Kumar v. Frisco Independent School District et al, No. 4:2019cv00284 - Document 124 (E.D. Tex. 2020)
Court Description: MEMORANDUM OPINION AND ORDER AND FINDINGS OF FACT AND CONCLUSIONS OF LAW. In accord with the Court's findings of fact and conclusions of law, the Court finds that Kumar failed to meet his burden with respect to his claims under Section 2 of the Voting Rights Act. The Court dismisses Kumar's claims with prejudice and will issue judgment accordingly. Signed by District Judge Amos L. Mazzant, III on 8/4/2020. (rpc, )
Download PDF
<![CDATA[
Kumar v. Frisco Independent School District et al Doc. 124 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 1 of 103 PageID #: 3120 United States District Court EASTERN DISTRICT OF TEXAS SHERMAN DIVISION SURESH KUMAR Plaintiff, v. FRISCO INDEPENDENT DISTRICT, ET AL. Defendants. SCHOOL § § § § § § Civil Action No. 4:19-CV-00284 Judge Mazzant MEMORANDUM OPINION AND ORDER AND FINDINGS OF FACT AND CONCLUSIONS OF LAW The present action concerns allegations of voting rights discrimination brought pursuant to Section 2 of the Voting Rights Act of 1965, 52 U.S.C. § 10301. Plaintiff Suresh Kumar (“Kumar”) challenges the at-large electoral system utilized by Frisco Independent School District (“FISD” or “the District”), and its Board of Trustees (“Board”) (collectively, “Defendants”) which he claims dilutes the voting strength of the African American, Hispanic, and Asian minority voting populations. Following a bench trial, and for the reasons that follow,1 the Court finds that Frisco ISD’s at-large electoral system does not violate Section 2 of the Voting Rights Act. Under the totality of the circumstances, FISD’s electoral system does not deny African American, Hispanic, or Asian minority voting populations an equal opportunity to participate in the electoral process or to elect representatives of their choice. LEGAL STANDARD Section 2 of the Voting Rights Act, 52 U.S.C. § 10301, provides in subsection (a) that: “No voting qualification or prerequisite to voting or standard, practice, or procedure shall be imposed or applied by any State or political subdivision in a manner which results in a denial or abridgement This Memorandum Opinion and Order constitutes the Court’s findings of facts and conclusions of law as mandated by Federal Rule of Civil Procedure 52(a)(1). 1 Dockets.Justia.com Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 2 of 103 PageID #: 3121 of the right of any citizen of the United States to vote on account of race or color . . . .” Pursuant to subsection (b): A violation of subsection (a) is established if, based on the totality of circumstances, it is shown that the political processes leading to nomination or election in the State or political subdivision are not equally open to participation by members of a class of citizens protected by subsection (a) in that its members have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice. The extent to which members of a protected class have been elected to office in the State or political subdivision is one circumstance which may be considered: Provided, [t]hat nothing in this section establishes a right to have members of a protected class elected in numbers equal to their proportion in the population. 52 U.S.C. § 10301(b). This amended version of the Act—which does not require proof of discriminatory intent but rather focuses on discriminatory results2—was first considered by the Supreme Court in the seminal case Thornburg v. Gingles, 478 U.S. 30 (1986). In Gingles, the Court established the current two-step framework for analyzing Section 2 cases via a three-part threshold test. Id. Thus, post-Gingles, to prevail on a Section 2 claim, a plaintiff must first demonstrate that: (1) the minority group is “sufficiently large and geographically compact to constitute a majority in a single member district[;]” (2) the minority group “is politically cohesive[;]” and (3) “the white majority votes sufficiently as a bloc to enable it-in the absence of special circumstances, such as the minority candidate running unopposed-usually to defeat the minority’s preferred candidate.” Benavidez v. Irving Ind. Sch. Dist., 2014 WL 4055366, at *4 (N.D. Tex. Aug. 15, 2014) (“Benavidez III”) (citing Gingles, 478 U.S. at 50–51). “Failure to establish any one of the Gingles factors precludes a finding of vote dilution, because “[t]hese circumstances are necessary preconditions for multimember districts to operate to impair minority voters’ ability to elect 2 League of United Latin Am. Citizens, Council No. 4434 (LULAC) v. Clements, 986 F.2d 728, 743 (5th Cir. 1993). 2 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 3 of 103 PageID #: 3122 representatives of their choice . . . .” Clements, 986 F.2d at 743 (citing Gingles, 478 U.S. at 50; Overton v. City of Austin, 871 F.2d 529, 538 (5th Cir. 1989)).3 Should a plaintiff meet the threshold set out by Gingles, the Court then proceeds to step two of its inquiry. During step two, “[t]he minority group must further demonstrate that, under the totality of the circumstances, ‘its members have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice.’” Reyes v. City of Farmers Branch, Texas, 2008 WL 4791498, at *2 (N.D. Tex. Nov. 4, 2008) (citation omitted). Per Gingles, the second phase requires courts to determine, based upon a searching practical evaluation of the ‘past and present reality,’ . . . whether the political process is equally open to minority voters.” Gingles, 478 U.S. at 79. Courts must take a “functional” view of the political process as it considers the facts of each unique case. Id.; see also NAACP v. Fordice, 252 F.3d 361, 367 (5th Cir. 2001) (quoting Magnolia Bar Ass’n, Inc. v. Lee, 994 F.2d 1143, 1147 (5th Cir. 1993) (“Before making its totality of the circumstances analysis, the district court correctly recognized that it was required to effect a flexible, fact-intensive inquiry predicated on ‘an intensely local appraisal of the design and impact of the contested electoral mechanisms . . . .’”). In making this determination, courts are guided by a set of factors drawn from the Senate Report accompanying the 1982 amendments. The so-called Senate Factors include: (1) the history of voting-related discrimination in the State or political subdivision; “A racial gerrymandering claim is ‘analytically distinct’ from an intentional vote dilution claim.” Anne Harding v. Cty. of Dallas, Texas, 948 F.3d 302, 312 (5th Cir. 2020). As the Supreme Court explained in Miller v. Johnson: 3 Whereas [an intentional] vote dilution claim alleges that the [government] has enacted a particular voting scheme as a purposeful device to maintain or cancel out the voting potential of racial or ethnic minorities, an action disadvantaging a particular race, the essence of [a racial gerrymandering claim] . . . is that the [government] has used race as a basis for separating voters into districts. Anne Harding, 948 F.3d at 312 (citing Shaw v. Reno, 509 U.S. 630, 652 (1993); Miller v. Johnson, 515 U.S. 900, 911 (1995)). 3 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 4 of 103 PageID #: 3123 (2) the extent to which voting in the elections of the State or political subdivision is racially polarized; (3) the extent to which the State or political subdivision has used voting practices or procedures that tend to enhance the opportunity for discrimination against the minority group, such as unusually large election districts, majority vote requirements, and prohibitions against bullet voting; (4) the exclusion of members of the minority group from candidate slating processes; (5) the extent to which minority group members bear the effects of past discrimination in areas such as education, employment, and health, which hinder their ability to participate effectively in the political process; (6) the use of overt or subtle racial appeals in political campaigns; (7) the extent to which members of the minority group have been elected to public office in the jurisdiction. Id. at 44–45. The Senate Report also noted, as the Gingles Court pointed out, that courts should consider whether there is: “[(8)] evidence demonstrating that elected officials are unresponsive to the particularized needs of the members of the minority group; [or (9) evidence] that the policy underlying the State’s or the political subdivision’s use of the contested practice or structure is tenuous may have probative value. Id. Courts additionally look to a tenth factor following Johnson v. De Grandy, 512 U.S. 997, 1006–07 (1994): “[(10)] whether the number of districts in which the minority group forms an effective majority is roughly proportional to its share of the population in the relevant area[.]” Fairley v. Hattiesburg, Miss., 584 F.3d 660, 672–73 (5th Cir. 2009) (quoting League of United Latin Am. Citizens (LULAC) v. Perry, 548 U.S. 399, 426 (2006) (hereinafter cited as “LULAC”). “Multimember districts and at-large election schemes . . . are not per se violative of minority voters’ rights. Minority voters who contend that the multimember form of districting violates § 2[] must prove that the use of a multimember electoral structure operates to minimize or cancel out their ability to elect their preferred candidates.” Benavidez v. Irving Indep. Sch. Dist., Texas, 690 F. Supp. 2d 451, 456 (N.D. Tex. 2010) (“Benavidez II”) (citing Gingles, 478 U.S. at 48). 4 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 5 of 103 PageID #: 3124 The burden of proof is on the plaintiff to demonstrate, by a preponderance of the evidence, that “all of the Gingles preconditions were satisfied and that based on the totality of the circumstances the at-large election system diluted the voting strength of [the minority group].” League of United Latin Am. Citizens #4552 (LULAC) v. Roscoe Ind. Sch. Dist., 123 F.3d 843, 846 (5th Cir. 1997) (citing Overton, 871 F.2d at 532). ANALYSIS The Court is presented with a challenge to Frisco ISD’s at-large electoral system pursuant to Section 2 of the Voting Rights Act of 1965, 52 U.S.C. § 10301.4 Based on the Court’s Rule 52 findings of fact and conclusions of law, the Court finds that Kumar failed to meet his burden with respect to his voter dilution claim. The Court’s Opinion proceeds as follows: I. Standing; II. Admissibility of Exhibits; III. Reliability of Mr. Ely’s Methodology; IV. Findings of Fact; and V. Conclusions of Law. Because standing is in question, the Court considers Defendants standing challenge first. I. Standing On October 17, 2019, Defendants filed Defendants’ Motion to Dismiss for Lack of Standing Pursuant to FED. R. CIV. P. 12(b)(1) (Dkt. #41). After considering arguments from both parties and finding that Kumar’s First Amended Complaint was unclear on whom Kumar was seeking to represent, the Court, in its discretion, ordered Kumar to amend his complaint (Dkt. #74).5 On March 20, 2020, Kumar filed his Second Amended Complaint (Dkt. #81). Unlike 4 Kumar initially included intentional voter dilution claims under the Fourteenth and Fifteenth Amendments to the United States Constitution. Those claims have since been voluntarily withdrawn. See Dkt. #123. 5 The Court only permitted such amendment because: (1) the Court found that, under his First Amended Complaint, Kumar had constitutional standing to represent himself; (2) Kumar’s pleadings were unclear on whether Kumar was attempting to represent parties not present in this litigation; (3) the Court required full knowledge of Kumar’s purported representation to consider any prudential standing concerns; and (4) while the Court stated that, at that time, Kumar’s third-party standing argument was foreclosed, the Court recognized that Kumar still had a viable argument that he was merely attempting to represent himself (Dkt. #74). 5 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 6 of 103 PageID #: 3125 Kumar’s First Amended Complaint, Kumar’s Second Amended Complaint unequivocally couches any allegations in terms of Kumar’s rights, not the rights of all minority groups located within FISD (Dkt. #81). Nonetheless, Defendants submit in Defendants’ Proposed Findings of Fact and Conclusions of Law that Kumar still does not have standing to assert a voting rights claim on behalf of African Americans or Hispanics (Dkt. #75).6 For the reasons discussed below, the Court disagrees. A. Constitutional Standing Because constitutional standing necessarily affects the prudential standing analysis a court must conduct under Kowalski v. Tesmer, 543 U.S. 125, 129–30 (2004), the Court begins with constitutional standing.7 Under the United States Constitution, the “judicial Power of the United States” extends only to “Cases” and “Controversies.” U.S. CONST. art. III, §§ 1–2. “No principle is more fundamental to the judiciary’s proper role in our system of government than the constitutional limitation of federal-court jurisdiction to actual cases or controversies.” Raines v. Byrd, 521 U.S. 811, 818 (1997) (quoting Simon v. E. Ky. Welfare Rights Org., 426 U.S. 26, 37 (1976)); cf. Sprint Communications, Inc. v. Jacobs, 571 U.S. 69, 77 (2013) (quoting Colorado River Water Conservation Dist. v. United States, 424 U.S. 800, 817 (1976)) (stating that “a federal court’s ‘obligation’ to hear and decide a case is ‘virtually unflagging.’”). The doctrine of standing 6 The parties concede that Kumar has standing to represent the Asian population in FISD. The Court already ruled that Kumar had standing to represent Asians under Benavidez v. City of Irving, Tex., 638 F. Supp. 2d 709, 710 (N.D. Tex. 2009) (“Benavidez I”) (Dkt. #74). Pursuant to Kumar’s Second Amended Complaint, however, it seems that Kumar no longer asserts standing to represent the Asian community as a whole. Rather, Kumar now couches his allegations as relating to himself only with the caveat that he claims, for Gingles evidentiary purposes, that he belongs to a minority coalition comprised of Africans Americans, Hispanics, and Asians. 7 The Court is aware that Defendants do not challenge constitutional standing. Nevertheless, as previously stated, the Court “must zealously guard against exercising its Article III power over a matter erroneously” (Dkt. #74). The Court is now presented with a Second Amended Complaint that, to a degree, amends the allegations levied at Frisco ISD and the Board. When the Court’s duty to guard against an improper assertion of its jurisdiction is coupled with the reality that constitutional standing necessarily affects a prudential standing analysis under Kowalski, the Court finds it appropriate to re-consider constitutional standing. 6 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 7 of 103 PageID #: 3126 is “rooted in the traditional understanding of a case or controversy,” and it “limits the category of litigants empowered to maintain a lawsuit in federal court . . . .” Spokeo, Inc. v. Robins, 136 S. Ct. 1540, 1547 (2016) (citation omitted); see also Lexmark Int’l, Inc. v. Static Control Components, Inc., 572 U.S. 112, 125 (2014) (clarifying that constitutional standing is limited to an analysis of “Article III’s limitation of the judicial power to resolv[e] ‘Cases’ and ‘Controversies’”). So, the Court must decide issues of standing before all other issues because it “determines the court’s fundamental power even to hear the suit.” Ford v. NYLCare Health Plan of Gulf Coast, Inc., 301 F.3d 329, 332 (5th Cir. 2002). To establish standing, a plaintiff must have: (1) suffered an injury in fact; (2) that is fairly traceable to the challenged conduct of the defendant; and (3) that is likely to be redressed by a favorable judicial decision. Spokeo, 136 S. Ct. at 1547 (citing Lujan v. Defs. of Wildlife, 504 U.S. 555, 560–61 (1992)). A plaintiff’s burden to establish standing depends upon the stage at which standing is challenged. See Lujan, 504 U.S. at 561 (“Since they are not mere pleading requirements but rather an indispensable part of the plaintiff’s case, each element must be supported in the same way as any other matter on which the plaintiff bears the burden of proof, i.e., with the manner and degree of evidence required at the successive stages of the litigation.”). At the pleading stage, “general factual allegations of injury resulting from the defendant’s conduct may suffice . . . .” Id. (citing Lujan v. Nat’l Wildlife Fed., 497 U.S. 871, 889 (1990)). At the summary judgment stage, however, “the plaintiff can no longer rest on such ‘mere allegations,’ but must ‘set forth’ by affidavit or other evidence ‘specific facts,’ FED. R. CIV. P. 56(e), which for purposes of the summary judgment motion will be taken to be true.” Id. And “at the final stage, those facts (if controverted) must be ‘supported adequately by the evidence adduced at trial.’” Id. (citing Gladstone, Realtors v. Village of Bellwood, 441 U.S. 91, 115 n.31 (1979)). Following a trial on 7 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 8 of 103 PageID #: 3127 the merits, the Court is satisfied that Kumar established constitutional standing through the evidence adduced at trial. Id. Kumar established an Article III injury. See Adar v. Smith, 639 F.3d 146, 150 (5th Cir. 2011) (“[S]tanding does not depend upon ultimate success on the merits . . . .”); Cole v. Gen. Motors Corp., 484 F.3d 717, 723 (5th Cir. 2007) (holding that a court is not concerned with the merits of a claim when assessing Article III standing); Hanson v. Veterans Admin., 800 F.2d 1381, 1385 (5th Cir. 1986) (“It is inappropriate for the court to focus on the merits of the case when considering the issue of standing.”). Kumar alleges that he has been stripped of the right to elect representatives of his choice due to an at-large electoral system which dilutes the voting power of minority populations. Specifically, Kumar avers that: (1) he is a member of a minority group— Indian (Asian); (2) he is a registered voter in FISD; and (3) because of Kumar’s minority status, Defendants have impermissibly diluted his voting power in FISD. Kumar proffered adequate evidence at trial establishing, for Article III purposes, that he suffered such injury in fact. See Lujan, 504 U.S. at 560–61.8 The right to a democratic process free of invidious discrimination is the cornerstone of the Voting Rights Act—and having that freedom denied or impaired due to an unjust system is the quintessential harm Congress sought to prevent. The Court finds that Kumar carried his burden in establishing an injury-in-fact. See Spokeo, 136 S. Ct. at 1547 (citing Lujan, 504 U.S. at 560–61). Kumar also demonstrated that his injury is fairly traceable to the challenged conduct of Defendants. Kumar claims that Defendants’ at-large electoral process, which requires a districtwide plurality for staggered positions, results in a discriminatory system that dilutes his vote and effectively denies him equal representation. Were Defendants to rid themselves of this at-large The Court is not holding that Kumar’s allegations are in any way factually correct at this juncture. See Truman v. United States, 26 F.3d 592, 594 (5th Cir. 1994). 8 8 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 9 of 103 PageID #: 3128 system, Kumar maintains that his voting rights would be restored. Yet, Defendants refuse to implement any change to FISD’s electoral scheme. As such, any lack of representation Kumar has experienced is derived directly from Defendants’ electoral scheme. Kumar proffered adequate evidence at trial establishing, for Article III purposes, that his injury is adequately traceable to the challenged conduct of Defendants. See Lujan, 504 U.S. at 560–61. Kumar accordingly established the traceability element of constitutional standing. See Spokeo, 136 S. Ct. at 1547 (citing Lujan, 504 U.S. at 560–61). Finally, Kumar demonstrated that if the Court were to grant Kumar the declaratory and injunctive relief he seeks, Kumar’s injury would be redressed. Kumar asks that the Court, among other things: (1) declare Defendants’ at-large electoral process unconstitutional; (2) grant a permanent injunction against Defendants prohibiting them from conducting further elections under said system; and (3) institute an order directing Defendants to devise an election plan that complies with Section 2 of the Voting Rights Act. Should the Court grant Kumar’s requests, the stain of Defendants’ flawed electoral system9 would be lifted and the harm Kumar complains of redressed. See id. (citing Lujan, 504 U.S. at 560–61). Kumar established the redressability element of constitutional standing. See id. (citing Lujan, 504 U.S. at 560–61). In sum, the Court finds that Kumar established constitutional standing to proceed on his Voting Rights Act claims. It remains to be determined, however, whether Kumar established prudential standing. B. Prudential Standing While it is axiomatic that a plaintiff must establish constitutional standing to proceed on his or her claims, a plaintiff is also required, at least for the time being, to establish prudential Again, the Court is not ruling on the merits of Kumar’s allegations. See Adar, 639 F.3d at 150; Cole, 484 F.3d at 723; Hanson, 800 F.2d at 1385. 9 9 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 10 of 103 PageID #: 3129 standing. See St. Paul Fire & Marine Ins. Co. v. Labuzan, 579 F.3d 533, 539 (5th Cir. 2009) (“Prudential standing requirements exist in addition to the immutable requirements of Article III as an integral part of judicial self-government.”) (internal quotation marks omitted). In one of the Supreme Court’s seminal cases on standing, Lexmark, Justice Scalia, writing for a unanimous Court, explained that while the Article III “Cases” and “Controversies” analysis established the absolute minimum for constitutional standing, “[i]n recent decades, [the Supreme Court] adverted to a ‘prudential’ branch of standing.” 572 U.S. at 125. The Supreme Court noted that while prudential standing has not been “exhaustively defined,” three principles characterized this judicial construction; namely, “the general prohibition on a litigant’s raising another person’s legal rights, the rule barring adjudication of generalized grievances more appropriately addressed in the representative branches, and the requirement that a plaintiff’s complaint fall within the zone of interests protected by the law invoked.” Id. at 126 (citing Elk Grove Unified School Dist. v. Newdow, 542 U.S. 1, 12 (2004) (quoting Allen v. Wright, 468 U.S. 737, 751 (1984)). Although the Court enumerated three principles that encompassed “prudential standing,” the Court held that the third principle—that a plaintiff’s complaint must fall within the zone of interests protected by the law invoked—is a matter of statutory construction, rather than prudential standing. Id. The Court also noted that the first principle—the bar on generalized grievances—is a constitutional strand of standing, not a prudential strand. Id. at 127 n.3 (citing Lance v. Coffman, 549 U.S. 437, 439 (2007) (per curiam); DaimlerChrysler Corp. v. Cuno, 547 U.S. 332, 344–346 (2006); Lujan, 504 U.S. at 573–574). After Lexmark, “the doctrine of third-party standing or jus tertii, still remains [for the time being] in the prudential category.” 33 CHARLES ALAN WRIGHT & ARTHUR R. MILLER, FEDERAL PRACTICE AND PROCEDURE § 8343, at 1 (2d ed. April 2019 Update). In the Fifth Circuit, for 10 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 11 of 103 PageID #: 3130 instance, the Fifth Circuit recognized the viability of third-party standing jurisprudence remains uncertain, yet still held that the “prudential requirement that a party must assert its own rights [still applies], see, e.g., Danos v. Jones, 652 F.3d 577, 582 (5th Cir. 2011), and [courts] are bound to follow [this requirement] until the Supreme Court squarely holds to the contrary, see Exelon Wind 1, L.L.C. v. Nelson, 766 F.3d 380, 394 (5th Cir. 2014).” Superior MRI Servs., Inc. v. All. Healthcare Servs., Inc., 778 F.3d 502, 504 (5th Cir. 2015). The requirements of third-party standing, then, are that a litigant ordinarily “must assert his own legal rights and interests and cannot rest his claim to relief on the legal rights or interests of third parties.” U.S. Dep’t. of Labor v. Triplett, 494 U.S. 715, 721 (1990) (citing Valley Forge Christian College v. Americans United for Separation of Church & State, Inc., 454 U.S. 464, 474 (1982) (quoting Warth v. Seldin, 422 U.S. 490, 499 (1975)); see also 33 CHARLES ALAN WRIGHT & ARTHUR R. MILLER, FEDERAL PRACTICE AND PROCEDURE § 8343, at 1 (2d ed. April 2019 Update) (“The problem of third-party standing arises where a plaintiff seeks to enforce the legal rights of some third party not before the court.”). The reasoning behind the third-party standing doctrine is plentiful, but two reasons promulgated by the Supreme Court in Singleton v. Wulff are sufficient for the present inquiry. 428 U.S. 106, 113–14 (1976). As articulated by the Supreme Court: The reasons are two. First, the courts should not adjudicate such rights unnecessarily, and it may be that in fact the holders of those rights either do not wish to assert them, or will be able to enjoy them regardless of whether the in-court litigant is successful or not. Second, third parties themselves usually will be the best proponents of their own rights. Id. at 113–14. The third-party standing rule, however, is not absolute. See Kowalski v. Tesmer, 543 U.S. 125, 129–130 (2004). The Supreme Court held in Kowalski that “a party that satisfies the requirements of Article III standing may seek to enforce the legal rights of a third-party where: (1) the party has a ‘close’ relationship with the possessor of the right; and (2) ‘there is a ‘hindrance’ 11 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 12 of 103 PageID #: 3131 to the possessor’s ability to protect its own interests.’” Id. at 130; see also Veasey v. Perry, 29 F. Supp. 3d 896, 908–09 (S.D. Tex. 2014) (applying the Kowalski two-step framework to determine whether plaintiffs had third-party standing to assert a Voting Rights Act violation); Greater Birmingham Ministries v. State, 161 F. Supp. 3d 1104, 1115 (N.D. Al. 2016) (same). The Supreme Court is “quite forgiving” in the context of the First Amendment and when “enforcement of the challenged restriction against the litigant would result indirectly in the violation of third parties’ rights.” Id. Otherwise, “the Court generally looks unfavorably upon third-party standing.” Id. (citations omitted). With the contours of standing recognized, the Court is now presented with the collision of two Goliaths in the law: prudential standing and the Voting Rights Act. While prudential standing conclusively states that a plaintiff “must assert his own legal rights and interests [] and cannot rest his claim to relief on the legal rights or interests of third parties[,]” Triplett, 494 U.S. at 721, the Voting Rights Act, per Gingles, requires the same plaintiff to establish a minority coalition. 