dividing1 - Dividing Light from Dark: Quantitative...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
Dividing Light from Dark : Quantitative Standards for Detecting Gerrymanders Micah Altman, Harvard University Michael McDonald, University of Illinois, Springfield DRAFT VERSION: 9/1/2000 ( Not for citation.) Introduction: The Political Decennial Pastime “It ’s tough to make predictions, especially about the future.” 1 In the area of redistricting, one is tempted to add, as a corollary, “it ’s tough to determine legislative intent, especially where no one meant for that to happen.” Despite these difficulties, predicting the consequences of redistricting is a decennial pastime, providing debate for scholars and advantage for politicians. The courts, following closely, engage in the intimately related activity of divining the intent of the legislature who created these district plans. In the area of redistricting intent and effect are intimately intertwined. Lacking “smoking gun” evidence, courts and scholars often look to the predicted effect of a redistricting plan in an attempt to divine the intent behind it In a book that accompanied the last round of redistricting, Mark Rush brought the predictive enterprise into question. Using a simple regression model to analyze the relationship between registration and voting in towns in Connecticut and Massachusetts, Rush argued that redistricting makes little difference in partisan terms, because the building blocks of a gerrymander are unstable: “Partisan blocs . .. appear to be in a constant state of flux.” (Rush 1993, 128). 2 In contrast, Gelman and King [1994] find that redistricting is quite predictable, and use simulation to predict the partisan bias and responsiveness of a particular plan. Unlike Rush's method, which is applied directly to registration and voting in individual precincts, Gelman and King apply their model to the district-level data to simulate the seats-response curves for hypothetical elections. The shape of the hypothetical seats-votes response curve associated with a 1 This aphorism is usually attributed to Yogi Berra. Mr. Berra, however, in personal communication (July 25, 1997) claims that this is a corruption of his better known statement “The future ain ’t what it used to be.”
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
particular plan is often thought to reflect partisan intent. For example, a plan that would award the Democrats 70% of the seats if they obtained 51% of the votes, but would award the Republicans with fewer seats, if they captured this same share, is often said to be biased against the Republican. This is usually regarded as (not incontrovertible) evidence of intent to gerrymander, or as a basis for comparing competing plans (see, e.g., Gronke & Wilson 1999) Responding to Rush and to King & Gelman, Kousser (1996) shows that the partisan results of an election can be more easily, and, he claims, more accurately, predicted using only partisan registration data. Using district registration totals and a unique variant of the ecological regression, he correctly predicts the district winner in 90% of the Congressional and legislative
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 26

dividing1 - Dividing Light from Dark: Quantitative...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online