Hines, Justice.
In accordance with the Unified Appeal Procedure, this Court granted the application for interim review to address pretrial issues in this case. Brandon Dwayne Smith is charged with malice murder and other crimes in Hall County, and the State is seeking the death penalty. Smith filed a challenge to the grand and traverse jury lists in Hall County, claiming that Hispanics are underrepresented in violation of the fair-cross-section requirement of the Sixth Amendment. After a hearing in May 2001 lasting several days, the trial court ruled against him on his challenge to the grand jury list and in his favor on his challenge to the traverse jury list. On appeal, we requested the parties address the following:
(1) Whether the trial court correctly found that Hispanics are a distinctive group under a Sixth Amendment analysis in Hall County.
(2) Whether the trial court correctly found that the Hall County jury commission’s use of 1990 Census statistics to compile the grand jury list applicable to this case was proper.
(3) Whether the trial court correctly found with regard to the traverse jury list that Smith had satisfied the second prong of the Sixth Amendment test by showing an over-18 Hispanic population of 17.1% without requiring Smith to show that these individuals were citizens and therefore eligible to serve on the jury.
(4) Whether, under a Sixth Amendment analysis, the trial court correctly found that Smith had met his burden of showing systematic exclusion of Hispanics on the traverse jury list by showing that Hispanics had not been identified historically on the list and the efforts of the jury commission had been insufficient to meet constitutional requirements.
(5) Whether the trial court correctly found that the State had failed to rebut Smith’s prima facie showing of a Sixth Amendment violation with the traverse jury list because the State did not present reliable evidence to show that many Hall County Hispanics counted in the 2000 Census are not U. S. citizens and ineligible for jury service.
[716] THE LAW AND THE TRIAL COURT’S ORDER
1. This Court has addressed the issue of Hall County Hispanics and their representation in Hall County jury pools before. In Mobley v. State, 262 Ga. 808 (2) (426 SE2d 150) (1993) and Mobley v. State, 265 Ga. 292 (6) (455 SE2d 61) (1995), this Court affirmed the trial court’s orders that Mobley had failed to prove that Hispanics in Hall County were a distinctive group or that they were underrepresented on the jury lists. The issue arose again in Morrow v. State, 272 Ga. 691 (1) (532 SE2d 78) (2000). In Morrow, this Court outlined the three-part test to be used for determining the existence of a prima facie Sixth Amendment fair-cross-section violation, which is: (1) whether the group alleged to be excluded is a “distinctive” group in the community; (2) whether the representation of this group in jury pools is not fair and reasonable in relation to the number of such persons in the community; and (3) whether this underrepresentation is due to systematic exclusion of the group in the jury selection process. Morrow, supra at 692; Duren v. Missouri, 439 U S. 357, 364 (II) (99 SC 664, 58 LE2d 579) (1979); Bowen v. Kemp, 769 F2d 672, 684 (11th Cir. 1985). The defendant has the burden of proving a prima facie case of constitutional error in the composition of the challenged jury pool. Morrow, supra at 693; Berryhill v. Zant, 858 F2d 633, 638 (11th Cir. 1988). Morrow claimed that Hispanics were underrepresented on the grand and traverse jury lists created in the middle and late 1990’s because the large influx of Hispanics into Hall County since 1990 had rendered racial/ethnic percentages in the 1990 Census obsolete. The jury lists must be based on the most recent census statistics for the county. Unified Appeal Procedure Rule II (C) (6) (b). Morrow argued that the trial court should have ordered the county to instead use population estimates derived from his “test census” of a small portion of the county and other estimates of Hispanics in Hall County, which allegedly showed much higher numbers of Hispanics. This Court affirmed the trial court’s refusal to accept Morrow’s statistics, finding them to be unreliable when compared to the more-comprehensive 1990 Census. Morrow, supra at 694-695. When using the 1990 Census numbers, the absolute disparity of Hispanics on the grand and traverse jury lists when compared with their percentage of the county population was less than 5% and therefore constitutional. See Morrow, supra at 695; Cook v. State, 255 Ga. 565 (11) (340 SE2d 843) (1986) (“As a general proposition, absolute disparities under 10% usually are sufficient to satisfy constitutional requirements.”). This Court did not reach the question of whether the constitutional analysis was affected by evidence that most of the Hispanics in Hall County are not U. S. citizens and therefore ineligible for jury service. Morrow, supra; OCGA § 15-12-40.1.
