Works in Progress

"Where and When to Use Areal Interpolation Matching Relative to Spatial Audits" with John Curiel. (Revise and resubmit at State Politics and Policy Quarterly)


Amos and McDonald (2020) demonstrate the process of conducting spatial audits of residential addresses using hierarchical geocoding to match individual voters to legislative districts. Hierarchically geocoding residential addresses is the ideal method for matching purposes. However, cost constraints can limit its applicability for many researchers. In this letter, we make use of the data produced by Amos and McDonald to measure the extent to which one can correctly assign individuals to their state legislative districts using areal interpolation matching methods–centroid, geographic overlap, and population overlap matching. We test the accuracy of these three matching methods using the smallest unit of geography publicly available–the ZIP Code. We find that when predicting onto hierarchically geocoded data, these matching methods can ensure 90 percent accuracy of assignment when an individual’s ZIP code is split between fewer than 1.3 effective districts. This is most true when using population overlap matching. These results indicate that researchers can more efficiently conduct similar analyses by first using areal interpolation, reserving hierarchical geocoding for observations located within areas where districts divide lower-level geographic units. Our results extend the work of Amos and McDonald by identifying where shortcuts to hierarchical geocoding can be made while maintaining accurate data.

"Surrogate Representation in the United States House of Representatives." (Under review)


While geographic-based representation for members of the U.S. House of Representatives is constitutionally mandated, the "lines" of American politics have become blurred. Citizens contact, volunteer, and donate to members of Congress from across the country despite having no territorial relationship with them. Surrogate representation—the representation of Americans by legislators outside their legislative district—is an understudied phenomenon that offers insights into how, and why, these relationships form. In this paper, I offer a psychological theory of representation where legislators make group-based representative claims that are accepted by citizens across political boundaries—thus facilitating surrogate representation. Using campaign contributions to U.S. House members from 2012-2018, I demonstrate that surrogate representatives see a significantly larger percentage of their total constituencies comes from outside their congressional district. The evidence presented here challenges the standard account of representation in American politics and calls for a more robust examination of surrogate representation in contemporary politics.