478 U.S. at 50–51. The Court must determine whether Kumar entered into the waters of third-party standing by offering Gingles evidence relating to a minority coalition of African Americans, Hispanics, and Asians when he, as the sole-plaintiff, is Indian-American. While the Court is unaware of any precedent encompassing a sole plaintiff challenging an election system via evidence of a multiple minority group coalition, the Court is of the opinion that Kumar nonetheless established standing to proceed with his claim. First, the Voting Rights Act does not require a class action for a plaintiff to proceed with a claim. 52 U.S.C. § 10302, formerly 42 U.S.C. § 1973(a), provides, in relevant part, that the Attorney General or “an aggrieved person” may institute a “proceeding under any statute to enforce the voting guarantees of the Fourteenth or Fifteenth Amendment . . . .” An “aggrieved person” is 12 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 13 of 103 PageID #: 3132 defined as a voter “whose rights have allegedly been denied or impaired.” ) Veasey, 29 F. Supp. 3d at 908–09 (citing Roberts v. Wamser, 883 F.2d 617, 621 (8th Cir. 1989)). The use of the singular term, “person,” denotes that a collective group of similarly situated individuals, or “persons,” is not statutorily required to maintain an action under the Voting Rights Act. See Person, WEBSTER’S THIRD NEW INTERNATIONAL DICTIONARY (1976) (defining person as “an individual human being”); see also Person, THE RANDOM HOUSE COLLEGE DICTIONARY (1973) (defining person as “a human being”). A class action is not mandated by the text. Moreover, the Court is unaware of any authority which prohibits individual representation in a Voting Rights Act case. On the contrary, the Court independently ascertained that individual plaintiffs previously brought Voting Rights Act actions absent class actions. See, e.g., Benavidez I, 638 F. Supp. 2d at 710. And any requirement that multiple plaintiffs of the same minority group must coalesce as a class to file suit would be an antithetical requirement that would disserve the overarching purpose of the Act—stomping out invidious racial discrimination. A review of the Gingles factors supports this conclusion. Under Gingles: • First, the minority group must be able to demonstrate that it is sufficiently large and geographically compact to constitute a majority in a single-member district. • Second, the minority group must be able to show that it is politically cohesive. If the minority group is not politically cohesive, it cannot be said that the selection of a multimember electoral structure thwarts distinctive minority group interests. • Third, the minority must be able to demonstrate that the white majority votes sufficiently as a bloc to enable it—in the absence of special circumstances, such as the minority candidate running unopposed—usually to defeat the minority’s preferred candidate. Gingles, 478 U.S. at 50–51. A review of the threshold Gingles factors reveals that a plaintiff must necessarily put on evidence of discrimination against the minority group to which they belong to demonstrate discrimination against themselves. This evidentiary burden remains whether a 13 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 14 of 103 PageID #: 3133 plaintiff is joined by a coalition of similarly situated plaintiffs or is individually representing himor herself. Because a class action is not mandated by the Voting Rights Act, it necessarily follows that, as in Benavidez I, a sole plaintiff may submit coalition evidence under Gingles absent a “coalition of plaintiffs.” 638 F. Supp. 2d at 710. The question, then, becomes the following: How does standing jurisprudence limit a sole plaintiff’s ability to submit coalition evidence including other minority groups to which the plaintiff does not belong? The Court previously held that Kumar would not have standing to represent entire minority groups to which he did not belong— i.e., African Americans and Hispanics—if that was what Kumar was attempting to do. Having amended his complaint, Kumar now contends that he is attempting to do no such thing; rather, Kumar maintains that he only seeks to represent himself. With this Amended Complaint before the Court, Kumar asserts that there is no third-party standing issue. Defendants counter, however, that Kumar is still asserting the rights and interests of third parties—African Americans and Hispanics living within FISD—without those parties’ participation. Because Kumar has not met the Kowalski exception to third-party standing, Defendants continue, Kumar’s claims must be dismissed for lack of standing. The Court agrees with Kumar that the Court need not dismiss the present matter for a third-party standing issue. First, Kumar is not asserting standing on behalf of all minority groups; on the contrary, Kumar only asserts standing to represent himself. A comparison of Kumar’s First Amended Complaint to Kumar’s Second Amended Complaint does well. In Kumar’s First Amended Complaint, each allegation revolved around one of two groups: (1) Kumar himself; or (2) Kumar along with all African Americans, Hispanics, Asians, and other minorities. Count 1 of Kumar’s First Amended Complaint is telling. In Count 1, Kumar originally argued as follows: 14 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 15 of 103 PageID #: 3134 55. The allegations set forth in paragraphs 1–52 above are hereby incorporated as if fully set forth herein. 56. Section 2 of the Voting Rights Act prohibits any standard, practice or procedure that results in the denial or abridgment of minority voting rights. Specifically, it forbids any electoral system that denies African Americans, Hispanics, Asians, and other minority groups an opportunity equal to that afforded to other members of the electorate to elect representatives of their choice. 57. FISD’s at-large electoral system for electing its Board unconstitutionally dilutes the voting strength of African Americans, Hispanics, Asians, and other minority voting populations and is not equally open to participation by FISD’s voters of color. Further, the electoral system results in African Americans, Hispanics, Asians, and other minority populations having less opportunity than other FISD voters to meaningfully participate in the electoral process and to elect representatives of their choice. 58. The African American, Hispanic, Asian, and other communities of color in FISD are sufficiently large, geographically compact, and constitute a politically unified group that votes cohesively as a bloc, such that a properly-apportioned single-member electoral district can be drawn in which minorities would constitute a majority of eligible voters. 59. FISD Board elections are characterized by racially-polarized voting in which the predominately White voting bloc votes in a way that regularly defeats the candidates of choice of African American, Hispanic, Asian, and communities of color and has a chilling and discouraging impact on participation from voters of color. Thus, based on the totality of past and present circumstances, the FISD electoral system impermissibly dilutes the minority vote and stymies that community’s ability to participate fully in the election process. 60. Accordingly, Mr. Kumar requests that the Court issue a declaratory judgment that the FISD electoral system violates Section 2 of the Voting Rights Act. (Dkt. #23) (emphasis added). In the same First Amended Complaint, however, Kumar asserted that he was the sole plaintiff (Dkt. #23). From this muddled Complaint, it was not readily apparent whom Kumar was attempting to represent. Kumar’s Second Amended Complaint is void of such confusion. In Count One of Kumar’s Second Amended Complaint, for instance, Kumar asserts: 55. The allegations set forth in paragraphs 1–54 above are hereby incorporated as if fully set forth herein. 56. Section 2 of the Voting Rights Act prohibits any standard, practice or procedure that results in the abridgement of a citizen’s voting rights through denying any aggrieved person an opportunity equal to that afforded to other members of the electorate to participate in the political process and elect representatives of their choice. 57. Mr. Kumar is a United States citizen and an aggrieved person entitled to the protection of the Voting Rights Act. FISD’s at-large system for electing its Board of Trustees violates 15 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 16 of 103 PageID #: 3135 the Voting Rights Act by diluting Plaintiff Suresh Kumar’s vote, which denies Mr. Kumar an equal opportunity to participate in the political process and elect representatives of his choice. 58. Accordingly, Mr. Kumar requests that the Court issue a declaratory judgment that the FISD electoral system violates Section 2 of the Voting Rights Act. (Dkt. #81). As evidenced by the amendments to Kumar’s Complaint, Kumar is not asserting that he has standing to represent Africans Americans, Hispanics, or even Asians now; rather, he claims that his rights have been violated and he simply needs to put on evidence of a minority coalition, per Gingles, to protect his rights. The Court is now presented with a single minority plaintiff who wishes to put on evidence of how a coalition to which he allegedly belongs to has been affected by a purportedly discriminatory electoral system. In other words, Kumar is asserting his own legal interests and, as required under Gingles, merely attempting to demonstrate how those personal interests were harmed. To do this, Kumar must employ statistical evidence demonstrating that a particular coalition exists. While the interplay of prudential third-party standing and the Gingles evidentiary burden do not seem compatible at first glance, the two are not mutually exclusive. In the end, both Gingles and the third-party standing doctrine must, and indeed can, coexist to permit meaningful actions under the Voting Rights Act. This leads the Court to its second finding. The third-party standing doctrine and Gingles evidentiary burden coexist because Gingles does not require a plaintiff to “rest” his or her claim on the rights of others. This is because Gingles is an evidentiary matter, not a standing matter. The Gingles framework sets out a threshold evidentiary burden, not a proxy test for standing. Only by equating the two does one make them incompatible. By recognizing that Gingles is not requiring a plaintiff to assert the rights of other unnamed parties, but rather requiring a plaintiff to offer evidence establishing a minority coalition to which he or she belongs, a court can adequately separate the tension between the two legal doctrines. Kowalski supports this conclusion. 16 In Kowalski, a group of indigent criminal Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 17 of 103 PageID #: 3136 defendants and criminal attorneys challenged a state court’s practice—and the statute which subsequently codified the state court’s practice—of denying the appointment of appellate counsel following a plea-based conviction. 543 U.S. at 127–29. On Appeal, the Sixth Circuit barred the indigent defendants from proceeding with the lawsuit. Notwithstanding its initial holding, the Sixth Circuit held, independently, that the attorneys had third-party standing to assert the indigent defendant’s rights. Id. The Sixth Circuit then held that the state statute was constitutional. Id. Sitting en banc, the Sixth Circuit reviewed its decision. Ultimately, the Sixth Circuit affirmed its third-party standing decision, but reversed course on the merits and found the state statute unconstitutional. Id. The Supreme Court subsequently granted certiorari. Writing for the majority, Chief Justice Rehnquist held that the attorneys lacked third-party standing to assert the rights of the indigent defendants. Id. at 568–69. Applying a two-part test, the Court held that the attorneys did not demonstrate a “close relationship” with their alleged clients nor demonstrate a hindrance to indigent defendants “advancing their own constitutional rights.” Id. More importantly for this Court’s purposes, however, was the Kowalski Court’s explanation of the third-party standing doctrine. According to the Supreme Court, the third-party standing doctrine: assumes that the party with the right has the appropriate incentive to challenge (or not challenge) governmental action and to do so with the necessary zeal and appropriate presentation. It represents a “healthy concern that if the claim is brought by someone other than one at whom the constitutional protection is aimed,” the courts might be “called upon to decide abstract questions of wide public significance even though other governmental institutions may be more competent to address the questions and even though judicial intervention may be unnecessary to protect individual rights.” Id. at 129 (internal citations omitted). The underlying purpose of the third-party standing doctrine is not implicated here like it was in Kowalski. Unlike the Respondents in Kowalski, Kumar is not a third-party attempting to litigate the merits of a constitutional protection aimed at a third party. Instead, Kumar is asserting that his 17 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 18 of 103 PageID #: 3137 personal constitutional rights have been infringed. The concerns of the Kowalski Court cannot reasonably apply here. Surely, a man who claims that he has suffered from voter discrimination at the hands of his government will have the “appropriate incentive,” “necessary zeal,” and “appropriate presentation” in challenging what he perceives as an affront to his basic, fundamental rights as a citizen. Id. Put simply, Kumar does not have to “rest his claim to relief on the legal rights or interests of third parties.” Warth, 422 U.S. at 499. It is his rights that Kumar is seeking to protect. And the fact that, under the Voting Rights Act and Gingles, Kumar must now offer evidence establishing a minority coalition to which he belongs—which necessarily encompasses minority groups not present before the Court—does not change the Court’s conclusion. The Court is not required to decide an abstract question here; the judiciary is the appropriate governmental institution to ferret out statutory violations; and Kumar is the best proponent of his own rights. See Kowalski, 543 U.S. at 129. What is more, Kumar is not actively seeking to litigate on behalf of an absent subset of people “at whom [the Voting Rights Act] is aimed.” Id. Quite the opposite is true. As previously alluded to, the current legal landscape surrounding the enforcement of the Voting Rights Act requires Kumar, an “aggrieved person” with constitutional standing, to put forth the very evidence which Defendants now fault him for putting forward. It cannot be reasonably said that Kumar “rests his claim” on the rights of third parties when the law mandates that Kumar meet an evidentiary burden that is, in a sense, concerned with third parties. If Gingles is to so intersect with the third-party standing doctrine such that courts read into the Voting Rights Act a requirement that a representative from each minority group must be present in litigation to confer standing, a new burden will arise, the text of the Voting Rights Act will be effectively amended from “aggrieved person” to “aggrieved persons,” and the Act’s individual safeguard will be fundamentally eviscerated. Justice—and the Voting Rights Act—demands a different conclusion. 18 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 19 of 103 PageID #: 3138 One may query how this conclusion is permitted when it allows for plaintiffs, such as Kumar, to bring actions that may not reflect the consensus of the community that he or she claims is part of the minority coalition he or she wishes to establish. The answer is this: a minority coalition requires cohesion. Should there be members of the community who “do not wish to assert [their rights],” Singleton, 428 U.S. at 113–14, those same members may speak out—as many did in this action—and condemn the action, and even the plaintiff, as not speaking on their behalf. And if that is not convincing, the Court’s conclusion does not require such a proactive stance by absent third parties. The defense—again, as occurred here—can proffer evidence that a coalition does not exist because of a lack of cohesion. In the end, each community members’ voice will be heard, and each plaintiff will still be afforded the opportunity to defend his or her individual rights. Having recognized the tension that Gingles and the third-party standing doctrine may elicit, the Court holds that the two legal doctrines are separate, compatible, and, in this case, do not prevent the Court from exercising jurisdiction. Kumar is not asserting standing on behalf of any minority population. Defendants’ arguments that Kumar lacks standing are therefore unpersuasive. Kumar wishes to offer evidence, under Gingles, to establish a minority coalition. The Court finds that Kumar has constitutional standing to proceed with his claim and there is no third-party standing implicated in this matter. Before the Court can reach the merits of Kumar’s claim, however, it must first address the remaining threshold issues: (1) the admissibility of exhibits; and (2) the reliability of certain expert reports. II. Admissibility of Exhibits The Court must now determine whether to admit certain exhibits that were conditionally admitted at trial. Among those exhibits are Plaintiff’s Exhibit 77, Defendants’ Exhibits 3–4, and Defendants’ Exhibits 8–13. The Court considers each exhibit below and holds that Plaintiff’s 19 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 20 of 103 PageID #: 3139 Exhibit 77 is admitted, Defendants’ Exhibit 3 is stricken, and Defendants’ Exhibits 4 and 8–13 are admitted. A. Plaintiff’s Exhibit 77 The Court first considers Plaintiff’s Exhibit 77. At trial, Kumar offered the testimony of Mr. Carlos Gallardo (“Mr. Gallardo”). Mr. Gallardo applied for a vacancy on FISD’s Board in 2014 and sought election to the Board in 2015. Mr. Gallardo was unsuccessful in both attempts. On direct examination of Mr. Gallardo, Kumar offered Plaintiff’s Exhibit 77. Plaintiff’s Exhibit 77 is comprised of: (1) Mr. Gallardo’s Form of Interest for the Frisco ISD Board of Trustees—i.e., his application to fill the 2014 vacancy; (2) an email correspondence—between Mr. Gallardo and Assistant to the Superintendent and Board of Trustees, Beverly Brunson (“Ms. Brunson”)— regarding a Dallas Morning News request for the Forms of Interest for each applicant; and (3) an email entitled “FISD Board of Trustees” which stated the Board’s choice for the position (Plaintiff’s Exhibit 77). Defendants objected to Plaintiff’s Exhibit 77 as hearsay. More specifically, Defendants claimed that the emails contained conversations between individuals not present at trial. The Court instructed Defendants that it did not see how Plaintiff’s Exhibit 77 constituted hearsay when the emails contained within Plaintiff’s Exhibit 77 were FISD documents sent by FISD employees. Without further elaboration by Defendants, the Court conditionally admitted Plaintiff’s Exhibit 77. The Court then instructed Defendants that the Court would fully admit Plaintiff’s Exhibit 77 post-trial if Defendants failed to establish that the emails were not FISD documents. Because Defendants failed to offer any argument that these documents are not FISD documents, the Court overrules Defendants’ objection and admits Plaintiff’s Exhibit 77. The Federal Rules of Evidence dictate that hearsay is not admissible unless it falls within an exception established by a rule or statute. FED. R. EVID. 802. Hearsay is “a statement that: (1) 20 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 21 of 103 PageID #: 3140 the declarant does not make while testifying at the current trial or hearing; and (2) a party offers in evidence to prove the truth of the matter asserted in the statement.” FED. R. EVID. 801. An opposing party’s statement, however, is excluded from the definition of hearsay when the “statement is offered against an opposing party and: (A) was made by the party in an individual or representative capacity.” FED. R. EVID. 801(d)(2)(A). Here, Defendants claim, without further elaboration, that Plaintiff’s Exhibit 77 is offered to prove the truth of the matter asserted. Further, Defendants assert that Plaintiff’s Exhibit 77 is an out-of-court statement because the Exhibit includes, in part, an email chain between multiple individuals not present at trial. Plaintiff’s Exhibit 77 is not hearsay. The emails included in Plaintiff’s Exhibit 77 are emails between Mr. Gallardo and Ms. Brunson. Ms. Brunson is serving in her role as Assistant to the Superintendent and Board of Trustees when emailing Mr. Gallardo. Indeed, Ms. Brunson’s signature in the “FISD Board of Trustees” email, which rejected Mr. Gallardo’s application, shows as follows: “Sincerely, FISD Board of Trustees” (Plaintiff’s Exhibit 77). Nevertheless, Defendants object and claim that these emails are hearsay. Defendants are incorrect. These emails are statements made by FISD’s Board through the Board’s Secretary. One of the emails officially notified Mr. Gallardo that the Board was releasing his Form of Interest to the Dallas Morning News. The other email notified Mr. Gallardo that the Board was selecting another applicant for the 2014 vacancy. Both emails necessarily constitute statements made by the Board in its official capacity. Therefore, the emails fall outside the scope of Rule 801’s definition of hearsay. See FED. R. EVID. 801(d)(2)(A). Defendants’ objection is accordingly overruled, and Plaintiff’s Exhibit 77 is admitted. 21 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 22 of 103 PageID #: 3141 B. Defendants’ Exhibit 3 The remaining exhibits the Court must consider are, to some degree, connected. Defendants’ Exhibit 3 is the Supplemental Expert Report of Dr. Peter Morrison (“Dr. Morrison”) (“Morrison’s Supplemental Expert Report”). Defendants’ Exhibits 4 and 8–13 are a set of maps that Ms. Lori Wassam (“Ms. Wassam”), FISD’s internal demographer, created (hereinafter, the “Wassam Exhibits”). While the Wassam Exhibits are used in Morrison’s Supplemental Expert Report, they are also independently offered. The Court considers Defendants’ Exhibit 3 first, then considers Defendants’ Exhibit 4 and 8–13. The procedural history concerning Defendants’ Exhibit 3 is as follows. On March 17, 2020, Kumar filed Plaintiff’s Objections to Defendants’ Exhibits and Witnesses in the Joint Final Pre-Trial Order (Dkt. #79). In his Non-Motion, Kumar objected to the admission of, among other things, Morrison’s Supplemental Export Report (Dkt. #79). Defendants responded on March 25, 2020, by filing Defendants’ Response to Plaintiff’s Objections to Defendants’ Exhibits and Witnesses in the Joint Final Pre-Trial Order (Dkt. #84). At the Court’s Final Pretrial Conference, the Court heard arguments from both parties regarding whether the Court should admit Morrison’s Supplemental Expert Report (Dkt. #96). The Court instructed the parties that if Morrison’s Supplemental Expert Report was truly a supplement, the Court would admit it; however, if the Report was not a supplement, the Court would strike it (Dkt. #96). The parties were further instructed that if they ultimately decided to offer the Report, Defendants should offer Morrison’s Supplemental Expert Report at trial and the Court would then hear arguments on its admissibility (Dkt. #96). The Court reiterated that it would hold the experts to the opinions within their reports (Dkt. #96). Finally, the Court ordered Defendants to produce any documents supporting Morrison’s Supplemental Expert Report (Dkt. #96). 