Partisan Agenda Control and the Dimensionality of Congress

Abstract

Recent scholarship has questioned the familiar characterization of congressional voting as unidimensional (Crespin and Rohde 2007; Roberts et al. 2008). We build on these efforts by showing how a simple type of agenda control can suppress a latent second dimension. We argue that attempts to keep the majority party unified can lead to the estimation of more vertical cut lines and prevent the revelation of ideological differences along a second dimension. We evaluate this argument by examining the relationship between cut line angles and various measures of party control for all recorded roll call votes in the House of Representatives from 1875 to 1997. We also examine the relationship between measures of a second dimension and majority party unity in congress level regression. Our findings help to explain why a single dimension might explain a large amount of the variance in voting data when latent ideology is multidimensional.