Opinion
6:22-cv-01274-EFM
07-10-2024
MIGUEL COCA and ALEJANDRO RANGEL-LOPEZ, Plaintiffs, v. CITY OF DODGE CITY, KANSAS, a municipal corporation, Defendant.
MEMORANDUM AND ORDER
ERIC F. MELGREN CHIEF UNITED STATES DISTRICT JUDGE
Silhouetted against the glow of legends and memories of the Wild West, Dodge City now exists as a multicultural, rapidly expanding community populated mostly by recent Hispanic immigrants and their families. Faced with this changing demographic, some residents of Dodge City seek to bring about a new method of election for City Commissioners through legal action. Plaintiffs Miguel Coca and Alejandro Rangel-Lopez claim that Defendant Dodge City's use of at-large elections for its City Commissioners violates their voting rights under Section 2 of the Voting Rights Act. Through this lawsuit, they seek to enjoin Defendant's current at-large voting scheme and instead implement district-based elections for City Commissioners.
After a year and a half of litigation, this matter came for bench trial on February 26, 2024. The Court has thoroughly considered the evidence and arguments presented at trial, the parties' post-trial submissions, and the relevant law, and makes the following findings of fact and conclusions of law as required by Rule 52(a) of the Federal Rules of Civil Procedure.
I. Findings of Fact
A. The Parties
1. Plaintiff Miguel Coca is a resident of Dodge City. He identifies as Latino/Hispanic,is a citizen of the United States, and is over the age of eighteen.
The Court will use Hispanic and Latino interchangeably throughout this Order.
2. Coca was born and raised in Dodge City. He first lived in a trailer park called Lazy Acres in south Dodge City, and now lives on the east side, south of Comanche Street.
3. Coca attended public school in Dodge City, specifically Beeson Elementary, Comanche Intermediate, Dodge City Middle School, and Dodge City High School.
4. Coca has voted in every Dodge City Commission election since 2019.
5. Coca has served on the Dodge City Library Board, a position to which he was appointed by the Dodge City City Commission (the “City Commission”).
6. Plaintiff Alejandro Rangel-Lopez is a resident of Dodge City, Kansas. Rangel-Lopez identifies as Latino/Hispanic, is a citizen of the United States, and is over the age of eighteen.
7. Rangel-Lopez was born and raised in Dodge City, Kansas, and is a first generation Mexican American. His parents immigrated to the United States in the 1990s.
8. Rangel-Lopez attended Dodge City schools for pre-kindergarten through high school.
9. Rangel-Lopez was selected for an internship in the Dodge City City Manager's Office in the Summer of 2022.
10. During this internship, Rangel-Lopez never voiced his concerns with the at-large voting method to any city employees.
11. Rangel-Lopez was also appointed to serve on the Steering Committee tasked with formulating Dodge City's “Strategic Plan” for welcoming and integrating immigrants into the Dodge City community.
12. Rangel-Lopez describes himself as “a little bit” litigious. This is his second lawsuit filed in which he alleges voting rights violations. The first, Rangel-Lopez et al v. Ford County, Kansas, Clerk, involved the relocation of Dodge City's polling location to outside Dodge City limits by Ford County. Rangel-Lopez voluntarily dismissed that suit without prejudice.
Case No. 2:18-cv-02572-DDC-TJJ.
13. Defendant is the City of Dodge City, Kansas, a municipal entity capable of being sued under Kansas state law.
B. Procedural History
14. Plaintiffs filed the present case on December 15, 2022.
15. Plaintiffs originally asserted claims against Defendant and each of the then-present City Commissioners in their official capacities for violating Section 2 of the Voting Rights Act (“VRA”), the Fourteenth Amendment's Equal Protection Clause, and the Fifteenth Amendment.
16. As relief, Plaintiffs sought to create a five-district voting system in Dodge City and remove the election cycles from November of odd-numbered years to November of even-numbered years.
17. On April 18, 2023, the Court dismissed Plaintiffs' claim under the Fifteenth Amendment.
18. On December 11, 2023, the Court granted in part and denied in part Defendants' Motion to Exclude Dr. Matthew Barreto. Specifically, the Court found that Dr. Barreto's homogenous precinct analysis was inapplicable and unreliable as applied to Dodge City. Soon after, the Court denied Defendants' Motion for Summary Judgment.
19. On February 23, 2024, the Court dismissed the individual Defendants from the case, leaving only the City as the remaining Defendant.
20. A trial before the bench commenced on February 26, 2024.
21. During trial, the Court heard from the following witnesses presented by Plaintiff: Plaintiff Alexandro Rangel-Lopez, Plaintiff Miguel Coca, Dr. Matthew A. Barreto, Dr. Kassra A. R. Oskooii, Dr. Christina Bejarano, Dr. Ruben Martinez, Monica Vargas, and Jan Scoggins.
22. After Plaintiffs rested their case, Defendant moved for judgment on partial findings.
23. The Court granted Defendant's motion as to Plaintiffs' claim under the Fourteenth Amendment. Furthermore, the Court granted Defendant's motion for judgment on Plaintiffs' request for moving elections to Novembers of even-numbered years. As stated more fully on the record, the Court did not find itself with the authority-or compelled by the evidence-to set aside Kansas state law governing election cycles.
24. Defendant presented the following witnesses: Dr. Jonathan Katz, Nickolaus Hernandez, Ernestor De La Rosa, Dr. Kimberly Nelson, Sharon Seibel Robb, Cherise Tieben, Melissa McCoy, Dave J. Rebein, and Rick Sowers.
Due to Seibel Robb's scheduling conflicts, Plaintiffs allowed Defendant to examine her as a witness during their case in chief.
25. The trial lasted five days, concluding with closing arguments on March 1, 2024.
26. At the close of trial, the Court instructed the parties to submit proposed findings of fact and conclusions of law by March 22, 2024, with responses due by April 5, 2024.
27. The parties have submitted their respective proposed findings of fact and conclusions of law. This matter is fully ripe for the Court's ruling under Federal Rule of Civil Procedure 52(a).
C. Dodge City Elections Background
28. Dodge City is a political subdivision within the meaning of, and subject to, the requirements of the VRA, 52 U.S.C. § 10301.
29. Dodge City is within Ford County, along with the townships of Bucklin, Spearville, and Ford City.
30. Dodge City began as a cow town, famous for gunslingers, saloons, and its Boot Hill cemetery. It has since refined its reputation to one for tourism and new businesses, including a meatpacking plant that employs several thousand workers.
31. Dodge City occupies approximately 25 square miles. Driving from one side of the city to the other takes 15 minutes or less.
32. Dodge City, Kansas operates under a commission-manager form of government.
33. The Commission is made up of five Dodge City residents who are elected using an at-large method of election.
34. In 2020, Dodge City's population was 27,788.
35. As a point of contrast, the City of Wichita, Kansas, has a population of roughly 400,000. It holds district-based elections for six districts, resulting in one city council member per roughly 70,000 people.
36. Dodge City adopted the at-large election method in 1971. There is no evidence Dodge City adopted this method of voting to discriminate against minority voters or dilute the Latino vote in Dodge City.
37. Every election cycle, three seats on the Commission are up for election.
38. Dodge City residents may vote for up to three different candidates.
39. Commissioners serve either four-year or two-year terms based on the number of votes that the top three finishers in an election receive. The winner with the least votes serves a two-year term, and the top two finishers serve four-year terms.
40. The elections for the Commission are non-partisan.
41. Pursuant to Kansas law, Commission elections are conducted by secret ballot.
42. Although Dodge City determines the method of election, Ford County administers each election.
43. In 1998, Ford County consolidated its seven polling locations within Dodge City to just one. Shortly thereafter, the Department of Justice sent monitors to ensure that Dodge City was meeting its obligations under the Voting Rights Act. Apparently satisfied, the monitors left after five or six years without any negative findings, charges, proposed consent decrees, or any other type of remedial action proposals.
44. Following a 2015 change to Kansas state law, K.S.A. § 25-2107, Dodge City changed the Commission election schedule from April of even-numbered years to November of odd-numbered years. The first election pursuant to the new schedule occurred in 2017.
45. In 2018, Ford County moved the polling location for a general election on relatively short notice to a location outside Dodge City's limit. Ford County stationed bilingual employees at the previous polling locations and offered free rides to take voters to the new location. Ford County also offered advance voting.
46. The move of the polling location created a backlash, including Rangel-Lopez bringing a lawsuit against Ford County. Although Rangel-Lopez voluntarily dismissed his case, a congressional oversight committee concluded that local officials did not hold meetings with concerned citizens or address Latino concerns before making the decision to move the polling location outside city limits.
D. Ethnic Makeup of Dodge City
47. Over the past twenty years, the Hispanic population in Dodge City has grown significantly.
48. Between 2000 and 2020, the total population of Dodge City increased from 25,152 to 27,788.
49. During the same period, the white population-not Hispanic or Latino-decreased from 13,060 in 2000 to 8,129 in 2020. In 2020, whites composed only 29.3% of the total Dodge City population.
50. Conversely, the Hispanic population increased from 10,784 in 2000 to 17,759, or 63.91% of the total population in 2020.
51. In 2020, other ethnic minorities-including Black and Asian-rounded out Dodge City's population with 1,900.
52. In 2000, there were 9,905 whites in the white citizens of voting age population (“CVAP”) in Dodge City.
53. By 2022, that number decreased to an estimated white CVAP of 6,468, which constitutes 43.87% of the total CVAP in Dodge City.
54. In 2000, there were 2,560 persons in Dodge City's Hispanic CVAP, or 19.53% of the total CVAP.
55. By 2021, that percentage was 46.13%, with 6,398 persons. And in 2022 the Hispanic CVAP in Dodge increased to an estimated 48.68%, a leap of 2.55%, with the total persons in Dodge City's Hispanic CVAP increasing to an estimated 7,176.
56. The single greatest reason for this rapid increase in the Hispanic population over the last few decades is a surge in immigration to Dodge City from Puerto Rico, Mexico, Guatemala, Cuba, and other Latin American countries.
57. Many of these immigrants are documented, but a sizable number are not.
58. One of the reasons for this rise in immigration is the meatpacking plant, which employs roughly 70% Latino workers.
59. The Latino population is considerably younger in Dodge City than the white population. For example, Latino students make up approximately 80% of the enrollment at Dodge City's public schools.
60. Based on the evidence of consistent trends presented to the Court, it is more likely than not that these trends-a decreasing white CVAP and rapidly increasing Hispanic CVAP- have continued past 2022 until the present and will continue going forward.
61. Today, Hispanics constitute the largest ethnic group by total population and by CVAP in Dodge City. Based on the clear trends presented to the Court, it is more likely than not that Hispanics presently comprise a majority of the total CVAP in Dodge City.
E. Geography of Dodge City
62. The Court heard multiple witnesses testify as to the layout of Dodge City. Most of these witnesses distinguished between “North Dodge” and “South Dodge.” In general, North Dodge incorporates the area north of Comanche Street. South Dodge occupies the area below Comanche Street.
63. North Dodge has significantly more whites than Latinos.
64. South Dodge boasts a heavily majority-Latino population.
65. Within South Dodge is Old Dodge City, which generally begins from Central Avenue to 14th Avenue, east-to-west, and goes from either around Hoover Pavilion or Wyatt Earp Boulevard to Comanche Street, south-to-north. This is a distinct, high-traffic area that contains government buildings, certain tourist attractions like the Boot Hill Museum, shops, and restaurants. It also has numbered streets, unlike those streets east of Central Avenue.
66. The area east of Central Avenue and south of Comanche may be referred to as Mideast Dodge City. This area is heavily residential, with small single-family housing units. Unlike the area west of Central Avenue, it has lettered streets, from A- to- P.
67. Another significant boundary is the Wyatt Earp Corridor, a major four-lane intersection that goes through the width of the City from west to east or east to west.
68. Although it forms a natural boundary in Dodge City, the Wyatt Earp Corridor is also home to many hotels, restaurants, shops, landmarks, and the Amtrak station and is a high-population area.
69. Turning to North Dodge, the area includes the city's country club, larger houses, middle-class housing development, several major retailers, and the City's community college and high school.
70. Sixth Avenue, running north and south, is another natural boundary which helps distinguish between Northeast Dodge and Northwest Dodge.
F. History of concerns with at-large voting in Dodge City
71. The issue of at-large elections negatively affecting Latino representation on the City Commission has never been raised before the City Commission by any concerned members of Dodge City's Latino community.
72. In April or May of 2011, Sharon Seibel Robb attended a conference in her capacity as the Ford County Clerk. An attorney, Bruce Adelson, hosted a presentation on the potential vote dilatation effects of at-large elections.
73. On June 29, 2011, the Department of Justice (“DOJ”)'s Civil Rights Division contacted Seibel Robb to request election information for Dodge City.
74. Seibel Robb informed Ford County Commissioners of the DOJ's letter and Adelson's presentation. The Ford County Commissioners directed Seibel Robb to contact Adelson. In an email to Seibel Robb, Adelson opined that the DOJ was conducting a Section 2 investigation and instructed Seibel Robb that “if the DOJ determines that a Voting Rights Act violation exists, DOJ will notify Dodge City of its findings and inform the City that it is planning an enforcement action.”
