Talk Talk raises the online voting myth

Another polling day, another chance for internet firms to push online elections.

According to Iain, a YouGov poll for TalkTalk claims that 50% of people would be more likely to vote if they could do so online.

As usual, this poll is complete bollocks*. Apart from the uber-geeks, no-one is going to vote purely because they can do so online. They will do so because they are inspired to vote for (or against) a particular candidate or party. All internet voting can do is to make voting a little bit simpler.

When I worked for the Electoral Reform Society, we used to get polls like this all of the time. I seem to recall one which claimed that the number more likely to vote online was 90% so at least the trend is in the right direction.

I have two fundamental concerns about this sort of statistic. First, it doesn't measure how likely they are to vote with or without online elections. Simply raising their tendency to vote from 'absolutely no way' to 'almost certainly not' doesn't get us very far, but they will fill the criteria that Talk Talk needs for their poll.

Second. It's just not borne out by the facts. There have been a few trials with online elections in the UK since 2000. Cities including Sheffield and Liverpool have tried it. Whilst the technology seemed to work ok, there was no magnificent rise in turnout. In fact, in those areas trialing e-voting (of various kinds) turnout actually rose by just 1.5% in 2004. Sounds like a bit of an improvement? Well not when you compare it with those areas where there were pilots of any kind where turnout rose by 2.5%. So you could claim that e-voting actually lowers relative turnout.

And what is more, e-voting is massively expensive. For an average local election, it costs about £2 per vote cast. For an e-election, that figure rises to about £102. Great value for money.

While the technology worked ok in the trials, that is not to say a more widespread use would not attract hackers of all descriptions - from those who want to rig the result to those who simply want to wreck the whole thing. No computer system is completely safe and a UK general election (for instance) would be a high profile target. And because it uses black box technology, elections are always going to be treated with suspicion by more than just the nutty conspriacy theorists.

*note to YouGov: I'm not criticising your polling abilities