Research on Partisan Symmetry
The SRR Project maintains a comprehensive, up-to-date list of published studies on partisan symmetry analysis. The studies are annotated with a summary and evaluation of each publication.
Publications on Partisan Bias (3)
A quantitatively sophisticated discussion of the methods used to estimate partisan bias, this article still operates at a slightly lower level of sophistication than the article below. The article does not spend much time discussing the definition of partisan bias or responsiveness, though it does apply the method to estimate partisan bias and responsiveness in the Ohio State House of Representatives between 1968 and 1990.
The most quantitatively sophisticated explication of partisan symmetry analysis, this article is notable for its brevity in developing the core statistical tenets of partisan bias calculations. Gelman and King use the method in order to estimate the partisan bias in lower house elections in Ohio, Connecticut and Wisconsin between 1968 and 1980.
The first article to relax the “uniform partisan swing” assumption in assessing partisan bias and responsiveness. King suggests that partisan symmetry analysis ought to draw district-level election outcomes from an underlying probability distribution, called the mean voter preference distribution; this step introduces stochastic variability into the model. Substantively, the article affirms that the concept of “vanishing marginals” – namely, that districts become significantly less responsive during the decade after redistricting – applies not just to U.S. Congress, but also to state legislatures.
Publications on Proposed Partisan Gerrymandering Standards Using Partisan Symmetry (2)
This amicus brief sets forth the argument that partisan symmetry defines fairness in the redistricting context. The brief notes that there are widely accepted methods for calculating partisan bias and that the Court should consider basing a partisan gerrymandering standard on partisan symmetry. The amici suggest several standards – subsequently expounded in greater depth in the article below – based on partisan symmetry, but they do not advocate a particular standard in the brief.
Describes the Supreme Court’s response to partisan symmetry analysis in LULAC v. Perry, and responds to the Justices’ concerns and misunderstandings about partisan symmetry. The article discusses several possible legal thresholds for prima facie unconstitutional partisan bias: 1) require plans to have as little partisan bias as practicable; 2) require plans to deviate from partisan symmetry by less than one seat; 3) require plans to have a bias that falls under a specific percentage point threshold; 4) require that redistricting plans must not translate a majority of votes into a minority of seats; and 5) require that a redistricting plan must not have a bias that is severe or greater than in the plan it replaces. As the authors note, “We suggest that one key element of such a process [for evaluating unconstitutional partisan gerrymandering] would be based primarily on prima facie evidence for violations of some form of partisan symmetry test, and rebuttal thereto based on claims of competing and overriding legitimate considerations.”