A Comparison of Constituency Influences in the Upper and Lower Chambers of a State Legislature

Abstract

T IS GENERALLY CONCEDED that the voting behavior of legislators is affected by both party and constituency influences. Some legislators identify themselves chiefly with their party, while others perceive their role as one of representing constituency interests.1 Many inquiries have been made into the relative merits of Burkean representation, accountability to constituency, and party responsibility, but few studies have bothered to investigate the actual extent of insurgency, constituency influences, and party voting in legislative bodies.2 The only published article to date which has examined the effect of constituency influences on party cohesion in the voting behavior of state legislators, is a study of the Massachusetts House of Representatives by Duncan MacRae, Jr.3 Working with roll-call votes on legislation which appeared to reflect socio-economic differences, MacRae found that legislators representing districts typical of those which generally elected members of their party on a home-owner occupancy index tended to display greater party loyalty than legislators elected by districts atypical of their party. MacRae also found that representatives whose previous election margins were narrow tended to cross party lines more often than representatives elected by wide margins. The study summarized by this paper produced similar findings regarding the 1957 session of the Pennsylvania House of Representatives, but, perhaps more significantly, found the patterns which described legislative voting behavior in the lower chambers in both Massachusetts and Pennsylvania to be inappropriate for describing voting behavior in the Pennsylvania State Senate. Variations in the voting behavior of members of the upper and lower chambers of a legislature should not confound readers of the American classics. The Federalist Papers discuss the probable behavior of members of the separate chambers and in so doing inadvertently suggest certain hypotheses capable of verification. These hypotheses will be examined later.