Abstract

<jats:p>In the infancy of a science the use even of fairly crude methods of analysis and description can produce surprisingly large increments of knowledge if new perspectives are brought to bear upon available data. Such perspectives not infrequently require both a combination of methodologies and a critical appraisal of the limitations of each. The emergence of American voting-behavior studies over the last two decades constitutes a good case in point. Studies based on aggregate election statistics have given us invaluable insights into the nature of secular trends in the distribution of the party vote, and have also provided us with useful theory concerning such major phenomena as critical elections. Survey research has made significant contributions to the understanding of motivational forces at work upon the individual voter. As it matures, it is now reaching out to grapple with problems which involve the political system as a whole.</jats:p>