It’s the question, stupid: Democracy experts respond to the EU referendum question proposals

The Electoral Commission has this week set out its advice on the wording of the question in a proposed referendum on the UK’s membership of the European Union. In this post, leading democracy experts share their views on the alternative questions, considering the implications for the campaign and drawing on evidence from previous referendums in the UK and elsewhere.

The question included in the EU (Referendum) Bill – a Private Members’ Bill brought by James Wharton MP and currently before Parliament – is ‘Do you think that the United Kingdom should be a member of the European Union?’, with answers ‘Yes or ‘No’.

The Electoral Commission’s new report – following consultation on the Wharton proposal – recommends two alternatives:

  • The first retains the traditional format: ‘Should the United Kingdom remain a member of the European Union?’, with answers ‘Yes’ or ‘No’.
  • The second moves away from the yes/no format: ‘Should the United Kingdom remain a member of the European Union or leave the European Union?’, with answers ‘Remain a member of the European Union’ or ‘Leave the European Union’.

imageDr Chris Gifford, Head of Behavioural and Social Sciences at the University of Huddersfield
A key argument made by hardline Eurosceptics for a UK ‘in/out’ referendum is that it is what the people want. They regularly refer to opinion polls in favour of a referendum and use petitions to claim extensive public support. The implication is that the UK’s membership of the EU is a question of fundamental importance to the British people.

The Report for the Electoral Commission on the intelligibility of  the referendum question tells a rather different story. Participants had little understanding of the EU. Some demonstrate no knowledge whatsoever, including not knowing that the UK was a member, which is consistent with polling evidence that shows the EU is not a salient issue for most voters. This is not to suggest that they are unconcerned but, as they reasonably tell researchers, they are simply ill-informed; perhaps unsurprising considering that much of the press continue to treat the EU as a foreign power intent on destroying British democracy and ‘our’ way of life.

In effect, the report challenges not just the proposed referendum question but the very legitimacy of holding a referendum. Citizens may be coerced in to making a decision on an issue of national importance on which they have been chronically misinformed, or do not consider of direct relevance to their lives. The quality of the campaign debate will be crucial if the result is to hold.


P.Whiteley 4Paul Whiteley, Professor of Government, University of Essex
There is always controversy over the wording of referenda on constitutional issues.  This happened in the referendum on the Alternative Vote system for Westminster elections in 2011 – so much so that the wording of the question was changed.  Equally there has been a lively debate about the wording of the referendum on Scottish independence due in 2014.  The Electoral Commission are nervous about the question proposed in James Wharton’s Bill because it presumes that people know that Britain is an EU member. They would prefer alternatives which make no such assumption.

At first sight these distinctions appear to be fairly trivial, and it might be argued that people who don’t know that Britain is a member are unlikely to vote in the referendum in any case.  But the Electoral Commission may have a point.  The National Evaluation of Policy Monitor, an internet survey conducted every month at the University of Essex, asks the following question:  Overall, do you strongly approve, approve, disapprove, or strongly disapprove of Britain’s membership in the European Union?  In the September 2013 survey some 43 per cent of respondents approved of membership and 40 per cent disapproved.  The interesting statistic, however, is that no less than 17 per cent stated that they ‘did not know’, which is fairly high for opinion surveys.  Clearly, many of these respondents will have known that Britain is an EU member, but just couldn’t decide on whether membership was a good thing or a bad thing.  But some of them could simply not have known that Britain is a member, and so did not answer for that reason.

I think the Electoral Commission needs to run an internet survey in which different random subsamples of respondents are given the three different versions of the question.  If we had such data we could easily find out if the wording made a difference.  This is an important constitutional issue and so if the referendum goes ahead we need to get it right.


imageDr Jane O’Mahony, Senior Lecturer in European Politics, University of Kent
The advice of the Electoral Commission to Parliament on the proposed EU referendum question highlights the difficulties in holding referendums on extremely complex political issues such as EU membership.  When it comes to considering James Wharton’s Bill, Parliament would do well to heed the lessons learned from the Irish experience of EU referendums (since 1973 there have been nine EU referendums in Ireland, most recently in 2012 on the EU’s Fiscal Compact).

The Irish experience shows that levels of knowledge can have an impact on the final referendum result.  The famous (or infamous) poster slogan from the 2008 referendum on the Lisbon Treaty by a number of groups advocating a No vote, ‘If you don’t know, vote no’, illustrates nicely a trend in Irish EU referendums: rather than abstaining, those lacking in knowledge on the specific issue/question often choose to vote No.  Eurobarometer opinion poll data has shown the British public’s levels of knowledge of the EU (both objective and subjective) to be low (as the Electoral Commission also found in their own research). In this context, at the very least it is vital that the eventual question asked is as unambiguous and clear as possible.


imageDr Matt Qvortrup, author of Direct Democracy and adviser on referendum administration to the US State Department and Elections Canada
The Electoral Commission refers to the possibility of moving away from Yes/No referendum questions. This is a bad idea. Referendums are about yes and no questions. The beauty of referendums is precisely that they ask a simple question. A referendum is the people’s opportunity to veto a decision reached by Parliament or another elected body. And a veto requires the voters to have the option of choosing between yes and no.

To date there has been no research showing that the wording of the question plays a role in referendums. The only slight statistical effect on the outcome is the length of the question. As I show in my forthcoming book Referendums and Ethnic Conflict (University of Pennsylvania Press 2014), a longer question increases the probability of a no vote, though at most by one or two per cent.

