Racially Polarized Voting Analysis of Voting Expert Witness found to satisfy the Daubert requirements in light of anticipated changes
Posted on November 9, 2023 by ewp_staff_writer
Plaintiff, Black Louisiana voters, alongside nonprofit organizations advocating for civic engagement and social equality, filed this lawsuit formally accusing the Defendants, specifically Secretary of State R. Kyle Ardoin, and Intervenor-Defendant, the State of Louisiana, through Louisiana Attorney General Jeff Landry (collectively, “Movants”) of having violated Section 2 of the Voting Rights Act of 1965. Section 2 requires the redistricting body to ensure that voters of color have an equal opportunity “to participate in the political process and to elect candidates of their choice.” The core contention revolved around the assertion that the implemented maps resulted in an unjust denial of a meaningful opportunity for Black voters in Louisiana to elect candidates of their choice to both the State Senate and House of Representatives.
The argument was underpinned by the application of the Gingles test, a legal framework established by the Supreme Court. According to this test, the State Legislative Maps were deemed dilutive based on three critical factors:
Firstly, it was argued that the Black population in Louisiana was sufficiently large and geographically compact to potentially constitute a majority in six to nine additional single-majority House districts and three more single-member Senate districts. Secondly, the lawsuit contended that voting patterns in Louisiana were characterized by high racial polarization, indicating that voters tended to align along racial lines. Lastly, the Plaintiffs argued that, under the State Legislative Maps, racially polarized voting would typically lead to the defeat of Black Louisianans’ preferred candidates in majority-white districts. The lawsuit stressed upon the cohesive political alignment of Black voters in Louisiana, which faced a serious setback with the white majority’s consistent bloc voting, which routinely worked against the candidates preferred by Black voters.
Movants demanded the exclusion of the testimony one of Plaintiff’s experts, Dr. Lisa R. Handley, under Federal Rules of Evidence 702 and 703, as well as Daubert and its progeny. The Movants did not contest Handley’s credentials as an expert in racially polarized voting. Instead, their challenge focused on the assertion that the methodologies employed by Handley were both unreliable and not entirely relevant to the case.
Voting Expert Witness
With over twenty-five years of experience in redistricting and voting rights, Lisa R. Handley is recognized nationally and internationally as an expert in these fields. She has served as a practitioner and academician, advising numerous jurisdictions, including the U.S. Department of Justice, and clients on redistricting. Handley’s expertise extends to her role as an expert in dozens of redistricting and voting rights court cases, working with state and local jurisdictions, redistricting commissions, and civil rights organizations.
Discussions by the Court
The Court was required to apply the familiar FRE 702 and Daubert analysis. It’s worth noting that a revision to Federal Rules of Evidence 702 was slated to become effective on December 1, 2023. The anticipated change clarified that expert testimony might not be admitted “unless the proponent demonstrated to the court that it was more likely than not that the proffered testimony met the admissibility requirements set forth in the rule.” Section 702(d) was amended to include language stating that the “expert’s opinion reflects a reliable application of the principles and methods to the facts of the case.” The intent of the proposed rule change was to focus and direct district courts to conduct the gate-keeping inquiry enunciated in Daubert and refrain from bypassing the admissibility determination in favor of a question of weight to be decided by a fact finder.
Handley had been engaged by the Plaintiffs with the specific task of analyzing “voting patterns by race” to establish the foundation for two of the three elements of the ‘results test’ outlined in Thornburg v. Gingles. These elements included conducting a racial bloc voting analysis to ascertain the political cohesion of the minority group and determining if whites were consistently voting as a bloc to typically thwart the candidates preferred by minority voters.
Handley relied primarily on the ecological inference RxC statistical technique to analyze voting patterns by race, using voter data at the precinct level to estimate such patterns. To accomplish this, she aggregated a substantial amount of election precinct data to form a comprehensive database for her statistical analyses. Movants contested Handley’s methodology, asserting that she utilized undisclosed sources for compiling her database, specifically the Voting and Elections Science Team for shapefiles and the ACLU for data aggregation.
Movants conceded that experts might seek assistance in gathering underlying data, acknowledging that such reliance is not inherently flawed. However, they argued that relying on “undisclosed persons with unknown credentials to process data is unreliable.”
The Court, in its findings, determined that the compilation of data by others, as relied upon by Handley, was simply that—a compilation of data. While others may have participated in gathering the data, the Court emphasized that this did not render the data insufficient. Handley testified that she directed the data gathering and its compilation, with no evidence indicating that others conducted the analysis. There was no evidence that Handley relied on the opinions or expertise of undisclosed experts beyond her own scope of expertise.
Upon reviewing Handley’s reports, the Court found that she performed the data analysis to arrive at her conclusions, and there was no challenge to her expertise in analyzing voting patterns by race. Additionally, the Court clarified that the Plaintiffs were not obligated to disclose the underlying data sources and gatherers under Rule 26(a)(B)(2) of the Federal Rules of Civil Procedure or the Court’s scheduling order.