[717] Smith challenged the grand and traverse jury lists on the same Sixth Amendment grounds as did Morrow. In Smith’s case, he has been indicted by a grand jury drawn from a grand jury list created in March 2000 using the 1990 Census numbers. The 2000 Census numbers, showing an increased Hispanic population, were not available to the jury commission until March 2001, about eight months after Smith’s indictment. After determining that Hall County Hispanics are a distinctive group, the trial court ruled, by following this Court’s holding in Morrow, that the 1990 Census numbers used by the jury commission when creating the grand jury list were the only reliable population measurement then available, and the jury commission did not err by using them. Using 1990 Census numbers, the absolute disparity of Hispanics between their percentage of the county population and their representation on the grand jury list was within constitutional limits. Thus, the trial court refused to quash the indictment.
The trial court viewed the traverse jury list differently. It determined that the 2000 Census numbers would be available to the jury commission when it compiles Smith’s traverse jury list, and that the master trial jury list1 had only 2.6% Hispanics2 while the 2000 Census showed Hispanics over the age of 18 to be 17.1% of the over-18 county population. The trial court found the resulting absolute disparity of 14.5% to be unconstitutional. With regard to the third prong of the Sixth Amendment test, whether the underrepresentation was due to systematic exclusion of Hispanics in the jury selection process, the trial court found the “most compelling evidence” of systematic exclusion was that Hispanics were not tracked as a separate category on the traverse jury list or recorded on jury application and voter registration forms; historically, there were only identity blocks for male, female, white, black, and other. The trial court acknowledged that the jury commissioners had tried to recruit more Hispanics for the jury list but found these efforts insufficient to preclude a finding of systematic exclusion. Therefore, the trial court found that Smith had proven a prima facie Sixth Amendment fair-cross-section violation. Although evidence from both parties showed that most Hall County Hispanics are not U. S. citizens, and therefore ineligible for jury service, the trial court faulted the State for failing to provide reliable citizenship numbers to rebut Smith’s prima facie case. The [718] trial court also found that the State could have rebutted Smith’s prima facie case by presenting more evidence that the jury commission had tried to increase the number of Hispanics on the jury list but failed due to lack-of-citizenship problems. The trial court granted Smith’s challenge to the traverse jury list and ordered the jury commission to recompile a traverse jury list that accurately reflected Hall County’s population.
THE DISTINCTIVENESS OF HALL COUNTY HISPANICS
2. The first question posed to the parties was whether the trial court correctly determined that Hispanics in Hall County are a distinctive group under a Sixth Amendment analysis. To show that a group is distinctive under the Sixth Amendment, a defendant must prove:
(1) that the group is defined and limited by some factor . . . ; (2) that a common thread or basic similarity in attitude, ideas, or experience runs through the group; and (3) that there is a community of interest among members of the group such that the group’s interests cannot be adequately represented if the group is excluded from the jury selection process. [Cit.]
Potts v. State, 259 Ga. 812 (1) (388 SE2d 678) (1990). Whether a group is sufficiently distinct under the Sixth Amendment fair-cross-section analysis is a question of fact. Id. At the hearing, almost every witness for the defendant and for the State testified that they believed Hispanics are a distinctive group in Hall County. Witnesses identified characteristics shared by Hall County Hispanics such as origin in Mexico or South or Central America, the speaking of Spanish, professing the Roman Catholic faith, a strong work ethic, and strong family traditions. Many courts have determined that Hispanics are a distinctive group for Sixth Amendment purposes. See, e.g., United States v. Lara, 181 F3d 183, 192, n. 1 (1st Cir. 1999); United States v. Rioux, 97 F3d 648, 654 (2nd Cir. 1996); United States v. Esquivel, 88 F3d 722, 726 (9th Cir. 1996); United States v. Garcia, 991 F2d 489, 491 (8th Cir. 1993). The trial court did not err by finding that Hall County Hispanics are a distinctive group. See id.
