Partisan identification had clearly allowed voters to organise information to allow them to make sense of the political world.
Non-partisans generally had a lower level of political interest. Russell Dalton argued that, with better education and an increase in the information available to voters through the media, a process of cognitive mobilization was taking place, especially among younger voters. Voters were now able to develop views on issues and the parties’ stances on these would influence their voting decisions (R. Dalton, The Journal of Politics, Vol. 46 No. 1, 1984).
If people were developing greater knowledge of politics and party policies then political scientists concluded that voters’ views of issues would become more important in determining how they voted.
Downs’s spatial model was used, not with just a left-right dimension but with a number of dimensions measuring where people put themselves in relation to issues where there were opposite alternative policies.
The assumption was that, if people understood their position on the issues that they saw as most important and understood the parties’ positions on these issues, then the closer the two were together, the more likely the voter was to choose the party.
Post-election surveys of British voters showed that by the 1980s they were much more likely to be able to name two issues that affected their vote (see Mark Franklin Electoral Studies Vol. 4 No. 1, 1985 for an explanation of issue voting theory). A modification of issue voting ideas was the directional voting theory of Rabinowitz and Macdonald in 1989 (see above).
The main problem with using issues in electoral studies was that they were not actually very good in explaining overall voting patterns. In the 1987 and 1992 elections Labour was seen as preferable to the Conservatives in the issues that voters saw as most important, such as health, but it was the Conservatives that won the elections.
Converse had found back in 1964 that voters mostly did not have a clear view of issues and would often answer the same question about their views in different ways at different times and also changed their answer to fit with the views of the party they supported. They did not have a coherent ideological view that structured their attitude to issues. (The nature of belief systems in mass publics, in Ideology and Discontent, ed. D. Apter, 1964).
Despite increases in information availability there is still doubt as to how clear a view of issues people have.
A new approach to issue voting comes from Dan Stevens who argues, taking the EU as an example, that issues do not obviously affect an individual election but as party elites slowly shift their position on issues, so the voters also slowly adapt their view of what the party stands for and this has an effect on their support for a party as well as their view of the issue (European Journal of Political Research Vol 52 No 4, 2013).
The problems with issue voting had already led Donald Stokes to talk about the important of valence issues, in contrast to the positional issues that the issue theorists had tended to use. Elections are increasingly about:-
- how impressive the party leaders are
- whether a party can achieve economic growth, low inflation, low unemployment and low interest rates
- how well they would run public services
These are valence issues that everyone thinks are important rather than issues on which there are opposite points of view such as greater European integration vs withdrawal from Europe (though a valence issue might be which party would negotiate the best for Britain within the EU). For the parties that have been in government, voters will reach a judgement on how well they have done in the past as well as how well they might do in the future, or how successful the Prime Minister has been as well as how good a new party leader would be as Prime Minister.
This version of rational choice theory is now the most favoured explanation of voting behaviour in Britain.
The book, Electoral Choice in Britain by Harold Clarke, David Sanders, Marianne Stewart and Paul Whiteley, 2004 take the valence voting point of view. It used the British Elections Studies from 1964 to 2001, opinion polls and other data and found that the valence model provided a better explanation of voting behaviour than either the social cleavages model or the issue voting model.
They also found that people’s identification with parties was not always lifelong but that people often supported a party for maybe three elections then switched to another party and changed their party identification in doing so.
Nevertheless, party identification still helped voters to see their chosen party’s policies favourably. Although Geoffrey Evans and a few others have kept up a rear-guard action in defence of the continued importance of class voting, often by using a more sophisticated division of classes than just middle and working, they are in a minority at present (G Evans, Annual Review of Political Science Vol. 3 No. 1, 2000).