75. Seibel Robb subsequently informed Dodge City officials of the DOJ's letter.
76. Adelson conducted his own investigation into Dodge City's elections. On August 26, 2011, Adelson sent an advertising letter to Seibel Robb, claiming that Dodge City appeared to violate the VRA and offered his services as an attorney. Three days later, Ford County executed a retainer agreement with Adelson. Adelson proceeded to interview Dodge City community members and meet with city officials. He again stated that Dodge City was potentially in violation of Section 2 and suggested further undefined studies, once again offering his further services. On December 11, 2011, Adelson sent an invoice to Ford County, which appears to be the last interaction between them.
77. After Seibel Robb complied with the DOJ request, the DOJ never contacted Ford County or Dodge City regarding voting rights issues again. As Adelson suggested in his email, the DOJ's failure to contact Dodge City again indicates that the DOJ did not find that Dodge City's voting practices violated Section 2 of the VRA.
78. In February 2019, Jan Scoggins-a City Commissioner at the time-emailed then-City Manager Cherise Tieben and asked to add the issue of at-large elections to the City Commission's meeting agenda. In her email, Scoggins stated that she represented a community group. Tieben instructed Scoggins that she could raise the issue herself during the meeting and the community group could attend and raise it during the visitors' section as well.
79. Shortly before the City Commission's meeting, Rangel-Lopez recalls discussing at-large elections with Scoggins and others at a Ford County Democrats' meeting.
80. The City Commission's meeting occurred on March 4, 2019. No members from Scoggins' community group attended. Near the end of the meeting, Scoggins suggested moving to a mixed election system where three Commissioners would be elected by district and two at large. She also asked the City Commission to consider term limits. There is no evidence that Scoggins disclosed the reason for her request to the City Commission.
81. After some discussion, the City Commission directed city staff “to get more information for further discussion.”
82. On April 15, 2019, pursuant to City Commission direction, City Attorney Brad Ralph discussed information relating to switching to a mixed voting system. Two City Commissioners, Brian Delzeit and Joyce Warshaw, stated they were opposed to looking further into the matter. No other Commissioners, including Scoggins, made any statements on the issue.
83. Beyond this isolated incident in 2019, there is no evidence that any member or affiliate of the Hispanic community in Dodge City has ever voiced opposition to the at-large method of elections to any City Commissioner-at least not until this lawsuit.
84. In contrast, several members of the Hispanic community-including City Manager Nickolaus Hernandez, a Latino-have voiced support for at-large elections, or at least considered a change to district-based elections unnecessary.
85. The most nuanced opinion came from Ernestor De La Rosa, a witness the Court finds to be very credible and highly persuasive.
86. Born in Mexico and having moved to Dodge City when he was 14, De La Rosa previously served as an Assistant City Manager in Dodge City. During that time, he was instrumental in many of the multicultural and welcoming programs described below.
87. De La Rosa stated, “looking at [district-based elections] from my own lived experience as a first-generation immigrant, Latino, whatnot, I just don't know if it helps us or hurts us.” He further stated that while he believed it worked well for larger towns like Wichita and Topeka, he was unsure if it would have any positive effect in Dodge City.
88. Explaining his opinion, De La Rosa expressly acknowledged the lack of Latino engagement in Dodge City politics. He believed this stemmed in large part from lack of education regarding local politics, particularly among first generation immigrants who might not understand local elections. He also stated that local Latino community leaders were generally unsupportive of Blanca Soto, a Latina, when she campaigned for City Commission. Because she was a single mom of two children working a full-time job, De La Rosa noted Soto was simply unable to put up signs or campaign as much with her home responsibilities. De La Rosa concluded that the infrastructure to support Latino candidates simply does not exist in the Dodge City Latino community at present.
89. He firmly stated that switching to district-based elections was not “the golden bullet per se.” Rather, he believed that if Latinos in Dodge were an active voter community, they “would flip any committee and commission in the community tomorrow.”
90. De La Rosa repeated this sentiment on cross-examination, stating, “If my community were to be educated and showed up to the polls, those commissions, those boards would look different tomorrow.”
91. Similarly, Plaintiff Miguel Coca testified that many in the Hispanic community are apathetic toward local elections.
G. Cultural relations in Dodge City
92. The Court received credible testimony that the Latino community in Dodge City is vibrant and active. Latinos own and operate many downtown businesses and have leadership positions in community and professional organizations, such as the Chamber of Commerce, local unions, charitable organizations, political parties, and educational fields.
93. To engage its Latino population, Defendant operates several organizations specifically targeted to address concerns from its Latino residents.
94. The first of these, the Cultural Relations Advisory Board, advises the City Commissions on policies and practices designed to increase intercultural awareness, education, and celebration. It also puts on events in the community. Several members of the Latino community, including Rangel-Lopez, serve on this board.
95. Other programs include the Welcoming Dodge City Strategic Plan, the International Festival held yearly during Welcoming Week, New Americans Dinner, and Engage Dodge.
96. Dodge City offers its bilingual employees extra pay. Part of the requirement for this pay increase is the employee's willingness to assist Spanish speaking residents with their concerns or serve as interpreters for persons interacting with city officials when necessary.
97. Dodge City also gives its bilingual employees the option to work for Ford County at the polling locations on election days without requiring them to take vacation time.
98. At the time of trial, the National League of Cities named Dodge City as a finalist for its cultural diversity award because of Dodge City's Engage Dodge program.
99. Dodge City also publishes all voting and business information online and via printed flyers in English and Spanish.
100. Dodge City has been instrumental in bringing mobile immigration services to southwest Kansas through the Southwest Kansas Coalition. It also provides free space at its facilities for naturalization ceremonies, working with Ford County after these ceremonies to help new citizens register to vote.
101. Despite Defendant's best efforts through the above programs, first-generation immigrants or mixed status families in Dodge City generally do not involve themselves in local politics.
102. Voter turnout among Latinos in Dodge City is very low, with Latinos comprising roughly 30% of voters in the 2022 general election despite being the largest ethnic CVAP group.
103. The most significant and prevalent complaint from Plaintiffs' witnesses was the perceived lack of upkeep, particularly with regards to roads and sidewalks, in South Dodge compared to North Dodge. Rangel-Lopez complained that Dodge City prioritizes road maintenance for North Dodge and provides better playground facilities. He also complained that many sidewalks in South Dodge have been buckled by tree roots. However, neither Rangel-Lopez nor any other of Plaintiffs' witnesses were able to testify as to how the City Commission makes decisions as to any of their complaints. Furthermore, Rangel-Lopez never mentioned the issue of road maintenance to any Dodge City official while interning for the City Manager's office.
104. City Manager Hernandez explained that under Kansas law, sidewalk maintenance is the responsibility of the adjacent homeowners. However, Dodge City does not pressure homeowners into sidewalk repair while at the same time offering financial aid to any wishing to offset the costs of sidewalk maintenance.
105. Similarly, Hernandez testified that in 2017, Dodge City created a “pavement condition index,” wherein Dodge City reviewed and scored every single street within city limits by level of deterioration. It then divided the city into six zones and currently prioritizes the zones with the lowest overall scores on a rotational basis. The scores for each road were recently updated in 2023. He credibly testified that Dodge City does not prioritize road maintenance in North Dodge over road maintenance in South Dodge.
106. Dodge City is also in the middle of a $15 million “streetscape” project, wherein much of historic Dodge City, i.e., areas south of Comanche, will experience renovation and expansion. This includes new playgrounds for three parks, new water features, and both renovation and reconstruction of entire streets.
H. Plaintiffs' Gingles I expert-Dr. Oskooii
107. To meet their burden under Gingles I-that is, to show that the Latino population in Dodge City is sufficiently large and geographically compact to constitute a majority in at least one single-member district-Plaintiffs relied on Dr. Kassra A. R. Oskooii to produce maps dividing Dodge City into five districts.
108. Dr. Oskooii is a tenured Professor of Political Science at the University of Delaware.
109. Dr. Oskooii received a Bachelor's Degree in Political Science in 2008 from the University of Washington and a Master's and Doctorate in Political Science from the University of Washington in 2013 and 2016, respectively.
110. Dr. Oskooii has taught classes related to voting rights and representation, redistricting, American politics and political behavior, and race and ethnic politics. As part of his courses on redistricting, Dr. Oskooii covers apportionment, the census, and traditional districting criteria. He also teaches students how to draw single-member district maps for jurisdictions that currently employ at-large election systems.
111. Dr. Oskooii has published 22 peer-reviewed journal articles on topics including Section 2 of the VRA, ecological inference, and Latino politics. His articles frequently contain surveys.
112. Dr. Oskooii has been retained as an expert witness or government consultant in voting matters “about a dozen times.” In addition to this case, Dr. Oskooii has submitted remedial maps for Washington State's legislative redistricting and drew maps for the Roswell Independent School District. A federal court recently adopted his map for the Washington State Legislature, and the Roswell Independent School District also adopted Dr. Oskooii's map, which remains in place today.
See Soto Palmer v. Hobbs, No. 3:22-cv-05035, Doc. #290, slip op. (W.D. Wash. Mar. 15, 2023).
113. Throughout Dr. Oskooii's live testimony, his opinions were clear, consistent, and forthright. He had no difficulty articulating the bases for his mapping decisions. He answered all questions presented to him clearly and respectfully.
114. The Court finds Dr. Oskooii highly credible.
115. In total, Dr. Oskooii submitted 14 illustrative maps.
116. For Maps 1-12, Dr. Oskooii considered total population, population equality, compactness, contiguity, reasonable shape, and communities of interest. He did not, however, consider race for any of these maps.
117. For Maps 13 and 14, Dr. Oskooii considered the above criteria and also considered race. He testified that race did not predominate at any time while drawing the maps.
118. In Map 1, District 1 covers the area south of Trail Street. District 2 covers the Mideast portion of Dodge City, while Center City is conceived as between Central Avenue and 14th Avenue and encompassed by District 3. North Dodge is conceived as above Comanche Street and is divided into two districts-Districts 4 and 5-by 6th Avenue. Demographically, District 1's Hispanic CVAP is 55.61%, District 2's Hispanic CVAP is 60.68%, District 3's Hispanic CVAP is 60.00%, District 4's Hispanic CVAP is 29.65%, and District 5's Hispanic CVAP is 18.12%.
119. Map 2 presents a different conception of District 1, the South Dodge district, by incorporating Wyatt Earp Boulevard as the upper boundary on the west side and Trail Street on the east side. Because District 1 extends farther northwest than in Map 1, District 3 loses some population in the southwest part of the district. To gain population back for population equality purposes, part of District 3 extends north of Comanche Street, taking part of District 5's population. Also, District 4 only extends as far south as West Wyatt Earp Boulevard, in light of District 1 taking the population in the southwest. Demographically, District 1's Hispanic CVAP is 55.45%, District 2's Hispanic CVAP is 59.38%, District 3's Hispanic CVAP is 59.66%, District 4's Hispanic CVAP is 28.98%, and District 5's Hispanic CVAP is 16.80%.
120. In Map 3, District 3 extends farther west of North 14th Avenue, instead of capturing the area north of Comanche Street as it does in Map 2. As a result, District 4 correspondingly captures some areas past North 14th Avenue, stopping at 9th and 6th Avenues for equal population purposes. Districts 1, 2, and 5 remain largely the same as they did in Map 2. Demographically, District 1's Hispanic CVAP is 55.41%, District 2's Hispanic CVAP is 59.43%, District 3's Hispanic CVAP is 61.22%, District 4's Hispanic CVAP is 27.37%, and District 5's Hispanic CVAP is 18.12%.
121. Map 4 presents a different conception of District 3, the Center Dodge District, by encompassing areas right below Wyatt Earp Boulevard and capturing the Hoover Pavilion and Wright Park Zoo. As a result of this change, District 4, the Western Dodge District, does not extend south past Division Street. Also, as a result of the new District 3, District 1 extends all the way north to the Wyatt Earp Boulevard on the east side and only up to the Arkansas River on the west side. District 2, the Eastern Dodge City District, starts at Central Avenue-instead of Avenue A as it does in Map 3-and does not extend farther south beyond Wyatt Earp Boulevard. Demographically, District 1's Hispanic CVAP is 55.59%, District 2's Hispanic CVAP is 60.99%, District 3's Hispanic CVAP is 59.44%, District 4's Hispanic CVAP is 28.88%, and District 5's Hispanic CVAP is 18.12%.
122. Map 5 presents a slightly different conception of District 3, the Center Dodge District, while still encompassing Hoover Pavilion and Wright Park Zoo as it did in Map 4. District 3 in Map 5 attempts to keep more of 6th through 9th Avenues within the Center City district than it did in Map 4. As a result, District 4 extends farther south than it did in Map 4, but remains north of the Wyatt Earp Corridor. Districts 1, 2, and 5 remain largely the same as they did in Map 4. Demographically, District 1's Hispanic CVAP is 55.59%, District 2's Hispanic CVAP is 60.99%, District 3's Hispanic CVAP is 60.88%, District 4's Hispanic CVAP is 26.33%, and District 5's Hispanic CVAP is 18.12%.
123. Map 6 presents a new conception of District 1, the South Dodge District, by incorporating the Arkansas River as its northeastern boundary. As Dr. Oskooii noted, the “area below [the] Arkansas River does not have enough population to be in one district.” Accordingly, the western part of District 1 extends above the Wyatt Earp Boulevard for the first time, in order to gain enough population to achieve population equality. Consequently, in Map 6, District 4 loses some population. In order to gain sufficient population, it captures some area directly above District 3's northern boundary, cutting across 6th Avenue and taking some population from District 5. District 3 reverts back to its almost-perfect-square configuration that it had in Maps 13, while District 2 extends farther south to capture the population District 1 ceded in the southeast. Demographically, District 1's Hispanic CVAP is 55.51%, District 2's Hispanic CVAP is 58.69%, District 3's Hispanic CVAP is 59.55%, District 4's Hispanic CVAP is 26.99%, and District 5's Hispanic CVAP is 16.60%.
