In Course 2, we studied the brutal litigation cascade of Alabama forcing its gerrymandered map to the Supreme Court. Here, we look under the hood of Allen v. Milligan (2023). Against all odds, the Supreme Court upheld the Voting Rights Act and struck down Alabama's map. Why? Because the data science was airtight. The plaintiffs successfully deployed the exact computational methods discussed in Modules 7-10. By examining the expert witness reports filed in the Alabama federal district court, we see the blueprint for securing democratic sovereignty through irrefutable data analysis.
In This Module
- Covers: The technical expert reports filed in Allen v. Milligan, integrating both Ecological Inference and MCMC Ensembles into a unified legal argument.
- Why it matters: State governments constantly invent new legal theories to claim their maps are "race neutral." Milligan proves that aggressive, multi-scale computational modeling can pierce that defense.
- After this module, the reader can: Understand how a finished Methodology Portfolio is converted directly into explosive, unassailable federal testimony.
Reading List
Conceptual
-
Read the majority opinion authored by Chief Justice Roberts. Uniquely for a Supreme Court decision, the opinion spends immense time describing the specific mapping algorithms and data estimates used by the plaintiffs. Roberts forcefully defends the rigors of the Gingles test, validating the decades of statistical methodology we have studied in this course.
Methods
-
The Ecological Inference masterclass. Dr. Liu processed dozens of endogenous Alabama elections through EI algorithms to prove Racially Polarized Voting. The state attempted to claim that Black Alabamians were simply voting for Democrats, not "minority-preferred" candidates. Dr. Liu's statistical outputs rigorously destroyed the "party, not race" defense.
-
The Ensemble masterclass. Dr. Duchin demonstrated that if you simply program an MCMC computer to draw thousands of random Alabama maps that roughly mirror the racial demographics of the state, almost all of them produce at least two Black opportunity districts. Because the state's enacted map only produced one, it was mathematically revealed as an artificial intent to crack the Black Belt.
Key Concepts
Why did Allen v. Milligan validate computational redistricting methods for the Voting Rights Act?
Chief Justice Roberts authored a majority opinion describing the specific algorithms and data estimates used by the plaintiffs. The Court upheld the VRA and struck down Alabama's map because the data science was airtight—MCMC ensembles and Ecological Inference successfully proved the state illegally diluted Black voting power. The decision validated decades of statistical methodology.
How did Dr. Liu's EI analysis destroy the 'party, not race' defense?
Alabama claimed Black voters were simply voting for Democrats, not "minority-preferred" candidates. Dr. Liu processed dozens of endogenous elections—including nonpartisan local races—through EI algorithms, demonstrating that Black voters cohesively supported specific candidates regardless of party label while white voters systematically opposed them. The pattern was racially polarized, not merely partisan.
How did Dr. Duchin's ensemble prove Alabama intentionally cracked the Black Belt?
Dr. Duchin generated thousands of random Alabama maps following all legal constraints. Almost all produced at least two Black opportunity districts. Because the enacted map only produced one, it was mathematically revealed as an artificial intent to crack the geographically concentrated Black Belt population across multiple districts, diluting their ability to elect representatives of their choice.
Goal: Fuse the data methods into the final chapters of your Methodology Portfolio.
A single algorithm is rarely enough. The strongest cases combine spatial simulation with behavioral modeling. Detail their intersection.
- The Hybrid Test: How will your Ensemble Maps (M7) interact with your Ecological Inference results (M10)? Enter a protocol into your Portfolio. For example: "Any hypothetical 'fair' map generated by the MCMC ensemble will then be stress-tested against the EI-derived Racially Polarized Voting equations to verify it functions as a true opportunity district."
- Legal Defense Check: The state of Alabama argued that algorithms like MCMC rely on "proportionality," which the VRA explicitly bans. Re-read the Roberts decision and write a two-sentence defense to place in your Portfolio explaining why your algorithm tests intent, rather than guaranteeing proportionality.