The doctoral dissertation research project will develop a new general method to measure the impact individual districts have on estimates of advantage associated with different configurations of legislative units. The doctoral student will empirically examine the relationship between existing indicators of district boundary configuration and new measures that he will develop and test. He will evaluate the fundamental assumptions common in the analysis of electoral advantage, and he will assess the validity of different measures as well as fundamental statistical assumptions in electoral modeling techniques in order to determine the robustness of estimates to spatial dependence. This project will expand basic understanding of electoral dynamics. It will improve the accuracy, fidelity, and usefulness of electoral models, and it will provide new tools for analyzing legislative district plans, thereby contributing to societal debates about the configuration of electoral unit boundaries. The simulation techniques developed in this project also will be applicable for sensitivity analysis of other kinds of spatial models. As a Doctoral Dissertation Research Improvement award, this award also will provide support to enable a promising student to establish a strong independent research career.
The doctoral student undertaking this project will extend existing symmetry measures to examine individual district contributions to the estimated asymmetry of a redistricting plan. These new local measures of district influence will be used to analyze the impact strangely shaped districts have on a plan's overall estimated advantage for one group over others. Two fundamental components of electoral models will be assessed. The student will interview redistricting officials and stakeholders to guide and inform the development of new measures of electoral advantage in order to ensure they are meaningful and useful in practice. He also will evaluate the impact of geospatial dependence in estimating the relationship between the share of seats a group wins in a legislature and the share of votes the group receives across all districts. He will assess legislative election models for potential spatial misspecification, and he will assess the degree to which an increase in aggregate support for a political group is modeled by a set of independent, randomly distributed district changes. The student also will examine the spatial structure of electoral swings to improve the models on which the new local influence measures depend.