In Louisiana, around 30% of voters consistently participated in early and absentee voting in statewide elections from 2012 onwards. Notably, during the November 2020 elections, 45.6% of the total votes cast were early or absentee. Similar trends were observed in the November 2019 and November 2022 elections, with 33.7% and 26.9% of overall votes cast being early and absentee, respectively. However, the Louisiana Secretary of State website only reported candidate-specific early and absentee votes at the parish-wide level, prompting Handley to disaggregate the data down to the precinct level for her racially polarized voting (RPV) analysis.
To achieve this, Handley employed a non-peer-reviewed allocation method, distributing early and absentee votes to specific precincts within a parish proportionally based on the votes received by each candidate on Election Day in the studied areas. The Defendant contested Handley’s allocation method, asserting that it failed to cap the number of early or absentee votes assigned to each precinct based on the actual voter turnout, leading to unreliable results. The Defendant argued that this lack of capping resulted in overestimation in some precincts and underestimation in others.
Handley justified her allocation method by emphasizing the statistically significant percentage of the total vote represented by early voting in Louisiana. She explained that, lacking specific early voting data by precinct, she assumed the same allocation of election day votes per candidate per precinct to distribute early votes. Movants and the Defendant’s expert, Dr. Tumulesh Solanky, argued that Handley’s allocation method introduced bias, resulting in over and underestimation of votes in some precincts. Movants did not argue that ecological inference (EI) was an inappropriate analysis for evaluating racially polarized voting (RPV). They acknowledged the necessity of precinct-level data for conducting the EI analysis. The crux of the dispute centered on the optimal method for de-aggregating or allocating parish-wide data to obtain usable precinct-level data.The Court recognized that certain challenges to expert testimony might pertain more to the weight of the evidence than to its admissibility. The Court, henceforth, disagreed, deeming the slight over and underestimate resulting from the subject allocation method statistically insignificant and not rendering Handley’s conclusions unreliable. The Court found that Handley’s data sets remained consistent, debunking the argument that she assumed homogeneity in voting across precincts. Contrary to this claim, day-of-voting data was accessible at the precinct level, providing a direct measure of votes cast and reflecting diverse voting patterns among precincts. Handley justified the logical proportionate allocation of early votes among precincts, emphasizing that she tested for the confidence of the data.In addressing Solanky’s criticism of the allocation method, Handley conducted tests for bias and found none, further reinforcing the reliability of her approach.
Movants further challenged the early vote allocation method, contending that it lacked peer review. Handley countered by testifying that other experts utilized the same methodology, and Solanky, despite disagreeing with Handley’s approach, used her database for his EI analysis. Handley also submitted a rebuttal report, conducting additional analyses consistent with and supporting her RPV conclusions. The Court concluded that Handley’s allocation method wasn’t biased, cited a lack of evidence to render the analysis infirm, and deemed the allocation assumptions challengeable through cross-examination.
Movants contended that Handley’s opinions lacked relevance and failed to aid the trier of fact because she did not conduct a district-specific racially polarized voting (RPV) analysis but focused only on seven “areas of interest.” The argument asserted that the ecological inference (EI) analysis should have been performed statewide. Movants specifically claimed that Handley’s classification of districts as “effective” or not lacked specificity regarding the required Black Voting Age Population (BVAP) threshold for effectiveness.
The contention further emphasized that Handley’s opinion was irrelevant as it did not express the “threshold level of BVAP” necessary to afford black voters a realistic opportunity to elect their preferred candidate. Plaintiffs asserted that the State Legislative Maps were dilutive, claiming that the Black Population in Louisiana was “sufficiently large and geographically compact to constitute a majority” in six to nine additional single-majority House districts and three additional single-member Senate districts. The contention highlighted that vote dilution claims were district-specific.
Vote dilution claims, according to Movants, required a RPV analysis specific to the areas of the state where the vote dilution is alleged to occur, as relying on statewide voting statistics to establish legally significant white bloc voting was deemed erroneous by the Fifth Circuit. Handley’s methodology involved using election results from 16 different statewide elections but confined her EI analysis to the specific election data for the voters who live within each of the seven areas of interest.
The Court, in its findings, determined that Handley conducted a sufficiently localized analysis of the challenged districts. It concluded that Plaintiffs had demonstrated by a preponderance that Handley’s opinion testimony would assist the Court as the trier of fact. The Court found that her opinions were based on sufficient facts and data, derived from reliable principles and methods, and that she reliably applied these principles and methods to the facts of the case.
The Court denied the motion to exclude the testimony of Plaintiff’s expert Lisa Handley. The Court has not arrived on an outcome for this case since the remaining issues involved in this case still await resolution.
The Court applied the familiar Rule 702 and Daubert analysis for determining admissibility of expert testimony. The Court noted that proposed amendments to Rule 702 clarify that the proponent of the testimony must demonstrate its admissibility. The Court found that the voting expert witness’ methodology and data sources were sufficiently reliable. The fact that assistants helped compile data did not render the data or analysis unreliable. The Court found the expert’s method of allocating the parish-wide early voter data proportionally to the precincts was reasonable and did not induce bias. The Court found that Handley performed a sufficiently local analysis of the challenged districts. The Court ruled that the slight imprecision resulting from the allocation methodology was insignificant and could be addressed on cross-examination. The Court found the expert’s opinions would assist the trier of fact.
Posted In: Expert Challenges, Voting Expert Witness
Tagged In: Election, Expert Testimony, Method of Allocation, Unreliable, Voting