The State of North Carolina appeals from an order of the district court issuing writs of habeas corpus on behalf of Charles Bruce Keeten, Bernard Avery, and Larry Darnell Williams, and granting Williams’ claim for relief from his death sentence. We conclude that the district court erred in issuing the writs and in granting Williams relief and, therefore, reverse.
I.
In 1968, the Supreme Court held that a venireman in a capital case may be excluded for cause if he is unwilling “to consider all of the penalties provided by state law.” Witherspoon v. Illinois, 391 U.S. 510, 522 n. 21, 88 S.Ct. 1770, 1777 n. 21, 20 L.Ed.2d 776 (1968). The Witherspoon decision left open the question of whether the Court might someday find that the Constitution required two separate juries in capital cases: one to determine guilt or innocence, and one to determine punishment. Under this bifurcated proceeding, the first jury may contain jurors who could not vote for the death penalty, although the second jury may not. The Court stated that for it to consider such a proceeding there would have to be a showing “that the exclusion of *131jurors opposed to capital punishment results in an unrepresentative jury on the issue of guilt or substantially increases the risk of conviction.” Id. at 518, 88 S.Ct. at 1775. The Court declined to so hold in Witherspoon because the data presented in that case was “too tentative and fragmentary.” Id. at 517, 88 S.Ct. at 1774. The appeals we consider today address the issue of whether, if such a showing is made, it necessarily mandates the adoption of the two-jury proceeding outlined in Wither-spoon. * * *
In 1976, Keeten was convicted of murder and sentenced to life imprisonment following a single-stage trial under North Carolina’s then customary non-statutory state practice. Under this system, a single jury determined both guilt and sentencing in a one-stage proceeding.1
In 1980, Avery was convicted of murder and sentenced to life imprisonment and, in 1982, Williams was convicted of murder and sentenced to death. Both men were sentenced under North Carolina’s current death penalty statute enacted in 1977, N.C. Gen.Stat. § 15A-2000, et seq., following jury trials conducted pursuant to § 15A-2000(a)(2). This statute provides that a single jury shall hear both the guilt phase and the sentencing phase, but shall do so in two stages.
In each of the three cases, during the voir dire of the prospective jurors, some persons were excluded because they would not consider returning the death penalty. The exclusion of these jurors was pursuant to the Supreme Court’s decision in Witherspoon, North Carolina case law, State v. Bowman, 80 N.C. 432 (1879), and, in Keeten’s and Avery’s cases, N.C.Gen.Stat. § 15(A)-1212(8).2 In addition, at Williams’ trial, one of the prospective jurors, Nancy Melton, was excluded for cause as a Witherspoon -excludable (“WE”) when, on each occasion that she was questioned, she stated that she was “not sure” she could follow the law if it required imposition of the death penalty.
In their habeas petitions in the district court, Keeten, Avery and Williams (“petitioners”) urged that recent studies had proven that the exclusion of death penalty opponents created a conviction prone jury. Petitioners introduced two types of studies as proof of their claim: attitudinal surveys and mock trial studies. The attitudinal surveys, which compared an individual’s responses to certain “prosecution-oriented” questions with his view on the death penalty, indicated that either strong death penalty opponents or WE’s gave fewer conviction prone answers to questions than did (1) death penalty proponents, (2) persons who were neutral on the death penalty, or (3) persons who moderately opposed the death penalty.3 The mock trial studies, *132which compared an individual’s view on the death penalty with his willingness to convict in a mock trial setting, indicated that WE’s and death penalty opponents convicted less frequently than death-qualified jurors (“DQ’s”),4 and those not opposing capital punishment.5 Petitioners claimed that the attitudinal surveys and the mock trial studies justified the need for a bifurcated jury proceeding in which one jury would determine guilt or innocence and, if the defendant were found guilty, a separate jury would determine the appropriate sentence.
