The Coefficient of Party Influence

In a 2000 article in American Journal of Political Science,James Snyder and Tim Groseclose develop and apply an innovativemethod for detecting and estimating the frequency and magnitudeof party influence in congressional roll call voting. This paperpresents a framework for assessing the coefficient that theauthors interpret as "party influence." The analysis revealsthat, and shows why, the coefficient manifests two troublesomecharacteristics. The coefficient cannot discriminate betweendisparate types of party influence because the mapping betweentypes of partison influence and signs of the coefficient isnot one-to-one. Similarly, the coefficient has a responsivenessproblem because a marginal increase in one party's influencecan cause the estimate of the coefficient to increase, decrease,or remain constant. Because the literature on parties in Congressemphasizes majority-party strength, the inability of the coefficientto isolate party-specific effects is a serious drawback in theongoing hunt for genuine party discipline.