Data Analysis for PolicyMaking in Georgia
Module 4 · Disparate Impact Analysis › 4.1 What disparate impact means
Subsection 4.1

What disparate impact means

~3 min

Reading

Disparate impact is a legal and statistical concept that describes a situation in which a facially neutral policy or practice produces unequal outcomes for a protected group—even without any evidence of discriminatory intent. It is distinct from intentional discrimination (also called "disparate treatment"). Understanding disparate impact analysis is essential for voting rights work, even after Callais v. Landry raised the evidentiary bar for redistricting claims.

The basic framework

A disparate impact claim follows a three-step structure:

  1. Identify the policy or practice: What did the government do? (e.g., consolidate polling places, change voter ID requirements, redraw district lines.)
  2. Quantify the differential effect: Does the policy affect a protected group at a statistically different rate than the comparison group? (e.g., Black voters wait longer, mail ballots from Black precincts are rejected at higher rates.)
  3. Evaluate necessity and alternatives: Was the disparate effect necessary to achieve a legitimate government purpose? Could the purpose be achieved with a less discriminatory alternative?

Why statistical evidence matters

Disparate impact analysis requires actual numbers. You cannot assert that a policy "probably" affects a group differently—you have to show it, with data. The analysis needs a comparison group, a baseline, and a measure of effect size that is large enough to be meaningful (not just statistically significant).

Disparate impact in voting rights

Common disparate impact claims in voting rights include: unequal wait times at polling places by race; higher mail ballot rejection rates in majority-minority counties; polling place consolidations that disproportionately increased travel distances for Black or Latino voters; and voter purge practices that removed minority voters at higher rates.

Post-Callais note

After Callais v. Landry, disparate impact evidence is still valuable—but its role has shifted. In redistricting, it is now evidence supporting an intent inference, not a standalone basis for a Section 2 remedy. In other voting rights contexts (polling place accessibility, ballot rejection), disparate impact claims under other statutes remain viable as before. Knowing exactly which legal theory your data is supporting is now more important than ever.

Disparate Impact Analysis — Three-Step Framework Step 1 Identify the policy What specific government action or policy is at issue? Example: county reduced polling places by 50% Step 2 Quantify differential effect Does the policy affect the protected group at a higher rate? Example: Black voters' avg. travel distance increased 2× vs. white voters Step 3 Evaluate necessity Was the effect necessary? Could a less discriminatory alternative work? Example: Could cost savings be achieved with fewer closures? Post-Callais: in redistricting, disparate impact is now circumstantial evidence of intent — not a standalone remedy basis. In other voting contexts (ballot rejection, polling access), disparate impact claims remain viable under other statutes.
Diagram 4.1 · Disparate impact analysis framework. Three steps from policy identification to necessity evaluation.