The Gingles factors and Section 2
From 1986 through April 29, 2026, the legal standard for evaluating majority-minority district claims under Section 2 of the Voting Rights Act was largely governed by the three-part test established in Thornburg v. Gingles (1986). Data analysts working on redistricting cases had to demonstrate each of the "Gingles factors" quantitatively.
What Section 2 required
Section 2 of the VRA prohibits any voting practice or procedure that "results in a denial or abridgement" of the right to vote on account of race. Under the Gingles framework, a plaintiff challenging a district map as diluting minority voting power had to show:
Gingles Factor 1 — Numerically sufficient and geographically compact
The minority group must be sufficiently large and geographically compact to constitute a majority in a single-member district. This is a quantitative test: analysts calculate CVAP percentages in proposed alternative districts to show a majority-minority district is possible. Geographic compactness is measured using standard scores (Reock, Polsby-Popper).
Gingles Factor 2 — Politically cohesive
The minority group must vote cohesively—that is, it must support the same candidates at high rates. Data analysts demonstrate cohesion through racially polarized voting (RPV) analysis, using ecological inference or ecological regression to estimate the candidate preferences of different racial groups from aggregate precinct-level returns (since individual ballots are secret).
Gingles Factor 3 — White bloc voting defeats minority-preferred candidates
The white majority must usually vote as a bloc in a way that defeats the minority-preferred candidate. This too is demonstrated through RPV analysis—specifically showing that the candidate preferred by the minority group loses in the challenged district because of racially polarized bloc voting among the majority group.
The totality of circumstances
Meeting all three Gingles preconditions did not automatically win a Section 2 case. Courts also considered "totality of circumstances"—the history of official discrimination, the degree of polarization, socioeconomic disparities, and other factors. But the Gingles factors were the quantitative threshold that analysts had to clear first.
Data implications (pre-Callais)
To litigate a Section 2 case before Callais v. Landry, analysts needed: CVAP data by race for proposed district configurations; RPV analysis (ecological inference results) for recent elections; and compactness scores for proposed alternative districts. The effects-based framework rewarded thorough, multi-election RPV analysis with consistent results across methods.