substantial underrepresentation of [defendant’s] race or of the identifiable group to which he belongs ... over a significant period of time.” Castaneda, 430 U.S. at 494, 97 S.Ct. at 1280. As noted in our discussion of Ramseur’s study of Essex County’s grand jury venire, see supra, a three year study which covers only half of the group at issue does not comport with cases in which courts have found violations of the standards set forth in Castaneda. Moreover, a sampling of only half of the jury forepersons who served could skew the results. First, it is possible that those contacted do not represent the entire three year period purportedly covered by the study. Second, it is possible that, even if such a sample did span the entire period, it might represent merely a series of sporadic periods of underrepresentation rather than one that continued “over a significant period of time.” Ramseur’s study is thus insufficient to establish a constitutional violation under Castaneda. Therefore, we hold that Ramseur has not made a successful equal protection challenge with respect to the selection of grand jury forepersons in Essex County. V.’ Finally, we address Ramseur’s contention that misconduct by the prosecutor denied him his right to a fair trial in violation of the Sixth Amendment. The conduct alleged to be improper includes: (1) the prosecutor’s reference to matters outside of the evidence; (2) the prosecutor’s testimony regarding his personal opinion as to the veracity of testimony and Ramseur’s guilt; (3) the prosecutor’s ridicule of defense experts that allegedly resulted in their impeachment; (4) the prosecutor’s misstatements about and improper use of evidence; and (5) the prosecutor’s mischaracterization of the defense. At the outset we make the observation that “not every trial error or infirmity which might call for application of supervisory powers ' correspondingly constitutes a ‘failure to observe that fundamental fairness essential to the administration of justice.’ ” Donnelly v. DeChristoforo, 416 U.S. 637, 642, 94 S.Ct. 1868, 1871, 40 L.Ed.2d 431 (1974) (quoting Lisenba v. California, 314 U.S. 219, 236, 62 S.Ct. 280, 289, 86 L.Ed. 166 (1941)). Instead, our review of a prosecutor’s conduct in a state trial on application for a writ of habeas corpus is limited to determining whether the prosecution’s conduct “so infect[ed] the trial with unfairness as to make the resulting conviction a denial of due process.” Greer v. Miller, 483 U.S. 756, 765, 107 S.Ct. 3102, 3108, 97 L.Ed.2d 618 (1987) (quoting Donnelly, 416 U.S. at 643, 94 S.Ct. at 1871); Lesko v. Lehman, 925 F.2d 1527, 1546 (3d Cir.1991), cert. denied, — U.S. -, 112 S.Ct. 273, 116 L.Ed.2d 226 (1991). While the line is sometimes a fine one, “[i]t is essential to distinguish between ordinary trial error and that sort of egregious misconduct which amounts to a denial of constitutional due process.” United States ex rel. Perry v. Mulligan, 544 F.2d 674, 678 (3d Cir.1976), cert. denied, 430 U.S. 972, 97 S.Ct. 1659, 52 L.Ed.2d 365 (1977). In evaluating whether the prosecutor’s misconduct rose to the level of constitutional violation, we must examine that conduct in the context of the trial as a whole. See Greer, 483 U.S. at 766, 107 S.Ct. at 3109 (determining whether “remarks, in the context of the entire trial, were sufficiently prejudicial to violate respondent’s due process rights”) (quoting Donnelly, 416 U.S. at 639, 94 S.Ct. at 1869); see also United States v. Adams, 759 F.2d 1099, 1111 (3d Cir.), cert. denied, 474 U.S. 971, 106 S.Ct. 336, 88 L.Ed.2d 321 (1985) (examining the “context of the entire trial”). We conclude, as did the New Jersey Supreme Court, that although there were instances in whic