(concurring). I join in the majority opinion, which reasonably applies the test governing the Sixth Amendment’s “fair cross section” requirement, as articulated by the United States Supreme Court in Duren v Missouri, 439 US 357, 364; 99 S Ct 664; 58 L Ed 2d 579 (1979). I write separately only because I have questions concerning both Duren’s test and the constitutional standard toward which this test is directed.
The Sixth Amendment guarantees criminal defendants the right to a trial “by an impartial jury . ...” In Taylor v Louisiana, 419 US 522, 526; 95 S Ct 692; 42 L Ed 2d 690 (1975), the Supreme Court determined that “the presence of a fair cross section of the community on venires, panels, or lists from which petit juries are drawn is essential to the fulfillment” of this constitutional guarantee. The “fair cross section” requirement is satisfied as long as “distinctive” groups are reasonably represented on the jury venire; however, it does not entitle a defendant to a jury whose composition is proportional to that group’s presence within the community from which the venire is chosen. As Taylor emphasized:
Defendants are not entitled to a jury of any particular composition, but the jury wheels, pools of names, panels, or *620venires from which juries are drawn must not systematically exclude distinctive groups in the community and thereby fail to be reasonably representative thereof. [Id. at 538 (citations omitted).]
Under the “fair cross section” analysis, it is unnecessary for a defendant to show that the lack of “reasonable representation” of a “distinctive” group is the result of discrimination in the jury-selection system, as would be required under the Equal Protection Clause of the Fourteenth Amendment. See, e.g., Castaneda v Partida, 430 US 482, 494; 97 S Ct 1272; 51 L Ed 2d 498 (1977). Rather, in “fair-cross-section cases, systematic disproportion itself demonstrates an infringement of the defendant’s interest in a jury chosen from a fair community cross section.” Duren, 439 US at 368 n 26 (emphasis added). So the critical constitutional inquiry appears to be directed toward the extent or magnitude of the “systematic disproportion.” While “proportional” representation of “distinctive” groups is not required, what constitutes “proportional” representation must nonetheless be constantly borne in mind so that the level of “disproportion” can be calculated because, at some uncertain point, a level of “disproportion” that is apparently constitutionally acceptable is transformed into a level of “disproportion” that breaches the Sixth Amendment. And it is the responsibility of this Court to determine on a “case by case” basis when that point of transformation occurs, principally through the application of myriad statistical tests, some of which have been given the explicit imprimatur of the United States Supreme Court and others of which have not, but at the same time have not been repudiated, in light of the apparent nonexclusivity of the approved tests.
In Duren, the Supreme Court set forth a three-part test to evaluate “fair cross section” challenges. Specifically, in order to establish a prima facie case, a defendant must show
*621(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.]
The dispositive question in this case concerns the second part of Duren’s test — i.e., whether the representation of African-Americans in venires from which juries were selected in Kent County during the period in which defendant was tried and convicted is “fair and reasonable in relation to the number of such persons in the community.” Id.
To determine whether representation is “fair and reasonable” under the Duren test, courts have applied yet more tests. In People v Smith, 463 Mich 199; 615 NW2d 1 (2000), this Court discussed three statistical tests that have been used to measure whether representation of a “distinctive” group in the jury pool is “fair and reasonable”: the “absolute disparity” test, the “comparative disparity” test, and the “standard deviation” test. Recognizing that all three tests are imperfect and susceptible to criticisms, Smith held:
We thus consider all these approaches to measuring whether representation was fair and reasonable, and conclude that no individual method should be used exclusive of the others. Accordingly, we adopt a case-by-case approach. Provided that the parties proffer sufficient evidence, courts should consider the results of all the tests in determining whether representation was fair and reasonable. [Id. at 204.]
After a decision on habeas corpus review by the United States Court of Appeals for the Sixth Circuit asserting that Smith constituted an “unreasonable” application of “clearly established federal law,” Smith v *622Berghuis, 543 F3d 326, 329, 334 (CA 6, 2008), the United States Supreme Court unanimously reversed the Sixth Circuit, holding that “neither Duren nor any other decision of this Court specifies the method or test courts must use to measure the representation of distinctive groups injury pools.” Berghuis v Smith, 559 US 314, 329; 130 S Ct 1382, 1393; 176 L Ed 2d 249 (2010). Noting that “[e]ach test is imperfect,” the Supreme Court declined “to take sides today on the method or methods by which underrepresentation is appropriately measured.” Id. at 329-330.