124. Map 7 presents the same conception of District 1, the South Dodge District, as Map 6 and incorporates the Arkansas River as its northeastern boundary. Yet rather than have District 4 gain population east of 6th Avenue in the central part of Dodge City, it gains that population east of 6th Avenue in the northern part of the city. This conception of northwest Dodge unites new housing developments with other soon-to-be areas of growth in North Dodge. As a result, District 5 loses population and needs to regain equal population by extending slightly below Comanche Street, eating a bit into some of District 3's population. Demographically, District 1's Hispanic CVAP is 55.51%, District 2's Hispanic CVAP is 58.69%, District 3's Hispanic CVAP is 59.04%, District 4's Hispanic CVAP is 24.92%, and District 5's Hispanic CVAP is 20.92%.
125. Map 8 presents a similar conception of Map 7's District 4 and adopts the version of District 3, the Center City district, that extends farther south to include Hoover Pavilion and the Wright Park Zoo (as existed in Maps 4-5). In order to do this while still abiding by equal population and other redistricting criteria, District 3 needs to extend a bit below the Arkansas River. District 1, in turn, extends farther north on the east side to East Trail Street, to pick up population, which cuts into District 2's southern population. Demographically, District 1's Hispanic CVAP is 56.30%, District 2's Hispanic CVAP is 60.68%, District 3's Hispanic CVAP is 58.01%, District 4's Hispanic CVAP is 26.66%, and District 5's Hispanic CVAP is 21.50%.
126. Map 9 presents a new conception of District 2 by recognizing the Wyatt Earp Corridor as its own community of interest. This conception is also reflected-with some variations-in Maps 10-12. As a result, District 3, the Center Dodge District, moves farther east in a way that grabs some population in Mideast Dodge City. Districts 1, 4, and 5 are largely consistent with prior illustrative maps. Demographically, District 1's Hispanic CVAP is 55.70%, District 2's Hispanic CVAP is 56.69%, District 3's Hispanic CVAP is 65.13%, District 4's Hispanic CVAP is 25.87%, and District 5's Hispanic CVAP is 16.80%.
127. Map 10 presents a slightly different conception of District 3 than was presented in Map 9. In Map 10, District 3 moves west back to North 14th Avenue, in an attempt to recapture some of the Center City Dodge area that it ceded in Map 9. Consequently, more of District 4- which loses population when District 3 moves west in Map 10-proceeds farther east of 6th Avenue than it did in Map 9. That forces District 5 to extend slightly below Comanche Street to regain some population. Districts 1 and 2 are largely similar to their configurations in Map 9. Demographically, District 1's Hispanic CVAP is 55.70%, District 2's Hispanic CVAP is 56.64%, District 3's Hispanic CVAP is 60.99%, District 4's Hispanic CVAP is 27.81%, and District 5's Hispanic CVAP is 16.51%.
128. Map 11 presents a slightly different conception of District 2 than in Maps 9 and 10. In Map 11, District 2 “more fully captures both sides of Wyatt Earp Corridor” by extending to the westernmost part of the city on both sides of the street. The other districts are largely similar to their configurations in either Map 9 or Map 10. Demographically, District 1's Hispanic CVAP is 55.90%, District 2's Hispanic CVAP is 55.15%, District 3's Hispanic CVAP is 66.14%, District 4's Hispanic CVAP is 26.73%, and District 5's Hispanic CVAP is 18.12%.
129. Map 12 presents a similar conception of District 2 as in Map 11, but District 2 here extends farther south than West Wyatt Earp Boulevard to capture more area on the western side of the corridor. Consequently, District 1, which loses population in the west, extends above Trail Street on the eastern side to make up for that population loss. The remaining districts are fairly similar to their configurations in Map 11. Demographically, District 1's Hispanic CVAP is 55.46%, District 2's Hispanic CVAP is 55.29%, District 3's Hispanic CVAP is 66.62%, District 4's Hispanic CVAP is 27.85%, and District 5's Hispanic CVAP is 18.12%.
130. Map 13 looks very similar to the previous maps-it largely reflects the same communities of interest of South Dodge, Mideast, Center Dodge City, and Northwest and Northeast Dodge. Demographically, District 1's Hispanic CVAP is 55.63%, District 2's Hispanic CVAP is 64.76%, District 3's Hispanic CVAP is 55.49%, District 4's Hispanic CVAP is 25.38%, and District 5's Hispanic CVAP is 21.33%.
131. Map 14 likewise looks quite similar to previous maps. District 4 is slightly different in Map 14 than in Map 13 in that it encompasses more of the newer housing developments in the northeast. Consequently, District 5 extends below Comanche Street to gain some population that it lost by virtue of District 4's configuration. Demographically, District 1's Hispanic CVAP is 55.69%, District 2's Hispanic CVAP is 70.49%, District 3's Hispanic CVAP is 52.18%, District 4's Hispanic CVAP is 19.57%, and District 5's Hispanic CVAP is 26.79%.
132. In conclusion, Dr. Oskooii's maps all indicate that the Hispanic CVAP in Dodge City is sufficiently large and geographically compact to constitute a majority of voters in at least one district.
I. Plaintiff's Gingles II and III expert-Dr. Barreto
133. Dr. Matthew A. Barreto is a tenured Professor of Political Science and Chicano Studies at the University of California, Los Angeles. Prior to his appointment at the University of California, Los Angeles, Dr. Barreto was a tenured professor at the University of Washington.
In fact, this case appears to have stemmed in part from a class project conducted by one of Dr. Barreto's students on voting patterns in Dodge City.
134. Dr. Barreto holds a Ph.D. from the University of California, Irvine, a Master of Science from the University of California, Irvine, and a B.A. from Eastern New Mexico University.
135. He is also a cofounder and faculty director of the Latino Policy and Politics Institute, and the cofounder and director of the UCLA Voting Rights Project.
Several members of the UCLA Voting Rights Project represent Plaintiffs as attorneys in this case.
136. Dr. Barreto has published over eighty peer-reviewed articles, primarily regarding race and ethnicity in the political system and voting patterns, ecological inferences, Latino politics, and survey methodology. He has also published four books and has won the 2014 American Political Science Association Best Book Award for Race, Ethnicity, and Politics.
137. Dr. Barreto has extensive experience serving as an expert witness in voting rights cases on topics such as social science methods, mapping, racially polarized voting, demographic changes, racial and ethnic politics, and Gingles preconditions II and III.
138. Because Dodge City does not track voter ethnicity, Dr. Barreto relied on a method called Bayesian Improved Surname Geocoding (“BISG”). Looking at actual voters' surnames (instead of merely the precinct's CVAP), BISG analyzes and reports the probability that the voter is of a certain ethnicity. Although undependable on an individual level, BISG may assist courts in determining whether Hispanic voters in the aggregate favored certain candidates. Several courts have previously held BISG to be a reliable scientific method that provides helpful data on voter ethnicity in Section 2 cases.
139. Using BISG to determine the overall likelihood of voters' ethnicity, Dr. Barreto employed two types of ecological inference (“EI”) models to identify racially polarized voting patterns in Dodge City: (1) King's Iterative Ecological Inference (“King's EI”) and (2) Rows by Columns (“RxC”).
140. According to Dr. Barreto, King's EI was originally developed by Harvard professor Gary King, who later collaborated with others to create RxC. Dr. Barreto stated that King's EI runs each candidate against the entire field iteratively, that is, one at a time. Because of its iterative process, Dr. Barreto testified that King's EI is more accurate in elections where voters have multiple votes or there are multiple candidates.
141. In contrast, RxC provides a single report for all candidates. Dr. Barreto stated that RxC tends to show a more averaged result, a statement borne out by the prepared charts.
142. Dr. Barreto testified that the two systems were “quite similar” and “almost identical.”
143. Dr. Barreto examined endogenous elections in 2014, 2017, 2019, and 2021, endogenous elections being those of the same type as the election at issue before the Court (i.e., City Commission elections). He also examined 20 exogenous elections, which are elections other than the type being challenged under the VRA. Only two of the exogenous elections examined by Dr. Barreto were multi-candidate/multiple votes election localized to Dodge City-the 2019 Dodge City Community College (“DCCC”) Trustee election and the 2021 School Board election. When analyzing Dodge City elections, Dr. Barreto structured his report on point estimates, i.e., the most likely percentage of the voter population which voted for a specific candidate. In his work, he did not include confidence intervals or a potential margin of error.
144. Dr. Barreto submitted a chart, Exhibit 121, purporting to show his analyses for 24 elections under both King's EI and RxC. The table is included below as it was presented to the Court.
Table 1 Dodge City Ecological Inference (EI) Candidate Choice Estimates
(Image Omitted.)
(Image Omitted.)
(Image Omitted.)
145. On the stand, Dr. Barreto testified that the shaded green cells represented the preferred candidates in each election. Dr. Barreto used red shading for all other cells. Furthermore, Dr. Barreto designated the winners of most elections with an asterisk.
146. The Court finds that Dr. Barreto's experience and expertise in the field of voting rights is unquestionable. Carefully observing his demeanor, the Court concludes that Dr. Barreto presented as likeable, straightforward, nonargumentative, and a master of both his craft and of being an expert witness. His answers were articulate, responsive to counsel's questions, and comprehendible. The Court considers Dr. Barreto a credible witness whose testimony and analyses are entitled to, for the most part, substantial weight.
147. Although Dr. Barreto presents very credibly on the whole, there are several issues which call aspects of his testimony into question. First, when it comes to identifying Latinopreferred candidates, the Court concludes Dr. Barreto's opinion is only somewhat helpful.
148. To begin with, Dr. Barreto's criteria for which candidates were “Latino-preferred” appears to be so thin as to be almost nonexistent. For example, Dr. Barreto states he identifies the White and Latino preferred candidates in Exhibit 121 with green shading. Even a cursory overview of the tables reveals that the green shading represents merely the highest percentage of the voter share.
149. For example, in the RxC analysis for the 2021 School Board election, Dr. Barreto applies green shading to Killion, who received 10.2% of the Latino vote, and red shading to Zortman, who received 10.1%. It belies credulity to conclude that Killion was a Latino-preferred candidate and Zortman was not based on 0.1% difference in the number of Latino votes they received.
150. Although the shading is helpful to quickly identify the top vote recipients, the Court will not rely heavily nor exclusively on Dr. Barreto's opinion-shown through green shading-to determine which candidates were Latino-preferred.
151. Dr. Barreto also appears to apply his own criteria inconsistently, without disclosing the facts or data underlying this irregularity. For example, Dr. Barreto testified that the 2019 City Commission election did not have any “obvious Latino candidates of choice.” But his King's EI analysis for that election shows three candidates-shaded green-who received 32.1%, 25.9%, and 18.2% of the Latino vote. The remaining candidates averaged 12.1% of the Latino vote. Yet, in the 2021 election, Scoggins received 14.2% of the Latino vote compared to an average of 8.5% for non-preferred candidates under an RxC analysis. This result left no doubt in, however, Dr. Barreto's mind that Scoggins was a Latino-preferred candidate.
152. To his credit, Dr. Barreto clarified, “I wouldn't say there was no racially polarized voting [in 2019]. I would say there was racially polarized voting in this election, but that there was no obvious in this particular election Latino candidate of choice ” But Dr. Barreto did not explain why the election in 2019 under King's EI was less probative of Latino preferences than elections with similar or less dramatic results. Such unexplained discrepancies in Dr. Barreto's opinion lead the Court to give only some weight to his opinion regarding which candidates were Latino-preferred.
153. Second, Dr. Barreto resorted to hyperbole on the stand, which undercuts his reliability as an objective expert in this case. Specifically, Dr. Barreto testified on the stand that in Dodge City “Hispanic preferred candidates have never won.” Such exaggeration-which is directly contrary to the evidence in this case-does not a convincing witness make. Still, this appears to be an isolated incident, and the Court does not discredit Dr. Barreto because of it.
154. Finally, the Court is aware of Dr. Barreto's consistent role with this case, from its inception through trial. His involvement-bordering on entanglement-goes far beyond the normative role of an unbiased expert. Regardless of whether a project assigned in Dr. Barreto's class influenced the inception of this lawsuit, Dr. Barreto is the cofounder and continuing faculty director of the UCLA Voting Rights Project, which has vigorously advocated for Plaintiffs throughout this litigation. Furthermore, Dr. Barreto was the only witness to remain throughout the entire week-long trial, demonstrating a personal interest beyond that normally held by an expert witness.
155. The Court does not consider Dr. Barreto's associations or role in this case a mark against his skill, knowledge, or expertise in his field. Neither does it render his testimony wholly incredible as Defendant urges. However, his role as a potential advocate for Plaintiffs calls into question his objectivity as a witness. Thus, where Dr. Barreto's opinions rely on less than hard data, the Court will consider explanations for his opinions based on his potential personal bias. For example, Dr. Barreto's preference for King's EI over RxC-based solely on his own opinion - could be explained by bias, especially considering that King's EI is dramatically more indicative of racially polarized voting and thus more helpful to Plaintiffs' cause. Therefore, the Court affords little weight to Dr. Barreto's opinion that King's EI is more accurate than RxC for multi-candidate/multiple vote elections.
