Robustness of DVC Scores to Alternative Density-Variation and Compactness Measures in Evaluating Redistricting Plans

• Activity Number: 164 - Social Statistics Speed Session
• Type: Contributed
• Date/Time: Tuesday, August 10, 2021 : 10:00 AM to 11:50 AM
• Sponsor: Social Statistics Section
• Author(s): Cameron T. Rider* and Thomas R. Belin
• Companies: C-FIT Redistricting (Coalition for Fairness, Integrity, and Transparency in Redistricting) and UCLA FSPH
• Keywords: redistricting; gerrymandering; seats-votes curve; compactness; sf file; shapefile

Abstract

Partisan gerrymandering has contributed to political polarization in the United States, but enforceable standards to protect against the practice remain elusive. As a one-number summary for evaluating redistricting plans, Belin, Fischer, and Zigler (2011 Statistics, Politics and Policy) proposed a “density-variation / compactness” (DVC) score, where plans with higher scores are expected to have fewer “safe seats”. Choices for input components in the framework motivate the present exploration of the robustness of DVC scores to alternative density-variation and compactness measures. In contrast to using mean absolute deviation and Reock compactness (area of district divided by area of circle encompassing district), as in Belin et al. (2011), we calculate DVC scores for multiple states using the standard deviation and a convex-hull compactness measure (area of district divided by area of convex region encompassing district, which yields larger numerical values than the Reock measure). We discuss implications of findings for the ability to provide valuable and timely information into the redistricting process and the prospect of protecting against partisan gerrymandering.