In two previous posts we discussed why asking people how they intend to vote in the general election is not enough to understand adequately how voters relate to political parties. One reason for this is that ‘vote intention’ does not reveal how one’s preference for this party compares to their preferences for other parties. The other reason is that we get little idea of how strong that person’s intention to vote for that party actually is. Both problems can be solved by asking voters, in addition to their vote intention, how likely or unlikely it is that they would vote for each party on offer, on a scale from 0 (‘very unlikely’) to 10 (‘very likely’). We use here the information from such questions as asked in the British Election Study at the onset of the campaign.
These questions allow us to gauge the potential support for every party in the electorate i.e. the proportion of voters who consider each party a sufficiently attractive choice that they may vote for them on polling day. As most voters have high preferences for more than one party [insert hyperlink to blog post 1], a part of each party’s ‘potential electorate’ overlaps with other parties. Parties’ competition for votes, therefore, focuses particularly on these groups of voters. It is unlikely that a given party would ever realise its entire voter potential (unless it had a particularly good campaign while its competitors had a particularly bad one), but nonetheless we can identify the potential support each party could conceivably obtain on election day, if everything went its way.
Based on this information, we can split each party’s potential pool of voters into five groups:
- Citizens with a very high electoral preference for the party in question, and low electoral preferences for all other parties (e.g. people who rate party X 8 or higher on the 0-10 scale, while rating all the other parties below 5). As electoral preferences tend to change only marginally during a campaign, it is unlikely that these ratings will shift to such an extent that these voters end up supporting another party. We can call this group the party’s strong core support.
- Citizens with a lukewarm preference for party X (say, with a rating of 6 or 7 out of 10) and low preferences for all other parties. Again, it is unlikely that the campaign will raise the low preferences for other parties to a level where they overtake the most preferred one, but the lack of real enthusiasm for party X means that these voters might end up staying at home on polling day. This group comprises a party’s weak core support.
- Citizens with a high electoral preference for party X, but who also hold approximately equally strong preferences for one or more other parties (i.e. they give party X and at least one other party a score of at least 8 out of 10). The strength of these people’s electoral preferences makes them almost certain to turn out and vote, but they are torn between different parties, only one of which they can actually choose on the ballot. Just because they say they will vote for party X in an opinion poll, therefore, does not mean that they will do so on polling day; they see party X as very appealing, but another party as very appealing as well, and they could easily switch between the two during the campaign. Conversely, this group could also contain people who do not state an intention to vote for party X, but nonetheless think very highly of it – giving party X the chance to win them over before the polls close. These voters we can call strong supporters subject to competition.
- Citizens with a lukewarm preference for party X (i.e. a rating of 6 or 7 out of ten), but who also have similarly lukewarm preferences for other parties. Like the previous group, these are potential voters who may switch to another party – but, given their support for either party is limited at best, they may still decide not to bother voting at all. These are weak supporters subject to competition.
- Finally, there are the remaining voters: people who have a fairly high preference for party X but who still have positive impressions of the alternative; or reluctant party supporters who say they will vote for party X on polling day, but do not do so because they find that party particularly attractive (for example, people who vote tactically for party X because they particularly dislike another party). These voters are not entirely immune to the possibility of switching their support for another party, but it would take something of an upset to make that happen.
If we apply this logic to tomorrow’s election, we can see how the potential support for the Labour and Conservative party’s breaks down. The table below shows the potential support for each party at the start of the election campaign, and also the proportion of that potential support which fits into the five groups described above.
|strong core support||46%||47%|
|weak core support||4%||5%|
|strong supporters, may still switch away||16%||14%|
|weak supporters, may still switch away||4%||6%|
|reluctant support and others||29%||28%|
|Share of all with a vote intention||33%||35%|
|support, may switch from other parties (as percentage of all voting)||6%||5%|
Each party has a similarly sized pool of voters currently intending to support it tomorrow, with 33% of voters saying they will support the Conservatives and 35% saying they will support Labour. Almost half of these groups constitute the parties respective core support i.e. 46% of people currently saying they will vote Tory, and 47% of those saying they will vote Labour, are almost certain to do so. The other half is less certain; they may be at risk of switching to another party (even if they see the Tories or Labour as a really good choice), or of staying at home because they are not particularly enthused by any party.
The two parties also have the chance to increase their support, however, based on the strong supporting potential switchers who are currently saying they will vote for another party. If either the Conservatives or Labour have a particularly good campaign today and tomorrow, they may be able to increase their pool of support by up to 6% and 5% respectively, producing a total vote share of around 40% for both. However, if we look at the size of those groups of voters currently saying they will for Conservative or Labour but who may yet switch to another party they find almost equally attractive (20% of each party’s current pool), we see that each party also has a great deal to potentially lose. It is perhaps unsurprising, therefore, that each party has fought a largely defensive, steady-as-she-goes campaign; they have a lot to lose, and far less to potentially gain.
Cees van der Eijk is a Professor of comparative politics, research methods and electoral studies at the Methods and Data Institute at the University of Nottingham. Stuart Fox is a PHD candidate who research focuses on political participation, alienation and apathy of young people in Western democracies.