J. Defendant's Gingles II and III expert-Dr. Katz
156. Defendant relies on the opinion of its own expert, Dr. Jonathan Katz, who analyzed the same four endogenous elections as Dr. Barreto. Unlike Dr. Barreto, Dr. Katz solely conducted an RxC analysis.
157. Dr. Katz testified that RxC was an EI analysis specifically adapted for multiple groups and multiple candidates. For that reason, he opined that King's EI was less effective than RxC when dealing with multiple candidates or multiple groups. Accordingly, Dr. Katz concluded King's EI analysis was unhelpful for Dodge City's City Commission elections.
158. Similarly, Dr. Katz opined that “the typical standard in social sciences is to use 95 percent confidence intervals.” He further stated that failure to include a 95% confidence interval meant that no conclusions could be drawn whatsoever.
159. In analyzing the four endogenous elections under an RxC model, Dr. Katz included confidence intervals in his own work. The table below-admitted within Exhibit 456-is one example:
Table 3: EI Results for November 2021 for Dodge City Commission
Latinos
Whites
Other
Burns
12.6
19.3
12.3
(3.8, 24.7)
(15.7, 22.5)
(2.4, 27.5)
Nuci
13.6
16.5
13.8
(4.4, 25.6)
(12.8, 19.6)
(3.1, 30.5)
Reinert
10.7
15.2
13.3
(2.5, 21.2)
(11.7, 18.1)
(2.9. 30.2)
Rhoten
5.8
3.9
11.6
(1.3, 12.7)
(1.7, 5.7)
(1.9, 26.1)
Salinas
10.9
9.3
12.1
(2.9, 20.6)
(6.0, 12.0)
(2.3, 27.6)
Scoggins
14.1
10.0
12.0
(5.1, 24.4)
(6.9, 12.8)
(1.9, 28.0)
Soto
14.9
10.1
11.5
(5.9, 26.2)
(6.5, 13.2)|
(1.6, 28.1)
Taylor
17.4
15.6
13.3
(6.9. 30.1)
(11.6, 19.1)
(2.8, 29.1)
160. Dr. Katz explained that the number in line with the candidate's name represents the point estimate as to the percentage of votes received from Latinos, Whites, or Other. The two numbers within the parentheses represent the confidence intervals. The data for the 2021 election supports finding with 95% confidence that Burns, for example, might have received anywhere from as little as 3.8% to as much as 24.7% of the Latino vote. Where confidence intervals overlap, statistical confidence drops. Dr. Katz opined that the overlap between the confidence intervals for each candidate meant that no candidate could be labeled Latino-preferred with any statistical certainty. Dr. Katz viewed the overlap between Nuci and Taylor as “a statistical tie.”
161. Some conclusions are possible. For example, Dr. Katz stated “[w]know with a high degree of statistical confidence that Rhoten was the least preferred candidate.” But because “the rest of them all overlap . . . it is a statistical dead heat.” Therefore, Dr. Katz stated it was impossible to know if any candidates were Latino-preferred.
162. Dr. Katz's analyses of the other elections followed a similar pattern: wide confidence intervals, occasionally greater than 30 points, and high degrees of overlap. One more example is helpful. In the table below, Dr. Katz analyzed the results of the 2017 City Commission election.
Table 4: EI Results for November 2017 for Dodge City Commission
Latinos
Whites
Other
Delzeit
16.1
26.0
17.4
(4.8, 31.7)
(21.8, 29.7)
(3.5, 36.3)
Gwaltney
15.0
18.5
16.1
(3.6, 29.3)
(14.6, 21.8)
(3.1, 35.9)
Seilens
16.5
9.4
16.2
(5.0. 30.7)
(5.9, 12.5)
(3.0, 35.4)
Smoll
16.9
18.2
16.4
(4.5, 33.2)
(14.1, 21.8)
(3.6, 33.7)
Warshaw
21.4
21.4
17.4
(6.3, 38.5)
(17.0, 25.6)
(3.2, 36.3)
Zuniga
14.0
6.4
16.4
(3.9, 27.8)
(3.0, 9.2)
(3.4, 36.4)
163. Once again, Dr. Katz stated that no conclusions were possible based on the high degree of uncertainty caused by significant overlap in the confidence intervals for each candidate. Interestingly, Dr. Katz originally concluded that “Zuniga was the least preferred” before retracting that statement because “they all overlap.”
164. For several reasons, the Court finds Dr. Katz's testimony should be entitled to some but little weight. First, the Court does not at all credit Dr. Katz's opinion that the lack of 95% confidence intervals renders Dr. Barreto's results completely useless. This is because the Court, unlike Dr. Katz, is not looking for statistical certainty. Rather, as Plaintiffs properly point out, the burden here is merely preponderance of the evidence, i.e., 51% certainty-not 95% certainty. Given that Dr. Katz and Dr. Barreto appear to agree that the point estimates represent the most likely scenario, the Court considers them relevant to establishing Gingles II and III. And yet, the Court credits Dr. Katz's testimony that the point estimate is exactly what it claims to be-an estimate that does not necessarily reflect reality.
165. For the same reasons, the Court does not credit Dr. Katz's opinion that no salient conclusions are possible due to the high degree of overlap in confidence intervals between each candidate. Furthermore, during cross-examination of Dr. Katz, Plaintiffs referenced an article by Dr. Katz in which he specifically stated that “high levels of overlap” do not upset the conclusion that point estimates represent the most likely scenario. Because Dr. Katz previously stated-in a peer-reviewed article, no less-that conclusions are still possible despite greatly overlapping confidence intervals, his contrary statement before the Court carries no weight whatsoever.
166. Third, Dr. Katz committed several errors in both his report and his deposition which Plaintiffs highlighted. Plaintiffs also cite multiple cases in which Dr. Katz's testimony was found to be uncredible or illogical. While the Court does not condemn innocent mistakes, multiple errors and the sheer number of times Dr. Katz has been found to lack credibility in court weigh against finding that Dr. Katz provides a persuasive statement in this case.
Vol IV, 176:7 (“My bad for not catching that”); 176:17 (“I forgot to cite it in the citation”); 179:9 (“I was mistaken, yes.”); 184:22 (stating when directed to opine on his own conclusion from a peer-reviewed journal “I might be wrong. I wouldn't bet large sums of money on it”).
See, e.g., Bethune-Hill v. Va. State Bd. of Elections, 326 F.Supp.3d 128, 150 (E.D. Va. 2018); Luna v. Cnty. of Kern, 291 F.Supp.3d 1088, 1125 (E.D. Cal. 2018) (finding Dr. Katz's testimony “undercut” by his contrary positions in other cases).
During his testimony, Dr. Katz stated that he seldom reads judicial opinions, even for cases in which he served as an expert witness. It is not surprising then that Dr. Katz has failed to address many of the issues which courts consistently raise with his work.
167. Finally, the Court carefully observed Dr. Katz's demeanor. On cross-examination, Dr. Katz appeared argumentative and evasive, often interrupting Plaintiffs' counsel, needlessly contradicting minor details, and seldom giving straight forward to difficult questions. His conduct was not conducive to finding that he acted as a neutral, reliable authority on the subject matter.
The interruptions may have been exasperated by the fact that Dr. Katz appeared remotely via Zoom, but it does not change the Court's overall assessment of his demeanor while testifying.
a. Endogenous elections
168. Before delving into the statistical data, the Court must wrestle with whether to rely on King's EI or RxC, a dilemma necessitated by the often-extreme differences between the two analyses. The Court first considers Dr. Katz's testimony that RxC was specifically developed for the type of elections at issue here. His testimony-uncontradicted on this issue-merits some weight. In contrast, Dr. Barreto “believes” King's EI is more accurate. Although the Court credits Dr. Barreto on the whole, his opinion on this issue lacks support and appears to stem from his own ipse dixit. Especially considering his intimate role in Plaintiffs' case, the Court needs more than his personal beliefs to credit his testimony on a given issue.
Because the differences between Dr. Barreto's and Dr. Katz's RxC analyses on the endogenous elections are slight, the Court will not distinguish between the two. As a practical matter, it will reference the figures from Dr. Barreto's RxC analysis throughout this Order.
169. Furthermore, in the 2014 City Commission election, Dr. Barreto's use of King's EI appears to result in inconsistencies with the actual outcome. There, Peters received the highest percentage of white votes with 19.3%, beating election winners Sowers and Smoll who had 19.1% and 17.8%, respectively. Peters also received a far higher share of the Latino vote than either Sowers or Smoll-16.3% compared to 5.1% and 10.6%, respectively. Therefore, King's EI effectively states that Peters should have been elected over both Sowers and Smoll. The record clearly indicates he did not.
170. A possible explanation for this discrepancy might be that the relatively scant number of Black and Asian voters in Dodge City favored Sowers and Smoll over Peters, but that is sheer speculation. Dr. Barreto did not address any possible explanations. Nor does his King's EI analysis appear to take into consideration other ethnic groups within Dodge City. The Court will avoid speculating and accept Plaintiffs' evidence as offered. As it stands, the 2014 election disfavors finding that King's EI more accurately examines multi-candidate/multiple vote elections than RxC.
171. For these reasons, the Court finds that RxC is the more reliable method for evaluating the endogenous elections and other multi-candidate/multiple vote elections and that its results are more persuasive. Nevertheless, the Court will consider both King's EI and RxC together in making its factual findings. But where King's EI and RxC conflict, the Court will make its findings according to the RxC analysis.
172. There is yet another difficulty presented by the type of endogenous elections at issue here-multi-candidate, multiple vote elections. There was not a single election where one candidate received more than 50% of the Latino vote. In fact, only three candidates ever received higher than 30% of the Latino vote-and that was only under King's EI. Because voters in these elections were able to cast up to three votes, it follows that the percentage of votes received by each candidate would be lower. Faced with this unique situation, the Court looks to the discrepancy between the highest and lowest recipients of Latino votes for indications of Latinopreferred candidates rather than simply whether any candidate received a majority of votes.
i. 2014 City Commission election
173. Beginning in 2014, the endogenous elections analyzed by Dr. Barreto provide some evidence of shared Latino preferences in voting. Under RxC, Zuniga and Scoggins received 20.0% of the Latino vote, and Peters received 13.8%. The other candidates received 12.9%, 12.1%, 11.6%, and 9.6%. Similarly, under King's EI, the top three vote-getters for Latinos were Zuniga, Scoggins, and Peters with 33.3%, 31.6%, and 16.3%. The rest of the field averaged 7.9%.
174. Because the RxC analysis shows Zuniga and Scoggins receiving roughly twice as many votes as the other candidates, the Court considers these numbers highly probative that Zuniga and Scoggins were the Latino-preferred candidates for this election. To the extent King's EI is considered, it also supports this finding. In contrast, the less than 1% degree of separation under RxC between Peters and the next highest recipient of the Latino vote does not support finding that Peters was Latino-preferred.
175. Scoggins was elected.
ii. 2017 City Commission election
176. The RxC analysis shows the highest recipients of Latino votes were Warshaw (19.2%), Zuniga, (18.2%), and Delzeit (17.6%). However, two other candidates scored remarkably close, namely Sellens (17.0%) and Smoll (16.9%). The only other candidate received 11.2% of the Latino vote. The King's EI for 2017 shows Sellens, Warshaw, and Zuniga received 25.8%, 22.3%, and 17.5% of the Latino vote, respectively. Zuniga did not campaign that year, although her name remained on the ballot. The next highest recipient of Latino votes was Smoll with 16%. The other two candidates received 8.1% and 9.4% of the Latino vote.
177. The evidence here somewhat favors finding that significant numbers of the Latino community preferred Warshaw as a candidate because he scored the highest in RxC and was the second highest vote recipient in King's EI. However, the rest of the point estimates in RxC are highly similar and reflect an evenly distributed Latino vote. Therefore, the Court does not find that any of the other candidates were Latino-preferred.
178. Warshaw was elected.
iii. 2019 City Commission election
179. For the 2019 election under RxC, Hessman (23.5%), Smoll (21.1%), and Sowers (20.6%) took the top three spots for Latino votes. The other two candidates-Nuci and Reeves- received 18.7% and 16.0%. King's EI tells a similar story, with Hessman, Smoll, and Sowers receiving 32.1%, 25.9%, and 18.2% of the Latino vote, compared to an average of 12.1% for the other candidates. Taken together, these statistics support finding that-to the extent a significant number of Latinos preferred any candidates in 2019-Hessman, Smoll, and Sowers were Latinopreferred.
180. Smoll and Sowers were elected.
iv. 2021 City Commission election
181. Finally, in the 2021 election, the RxC reveals the top three recipients of Latino votes as Taylor (17.1%), Soto (15.1%), then Scoggins (14.2%). The rest received on average 10.7%. However, an even cursory glance at the data reveals Rhoten significantly brought down that average with just 6.1%, the only candidate to score a single-digit percentile. King's EI keeps the same three candidates but flips the order, with Scoggins, Soto, and Taylor receiving 29.2%, 20.6%, and 18.1% of the Latino vote. The other five candidates averaged 6.2%. On the higher end of (according to Dr. Barreto) the “nonpreferred candidates” is Burns with 13.6% of the Latino vote- merely 0.6% less than Scoggins. Once again, it belies credulity to hold that Scoggins was a Latinopreferred candidate and Burns was not based on a 0.6% difference in the number of Latino votes they received.
182. Given the massive conflict between RxC and King's EI when it comes to Scoggins (29.2% compared to 14.2%), the Court will rely on RxC instead of King's EI. And Dr. Barreto's RxC analysis does not support finding that Scoggins was Latino-preferred over Burns or the other candidates. However, because RxC and King's EI consistently show Taylor and Soto receiving a significant share of Latino votes, Dr. Barreto's analysis supports finding they were Latinopreferred.