THE GRAND JURY LIST
3. The trial court correctly denied Smith’s motion to quash his indictment after finding that the Hall County jury commission did not err by using 1990 Census numbers to compile the 2000 grand jury list. The Hispanic population in Hall County, and the overall [719] county population, grew considerably from 1990 to 2000. The 1990 Census showed that there were 3,252 Hispanics over age 18 out of a total over-18 county population of 70,969 (4.6%). See Morrow, supra at 695. The 2000 Census showed that there are 17,424 over-18 Hispanics out of a total over-18 county population of 101,760 (17.1%). However, the 2000 Census numbers showing an increased percentage of Hispanics did not become available until a year after the Hall County jury commission compiled the grand jury list in March 2000. Smith was indicted in June 2000. Smith claims that the jury commission was aware of the increasing Hispanic immigration into Hall County and should not have relied on the 1990 Census. Instead, he argues that the jury commission should have deviated from the 1990 Census numbers, estimated a larger number of Hispanics in Hall County, and raised the percentage of Hispanics on the grand jury list.
This is essentially the same argument that failed in Morrow. The Unified Appeal Procedure requires that the group percentages on the grand jury list be based on the county statistics from the most recent decennial census. Unified Appeal Procedure Rule II (C) (6) (b). Obviously, the census conducted by the federal government is not perfect, and county populations are not static. People of all groups constantly move into and out of counties, especially during growth periods such as that currently being experienced in the metropolitan Atlanta area. But the census, a comprehensive county-wide head count, is clearly more accurate than case-by-case population estimates like those found to be unreliable in Morrow. Courts and jury commissions need a valid population benchmark upon which to calculate the appropriate group percentages on the jury list, and that benchmark cannot be a moving target. Smith has not shown how interim population estimates would be more accurate in the compilation of the percentages on the jury list. Indeed, arbitrarily adjusting the group percentages on the grand jury list to accommodate a particular defendant would invite fair-cross-section challenges to the list by other defendants. When compiling the March 2000 grand jury list, the Hall County jury commission properly relied on the most recent decennial census numbers that were available to it. Morrow, supra at 694-695; Unified Appeal Procedure Rule II (C) (6) (b). The absolute disparity between the percentage of Hispanics on the 2000 grand jury list and the percentage of jury-eligible Hispanics in the county was within constitutional bounds. See Morrow, supra at 695; Cook, supra.
[720] THE TRAVERSE JURY LIST
4. The third question posed to the parties concerned the second prong of the Sixth Amendment test: whether the representation of Hispanics in the Hall County traverse jury pool is not fair and reasonable in relation to the number of Hispanics in Hall County. Duren, supra. “Generally speaking with regard to the second prong ... , an absolute disparity between the percentage of a group in the population and its percentage in the jury pool of less than 5% is almost always constitutional; an absolute disparity between 5 and 10% is usually constitutional; and an absolute disparity of over 10% is probably unconstitutional.” Morrow, supra at 692. The trial court determined that the 2000 Census, which will be available to the jury commission when it compiles Smith’s traverse jury list, showed that 17.1% of the over-18 county population is Hispanic. Only 2.6% of the master trial jury list is Hispanic. The trial court found the resulting 14.5% absolute disparity sufficient to satisfy the second prong of the Sixth Amendment test. The State argues that it was error for the trial court to use total over-18 population figures for Hispanics in the county, instead of jury-eligible figures, because the evidence showed that the overwhelming majority of Hall County Hispanics are not U. S. citizens and therefore are ineligible to serve on juries. OCGA § 15-12-40.1 (no person is qualified to serve on a jury unless that person is a U. S. citizen). The extent and effect of any alleged underrepresentation are mixed questions of law and fact: the extent of underrepresentation is a question of fact to be determined by the trial court, and the effect of the disparity, once its extent has been determined, is a question of law. Morrow, supra at 693. “With mixed questions of fact and law, this Court accepts the trial court’s findings on disputed facts and witness credibility unless clearly erroneous, but independently applies the legal principles to the facts.” Id.
Witnesses testified that Hispanic immigration into Hall County, primarily from Mexico, accelerated after 1990 and greatly accelerated after 1995. Like the immigration of other ethnic groups into this country, the first arrivals were often males who obtained jobs at higher pay than they were able to obtain in their native country. They worked for limited periods of time, sent money home to their families, and then returned to Mexico. Eventually, as their earnings increased, their stays became longer and some sent for their families and settled permanently. Word spread among their former neighbors in Mexico that Hall County was a good place to live and work, and thus the cycle continued.
