Social Media as Oracle – Predicting the Election Result

A lot has been written over the past four weeks about the impact of social media on this election campaign. At the start we were all abuzz with the thought of the UK’s first digital election, expecting an Obama-esque online surge to number 10. However, once the TV debates started it soon became clear that they would be the key influencer in this election (though data from the Sun’s “save page 3” campaign is yet to come in). This is not to say that social media has had no effect on people’s voting behavior, or that it does not reflect their voting intentions. So, with a bit of help from the BBC, I’m going to stick my neck out and try to predict the political landscape we will wake up to on Friday morning.

In all of the below models I have tried to remember that the three leaders do have parties behind them, but I have conveniently forgotten that there are other parties out there. If Nigel Farage is Prime Minister by the end of the week, I’ll come back and run some of this data again. In all of these models I have assumed a uniform swing across the country.

Model 1: Facebook

Facebook’s Democracy UK poll has 43% backing the LibDems, 24% the Conservatives and 23% Labour. If this is an accurate reflection of the general population this would give the LibDems an outright majority of 182 seats. I haven’t got a Venn diagram to show you the overlap between the people who think it is important to vote in Facebook polls and those who think it is important to vote in a general election, but I think that it would demonstrate the flaw in this model.

Model 2: Homepage Traffic

As eloquently pointed out by Robin Goad of Hitwise the levels of inbound traffic to the websites of the three main parties are now pretty much equal in stark contrast to the levels a month ago. In the week leading up to May 1^st, the Conservatives had 36% of traffic, Labour 33% and the LibDems had fallen away slightly to 31%. This model gives a hung parliament, with Labour as the largest party, winning 296 seats.

Model 3: Social Media Chatter

Even if the “digital election” didn’t live up to billing, levels of social media activity around the election have been huge, particularly the Twitter response the leaders’ debates. The chart below shows a comparision of the number of mentions in social media of the three main parties and their leaders over the last week.

If (and once again it’s a big if) this is reflected at the polls, we get a hung parliament, with the Conservatives as the largest party, winning 317 seats.

Model 4: Social Media Sentiment

Finally, if we look at the same social media chatter, but only consider the positive statements the chart changes. This model also gives a hung parliament, with Labour as the largest party, winning 287 seats.

Some of these predictions seem more likely than others, but we will have to wait until Friday to see which is the most accurate. They mostly suggest that it is going to be close, which at least chimes with the other predictions out there, but as Peter Snow used to say – “it’s just a bit of fun, that’s all it is”.