183. Taylor was elected.
v. Summary of Latino-preferred candidates
184. In summary, the Court finds that the statistical date indicates:
• In the 2014 election, significant numbers of the Latino community in Dodge City demonstrated a preference for Zuniga and Scoggins.
• In the 2017 election, significant numbers of the Latino community in Dodge City demonstrated a preference for Warshaw.
• In the 2019 election, significant numbers of the Latino community in Dodge City demonstrated a preference for Hessman, Smoll, and Sowers.
• In the 2021 election, significant numbers of the Latino community in Dodge City demonstrated a preference for Soto and Taylor.
b. Exogenous elections
185. Dr. Barreto testified that “[f]our elections is just far too few for us to draw a conclusion,” a sentiment echoed by Dr. Katz. The problem is that both Dr. Barreto and Dr. Katz analyzed only four endogenous elections. The rest of the elections analyzed by Dr. Barreto in Exhibit 121 are exogenous. Of those, only two-the 2021 Dodge City School Board election and the 2019 DCCC Trustee election-bear any resemblance to City Commission elections. Like the City Commission elections, voters in these two elections were able to cast up to three votes in a local, multi-candidate, nonpartisan election taking place in an odd-numbered year. Because of their similarity to the endogenous elections, the Court considers them highly relevant to the Gingles analysis.
186. The 2021 School Board election under RxC shows a more or less evenly distributed vote, with the four highest recipients being 13.1%, 11.2%, 10.5%, and 10.2%, compared to an average of 9.2% for the remaining candidates. As usual, King's EI is more polarized, with the top four vote-getters for Latinos receiving 19.8%, 17.8%, 11.2%, and 9.4% of the Latino vote. On average, the other six candidates received 5.4% of the Latino vote.
Dr. Barreto does not explain why he designated the top four vote recipients with green shaded cells instead of the top three like he did with every other multi-candidate/multi-vote election.
Because there is no evidence before the Court of who won the 2021 School Board election, the identities of the Latino-preferred candidates are irrelevant under Gingles III. Therefore, the Court need not determine them here.
187. In the 2019 DCCC election, RxC shows Garcia (18.2%), Henrichs (15.3%), and Lewis (14.8%) receiving the most votes. The other candidates averaged 12.9%. Under King's EI analysis, the top three recipients of Latino votes were Garcia, Turley, and Henrichs with 29.5%, 24.2%, and 21.2% of the Latino vote share. Of the remaining candidates, Lewis received 15.6% compared to the other three candidates receiving 8.3%, 5.1%, and 4.2%.
188. Garcia was easily the top recipient of Latino votes in both RxC and King's EI. Therefore, the Court considers him to be Latino-preferred candidates in this election. Likewise, Henrichs stands out from the crowd in both King's EI and RxC, albeit not to the same extent as Garcia. Accordingly, the Court also considers him a Latino-preferred candidate for this election.
189. Lewis is a closer call. Although his vote share under RxC is close enough to Henrichs' to be practically indistinguishable, he was not within the top three under King's EI. Nevertheless, he still scored substantially higher than the other three nonpreferred candidates. Taken together, the election data indicates that Lewis was a Latino-preferred candidate. The Court concludes, therefore, that the 2019 DCCC Trustee election appears to support finding that significant numbers of Latinos choose to vote for the same candidates.
190. The rest of the exogenous elections are far less probative because they deal with partisan elections with only two candidates where voters could cast only one vote. Nevertheless, they comprise the most consistent evidence of racially polarized voting. In every election analyzed by Dr. Barreto under by King's EI and RxC, the majority of Latinos voted for the same candidate.
c. Lay evidence of Hispanic voting preferences
191. Along with the statistical analysis provided by the experts, courts may look to lay witness testimony as further evidence of racially polarized voting. Monica Vargas testified that the meatpackers' union endorsed Blanca Soto in the 2019 City Commission election and Angie Gonzalez in the 2020 County Clerk election. This is potentially relevant because Vargas testified that 75% of union members are Latino. However, when asked about the level to which Dodge City meatpacking workers were engaged in local politics, Vargas responded, “Very little.” Given the lack of political involvement by union members, the Court does not find Vargas' testimony that the union “endorsed” certain candidates probative that Latinos in general endorsed Soto or Gonzalez. Nevertheless, Vargas's testimony that Soto and Gonzalez showed commitment to working with the Latino communities supports finding that they were Latino-preferred candidates.
See, e.g., Sanchez, 97 F.3d at 1320 (“Second, while lay testimony is relevant to determine who is the candidate of choice, it is not alone dispositive.... However, absent the finding the statistical evidence is unreliable, insufficient, or irrelevant, lay testimony should not eclipse the [statistical] analysis.”). Indeed, the Tenth Circuit instructs courts to review lay evidence when analyzing whether white candidates are nevertheless minority preferred. See id. at 1321 (listing factors).
192. Plaintiff Rangel-Lopez testified that he voted for Scoggins and Soto, without specifying which elections he participated in. He stated that both Scoggins and Soto listened to the Hispanic community. Given his age, the Court infers that Rangel-Lopez refers to the 2021 election where both Scoggins and Soto were not elected. Rangel-Lopez stated that he has never had a City Commissioner knock on his door to ask his opinion on issues. However, he also stated “I can't speak for any other voter than myself.” Therefore, Rangel-Lopez's testimony is not indicative of racially polarized voting in the Hispanic community.
193. Plaintiff Miguel Coca echoed Rangel-Lopez's opinion of Scoggins and Soto, saying that they “have the interests of the Hispanic population at heart.” He further testified that he has seen Soto visit South Dodge.
194. De La Rosa testified that Soto was generally unsupported by the Latino community when it came to assistance during her campaign.
195. Soto lives in North Dodge.
196. Jan Scoggins similarly lives in North Dodge. She also testified that while campaigning, she went door to door and spoke to civic, social, and church groups. She stated that the people in those groups were all Caucasian. Similarly, for her campaign in 2021, Scoggins went door to door in both North and South Dodge as well as Hispanic businesses. Scoggins emphasized that she funded her own campaign.
197. Chuck Taylor lives south of Comanche in a neighborhood identified by Dr. Oskooii as predominantly Hispanic. Based on other evidence that candidates in Dodge City must necessarily engage with their neighbors and campaign door to door, this supports finding that Taylor engages with the Hispanic community.
198. Joseph Nuci, who won election to the City Commission in 2021, is Hispanic. However, the Court heard no other evidence that Nuci was a Hispanic-preferred candidate. In fact, Coca testified that he has never seen Nuci involved in the Hispanic community or at community events. Because the statistical data in the King's EI analysis does not show a high level of Latino support for Nuci in either of his elections, the Court does not consider him to be a Latino-preferred candidate despite his ethnic background.
199. There is no evidence of minority communities sponsoring any candidate. This makes sense, given the testimony that campaigning in Dodge City is very inexpensive due to its small size.
200. Likewise, the Court heard multiple witnesses testify that whites and Latinos have not fought over any issues that were divisive between whites and Latinos in Dodge City. In other words, the Court has no evidence that any candidate has given attention to issues concerning minorities, beyond Rangel-Lopez's vague statement that Soto and Scoggins listen to the Hispanic community.
201. The Court has no evidence of differing minority turnout for each election.
202. Reviewing the lay testimony, the Court concludes it supports finding that Blanca Soto and Chuck Taylor were Latino-preferred candidates in 2021. Likewise, Scoggins' own testimony of campaigning door to door in Hispanic neighborhoods and businesses-along with Rangel-Lopez's testimony that Scoggins listens to the Latino community-supports finding that she is a Latino-preferred candidate, even though Vargas did not identify Scoggins as a candidate supported by the meatpackers' union.
203. Clearly, the most probative evidence of a cohesively voting Latino community in Dodge City comes from the statistical data of the endogenous elections and the two similar exogenous elections. For the most part, that data comports with the lay witnesses' testimony with one exception-Scoggins in 2021. Simply put, the RxC analysis does not show her garnering enough votes to qualify as Latino-preferred for that election. However, the lay testimony shows that she cared for issues affecting the Latino community and campaigned door to door with Hispanic neighborhoods. Under Sanchez, these factors support finding that she is a Latinopreferred candidate. However, they do not override the statistical data, which suggests that despite her best efforts, Latinos did not vote for her over other candidates in significant numbers. The matter is simply too close to call, suggesting it is as likely as not that Scoggins was Latino-preferred in 2021. Therefore, Plaintiffs have failed to meet their burden to show by a preponderance of the evidence that Scoggins was a Latino-preferred candidates in 2021.
See Abbott v. Perez, 585 U.S. 579, 619 (2018) (“Courts cannot find § 2 . . . violations on the basis of uncertainty.”).
204. Ultimately, the Court finds Plaintiffs have established by a preponderance of the evidence that:
• In the 2014 City Commission Election, the Latino-preferred candidates were Scoggins and Zuniga.
• In the 2017 City Commission Election, the Latino-preferred candidate was Warshaw.
• In the 2019 City Commission Election, the Latino-preferred candidates were Sowers, Hessman, and Lewis.
• In the 2021 City Commission election, the Latino-preferred candidates were Soto and Taylor.
• In the 2019 DCCC Trustees election, the Latino-preferred candidates were Garcia, Henrichs, and Lewis.
The Court makes no finding as to the preferred candidates for the 2021 School Board because, as discussed further below, Dr. Barreto did not indicate the winners of that election-therefore, it is irrelevant to a Gingles III analysis.
205. In total, the four endogenous elections support finding that Latinos in Dodge City vote in significant numbers for the same candidates. Likewise, the exogenous elections support the same conclusion.
K. Other evidence presented at trial
1. Plaintiffs' remaining experts
206. Along with Dr. Oskooii and Dr. Barreto, Plaintiffs called Dr. Christina Bejarano and Dr. Ruben Martinez to offer testimony relevant to a totality of the circumstances analysis under the second part of a Gingles analysis.
a. Dr. Bejarano
207. Dr. Bejarano is a Professor of Political Science with tenure at Texas Women's University and specializes in American electoral politics, specifically the political representation and political behavior of Latinos and women. She has published both peer-reviewed articles and books on these topics.
208. During Dr. Bejarano's testimony, the Court observed her demeanor and concludes that she presented as an unreliable witness. Upon being cross-examined by Defendant's counsel, Dr. Bejarano was argumentative and evasive. Defense counsel consistently asked her simple yes or no questions, which Dr. Bejarano obstinately refused to answer on the pretense that they were not clear.
209. This issue was so pervasive and disruptive that the Court gently informed Dr. Bejarano that she was “unnecessarily quibbling with the attorney” and asked “that you not do that.” Although she reformed her answers slightly, she remained evasive and unwilling to engage with questions as asked. On the whole, her conduct did not suggest that she acted as a neutral, unbiased, and reliable expert, and thus the Court gives no weight to her opinions in this case.
210. Dr. Bejarano first purported to testify about the extent to which Latinos have been underrepresented in elected positions in Dodge City.
211. In reviewing the number of Latinos who have held elected office in Dodge City, Dr. Bejarano relied on a directory provided by the National Association of Latino Elected Officials.
212. As reviewed by Dr. Bejarano, the NALEO directory indicated that no Latino had ever been elected to the City Commission.
213. However, Dr. Bejarano conducted her own research which revealed a Latino had served as a City Commissioner between 1999 and 2001. Furthermore, she discovered that Soto had been appointed to the City Commission in 2021.
214. Dr. Bejarano was aware of Nuci and that he identifies as Latino. However, he was not listed in NALEO and she did not include him as an elected Latino official.
215. Dr. Bejarano concluded that in all elected positions within Dodge City and Ford County, Latinos were underrepresented.
216. Next, Dr. Bejarano opined that at-large off-cycle elections negatively impacted Latino representation.
217. Dr. Bejarano also claimed that Dodge City elected officials were not responsive to Latinos' particularized concerns. In reaching this conclusion, she first relied on the fact that the City Commission had not changed to a district-based election system even after Latino citizens had expressed dissatisfaction with the at-large method of voting.
218. The Court finds Dr. Bejarano's reasoning here wholly unpersuasive and confusing. To begin with, there is zero evidence that Latinos ever directly asked the City Commission to consider district-based elections. Dr. Bejarano cannot seriously opine that the City Commission's continued use of the at-large election system demonstrates unresponsiveness to Latinos' concerns when Latinos never voiced those concerns to the City Commission.
219. Dr. Bejarano also testified that voter access was an area in which Dodge City officials showed a lack of responsiveness to Latinos. She testified that the single polling location in Dodge City and the moving of that location in 2018 both showed unresponsiveness by city officials to Latino concerns.
220. Dr. Bejarano, however, did not address the fact that Ford County was the entity who decided to move the polling location in 2018. She similarly did not acknowledge that Dodge City provides door-to-door transportation to its polling location, posts all voting information in Spanish and English, allows mail-in voting, or the many other services Dodge City provides that cater to its Hispanic population.
221. Next, Dr. Bejarano characterized then-President Donald Trump's decision during the COVID-19 crisis to keep meatpacking plants open as an example of Dodge City's unresponsiveness to Latinos. The Court recognizes that Latinos comprise the majority of meatpackers at the plants in Dodge City. But Dr. Bejarano's attempt to connect then-President Trump's decision to keep meatpacking plants open during the COVID pandemic with Dodge City's “unresponsiveness” rings hollow at best and ridiculous at worst.