Although question wording does not itself affect the outcome, allowing Parliament to set the question may create the impression of a bias. This is likely to predispose voters to saying no (we saw an example of this in Denmark in the early 1990s when there was a perception that the question was biased in favour of the government, as a consequence the government lost credibility, and indirectly support). For these reasons there is a strong case for the question to be set by an independent body.


imageTim Bale, Professor of Politics at Queen Mary, University of London
A referendum on the UK’s membership of the European Union is not going to be an end to all our problems, whichever way it goes.  If we vote to leave, the economic and diplomatic consequences are incalculable.  If we vote to remain, who honestly thinks that the Eurosceptics will simply accept it with good grace?  To quote the Clash, ‘If I go there will be trouble and if I stay it will be double.’

But how the question we actually end up voting on is worded probably won’t make that much difference to the result – unless that result turns out to be a lot closer than I suspect it will be.  That in turn will depend on who is in power when and if the referendum is held. If Labour doesn’t manage to hold its nerve and, in the run up to the general election, commits to a referendum should it win power, then all bets are off.  If, on the other hand, the Conservatives are returned to office, with or without the Liberal Democrats, the smart money is on them being able, without too much difficulty, to persuade the electorate to endorse whatever deal David Cameron manages to do with the EU.

That said, it’s important, of course, to ensure that the question put taps, as objectively as possible, into what the real issue is.  Best practice – and tradition – is to try to produce a Yes or No answer.  However, as the Electoral Commission has found, this could be difficult this time round.  It strikes me, then, that this may be one of those occasions on which we need to depart from precedent and provide people with a more explicit choice.  After all – and back to the Clash – in the end, the real issue is ‘Should I stay or should I go now?’


imageCas Mudde, Assistant Professor in the Department of International Affairs, University of Georgia (USA)
Academic studies have shown over and over again that how you ask a question influences what answer you get. Consequently, the formulation of the question of the British referendum on EU membership is of crucial importance. The Electoral Commission was right to advice against the original question, which was unclear about the current membership status of the United Kingdom. It is also unwise to use a question that leads to the answers “yes and “no,” as survey research has shown that (uninformed) people tend to favour “yes” over “no,” irrespective of the question, simply because they prefer to agree rather ‘than disagree. This makes the third option, ‘Should the United Kingdom remain a member of the European Union or leave the European Union?, with the answer options ‘Remain a member of the European Union’ and ‘Leave the European Union,’ the best, that is, the least biased, question.

This question should be asked without any ‘explanatory’ introduction. Whatever introduction is finally agreed upon, it will undoubtedly be both biased and confusing to some voters. There is no doubt that a large portion of the British electorate is ill-informed about the consequences of (continued) EU membership. This is a consequence of both the immense complexity of the issue (on which even experts will disagree) and the lack of a serious debate in the media. Only a long and balanced debate in the months preceding the referendum can overcome this lack of knowledge. It is crucial that the British government, and the European Union, either finance the campaigns of both camps or of none of the two camps. However, they should realize that (even) a relatively well informed electorate can vote against their preferred option – as the French showed in the 2005 referendum on the ‘European Constitution.’


imageDr. Stephen Quinlan is a Research Fellow at the University of Strathclyde, researching the Scottish independence referendum, following previous work on Irish referendums.
“My God – what a stupid question!” Pierre Bourgault, when told of the Québec separatist government’s choice of wording for the province’s secession referendum in 1980. Bourgault, a strong separatist himself was incandescent about the question being put to Quebeckers, which was lengthy (133 words), and asked voters to give the provincial government ‘a mandate to negotiate” ‘economic association’ with the rest of Canada. The question was widely pilloried as ambiguous about the consequences of a ‘yes’ vote, and the proposal went down to defeat, in large part because the debate concentrated on issues beyond the specific question that was being put to voters. The Australian referendum on retention of the British Monarch as Head of State in 1999 is another instance of a referendum question that came in for strong criticism. This was primarily because the question asked Australians not only to decide upon retention of the monarchy but also simultaneously to choose the selection mechanism for a new Australian President. Here too the referendum proposal went down to defeat, with the referendum question considered to have influenced many voters’ vote choice.

The above two examples bring into sharp focus the crucial importance of referendum questions. Ambiguity, overly technical/legal constructions, and simultaneous decisions can lead to confusion among the electorate, which can result in the debate becoming less focused on the issues that voters are being asked to decide upon and can also have implications for voter turnout. Whatever question is put to British voters in the event of a referendum on the country’s membership of the EU, architects of the question would be wise to keep in mind the above experiences of referendums in other countries and endeavor to ensure clarity in the question that is put to the electorate.


imageDr Lee Rotherham, Research Fellow at the TaxPayers’ Alliance
Ah: conditional tenses, jussive subjunctives, litotes and all the other extravagances of language that our post-Grammar and post-grammar schooling system seemingly no longer exports. The Electoral Commission is right on the need to be precise about the wording that is chosen for a referendum. It needs to be unambiguous, and it also needs to be impartial – not like the infamous 1936 Rhineland question. But in this instance, it’s generating a distracting dispute about whether it’s better to drive home in a Toyota or a Ford.

In the case of the Wharton wording, some common sense could usefully be applied. The question might imply a future direction not currently undertaken. But given it adopts a clearly unpartisan baseline of expression, dealing with first level principles, and given the author will have received expert legal advice from the Clerks of the House, we might possibly let the margin slide. Anyone who is unaware the EU exists is so far off the grid they probably live in a disused shaft tucked into some remote river bed. Meanwhile, anyone who does know, but thinks the UK is still part of the European Free Trade Association, will likely have views on wanting to join in 2015 even if it’s forty two years late.

Note: This post represents the views of the authors, and not necessarily those of Democratic Audit of the London School of Economics. Please read our comments policy before commenting. Shortlink:

Home page image credit: Justinc (CC BY-SA 2.0)

Similar Posts

Comments are closed.