22 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 23 of 103 PageID #: 3142 At trial, Defendants offered Morrison’s Supplemental Expert Report (Defendants’ Exhibit 3) during their case-in-chief to argue, among other things, that Kumar could not establish a voting dilution claim. Kumar objected and claimed that Morrison’s Supplemental Expert Report was not a supplement. Rather, Kumar maintained that Morrison’s Supplemental Expert Report was an untimely new report masquerading as a “supplement” to the Expert Report of Peter Morrison (“Morrison’s Initial Expert Report”) (Defendants’ Exhibit 2). Finding it necessary to compare the two Reports, the Court conditionally admitted Defendants’ Exhibit 3 and stated that it would determine the admissibility of Morrison’s Supplemental Expert Report post-trial. Having reviewed Morrison’s Initial Expert Report, Morrison’s Supplemental Expert Report, and the governing Scheduling Order, the Court agrees with Kumar that Morrison’s Supplemental Expert Report should be stricken. Federal Rule of Civil Procedure 26(a) requires the production of certain initial disclosures. Rule 26(e) then imposes a duty to supplement Rule 26(a) disclosures when it states: (e) Supplementing Disclosures and Responses. (1) In General. A party who has made a disclosure under Rule 26(a)—or who has responded to an interrogatory, request for production, or request for admission—must supplement or correct its disclosure or response: (A) in a timely manner if the party learns that in some material respect the disclosure or response is incomplete or incorrect, and if the additional or corrective information has not otherwise been made known to the other parties during the discovery process or in writing; or (B) as ordered by the court. (2) Expert Witness. For an expert whose report must be disclosed under Rule 26(a)(2)(B), the party’s duty to supplement extends both to information included in the report and to information given during the expert’s deposition. Any additions or changes to this information must be disclosed by the time the party’s pretrial disclosures under Rule 26(a)(3) are due. FED. R. CIV. P. 26(e). Parties must make these expert-witness disclosures within the deadlines set by the Court’s Scheduling Order. State Auto. Mut. Ins. Co. v. Freehold Mgmt., Inc., 2019 WL 23 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 24 of 103 PageID #: 3143 1436659, at *21 (N.D. Tex. Mar. 31, 2019) (citing FED. R. CIV. P. 26(a)(2)(D)). Per Rule 26(a)(2), parties “must supplement these disclosures when required under Rule 26(e).” FED. R. CIV. P. 26(a)(2). For expert witnesses, this means that “[a]ny additions or changes [to an expert’s report or the information given during an expert’s deposition] must be disclosed by the time the party’s pretrial disclosures under Rule 26(a)(3) are due.” FED. R. CIV. P. 26(e)(2). Pretrial disclosures must be made at least 30 days prior to trial “unless the Court orders otherwise.” FED. R. CIV. P. 26(a)(3)(B) (emphasis added). With that being said, “[a] district court may grant a party leave to supplement an expert’s report after the deadline in the scheduling order has expired, but only if good cause is shown under Rule 16(b).” Freehold Mgmt., 2019 WL 1436659, at *21. The Court considers the same four factors under Rule 16(b) that it considers under Rule 37(c). See Geiserman v. MacDonald, 893 F.2d 787, 791 (5th Cir. 1990). Under Rule 37(c), “[i]f a party fails to provide information or identify a witness as required by Rule 26(a) or (e), the party is not allowed to use that information or witness to supply evidence on a motion, at a hearing, or at a trial unless the failure was substantially justified or harmless.” Torres v. City of San Antonio, 2014 WL 7339122, at *1 (W.D. Tex. Dec. 23, 2014). When evaluating whether a violation of Rule 26 is harmless for purposes of Rule 37(c)(1), the Court looks to four factors: (1) the explanation for the failure to disclose; (2) the importance of the testimony/evidence; (3) potential prejudice to the opposing party in allowing the testimony/evidence; and (4) the possibility of a continuance to cure such prejudice. Torres, 2014 WL 7339122, at *1; Hamburger v. State Farm Mut. Auto Ins. Co., 361 F.3d 875, 883 (5th Cir. 2004). Here, the Court’s Scheduling Order required the parties to provide a Joint Final Pretrial Order on February 28, 2020 (Dkt. #21). A trial date was initially set between May 1, 2020, and May 29, 2020 (Dkt. #21). Finally, the same Scheduling Order required Defendants to submit any 24 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 25 of 103 PageID #: 3144 expert reports by October 17, 2019 (Dkt. #21). On November 7, 2019, pursuant to an agreement between Defendants and Kumar extending the expert report deadline, Defendants filed Defendants’ Disclosure of Expert Witnesses (Dkt. #45). Attached as Exhibit A to Defendants’ Disclosure was Morrison’s Initial Expert Report (Dkt. #45, Exhibit A) (hereinafter cited as, “Defendants’ Exhibit 2”). Morrison’s Initial Expert Report is ten pages, includes one table (“Table 1”), no maps, and a one-page appendix further detailing Table 1 (Defendants’ Exhibit 2). On February 27, 2020, one day before the parties were to file a Joint Final Pretrial Order, Defendants produced Morrison’s Supplemental Expert Report (Defendants’ Exhibit 3). Morrison’s Supplemental Expert Report is twenty pages, contains thirteen maps, and is grounded in the purported need for further demographic evaluation. Having reviewed both reports, the Court finds it appropriate to strike Morrison’s Supplemental Expert Report because: (1) the Report is untimely; and (2) Defendants have not established good cause for the Court to grant leave under Rule 16(b). Morrison’s Supplemental Expert Report is untimely. Defendants claim that the Report is timely because it was filed more than 30 days prior to trial. Defendants are mistaken. Thirty days prior to the trial date is not the operative deadline in this matter. Under Rule 26(a)(3)(B), the operative deadline for supplementing an expert report is 30 days prior to trial absent a court order. In other words, the 30-day limit is a default subject to amendment by court order. Here, the Court ordered all expert reports to be filed by October 17, 2019. Any filing after said date—save the agreed-upon extension of the original date to November 7, 2019—required leave of court. See FED. R. CIV. P. 26(a)(3)(B). Defendants did not seek leave of court. Instead, Defendants unilaterally submitted Morrison’s Supplemental Expert Report the night prior to the deadline to submit the Joint Final Pretrial Order. Morrison’s Supplemental Expert Report is consequently 25 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 26 of 103 PageID #: 3145 untimely. As such, Defendants must carry their burden under Rule 16(b) and establish good cause for the Court to grant leave for this supplement. Defendants have not carried their burden. The Court finds that leave should not be granted because Morrison’s Supplemental Expert Report is a new expert report premised upon new opinions and conclusions based on facts available to Defendants “from the start of the case.” Newberry v. Disc. Waste, Inc., 2020 WL 363775, at *2 (E.D. Tex. 22, 2020). A review of the Rule 16(b) factors supports the Court’s conclusion. i. The Explanation for the Failure to Disclose Defendants maintain that Morrison’s Supplemental Expert Report was filed to respond to the deposition of Mr. David Ely (“Mr. Ely”)—one of Kumar’s experts. Defendants assert that it was only possible for them to issue this supplement post-deposition when they became aware of a need to respond to certain testimony. The Court is unconvinced. As previously discussed, the federal rules allow experts to supplement expert reports to account for otherwise incomplete reports. See FED. R. CIV. P. 26(e)(2). Supplemental disclosures, however, “are not intended to provide an extension of the deadline by which a party must deliver the lion’s share of its expert information.” In re Complaint of C.F. Bean L.L.C., 841 F.3d 365, 371 (5th Cir. 2016) (citing Sierra Club, Lone Star Chapter v. Cedar Point Oil Co., 73 F.3d 546, 571 (5th Cir. 1996)). In keeping with this principle, supplemental disclosures are only permissible as a means of “correcting inaccuracies[] or filling the interstices of an incomplete report based on information that was not available at the time of the initial disclosure.” Diaz v. Con-Way Truckload, Inc., 279 F.R.D. 412, 421 (S.D. Tex. 2012) (citing Cook v. Rockwell Int’l Corp., 580 F. Supp. 2d 1071, 1169 (D. Colo. 2006)) (emphasis in original). On cross examination, Dr. Morrison testified that, in his initial expert report, he failed to review whether the proffered illustrative district (hereinafter “Kumar’s Illustrative District” or 26 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 27 of 103 PageID #: 3146 “District”) respected communities of interest because he was unaware that such analysis was customary (TR4: 144–45).10 Following Defendants filing Morrison’s Initial Expert Report, Defense Counsel notified Dr. Morrison that Texas law existed which bore directly on respecting communities of interest in the rendering of illustrative districts. Consequently, following the discussion of communities of interest at Mr. Ely’s deposition, Dr. Morrison requested more information from FISD so he could offer a “supplement” to his initial report. Dr. Morrison admitted that the only reason the supplement was necessary was because he did not see the relevance of a communities of interest analysis when he constructed his initial report. Indeed, Dr. Morrison testified that his supplement was not predicated upon the existence of some new facts. The supplement was merely required because Dr. Morrison was unaware of law that existed prior to the filing of his initial report—law that was made known to him by Defense Counsel. With these underlying facts, it is apparent that Dr. Morrison was not attempting to correct an inaccuracy nor complete an outstanding issue in his report. See Diaz, 279 F.R.D. at 421. Dr. Morrison was instead taking a second crack at his report utilizing information available at the time of his initial disclosure as Diaz forbids. Id. This factor weighs against permitting Defendants’ untimely filing. ii. The Importance of the Testimony/Evidence Defendants claim that Morrison’s Supplemental Expert Report is important because it rebuts the expert opinions Kumar proffered. Kumar offers no counterargument as to the importance of the Report. Any such argument would be unconvincing regardless. As will be discussed below, Morrison’s Supplemental Expert Report offers new conclusions based upon new information—albeit, information that previously existed—which could strengthen FISD’s defense. It would blink reality for the Court not to find Morrison’s Supplemental Expert Report as an “TR” refers to Trial Transcript, 4 means day four, and 144–45 signifies the page numbers where the information can be found. Each transcript cite follows this pattern. 10 27 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 28 of 103 PageID #: 3147 important piece of potential testimony. As such, this factor weighs in favor of permitting Defendants’ untimely filing. iii. The Potential Prejudice to the Opposing Party in Allowing the Testimony/Evidence The potential prejudice to Kumar in allowing Morrison’s Supplemental Expert Report convinces the Court that the Report, irrespective of its importance, must be stricken. A comparison of Morrison’s Initial Expert Report to Morrison’s Supplemental Expert Report evidences why the Court’s conclusion is necessitated. In Morrison’s Initial Expert Report, Dr. Morrison claims that, among other things, Mr. Ely: (1) employed an unreliable hybrid methodology for determining the minority citizen voting age population; (2) cannot accurately distinguish Asian from non-Asian registered voters; (3) proffered a report that includes “logical mismatches between the number of persons 18 and older (in 2010) and his estimates of the number of citizens 18 and older (in years thereafter)”; (4) failed to use the Iterative Proportional Fitting method for disaggregating and reaggregating census block data; (5) misapplied the Lauderdale-Kestenbaum Asian surname lists which resulted in, at best, a “guesstimate” of the actual number of Asian voters in the District; and (6) damaged the reliability of Dr. Kenneth Mayer’s (“Dr. Mayer”) expert report which relied upon the estimates provided by Mr. Ely (Defendants’ Exhibit 2). In contrast to Morrison’s Initial Expert Report, Morrison’s Supplemental Expert Report asserts that: (1) Mr. Ely has not satisfied the first Gingles factor because he failed to respect traditional redistricting principles—such as maintaining communities of interest and traditional boundaries; (2) Mr. Ely disregarded the most obvious communities of interests in creating Kumar’s Illustrative District when he split numerous existing school attendance zones and local neighborhoods within FISD; (3) Kumar’s Illustrative District, as it stands, makes it mathematically impossible to draw a properly apportioned remedial proposal 28 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 29 of 103 PageID #: 3148 which allows for seven districts; and (4) Kumar’s Illustrative District fails to satisfy Gingles because it offers no lawful remedy under the first Gingles factor (Defendants’ Exhibit 3). As evidenced by a comparison of the two reports, Morrison’s Supplemental Exhibit Report presents new conclusions and opinions centered around Dr. Morrison’s new understanding of communities of interest analyses within the Fifth Circuit. Permitting Defendants to offer this new report as a “supplement” would prejudice Kumar. To be sure, Kumar lacked the appropriate time to properly consider the new report and prepare an adequate rebuttal due to Defendants’ untimely production of Morrison’s Supplemental Expert Report. Striking a report for these circumstances is not unheard of. On the contrary, courts within the Fifth Circuit recognize that making untimely, substantive changes to expert reports presents real prejudice to the other side, and courts routinely disallow litigants from making untimely disclosures. See, e.g., Reliance Ins. Co. v. La. Land & Expl. Co., 110 F.3d 253, 257–58 (5th Cir. 1997) (finding no abuse of discretion where the district court “concluded that ‘[t]o allow plaintiff to add more material now and create essentially a new report would prejudice the defendants, who would then have to get an expert to address these lastminute conclusions . . . .’”); see also Alexander J. Chern, “Hybrid Witnesses” Are Not Entitled to “Hybrid Rules”, 38 REV. LITIG. 333, 348 (2019) (internal quotations omitted) (“Disclosure of expert witnesses is more than mere procedure: The advance disclosure concerning expert witnesses protects the opposing party. Having received pretrial notice, the opposing party can prepare to respond to the expert testimony.”) (quoting Anne Bowen Poulin, Experience-Based Opinion Testimony: Strengthening the Lay Opinion Rule, 39 PEPP. L. REV. 551, 576 (2012)). Defendants’ attempt to make untimely, substantive changes to their expert report likewise threatens to prejudice Kumar. See Reliance, 110 F.3d at 257–58 (examining and affirming the district court’s decision 29 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 30 of 103 PageID #: 3149 to deny a motion to file a supplemental expert report). The Court finds this factor favors striking Morrison’s Supplemental Expert Report. iv. The Possibility of a Continuance to Cure Such Prejudice Finally, it is readily apparent that the possibility of a continuance to cure any prejudice that might befall Kumar has come and gone. A continuance is simply not a viable option. See Reliance, 110 F.3d at 258 (“District judges have the power to control their dockets by refusing to give ineffective litigants a second chance to develop their case.”) (citation omitted); State Auto., 2019 WL 1436659, at *26. Accordingly, these factors weigh against allowing Defendants’ untimely filing. Because three of the four good cause factors weigh against allowing Defendants to file their untimely expert report, Defendants’ Exhibit 3 is hereby stricken. C. Defendants’ Exhibits 4 and 8–13 The Court is finally presented with Defendants’ Exhibits 4 and 8–13. As previously discussed, the Court heard argument surrounding Plaintiff’s Objections to Defendants’ Exhibits and Witnesses in the Joint Final Pre-Trial Order (Dkt. #79) at the Court’s Final Pretrial Conference (Dkt. #96). In addition to considering the admissibility of Morrison’s Supplemental Expert Report, the Court also considered whether the Court should independently admit the Wassam Exhibits (Dkt. #96). After hearing arguments from both parties, the Court granted Kumar leave to depose Ms. Wassam to properly prepare for trial (Dkt. #96). The Court also permitted the parties to resume discovery and ordered that any documents supporting these opinions or exhibits be produced immediately (Dkt. #96). Any objections to the Wassam Exhibits, the Court concluded, were to be raised at trial (Dkt. #96). 30 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 31 of 103 PageID #: 3150 Defendants first witness at trial was Ms. Wassam. On direct, Ms. Wassam testified that she currently serves as the internal demographer for FISD and created each of the relevant exhibits for Defendants. Ms. Wassam explained that she relied upon Kumar’s Illustrative District to create these exhibits. Specifically, Ms. Wassam elaborated that she superimposed the Illustrative District over various FISD graphics to demonstrate how the Illustrative District affected certain communities of interest. Kumar objected claiming that the Wassam Exhibits were untimely. Defendants disagreed and claimed that the Wassam Exhibits were produced as soon as they were created. Defendants further countered that Defendants timely disclosed Ms. Wassam as a lay witness who had knowledge regarding communities of interest in FISD. The Court ultimately agrees with Defendants, and for the reasons discussed below, overrules Kumar’s objection. The Wassam Exhibits are untimely. The Court’s Scheduling Order mandated that discovery “shall” be completed by December 26, 2019. Nonetheless, Defendants produced the Wassam Exhibits on February 27, 2020—one day before the Joint Final Pretrial Order was due. Without leave of Court—which Defendants did not request—Defendants production of the Wassam Exhibits is untimely. With that being said, the Court finds that Defendants have established good cause to admit the evidence under Rule 37(c). i. The Explanation for the Failure to Disclose Defendants maintain that the Wassam Exhibits were produced as soon as they were created. Notably, Ms. Wassam and Defense Counsel contradicted one another in their representations to the Court on when the Wassam Exhibits were created. The Court takes Defense Counsel’s assertion, as an Officer of the Court, that he produced the Wassam Exhibits as soon as he obtained 31 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 32 of 103 PageID #: 3151 them as true. While Defendants ideally should have created the Exhibits earlier, the Court finds this factor neutral, or, at worst, slightly disfavoring admittance. ii. The Importance of the Testimony/Evidence The Wassam Exhibits aid the Court in understanding the demographics of the FISD community. Further, the Exhibits allow the parties to flesh out the testimony of each expert— whether on direct or cross examination—to allow the Court to make an educated decision in a complex, statistic-heavy case. This factor weighs in favor of admitting the Wassam Exhibits. iii. The Potential Prejudice to the Opposing Party in Allowing the Testimony/Evidence The potential prejudice to Kumar is minimal. At the Court’s Final Pretrial Conference, the Court granted Kumar leave to depose Ms. Wassam to fully understand, and prepare for, the Wassam Exhibits. The Court further ordered that Defendants produce any relevant documents underlying Morrison’s Supplemental Expert Report or the Wassam Exhibits. Importantly, the Wassam Exhibits were not created by an expert. Ms. Wassam is a lay witness who superimposed Kumar’s Illustrative District over a variety of previously produced maps held by FISD. The Wassam Exhibits do not include any conclusions, statistics, or opinions; they are simply maps. Unlike Morrisons’ Supplemental Expert Report, which would require a great deal of time, resources, and expert analysis to adequately prepare a rebuttal for, the Wassam Exhibits could be adequately addressed through a deposition and minimal expert review. Because the Court granted Kumar leave to depose Ms. Wassam to account for any prejudice, and the Wassam Exhibits— independent of Morrison’s Supplemental Expert Report—are not offered as expert testimony, the Court finds this factor favors admitting the Wassam Exhibits. 32 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 33 of 103 PageID #: 3152 iv. The Possibility of a Continuance to Cure Such Prejudice The possibility of a continuance to cure any prejudice that might befall Kumar has come and gone. What is more, the Court already mitigated the possibility of prejudice by permitting Kumar to depose Ms. Wassam prior to the commencement of trial. Having considered each of the above factors, the Court finds that the Wassam Exhibits should be admitted. Kumar’s objections as to each Exhibit are consequently overruled.11 III. Reliability of Mr. Ely’s Methodology The Court is next tasked with determining whether Kumar’s experts employed reliable methods when producing their opinions. The Court finds Mr. Ely’s methods are all acceptable save his reliance on the Lauderdale & Kestenbaum List of Asian Surnames which is unreliable and his probabilistic estimate of African American voter turnout which goes to the weight of the evidence, not its admissibility On September 26, 2019, Kumar filed the Expert Report of David Ely (the “Ely Report”) (Dkt. #34) (hereinafter cited as, “Plaintiff’s Exhibit 1”). In the Ely Report, Mr. Ely: (1) constructed an illustrative district; and (2) provided Dr. Mayer with a working data set for the second and third Gingles factors. In Section C of the Ely Report, Mr. Ely discussed the methodology he employed to construct Kumar’s Illustrative District. Mr. Ely used 2010 Census PL94-171 data (“2010 Census data”) and the American Community Survey’s (“ACS”) 2017 five-year data set (“20132017 ACS data”) to construct the illustrative district. Because the 2010 Census only provides citizen voting age population (“CVAP”) data at the Block Group level, Mr. Ely estimated the The Court acknowledges Kumar’s objection that the Wassam Exhibits are inaccurate. But Ms. Wassam is not an expert in this case, she is a lay witness. The Court will therefore consider Kumar’s objections as weight objections, not admissibility objections. 11 33 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 34 of 103 PageID #: 3153 African American and Hispanic share of CVAP at the Census Block (“Census Block”) level.12 The Ely Report also stated, which will be discussed further below, that Mr. Ely’s illustrative district analysis included “a breakdown of registered voters as described above” (Plaintiff’s Exhibit 1). This final statement references Section A of Mr. Ely’s Findings and Analysis which is entitled “Registered Voter Analysis.” In Section A, Mr. Ely discussed the methodology used to provide Dr. Mayer with a working data set for Dr. Mayer’s election polarization analysis. Mr. Ely used the Census Bureau’s List of Spanish Surnames and the Diane S. Lauderdale and Bert Kestenbaum Asian surname list (“Lauderdale & Kestenbaum List of Asian Surnames”)13 to estimate each ethnicity’s share of total voters in a given precinct. For African American surnames, Mr. Ely overlaid precinct geography over census geography to “produce estimates of CVAP composition for each precinct” (Plaintiff’s Exhibit 1). After determining which eligible voters did not have Spanish or Asian surnames, Mr. Ely used CVAP data to estimate the probability that the voter was African American.14 In the end, Mr. Ely created a proposed single-member district with a “combined Minority (including Multi-race and Other race) population of 55.2%, and a combined [CVAP] of 51.2%” “Blocks (Census Blocks) are statistical areas bounded by visible features, such as streets, roads, streams, and railroad tracks, and by nonvisible boundaries, such as selected property lines and city, township, school district, and county limits and short line-of-sight extensions of streets and roads. Generally, census blocks are small in area; for example, a block in a city bounded on all sides by streets . . . .” (Plaintiff’s Exhibit 1) (citing U.S. CENSUS BUREAU, P. L.94171 REDISTRICTING DATA FOR THE YEAR 2010 CENSUS: GEOGRAPHIC TERMS AND CONCEPTS). “Block Groups [referred to as “BG” in this footnote] are statistical divisions of census tracts, are generally defined to contain between 600 and 3,000 people and are used to present data and control block numbering. A block group consists of clusters of blocks within the same census tract that have the same first digit of their four-digit census block number . . . . A BG usually covers a contiguous area. Each census tract contains at least one BG, and BGs are uniquely numbered within the census tract. Within the standard census geographic hierarchy, BGs never cross state, county, or census tract boundaries but may cross the boundaries of any other geographic entity.” (Plaintiff’s Exhibit 1) (citing U.S. CENSUS BUREAU, P. L.94-171 REDISTRICTING DATA FOR THE YEAR 2010 CENSUS: GEOGRAPHIC TERMS AND CONCEPTS). 13 Labeling the Asian surname list the “Lauderdale & Kestenbaum List of Asian Surnames” is a bit of a misnomer. As will be discussed, Lauderdale and Kestenbaum actually produced 24 lists across the 6 major Asian American populations. For simplicity’s sake, however, the Court uses the singular “list” to refer to Lauderdale and Kestenbaum’s work in general. The Court discusses individual “lists,” however, when relevant. 14 Census data and ACS data include self-identification. Surname registries do not. See TR2: 95–96; 129. 12 34 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 35 of 103 PageID #: 3154 (Plaintiff’s Exhibit 1). Absent the Multi-race and Other-race populations, Kumar’s Illustrative District registered a CVAP of 50.03 as seen below in Table 1 of the Ely Report. Table 1: Illustrative District Demographics On November 7, 2019, Defendants filed Defendants’ Objections to Plaintiff’s Experts (Dkt. #46). Defendants asserted that Mr. Ely’s opinions were unreliable and should be excluded (Dkt. #46). More specifically, Defendants claimed that Mr. Ely’s opinions were unreliable because Mr. Ely: (1) utilized Hispanic and Asian surname lists which have a high error rate; (2) speculated on which voters in FISD are Black; and (3) employed a hybrid methodology to classify voters by race/ethnicity (Dkt. #46). On November 21, 2019, Kumar filed Plaintiff’s Response to Defendants’ Objections to Plaintiff’s Experts (Dkt. #53). Kumar countered that Defendants were mistaken and Mr. Ely’s methodology was reliable (Dkt. #53; Plaintiff’s Exhibit 4; Plaintiff’s Exhibit 5). On March 27, 2020, the Court entered its Memorandum Opinion and Order overruling Defendants’ Objections (Dkt. #85). In overruling Defendants’ Objections, the Court did not 35 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 36 of 103 PageID #: 3155 discuss the reliability of Mr. Ely’s methodology. Rather, the Court grounded its holding in the Court’s ability to “admit questionable technical evidence [while not giving that evidence] more weight than it deserves” (Dkt. #85) (citing Harding v. Cty of Dallas, Tex., 2018 WL 1156561, at *1 (N.D. Tex. Mar. 5, 2018) (citing Gibbs v. Gibbs, 210 F.3d 491, 500 (5th Cir. 2000))). The Court accordingly proceeded to trial without excluding Mr. Ely’s expert opinions and reserved the right to determine the reliability of the evidence post-trial. At trial, Defendants again raised the issue of Mr. Ely’s: (1) use of Hispanic and Asian surname lists; (2) purported speculation in estimating which voters are Black; (3) use of a supposed “hybrid methodology” to estimate the vote share among each ethnic group; and (4) alleged failure to use “best practices” when forming Kumar’s Illustrative District. The Court now considers whether the methods Mr. Ely employed are reliable. Federal Rule of Evidence 702 provides for the admission of expert testimony that assists the trier of fact to understand the evidence or to determine a fact in issue. FED. R. EVID. 702. In Daubert v. Merrell Dow Pharmaceuticals, Inc., the Supreme Court instructed courts to function as gatekeepers when determining whether expert testimony should be presented to the jury. 509 U.S. 579, 590–93 (1993). Courts act as gatekeepers of expert testimony “to make certain that an expert, whether basing testimony upon professional studies or personal experience, employs in the courtroom the same level of intellectual rigor that characterizes the practice of an expert in the relevant field.” Kuhmo Tire Co. v. Carmichael, 526 U.S. 137, 152 (1999). The party offering the expert’s testimony has the burden to prove that: (1) the expert is qualified; (2) the testimony is relevant to an issue in the case; and (3) the testimony is reliable. Daubert, 509 U.S. at 590–91. A proffered expert witness is qualified to testify by virtue of his or her “knowledge, skill, experience, training, or education.” FED. R. EVID. 702. Moreover, to be 36 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 37 of 103 PageID #: 3156 admissible, expert testimony must be “not only relevant but reliable.” Daubert, 509 U.S. at 589. “This gate-keeping obligation applies to all types of expert testimony, not just scientific testimony.” Pipitone v. Biomatrix, Inc., 288 F.3d 239, 244 (5th Cir. 2002) (citing Kuhmo, 526 U.S. at 147). In deciding whether to admit or exclude expert testimony, the Court should consider numerous factors. Daubert, 509 U.S. at 594. In Daubert, the Supreme Court offered the following, non-exclusive list of factors that courts may use when evaluating the reliability of expert testimony: (1) whether the expert’s theory or technique can be or has been tested; (2) whether the theory or technique has been subjected to peer review and publication; (3) the known or potential rate of error of the challenged method; and (4) whether the theory or technique is generally accepted in the relevant scientific community. Id. at 593–94; see also Pipitone, 288 F.3d at 244 (same). When evaluating Daubert challenges, courts focus “on [the experts’] principles and methodology, not on the conclusions that [the experts] generate.” Daubert, 509 U.S. at 595. The Daubert factors are not “a definitive checklist or test.” Id. at 593. As the Supreme Court has emphasized, the Daubert framework is “a flexible one.” Id. at 594. The test for determining reliability can adapt to the particular circumstances underlying the testimony at issue. Kuhmo, 526 U.S. at 152. The decision to allow or exclude experts from testifying under Daubert is committed to the sound discretion of the district court. St. Martin v. Mobil Expl. & Producing U.S., Inc., 224 F.3d 402, 405 (5th Cir. 2000) (citations omitted). Defendants challenge four of Mr. Ely’s methods. The Court considers each method separately.15 15 Boilerplate objections not renewed at trial are denied. See United My Funds, LLC v. Perera, 2020 WL 1225042, at *5 (E.D. Tex. 2020) (stating that the Court denies boilerplate objections universally). 