See Exec. Order No. 13917 (admitted as Ex. 499).
Although Dr. Bejarano also testified that there were safety concerns voiced by Latinos working at the meatpacking plants, she admitted that she “did not go through all the safety protocols in terms of what was being done.” In other words, she merely repeated complaints she heard from undisclosed sources. This is not the testimony of an expert applying reliable methodology to testify as to a matter within her expertise, and the Court will not credit it.
222. Dr. Bejarano then addressed a shooting that occurred in 2010 in which a Hispanic man was killed by two white men. Despite the City's creation of the Cultural Advisory Board in response to complaints of racial unrest, Dr. Bejarano opined that the City was unresponsive to Latinos' concerns.
223. As a part of her responsiveness testimony, Dr. Bejarano testified as to the data for crime arrests and citations in Dodge City based on race. A chart showing this data was admitted as Exhibit 61.
224. Between September 2017 and January 2020, 57.1% on average of all police stops in Dodge City involved Hispanic individuals. Similarly, 64.2% on average of all citations were issued to Hispanics, with 48.5% on average of all warnings issued to Hispanics. The average percentage of Hispanics versus other ethnicities who were arrested during that time period was 66.2%.
225. Dr. Bejarano interpreted these figures as showing a racial profiling against Latinos in Dodge City. She arrived at this conclusion by comparing the voting age population of Latinos with the percentages of stops involving Latinos. Dr. Bejarano did not consider the total Latino population, nor did she consider that the Latino population is overall significantly younger that the white population in Dodge City.
226. The Court does not credit Dr. Bejarano's opinion that these figures show any racial profiling or unresponsiveness. In 2020, for example, Hispanics constituted 63.91% of the documented total population in Dodge City. Therefore, the percentage of stops involving Latinos is actually less than the percentage of Latinos living in Dodge City. Furthermore, the percentages of citations and arrests is fairly similarly to the Latino population percentage. While the percentage of warnings is significantly less for Latinos, the Court has no reason to assume that this is due to racial animus instead of some other reason, such as the Latino population being significantly younger.
The Court is also aware that undocumented immigrants form a significant portion of Dodge City's population.
227. As the final consideration under this factor, Dr. Bejarano testified that Dodge City's cooperation with U.S. Immigration and Customs Enforcement to identify and deport illegal immigrants evidenced a lack of responsiveness to the Latino community. In essence, Dr. Bejarano would have this Court conclude that Defendant's efforts to help enforce federal law are unlawful. Given the Court's constitutionally defined role in enforcing federal law, Dr. Bejarano could not have picked an example more ill-suited to her target audience.
During the trial, the Court expressed its observation that Dr. Bejarano appears to believe that anything law enforcement does is suspect. That finding stands here and cuts sharply against Dr. Bejarano's credibility.
228. Moreover, Dr. Bejarano purported to address the effects of discrimination faced by Latinos in Dodge City. She listed off various statistics regarding differences in pay, education levels, unemployment, food stamp eligibility, housing, and health insurance between Hispanics and whites. Each category demonstrated significant improvement between 2010 and 2021.
229. The Courts finds that there are general economic disparities in Dodge City between Hispanics and whites. However, there is no credible evidence before the Court that these disparities stem from discrimination.
230. Furthermore, Dr. Bejarano admitted that Dodge City's Latino population is heavily comprised of immigrants and first-generation citizens. She also admitted that new migrants in general lack socioeconomic resources. It naturally follows that new immigrants would not have the same education or income as persons growing up in Dodge City. Indeed, the significant improvements in Latinos' education level and income between 2010 and 2021 demonstrate that these immigrants and their descendants have been able to fully take advantage of the benefits offered to all residents. The same is true of the other categories referenced by Dr. Bejarano.
231. In short, Dr. Bejarano's opinion that these disparities show the effects of discrimination is without support and entitled to no weight.
b. Dr. Martinez
232. Plaintiffs' final expert was Dr. Ruben Martinez. Plaintiffs retained Dr. Martinez to analyze the history of official discrimination in Dodge City and the historical background of the at-large method of election in Dodge City.
233. Dr. Martinez has a Bachelor of Science from the University of Southern Colorado, a Master's in Sociology from Arizona State University, and a Doctorate in Sociology from the University of California Riverside. Now retired from teaching, Dr. Martinez has been a professor of sociology at multiple universities, a sociology department chair, and a director of multiple sociological research institutes.
234. In preparing his report for this case, Dr. Martinez reviewed scholarly articles dealing with race relations in Kansas at large and oral histories by Luiz Sanchez, the first Latino Commissioner on the City Commission and Dodge City's first Latino mayor, and an individual named Fred Rodriguez.
235. Dr. Martinez also visited Dodge City to inspect records at the local diocese and the city clerk's office. However, the diocese was not open to visitors as it was holy week and the city clerk's office was in the process of moving and could not provide any records.
236. Dr. Martinez testified that Sanchez grew up in the Mexican Village-a railroad workers company town within the borders of Dodge City-and was denied access to the public swimming pool during his youth. Sanchez was ultimately elected to the City Commission in 1982.
237. Dr. Martinez also testified that Latinos were subjected to segregated theaters at some point in the City's history. He did not specify when.
238. Finally, Dr. Martinez opined that before 1964, Dodge City had restrictive covenants prohibiting the sale of houses within certain neighborhoods to Hispanics. Dr. Martinez did not review those covenants, nor does he know when they were eliminated.
239. On cross-examination, Dr. Martinez admitted that he had not found evidence of official discrimination in Dodge City's housing, education, or segregation in public areas or public schools. Furthermore, Dr. Martinez found no evidence that Dodge City had ever been a “sundown town” even though he testified about the effect of sundown towns in other areas. Lastly, Dr. Martinez uncovered no evidence that Dodge City had ever discriminated against Latinos by policy.
A “sundown town” refers to a town known for directing minorities, usually African Americans, to leave town by sundown. See Thomas v. Sch. Bd. St. Martin Par., 2023 WL 4926681, at *9 n.60 (W.D. La. July 31, 2023) (defining “sundown town”).
240. Instead of testifying about Dodge City, Dr. Martinez elaborated on segregation concerning African Americans in other parts of Kansas, such as Hays or Kansas City. He even referenced Brown v. Board of Education, the seminal Supreme Court dealing with a segregated school in Topeka, Kansas.
241. Dr. Martinez did mention that a chapter of the American G.I. Forum, a Latino civil rights organization, was established in Dodge City in 1954. Beyond the date of establishment, he had no information about it or its activities.
242. The rest of Dr. Martinez's testimony involved broad, generalized statements about race relations nationally or historically.
243. For example, Dr. Martinez noted that Dodge City had worked with a national organization called National Day of the Cowboy. This organization has a code of conduct that largely mirrors a 10-point code of conduct created by Gene Autry. However, the National Day of the Cowboy's code does not include the language in Gene Autry's code that one must not be racially intolerant. Rather, it just says its members should be “tolerant” with specifying race. From this, Dr. Martinez appears to draw the tenuous inference that Dodge City as a municipality is racially intolerant, even if it was not direct evidence of an official discriminatory policy.
244. Dr. Martinez opined-once again based on national observances with no ties to Dodge City in particular-that at-large voting systems stemmed from racial animus during the early twentieth century. In his opinion, such systems are essentially institutionalized racism.
245. Dr. Martinez did not offer any testimony concerning the adoption of Dodge City's at-large voting system in 1971 or the reasons therefor.
246. Dr. Martinez ultimately concluded that institutionalized racism negatively impacts Hispanics in Dodge City in day-to-day practices.
247. He also concluded that at-large voting methods per se negatively impact Latino populations.
248. The Court finds that Dr. Martinez's testimony provides no evidence of discriminatory practices in Dodge City bearing on Latinos' rights or ability to vote. Furthermore, the vast majority of Dr. Martinez's testimony fails to establish any relevant nexus to the facts of this case. Only speculation connects Dr. Martinez's generalized platitudes about national and statewide tendencies over the past few hundred year to Dodge City itself. And beyond discussing once existing segregated theaters and a segregated public pool in the early twentieth century, Dr. Martinez has offered no specific facts applicable to Dodge City which show official discrimination.
249. To conclude, the Court finds Dr. Martinez's opinions unsupported by the evidence and wholly unhelpful to this case. Therefore, the Court accords it no weight.
2. Defendant's remaining expert witness-Dr. Kimberly Nelson
250. Dr. Kimberly Nelson is a tenured Professor of Public Administration and Government at the University of North Carolina at Chapel Hill. In that position, Dr. Nelson serves as the co-director for the Center for Public Leadership and Government. Dr. Nelson possesses an undergraduate degree from Virginia Tech, a master's degree from the University of Texas San Antonio in public administration and a Ph.D. in public administration from North Carolina State University. She has published more than a dozen peer-reviewed articles and book chapters, served as a co-editor of a peer-reviewed journal, and has authored three books, two of which as local government management textbooks.
251. Dr. Nelson specializes in assisting local government and governments officials is choosing the best form of government for each situation. Included with the form of government, Dr. Nelson also analyzes the types of elections associated with that form.
252. Dr. Nelson conducted a review of all cities in the United States with populations between 20,000 and 50,000. Within the 1,067 cities fitting this description, 597 (or 56%) utilize a Council-Manager form of government, identical to the Commission-Manger form of government employed in Dodge City. Of those, 66.2% utilize at-large elections for council or commission members.
Dr. Nelson stated that a “Commission” form of government is different than Dodge City's “CommissionManager” form of government in that each commission member acts as a department head without a need for a city manager.
253. Dr. Nelson testified that many benefits accompany a Council-Manager form of government, including less dysfunctional conflict, fewer problematic spending issues, higher reserves which leads to greater budget solvency, more innovation, and less risk of corruption.
254. Based on her research, Dr. Nelson testified that the at-large voting method was widely adopted in the early 1900s to combat government corruption. At that time, corrupt political dynasties or political machines' power base resided in wards. Thus, removing wards better enabled public choice in elections and resolved many problems caused by corrupt politicians.
255. Throughout her research, Dr. Nelson has never seen any indication that at-large elections were designed to dilute or suppress minority voters' right to vote. Rather, Dr. Nelson testified that the originator of the Council-Manager form of government, Richard Childs, wrote extensively about the benefits of at-large elections for smaller communities.
256. One of the benefits described by Dr. Nelson is that at-large voting schemes help prevent single-issue candidates because candidates must appeal to the city at large. Dr. Nelson also testified that single-issue candidates tend to create dysfunctional conflict within the governing body. Based on her experience and research, Dr. Nelson opined that district-based “boards generally don't function as well as they would in an at-large system.”
257. Dr. Nelson next stated that at-large elections do not tend to enhance the opportunity for Latinos to be elected to city councils. In contrast, she found that moving from at-large elections to district-based elections was very helpful in electing African Americans to city councils, albeit less so now than in the 1970s and 1980s.
258. Dr. Nelson explained that, unlike African Americans, “Latinos” includes persons from many different countries, cultures, and backgrounds which all can lead to ideological differences. Thus, Latinos do not necessarily look for the same things in their elected officials, causing a disparity in their candidate preferences.
259. To the extent Dr. Nelson concludes that switching to at-large elections here would not benefit Latinos, the Court does not find her testimony more persuasive than the statistical analysis performed by Dr. Barreto and Dr. Katz. In essence, the issue addressed by Dr. Nelson is Gingles II-whether a substantial number of Latinos prefer the same candidates. Although Latinos may come from various cultural backgrounds and differ ideologically, the issue of whether Latinopreferred candidates exist is intensely localized and factually based. Regardless, Dr. Nelson on cross-examination specifically stated she was not offering an opinion on the effects of minority representation due to switching from at-large to district-based elections.
260. However, Dr. Nelson's opinion is relevant as to Defendant's interests in maintaining its at-large elections for its Council-Manager form of government. Based on Dr. Nelson's uncontradicted testimony, the Court finds that a city may value the at-large election for its usefulness in uniting the population, providing residents access to each Commissioner, and preventing dysfunctional conflict caused by single-issue candidates.
II. Conclusions of Law
Plaintiffs' claims arise under 52 U.S.C. § 10301, commonly known as Section 2 of the Voting Rights Act. In Thornburg v. Gingles, the Supreme Court created a two-step test for plaintiffs seeking relief under Section 2. Plaintiffs must first show that the evidence could support finding the “necessary preconditions” set out in Gingles. Namely, Plaintiffs must establish: (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 . . . to defeat the minority's preferred candidate.”After doing so, Plaintiffs must demonstrate under the totality of the circumstances that the type of election at issue “result[s] in unequal access to the electoral process.” The totality of the circumstances analysis incorporates “the nonexclusive list of factors known as the senate factors, as well as any other material facts.”
478 U.S. 30, 50 (1986).
Id.
Id. at 50-51.
Id. at 46.
See United States v. Alamosa Cnty., 306 F.Supp.2d 1016, 1028 (D. Colo. 2004) (citing S. Rep. No. 97417, 1982 U.S.C.C.A.N. 177 at 206-07); see also Sanchez v. Bond, 875 F.2d 1488, 1491-92 (10th Cir. 1989) (listing factors).
The Court, therefore, must first determine whether Plaintiffs have succeeded in meeting the Gingles preconditions before proceeding to the second step if warranted.
