Rise of the election machines

No, this post isn't about the dangers of electronic voting. TSN Gallup has conducted a poll on the use of election machines - i.e. online surveys that ask you various political questions and then recommend candidates and/or parties. Helsingin Sanomat reports on it here (fi) and Verkkouutiset offers more data.

About a third of people eligible to vote used election machines. That's up from about a quarter in 2003. 71 percent of election machine users voted. I wonder, does that mean that a sizable chunk of the rest would have voted if online voting had been possible? Overall the turnout was 68 percent. If my math is correct, roughly 35% of voters used election machines.

For their users election machines were the most important way of acquiring information about the election, narrowly beating TV news programs and newspaper articles. Lest you despair, users followed the election more closely than non-users using other information sources, too. For the population as a whole, election machines tied for third with election-related TV programs. They were behind news programs and newspaper articles, but ahead of candidate web sites, real life acquaintances, election ads in newspapers, and radio.

Interestingly, Internet forums - the category under which blogs are counted, at a guess - were the most important source of information for a big fat zero percent of voters in general. Among election machine users, three percent considered Interent forums as their most important information source. I think it's safe to say that at the moment Finnish political blogs are not a force with which anyone needs to reckon.

15 percent of users wound up voting for the candidate recommended by an election machine because they didn't have a favorite of their own. Three percent switched from a previous favorite candidate to a recommended candidate. An election machine helped one in five users to exclude a candidate. 53 percent of users voted for a candidate that no election machine had recommended. By my count roughly six percent of voters took their cues from an election machine.

I find it a bit odd that 47 percent of users voted for a candidate that was recommended to them by an election machine, but relatively few admit to following the recommendations. In my experience the candidate for whom I'm thinking about voting is almost never recommended to me as the top choice. Perhaps I know less about candidates' issue positions than the average user, but I suspect that respondents aren't being entirely honest about how important election machines were in their voting decisions.

I'm one of the people who used election machines to exclude candidates and I found that they work pretty well for that purpose. I also checked out a few candidates who I would have otherwise ignored. If memory serves, the candidate I voted for ranked high in at least one list of recommendations, but not at number one.

Users were more likely than other voters to opt for the National Coalition Party (31% versus 22%) and the Green League (17% versus 8%). The Centre Party (17% versus 23%) and the Social Democratic Party (12% versus 21%) were less popular among users than non-users. Presumably this is due to Coalitionists and Greens being more plentiful online than among the general populace.

The most popular election machines belonged to YLE (fi), MTV3 (fi), and Helsingin Sanomat (fi), in that order.