Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Political Science

First Advisor

James C. Garand


This dissertation addresses a simple question: Why are some members of congress very responsive to the ideological preferences of their constituency while others appear to ignore their constituency? Although numerous scholars have explored this issue, they have consistently: (1) limited their analyses to the influence of member characteristics while ignoring the potentially significant influence of constituent factors and (2) failed to test a single comprehensive model of congressional responsiveness that considers the full range of member and constituent factors hypothesized to influence variation in congressional responsiveness. This dissertation directly addresses these two problems by developing and testing a comprehensive model of congressional representation that focuses on both the influence of constituent and member conditions on variation in congressional representation. Specifically, I examine the influence of electoral margins, congressional retirement, electoral proximity, congressional seniority, constituency homogeneity, and constituency political engagement. My analyses suggest that constituent conditions play a significant role in explaining variation in congressional responsiveness. In particular, I find that members from relatively homogeneous and politically engaged constituencies are significantly more responsive to the preferences of their constituents. I argue that this is because relatively homogeneous constituencies send relatively consistent signals regarding their preferences, while politically engaged constituencies are more likely to make their preferences known and punish members of congress who are not responsive to them. Surprisingly, I find that the influence of member conditions on variation in congressional responsiveness appears to be minimal and their influence varies between chambers of congress. Ultimately, I conclude, much of the variation exhibited in congressional representation is not random, but rather systematically associated with various member and constituent characteristics. Moreover, constituent factors play a vital role in this process and future research by congressional scholars must recognize this role if they are to fully understand the factors that influence congressional representation.