The State’s expert witnesses testified that these studies were flawed. According to these witnesses, the most serious flaws were the lack of random samples in most of the studies, poor test design, and failure to check for consistency a person’s responses to related survey questions. The State’s experts also expressed the view that the attitudes asked about in the surveys were too general to predict behaviors, and that past behaviors caused attitudes rather than vice versa. These conclusions were supported by the research of Dr. Steven Pen-rod, a psychologist specializing in jury research. Dr. Penrod found almost no correlation between attitudes about the criminal justice system and verdicts in mock trials which he conducted, between the verdicts themselves, or between the several attitudes about which he inquired. Finally, the State relied on statistics from petitioners’ own studies, which established that the impact of conviction prone differences between groups, to the extent it exists, is minimal, except in one respect: opposition to the death penalty strongly increases the likelihood of juror nullification. This nullification occurs when a juror so strongly opposes the death penalty that he will refuse to convict if the death penalty can be imposed as a result of the conviction.
Based on the statistical information presented to it, the district court found that: (1) WE’s are a distinctive group in the community and their exclusion violates the requirement under the Sixth Amendment to the United States Constitution that a jury panel be drawn from a fair cross-section of the community;6 (2) the exclusion of WE’s created a conviction prone jury in violation of the due process clause of the *133Fourteenth Amendment; and (3) in Williams’ case, prospective juror Melton was challenged for cause in violation of Wither-spoon.
The State appeals.
II.
On appeal, the State contends that the district court erred in finding that the exclusion of WE’s created a conviction prone jury in violation of due process and petitioners’ right to a jury selected from a fair cross-section of the community. The State also contends that the district court erred in finding that, in Williams’ case, prospective juror Melton was challenged for cause in violation of Witherspoon. We agree with each of these contentions.
Assuming that the evidence petitioners presented to the district court does indeed supply the persuasiveness found lacking by the Supreme Court in Witherspoon,7 we nonetheless conclude that the district court erred in granting petitioners relief under this argument.
A. The Fair Cross-Section Requirement
Petitioners urged below and the district court held that jurors opposed to the death penalty comprise a distinctive group in the community whose exclusion violates the Sixth Amendment. We disagree.
Although the right to a jury trial includes the right to a jury venire drawn from a representative cross-section of the community, it does not include the right to be tried by jurors who are unable or unwilling to follow the law and the instructions of the trial judge in a capital case. Lockett v. Ohio, 438 U.S. 586, 596-97, 98 S.Ct. 2954, 2960, 57 L.Ed.2d 973 (1978). The state as well as the defendant has legitimate and vital interests at stake at both the guilt and penalty stages of a capital case.
The Supreme Court recently reaffirmed this principle in Adams v. Texas, 448 U.S. 38, 100 S.Ct. 2521, 65 L.Ed.2d 581 (1980):
[A] juror may not be challenged for cause based on his views about capital punishment unless those views would prevent or substantially impair the performance of his duties as a juror in accordance with his instructions and his oath. The State may insist, however, that jurors will consider and decide the facts impartially and conscientiously apply the law as charged by the court.
Id. at 45, 100 S.Ct. at 2526. Furthermore, as the Fifth Circuit reasoned in Spinkellink v. Wainwright, 578 F.2d 582, 597 (5th Cir.1978), cert. denied, 440 U.S. 976, 99 S.Ct. 1548, 59 L.Ed.2d 796 (1979),8 members of the venire who are irrevocably opposed to capital punishment would engage in jury nullification if permitted to sit at the guilt stage of a capital case and, therefore, would not be impartial fact finders.9 Under these cases, the state may thus exclude *134such jurors without violating a defendant’s right to a fair cross-section of the community. Accordingly, petitioner’s Sixth Amendment claim must fail.10
B. The Due Process Requirement
Petitioners also urged below and the district court held that the death-qualification process impermissibly creates conviction prone juries in violation of the Fourteenth Amendment’s due process clause. We do not agree. The fact that a death-qualified jury may be more conviction prone than one which includes jurors strongly opposed to the death penalty does not demonstrate which jury is impartial. Rather, “[i]t indicates only that a death-qualified jury might favor the prosecution and that a nondeath-qualified jury might favor the defendant.” Spinkellink v. Wainwright, 578 F.2d at 594.