37 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 38 of 103 PageID #: 3157 A. Hispanic & Asian Surnames Defendants first challenge Mr. Ely’s use of Hispanic and Asian surname data in his report. Before considering whether the use of such data is reliable, the Court finds it prudent to discuss how Mr. Ely used the surname data. Mr. Ely utilized Hispanic and Asian surname data to construct a data set that Dr. Mayer could use for his election polarization analysis. While the Ely Report states that Mr. Ely used surname data to construct an illustrative district, that statement has since been corrected. Mr. Ely provided a Declaration, now Plaintiff’s Exhibit 4, which clarified that the mention of surname data in Section C of his Report was inadvertently included. Any reference to Section A in Section C should thus be disregarded because Mr. Ely’s illustrative district analysis “did not include a review of any voter registration list” (Plaintiff’s Exhibit 4). The exclusion of surname data in the construction of Kumar’s Illustrative District was reiterated at trial: Q. In creating your illustrative district, you only used census and ACS data to identify the CVAP of Asians, Blacks and Hispanics in Frisco ISD, is that correct? A. That’s correct. Q. But instead of just using census data, like you did to create your district, you used surnames for Asians and Hispanics from a list of registered voters, combined with census and ACS CVAP data for Blacks to create the database for Dr. Mayer to use for his analysis, correct? A. Correct. (TR2: 92). As evidenced, Mr. Ely’s use of Hispanic and Asian surname data was limited to constructing a working data set for Dr. Mayer. With the scope of Mr. Ely’s surname use understood, the Court turns to the reliability of each data set. i. Census Bureau’s List of Spanish Surnames Defendants argue that Mr. Ely’s opinions are, in part, unreliable due to his use of the Census Bureau’s List of Spanish Surnames. Defendants’ objection is directed toward the data that 38 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 39 of 103 PageID #: 3158 underlies Dr. Mayer’s opinions as to the second and third Gingles factors. The Court recognizes that the use of surname registries is questionable and entitled to lesser weight. Nevertheless, the Court finds that Kumar has demonstrated that the data Mr. Ely used to construct Dr. Mayer’s data set is sufficiently probative. The Court consequently declines to exclude Dr. Mayer’s testimony and instead permits the testimony while affording the evidence he relies upon reduced weight. In the Fifth Circuit, there is a preference for experts to utilize Census data in voting rights cases. See Reyes, 2008 WL 4791498, at *9. Census data, however, is not always used. In its place, experts sometimes use Spanish surnames as a surrogate for Hispanic origin. See Clements, 986 F.2d at 774, 868–73. The Fifth Circuit is dubious, at best, when it comes to the use of surname registries. Indeed, the Fifth Circuit has stated that the use of Spanish surname registration: (1) tends to “misidentify Hispanic persons as non-Hispanic and vice-versa”; (2) is “novel”; (3) is “highly problematic”; and (4) is “disfavored.” Rodriguez v. Bexar County, 385 F.3d 853, 867 n.18 (5th Cir. 2004) (citing United States v. Alamosa Cnty., 306 F. Supp. 2d 1016, 1022 (D. Colo. 2004) (Krieger, J.)). The Fifth Circuit concluded that “census data based upon self-identification provides the proper basis for analyzing Section 2 vote dilution claims in the future.” Id. Nonetheless, an expert’s use of Spanish surname registration is not per se fatal. Id. Should the party offering the evidence demonstrate a “strict showing of probativeness,” a court may rely upon the evidence while affording the evidence “reduced weight” and noting “that self-identification data provides a more reliable means of determining ethnicity.” Id. (citing Alamosa Cnty., 306 F. Supp. 2d at 1022); see also Reyes, 2008 WL 4791498, at *9 (requiring the SSRV data utilized by the TLC to be “sufficiently probative” before it could be considered by the court). Surname registries are not the only tools that courts, at times, permit experts to consider. For instance, some courts find ACS CVAP data sufficiently probative to be considered. ACS 39 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 40 of 103 PageID #: 3159 CVAP data is found, at times, to be sufficiently probative because: (1) “the Census Bureau no longer collects [CVAP] data on the official Census forms”; and (2) “ACS data is perhaps the best measure of citizen voting age data currently available; it is collected by the Census Bureau and the Census Bureau’s publication of and reliance on ACS data ‘suggests that the Bureau considers ACS data reliable and intends for it to be relied upon in decisions such as Voting Rights Act compliance.’” Rodriguez v. Harris Cnty., Tex., 964 F. Supp. 2d 686, 727–28 (S.D. Tex. 2013) (“In this case, the Court finds the ACS data sufficiently probative on the issue of [CVAP].”); see also Harding v. Cnty. Of Dallas, Tex., 336 F. Supp. 3d 677 (N.D. Tex. 2018) (same); Fabela v. City of Farmers Branch, 2012 WL 3135545, at *4–8 (N.D. Tex. Aug. 2, 2012) (same); Benavidez I, 638 F. Supp. 2d at 721 (same). Before addressing the evidence in question, the Court first considers why Mr. Ely relied upon the Census Bureau’s List of Spanish Surnames. Mr. Ely employed surname matching to estimate the total Hispanic voting population within Kumar’s Illustrative District. Mr. Ely claimed this methodology was required because “voters in [FISD] were not required to self-identify race when voting” (Plaintiff’s Exhibit 5). Mr. Ely further clarified that his task in constructing Dr. Mayer’s data set was not to estimate the percentage of eligible voters—which would be inadequate to analyze election results. Rather, Mr. Ely was tasked with estimating the ethnicity of the actual voters in each relevant election. Given his task, Mr. Ely chose to employ the Census Bureau’s List of Spanish Surnames rather than ACS CVAP data. 16 According to Mr. Ely, ACS CVAP data is directed toward eligible voters which makes it an unreliable source for determining actual voters. If Mr. Ely had used ACS CVAP data to determine actual voters, then he would have compiled a data set which overestimated the actual voters in each ethnic group examined. And without reliable 16 Neither Mr. Ely nor Dr. Morrison discussed why Mr. Ely did not consult Census data from 2010. The Court discusses why the use of this data would be, to a degree, unreliable, infra. 40 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 41 of 103 PageID #: 3160 turnout rates for each ethnic group to attempt to remedy any inaccuracies, any data set would be plagued with inaccuracies. This issue would be further exacerbated by the fact that ACS CVAP data is less reliable for smaller geographic precincts. Rather than employ a methodology riddled with discrepancies, Mr. Ely opted to consult surname data while looking at the actual voters who voted in given precincts to determine the ethnicity of those voters. Although Dr. Morrison contends that Mr. Ely should have utilized available ACS CVAP data, the Court is satisfied that Mr. Ely’s decision to forgo ACS CVAP data was well-reasoned. While Dr. Morrison takes issue with Mr. Ely’s decision to not utilize ACS CVAP data, nothing compels an expert to use said data. On the contrary, the Fifth Circuit “has not decided whether the five-year aggregation of ACS data is always sufficient to establish the [CVAP].” Rodriguez, 964 F. Supp. 2d at 727. And courts that have permitted reliance on ACS CVAP data have not indicated that it is the only data that may be used. Id. (finding that ACS CVAP data is the “only reliable indicia of [CVAP] in the record”) (emphasis added). The Court now looks at how Mr. Ely employed his methodology. Mr. Ely consulted the Census Bureau’s List of Spanish Surnames, the Lauderdale & Kestenbaum List of Asian Surnames, and ACS CVAP data for African American voters—which will be discussed in Part IV(B), infra—to code each registered voter based off his or her ethnicity within a precinct. Hispanic and Asian voters were coded as either: (1) a Spanish surnamed registered voter (“SSRV”); or (2) an Asian surnamed registered voter (“ASRV”) (Plaintiff’s Exhibit 1). Mr. Ely then aggregated each registered voter who voted in a given election “to produce a count for each precinct in each election of total voters” (Plaintiff’s Exhibit 1). Hence, voters were next coded at the precinct-level as: (1) Spanish surnamed voters (“SSV”); or (2) Asian surnamed voters (“ASV”). While Mr. Ely’s methods do not include Census or ACS CVAP data, 41 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 42 of 103 PageID #: 3161 the Court finds the data produced by Mr. Ely is sufficiently probative such that it may be considered. The Court reaches its conclusion for a number of reasons. First, like Reyes, this is a case that was filed well after the most recent Decennial Census. In Reyes, the case was filed in 2007, “many years after the 2000 Census data [was] collected.” Reyes, 2008 WL 4791498, at *9. Here, Kumar filed this action in 2019, some nine years post-collection of the 2010 Census data. As Reyes recognized, “[c]ourts in the Fifth Circuit have noted that Census data [is] less probative the further away from the Census cycle one gets.” Id. (citing Perez v. Pasadena Indep. Sch. Dist., 958 F. Supp. 1196, 1212–13 (S.D. Tex. 1997)). Second, numerous courts have found the Census Bureau’s List of Spanish Surnames sufficiently reliable to consider absent recent Census data. See, e.g., Luna v. Cnty. of Kern, 291 F. Supp. 3d 1088, 1119 (E.D. Ca. 2018); Patino v. City of Pasadena, 230 F. Supp. 3d 667, 668 (S.D. Tex. 2017), Fabela, 2012 WL 3135545, at *4–8; see also Reyes, 2008 WL 4791498, at *9 (“In addition, the evidence shows that the Spanish surname list used by the TLC in identifying SSRVs has been published and heavily tested by the U.S. Census Bureau.”). Third, Dr. Morrison recognized the Census Bureau’s List of Spanish Surnames as a list that meets “minimum scientific standards of construction and validation” (Plaintiff’s Exhibit 5) (citing ALLAN F. ABRAHAMSE, PETER A. MORRISON, AND NANCY MINTER BOLTON, 19 POPULATION RES. AND POL’Y REV., 283–300 (2000) (“Surname Article”)). Dr. Morrison went on to say that the Census Bureau’s List of Spanish Surnames “has been carefully developed and has undergone thorough evaluation” (Plaintiff’s Exhibit 5) (citing Surname Article)). The Census Bureau’s List of Spanish Surnames, while disfavored, has been tested, subjected to peer review, determined to have an acceptable rate of error, 17 and accepted in the relevant scientific community. The Court recognizes that the Census Bureau’s List of Spanish Surnames produces false positives and false negatives. These false identifications can be attributed to: (1) the surname commission (“SCOM”) rate—the 17 42 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 43 of 103 PageID #: 3162 Daubert, 509 U.S. at 594; Pipitone, 288 F.3d at 244. “The Court finds that it can consider this type of data when, as here, Census data [is outdated] and therefore less likely to be accurate.” Reyes, 2008 WL 4791498, at *9 (citing Rodriguez, 385 F.3d at 867). ii. Lauderdale & Kestenbaum List of Asian Surnames Defendants also contest Mr. Ely’s reliance upon the Lauderdale & Kestenbaum List of Asian Surnames. Like the Census Bureau’s List of Spanish Surnames, Defendants claim that the Lauderdale & Kestenbaum List of Asian Surnames only provides for “guesstimates” of the Asian population. The Court finds that Kumar has not made a strict showing of probativeness as to this evidence. The Lauderdale & Kestenbaum List of Asian Surnames was created after Lauderdale and Kestenbaum recognized that “[t]here [were] no Asian surname lists with a similar level of acceptance or recognition [to the Census Bureau’s List of Spanish Surnames]” (Plaintiff’s Exhibit 1, Appendix B). Lauderdale and Kestenbaum thus embarked to establish a set of acceptable Asian surname lists that would enable demographers, and the like, to accurately use surnames to infer ethnic identification. Lauderdale and Kestenbaum created six lists based on the six largest Asian American ethnic groups: Chinese, Korean, Japanese, Vietnamese, Asian Indian, and Filipino. These lists were derived for two data contexts. First, there are conditional lists. Conditional lists limit the inferences on ethnicity one can make based upon race data availability. Second, there are unconditional lists. Unconditional lists can be used absent records including race classification. The lists were further categorized as either predictive or strongly predictive. Predictive lists percentage of citizens whose surnames appear on the Census Bureau’s List who otherwise reported as “Non-Hispanic” on the Decennial Census Spanish Origin Research (“SOR”) file; and (2) the surname omission (“SOM”) rate—the percentage of citizens who reported themselves as Hispanic but nonetheless do not appear on the List (Defendants’ Exhibit 2). According to Dr. Morrison, the List has a “10.02 percent SCOM rate and a 20.55 percent SOM rate” (Defendants’ Exhibit 2). The Court accounts for these recorded statistical discrepancies when considering the evidence before it. See Rodriguez, 385 F.3d at 867 n.18. 43 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 44 of 103 PageID #: 3163 include surnames from each list where “at least 50 percent of persons with that surname [are] associated with an origin (e.g., Korea) and less than 50 percent with any of the other countries” (Plaintiff’s Exhibit 1, Appendix B). Strongly predictive lists include surnames from each list where at least 75 percent of persons are associated with an Asian origin. This conditionspecific/predictive or strongly predictive breakdown resulted in the creation of 24 lists: “two sets (predictive and strongly predictive) of two types (conditional and unconditional) for six Asian American groups” (Plaintiff’s Exhibit 1, Appendix B). A list that is strongly predictive has reduced coverage, and vice versa. “Thus[,] the two sets of lists are suited to different applications, dictated by the importance of accuracy (e.g., being surer of a person’s Chinese ethnicity versus detecting a higher proportion of Chinese persons)” (Plaintiff’s Exhibit 1, Appendix B). As in the case of Mr. Ely’s use of the Census Bureau’s List of Spanish Surnames, the Court finds that Mr. Ely’s decision to forgo ACS CVAP data was well-reasoned. Deciding whether the Lauderdale & Kestenbaum List of Asian Surnames is reliable, and whether Mr. Ely’s means of using said List was permissible, is a more difficult issue. Mr. Ely did not specify which lists he used when constructing the data set he provided Dr. Mayer. And even after Dr. Morrison critiqued Mr. Ely’s methodology and called into question which lists Mr. Ely used, Mr. Ely failed to clarify which lists he relied upon. Instead, Mr. Ely responded by stating that he used the lists before with a conclusory statement that these lists are peer-reviewed, tested, and widely accepted. Mr. Ely provided no authority corroborating any of his statements. The Court is subsequently left to guess which lists Mr. Ely used to determine their reliability. This is not difficult. There are only a finite number of lists. At best, the lists Mr. Ely relied upon are 75 percent accurate. At worst, Mr. Ely used lists that produced a correct estimation only 50 percent of the time. This leaves a great deal to be desired—as do the lists themselves. 44 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 45 of 103 PageID #: 3164 Lauderdale and Kestenbaum explicitly recognize the inherent limitations in their work. For instance, the authors note that, for the Indian and Korean origins, the sensitivity of their predictive lists is between 60 and 64 percent. Lauderdale and Kestenbaum then compare their lists to the Census Bureau’s List of Spanish Surnames and state that their “conditional Chinese, Japanese, and Vietnamese predictive lists have accuracy similar to and coverage somewhat lower than the Spanish list, while the coverage for the other three lists is considerably poorer” (Plaintiff’s Exhibit 1, Appendix B). “Without race information,” the authors continue, “the Asian lists do not identify as high a percentage of the population as does the Spanish surname list. The primary reason is that many frequently-occurring Asian surnames either are not unique to a single Asian ethnic group or are not uniquely Asian” (Plaintiff’s Exhibit 1, Appendix B) (emphasis added). Further exacerbating the issue of reliability is the authors’ inability to quantify how marriage outside one’s specific ethnic group affects the data pool due to concomitant name changes. And, if that is not enough, the authors note that varying socio-economic status among Asian groups can affect cohesion within the aggregate. This leads the authors to recommend that future studies relying upon their work be limited to studies focused on separate Asian ethnic groups. Despite the panoply of issues with the Lauderdale & Kestenbaum List of Asian surnames, Mr. Ely claims that the Court should be undeterred from considering his data. The Court disagrees. The Lauderdale & Kestenbaum List of Asian Surnames may be acceptable in some fields, but the pros of the List are simply too outweighed by the cons for the Court to find the data the List produces reliable. The Lauderdale and Kestenbaum List of Asian Surnames consequently cannot meet the strict showing of probativeness that the Fifth Circuit mandates. See Rodriguez, 385 F.3d at 867 n.18. Mr. Ely’s failure to specify which lists he used, coupled with the error rate that each list reports, is enough to be fatal. But if that is not enough, Mr. Ely’s failure to account for differing 45 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 46 of 103 PageID #: 3165 sensitivities among ethnicities compels the Court’s conclusion. The Court is a jack of all trades and master of none. In other words, the Court is a generalist—a generalist which relies on specialists where the Court otherwise lacks expertise. When the Court is presented with a technical issue, such as one concerning demographic data, the Court necessarily turns to expert opinions proffered by the parties to properly interpret the issues before it. In considering these expert opinions, the Court cannot consider mere guesswork. Each opinion offered, method employed, and expert who testified must be reliable. See Daubert 509, U.S. at 594. Here, the Lauderdale & Kestenbaum List of Asian Surnames warns that those who utilize the surname lists to identify an aggregated group of Asian Americans—such as a group of Asian Americans within a specific electoral precinct—must adjust “for differences in the sensitivity of the six lists” (Plaintiff’s Exhibit 1, Appendix B). Failure to account for these sensitivities results in ethnicities with greater surname sensitivity being over-represented in the aggregate group. Mr. Ely ignored Lauderdale and Kestenbaum’s stark warning. Mr. Ely’s testimony at trial is telling: Q. Instead, for the Gingles 2 and 3 database, you used surname lists, correct? A. I used actual voter lists and surname analysis on those lists. Q. And the surname -- and you used the Asian surname list that we just talked about that has a 60 to 64 percent sensitivity rate, correct? A. That’s correct. Q. That you made no adjustments for, despite the authors’ recommendation that you do so, correct? A. I made no adjustments. (TR2: 101–02). The Ely Report lumps all six Asian American groups together without accounting for any sensitivity among the groups. Mr. Ely’s decision to ignore Lauderdale and Kestenbaum’s best-practice advice greatly undermines the reliability of Mr. Ely’s work and further supports the Court’s conclusion. 46 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 47 of 103 PageID #: 3166 The Court finds the Lauderdale & Kestenbaum List of Asian Surnames error-prone and Mr. Ely’s methodology in applying the List faulty. The Court accordingly declines to consider Dr. Mayer’s opinions to the extent they rely upon data derived from Mr. Ely’s use of the Lauderdale & Kestenbaum List of Asian Surnames. B. African American Voters Defendants also object to Mr. Ely’s methodology to determine African American voters. Specifically, Defendants allege that Mr. Ely provided little explanation on how he classified voters as African Americans. Defendants go so far as to claim that Mr. Ely merely speculated on the identity of African American voters in FISD. Defendants conclude that the Court should exclude Mr. Ely’s estimates as unreliable. The Court disagrees. As previously stated, Mr. Ely consulted the Census Bureau’s List of Spanish Surnames and the Lauderdale & Kestenbaum List of Asian Surnames to estimate eligible Hispanic and Asian voters. Mr. Ely then aggregated each Hispanic and Asian registered voter who voted in a given election “to produce a count for each precinct in each election of total voters” (Plaintiff’s Exhibit 1). This allowed Mr. Ely to determine the actual voters in each precinct election. As for African American voters and “Other” voters—“Other” is defined to include non-Asian, non-Hispanic, and non-African American voters—Mr. Ely consulted ACS CVAP data to create a probabilistic estimate of eligible African American and Other voters. Mr. Ely resorted to probabilistic estimates of ACS CVAP data because African Americans do not have unique ethnic surnames. Mr. Ely took these estimates, in the form of percentages, and applied the percentages to actual voter data, at the individual level, without Spanish or Asian surnames. Critically, this meant that Mr. Ely was not estimating actual voter turnout, but rather what race/ethnicity each voter was based off turnout data. By using this method, Mr. Ely claimed he was able to estimate the actual amount of African 47 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 48 of 103 PageID #: 3167 American and Other voters in relevant elections once the data was aggregated at the precinct level while maintaining the integrity of actual voter lists. Finally, Mr. Ely, combined the estimates of actual African American, Hispanic, Asian, and Other voters and provided the actual voter estimates to Dr. Mayer. As an initial matter, the Court finds the ACS CVAP data used here sufficiently probative for the reasons set forth in Rodriguez. 964 F. Supp. 2d at 727–28; see also Fabela, 2012 WL 3135545, at *4–8 (finding ACS CVAP data the “most accurate data readily available”). The troubling aspect of Mr. Ely’s methods, however, is his probabilistic estimate that is predicated upon an assumption that African Americans and Whites turn out in elections in similar percentages. Mr. Ely readily admits that after he coded each voter as Hispanic or Asian, he was only able to calculate the probability that each remaining voter was African American. In estimating the race/ethnicity of each of the remaining voters, Mr. Ely proceeded with the underlying assumption that African American voters and Other voters turn out in elections in similar percentages. Mr. Ely expanded upon this at trial: Q. Okay. Can you briefly explain to us, very briefly, how you compiled that data? … A. So for each of the elections that I have selected election data, I select out of the registered voter file those voters who are flagged as having voted in that election. I use the surname analysis that I described earlier to identify particular voters as either Spanish surname or Asian surname. And for the voters who are not flagged as being either Spanish surname or an Asian surname, I divide -- well, I give each individual, in a sense, a probability of being Black or White based on the composition of the precinct citizen voting age population. Basically, it’s been my experience in this kind of analysis, the turnout rates among different ethnic groups can vary pretty dramatically, and that’s part of the reason why, especially for Asian and Latino voters, using the surname matching methodology is by far the most reliable method to get an accurate estimate or a reliable estimate of the actual voters and why I would -- I would only use citizen voting age based estimates for those groups if I did not have the availability of surname matching the individual voters. 