Published online by Cambridge University Press: 27 January 2009
One of the most striking features of British general election results is the large variation in turnout across constituencies. In 1983, for instance, the mean turnout was 72.6 per cent of the registered electorate but it ranged from 82.1 per cent in Carmarthen to 51.8 per cent in the City of London and Westminster. Moreover, these variations tend to be reproduced in successive elections – the bivariate correlation between constituency turnout in the 1983 and 1987 elections for all 633 mainland Britain constituencies is +0.94.
1 Crewe, I., Fox, A. and Alt, J., ‘Non-voting in British General Elections 1966–October 1974’, in Crouch, C., ed., British Political Sociology Yearbook, Volume 3: Participation in Politics (London: Croom Helm, 1977), pp. 38–109Google Scholar; Swaddle, K. and Heath, A., ‘Official and Reported Turnout in the British General Election of 1987’, British Journal of Political Science, 19 (1989), 537–51.CrossRefGoogle Scholar
2 Denver, David and Hands, Gordon, ‘Marginality and Turnout in British General Elections’, British Journal of Political Science, 4 (1974), 17–35CrossRefGoogle Scholar; Denver, David and Hands, Gordon, ‘Marginality and Turnout in General Elections in the 1970s’, British Journal of Political Science, 15 (1985), 381–8.CrossRefGoogle Scholar
3 Eagles, M. and Erfle, S., ‘Community Cohesion and Voter Turnout’, British Journal of Political Science, 19 (1989), 115–25.CrossRefGoogle Scholar
4 Mughan, Anthony, Party and Participation in British Elections (London: Frances Pinter, 1986).Google Scholar
5 It should be noted that parliamentary constituency boundaries were redrawn between 1979 and 1983. As a consequence, the 1979 marginality of 1983 constituencies can only be estimated for most cases. These estimates are based on The BBC/ITN Guide to the New Parliamentary Constituencies (Chichester: Parliamentary Research Services, 1983).Google Scholar
6 Initially, partial correlation coefficients were calculated for all variables thought likely to influence turnout. Variables whose partial correlation coefficients had a significance of less than 90 per cent were discarded.
7 See Lovett, A. and Flowerdew, R., ‘Analysis of Count Data Using Poisson Regression’, Pro fessional Geographer, 41 (1989), 190–8.CrossRefGoogle Scholar
8 We also undertook an Ordinary Least Squares (OLS) regression analysis which resulted in conclusions very similar to those reported here.
9 Poisson regression yields a ‘pseudo’ R2 statistic which measures the percentage deviance in the dependent variable reduced by the addition of explanatory variables. It is analogous to the R2 statistic in Ordinary Least Squares regression.