A. Plaintiffs sufficiently establish Gingles I.
The first Gingles precondition requires Plaintiffs to show that the Hispanic population in Dodge City “is sufficiently large and geographically compact to constitute a majority in a single- member district.” This inquiry, “simply asks whether any remedy is possible in the first instance.” “Unless minority voters possess the potential to elect representatives in the absence of the challenged structure or practice, they cannot claim to have been injured by that structure or practice.”
Id. at 50 (“If it is not, as would be the case in a substantially integrated district, the multi-member form of the district cannot be responsible for minority voters' inability to elect its candidates.”).
Sanchez, 97 F.3d at 1311.
Id. (quoting Gingles, 478 U.S. at 50 n.17).
The sufficiently large standard is a simple test, merely requiring that “a party asserting § 2 liability . . . show by a preponderance of the evidence that the minority population in the potential election district is greater than 50 percent.” Turning to geographic compactness, “the inquiry should take into account traditional districting principles such as maintaining communities of interest and traditional boundaries.”
Bartlett v. Strickland, 556 U.S. 1, 19-20 (2009).
League of United Latin Am. Citizens (LULAC) v. Perry, 548 U.S. 399, 433 (2006) (further citations and quotations omitted).
Before turning to the legal arguments, Defendant argues that the Court must at this time determine whether the proposed districting plan truly enhances Latinos' ability to elect their own representatives in Dodge City. Quoting the Tenth Circuit, Defendant claims “the first prerequisite asks about the existence of a legally cognizable injury.” It argues that Latinos have suffered no cognizable injury because by 2022 they already formed the largest CVAP plurality in Dodge City, and now likely comprise a majority of Dodge City's CVAP. From this, Defendant argues Latinos already have the capacity to elect any candidates of their choice should they so choose.
Sanchez, 97 F.3d at 1311.
But the Supreme Court has postulated that “it may be possible for a citizen voting-age majority to lack real electoral opportunity.” Therefore, the fact that Latinos are likely the majority CVAP in Dodge City does not automatically prevent Plaintiffs from prevailing on their Section 2 claim.
LULAC, 548 U.S. at 428.
Returning to Gingles I, Defendant's request for a “performance analysis” essentially asks this Court to look beyond the scope of Gingles I and to examine the overall circumstances of the present at-large voting scheme.
See, e.g., Salas v. Sw. Tex. Jr. Coll. Dist., 964 F.2d 1542, 1555 (5th Cir. 1992) (“Although a registered voter majority class faces an obvious, difficult burden in proving that their inability to elect results from white bloc voting, they are not precluded, as a matter of law, from seeking to prove such a claim. In deciding such a majority's claim, the district court looks to the totality of circumstances.”).
In Abbott v. Perez, the Supreme Court clarified:
585 U.S. 579 (2018).
Under Gingles, the ultimate question is whether a districting decision dilutes the votes of minority voters, and it is hard to see how this standard could be met if the alternative to the districting decision at issue would not enhance the ability of minority voters to elect the candidates of their choice.
Id. at 617 (emphasis added).
At first blush, the “ultimate question” mentioned in Abbott seems to refer to the totality of the circumstances analysis. Defendant provides little reason to believe otherwise. Although, as described by the Tenth Circuit in Sanchez, “[t]he inquiries into remedy and liability . . . cannot be separated,” the Tenth Circuit did not collapse the totality of the circumstances analysis into Gingles I. Rather, both the Tenth Circuit and the Supreme Court have treated the intertwined inquiries of injury and remedy under Gingles I in a very specific way. That is to say, “the first Gingles requirement is needed to establish that the minority has the potential to elect a representative of its own choice in some single-member district. Without such a showing, there neither has been a wrong nor can be a remedy.”
Sanchez, 97 F.3d at 1311.
See id. at 1322-26 (performing totality of the circumstances analysis).
Strickland, 556 U.S. at 15 (emphasis added) (further citations and quotations omitted).
Thus, the inquiry under Gingles I is limited to whether the proposed solution would have its intended effect of allowing the minority population to elect their chosen representatives. In one sense, Gingles I asks not if there is an injury but whether the proposed solution would fix a hypothetical injury (to be proven later in the analysis). If not, then it is legally impossible to find that the minority can suffer any injury from the present voting scheme. However, as courts have repeatedly recognized, finding that a new districting plan will permit election of minorities' representatives is insufficient to show that the plaintiffs in fact suffer an injury from the present voting scheme. Thus, it is the “[f]ailure to prove the totality of circumstances [that] establishes the minority is not harmed by the challenged practice.”
See, e.g., Abbott, 585 U.S. at 618 (holding inability to craft a permissible remedy necessitates finding there is no injury).
See id.
See, e.g., Alamosa Cnty., 306 F.Supp.2d at 1040 (finding that although plaintiffs met all three Gingles preconditions, they had not satisfied the totality of the circumstances analysis under the second Gingles step).
Sanchez, 97 F.3d at 1311; see also Rose v. Sec'y, State of Ga., 87 F.4th 469, 475-76 (11th Cir. 2023) (“Thus, especially in a case like this one, where plaintiffs offer only a single, dramatic remedy . . . it makes no difference whether a claim fails for the lack of a permissible remedy at the precondition stage or after the totality of the circumstances analysis.”).
For these reasons, the Court will limit its analysis under the Gingles I factor to whether the Latino CVAP in Dodge City is sufficiently large and geographically compact to elect representatives of their choice in at least one proposed district. The answer to this question is simply yes, based on maps submitted by Plaintiff's expert, Dr. Oskooii.
See Fusilier v. Landry, 963 F.3d 447 (5th Cir. 2020) (“The first Gingles precondition, however, does not require a showing that a majority of the voters in a future election will be black. All the plaintiffs have to show is that the black minority is numerous enough to constitute a majority in a single-member district. That is not to say that the probability of blacks electing their preferred candidates in future elections is irrelevant. To the contrary, this bears on the totality of the circumstances, which we discuss later.” (further citation and quotations omitted)); but see Anne Harding v. Cnty. of Dallas, 948 F.3d 302, 311 (5th Cir. 2020) (characterizing Abbott as holding that “it was dispositive that plaintiffs' alternative map did not provide Latinos with an improved opportunity to elect another Latino-preferred candidate” (emphasis added)).
Dr. Oskooii drew 14 maps, considering total population, compactness, contiguity, and communities of interest. He testified that he did not consider race except in drawing Maps 13 and 14. In describing traditional districting criteria, Dr. Oskooii cited a handout published by the Kansas Office of Revisor of Statutes dealing with state legislative redistricting. The guidelines stated that districts should be compact, continuous, preserve the integrity of existing political subdivisions to the extent possible, recognized similarities in interests, and be easily identifiable and understandable by voters.
Admitted as Pls.' Ex. 145.
Defendant objects to Dr. Oskooii's maps, claiming that Dr. Oskooii relied upon race as the dominant factor in drawing his maps and ignored traditional districting principles. Of course, “race may not be the predominant factor in drawing district lines unless there is a compelling reason.”But “[t]he question whether additional majority-minority districts can be drawn, after all, involves a ‘quintessentially race-conscious calculus.'” Courts must avoid irrational arguments that necessitate concluding that “racial predominance plagues every single illustrative map ever adduced at the first step of Gingles.” Furthermore, the Supreme Court has clearly stated that “[t]he contention that mapmakers must be entirely ‘blind' to race has no footing in our § 2 case law.”
Allen v. Milligan, 599 U.S. 1, 31 (2023) (further citation, quotations, and brackets omitted).
Id. (quoting De Grandy, 512 U.S. at 1020).
Id. at 33.
Id.; see also Bethune-Hill v. Va. State Bd. of Elections, 580 U.S. 178, 196 (2017) (upholding intentional demographic target of 55% black voting age population for 11 new districts because race, although considered, did not predominate the legislature's decision).
Here, Dr. Oskooii specifically testified at trial that race did not predominate as to any map and was only considered for Maps 13 and 14. Therefore, Defendant must present “circumstantial evidence of a district's shape and demographics” to show Dr. Oskooii was lying. Defendant extrapolates “nine constants” from Dr. Oskooii's 14 maps, characterizing these constants as circumstantial evidence of racial predominance. These are:
Miller v. Johnson, 515 U.S. 900, 916 (1995).
1. In Districts 1-3, there is a majority of Hispanic citizens of voting-age population;
2. In Districts 1-3, Hispanics make up more than 55% of the citizens of votingage population;
3. In Districts 1-3, Whites make up less than 40% of the citizens of votingage population;
4. In Districts 4-5, Whites have a majority of the citizens of voting age population;
5. In Districts 4-5, the White citizens of voting-age population exceeds 60%;
6. In Districts 4-5, the Hispanic voting-age population is less than 30%;
7. In Districts 1-3, there are approximately one-and-a-half times more Hispanic eligible voters than White eligible voters;
8. In Districts 4-5, there are approximately two times more White eligible voters than Hispanic eligible voters; and
9. Districts 4 and 5 always have approximately two times more White eligible voters than the number of White eligible voters in Districts 1-3.
From these “constants,” Defendant argues that Dr. Oskooii intentionally packed racial groups to give Latinos three seats on the City Commission in each map. Although the Court admires Defendant's vigorous advocacy and detailed review of the record, its arguments are not persuasive. Dr. Oskooii credibly testified that for Maps 1-12 he considered population equality, compactness, contiguity, reasonable shape, and communities of interest-not race. And for Maps 13 and 14, Dr. Oskooii testified that he also considered race, but it did not predominate.
Defendant presents no cogent argument as to how Dr. Oskooii's proposed maps are at odds with the factors Dr. Oskooii claims to have relied upon. Indeed, these factors provide a cogent and persuasive explanation as to the “constants” of which Defendant complains. Further, they accurately reflect the reality of Dodge City's geodemographics. Beyond Dr. Oskooii's maps, the evidence is clear that Hispanics dominate the overall population and comprise the largest ethnicity as far as CVAP in Dodge City. Although hardly monochromatic, testimony from Dodge City residents confirms that neighborhoods south of Comanche tend to be predominately Hispanic and the areas north tend to be predominantly White. This testimony affirms both Dr. Oskooii's maps and the nine “constants.”
Defendant briefly quibbles with Dr. Oskooii splitting up the “community of interest” along Wyatt Earp Boulevard in Maps 1-8, 13, and 14. But Defendant also protests against Maps 9-12 keeping the Wyatt Earp Boulevard community together in District 2 because “it is stretched like a thin slice of bacon from the Western-most parts of Dodge to the Eastern-most parts.” These multiplicious and contradictory arguments are not persuasive.
Defendant also claims that Dr. Oskooii ignored traditional districting principles, namely preserving the existing precincts. This argument likewise misses the mark. The Kansas guidelines cited by Dr. Oskooii state that redistricting plans should preserve prior existing political boundaries “to the extent possible.” Dr. Oskooii testified that it was not possible to preserve the existing precincts because not a single precinct had sufficient population to comprise a district. Defendant offered no evidence to rebut Dr. Oskooii's testimony. Therefore, Dr. Oskooii did not offend traditional districting principles by splitting the preexisting precincts.
Defendant's arguments regarding Dr. Oskooii's maps obfuscate the issue in Gingles I. It is a simple inquiry-whether Hispanics are sufficiently large and geographically compact enough that in a legally permissible districting they could form a single performing district. Based on the size and compactness of the Dodge City Hispanic population, Defendant cannot seriously argue that Plaintiffs fail to meet Gingles I's requirements. They concede as much in their proposed findings of fact. With this admission, it seems clear that despite Defendant's efforts to the contrary, Gingles I is a non-issue in this case.
Regardless, Plaintiffs have satisfied their burden to postulate a reasonable and legally permissible solution that would allow Latinos to elect at least one representative of their choice to the Commission. Whether the Court were to choose Map 1 or Map 11, the result remains the same. Dr. Oskooii has submitted legally permissible maps which would likely allow Latinos to elect at least one candidate of their choice to the City Commission. Under this Court's understanding of the Supreme Court and Tenth Circuit caselaw, that is all the Gingles I precondition requires.
B. Plaintiffs sufficiently establish Gingles II.
To establish the second Gingles precondition, Plaintiffs must prove that Latinos in Dodge City vote cohesively as a group. To be politically cohesive, the minority group must have “expressed clear political preferences that are distinct from those of the majority.” That is, Plaintiffs must show “that a significant number of minority group members usually vote for the same candidates.”
Sanchez, 97 F.3d at 1312 (further citation and quotations omitted).
Gingles, 478 U.S. at 56.
Courts “may not assume from a group of voters' race that they think alike, share the same political interests, and will prefer the same candidates at the polls.” Likewise, “only the race of the voter, not the race of the candidate, is relevant to vote dilution analysis.” Furthermore, the reasons underlying the minority population's voting habits are irrelevant to this inquiry.
LULAC, 548 U.S. at 433 (further citation, quotations, and brackets omitted).
Gingles, 478 U.S. at 68.
See Large v. Fremont Cnty., 709 F.Supp.2d 1176, 1192 (D. Wyo. 2010) (citing Bond, 875 F.2d at 1493).
Lay testimony is relevant but insufficient to establish political cohesion in voting dilution cases. Usually, statistical evidence is vital to a plaintiffs' case on Gingles II and III. Even so, the question is not whether bloc voting is “statistically significant,” but rather whether it is “legally significant.” As stated by the Supreme Court:
See, Sanchez, 97 F.3d at 1320 (“[W]hile lay testimony is relevant to determine who is the candidate of choice, it is not alone dispositive.”).