Given this state of the law, I join the majority opinion because it engages in a reasoned application of the relevant decisions of the United States Supreme Court and this Court. Consistently with the approach outlined in our decision in Smith, the majority opinion considers the results of all three tests for which the parties have proffered evidence in determining whether the representation of African-Americans, the “distinctive” group in question, was “fair and reasonable” in Kent County venires. Specifically, the majority opinion considers the results of the “absolute disparity” and the “comparative disparity” tests, as well as those of an additional test, the “disparity of risk” test,1 and concludes that the results are insufficient to warrant a finding that African-American representation in the venires during the relevant period was not “fair and reasonable.” Thus, defendant has failed to establish a prima facie violation of the Sixth Amendment’s “fair cross section” requirement. Although the Court of Appeals’ opinion and the dissenting opinions of this Court also, in my *623judgment, reflect reasonable efforts to apply Duren, their use of only the results from the “comparative disparity” test to ascertain a “fair cross section” violation, their decisions not to use data from multiple venires over time, and their decisions not to fully consider the results of the “absolute disparity” test cause me to prefer the majority’s analysis. See Smith, 463 Mich at 204 (“[N]o individual method should be used exclusive of the others.”).
That said, the fact that both sides have sought reasonably and in good faith to apply Duren underscores questions concerning Duren’s test itself. These largely arise from the sense that in applying Duren, this Court seems to me engaged more in the judicial equivalent of a Rorschach test, an essentially standardless inquiry in which judicial conclusions are indicative more of personal judgments concerning the “fairness and reasonableness” of the Kent County venire than in the application of any discernible constitutional command.
In particular, I am concerned about the statistical tests used to determine whether Duren’s second part has been satisfied. The limitations of these tests have been widely noted, see, e.g., Berghuis, 559 US at 329; ante at 603-611; post at 637-638, and need not be revisited here. It suffices to say that when, as here, members of the “distinctive” group comprise only a relatively small percentage of the community’s jury-eligible population, one test arguably makes it difficult for a defendant to ever satisfy the requisite showing of “underrepresentation,” another arguably exaggerates this “underrepresentation,” and the third appears to be generally disfavored because it does not constitute an appropriate measure of anything obviously relevant to a determination whether the level of *624representation on the venires was “fair and reasonable.” In light of these deficiencies, how do the results of these tests, either considered individually or collectively, usefully illuminate whether representation was “fair and reasonable”? How do the bench and bar draw a meaningful legal conclusion from the application of these tests to the available statistical data? How rational, and how flexible, are the statistical thresholds that have been established by some courts in distinguishing between “underrepresentations” that are compatible with a “fair cross section,” and those that are not? To what extent, if any, may these thresholds be raised or lowered, as a function of the nature or the degree of any “systematic exclusion” under the third part of Duren, or must these parts be analyzed entirely discretely? To what extent, if any, should these thresholds be raised or lowered by Fourteenth Amendment considerations of discriminatory purpose or intention, or are those considerations simply irrelevant to the “fair and reasonable” analysis under the Sixth Amendment? To what extent are these thresholds emblematic of what the dissent in Duren predicted would become a mere “constitutional numbers game,” Duren, 439 US at 375 (Rehnquist, J., dissenting), or do these effectively communicate some independent reality as to what is required by the Constitution? Existing statistical tests and thresholds certainly provide one means by which to address Duren’s second part, but is it now the law of the land that the fate of criminal defendants, such as Ramon Bryant, as well as the effectiveness of the criminal justice system in communities, such as Kent County, in maintaining the security of their citizens are to be determined as a function of whether the data emerging from a host of statistical tests are to be “rounded up” or “rounded down,” the number of decimal points considered, and whether the denominator *625reflecting the presence of the “distinctive” group within the community has been determined by the most recent census figures, by mid-census estimates, or by the latest moving-van rental figures? In summary, which specific statistical tests best communicate whether jury representation of “distinctive” groups is constitutionally “fair and reasonable,” and under what circumstances, and by the application of which thresholds of deviation from the “proportional” representation standard? If the fate of individual defendants, and the ability of individual communities to carry out the enormous responsibility of protecting their citizens from criminal predators, is to be dependent on statistical testing, then there should be no uncertainty regarding either the relevance of a particular test in a particular circumstance or the standards for assessing, and thereby according legal and constitutional significance to, the results of those tests.