48 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 49 of 103 PageID #: 3168 For White and African-American voters, typically, although there are large turnout differences in some elections, over a large area, usually those turnout differences are localized, and within a given local area, the turnout rates between Whites and AfricanAmericans is pretty close to the same. And so when I’m looking at the actual voters in a precinct who do not have either an Asian or a Spanish surname, I use the percentage breakdown between Black and White voters -Black and White citizen voting age population to assign a probability of this voter being either -- either Black or White. And the result of this is that I can aggregate at the precinct level an estimate of -- of Spanish surname, Asian surname, Black and White voters. There are small – you know, other categories that are in there, but they’re too small statistically to be able to meaningfully include in this analysis. --Q. Well, with Blacks in the dataset that you gave to Dr. Mayer, you estimated how many of the people who voted and don’t have ethnic surnames are Black, correct? A. That’s correct. Q. And this I believe you call a probabilistic estimate of Black voters, correct? A. Yes. Q. And that probabilistic estimate assumes that Black voters and White voters vote in the same percentage as their CVAP, correct? A. As their percentage of the non-Asian, non-Hispanic CVAP, that’s correct. Q. And I believe in your direct exam with Mr. Collins, you said that the -- the Whites and Black voting percentages, their turnouts were pretty close, is that correct? A. I said that they are often significantly different for a jurisdiction, but that at the local level they generally are pretty close. So you have high turnout precincts and low turnout precincts, but the Blacks and whites within each of those precincts tend to turn out at roughly the same rate. Q. Did you do that specific analysis for Black voters in Frisco ISD? A. No, I did not. Q. So your statement that Black voters and White voters show up in equal percentage, roughly equal percentages to their CVAP is just an assumption on your part, not based on any data or study of Frisco ISD voters, correct? A. Not of Frisco ISD voters. I don’t believe this data exists to do that analysis. (TR2: 66–68; 93–94). Based on Mr. Ely’s testimony, the Court is presented with an expert opinion that is predicated upon, to a degree, some unsupported, generalized assumptions about voter 49 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 50 of 103 PageID #: 3169 turnout at precinct levels. While the Court finds this portion of Mr. Ely’s methodology troubling, it does not find his probabilistic estimates incurable. The Court generally finds Mr. Ely’s estimate of African American voters is based on reliable ACS CVAP data and well-reasoned methodology, save Mr. Ely’s turnout assumption. Rather than excluding Dr. Mayer’s opinion to the degree that it is based on this evidence, the Court finds it appropriate to attribute a lesser weight to this evidence to account for any discrepancies in Mr. Ely’s probabilistic estimates.18 See Rodriguez, 385 F.3d at 867 n.18. C. Hybrid Methodology Defendants next allege that Mr. Ely departed from standard practice and employed an untested, hybrid methodology to combine surname data with ACS CVAP data into a working data set for Dr. Mayer. This “apples and oranges” approach, Defendants continue, “has not been published, has not been peer reviewed, and has never been successfully used in a voting rights case involving either Asian plaintiffs or a coalition of more than two minority groups” (Dkt. #122). The Court does not share in Defendants’ dismay. Setting aside the Court’s decision regarding the Lauderdale & Kestenbaum List of Asian Surnames for a moment, the Court does not find Mr. Ely’s “hybrid methodology” necessitates the exclusion of Dr. Mayer’s testimony. The Court need not labor to reach this conclusion. At trial, Dr. Mayer responded to the claim that Mr. Ely could not combine surname and census data with a The Court finds it appropriate to rely on Mr. Ely’s African American voter evidence while attributing a lesser weigh to it, in part, because of Dr. Mayer’s testimony. Dr. Mayer opined that based upon his understanding of the academic literature surrounding voter turnout, groups of individuals of similar socioeconomic status tend to have equivalent voter turnout. While at the national- or state-level this academic conclusion does not always reflect reality, at the precinct level this conclusion finds more teeth. Dr. Mayer testified that data from the same academic literature reveals that socioeconomic factors among varying ethnicities are often similar at the precinct level. Given this finding, it naturally follows that Whites and African Americans are more likely to share similar socioeconomic status and thus turnout to vote in equivalent numbers when at the precinct level. Mr. Ely’s work, consequently, should not be entirely disregarded as mere guesswork; rather, it should merely receive lesser weight. 18 50 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 51 of 103 PageID #: 3170 simple response: “that is done all the time in social science” (TR4: 151). Dr. Mayer then elaborated: In this particular case when you’re combining surname data when you have it, which is -is known to be more accurate because it’s an actual measure of people who vote. When you lack that data, you have a couple of choices. One is that you can abandon the accurate data that you do have and use a less accurate form of data, which would be just a broad estimate of the citizen voting age population, or you can treat it as what would be called the missing data or amputation problem and look for alternative ways of measuring the quantities that you need. And in this case, using voter turnout data at the precinct level, based on the demographics of that precinct, in cases where you lack surname analysis, which would be true in the case of African-American and non-Hispanic Whites in this case, that’s an entirely appropriate methodology. And the -- the position that you would need to do it entirely with citizen voting age population, that is a method that would be known to produce incorrect results, and what absolutely makes no sense is to reject accurate data that you do have because you don’t have accurate data for the entire scope of information that you need. Again, the -- the practice of combining data from different sources based on what’s available and what you’re trying to measure, this is something that is done all of the time. And the notion that it is somehow unacceptable in this case I think -- I don’t quite understand what that position is, because it’s just so inconsistent with social science standards and practices in this area. (TR4: 151–52). It is apparent from Dr. Mayer’s testimony that the Court does not need a journal article to find Mr. Ely’s methodology reliable in this instance. Mr. Ely adequately demonstrated why he used surname data for Hispanics and Asian Americans on the one hand and ACS CVAP data for African Americans on the other. It naturally follows that Mr. Ely had to combine his data sets to provide the working data set for Dr. Mayer’s election polarization analysis. While Dr. Morrison and Dr. Alford may have some consternation about Mr. Ely’s methods, the Court does not find Mr. Ely’s methods detestable. Expert testimony from Dr. Mayer established that it is commonplace to combine different types of data in social science research. The Court will not scoff at that practice here. 51 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 52 of 103 PageID #: 3171 This decision in hand, it remains to be seen how the Court’s prior decision regarding the Lauderdale & Kestenbaum List of Asian Surnames affects the Court’s analysis. Mr. Ely’s combination of surname data with ACS CVAP data is acceptable save one major flaw: the combined data set contains the data derived from the Lauderdale & Kestenbaum List of Asian Surnames. As the Court held, the Lauderdale & Kestenbaum List of Asian Surnames is an unreliable proxy for ascertaining which voters are Asian American. The data provided to Dr. Mayer cannot be relied upon to the degree that it is plagued with reliance on this flawed work. The whole of Mr. Ely’s work is admirable, but the sum is only reached by relying on a broken figure. Without the Asian surname data, Mr. Ely cannot produce a complete data set encompassing all minority groups within Kumar’s coalition. The Court, however, will not stop here and render the suit null upon a technicality. Rather, the Court demonstrates below that even assuming arguendo that the data set Mr. Ely provided Dr. Mayer was reliable, Kumar still failed to carry his burden. Before the Court reaches its Gingles analysis, however, it must still consider whether Mr. Ely used a reliable method to render Kumar’s Illustrative District. D. “Best Practices” in Constructing an Illustrative District Finally, Defendants claim that Mr. Ely failed to use the Census Bureau’s best practice method—iterative proportional fitting (“IPF”)—to resolve “pervasive logical inconsistencies” in Mr. Ely’s Census Block data (Defendants’ Exhibit 2). Kumar counters that Mr. Ely used said raking method in drawing Kumar’s Illustrative District. The Court finds the methods Mr. Ely employed to create Kumar’s Illustrative Districts reliable. Iterative proportional fitting is the Census Bureau’s “best practice” method for disaggregating and reaggregating census block data to form a district using individual block data. (Defendants’ Exhibit 2). One of a number of methods of raking, IPF is a means to take one data 52 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 53 of 103 PageID #: 3172 set at a point in time or level of geography and evaluate that same data set at a different point in time or geography while maintaining the integrity of the data set as it is re-aggregated to the next level. (Defendants’ Exhibit 2). In simpler terms, raking is a means of ensuring that a data set that is being altered to some degree will exactly match the initial input once it has been re-calibrated. By using the IPF raking method, an expert “mitigates internal inconsistencies and avoids reliance on several technical assumptions that are questionable” (Defendants’ Exhibit 2) (citations omitted). Dr. Morrison opines that this method also permits one to “match[] the total CVAP exactly, fit[] the distributions of minority CVAPs as closely as possible to the 2010 decennial census actual full-count distributions, and further ensure[] that the sum of minorities always adds up exactly to the respective total” (Defendants’ Exhibit 2). The parties dispute whether Mr. Ely employed this raking method while pointing to inconsistencies within his data. A review of how Mr. Ely constructed Kumar’s Illustrative District is helpful. Mr. Ely used 2010 Census data and the 2013–2017 ACS data to construct Kumar’s Illustrative District. The 2010 Census data was used to analyze the voting age population (“VAP”) by race and Hispanic Origin. The 2013–2017 ACS data was used to analyze the CVAP. Because the 2010 Census only provides CVAP data at the Block Group level, Mr. Ely estimated the African American and Hispanic share of CVAP at the Census Block level. Mr. Ely then calculated each Census Block’s share of VAP based on each ethnic group within the corresponding Block Group. This data was then reaggregated at the Census Block-level to establish the “appropriate share of each ethnic group’s CVAP within the corresponding [Block Group]” (Plaintiff’s Exhibit 1). Mr. Ely’s Expert Report makes no mention of raking during reaggregation. Table 1, supra, is based on this reaggregation of Census Block-level estimates. 53 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 54 of 103 PageID #: 3173 In reviewing Mr. Ely’s Expert Report, Dr. Morrison uncovered what he believed were “major inconsistencies” (Defendants’ Exhibit 2). These inconsistencies, Dr. Morrison went on, included “logical mismatches between the number of persons 18 and older (in 2010) and his estimates of the number of citizens 18 and older (in years thereafter)” (Defendants’ Exhibit 2). Dr. Morrison’s findings are produced below in Table 2. Table 2: Dr. Morrison’s Examples of Logical Inconsistencies in Mr. Ely’s Data Set 54 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 55 of 103 PageID #: 3174 Based on his findings, Dr. Morrison reasons that Mr. Ely failed to employ the IPF raking method when deriving his estimates of the numerical values at the Census Block-level. The reliability of Mr. Ely’s data set, Dr. Morrison concludes, is fatally tarnished. The Court disagrees. Despite not mentioning raking in his expert report, Mr. Ely stated at trial that when he allocated the Block Group data to Census Blocks, he used the raking process Dr. Morrison discussed to ensure that his data matched the origin input data (TR2: 46). In using this process, Mr. Ely relied on the “proportional relationship that exists between the [B]lock [G]roup voting age population and the [B]lock [G]roup citizen voting age data, and also the proportional relationship between the [B]lock [G]roup population data and the [B]lock population data” (TR2: 47). By using these proportional relationships, Mr. Ely maintains that he was able to employ a raking process which maintained said relationships from “the [B]lock [G]roup level to the data at the [Census] [B]lock level” (TR2: 47). Mr. Ely did distinguish the process he used, however, from the IPF process. According to Mr. Ely, the iterative part of IPF “requires additional data beyond what we have available here” (TR2: 82). Nonetheless, Mr. Ely testified that “the raking portion, which is what I have done, is exactly what this methodological paper requested and that [is] how I have referred to it as raking” (TR2: 82). So, the only difference between the IPF approach and Mr. Ely’s approach is that Mr. Ely employed a pure single-stage form of raking because he lacked the data necessary for IPF raking. In the end, Dr. Morrison conceded that Mr. Ely did in fact “perform[] a raking process with the data” (TR3: 144). Dr. Morrison went further and effectively stated that he no longer opposes the method of raking Mr. Ely employed. All that remains then is to determine whether the logical inconsistencies in Mr. Ely’s data should give the Court pause. The Court finds no need to exclude Mr. Ely’s work or attribute lesser weight to it. Because Mr. Ely raked his data set, Mr. Ely believes that another explanation for any logical inconsistencies 55 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 56 of 103 PageID #: 3175 exists. Any inconsistencies, Mr. Ely testified, are due to the inconsistencies that already existed in the two sources of data that came from the Census Bureau. The Census Bureau’s data has inconsistencies built in because “the 2010 census data is reflecting an enumeration that was done in 2010, whereas [the ACS] data is based on population estimates and surveys done between 2013 and 2017” (TR2: 82–83). Put simply, inconsistencies are bound to exist when the Block Groups are reflected in different time periods. The Court is not troubled by those inconsistencies and finds no reason to doubt the reliability of Mr. Ely’s methods. The Court determined that: (1) Kumar has constitutional standing to proceed with his claim; (2) there is no third-party standing implicated in this matter; (3) Plaintiff’s Exhibit 77 is admitted, Defendants’ Exhibit 3 is stricken, and Defendants’ Exhibits 4, and 8–13 are admitted; and (4) Mr. Ely’s methods are all acceptable save his reliance on the Lauderdale & Kestenbaum List of Asian Surnames which is unreliable and his probabilistic estimate of African American voter turnout which goes to the weight of the evidence, not its admissibility. With those findings, the Court now proceeds to its findings of fact and conclusions of law under Rule 52(a)(1). IV. Findings of Fact Frisco Independent School District is a fast-growing Texas Independent School District comprised of 72 campuses, 8,000 employees, and over 63,000 students. The District provides a broad, diverse curriculum that affords students a variety of educational opportunities including everything from the most basic curriculum, to career and technical certifications, Advanced Placement (“AP”) courses, and dual credit opportunities. Encompassing both the Collin and Denton Counties, FISD spreads across four major North Texas cities: Frisco, Plano, Little Elm, and McKinney. Because of the number of students FISD serves, FISD is comprised of a diverse 56 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 57 of 103 PageID #: 3176 population. As of 2020, FISD enrolled approximately 63,015 students: 58.8% minority, 41.2% White. As required by the Texas Education Code, FISD operates under a board of trustee’s form of government. FISD’s Board is comprised of seven Trustees. Trustees serve three-year terms with no term limits. To serve on the Board, one must win a seat through a non-partisan election. Trustees are elected via an at-large system to a specific “Place.” The Place that a candidate runs for does not have any geographic significance; rather, each Place serves as a measure to stagger elections. By staggering elections, FISD prevents a complete turnover of the Board in a single election. Accordingly, Places 1, 2, and 3 are elected in year one; Places 4 and 5 are elected in year two; and Places 6 and 7 are elected in year three. To win a Place, a candidate only needs to accumulate a plurality of the vote. Voters may vote in the election for each Place, but they may only cast one vote per Place.19 This brings the Court to the present action. On April 16, 2019, Plaintiff Suresh Kumar filed this action against FISD and FISD’s Board.20 Kumar alleges that FISD’s electoral system dilutes the voting power of the African American, Hispanic, and Asian communities within FISD such that minorities cannot meaningfully participate in the electoral process by electing candidates of their choice. Kumar points out that since the Board’s inception in 1907–08, there has only been one minority Trustee. And, according to Kumar, this is not for a lack of effort. Between 2010 and 2019, nine minority candidates sought election to the Board. All but one minority candidate, Mr. Gopal Ponangi, lost. FISD and the Board oppose Kumar’s characterization of their electoral scheme and counter that it allows for a representative form of governance that serves the FISD population well. After the FISD’s Board is independent of all other City of Frisco governance. The current Trustees are Rene Archambault, John Classe, Debbie Gillespie, Natalie Hebert, Anne McCausland, Gopal Ponangi, and Chad Rudy. 19 20 57 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 58 of 103 PageID #: 3177 Court denied Defendants’ Motion for Summary Judgement (Dkt. #82), the Court proceeded to a trial on the merits.21 At trial, Kumar offered the expert testimony of Mr. Ely and Dr. Kenneth Mayer (“Dr. Mayer”). Mr. Ely is the founder of Compass Demographics, a “consulting and database management firm specializing in projects involving Census and Election data” (Plaintiff’s Exhibit 1). Mr. Ely holds a B.S. in Mechanical Engineering and Social Sciences from the California Institute of Technology. Over the course of his career, Mr. Ely has served as a Special Master, expert, and/or consultant in a multitude of voting rights matters. At trial, Mr. Ely testified to Kumar’s ability to satisfy the first Gingles factor as well as the Senate Factors. Mr. Ely further testified as to the data he provided Dr. Mayer for Dr. Mayer’s expert analysis the second and third Gingles factors. Dr. Mayer is a professor of political science at the University of Wisconsin-Madison (Plaintiff’s Exhibit 6). He holds a degree in political science with a minor in mathematics from the University of California, San Diego, and a Ph.D. in political science from Yale University. Dr. Mayer testified to Kumar’s ability to satisfy Gingles factors 2 and 3 as well as the Senate Factors. The parties stipulate that Mr. Ely is qualified to express opinions related to the first Gingles factor. The parties further stipulate that Dr. Morrison is qualified to express opinions related to the second and third Gingles factors. Both Mr. Ely and Dr. Mayer, the parties agree, are qualified to express opinions related to the Senate Factors. The Court finds that Mr. Ely and Dr. Mayer are qualified, 21 The Court has subject matter jurisdiction over this action pursuant to: (1) 28 U.S.C. § 1331; (2) 28 U.S.C. § 1343(a)(3); (3) 28 U.S.C. § 1343(a)(4); and (4) Section 2 of the Voting Rights Act of 1965, 52 U.S.C. § 10301. There is no dispute that this Court has personal jurisdiction over FISD or over the Board of Trustees. Finally, venue is proper in the Court pursuant to 28 U.S.C. § 1391(b)(1) because all Defendants reside in the Eastern District of Texas. 58 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 59 of 103 PageID #: 3178 due to their education and experience, to offer expert opinions on whether Kumar can satisfy the respective Gingles factors and the Senate Factors. Kumar also offered the lay testimony of Ms. Grace Wang, Dr. Marie Walters, Mr. Muniraj (“Muni”) Janagarajan, Dr. Marc Payne, Mr. Carlos L. Gallardo, and himself. The following credentials and testimony were offered by each lay witness: Ms. Grace Wang (“Ms. Wang”) Ms. Wang is an Asian resident of Plano who ran for the Board in 2017 against a White incumbent and White challenger. Ms. Wang testified that she entered the race following the 2016 Presidential Election of Donald J. Trump which made her feel that she needed to “get involved.” She received approximately 15% of the popular vote. Ms. Wang further testified that she has applied for multiple FISD committee positions. She was denied each opportunity. Ms. Wang concluded that she believes Asians in FISD lack representation and that she supports the creation of a new district. Dr. Marie Walters (“Dr. Walters”) Dr. Walters is a Lieutenant Colonel who formerly served in the United States Air Force where she used her Ph.D. in systems engineering to work in electronic warfare. She is an African American who now resides in Frisco. Dr. Walters testified that she helped create the organization “OneVoice.” OneVoice was created, in partnership with FISD, to address the disparity among racial groups in testing success. Dr. Walters further testified that she applied for a vacancy on the Board, but her application was rejected. She concluded her testimony by stating that there are issues in FISD which must be resolved. 59 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 60 of 103 PageID #: 3179 Mr. Muniraj (“Muni”) Janagarajan (“Mr. Janagarajan”) Mr. Janagarajan is an information technology professional who immigrated from India to the United States in 1995. He eventually moved his family to Frisco. Because of his dissatisfaction with FISD on a variety of issues—mainly: (1) taxes; (2) fiscal responsibility; (3) campus safety; and (4) teacher pay—Mr. Janagarajan testified that he ran for election in May of 2019 against a White incumbent. Mr. Janagarajan lost but re-filed for election in 2020. He claims that his family is not receiving the equal opportunity and representation in FISD they deserve. Dr. Marc Payne (“Dr. Payne”) Dr. Payne is the Chairman of OneVoice, founder of FISD’s Diversity Task Force Committee, and the Vice President of the National Association for the Advancement of Colored People (“NAACP”) Chapter in Frisco. Dr. Payne is an African American who now lives outside FISD borders. Nonetheless, Dr. Payne testified that he is involved in FISD politics and formerly supported Dr. Walters’ potential candidacy for the Board. Dr. Payne concluded his testimony by stating that he is concerned that minorities are not receiving equal treatment in FISD. Mr. Carlos L. Gallardo (“Mr. Gallardo”) Mr. Gallardo is a former Officer in the United States Air Force and employee of the Central Intelligence Agency. He is Hispanic and lives in FISD. Mr. Gallardo testified that in 2014 he applied to fill a vacancy on the Board. His application was denied. In 2015, Mr. Gallardo ran for election for Place 7 against a White incumbent and Indian challenger. Mr. Gallardo was unsuccessful. Mr. Suresh Kumar (“Kumar”) Kumar, the named Plaintiff, immigrated to the United States from India and moved to Frisco, Texas. Kumar has two children, both of whom attended, or are attending, FISD schools. 60 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 61 of 103 PageID #: 3180 Kumar testified that he believes all people of color are entitled to equal representation; hence, he decided to file this lawsuit. The Court finds that Ms. Wang, Dr. Walters, Mr. Janagarajan, Dr. Payne, Mr. Gallardo, and Kumar are all fact witnesses regarding elections in FISD. This brings the Court to Defendants’ witnesses. Defendants offered the expert testimony of Dr. Peter Morrison and Dr. John Alford (“Dr. Alford”). Dr. Morrison is an applied demographer who is retired from The RAND Corporation (Defendants’ Exhibit 2). Dr. Morrison holds a B.A. in sociology from Dartmouth College and a Ph.D. in sociology from Brown University. Prior to retiring, Dr. Morrison served as a professor at the University of Pennsylvania and also lectured at UCLA, the RAND Graduate School, and the Helsinki School of Economics. Currently, Dr. Morrison serves as the President of Peter A. Morrison & Associates, Inc. He has testified as an expert in approximately twenty voting rights cases since 2012. At trial, Dr. Morrison provided a critique of Mr. Ely’s Expert Report and illustrative district. In so doing, Dr. Morrison focused on the first Gingles factor. Defendants also offered the expert testimony of Dr. Alford. Dr. Alford is a tenured professor at Rice University (Defendants’ Exhibit 1). Dr. Alford holds a B.S. in political science from the University of Houston, an M.P.A. from the University of Houston, an M.A. in political science from the University of Iowa, and a Ph.D. in political science from the University of Iowa. Dr. Alford previously provided expert reports and/or testified as an expert witness in a variety of voting rights matters at the state and federal level. At trial, Dr. Alford criticized the methodology Mr. Ely used which underpinned Dr. Mayer’s report. He then critiqued Dr. Mayer’s opinions and conclusions relating to the second and third Gingles factors. The parties stipulate that Dr. Morrison is qualified to express opinions related to the first Gingles factor. The parties further stipulate that 61 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 62 of 103 PageID #: 3181 Dr. Alford is qualified to express opinions related to the second and third Gingles factors. Like Plaintiff’s Experts, the parties agree that both Dr. Morrison and Dr. Alford are qualified to express opinions related to the Senate Factors. The Court finds that Dr. Morrison and Dr. Alford are qualified, due to their education and experience, to offer expert opinions on whether Kumar can satisfy the respective Gingles factors and the Senate Factors. Defendants also offered the lay testimony of Ms. Lori Wassam, Dr. Michael Waldrip, Mr. John Classe, Mr. Chad Rudy, Ms. Charis Hunt, Ms. Anjali Shirvaikar, Ms. Geneva Polster, Mr. Peter Burns, and Ms. Mufyn Lynn. The following credentials and testimony were offered by each lay witness: Ms. Lori Wassam (“Ms. Wassam”) Ms. Wassam is the internal demographer for FISD who created the Wassam Exhibits. Ms. Wassam testified as to how/why she created the Wassam Exhibits and what the Wassam Exhibits demonstrate. Dr. Michael Waldrip (“Dr. Waldrip”) Dr. Waldrip is the Superintendent of FISD. Dr. Waldrip testified to the performance of FISD, the demographics of the District, and how he has sought to address any disparities among races in the District. In concluding, Dr. Waldrip stated that he believes the current at-large system is a successful electoral scheme. He believes that the Board should take a great deal of pride in the success of FISD. Mr. John Classe (“Mr. Classe”) Mr. Classe is a Board Trustee who was first appointed to the Board in 2014. He was then elected in 2015 and re-elected in 2018. Mr. Classe testified to the general parameters of the 62 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 63 of 103 PageID #: 3182 electoral system, the growth in the District, and his belief that the Board meets the needs of minorities in FISD. Mr. Chad Rudy (“Mr. Rudy”) Mr. Rudy is the current President of the Board. He was first elected in 2015. Mr. Rudy testified as to the increasing politicization of what are supposed to be non-partisan Board elections. More specifically, Mr. Rudy pointed toward Empower Texans and Frisco United—two conservative organizations—as well as the Frisco Tea Party for their consistent opposition of FISD-initiatives. Despite the politicization caused by certain groups, Mr. Rudy testified that he believes the Board is responsive to all minority needs. Ms. Charis Hunt (“Ms. Hunt”) Ms. Hunt is the Managing Director of Human Resources for FISD. Ms. Hunt testified to FISD’s recruitment, retention, and hiring practices. More specifically, Ms. Hunt discussed various FISD’ policies and initiatives directed toward prohibiting discrimination and increasing diversity. Ms. Anjali Shirvaikar (“Ms. Shirvaikar”) Ms. Shirvaikar is an Asian American software programmer who resides in a portion of Plano encompassed within FISD. Ms. Shirvaikar testified that she ran for the Board in 2016 in a 3-person race where she lost by some-300 votes. She believes that she lost the election due to, among other things, an onslaught of negative campaign ads run against her by the Frisco Tea Party. Further, Ms. Shirvaikar testified that she received no aid from LULAC or the NAACP in her campaign. While Ms. Shirvaikar qualified that she could only speak for herself, Ms. Shirvaikar stated that she believes the at-large election system and Board support her and her child’s needs. 