The long shadow of attitudes: differential campaign effects and issue voting in EU referendums

Abstract

Can voters be persuaded by referendum campaigns? This article develops a theoretical model that synthesises the existing literature on campaign effects and issue-voting by arguing that the strength of pre-existing attitudes conditions voter receptivity to campaign arguments, thereby also determining their eventual vote choice. Using original panel data for the 2015 Danish opt-out referendum, there is evidence that attitude strength matters for whether voters are responsive to persuasion during campaigns. The article finds that voters with the most strongly-held attitudes felt well informed and certain about the consequences of the vote even before the start of the campaign, whereas voters with moderately-held attitudes are found to be more prone to believe those campaign arguments that are consistent with their EU attitudes, changing their vote intentions accordingly. Finally, voters with weakly-held attitudes were equally persuadable for the No and the Yes side of the campaign, but they are also the least pre-disposed to pay attention to campaign messages. The conclusions discuss the broader implications of the findings for our understanding of EU referendum campaigns.

Keywords:

• EU referendums
• issue-voting
• motivated reasoning
• campaign effects

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 We therefore do not necessarily expect the same learning dynamics to be at play in ‘repeat’ referendums held after a first no vote, as in Denmark in 1992/93, and Ireland in 02/03.

2 Attitudes towards the EU are not the only factor impacting voter choice, with other factors including risk propensity (Steenbergen and Siczek 2017) and the role of emotions (Garry 2014) also potentially shaping how voters process information and make choices.

3 See Online appendix 2 for more information about the context and the campaign of the 2015 JHA referendum.

4 This can be due to news coverage of EU affairs, but also because of past EU referendums and EP elections. There is evidence that most voters in the Danish context hold stable and consistent attitudes towards the EU (see Hobolt 2009). Theoretically, there is the potential that voter attitudes towards the EU could be affected by information in a referendum campaign, raising the risk of endogeneity. However, in our panel analysis we found stability in the underlying EU attitudes of voters (see Online appendix – A1).

5 In the period covered by the first panel sample, only 10% of front-page stories dealt with EU-related issues, whereas the figure increased to 12.5% from four weeks to two weeks out, and 16% in the final two weeks. The newspapers where a media content analysis was done were the two main Danish dailies (Politiken and Jyllands-Posten), and the most circulated tabloid daily (BT).

6 Note that confidence about how one will vote is not the same as certainty about underlying attitudes.

7 In this type of attitudinal question, there is some ambiguity about whether respondents chose more extreme answers in relation to questions such as ‘is the EU a good thing’ because they believe that the EU is really a very good thing, or because they are expressing the strength of their attitude. The implication of this is that for voters expressing sincere preferences based on positionality, extremity is only an indirect proxy for attitude strength, whereas for other voters, extremity is a direct measure of attitude strength because this is the question that they were actually answering.

8 With the existing opt-out, joining the supranational legal acts related to the common asylum and immigration policy was impossible. In contrast, with the proposed opt-in, a majority in parliament could at a later date decide to opt-in to any of these supranational acts. Many Danish voters (especially on the right) strongly opposed joining the common asylum and immigration policy.

9 First, while British voters had not been asked to vote in an EU referendum since 1975, EU issues were discussed frequently in the media – especially in the years prior to the referendum due to the eurocrisis and 2015 refugee crisis. Additionally, Eurobarometer responses to questions on the merits of the EU suggest that UK voters do have relatively stable (and strong) underlying attitudes towards the EU.

Additional information

Notes on contributors

Derek Beach

Derek Beach is a Professor of Political Science at the Aarhus University. He does research on EU integration, including crisis decision making and voter behaviour, and on research methodology. His most recent book is Process-Tracing Methods (University of Michigan Press). [[email protected]]

Daniel Finke

Daniel Finke is a Professor of Political Science at Aarhus University. His interest comprises EU politics, international organisations and legislative studies. His present research is on coordination inside the EU Commission and agenda setting in the UN General Assembly. [[email protected]]