What do we think of VoteMatch?

May 30th, 2009

Communicating manifesti to the masses?

A couple of weeks ago I attended the launch of VoteMatch at the Apple Store on Regent’s Street. A glitzy affair, the evening was compered by Steven Fry (replete in Black Tie) and Simon Hoggart of the Guardian, with Iain Dale giving a demonstration of the tool.

VoteMatch is an online application (accessable as an iPhone app) which asks you thirty questions, and directs you to which party most closely matches your responses. It has been (re)launched with a view to educating the public about where the parties stand in the run-up to the European election. It doesn’t seek to tell people how to vote, but rather seeks to allow for greater awareness and understanding of the parties’ positions, and to encourage people to learn more about the EU. Launched by Unlock Democracy Its motives are, I think, honourable and the idea is a Good Thing.

That said, I have some personal quibbles. The decision was made to focus entirely on European politics – ie What the EU should and should do/be. Whilst that is perhaps theoretically sensible, it means some disconnect between what people will actually vote upon in the European elections (including national issues or areas of competance not yet under the direction of the EU) and how their score matches that of the parties. I also retain a slight suspicion that those not weighting the issues they care about will get a very close result between several parties – perhaps reflective of how little difference there is between many parties on EU affairs. However, this is perhaps understandable, and I have no doubt that the General Election widget to follow will allow for results that are closer to how people will actually vote, rather than simply being educative of where the parties stand.

My bigger problem is with some of the questions. I have some experience of designing polling questions, and there are several that I though were (if not unanswerable) then perhaps not best designed to elucidate differences in political position given the Agree/Disagree/No Opinion format. Iain Dale picked up on one in particular: “In a recession, national governments should have the freedom to subsidise their own businesses.” – he thought national governments should have the freedom, but he is absolutely against subsidies.

I had more problem with two other questions: “Allowing workers to work more than 48 hours a week, even voluntarily, is open to abuse by employers” – of course it is ‘open’ to abuse, but I don’t think it would be widespread enough to justify the WTD over which Britain has just surrendered its opt-out. Similarly, “The President of the European Commission should be elected by the European Parliament.” seemed problematic – answering ‘Disagree’ could be either a belief in an unelected President of the Commission, or a belief in a directly-elected President (by the people). An ‘Agree’ answer could be a vote in favour of the European Parliament having that power, or an aspiration for directly-elected President and this as a first step.

The worst example of a mixed issue question was: “The EU should make it possible for any EU citizen go to any hospital in the EU and have the doctor treating them able to access their health records instantly.” – I am absolutely in favour of being able to visit any doctor in Europe and ready to press ‘Agree’ until the question becomes less about universal pan-European access to healthcare, and veers off into the merits (or lack thereof) of a pan-European health database, which I categorically oppose.

Combining issues might have allowed VoteMatch to keep down to 30 questions (a maximum that people will engage in answering apparently), but that’s no good if it renders the answers non-sensical.

After thirty questions, you indicate which issue matter a lot, and which you care less about, followed by a list of the parties you would consider supporting (default being all). The result (at least in my case) was very surprising, and indicated a party I would never consider voting for – perhaps reflective of my own confused political stance!

Personally, I think the questions need revision to ensure that each pertains to only a single-issue, and that the answers should allow for more than ‘Agree’ and ‘Disagree’, especially in questions which suggest such a contentious middle-ground (the election of the President of the Commission). For clarity of questions, I would recommend the famous Political Compass as a model. The survey could have included a Left-Right axis (ignoring Political Compass’ innovation) and superimposed a pro-EU/anti-EU axis – this I think would have given more meaningful results.

That said, anything that encourages people to learn more about their elections or about the political system is to be encouraged – if it should manage to re-engage the 70% of the electorate who are unlikely to vote on June 4th, then that would be a good thing too: unless you have money on the turnout markets.

UPDATE: Chris Took has posted another of these simulators on last night’s thread, called EUprofiler which can be found here. This answers a lot of the questions I raised in this article, though I still came out with the (unthinkable) result that I got from VoteMatch!

UPDATE: Peter Facey from Unlock Democracy, who created VoteMatch, has answered a question about where the parties’ positions were taken from (largely from questionnaires – some parties, like the BNP, refused to take part). His comment can be found at number 145. Thanks Peter.


NOTE: Over the next week, we’ll be announcing plans for the European Election Results night on Sunday 7th June 2009 – we’ll be covering all the results from across the Continent, with particular focus on the UK, and trying out some experimental tools and widgets.

If you are planning to reference the results (as a sub-agent at a count, or in some other capacity) on Twitter (I’m on @Morus1516) I’d recommend using the hashtag #eu09 (plus the #uk hashtag etc).

Comments are closed.