Witnesses for both Smith and the State testified that few of these new arrivals have obtained U. S. citizenship. This is not surprising, because one of Smith’s experts testified about national sta[721] tistics that show only 23.9% of the immigrants who arrived in this country between 1980 and 1989 have become U. S. citizens, and only 6.7% of the immigrants who have arrived since 1990 have become citizens. This expert, Dr. Bohon, a sociology professor at the University of Georgia, further testified that Hall County Hispanics are disproportionately immigrant and that the fastest rate of Hispanic immigration into Hall County occurred after 1995. She estimated that “at least 10%” of the Hispanics in Hall County are U. S. citizens.3 Father Jorge Christancho, a Catholic priest in Hall County whose church has a large number of Hispanic parishioners, testified that the language barrier makes first-generation Hispanic immigrants slow to become assimilated into the non-immigrant community and that there is not a large enrollment in citizenship classes. Richard Beamish, an Immigration and Naturalization Service agent assigned to Hall County, testified that he believed about 5% of Hall County Hispanics are U S. citizens. Cecilia Perra, a bilingual Hispanic resident of Hall County and a jury commissioner from 1990 to 1996, testified that she tried to recruit Hispanics for the jury list at her predominantly-Hispanic church by passing out jury list application forms, but “a lot of them just gave them back, [and] said they weren’t citizens.” The jury list application form contains a self-affirming oath attesting to U. S. citizenship; the voter registration form also contains this citizenship requirement. The voters’ registration list was the primary source of names for the jury list.
When alleging underrepresentation of a distinctive group, a defendant “must, to establish a prima facie case, present data showing that the percentage of persons in that group [on the jury list] is significantly lower than the percentage eligible to serve on juries.” (Emphasis supplied.) United States v. Artero, 121 F3d 1256, 1262 (9th Cir. 1997). “[A] comparison of percentages in [the jury pool] and ‘the gross population’ is ‘irrelevant,’ because ‘the pertinent inquiry is the pool of [the group claimed to be underrepresented] in the district who are eligible to serve as jurors.’ [Cit.]” Artero, supra. See also United States v. Grisham, 63 F3d 1074, 1078 (11th Cir. 1995) (“To examine the second element [of the Sixth Amendment fair-cross-section test], we must compare the difference between the percentage of the distinctive group among the population eligible for jury service and the percentage of the distinctive group on the (jury list].”); United States v. Pion, 25 F3d 18, 23, n. 5 (1st Cir. 1994); United [722] States v. Rodriguez, 776 F2d 1509, 1511, n. 6 (11th Cir. 1985). “The decisions of this Court suggest, and common sense demands, that eligible population statistics, not gross population figures, provide the relevant starting point.” Castaneda v. Partida, 430 U. S. 482, 504 (97 SC 1272, 51 LE2d 498) (1977) (Burger, C. J., dissenting).
The Ninth Circuit Court of Appeals confronted an almost identical situation in Artero, supra at 1260-1262, as the case now before us. Artero argued that the grand jury that indicted him was not drawn from a fair cross-section of the community because Hispanics were underrepresented on the grand jury list for the Southern District of California. He claimed that Hispanics comprised 24.2% of the population of that district, but only 9.7% of the names on the jury wheel. This resulted in an absolute disparity of 14.5%. The Ninth Circuit determined that Artero’s numbers were insufficient to establish a prima facie Sixth Amendment fair-cross-section violation because he did not account for the large number of southern California Hispanics who were not U. S. citizens and thus ineligible to serve as grand jurors. To serve on a federal jury, a person must be a U. S. citizen. 28 USC § 1865 (b) (1). When adjusting the statistical evidence to account for citizenship, the absolute disparity for jury-eligible Hispanics in the Southern District of California was 4.9%, which was insufficient to establish the second prong of the Sixth Amendment test. Artero, supra at 1261; Esquivel, 88 F3d at 727. See also Morrow, supra at 695; Cook, supra.