The process of voir dire is designed to remove from the venire individuals who cannot be fair to either side of the case. As the Fifth Circuit recently stated in Smith v. Balkcom, 660 F.2d 573, 579 (5th Cir.1981), cert. denied, 459 U.S. 882, 103 S.Ct. 181, 74 L.Ed.2d 148 (1982):
The guarantee of impartiality cannot mean that the state has a right to present its case to the jury most likely to return a verdict of guilt, nor can it mean that the accused has a right to present his case to the jury most likely to acquit. But the converse is also true. The guarantee cannot mean that the state must present its case to the jury least likely to convict or impose the death penalty, nor that the defense must present its case to the jury least likely to find him innocent or vote for life imprisonment____ The logical converse of the proposition that death-qualified jurors are conviction prone is that nondeath-qualified jurors are acquittal prone, not that they are neutral.
In this case, as in Balkcom, petitioners urge in effect that they have a constitutional right to a jury more likely to find them innocent than a death-qualified jury. Because petitioners are entitled only to an impartial jury, we conclude that the exclusion of WE’s from their juries did not violate their fights under the due process clause.11
III.
The State further contends that the district court erred in finding that, in Williams’ case, prospective juror Nancy Melton was excluded for cause in violation of Witherspoon. We agree.
Under Witherspoon, a member of the venire may be stricken for cause if he *135is unwilling “to consider all of the penalties provided by state law.” 391 U.S. at 522 n. 21, 88 S.Ct. at 1777 n. 21. The determination of whether a juror may be stricken is committed to the trial court’s discretion. Ristaino v. Ross, 424 U.S. 589, 594-95, 96 S.Ct. 1017, 1020, 47 L.Ed.2d 258 (1976).
When questioned as to whether she could follow North Carolina law, which provided for imposition of the death penalty, Melton stated on several occasions that she was “not sure.” She also expressed doubt as to whether she could vote for the death penalty and live with [her] conscience, and admitted she had certain religious beliefs about the death penalty.12 No rehabilitation was attempted, and Melton was stricken for cause. The district court nevertheless concluded that Melton had been improperly stricken because her statements did not clearly indicate an inability or unwillingness to consider the penalties imposed by North Carolina law.
Unlike the district court, we can find no evidence that the state trial judge abused his discretion in determining that Melton should be stricken. It is not unlikely that the colloquialism “I’m not sure,” bears a different connotation in different parts of the country. Moreover, it is beyond dispute that it could mean either “yes,” or “no,” depending upon the inflection of the words and the demeanor of the speaker, The trial judge was in the best position to observe Melton and to determine the true meaning of the words she used in her answers to the court’s questions. AccordingJy> we conclude that the trial judge’s decislon exclude Melton cannot be the basis f°r habeas relief.13
IV.
For the foregoing reasons, we reverse the order of the district court granting habeas relief in these cases.
. From 1949 until 1972, this system provided that the jury’s recommendation of either life imprisonment or death was binding on the trial judge, who imposed the sentence. In 1973, the system was changed to require a sentence of death upon conviction for first-degree murder. During Keeten’s trial, however, North Carolina’s then existing death penalty statute was declared unconstitutional in Woodson v. North Carolina, 428 U.S. 280, 96 S.Ct. 2978, 49 L.Ed.2d 944 (1976), and, consequently, life imprisonment was substituted as the appropriate sentence in capital cases.
. Pursuant to § 15A-1212(8),
"[a] challenge for cause to an individual juror may be made by any party on the ground that the juror: ... [a]s a matter of conscience, regardless of the facts and circumstances, would be unable to render a verdict with respect to the charge in accordance with the law of North Carolina.”