See, e.g., Growe v. Emison, 507 U.S. 25, 41 (1993); Sanchez, 97 F.3d at 1312 (“Necessarily, when we examine the evidence of political cohesiveness as voting preferences, we look to the same statistical evidence plaintiffs must offer to establish vote polarization.”).
See Gingles, 478 U.S. at 56.
Significant differences are not evidence that what is at work is legally or practically important. Statisticians distinguish between statistical and practical significance to make the point. When practical significance is lacking-when the size of a disparity is negligible-there is no reason to worry about statistical significance.
Brnovich v. Democratic Nat'l Comm., 141 S.Ct. 2321, 2343 n.17 (2021) (further quotations, alterations, and ellipses omitted) (quoting Federal Judicial Center, Reference Manual, at 252) (discussing statistical versus legal difference in context of § 2 disparate impact claim); see also Cooper v. Harris, 581 U.S. 285, 304 (2017) (acknowledging expert's determination of “statistically significant” racially polarized voting while concluding it failed to address the relevant legal question-that is, whether there was sufficient white voting to cancel African Americans' ability to elect representatives of their choice).
When analyzing statistical data in the voting rights context, courts distinguish between endogenous and exogenous elections. Endogenous elections are those involving the same office as the election being challenged under the VRA. Exogenous elections consist of any other elections where, here, Dodge City residents voted. Courts appear unanimous in concluding that the relative probative value of exogenous elections depends on their similarity to the election at issue.
See, e.g., Alamosa Cnty., 306 F.Supp.2d at 1021 (D. Colo. 2004); see also Sanchez, 97 F.3d at 1317 (recognizing expert analyzed both endogenous and exogenous elections when opining on Gingles II and III).
See Alamosa, 306 F.Supp.2d at 1021.
See id.
See Mo. State Conf. of NAACP v. Ferguson-Florissant Sch. Dist., 201 F.Supp.3d 1006 (E.D. Mo. 2016), aff'd, 894 F.3d 924 (8th Cir. 2018); Cisneros v. Pasadena Indep. Sch. Dist., 2014 WL 1668500 (S.D. Tex. Apr. 25, 2014) (“These partisan exogenous elections cannot be used to overcome the evidence supplied by the non-partisan endogenous elections.”).
To address Gingles II specifically, Plaintiffs must “establish by a preponderance of the evidence who is the preferred minority candidate in each election.” It is not enough for plaintiffs to generally show racially polarized voting or simply point to minority candidates running for office. More evidence, “including lay testimony or statistical analyses of voting patterns,” is necessary to show that an individual is a minority-preferred candidate.
Sanchez, 97 F.3d at 1320.
Id. at 1321.
Id.
Recognizing that minorities may also prefer white candidates, the Tenth Circuit has instructed courts to “engage in a detailed, practical evaluation of the extent to which any particular white candidate was, as a realistic matter, the minority voters' representative of choice.” It cautioned that when confronted “with a history of Anglo/Anglo contests,” courts should be hesitant of “the myopic presumption there is a minority preferred candidate in any race in which the minority votes.” In determining whether white candidates were minority-preferred, the Tenth Circuit listed several factors for courts to consider, including:
Id. (quoting and adopting Jenkins v. Red Clay Consol. Sch. Dist. Bd. of Educ., 4 F.3d 1103, 1129 (3d Cir. 1993)
Id. at 1320.
[1] the extent to which the minority community can be said to have sponsored the candidate . . . [;] [2] the candidate's attention to the issues concerning minorities; [3] the extent the candidate campaigned in the minority's neighborhoods or addressed
predominantly minority crowds . . . [;] [4] minority turnout for the election . . . [; and] [5] evidence of the Anglo candidate's ties to the minority community.
Id. (further citations and quotations omitted); but see Ruiz v. City of Santa Maria, 160 F.3d 543, 552 (9th Cir. 1998) (rejecting Sanchez's totality of the circumstances theory in favor of a “bright-line rule,” based on whether “a candidate who receives sufficient votes to be elected if the election were held only among the minority group in question”).
“This list, of course, is not exclusive, and the district court must be sensitive to the quality of the evidence plaintiffs and defendants offer to assist in this factfinding.” Practically speaking, however, statistical evidence remains key to determining which candidates are Latino-preferred.
Sanchez, 97 F.3d at 1321.
See id. at 1321-22 (reversing district court for not properly considering statistical evidence without any reference to the prior articulated factors).
As discussed in the Court's findings of fact, Plaintiffs have established by a preponderance of the evidence that Latinos in Dodge City vote in significant numbers for the same candidates. Therefore, the Court concludes as a matter of law that Gingles II is met in this case.
C. Plaintiffs fail to sufficiently establish Gingles III.
The Court now turns to the third Gingles precondition. In analyzing the evidence under this precondition, the Court remains cognizant of Plaintiffs' legal burden-that is, Plaintiffs must show that white bloc voting usually prevents Latino-preferred candidates from being elected in Dodge City's City Commission elections. The Supreme Court has been clear that Gingles III relates to “the [current] district as actually drawn.” Based on the Court's findings as to which candidates were Latino-preferred, Plaintiffs have not met their burden.
Gingles, 478 U.S. at 49.
Cooper, 581 U.S. at 302.
To describe legally significant racially polarized voting, “[t]he terms used by the Gingles Court are ‘usually,' ‘normally,' and ‘generally.'” However, neither the Supreme Court nor the Tenth Circuit has defined these terms. Other circuit courts, however, appear to fall into two camps when defining “usually,” “normally,” or “generally” under Gingles. For example, the Fourth Circuit decided it need not “specify a meaning for these terms; suffice it to say that they mean something more than just 51%.” In contrast, the Ninth Circuit mentioned that “usually” meant simply “more than half of the time.” However, despite an in depth review of related caselaw, the Court is unaware of any other court, or even subsequent Ninth Circuit opinions, to adopt the Ninth Circuit's definition of “usually.”
Lewis v. Alamance Cnty., N.C. , 99 F.3d 600, 606 n.4 (4th Cir. 1996) (citing Gingles, 478 U.S. at 49, 51, 56).
See Pope v. Cnty. of Albany, 94 F.Supp.3d 302, 335 (N.D.N.Y. 2015) (“In this largely fact-driven inquiry, different circuits have developed divergent acceptable thresholds of minority success.”).
Lewis, 99 F.3d at 606 n.4 (emphasis added); see also Uno v. City of Holyoke, 72 F.3d 973, 980 (1st Cir.1995).
Old Pers. v. Cooney, 230 F.3d 1113, 1122 (9th Cir. 2000).
See, e.g., Yumori-Kaku, 273 Cal.Rptr.3d at 459-60 (mentioning and criticizing the Ninth Circuit's holding in Old Person because “Gingles's third factor is not measured by mathematical formula but by the trial court's searching assessment of statistical and other evidence presented.”).
Even though it has not addressed the issue directly, the Tenth Circuit's decision in Sanchez v. Colorado leads the Court to reach the same conclusion as the Fourth Circuit. When the Tenth Circuit analyzed Gingles III in Sanchez, it recognized the importance of a “legally significant” racial voting block and noted the need for a flexible approach in the absence of a specific doctrinal test. The Tenth Circuit cautioned, however, “while legally significant white bloc voting enables the majority in the ordinary course[] to trounce minority-preferred candidates most of the time, its presence may be more subtle requiring close inquiry over time.” Thus, “the isolated success of a minority candidate in a district that usually exhibits vote polarization will not alone negate plaintiffs' showing.”
97 F.3d 1303 (10th Cir. 1996).
Id. at 1312-13.
Id. at 1313 (further citations and quotations omitted).
Id.
From the Tenth Circuit's juxtaposition of “most of the time” with “isolated success,” the Court infers a (comparatively) simple test. If white bloc voting prevents the election of Latinopreferred candidates “most of the time,” then Gingles III is met. This standard is necessarily higher than simply “a majority of the time”-or 51%. After all, if Latino-preferred candidates experienced success 49% of the time, a Latino-preferred candidate's success in any one election could hardly be considered “isolated.” To be “isolated,” that candidate's success must be the exception to the rule, with the rule reflecting the normative outcome in that locality. Accordingly, if a plaintiff cannot show that white-preferred candidates prevail over Latino-preferred candidates “most of the time,” then he has not met Gingles III.
However, this does not end the inquiry. If plaintiffs cannot show that the white-preferred candidate prevails “most of the time,” particularly in recent elections, they may rely on a more far-reaching analysis-i.e., the “close inquiry over time” described by the Sanchez court. What this inquiry entails is unclear. Nevertheless, it must comprise a more comprehensive approach under which plaintiffs show a recognizable pattern of white bloc voting “usually,” “normally,” or “generally” defeating Latino-preferred candidates within the locality.
See id.
Here, Plaintiffs have not shown that white bloc voting in Dodge City Commission elections defeats the Latino-preferred candidates of choice “most of the time.” Across the four endogenous elections analyzed, the Court identified eight Latino-preferred candidates. Of those, five were elected. This represents a 62.5% success rate for Latino-preferred candidates.
This analysis remains the same even if the Court had found facts perhaps more favorable to Plaintiffs. For example, if the Court had found that Scoggins was a Latino-preferred candidate in 2021, Latino-preferred candidates' success rate would still be 55.6%. Conversely, if there were no Latino-preferred candidates in 2019, then there were still five remaining Latino-preferred candidates. Of these, three were elected, representing a 60% success rate. Finally, if the Court had found (to be clear, it did not) both that there were no Latino-preferred candidates in 2019 and that Scoggins was a Latino-preferred candidates in 2021, then three of the six Latino-preferred candidates were elected-exactly 50%. In none of these hypothetical scenarios can Plaintiffs show that white bloc voting usually defeats Latino-preferred candidates.
Thus, the success rate for Latino-preferred candidates fails to indicate white bloc voting. The Court cannot conclude white bloc voting prevents Latino-preferred candidates from being elected most of the time when Latino-preferred candidates win at least half the time. In other words, although statistically significant, white bloc voting for these endogenous elections is not legally significant under Gingles III. Therefore, Plaintiffs must rely on a close inquiry over time to show that white bloc voting nevertheless “usually” prevents Latino candidates from being elected.
However, Dr. Barreto's analysis of elections in Dodge City dates back solely to 2014, covering only the endogenous elections discussed above. Likewise, Dr. Barreto's analysis of exogenous elections only analyzes elections between 2014 and 2022. And Dr. Katz only analyzed the same four endogenous elections as Dr. Barreto.
Turning to the exogenous elections, it becomes clear they do not assist Plaintiffs in showing legally significant white bloc voting. First, the two similar exogenous elections-2019's DCCC Trustee election and 2021's School Board election. Regarding the latter, Dr. Barreto did not indicate which candidates were elected. Therefore, it is entirely unhelpful to Gingles III, as Plaintiffs cannot show that white-preferred candidates beat Latino-preferred candidates where there is no indicated winner. As to the DCCC Trustee election, two Latino-preferred candidates won election to three available positions. This two-thirds success rate only supports the finding that white bloc voting does not normally prevent Latinos from electing their candidates.
The other exogenous elections are similarly unhelpful, although for a different reason. These county-wide, state-wide, and nation-wide elections do not show that Latinos usually are prevented from electing candidates of their choice by the white citizens of Dodge City. Even if the Court reached the “generalized conclusion” that racially polarized voting exists in Dodge City, that still would “fail[] to meaningfully (or indeed, at all) address the relevant local question.”That local question here is whether white bloc voting usually prevents Latinos from electing representatives of their choice within Dodge City. In other words, the Court is not interested in statistically significant evidence of racially polarized voting but legally significant evidence of the same. In general, dissimilar exogenous elections provide nothing helpful to this question.
Cooper, 581 U.S. at 304 n.5.
The only exogenous election which possibly helps determine Gingles III is the 2020 County Clerk election for Ford County. As a local election, it at least is relevant to whether racially polarized voting in Dodge City prevents Latinos from being elected. To that end, it reveals under both Dr. Barreto's King's EI and RxC analyses that the heavily preferred white candidate won election over the heavily preferred Latino candidate.
However, it is not very probative of Gingles III for two reasons. First, Dodge City is not the only municipality within Ford County. Indeed, Ford County also includes the townships of Bucklin, Spearville, and Ford City, meaning these outside votes affected the result of the election. Second, this is an isolated incident hardly indicative of the “close inquiry over time” Plaintiffs must perform. Therefore, the Court concludes that the 2020 Ford County Clerk election comprises some, albeit very little, evidence Gingles III is met in this case. But this election alone is nowhere near sufficient to carry Plaintiff's burden. As such, Plaintiffs have failed to present sufficient evidence showing that white bloc voting usually prevents Latino-preferred candidates from winning elections.
In sum, the endogenous elections between 2014 and 2022 show Latino-preferred candidates winning at least half of the available seats on the City Commission. The 2019 DCCC Trustee election resulted in Latino-preferred candidates winning two of the three available spots. Finally, the 2020 Ford County Clerk election show the heavily white-preferred candidate winning over the heavily Latino-preferred candidate. Whether considered alone or together, none of these facts establish, by a preponderance of the evidence, that white bloc voting in Dodge City “usually,” “normally,” or “generally” prevents the election of Latino-preferred candidates. Therefore, the Court concludes that Plaintiffs fail to meet the third Gingles precondition.
Because Plaintiffs failed to meet one of the Gingles preconditions, the matter is over, and the Court need not address the totality of the circumstances through the Senate factors under the second step of Gingles.
IT IS THEREFORE ORDERED that judgment be entered in favor of Defendant City of Dodge City, Kansas.
IT IS SO ORDERED.
This case is closed.