However, perhaps an even more fundamental question is also raised here — why certain statistical tests and not others? The United States Supreme Court has acknowledged that the three tests described in Smith are each imperfect, Berghuis, 559 US at 329, and has declined “to take sides today on the method or methods by which underrepresentation is appropriately measured,” id. at 329-330. Doubtless, there is no end to statistical tests by which a court might seek to compare various-sized populations of “distinctive” groups within a community and their representation on venires. Equally doubtless, as evidenced in this very case, tests can be devised that will tend both toward sustaining and repudiating a finding of “underrepresentation.” Is the new “disparity of risk” test a genuinely valid means of adducing the existence of a Sixth Amendment violation, or are the dissenting justices correct that it “neither improves nor *626clarifies this area of the law”? Post at 633. What are the standards by which this Court can discern which tests are relevant in identifying Sixth Amendment violations? And what is the relevance of the fact that some tests might point in one direction regarding the second Duren part, and others might point in the opposite direction? Does this suggest that these tests are asking and answering different questions, or that one test is asking and answering the wrong question? How do judges test the tests to ensure that the right question is being asked? Wdien tests differ in their results, how are these results to be reconciled in answering the ultimate constitutional question? May the court compare and contrast the degree or extent to which different tests deviate from thresholds distinguishing acceptable and unacceptable levels of statistical disparity? Is the court simply free to choose at its discretion among such conflicting tests? If there is some actual decision-making standard in selecting among conflicting tests, what is it? If such a standard has anything to do with determining which test better identifies “fair and reasonable” representation of “distinctive” groups in venires, then is this not a Catch-22 tautology, to wit, in choosing among tests that best identify the absence of “fair and reasonable” representation, a court must employ the test that best identifies “fair and reasonable” representation? What if multiple tests are applied, as in the instant case, and these produce split results of 2-1 or 3-1 or 7-6 in favor of the plaintiff or the defendant? Is there some “majority rule” that requires that we resolve conflicts in favor of the outcome of the majority of statistical tests applied? If so, does this not render all-important the court’s initial determination of which tests are going to be considered, and how that is to be determined? And if the “majority rule” does not apply, how do courts distinguish among conflicting tests *627in determining which of these will be dispositive in concluding that the Sixth Amendment has or has not been breached?2
These and related questions concern the meaning of Smith’s directive that courts must “consider all. . . approaches to measuring whether representation was fair and reasonable . . . Smith, 463 Mich at 204 (emphasis added). Indeed, very different conceptions of this obligation are reflected in the majority and dissenting opinions. Justice MARILYN KELLY argues that the Court of Appeals below “properly considered the results of all tests [including the absolute-disparity test], but decided that the comparative disparity test was ‘the most appropriate test to measure underrepresentation in this case,’ ” post at 638, quoting People v Bryant, 289 Mich App 260, 271; 796 NW2d 135 (2010), while the majority concludes that Smith requires more than simply alluding to a test and then failing to “consider” it. One might think that such a difference of opinion could be easily resolved if there were some clear sense regarding why a particular test is or is not “appropriate” in furthering our understanding of whether “fair and reasonable” representation has been achieved, which, of course, would require a clear understanding of what is meant by “fair and reasonable” representation, which in turn would require a clear understanding of whether the constitutional task at hand is simply to calculate the divergence of actual representation on the venire from the “ideal” of proportional representation (an ideal *628certainly implied by the Duren concept of “underrepresentation”)3 and then apply some statistical equivalent of the “I know it when I see it” test once articulated by former United States Supreme Court Justice Potter Stewart in the realm of obscenity law. Jacobellis v Ohio, 378 US 184, 197; 84 S Ct 1676; 12 L Ed 2d 793 (1964) (Stewart, J., concurring).4
That is, even if I could clearly answer each of the aforementioned questions, and knew which tests to “consider” and how to give legal import to their results, it still would be difficult to apply Duren because the ultimate constitutional standard to which it is directed remains unclear. I know what the constitutional standard in “fair cross section” cases is not. It is not an equal-protection standard under which any “underrep*629resentation” resulting from intentional or purposeful discrimination in the jury venire is prohibited by the Constitution. See Taylor, 419 US at 526-528; Duren, 439 US at 368 n 26. And it is not a “proportional” representation standard under which any systematic exclusion that results in the “underrepresentation” of a “distinctive” group is prohibited by the Constitution, Taylor, 419 US at 538; Duren, 439 US at 364 — although references by the Supreme Court to “underrepresentation,” “disproportion,” and “representative[ness]” would certainly cause some judges to look in precisely that direction, absent the Court’s admonition to the contrary. Thus, the “fair cross section” requirement, which purports to eschew both principles of nondiscrimination and proportional representation, must be premised on some alternative standard drawn from the Sixth Amendment’s guarantee of an “impartial jury.” See Taylor, 419 US at 526. But see Berghuis, 559 US at 334; 130 S Ct at 1396 (Thomas, J., concurring) (arguing that the fair-cross-section requirement “rests less on the Sixth Amendment than on an ‘amalgamation of the Due Process Clause and the Equal Protection Clause of the Fourteenth Amendment’ ”) (citations omitted).
