Electoral Nationalisation,Dealignment and Realignment: Australia and the US, 1900-88

Has the behaviour of Australian voters become more 'nationalised' over time? Kemp (1978) and to a lesser extent Aitkin (1968) conclude that it has. Using more precise and comprehensive measures of nationalisation, a more standard (variance-components) statistical method and longer and directly comparable series of data, I qualify this conclusion heavily and place it on a firmer theoretical footing. I find that voting behaviour has generally been more spatially uniform in Australia than in the US, and that a sustained homogenisation-but not a 'nationalisation'-has been underway since the realignment of the late 1930s to early 1940s. Exceptionally and continually high levels of party identification, together with class-based alignments (and realignments) of electoral forces, appear to underlie the relatively high spatial uniformity of voting behaviour in Australia.