voting behaviourthe decision-making processes and the social factors influencing patterns of voting.
Studies of voting behaviour have been of four main types: constituency studies, nationwide studies, cross-national studies, and those focused on particular categories of voters or the political implications of particular class locations. The seminal studies which influenced most later work were by LAZARSFELD et al. (1944) and Berelson et al. (1954) in the USA. These studies established the importance of socioeconomic variables such as socioeconomic status, religion, age, and gender as determinants of voting behaviour. They also made clear the part played by ‘group pressures’ and OPINION LEADERSHIP as influences on voting behaviour. The dominant paradigm which resulted was that most voting behaviour could be accounted for in terms of the PARTY IDENTIFICATION of voters, which for most individual voters was assumed to be relatively stable.
This model also informed the important study of the British electorate by Butler and Stokes (1969). These researchers painted a picture of voting behaviour in Britain in which, as well as being influenced by CLASS, voters tended to ‘inherit’ their party identification from their parents, where this had been 'strong’, was shared by both parents, and was for main parties. With increased volatility of voting behaviour (‘an erosion of partisanship’, LIFESTYLE politics, etc.) this model of voting behaviour is now more limited in its scope (see also CLASS DEALIGNMENT). Since many focused studies of voting behaviour and political attitudes (e.g. McKenzie and Silver, 1968, on WORKING-CLASS CONSERVATISM, and GOLDTHORPE and LOCKWOOD, et al., 1968b, on AFFLUENT WORKERS) assumed the predominance of‘class-based voting’, seeking to explain departures from this, an overall decline in ‘class’ voting also has implications for the interpretation of‘cross-class’ patterns of voting, which are longer exceptional. See also PARTY IMAGE, POLITICAL ATTITUDES, STABLE DEMOCRACY, MIDDLE-CLASS LEFT.