63 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 64 of 103 PageID #: 3183 Ms. Geneva Polster (“Ms. Polster”) Ms. Polster is a Hispanic retiree from the City of Frisco. Both of her sons are FISD grads. Ms. Polster testified that she has never experienced racism nor discrimination within FISD. She further testified that she believes the at-large system works properly and permits voters to elect the candidate of their choice. Mr. Peter Burns (“Mr. Burns”) Mr. Burns is one of the founders of OneVoice. He is an African American who resides within FISD’s borders. Mr. Burns testified that he has experienced no racism or discrimination in FISD. He further stated that he believes the at-large election system is “great.” Ms. Mufyn Lynn (“Ms. Lynn”) Ms. Lynn is a self-employed political consultant who has worked on numerous campaigns for the Board. She is an African American who lives in Little Elm. Ms. Lynn testified that she has experienced no prejudice or discrimination in FISD. She also stated that she believes the atlarge election system works well. The Court finds that Ms. Wassam, Dr. Waldrip, Mr. Classe, Mr. Rudy, Ms. Hunt, Ms. Shirvaikar, Ms. Polster, Mr. Burns, and Ms. Lynn are all fact witnesses regarding elections in FISD. With the lay and expert witnesses determined, the Court now turns to the merits of Kumar’s Voting Rights Act claim. A. Gingles I Under the first Gingles factor, Kumar must offer evidence that the minority group challenging FISD’s at-large election system—here, a purported African American, Hispanic, and Asian coalition—is “sufficiently large and geographically compact to constitute a majority in a single member district.” Gingles, 478 U.S. at 50–51. To satisfy this first factor, plaintiffs generally 64 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 65 of 103 PageID #: 3184 submit evidence of a “hypothetical redistricting scheme[] in the form of an illustrative plan[].” Gonzalez v. Harris Cnty., 601 F. App’x 255, 258 (5th Cir. 2015); see also Gingles, 478 U.S. at 88 (O’Connor, J., concurring) (“The phrase vote dilution itself suggests a norm with respect to which the fact of dilution may be ascertained . . . . In order to decide whether an electoral system has made it harder for minority voters to elect the candidates they prefer, a court must have an idea in mind of how hard it should be for minority voters to elect their preferred candidates under an acceptable system.”) (internal quotation marks omitted). In submitting an illustrative district, a plaintiff must prove that the minority citizen voting age population (“MCVAP”) of the illustrative district exceeds 50% of the total CVAP within the district. See Valdespino v. Alamo Heights Indep. Sch. Dist., 168 F.3d 848, 852–53 (5th Cir. 1999); Westwego Citizens for Better Gov’t v. City of Westwego, 946 F.2d 1109, 1117 n.7 (5th Cir. 1991). This majority-minority requirement ensures that absent an at-large electoral scheme, “minority voters possess the potential to elect representatives . . . .” Clements, 986 F.2d at 743 (emphasis in original) (citing Gingles, 478 U.S. at 50 n.17). Because of its threshold purpose, the 50% minority requirement “is a bright line test.” Benavidez I, 690 F. Supp. 2d at 457. Defendants claim that Kumar failed to offer an illustrative district that was sufficiently large and geographically compact. Defendants aver that Kumar’s Illustrative District does not meet the 50 percent threshold, is bizarrely shaped, and contains no concentrations of minority groups. Additionally, Defendants that claim Mr. Ely—Kumar’s expert for the first Gingles factor—failed to consider traditional districting principles when drawing Kumar’s Illustrative District. For the reasons that follow, the Court disagrees with Defendants and finds that Kumar satisfied the first Gingles factor. 65 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 66 of 103 PageID #: 3185 i. Sufficiently Large Defendants assert that Kumar’s Illustrative District fails to meet the bright line 50% threshold because the CVAP in the District is just 0.03% above 50%. Given that 50.03% is purportedly within the margin of error, Defendants conclude that Kumar’s Illustrative District fails at the outset. Kumar counters that the minority population of the District is 55.2% and that the MCVAP is actually 51.2%. Kumar further avers that the Court should disregard Defendants new strategy of labeling the CVAP as only 50.03%. According to Kumar, this distortion of the record is merely an eve of trial decision made to undermine Kumar’s figures. This late, game time decision, Kumar continues, is demonstrated by Dr. Morrison’s failure to mention any margin of error argument in his initial expert report. It was only at trial, Kumar goes on, that Dr. Morrison asserted a new opinion: “[A]nd I would also just as an addendum say that 50.03 is the estimate, the point estimate, what we call the estimated percentage, with a margin of error of plus or minus something. So it’s right on the edge of 50.0000” (TR3: 125) (emphasis added). Irrespective of the actual MCVAP, Kumar concludes that even a slight numerical majority satisfies the first Gingles factor. The Court agrees with Kumar and finds that Kumar satisfied the 50% minority requirement. The majority-minority rule “relies on an objective, numerical test: Do minorities make up more than 50 percent of the voting-age population in the relevant geographic area?” Bartlett v. Strickland, 556 U.S. 1, 18 (2009). While the Gingles factors generally should not be applied mechanically, “[i]t remains . . . that a party asserting § 2 liability must show by a preponderance of the evidence that the minority population in the potential election district is greater than 50 percent.” Id. at 19–20. The majority-minority rule, and its “clear-edged” application, has “its foundation in principles of democratic governance.” Id. at 19. “The special significance, in the 66 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 67 of 103 PageID #: 3186 democratic process, of a majority means it is a special wrong when a minority group has 50 percent or more of the voting population and could constitute a compact voting majority but, despite racially polarized bloc voting, that group is not put into a district.” Id. Applying this clear-edged rule, the Court finds that Kumar met his burden. First, the Court does not find this to be a situation like that of Brewer v. Ham, 876 F.2d 448, 451–52 (5th Cir. 1989) or Reyes, 2008 WL 4791498, at *10. In both of those cases, proposed districts were discredited despite initially showing plausible slim majority-minority districts. In Brewer, for instance, the Fifth Circuit affirmed the district court which found that the evidence supporting a minority total VAP of 55.91% was, at best, inconclusive in demonstrating that a majority-minority district existed. Brewer, 876 F.2d at 451–52. Likewise, in Reyes, Judge O’Connor found that a proposed district with a potential Hispanic CVAP of 52.5% failed to establish, by a preponderance of the evidence, that the plaintiff met the first Gingles factor. Reyes, 2008 WL 4791498, at *10. Judge O’Connor held that the plaintiff failed to establish a majority-minority district after finding that the evidence supporting a 52.5% Hispanic CVAP was unreliable. Id. Here, Kumar employed CVAP data, not total VAP data like the Appellants in Brewer. Further, unlike in Reyes, the evidence supporting Kumar’s CVAP of either 51.2% or 50.03% is not credibly disputed. Reliability is simply not at issue. This leads the Court to its second finding. The Court finds that any argument that Kumar’s Illustrative District is within the margin of error is weak. At trial, Dr. Morrison stated that it is likelier than not that a majority-minority district exists. The exchange between Defense Counsel and Dr. Morrison on this subject is telling: Q. Within the margin of error, could it be below 50.000? A. Very possibly. The only thing I can say is that in terms of standards of confidence, one can say that it is likelier than not that the majority exists. That is to say, it’s likelier than not that the true number, if we knew it, would be over 50 percent. But it might be like saying I flipped a coin three times and it came up heads twice, so if you tell me the coin is biased, I would have to guess it’s biased in favor of heads, not tails. 67 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 68 of 103 PageID #: 3187 Q. If you were to remove just one of those minority groups from his illustrative district, could you or Mr. Ely create a majority-minority district with just two of those groups? A. The honest answer is I don’t know . . . . (TR3: 125). Defendants ask that the Court enforce an absolute clear and convincing evidence standard for meeting the 50% threshold requirement. Unlike a Sith, however, the Court does not deal in absolutes. While the 50% threshold requirement is a bright line rule, the Court will not discredit Kumar’s Illustrative District based upon an expert whose testimony consisted of statements like “very possibly” and “I don’t know.” Absent reasoned arguments that cast genuine doubt upon this threshold requirement, the Court finds that Kumar met his burden. In reaching this determination, the Court need not decide whether the disputed CVAP is 51.2% or 50.03%. The 50% threshold requirement is just that—a 50% requirement. While a plaintiff’s ability to meet that requirement can be reasonably disputed in certain contexts—such as in Brewer and Reyes—that context is not present here. The Court accordingly turns to the next issue before it: geographical compactness. ii. Geographically Compact The parties next dispute whether Kumar’s Illustrative District is geographically compact. Having considered each parties arguments, the Court finds that the District is geographically compact. “[T]he Gingles compactness requirement ‘refers to the compactness of the minority population, not to the compactness of the contested district.’” Benavidez I, 638 F. Supp. 2d at 721 (citing LULAC, 548 U.S. at 433). “[S]hape is a significant factor that courts can and must consider in a Gingles compactness inquiry.” Sensley v. Albritton, 385 F.3d 591, 596 (5th Cir. 2004). When considering the shape of a proposed district, however, a court does not measure the aesthetics before it, but instead analyzes the “functional realities of the minority population.” Benavidez III, 68 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 69 of 103 PageID #: 3188 2014 WL 4055366, at *9; see also Clark v. Calhoun Cnty., Miss., 21 F.3d 92, 95 (5th Cir. 1994) (“The first Gingles precondition does not require some aesthetic ideal of compactness, but simply that the [minority] population be sufficiently compact to constitute a majority in a single-member district.”); Rodriguez, 964 F. Supp. 2d at 738 (“Geographical symmetry or attractiveness is a desirable consideration for districting, but only to the extent it facilitates the political process.”); cf. Sensely, 385 F.3d at 596 (“While . . . a compactness determination should not hinge on the shape of a district, the shape of a district certainly cannot be disregarded in a compactness inquiry. . . . [I]t is clear that shape is a significant factor that courts can and must consider in a Gingles compactness inquiry.”). “While no precise rule has emerged governing § 2 compactness, the inquiry should take into account traditional districting principles such as maintaining communities of interest and traditional boundaries.” LULAC, 548 U.S. at 433 (citations and internal quotation marks omitted). If a court fails to consider traditional districting principles, it runs the risk of assuming “from a group of voters’ races that they ‘think alike, share the same political interests, and will prefer the same candidates at the polls.’” Id. (citing Miller, 515 U.S. at 920; Shaw, 509 U.S. at 647). Failure to be mindful of such prohibited assumptions lends a real risk to diseasing what is otherwise meant to be a process that seeks to prevent discrimination. To be sure, “[t]he practical consequence of drawing a district to cover [] distant, disparate communities is that one or both groups will be unable to achieve their political goals.” Id. at 434. “Compactness is, therefore, about more than ‘style points’; it is critical to advancing the ultimate purposes of § 2, ensuring minority groups equal ‘opportunity . . . to participate in the political process and to elect representatives of their choice.’” Id. (citations omitted). While there may be no precise rule governing compactness, this much is true, in assessing compactness a court should consider: (1) the dispersion of the relevant 69 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 70 of 103 PageID #: 3189 minority community; (2) the shape of the illustrative district and the causes underlying that shape; and (3) the illustrative district’s compliance with traditional districting principles including maintaining communities of interest and traditional boundaries. See Rodriguez, 964 F. Supp. 2d at 737–54. Defendants assert that Kumar’s Illustrative District is not compact. Specifically, Defendants take issue with: (1) the shape of Kumar’s Illustrative District; (2) the dispersion of minorities within Kumar’s Illustrative District; and (3) Kumar’s compliance with traditional districting principles. In other words, Defendants take issue with each relevant metric of compactness. Kumar counters that his District satisfies the first Gingles factor’s compactness requirement. The Court considers each issue presented in turn. In considering these issues, the Court looks to Kumar’s Illustrative District, which is set out below in Appendix 1. 70 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 71 of 103 PageID #: 3190 Appendix 1: Illustrative District 1. Shape Defendants first assert that Kumar’s Illustrative District is so bizzarely shaped that it is necessarily non-compact. Citing a plethora of redistricting cases, rather than dilution cases, Defendants ask that the Court find that Mr. Ely disregarded traditional districting principles. The Court declines Defendants’ invitation. As previously stated, “[t]he first Gingles precondition does not require some aesthetic ideal of compactness, but simply that the [minority] population be sufficiently compact to constitute a majority in a single-member district.” Clark, 21 F.3d at 95. Indeed, “[i]n evaluating the compactness of the minority population, considerations of the dispersion of the territory of the district and the regularity or length of the perimeters of the district become subsidiary to 71 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 72 of 103 PageID #: 3191 considerations of the minority group’s compactness.” Benavidez I, 638 F. Supp. 2d at 721 (citing LULAC, 548 U.S. at 433). Here, Defendants make half-hearted attempts to disparage Kumar’s Illustrative District based on shape. Defendants provide the Court with no analysis of any compactness metrics and zero discussion of geometric tests. Rather, Defendants cite authorities that say courts should stand vigilant against bizarrely shaped districts—Defendants then summarily conclude that Kumar’s Illustrative District is bizarrely shaped. Defendants’ barebones argument holds no water. As Rodriguez discussed, “[t]here are many methods by which to assess the shape of the district. One recognized method is simply to examine the physical boundaries of the maps and the proposed districts, and based on that visual examination, determine if the district is strangely shaped.” 964 F. Supp. 2d at 738–39 (citing Sensley, 385 F.3d at 596). “This geometric test, though imprecise, offers some threshold assessment of compactness.” Id. Here, the Court need not go beyond the “Geometric Test.” Kumar’s Illustrative District is not plagued by tentacles and appendages like many districts struck down for their bizarre shapes in the redistricting context. See, e.g., Bush v. Vera, 517 U.S. 952, 965 (1996) (discussing a redistricting plan and assessing whether the district in issue had “bizarrely shaped tentacles”); Miller v. Johnson, 515 U.S. 900, 910 (1995) (discussing a redistricting plan and assessing whether the district in issue had “appendages”).22 And any gerrymandered features or jagged qualities that the District might have are grounded in objective, measurable foundations. Kumar’s Illustrative District is shaped the way it is because it is comprised of undivided Census Blocks. By combining the minority coalition concentrations within the District and not dividing the Census Blocks those concentrations are found in, the shape seen in Appendix 1, supra, comes into existence. This shape is further 22 The Court recognizes that these cases concern different legal issues. Nonetheless, the Court cites these cases to demonstrate shape considerations that the Supreme Court has previously considered, even in a different context. 72 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 73 of 103 PageID #: 3192 grounded in Mr. Ely’s use of recognizable boundaries, such as streets, to develop Kumar’s Illustrative District. Because: (1) the shape of Kumar’s Illustrative District is the result of physical boundaries, such as streets, and undisturbed Census Blocks; and (2) Defendants provide little argument surrounding shape, the Court finds the shape of Kumar’s Illustrative District is of no concern. More important than shape, is the dispersion of the minority groups within said shape. The Court turns to that next. 2. Dispersion of the Relevant Minority Community Defendants next challenge whether the relevant minority communities are properly dispersed in Kumar’s Illustrative District. More specifically, Defendants claim that aside from the shape of the District, the relevant minority groups in this action are dispersed throughout FISD without any particular concentrations. Kumar counters that there are three concentrations of the relevant minority communities which are accounted for in his District. The Court agrees with Kumar. As previously stated, “the Gingles compactness requirement ‘refers to the compactness of the minority population, not to the compactness of the contested district.’” Benavidez I, 638 F. Supp. 2d at 721 (citing LULAC, 548 U.S. at 433). The Court already determined that the shape of Kumar’s Illustrative District is a non-issue. Whether the minority communities within the District are compact is a different question. Kumar’s Illustrative District was created using entire Census Blocks. Contrary to Defendants’ assertion that minorities are dispersed throughout FISD, the Court finds that Kumar’s Illustrative District comprises three concentrations of minorities. Those concentrations can be viewed below in Appendix 2. 73 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 74 of 103 PageID #: 3193 Appendix 2: Plaintiff’s Exhibit 94 – Student Ethnicity Map As Appendix 2 illustrates in its legend, Hispanics are represented by green circles, Asians are represented by red triangles, and African Americans/ Black citizens are represented by pink circles. The first minority concentration is located in the North-Eastern portion adjacent to Independence Parkway and Panther Creek Parkway. This location contains a dense Asian population along with essentially average African American/ Black and Hispanic populations. The second minority concentration is in the downtown Frisco area North of the curve in the Dallas Tollway—which is identified as the Dallas Parkway in Appendix 1. This location contains a dense Hispanic population, average African American/ Black population, and below average Asian population. The final minority concentration is located South of 121 in Plano. This location 74 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 75 of 103 PageID #: 3194 contains a dense Asian population and essentially average African American/ Black and Hispanic populations like the first concentrated location. As previously discussed, the District was created using 2010 Census data; therefore, the District may not reflect new communities that may have since arisen within FISD. With that caveat in mind, however, it is apparent, even from a mere eyeball test, that there are at least three concentrations of minority communities within Kumar’s Illustrative District. Kumar’s Illustrative District is not simply comprised of three minority concentrations; rather, Kumar’s Illustrative District is comprised of three compact minority concentrations. At trial, Mr. Ely testified that the District stretches from the extreme north of FISD to the extreme South. Although this testimony makes it sound like the District encompasses a large swath of land at first blush, nothing is further from the truth. Mr. Ely testified that the greatest distance between any two residences within Kumar’s Illustrative District is “less than ten miles” (TR 2: 111). Surely a minority coalition that is separated by a mere ten miles, at the extreme, initially satisfies the compactness requirement of the first Gingles factor. See Benavidez I, 638 F. Supp. 2d at 721 (citing LULAC, 548 U.S. at 433). The Court therefore finds Defendants’ dispersion argument unavailing.23 While the Court finds that the shape of the District is a non-issue and the minority communities within said District are initially compact enough, the Court must turn to the final, and most important inquiry: Whether the minority communities are sufficiently compact while accounting for traditional districting principles. 23 Defendants seem to make a passing argument that because some of the Asian community South of 121 is not encompassed within Kumar’s Illustrative District, there must be a flaw in the District. Not so. See Campos v. City of Baytown, Tex., 840 F.2d 1240, 1244 (5th Cir. 1998) (“The fact that there are members of the minority groups outside the minority district is immaterial.”). Further, it should be noted that Appendix 2 depicts students, not minority citizens of voting age. While Appendix 2 is still useful, any argument Defendant places on it is, to a degree, undermined as it does not give a full picture. With that said, even if Appendix 2 omits some minority citizens, it still evidences concentrations. 75 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 76 of 103 PageID #: 3195 3. Traditional Districting Principles Defendants final objection to Kumar’s Illustrative District is their strongest. Defendants assert that in creating Kumar’s Illustrative District, Mr. Ely neglected communities of interest and traditional boundaries. By only focusing on race, the argument goes, it became easier for Mr. Ely to create a “compact” district at the cost of everything else. For what it is worth, Mr. Ely concedes that he did not engage in a “communities of interest analysis.” At the same time, however, Mr. Ely counters that he did perform a “functional compact analysis” which he claims is a basic communities of interest analysis rather than a complete one. Anything beyond this basic analysis, Mr. Ely concludes, is best reserved for the remedial phase of this litigation, not the liability phase. To consider whether Kumar’s Illustrative District neglects communities of interest in a fatal fashion, the Court must first determine what constitutes a community of interest. The “communities of interest” term of art is employed in a myriad of legal contexts. See Stephen J. Malone, Note, Recognizing Communities of Interest in a Legislative Apportionment Plan, 83 VA. L. REV. 461, 464–65 (1997) (“For instance, it has been used to describe the shared interests of a plaintiffs’ class in a class action lawsuit, a bargaining unit in labor law, a factor considered when a city wishes to annex unincorporated territory, and a requirement of a group challenging exclusion from jury selection.”). Clearly, the term is also employed in Voting Rights Act matters. See id.; see also LULAC, 548 U.S. at 433 (“While no precise rule has emerged governing § 2 compactness, the inquiry should take into account traditional districting principles such as maintaining communities of interest and traditional boundaries.”). Despite the terms widespread use, the term has remained difficult to define. That being said, the Court is not without guidance. In Carstens v. Lamm, a district court in Colorado put forth the following definition: “[C]ommunities of interest represent distinctive units which share common concerns with respect 76 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 77 of 103 PageID #: 3196 to one or more identifiable features such as geography, demography, ethnicity, culture, socioeconomic status or trade.” 543 F. Supp. 68, 91 (D. Colo. 1982) (discussing communities of interest in the redistricting context). Following Carstens, the Virginia Law Review published a Note which collected definitions promulgated by a variety of states including: Alaska, California, Colorado, Hawaii, Maine, Oklahoma, Oregon, Washington, West Virginia, and Wisconsin. See Malone, supra, at 466–67. Those state definitions focused on socio-economic interests. More recently, in Terrebonne Branch NAACP v. Jindal, Mr. Ely defined communities of interest as “reflecting some characteristic that has an impact on the representational interests or priorities of voters, or on the way voters interact with each other to elect or communicate with a representative.” 