Thus, some amount of “underrepresentation” is “fair and reasonable” and some amount is not, and the courts are to choose where the line is to be drawn. The problem is not that judges are ill equipped to determine what is “fair and reasonable,” as such inquiries are made daily by judges in other constitutional contexts. In the context of the Duren test, however, there is no agreed-upon standard or “ideal” by which to measure the constitutional mandate of an “impartial jury” at the venire stage. The United States Supreme Court has stated that proportionality constitutes one relevant measure, but that it is also not required in order to satisfy the Constitution, which leads to uncertainty *630because there is also no agreed-upon standard by which to measure how close a venire must come to proportionality, or indeed even how one measures proportionality. What “systemic,” but nondiscriminatory, deviations are acceptable under the Constitution, and what “systemic,” but nondiscriminatory, deviations are not? Given this lack of clear external standards and the wealth of divergent statistical measurements available, how can a judge ensure that his or her own private sensibilities concerning what is “fair and reasonable” in the make-up of the venire do not come to prevail over what is required by the Constitution and that statistical tests do not come to be selected, and standards for evaluating their results not come to be adopted, that merely tend to match those sensibilities? Focusing exclusively on the merits of the various tests obscures the forest for the trees, for without some clear sense of what the constitutional guarantee of an “impartial jury” requires at the venire stage, it will prove difficult, if not impossible, to achieve uniformity in the analyses of the composition of different venires. This, in turn, incurs the risk of making judicial determinations regarding “impartial juries” fraught with partiality and mathematical gamesmanship, while lending credence to Justice Rehnquist’s concerns about a “constitutional numbers game.” Duren, 439 US at 375 (Rehnquist, J., dissenting).
In the end, many trial and appellate judges have reviewed this case, and the question whether the venire here was “fair and reasonable” has closely divided them in favor of a negative response. While I have no doubt that each of these judges has addressed the question in this case “fairly and reasonably,” and in accordance with his or her own best understanding of Duren, there seems to be little in the way of a coherent constitutional standard that distinguishes between “systematic exclu*631sions” that violate the Constitution and “systematic exclusions” that do not, much less a clear statistical method for giving effect to this constitutional standard. And as a result, I believe that our decision-making in this realm resembles uncomfortably a judicial Rorschach test, in which the judge is ultimately required to look inward in determining what is “fair and reasonable,” rather than outward to a comprehensible constitutional rule of law. In joining the majority opinion, and despite what I believe to be confusion concerning aspects of the Duren test, I have sought to the best of my understanding of what is required by the United States Supreme Court to give reasonable meaning to this test and to the guarantees of the Constitution.
The majority opinion does not analyze the results of the “standard deviation” test because the only expert whom the trial court found credible, Dr. Paul Stephenson, testified that the test was “not appropriate” in the present circumstances.
While these questions may seem a mere quibble to some, when judges are free to pick and choose among disparate tests, pointing to disparate constitutional conclusions, defendants and communities that are not significantly disparate may end up being treated in a disparate manner as a function of the judicial decision-making involved in: (a) choosing appropriate tests; (b) evaluating or considering such tests; and (c) reconciling such tests when they produce conflicting results.
See also Duren, 439 US at 372 n * (Rehnquist, J., dissenting) (observing that if the fair-cross-section requirement “were truly an essential element of the due process right to trial by an impartial jury, a defendant would be entitled to a jury composed of [distinctive groups] in perfect proportion to their numbers in the community”). For a similar perspective, see Leipold, Constitutionalizing jury selection in criminal cases: A critical evaluation, 86 Geo L J 945, 965 (1998):
A defendant is thus placed in a strange position: he is entitled to a jury drawn from a fair cross section specifically because it increases the odds that different groups and perspectives will be represented in the jury pool, which in turn helps ensure that the panel is impartial; when actually seating a jury, however, he may not take those same characteristics into account. He may not base his peremptory strikes on the very same proxy for viewpoints that the Court has already used to justify the cross-section requirement, even if his efforts are designed to bring about the exact benefit that the cross-section requirement provides. An attempt to support the cross-section requirement on impartiality grounds thus runs headlong into the rule that race or gender may not be used as a substitute for inclinations, biases, or possible votes.
As did Justice Stewart, I also recognize that both this Court and the United States Supreme Court are quite possibly faced here with “the task of trying to define what may be indefinable.” Jacobellis, 378 US at 197 (Stewart, J., concurring).