Did you see the reports about the public getting keener to vote in European elections?

No, I didn’t either. But the odd thing is, there’s plenty of evidence that the public were keener to vote in European elections (which used a form of PR, natch) than previously.

The evidence is certainly patchy and incomplete, but the uniformity of the, often wrong, turnout gloom and doom stories seems to me to say rather more about the media’s fixed image (’turnout? must be down’) than about the actual evidence.

The key is to compare like-for-like data. For example, in several parts of England the last European election was run using an all-postal ballot, in which all possible voters were sent a ballot paper that they could then post back. Whatever the drawbacks of all-postal ballots (and there are many), one thing they do is to raise turnout. So if you have an all-postal ballot one time and then a traditional election the next, one would expect a fall in turnout – and that fall in itself doesn’t tell us anything about the public’s increasing or decreasing interest in the election or any change in their willingness to vote.

The other complication is that some of the time, in some parts of the country, European elections have occurred at the same time as local elections. As local elections usually have higher turnouts, putting the two together pulls up the European election turnout. Again, it means that you can’t conclude that the public has got more/less interested in the elections if, between the two you are comparing, the elections got combined/split.

One final complication: the local elections that took place on the same day as the European elections this year were not in exactly the same areas as those that took place on the same day as the European elections last time.

So it’s no surprise that the overall turnout picture is complicated and, even a couple of weeks after polling day, it’s hard to draw an exact picture.

There are though positive straws in the wind. First, of the eleven regions in Great Britain, three saw turnout rise on five years ago, three saw it fall and a further five are clearly special cases. Four saw it fall but had all-postal voting last time, yet not this time, whilst the fifth (London) had the London Mayor and GLA elections on the same day last time, but not this. That’s certainly not a picture of unmitigated joy, but imagine if the journalist cliche was ‘turnout is on the up’. Don’t you think something would have been made of these changes, rather than the near-uniform silence?

Now, you might object that these figures don’t fully allow for the changing pattern of where there were local elections. Fair enough, but then let’s take two clear examples. In both London and Wales there were no local elections this time, there were local elections last time, but there were none again the time before. And how does turnout compare?

London 1999 23.0%
London 2009 33.3% (+10.3%)

Wales 1999 28.1%
Wales 2009 30.4% (+2.3%)

The London figure is particularly striking and (as far as I’m aware) has gone completely unmentioned in election reporting. It’s all the more impressive as there is good evidence from political science research that the more reports say people aren’t voting, the fewer the number of people who vote. Just imagine what the increase might have been had the media story been about increasing turnout.

As I said, there is plenty of nuances to the overall picture, so if you’ve got any further figures to throw in that help make the actual picture clearer, please do post up a comment.