With regard to Smith’s argument that courts do not scrutinize other distinctive groups, like males or African-Americans, for citizenship eligibility, the Artero court stated:
Sometimes a distinctive group’s proportion of the population is an adequate substitute for its proportion of those eligible to serve on federal juries. For example, in Duren, there was no reason to doubt the usefulness of comparing tire percentage of women summoned for jury service to the percentage in the district, because there is no reason to think women would be disproportionately ineligible to serve on juries. [Cit.] Probably a higher percentage of women than men are jury-eligible, because women on average live longer and get convicted of felonies less than men. But in our nation of immigrants, it stands to reason that border counties and ports of entry would have significant numbers of immigrants not yet eligible to serve on federal juries. It took many of our ancestors a while to learn English and become citizens. [723] Where there is no reason to suppose that the percentage of persons in that group in the population is higher than the percentage eligible to serve, then the former may adequately support an inference as to the latter. Where such an inference is not reasonable, then disparity of percentages in the general population and in the jury [pool] cannot suffice, because the general population ratio does not imply the jury-eligible ratio.
Artero, supra at 1262.
To establish a prima facie case of a fair-cross-section Sixth Amendment violation, Smith had to show an actionable disparity between the percentage of Hispanics on the traverse jury list and the percentage of Hispanics in Hall County who are jury-eligible. See id. at 1260; Duren, supra. He based his percentage of jury-eligible Hall County Hispanics on census numbers showing that the number of over-18 Hall County Hispanics was 17,424, but the evidence showed that the over-18 Hispanics who are U. S. citizens, and thus eligible to serve on juries, amounts to a small fraction of that number. Even a generous assumption, based on testimony, that a fifth of the over-18 Hispanics in Hall County are U. S. citizens, still leaves the absolute disparity between the percentage of Hispanics on the traverse jury list and the percentage of jury-eligible Hispanics in Hall County at only 0.82%, well within constitutional bounds.4 See Morrow, supra at 695; Cook, supra. The trial court erred by finding that Smith had satisfied the second prong of the Sixth Amendment fair-cross-section test.
5. The fourth question posed to the parties involves the third prong of the Sixth Amendment test: whether the trial court correctly found that Smith had met his burden of showing systematic exclusion of Hispanics on the traverse jury list. See Duren, supra; Morrow, supra at 692. The trial court determined that the fact that Hispanics had historically not been separately tracked on the traverse jury list amounted to systematic exclusion. The trial court also found that the Hall County jury commission had made insufficient efforts to recruit Hispanics for the jury list.
The evidence showed that the voters’ registration list was historically the primary source of names for the traverse jury list.5 The per[724] son in charge of Hall County voter registration, Anne Phillips, testified that people can register to vote at the voters’ registration office, by mail, and at libraries, welfare offices, and driver’s license offices. In fact, most new Hall County voters now register at driver’s license offices. Some government agencies are also required to ask applicants if they would like to register to vote. The voter registration forms, which are supplied by the State of Georgia, contain a self-affirming oath attesting to U. S. citizenship. The forms have identity blocks to check for male, female, white, black and other; multiracial was added in 1998; Hispanic has been recently added. Before the Hispanic block was added, some Hispanics checked the “other” block and wrote Hispanic on the line next to it. Ms. Phillips testified that there have been voter registration drives by various groups in Hall County with the specific goal of encouraging Hispanics to register to vote. She estimated that approximately 1,400 Hispanics/Others are registered to vote out of 64,790 registered voters in the county (about 2.2%).
The jury selection clerk for Hall County testified regarding the jury list application form, which allows people to apply directly for inclusion on the jury list. The form contains check-off blocks for male, female, black, white and other. It also contains a question about U. S. citizenship and a self-affirming oath attesting to the truth of all answers on the form. Several jury commissioners and the jury selection clerk testified about their efforts to recruit additional Hispanics for the jury list. The jury selection clerk, knowing that the jury commission wanted to recruit additional Hispanics for the jury list, personally asked 20 to 50 Hispanics if they wanted to apply for inclusion on the jury list. None responded affirmatively. Some current and former jury commissioners also tried to recruit Hispanics for the jury list, with little success. As previously mentioned, a Hispanic jury commissioner in the 1990’s had many Hispanics hand back application forms and say that they were not U. S. citizens; she testified that it was “hard to find” a Hispanic U. S. citizen in Hall County at that time who was not already on the jury list. The jury commission also obtained Hispanic names from a charity’s list, but some of the people on the list had moved out of the county. Additionally, they placed an advertisement in a local newspaper seeking Hispanic volunteers for the jury list, but only received about ten responses.