2019 WL 4398509, at *5 (M.D. La. Apr. 29, 2019), report and recommendation adopted sub nom. Terrebonne Par. Branch NAACP v. Edwards, 399 F. Supp. 3d 608 (M.D. La. 2019). Finally, Mr. Ely testified here as to the definition he most recently attributes to communities of interest: Basically any sort of interest, representational interest that affects what people’s policy preferences are, how people connect with other people, how people interact, any of the things that might affect how they vote in an election, whether they vote in an election, who else they interact with in the process of deciding about those kinds of things. Anything that has a representational interest can be a community of interest, so it can be a neighborhood. It can be people who are members of a certain organization. It can be people who attend the same schools. It can be people who have the same political opinions. It can be people who are of the same socioeconomic status. There’s not -- it’s not infinite, but a very great range of possibilities of determining a community of interest, and that’s one thing that is usually part of a hearing process or analysis for a jurisdiction redistricting or creating a remedial plan. (TR2: 144). With “communities of interest” more solidly defined, the Court now considers whether Kumar’s Illustrative District preserves the communities of interest within FISD. While the Court acknowledges that Mr. Ely did not consider specific communities of interest, the Court nevertheless finds that Kumar’s Illustrative District sufficiently preserves communities of interest at the liability stage. 77 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 78 of 103 PageID #: 3197 Defendants claim Mr. Ely failed to preserve a number of communities of interest when constructing Kumar’s Illustrative District. Namely, Defendants point to: (1) neighborhoods; (2) subdivisions; (3) school attendance zones; (4) congressional districts; (5) the City of Plano; and (6) Census Blocks. Mr. Ely, as previously stated, admits that he did not analyze communities of interest but counters that he did perform a functional compact analysis. By performing a functional compact analysis, Mr. Ely means that he considered the boundaries and shape of the District and decided that the District would allow a resident to “relatively easily recognize which district he and his neighborhood are in and would allow the people -- the various communities that are within the district to interact with each other and to interact with a representative for the entire district” (TR4: 137–38). Mr. Ely claims that despite his failure to consider specific communities of interest, his functional compact analysis complied with the requirement that he use traditional districting criteria and enabled him to preserve communities of interest not specifically considered. The Court recognizes that communities of interest must be preserved. And although Mr. Ely did not consider specific communities of interest when drawing Kumar’s Illustrative District, and was simply incorrect when he said that his task was not to consider communities of interest, the Court still finds Defendants objections lacking.24 Defendants hurl a horde of objections at Kumar’s Illustrative District that, when considered in their totality, would mount an impossible task for Kumar. The Voting Rights Act, and Gingles, does not require that a plaintiff prove a perfect case in order to stomp out invidious discrimination. Were that the case, no plaintiff would succeed, and the stain of this Nation’s original sin would live on with no amends. While the Court in the end finds that there is no Voting Rights Act One of those objections concerns Mr. Ely’s considerations of race. While Mr. Ely considered race while developing Kumar’s Illustrative District, it cannot be said that Mr. Ely disregarded everything save race. Mr. Ely testified that he considered Census Blocks and street boundaries while constructing the District and also performed a functional compact analysis. Thus, the objection that Mr. Ely only considered race when drawing the District is disregarded. 24 78 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 79 of 103 PageID #: 3198 violation here, the Court will not up the burden on Kumar in the first Gingles factor merely because Kumar cannot establish the second. It is true that Kumar’s Illustrative District splits some neighborhoods, subdivisions, congressional districts, and the like. But it is equally true that “[t]he ultimate end of the first Gingles precondition is to prove that a solution is possible, not necessarily to present the ultimate solution to the problem.” Rodriguez, 964 F. Supp. 2d at 746. If the Court were to require an ultimate solution, it would conflate the liability phase with the remedial phase. And, if the Court were to enforce Defendants’ unreachable community of interests’ standards, no plaintiff would ever succeed in disputing potentially unlawful electoral schemes. The Voting Rights Act does not permit either result. Kumar only needs to show a “well-developed, legallyadequate plan that can be adjusted during [the remedial stage].” Id. (citing Fairley, 584 F.3d at 671 n.14). As discussed below, the Court finds that Kumar proffered a “well-developed, legally adequate plan,” and the Court is not concerned with any perceived flaws in said District. Id. The Court does not need to spill much ink to make its point. Defendants cite a variety of perceived failures in Kumar’s Illustrative District, yet many of those “failures” do not reach the threshold Defendants allege that they meet. Take neighborhoods and subdivisions, for instance. Defendants claim that Kumar’s Illustrative District dismembered numerous divisions and subdivisions in an “egregious fashion.” (TR3: 127–28). Defendants’ Exhibit 11, which is one of the Wassam Exhibits and is now labeled Appendix 3, is illustrative. 79 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 80 of 103 PageID #: 3199 Appendix 3: Defendants’ Exhibit 11 – Digital Map Overlapping Plaintiff’s Illustrative District over Neighborhoods and Subdivisions Appendix 3 demonstrates that there are seven (7) areas within Kumar’s Illustrative District where neighborhoods are split to some degree. While some neighborhoods and/or subdivisions are split, Mr. Ely testified that 96 percent of subdivisions were not split by the District (TR4: 139). The Court will not fault Kumar for a 4 percent discrepancy. What is more, the number of neighborhoods within FISD make it practically impossible for any illustrative district to not split at least a handful of neighborhoods. Again, the liability stage is not meant to be an insurmountable hurdle. The Court consequently finds Defendants neighborhood/subdivision argument unpersuasive. 80 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 81 of 103 PageID #: 3200 Similarly, Defendants’ objections regarding congressional districts, school attendance zones, and the City of Plano are non-starters. As the Fifth Circuit has generally recognized when it comes to school boards, school boards “act as a unit. [Boards have] the responsibility for operating all of the schools [within their district]. [A] Board’s function and responsibility is not decentralized into [certain attendance zones or other] segments.” Ferguson v. Winn Par. Police Jury, 528 F.2d 592, 595 (5th Cir. 1976). The same logic applies here. FISD’s school attendance zones should be recognized to the degree that Kumar can recognize them. But failure to completely account for them should not require the dismissal of Kumar’s claim. The FISD Board will work as a whole to represent all students in FISD, regardless of which school attendance zone the child is in. Furthermore, any attempt to comply with these zones would only be a temporary success. Testimony at trial showed that these attendance zones change often—due to FISD’s commitment to a small school model—and that students zoned in certain middle school zones are presently divided into different high school zones. See Defendants’ Exhibits 6–7. In fact, Kumar’s daughters, who lived at the same home, attended two different high schools due to FISD’s frequent redistricting. FISD cannot hold Kumar to a community of interest standard that it does not hold itself to. As to congressional districts, Mr. Ely testified that any congressional district would be larger than any illustrative district he could create. Consequently, it is impossible to create any FISD district that does not split a congressional district. Lastly, as to the City of Plano, FISD argues that Kumar splits the Plano boundary at the southern border. Testimony at trial, however, demonstrated that this objection is misplaced. According to Mr. Ely on direct examination, FISD’s current attendance areas currently split the Plano border in nearly the exact same way as Kumar’s Illustrative District (TR4: 141). Defendants’ arguments are accordingly unpersuasive. 81 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 82 of 103 PageID #: 3201 The Court recognizes the distinct need to afford communities of interest the respect they deserve. Kumar’s Illustrative District preserves those communities, even if unintentionally, to a large degree. As the inquiry into illustrative districts is a flexible one that permits room to grow at the remedial phase rather than calling for perfection at the liability phase, the Court finds Kumar’s Illustrative District does not offend traditional districting principles.25 Having found that the District is sufficiently large and geographically compact, the Court consequently finds that Kumar satisfied the first Gingles factor. B. Gingles II To carry his Section 2 claim, Kumar must next establish that his purported minority coalition is politically cohesive. Kumar cannot carry that burden. The data set Mr. Ely provided Dr. Mayer is fatally flawed. The Court consequently cannot determine whether Kumar’s minority coalition is cohesive when it does not have a reliable estimate of who is in that coalition in the first place. As the Court previously stated, however, the Court will not stop here and render the suit null upon a technicality. Rather, the Court demonstrates below that even assuming arguendo that the data set Mr. Ely provided Dr. Mayer was reliable, Kumar still failed to carry his burden. To succeed in a Voting Rights Act action at the threshold Gingles stage, a plaintiff must demonstrate that the minority group he or she is espousing is “politically cohesive.” Clements, 986 F.2d at 743 (citing Gingles, 478 U.S. at 51). While political cohesiveness is similar to the concept of racially polarized voting (“RPV”), it is a distinct concept. “The notion of political cohesiveness contemplates that a specified group of voters shares common beliefs, ideals, Defendants’ also make fleeting arguments regarding Kumar’s purported failure to preserve Census Blocks or respect traditional boundaries. The Court need not address unsupported, periphery arguments especially when said arguments are not discussed in post-trial briefing. Similarly, Defendants, citing the Texas Education Code, raised equal protection arguments at trial with little support. Those arguments were quickly refuted as Mr. Ely testified, and stated in his expert report, that the District is compact enough to form a single member district with 1/7 of the population of FISD, thus enabling the creation of the remaining six districts. The Court finds that Kumar’s Illustrative District therefore complies with the equal total population requirement. 25 82 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 83 of 103 PageID #: 3202 principles, agendas, concerns, and the like such that they generally unite behind or coalesce around particular candidates and issues.” Id. at 744 (citing Monroe v. City of Woodville, 881 F.2d 1327, 1331 (5th Cir.1989), modified, 897 F.2d 763, cert. denied, 498 U.S. 822 (1990)). On the other hand, RPV “describes an electorate in which white voters favor and vote for certain candidates or propositions, and minority voters vote for other candidates or propositions.” Id. (citing Gingles, 478 U.S. at 53 n.21 (“‘[R]acial polarization’ exists where there is ‘a consistent relationship between [the] race of the voter and the way in which the voter votes . . . .”)). As such, “while a showing of racially polarized voting will frequently demonstrate that minority voters are politically cohesive, a showing that minority voters are politically cohesive will not, by itself, establish racially polarized voting.” Id. In other words, evidence of RPV does not “automatically establish minority political cohesiveness.” Id. at n.5. An example is helpful. Imagine a town. In that town, White citizens only vote for candidates that are of type A. Minority citizens, on the other hand, split their support among candidates of types B, C, and D. In this hypothetical community, RPV clearly exists, but there is no evidence of minority political cohesiveness. Id. This is because minorities do not coalesce around one particular type of candidate—they are split, not cohesive. There are a number of ways to prove political cohesion. One way is to demonstrate “that a significant number of minority group members usually vote for the same candidates . . . .” Id. at 743 (citing Gingles, 478 U.S. at 743). This can be demonstrated through statistical evidence or through testimony “from persons familiar with the community in question, provided that such testimony is reinforced with other evidence or is not otherwise rebutted.” Id. at 743–44 (citing Brewer, 876 F.2d at 453–54; Overton, 871 F.2d at 536). Another means of proving cohesion— albeit, a disputed means—is to establish the four factors gleaned from Nixon v. Kent Cty., Mich., 790 F. Supp. 738, 743–44 (W.D. Mich. 1992). In Nixon, the court enumerated four factors that 83 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 84 of 103 PageID #: 3203 could be considered and balanced when determining whether an aggregate minority group experienced the “common disability of chronic bigotry.” Id. at 743. Those factors include: (1) whether the members have similar socioeconomic backgrounds resulting in common social disabilities and exclusion; (2) whether members have similar attitudes toward significant issues affecting the challenged entity; (3) whether members have consistently voted for the same candidates; and (4) whether the minorities consider themselves “one” even in situations in which they would benefit independently Id. at 743–44; see also Clements, 999 F.2d at 897 n.8 (Jones, J., concurring) (citing Nixon and its four factors as persuasive authority for determining whether a minority coalition exists). Irrespective of which route a plaintiff takes to prove cohesion, when considering whether political cohesion exists, the central focus of a court must be on voting patterns. See Campos, 840 F.2d at 1245; see also Overton, 871 F.2d at 536 (stating that courts should be cautious about reaching conclusions on minority cohesion absent an inquiry into the political dynamics of the relevant community). Minority voters need not “prove that the reason they vote for the same candidates is linked to race or color”—all that is required is a voting pattern. Clements, 986 F.2d at 744; see also Teague v. Attala Cty., Miss., 92 F.3d 283, 290 (5th Cir. 1996) (citing Nipper v. Smith, 39 F.3d 1494, 1524 (11th Cir. 1994)) (“Plaintiffs are to present evidence of racial bias operating in the electoral system by proving up the Gingles factors. Defendants may then rebut the plaintiffs’ evidence by showing that no such bias exists in the relevant voting community.”).26 26 The Fifth Circuit has clarified the issue of political cohesion in multi-minority group cases: Baytown argues that, in order to show cohesion when there are two minorities that make up the minority group, plaintiffs must show that Blacks are cohesive, that Hispanics are cohesive and that Blacks and Hispanics are together cohesive. We think that that burden is too great, if not impossible, in certain situations. Instead, the proper standard is the same as Gingles: whether the minority group together votes in a cohesive manner for the minority candidate. That can be done by examining elections when there was a minority candidate. The key is the minority group as a whole. Of course, if one part of the group cannot be expected to vote with the other part, the combination is not cohesive. If the evidence were to show that the Blacks vote against a Hispanic candidate, or vice versa, then the minority group could not be said to be cohesive. But if the statistical evidence is that Blacks and Hispanics together vote for the Black or Hispanic candidate, then cohesion is shown. 84 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 85 of 103 PageID #: 3204 Political cohesion does not mean that minorities vote for the same candidates at all times. Indeed, “the fact that racially polarized voting is not present in one or a few individual elections does not necessarily negate the conclusion that the district experiences legally significant bloc voting.” Gingles, 478 U.S. at 57. “Furthermore, the success of a minority candidate in a particular election does not necessarily prove that the district did not experience polarized voting in that election; special circumstances, such as the absence of an opponent, incumbency, or the utilization of bullet voting, may explain minority electoral success in a polarized contest.” Id. Courts look for patterns among voters, not ironclad alliances. Although use of the phrase “pattern” may connote a light burden, cohesion, especially cohesion among various races, is not easy to prove. See Clements, 999 F.2d at 896–97 (Jones, J., concurring) (“[T]he problem of determining minority political cohesiveness under Gingles may be difficult even when the claims of one minority group are at issue. But it should be self-evident that the problem is compounded when different minority groups, with radically different cultural and language backgrounds, socioeconomic characteristics and experiences of discrimination seek Section 2 coalition status.”). “If one part of the [combined minority] group cannot be expected to vote with the other part the combination is not cohesive.” Brewer, 876 F.2d at 453. Thus, here, like in Brewer, “the determinative question is whether [B]lack-supported candidates receive a majority of the Hispanic and Asian vote; whether Hispanic-supported candidates receive a majority of the [B]lack and Asian vote; and whether Asian-supported candidates receive a majority of the [B]lack and Hispanic vote in most instances in the KISD area.” Id. If there is no evidence of political cohesion, then it “cannot be said that the [electoral scheme] thwarts distinctive minority group interests.” Clements, 986 F.2d at 744. Campos, 840 F.2d at 1245 (internal footnotes omitted). 85 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 86 of 103 PageID #: 3205 At trial, Kumar presented expert testimony, through Dr. Mayer, regarding voter behavior in eleven elections within FISD. Seven of the elections were endogenous elections. Endogenous elections concern former races for the FISD Board. The other four elections were exogenous elections. Exogenous elections concern former races that were held within the FISD geographic area for non-FISD Board positions: i.e., Governor, Senator, President. Dr. Mayer concluded that the endogenous and exogenous elections revealed a pattern of minority voter cohesion. Defendants, through Dr. Alford, counter that the evidence at trial established quite the opposite of what Kumar claims. Defendants argue that the evidence at trial “conclusively shows that the minority group at issue here—a coalition of Asian, Black, and Hispanic voters in FISD—is not politically cohesive” (Dkt. #122). The Court agrees with Defendants and finds that Kumar’s evidence falls short.27 i. Endogenous Elections The Court begins its analysis by first reviewing the endogenous elections. To understand Dr. Mayer’s conclusions, it does well to first understand Dr. Mayer’s methods. Dr. Mayer relied upon the data set provided by Mr. Ely—which combined surname and ACS data—to analyze endogenous and exogenous elections. Rather than employing an ecological regression analysis to estimate individual voting behavior from this data, Dr. Mayer relied on an ecological inference analysis. For the Court’s purposes, this simply means that Dr. Mayer used a technique which recognizes that certain “statistical assumptions underlying regression analysis do not hold when one has aggregate data rather than individual data . . . .” (Defendants’ Exhibit 1). Dr. Alford agrees that ecological inference techniques “use the available data in a more efficient way and so provide estimates of the relationship between the size of demographic groups and candidate support that 27 Again, this is assuming arguendo that the data set underlying Dr. Mayer’s conclusions is not unreliable. 86 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 87 of 103 PageID #: 3206 are potentially less biased and more certain than previous methods” (Defendants’ Exhibit 1). Thus, the Court need not question the reliability of Dr. Mayer’s methods. In the end, Dr. Mayer simply used a more reliable method for estimating the voting behavior of subgroups in the relevant electorate in FISD. With this understanding in mind, the Court now looks to the seven endogenous elections Dr. Mayer analyzed. 1. 2019 FISD Place 1 In 2019, Gopal Ponangi (“Ponangi”), an Asian candidate, ran against Nate Adams (“Adams”), a non-Hispanic White candidate. Ponangi won the election with 56.9% of the vote while Adams received 43.1% of the vote. According to the ecological inference results, the breakdown of the votes among the relevant races is as follows. Ponangi received 48.4% of the Latino vote, 57.3% of the African American vote, 65.7% of the Asian vote, and 55.8% of the White vote (Plaintiff’s Exhibit 6). Adams, on the other hand, received 51.6% of the Latino vote, 42.7% of the African American vote, 34.3% of the Asian vote, and 44.2% of the White vote (Plaintiff’s Exhibit 6). 2. 2019 FISD Place 3 Also in 2019, Muni Janagarajan, an Asian candidate, ran against Chad Rudy, a nonHispanic White candidate. Rudy won the election with 63.0% of the vote while Janagarajan received 37.0% of the vote. According to the ecological inference results, the breakdown of the votes among the relevant races is as follows. Janagarajan received 53.8% of the Latino vote, 60.8% of the African American vote, 78.8% of the Asian vote, and 26.0% of the White vote (Plaintiff’s Exhibit 6). Adams, on the other hand, received 46.2% of the Latino vote, 39.2% of the African American vote, 21.2% of the Asian vote, and 74.0% of the White vote (Plaintiff’s Exhibit 6). 87 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 88 of 103 PageID #: 3207 3. 2017 FISD Place 4 In 2017, there was a three-way race for Place 4. The candidates were Asanga Jayatilaka (“Jayatilaka”), an Asian candidate, Anne McCausland (“McCausland”), a non-Hispanic White incumbent, and Jeff Snowden (“Snowden”), a non-Hispanic White candidate. McCausland won the election with 55.0% of the vote; Snowden came in second with 36.7% of the vote; and Jayatilaka finished last with 8.3% of the vote. According to the ecological inference results, the breakdown of the votes among the relevant races is as follows. McCausland received 34.2% of the Latino vote, 33.6% of the African American vote, 28.4% of the Asian vote, and 60.6% of the White vote (Plaintiff’s Exhibit 6). Snowden received 32.6% of the Latino vote, 36.6% of the African American vote, 27.7% of the Asian vote, and 37.7% of the White vote (Plaintiff’s Exhibit 6). Finally, Jayatilaka received 33.3% of the Latino vote, 29.8% of the African American vote, 43.9% of the Asian vote, and 1.8% of the White vote (Plaintiff’s Exhibit 6). 4. 2017 FISD Place 5 Like the Place 4 election, there was a three-way race in 2017 for Place 5. The candidates were Grace Wang, an Asian candidate, Debbie Gillespie (“Gillespie”), a non-Hispanic white incumbent, and Brian Powell (“Powell”), a non-Hispanic White candidate. Gillespie won the election with 53.6% of the vote; Powell came in second with 30.5% of the vote; and Wang came in last with 15.8% of the vote.28 According to the ecological inference results, the breakdown of the votes among the relevant races is as follows. Gillespie received 29.8% of the Latino vote, 33.6% of the African American vote, 15.5% of the Asian vote, and 60.5% of the White vote (Plaintiff’s Exhibit 6). Powell received 26.2% of the Latino vote, 31.1% of the African American vote, 13.5% of the Asian vote, and 32.2% of the White vote (Plaintiff’s Exhibit 6). Finally, Wang 28 28 The Court notes that the vote breakdown Dr. Mayer provided the Court equals 99.9%. 88 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 89 of 103 PageID #: 3208 received 44.0% of the Latino vote, 35.3% of the African American vote, 71.0% of the Asian vote, and 7.4% of the White vote (Plaintiff’s Exhibit 6). 5. 2016 FISD Place 2 In 2016, there was a three-way race for Place 2. The candidates were Anjali Shirvaikar (“Shirvaikar”), an Asian candidate, Steve Noskin (“Noskin”), a non-Hispanic White candidate, and Phil Ramirez, a Hispanic candidate. Noskin won the election with 39.2% of the vote; Shirvaikar came in second with 35.9% of the vote; and Ramirez came in last with 24.9% of the vote. According to the ecological inference results, the breakdown of the votes among the relevant races is as follows. Noskin received 31.3% of the Latino vote, 32.8% of the African American vote, 24.9% of the Asian vote, and 41.3% of the White vote (Plaintiff’s Exhibit 6). Shirvaikar received 31.4% of the Latino vote, 38.4% of the African American vote, 46.4% of the Asian vote, and 35.0% of the White vote (Plaintiff’s Exhibit 6). Finally, Ramirez received 37.3% of the Latino vote, 28.8% of the African American vote, 28.7% of the Asian vote, and 23.7% of the White vote (Plaintiff’s Exhibit 6). 6. 2015 FISD Place 3 In 2015, there was a three-way race for Place 3. The candidates were Srikanth Gurrapu (“Gurrapu”), an Asian candidate, Chad Rudy, a non-Hispanic White candidate, and Bryan Powell (“Powell”), a non-Hispanic White candidate. Rudy won the election with 46.6% of the vote; Powell came in second with 39.3% of the vote; and Gurrapu finished last with 14.1% of the vote. According to the ecological inference results, the breakdown of the votes among the relevant races is as follows. Rudy received 29.9% of the Latino vote, 26.4% of the African American vote, 21.6% of the Asian vote, and 51.1% of the White vote (Plaintiff’s Exhibit 6). Powell received 32.9% of the Latino vote, 33.3% of the African American vote, 22.7% of the Asian vote, and 41.1% of the 89 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 90 of 103 PageID #: 3209 White vote (Plaintiff’s Exhibit 6). Finally, Gurrapu received 37.2% of the Latino vote, 40.3% of the African American vote, 55.7% of the Asian vote, and 7.8% of the White vote (Plaintiff’s Exhibit 6). 7. 