On Measuring Partisanship in Roll Call Voting: The U.S. House of Representatives, 1877-1999

One way in which parties in the U.S. House of Representatives might affect legislative outcomes is if they pressure some of their members into voting the party line, when those members would prefer not to. How well do traditional measures of party voting, such as the Rice index of party difference, reflect party pressure? All traditional measues sufer from a significant and well-known problem--namely, they increase in size not only when parties devote more resources to influencing their members' votes but also when preferences within parties simply become more homogeneous. In this paper we disentangle these effects within a simple multidimensional spatial model of voting.