As regards the third prong of the Sixth Amendment test, Smith “must show that [Hispanics are] underrepresented in the jury-selection process due to systematic exclusion.” Garcia, supra at 491. [725] See also Duren, 439 U. S. at 364. To prove systematic exclusion, Smith “must show the exclusion is ‘inherent in the particular jury-selection process utilized.’ ” Garcia, supra, quoting Duren, supra at 366. For example, in Duren, the underrepresentation of women in the jury pool resulted from a state law that allowed eligible women to exempt themselves from jury service. Duren, supra at 366-367. Smith has shown no such inherent exclusion in the jury selection process utilized in Hall County. Courts have held that when the voter registration list is the primary source of names for the jury list, “ ‘the mere fact that one identifiable group of individuals votes in a lower proportion than the rest of the population does not make a jury selection system illegal or unconstitutional.’ [Cit.]” Garcia, supra at 492. See also United States v. Ireland, 62 F3d 227, 231 (8th Cir. 1995); Cook, 255 Ga. at 574. There was no showing of any effort to impede Hispanic voter registration in Hall County, although Hispanics comprise only about 2.2% of the county’s registered voters. There was also no evidence that the jury commission acted in a discriminatory manner by limiting or excluding Hispanic participation in the Hall County jury pool, although Hispanics comprise only 2.6% of the persons on the traverse jury list. See Garcia, supra. In fact, the evidence showed that the Hall County jury commission made efforts, albeit sporadic at times, to recruit additional eligible Hispanics for the jury list, and Smith’s counsel conceded at oral argument before this Court that those efforts continue to this day. The relatively small number of Hispanic U. S. citizens in Hall County explains the disparity between the percentage of Hispanic voters and jurors on those respective lists and the percentage of over-18 Hispanics in the county population. “If Hispanics in [Hall County] were less likely than others to be citizens, then non-citizenship rather than systematic exclusion of qualified individuals would explain both lower percentages of registered voters and lower representation in the jury [pool].” Artero, supra at 1262.
With regard to the trial court’s order stating that systematic exclusion resulted from the absence of a category for Hispanics on the jury list, there is no evidence that the failure to classify Hispanics prevented eligible Hispanics from registering to vote or applying for the jury list. In fact, cases involving Hispanic representation in the jury pool frequently involve the use of expert witnesses whose role is to analyze the jury list or wheel to determine the number of Hispanics included therein. See Artero, supra at 1261 (defense expert applied Spanish surname search to jury wheel to determine number of Hispanics on list); Esquivel, supra at 726; United States v. Esle, 743 F2d 1465, 1471 (11th Cir. 1984) (defense expert estimated the number of Hispanics on the jury wheel). Moreover, as shown in Division 4, the percentage of jury-eligible Hispanics in Hall County is close to the percentage of Hispanics on the traverse jury list, and [726] there is no evidence that significant disparities between jury-eligible Hispanics in the county and Hispanics on the traverse jury list existed previously. See Duren, supra at 366. We conclude that Smith failed to demonstrate the systematic exclusion of jury-eligible Hispanics in Hall County’s jury selection system. Accordingly, the trial court erred by finding that Smith met his burden of proving the third prong of the Sixth Amendment test.
6. Because Smith failed to carry his burden of proof regarding the second and third prongs of the Sixth Amendment test, he did not establish a prima facie fair-cross-section Sixth Amendment violation. See Duren, supra at 364; Morrow, supra at 692. Therefore, we need not consider the fifth question posed to the parties concerning the State’s rebuttal of Smith’s prima facie case. Juries can only be composed of people who are by law eligible to serve, and the evidence does not show that eligible Hispanic residents of Hall County are constitutionally underrepresented in the jury pool. However, Hispanic representation in Hall County’s jury pool will continue to increase. As noted by Dr. Bohon’s statistics, the percentage of immigrants who become United States citizens grows over time, and their children will be citizens by virtue of their birth in this country. As these Americans seize opportunity, the result will be greater Hispanic participation in the judicial and political processes in Hall County and the rest of the state. In our nation of immigrants, it has always been this way.
Judgment affirmed in part and reversed in part.
All the Justices concur, except Benham, J., who dissents.