A New Approach to Measuring the Racial Impact of Redistricting

At least since the passage of the Voting Rights Act, an important component in evaluating any new legislative districting plan requires examining its likely impact on racial and ethnic minorities to fully participate in the electoral process.b The prevailing standard enshrined both in legal precedent and the statistical methods used to evaluate redistricting plans is outcome based: we ask is the minority group electorally cohesive and if so can they elect candidates of their choice under a give plan? We propose an alternative standard and statistical methodology that is opportunity based. That is, we ask does a redistricting plan treat voters of political relevant groups equally in terms of their ability to affect the electoral outcome? The metric of political influence we use is the probability that a voter is decisive in an election. We develop an estimator of this probability for voters of each of the relevant groups for a given plan from (aggregate) historical election data. We, then, provide some examples from actual redistricting case