. Five of these surveys indicated the following differences among these groups: (1) the 1970 Bronson non-random study showed a 40%-60% difference in conviction prone answers between strong opponents of the death penalty and all other groups, see, Bronson, "On the Conviction Proneness and Representativeness of the Death-Qualified Jury: An Empirical Study of Colorado Veniremen,” 42 Colo.L.Rev. 1 (1970); (2) the 1960’s Bronson non-random poll indicated a 19%-51% difference between strong opponents of the death penalty and all other groups, see, Bronson, "Does the Exclusion of Scrupled Jurors in Capital Cases Make the Jury More Likely to Convict? Some Evidence from California,” 3 Woodrow Wilson L.Rev. 11 (1980); (3) the 1970’s Bronson non-random poll indicated a 28%-49% difference between strong opponents *132and all others, see, id.; (4) the 1971 Harris Poll indicated a 46%-52% difference between death qualified jurors ("DQ's”) and WE’s, see, Louis Harris & Associates, Inc., National Opinion Survey Section of Study No. 2016 (1971); and (5) the 1978 Ellsworth/Fitzgerald study indicated a 38%-50% difference between WE’s and DQ's, see, Ellsworth and Fitzgerald, "Due Process vs. Crime Control: Death Qualification and Jury Attitudes,” 8 Law and Human Behavior, 31 (1984).
. A "death qualified juror" is a juror who will consider imposing the death penalty and who is currently permitted to serve in capital cases under Witherspoon.
. The mock jury studies came to the following conclusions: the Wilson study, completed in 1964, indicated, without giving a specific statistical analysis, that death penalty opponents convicted less frequently than jurors not opposing the death penalty, see, Cody Wilson, "Belief in Capital Punishment and Jury Performance,” unpublished (1964); (2) the Goldberg study, completed in 1966, indicated that jurors opposing the death penalty are 6% less likely to convict than jurors not opposing the death penalty, see, Goldberg, "Toward Expansion of Witherspoon: Capital Scruples, Jury Bias, and the Use of Psychological Data to Raise Presumptions in the Law,” 5 Harv.C.R.-C.L.L.Rev. 53 (1970); (3) the Jurow study, completed in 1970, indicated that WE’s are 15% less likely to convict than DQ’s, see, Jurow, “New Data on the Effects of a ‘Death-Qualified’ Jury on the Guilt Determination Process," 84 Harv.L.Rev. 567 (1971); (4) the Harris Poll, completed in 1971, indicated that WE’s are 7% less likely to convict than DQ’s, see, Louis Harris and Associates, Inc., Study No. 2016 (1971); and (5) the Ells-worth/Thompson/Cowan (ETC) study, completed in 1979, revealed that WE’s were 25% less likely to convict than DQ’s, see, Ellsworth, Thompson & Cowan, "The Effects of Death Qualification on Jurors’ Predisposition to Convict and on the Quality of Deliberation," 8 Law and Human Behavior (1984).
. The Sixth Amendment provides that ‘‘[i]n all criminal prosecutions, the accused shall enjoy the right to a speedy and public trial, by an impartial jury of the State and district wherein the crime shall have been committed____” In Taylor v. Louisiana, 419 U.S. 522, 95 S.Ct. 692, 42 L.Ed.2d 690 (1975), the Supreme Court held that a jury panel reasonably drawn from a fair cross-section of the community is fundamental to the constitutional right to a jury trial.
. Although we are willing to decide the case upon this assumption, we do not necessarily agree with petitioners’ view-point that attitudinal surveys imply a relationship between attitudes toward the criminal justice system and behavior as jurors. Both law and trial practice evidence a belief that attitudes can often be put aside by prospective jurors. Examples of this include permitting jurors to serve despite exposure to prejudicial pretrial publicity, Stroble v. California, 343 U.S. 181, 72 S.Ct. 599, 96 L.Ed. 872 (1952), and permitting jurors to serve despite employment affiliation with or affinity to the prosecution, Smith v. Phillips, 455 U.S. 209, 102 S.Ct. 940, 71 L.Ed.2d 78 (1982). Moreover, in Adams v. Texas, 448 U.S. 38, 49-50, 100 S.Ct. 2521, 2528-29, 65 L.Ed.2d 581 (1980), the Supreme Court stated that a juror is not impartial in an unconstitutional sense just because a general attitude "affects" his decision-making process.