2015 FISD Place 7 Finally, in 2015, there was a three-way race for Place 7. The candidates were Rajesh Singh (“Singh”), an Asian candidate, Carlos Gallardo, a Hispanic candidate, and John Hoxie (“Hoxie”), a non-Hispanic White incumbent. Hoxie won the election with 57.6% of the vote; Gallardo came in second with 32.4% of the vote; and Singh finished last with 10.1% of the vote.29 According to the ecological inference results, the breakdown of the votes among the relevant races is as follows. Hoxie received 35.1% of the Latino vote, 37.9% of the African American vote, 22.9% of the Asian vote, and 62.6% of the White vote (Plaintiff’s Exhibit 6). Gallardo received 33.5% of the Latino vote, 39.1% of the African American vote, 20.0% of the Asian vote, and 32.1% of the White vote (Plaintiff’s Exhibit 6). Finally, Singh received 31.4% of the Latino vote, 23.1% of the African American vote, 57.2% of the Asian vote, and 5.3% of the White vote (Plaintiff’s Exhibit 6). ii. Exogeneous Elections In addition to the seven endogenous elections, Dr. Mayer also reviewed four exogenous elections. Dr. Mayer utilized the same ecological inference methods to analyze these elections. The results and breakdown of those elections are as follows. 1. 2018 Democratic Gubernatorial Runoff In 2018, Lupe Valdez (“Valdez”), a Hispanic candidate, ran against Andrew White (“White”), a non-Hispanic White candidate in the Democratic Gubernatorial Primary. Valdez won the contest within FISD, and the eventual nod by the Democratic Party, with 57.2% of the FISD 29 The Court notes that the vote breakdown Dr. Mayer provided the Court equals 100.1%. 90 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 91 of 103 PageID #: 3210 vote. White received the remaining 42.8% of the vote. According to the ecological inference results, the breakdown of the votes among the relevant races is as follows. Valdez received 53.9% of the Latino vote, 54.6% of the African American vote, 45.4% of the Asian vote, and 60.3% of the White vote (Plaintiff’s Exhibit 6). White, on the other hand, received 46.1% of the Latino vote, 45.4% of the African American vote, 54.6% of the Asian vote, and 39.7% of the White vote (Plaintiff’s Exhibit 6). 2. 2018 Gubernatorial General Following the 2018 Democratic Gubernatorial Primary, Valdez challenged Republican incumbent Greg Abbott (“Abbott”) and the Libertarian Party’s candidate, Mark Tippetts (“Tippetts”), in a three-way race. Both Abbott and Tippetts are White. Abbott won the contest, and eventually the election, with 57.2% of the FISD vote; Valdez came in second with 42.0% of the vote; and Tippetts came in last with 1.8% of the vote. According to the ecological inference results, the breakdown of the votes among the relevant races is as follows. Abbott received 43.4% of the Latino vote, 33.3% of the African American vote, 32.8% of the Asian vote, and 64.7% of the White vote (Plaintiff’s Exhibit 6). Valdez, on the other hand, received 56.6% of the Latino vote, 66.7% of the African American vote, 67.2% of the Asian vote, and 35.3% of the White vote (Plaintiff’s Exhibit 6).30 3. 2018 Democratic Senate Primary Along with Democratic Gubernatorial Primary, 2018 was also home to the 2018 Democratic Senate Primary. There were three hopefuls: (1) former Congressman, Beto O’Rourke (“O’Rourke”), a non-Hispanic White candidate; (2) Latina community activist Sema Hernandez (“Hernandez”); and (3) retired African American businessman Edward Kimbrough 30 The parties did not provide the Court with ecological inference results for White. 91 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 92 of 103 PageID #: 3211 (“Kimbrough”). O’Rourke won the contest, and eventually the Democratic nomination, with 63.4% of the FISD vote; Hernandez came in second with 24.3% of the vote; and Kimbrough came in third with 12.3% of the vote. According to the ecological inference results, the breakdown of the votes among the relevant races is as follows. O’Rourke received 31.4% of the Latino vote, 57.6% of the African American vote, 36.3% of the Asian vote, and 74.4% of the White vote (Plaintiff’s Exhibit 6). Hernandez received 32.8% of the Latino vote, 28.2% of the African American vote, 34.8% of the Asian vote, and 19.7% of the White vote (Plaintiff’s Exhibit 6). Finally, Kimbrough received 35.8% of the Latino vote, 14.1% of the African American vote, 29.0% of the Asian vote, and 5.9% of the White vote (Plaintiff’s Exhibit 6). 4. 2012 Presidential General Election The last election analyzed concerns the 2012 Presidential General Election which was a race between the first African American President, and then-incumbent, Barack Obama (“President Obama”) and the non-Hispanic white, and former Governor of Massachusetts, candidate, Mitt Romney (“Romney”). Romney won the vote in FISD, but eventually lost the election, with 65.3% of the vote. President Obama received 34.7% of the vote. According to the ecological inference results, the breakdown of the votes among the relevant races is as follows. Romney received 39.9% of the Latino vote, 26.5% of the African American vote, 37.7% of the Asian vote, and 74.8% of the White vote (Plaintiff’s Exhibit 6). President Obama, on the other hand, received 60.1% of the Latino vote, 73.5% of the African American vote, 62.3% of the Asian vote, and 25.2% of the White vote (Plaintiff’s Exhibit 6). With the data for each election now before the Court, the Court continues to its findings. 92 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 93 of 103 PageID #: 3212 iii. Findings The data does not support Kumar’s conclusion that African Americans, Asians, and Hispanics are politically cohesive.31 Kumar argues that “Dr. Mayer’s factual findings from endogenous races demonstrated that Latino voters preferred a minority candidate in four of seven races, Black voters in six of seven races, and Asian voters in seven of seven” (Dkt. #123). True enough—but Kumar misses the forest for the trees. The Court is not concerned with whether minority voters prefer minority candidates; rather, the Court is concerned with whether all three minority groups are cohesive such that they usually supported the same candidate. While the data may show that minority voters prefer minority candidates at times, the data does not show cohesion. African Americans, Asians, and Hispanics are not politically cohesive in FISD. In five of the seven endogenous elections, African American voters did not vote for the Asian supported candidate. Likewise, Hispanic voters voted for White candidates over Asian-supported candidates in over half of the endogenous elections. It cannot be said that African Americans or Hispanics usually vote the same as Asians looking at this data. Similarly, in the two elections involving Hispanic candidates, neither Hispanic candidate garnered sufficient African American support— or Hispanic support, for that matter. While a majority is not required, it is telling that the African American vote in those cases was practically split in thirds. This means that: (1) African Americans were not voting cohesively among themselves; and (2) African Americans were not necessarily supporting Hispanic candidates. Those same Hispanic candidates also failed to garner significant support from Asian voters, and, as previously stated, Hispanic voters. While two See Teague, 17 F.3d at 978 (citing Magnolia Bar Ass’n, Inc. v. Lee, 994 F.2d 1143, 1159 (5th Cir. 1993)) (“The district court is not obliged to accept statistical evidence as conclusive on the question whether racially polarized voting exists.”). 31 93 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 94 of 103 PageID #: 3213 elections do not necessarily establish a pattern, or lack thereof, it is hard to say that Asian voters are voting cohesively with Hispanic and African American voters when the votes are breaking down into nearly equivalent thirds across Asian, Hispanic, and White candidates. This trend of minority voters breaking down into thirds is not cabined to these two elections. In the five threeway races, each minority group consistently fractured its vote into roughly thirds, save the Asian vote which was the closest to anything resembling cohesion. With this data in hand, it cannot be inferred that minority voters are uniting together behind particular candidates; rather, minority voters are fracturing into nearly equivalent percentages at most turns with no clear minority supported candidates emerging. A voting pattern fails to emerge in exogenous elections, too. First, none of the selected exogenous elections are particularly enlightening for the Court’s purposes. Unlike FISD Board elections, two of the four exogenous elections selected are partisan contests with party labels. Further, unlike the local nature of the endogenous elections, the exogenous elections are top-ofthe-ballot statewide or national elections. These elections present very different circumstances that color the data—most importantly being the partisan nature of the elections. With that caveat in mind, the Court finds that the exogenous elections also fail to establish political cohesion. In the 2018 Gubernatorial Primary Runoff, Hispanic and African American voters preferred the Hispanic candidate, Valdez, while Asian voters preferred the non-Hispanic White candidate, White. Voter cohesion among the three races was lacking in this election. Likewise, in the 2018 Democratic Senate Primary, Hispanic voters preferred African American candidate, Kimbrough, while African American and Asian voters flocked to non-Hispanic White candidate, O’Rourke. Once again, voter cohesion was not found among the three races. 94 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 95 of 103 PageID #: 3214 The other two exogenous elections are the least helpful because they are the most partisan and thus least representative of the electoral scheme at issue. In the 2018 Gubernatorial Election, Hispanics, African Americans, and Asians supported Valdez while Whites supported Abbott. Similarly, in the 2012 Presidential Election, Hispanics, African Americans, and Asians supported President Obama while Whites supported Romney. It is apparent that there was cohesion in these elections between the three relevant races at the outset; however, there is more to this story. In an unselected race which Defendants offered as a counterpoint, Hispanic candidate Ted Cruz ran as a Republican incumbent against non-Hispanic White candidate, Beto O’Rourke, for the United States Senate. O’Rourke received the support of the African American, Hispanic, and Asian voters while Cruz received the support of White voters. What becomes clear from this picture is that in highly partisan elections, race becomes less of a motivating factor when politics becomes the end all. Thus, while cohesion may be present in these two partisan, exogenous elections, this cohesion has an asterisk by it. Simply put, the cohesion in these two elections is likely attributable to a political factor—Republican versus Democrat—that is at least not formally present in the FISD electoral scheme being challenged. The Court’s conclusion that minority voters are not cohesive in FISD is further supported from lay testimony at trial. Testimony at trial established that elections for the FISD Board are becoming increasingly political as organizations such as Empower Texans, Frisco United and the Frisco Tea Party attempt to derail FISD initiatives. Elections have consequently become hostile as these organizations, and their followers, produce attack ads, disrupt Board meetings, and progressively turn non-partisan elections into battlegrounds. One particular issue that these organizations home in on concerns taxes. 95 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 96 of 103 PageID #: 3215 FISD is limited on the tax level it can seek from its voters to run the District. To exceed said limit, and therefore capture an additional tax break, the District must hold a tax ratification election and receive voter approval. Enter Empower Texans, Frisco United, the Frisco Tea Party, and the like. Opposed to public education, or at least an increase in taxes funding public education, these organizations consistently take an anti-FISD initiative stance and rally likeminded citizens to follow suit. The merits of public education aside, these tax ratification elections have consequently become heated, divisive affairs where political affiliation matters. In fact, the Collin County GOP Chapter recently began endorsing candidates despite the non-partisan nature of these races. The politicization of these tax ratification elections evidences a split among minority voters. According to Board Trustee Chad Rudy, candidate Muni Janagarajan garnered strong support from Empower Texans as he opposed raising taxes. Likewise, Gallardo ran on fiscal responsibility and strongly opposed tax increases. Opposition to taxes is not limited to candidates. Kumar, who serves as the Collin County Precinct Chair for the GOP, consistently supports antitax candidates and, according to Board Trustee John Classe, became combative at a Board meeting when tax ratification was discussed. On the other side of this debate are Grace Wang and Anali Shirvaikar. Wang and Shirvaikar both testified that they supported raising operational funds. Shirvaikar went on to say that she supports public education, teachers, and ensuring that students have what they need to succeed (TR4: 114). As such, Shirvaikar says she supports candidates who are likeminded such as: Natalie Hebert, Anne McCausland, Chad Rudy, and Debbie Gillespie. Each of those candidates is White. It is apparent from this evidence that Kumar’s purported coalition is not as cohesive as he asserts. Minority voters are splitting along the conservative/liberal line that is forming in FISD and aligning not with one another, but with where each respective voter’s ideology rests on the 96 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 97 of 103 PageID #: 3216 public education issue. This split not only affects the policies of those who run for office, but the candidates each citizen supports. Kumar only supports conservative candidates; Gallardo supported Abbott over Valdez because of politics; Shirvaikar only supports pro-education candidates, and so on. Further evidence established that while Empower Texans and the like attempt to infiltrate FISD elections and amass a base, that base is sufficiently opposed by the propublic education electorate. According to Rudy and Classe, the Frisco Tea Party and Frisco Republican Club have a great deal of supporters within FISD; however, generally speaking, any candidate who opposes the District, or receives support from organizations like Empower Texans, loses to candidates who support public education (TR4: 71, & 87–88). As Dr. Alford noted, Kumar cannot meet the second Gingles factor when this is the reality of FISD. Testimony at trial demonstrated that minority voters are not coalescing; rather, they are splintering over tax/public education issues and, relatedly, whether they support Empower Texans, Frisco United, and the Frisco Tea Party.32 FISD Board elections simply are not the cohesive affairs that Kumar claims they are. Another example of the lack of cohesion between this alleged coalition came in the form of a sentiment shared by some witnesses throughout trial. This sentiment can be reduced to the following phrase: “This *fill-in-the-blank with person, race, ethnicity* does not represent me.” This sentiment was shared at least three times throughout trial. First, there are the text messages sent by Kumar. In Defendants’ Exhibit 15, a string of texts shows Kumar discussing the candidacy of Gopal Ponangi. In reference to Ponangi, Kumar texted the following: (1) “[Ponangi] is the most incompetent among the candidates in this election cycle”; and (2) “Gopal does not represent me. For that matter, he does not represent the Indian 32 Again, this is in contrast to the partisan exogenous elections which hosted a variety of complex political issues rather than one finite issue. 97 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 98 of 103 PageID #: 3217 community” (Defendants’ Exhibit 15). When questioned about his statement, Kumar doubled down on his texts and stated that Ponangi “definitely [did] not” represent him because of “[d]ifferent values” (TR2: 21). Second, there is the testimony of Dr. Payne. Dr. Payne’s testimony is illustrative of the commonsense notion that various ethnic groups are culturally, theologically, and politically diverse. On direct, Mr. Payne gave the example of a Christian man and a Muslim man. The Christian man, Mr. Payne stated, could not adequately represent the needs of a Muslim man, because these two men have fundamentally different concerns, histories, and life experiences. Like the Christian and the Muslim men, Dr. Payne concluded, a White man cannot adequately represent a Black man—they also have far too many different concerns, histories, and life experiences. In other words, they are not cohesive. Finally, there is the testimony of Anjali Shirvaikar. Ms. Shirvaikar lost a three-way race against a non-Hispanic White candidate and noted that she received no support in her candidacy from the NAACP or LULAC. When asked whether Kumar spoke for the Asian community in FISD Shirvaikar answered: “I know he does not speak for me . . . . (TR3: 116). A common theme arises from Kumar, Dr. Payne’s, and Ms. Shirvaikar’s testimony: one size does not fit all in FISD. The underlying philosophy found in Dr. Payne’s testimony fits the narrative heard across all three witnesses. Dr. Payne admitted that he could not possibly represent a Muslim man because, as a Christian, he did not share common beliefs with his Muslim neighbor. In other words, Dr. Payne acknowledged that he and the hypothetical Muslim man are not cohesive in certain regards. Kumar would have the Court embrace the truth that minorities need adequate representation—such as a Muslim man representing Muslim men and a Christian man representing Christian men—yet disregard Dr. Payne’s conclusion that people of different backgrounds cannot 98 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 99 of 103 PageID #: 3218 adequately represent one another. In other words, Kumar would have the Court say that while a Christian man cannot represent a Muslim man, a Muslim man can represent Jewish and Buddhist men. Kumar cannot have it both ways. Dr. Payne’s testimony is commonsense: humanity is composed of people of all walks of life and it is difficult, though not impossible, for people from “Walk of Life A” to represent people from “Walk of Life B.” The key here is that it is difficult, not impossible. What matters is who the people choose to represent themselves. This principle is one of the cornerstones of a democratic republic. By foregoing governance by the masses and electing to institute a system based in popular sovereignty, the Founders implemented a political scheme that requires leaders to be beholden to the electorate which entrusts them. Consequently, voters choose leaders who they believe represent their interests. Race, color, and creed do not necessarily preclude a member of the electorate from selecting a particular candidate to represent him or her, but these factors certainly can affect the votes of various citizens. To prevail on the second Gingles factor, Kumar must establish that the purported minority coalition approaches its political decisions cohesively, contrary to the conclusion of Dr. Payne’s hypothetical. Dr. Payne, Ms. Shirvaikar, and even Kumar’s testimony support the conclusion that Kumar falls short of his burden. Kumar would have the Court believe that either: (1) African Americans, Hispanics, and Asians all weigh the various political factors before them the same; or (2) African Americans, Hispanics, and Asians disregard these considerations all together and merely focus on the race of each candidate. The Court finds neither option tenable. Testimony evidences that the voters in FISD consider a panoply of issues and disagree, naturally, on many of the issues before them. This discord is not limited to White voters versus everyone else. It exists among all races. Testimony from Kumar, Dr. Payne, and Shirvaikar—not to mention the endogenous and exogenous election 99 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 100 of 103 PageID #: 3219 data the Court already reviewed—demonstrates this. Thus, the Court cannot agree with Kumar that, contrary to the conclusion of Dr. Payne’s hypothetical, the African American man (and woman) necessarily speaks for the Asian and Hispanic men (and women). While it is certainly possible for different races to be cohesive, the evidence here shows that such cohesion does not exist in FISD. This leads to the Court’s last finding supporting its decision. The most inclusive groups that work toward equality for all, OneVoice and the Frisco Mayer Ad Hoc Committee on Diversity and Inclusion, are not politically cohesive across all three races. As previously stated, OneVoice was started in partnership with FISD by Dr. Walters and others. The group is composed of African American and Hispanic citizens. During Dr. Walter’s tenure with the organization, however, OneVoice never partnered with the Asian community or any Asian organizations. It is unclear whether the Asian community ever joined OneVoice. Similarly, Dr. Payne expressed that, as a member of the Frisco Mayer Ad Hoc Committee on Diversity and Inclusion, it is difficult to garner Hispanic involvement because language serves as a barrier that makes it difficult to work together. This anecdotal evidence alone is not enough to support the Court’s conclusion, but when considered in conjunction with: (1) the endogenous and exogenous data; (2) the discord among minority voters over tax initiatives; and (3) the admissions by witnesses—including the named Plaintiff—that other minority citizens do not represent them, the Court necessarily concludes that Kumar failed to establish the second Gingles factor.33 C. Gingles III “Failure to establish any one of the Gingles factors precludes a finding of vote dilution . . . .” See Clements, 986 F.2d at 743 (citing Gingles, 478 U.S. at 50; Overton, 871 F.2d 33 See Nixon, 790 F. Supp. at 743–44. 100 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 101 of 103 PageID #: 3220 at 538). Because Kumar failed to establish the second Gingles factor, the Court need not address the third. 34 V. Conclusions of Law To prevail on a Section 2 claim post-Gingles, a plaintiff must first demonstrate that “the minority group is “sufficiently large and geographically compact to constitute a majority in a single member district.” 478 U.S. at 50–51. Here, Kumar established a majority-minority District with a MCVAP of, at least, 50.03 percent. Further, despite Defendants’ objections, Kumar’s Illustrative District is geographically compact and respects traditional boundaries and communities of interest. While Kumar carried the day on the first Gingles factor, he fell short of his burden on the second. Following the first Gingles factor, a plaintiff must demonstrate that the minority group coalition he or she espouses is “politically cohesive.” Id. The Court concludes that Kumar failed to do that for two reasons. First, the data set Mr. Ely provided Dr. Mayer—Kumar’s expert for the second and third Gingles factor—is unreliable and thus does not provide the Court with an accurate picture of voter cohesion in FISD. Mr. Ely relied on the Lauderdale & Kestenbaum List of Asian Surnames to estimate the Asian population. The Lauderdale & Kestenbaum List of Asian Surnames, however, is unreliable. Not only is the List unreliable, but Mr. Ely failed to specify which lists he used and admitted that he did not account for differing sensitivities among the various ethnic groups within the List. Consequently, when Mr. Ely combined Asian, Hispanic, and African American voter data for Dr. Mayer, he produced an infected data set with at least a third of the data being not only suspect, but unreliable. This flaw in Mr. Ely’s methodology Defendants aver that following Judge Ho’s concurring and dissenting opinion in Anne Harding, 948 F.3d at 315– 16 (2020), Kumar is now required to prove an “additional” Gingles factor: “[Plaintiffs] must additionally prove that the proposed district will in fact perform as plaintiff’s hope.” The Court need not address Judge Ho’s opinion construing Abbott v. Perez, --- U.S. ---, 138 S. Ct. 2305 (2018). Kumar failed to carry his burden on the original Gingles factors. The Court accordingly does not need to consider whether a fourth factor exists. 34 101 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 102 of 103 PageID #: 3221 undermined all of Dr. Mayer’s subsequent findings and rendered any sound conclusion on voter cohesion inconceivable. Kumar, accordingly, could not prove the second Gingles factor from the outset. Even if Dr. Mayer’s findings were not skewed from the outset based on Mr. Ely’s reliance on the Lauderdale & Kestenbaum List of Asian Surnames, however, Kumar still would not have proved that his purported minority coalition was politically cohesive. The endogenous and exogenous data evidences that minority voters are not coalescing around particular candidates but rather fracturing into nearly equivalent percentages at most turns. Further, testimony at trial revealed: (1) significant discord among FISD voters—minority voters included—over tax initiatives; (2) a theme of lay witnesses qualifying who represents them; and (3) a lack of cohesion in the most diverse, minority supportive organizations. Invidious discrimination must be stomped out in this Nation. No reasonable argument exists that could support a system that permits such evil. Yet simply because someone claims discrimination exists does not make it so. Ours is a Nation premised upon the rule of law where each citizen is entitled to his or her day in court should they choose to bring forward an allegation. But before that allegation becomes true, that same citizen must carry his or her burden. Here, Kumar received his day in court, but he did not carry his burden. The Court consequently concludes that Frisco ISD’s at-large electoral system does not violate Section 2 of the Voting Rights Act. 102 Case 4:19-cv-00284-ALM Document 124 Filed 08/04/20 Page 103 of 103 PageID #: 3222 . CONCLUSION In accord with the Court’s findings of fact and conclusions of law, the Court finds that Kumar failed to meet his burden with respect to his claims under Section 2 of the Voting Rights Act. The Court dismisses Kumar’s claims with prejudice and will issue judgment accordingly. SIGNED this 4th day of August, 2020. ___________________________________ AMOS L. MAZZANT UNITED STATES DISTRICT JUDGE 103
]]>
Some case metadata and case summaries were written with the help of AI, which can produce inaccuracies. You
should read the full case before relying on it for legal research purposes.
This site is protected by reCAPTCHA and the Google
Privacy Policy and
Terms of Service apply.
