Fair Division in Political Redistricting

Xiao Wang

Document Type

Thesis

Publication Date

2012

Disciplines

Computer Sciences | Mathematics

Abstract

Redistricting is the process of dividing states into congressional districts of equal population according to the decennial census. How the boundaries of districts are drawn will largely affect how many districts can be won by a party. The Constitution gives state legislatures the discretion to draw those boundaries. Thus a party in control of the legislature will likely bring about a voting outcome favorable to them by manipulating those boundaries. However, the redistricting law does not say anything explicitly about how the represented district should be determined geographically. The purpose of this paper is to bring about a fairer redistricting plan by quantifying the fair division of districts by using Computer Science and Mathematics. Instead of visually estimating the compactness of a congressional district, this paper describes new algorithms for computing the size of the largest inscribed circle and the smallest circumscribed circle and uses the ratio of those two as a measure of compactness. In addition, this paper defines a method for determining a compactness threshold for a qualified congressional district in a state by calculating the average compactness among all possible districts generated by census tracts in the state. Finally this paper introduces a novel protocol for creating a fair electoral map.

Comments

Advisors: Gary Brown, Lynn Ziegler

Share

COinS