"Spatial Constraints on Gerrymandering: A Practical Comparison of Methods" will be presented by James Saxon, postdoctoral fellow at Harris School of Public Policy at the University of Chicago.
Saxon studies compactness as a tool for districting reform (gerrymandering), the graph structure of American neighborhoods, and competing definitions of rurality--constrained clustering algorithms.
Compactness has long been proffered as a constraint on gerrymandering of legislative districts.
Along with equality of population and contiguity, both Congress and the courts have appealed to it as a `traditional districting principle.' But a surfeit of definitions has stymied its adoption and application: which one to use?
By implementing a flexible automated districting procedure, Saxon evaluates the practical impacts of these definitions - the seat shares that they imply for the two parties and for racial and ethnic minorities. Using a collection of historical districts, he measures the consistency of compactness scores and ranks, among the diferent definitions.
Saxon's paper argues that for automated districting, a choice among compactness measures does not affect many of the most critical, practical outcomes. For enacted Congressional districts, the many existing measures of compactness are markedly consistent. Compactness would be an effective legislative remedy to gerrymandering.
Read more about Saxon's project, Gerrymandering and Compactness.