. The reasoning in Spinkellink has already been adopted within this Circuit. See Barfield v. Harris, 540 F.Supp. 451, 464 (E.D.N.C.1982), affirmed in Barfield v. Harris, 719 F.2d 58 (4th Cir.1983).
. The research that exists in this area supports this conclusion. See, Harris Poll, supra, note 3; ETC study, supra, note 3; and Bronson study, supra note 2. In Harris, 56% of the WE’s would not vote to convict a person they believed guilty, even where they could also vote “no” on the death penalty at the sentencing phase and thereby prevent its imposition. Additionally, 38% of them would nullify if sentencing were in the judge’s discretion; and 57% would do so if the death penalty were automatic.
. We find it unnecessary to address the issue of whether the death-qualification process violates the Sixth Amendment by disproportionately excluding women and blacks. This claim, if colorable at all, can only be raised by a prisoner who has allegedly been injured by a disproportionately low number of women and/or blacks on his jury. Here, the fair cross-section requirement with regard to women and blacks was met in each of the cases before the Court: Keeten had six female jurors, Avery had eight, and Williams had five; in addition, Keeten had a 25% black jury against an expectation of a 16% black jury, Avery had an 8% black jury against an expectation of 15%, and Williams had an 8% black jury against an 8% expectation. Because no substantial disproportion has been shown, this claim is an inappropriate basis for relief.
. In addition to claiming that the death-qualification process created conviction-prone juries in violation of the due process clause, petitioners also claimed below that the State should have been prohibited from questioning prospective jurors about their death penalty views prior to the guilt/innocence phase of the trial, due to the biasing effect of the questions. The district judge did not address this latter complaint separately from petitioners' general due process argument. On appeal, respondents complain that it should have been ruled on as an independent issue. We disagree.
Even if this complaint were viewed separately, however, we would not find it a sufficient ground upon which to grant petitioners relief, as the sole study upon which this argument is based is, in the words of the' Supreme Court, far "too tentative and fragmentary" to establish a legal conclusion.
Moreover, were this complaint viewed separately, the Supreme Court's decision in Rose v. Lundy, 455 U.S. 509, 102 S.Ct. 1198, 71 L.Ed.2d 379 (1982), would preclude both Keeten and Avery from raising it, as they have not exhausted their state court remedies as to this particular issue.
. The following colloquy took place between the trial judge and Melton:
Q. Now, if you serve as a juror in this case, could and would you if called upon to do so make a sentence recommendation of life imprisonment or death in accordance with the law ... or would you be unable to do so regardless of the law ... because of some conscientious belief as to the proper punishment for first-degree murder; that is, that you conscientiously feel that it should in all cases be life in prison or you feel in all cases it should be the death penalty?
A. Your Honor, I’m not sure I could say that someone else had to die. I’m not sure I could do that.
Q. [T]his is something we have to determine at this stage — whether you if you serve as a juror could follow the law of North Carolina and if you are satisfied beyond a reasonable doubt of those things which the law requires you to be satisfied you could then return a recommendation of the death penalty.
A. I’m not positive I could do that. I’ve never been called on to do that, and I'm not sure that I could live with my conscience.
Q. Well, do you have conscientious beliefs about the death penalty — religious beliefs about it?
A. Yes.
Q- And you do not feel that you could follow the instructions of the court if you were satisfied beyond a reasonable doubt of the things of which you must be satisfied. If those conclusions would call for the death penalty, you don’t feel you could make such a recommendation?
A. I’m not sure that I could,
. Our holding today is in no way inconsistent with Adams v. Texas, 448 U.S. 38, 100 S.Ct. 2521, 65 L.Ed.2d 581 (1980). In Adams, the Supreme Court specifically stated that "neither nervousness, emotional involvement, nor inability to deny or confirm any effect whatsoever is equivalent to an unwillingness ... to follow the court's instructions.” Id. at 50, 100 S.Ct. at 2529 (emphasis added). Here, after observing Melton and listening to her testimony, the trial judge decided that, rather than being unable "to deny or confirm” her convictions, that Melton had, in fact, expressed an unwillingness to consider the death penalty.