Data Labs: The Labour Leadership Vote According to Social Media

Social media in the UK has a left-wing bias. We know this.

If there’s one thing our previous UK election prediction experiments could conclude indisputably, it’s that the way the world looks out of our internet windows is that little bit more groovy and socially conscious than the way it looks out in the street.

The reasons for this are twofold.

Partly it’s presentation bias – we may want to be seen to be more liberal than we actually are.

Mostly it’s about demographics though; social media, a relatively new form of media, will naturally be more popular amongst less conservative (with a small C) folk, those more ready to embrace the new, less tied to tradition, less likely to vote for the Conservative Party or the Republicans.

So, what happens when dealing with an election specifically of the left, one in which all candidates are of varying degrees of ‘lefty’?

How does the volume of social media attention then look?

---

It might look a bit like this, actually

Below we see the volume of mentions of each Labour leadership candidate, since May, for Brandwatch React.

The humps displayed are:

1) 15th June – Corbyn wins a last minute place on the ballot.

2) 20th-22nd July – Corbyn is the sole candidate to vote against the Tory Welfare Bill, followed by a YouGov poll suggesting Corbyn could win.

3) Early August – Corbyn becomes bookmaker’s favorite after a second poll.

Two months ago the Labour leadership election felt little more than an internal restructuring, with three barely distinguishable candidates looking to succeed Ed Miliband on similarly Centerist platforms.

But then Jeremy Corbyn, a politician few had heard of outside of Islington North, entered the race. And suddenly things got interesting, as his candidacy proved unexpectedly popular representing, as he does, a further left side of the Labour Party, and UK politics as a whole.

In the wake of the UK Election result, in which a Conservative Party majority defied the expectations of pretty much all pollsters, the Labour leadership contest has become a debate about the landscape of UK politics.

Corbyn squeezed onto the ballot paper at the very last minute, reaching the required number of nominations with only minutes to spare. And, it appeared, many nominated him with no intention of voting for him, they did it purely to “broaden the debate“.

There is a shopping psychology to this. Faced with a choice, consumers tend to go for the one in the middle.

So all candidates, seeing themselves as the most Centrist, would imagine themselves benefiting from outliers to either extreme.

---

The sentiment breakdown on social

Above: Sentiment analysis on the Labour leadership candidates over the last 12 weeks, for Brandwatch React

---

Can we predict the Labour leadership result?

At the time of writing, we can of course only speculate on the result.

On matters of futurism it’s wise to place faith in the bookies, moreso than pollsters or media monitors.

For reasons of Darwinian Capitalism – any bookie that proved less effective at predicting the future than their gamblers would not be able to survive as bookies for very long, so their very survival depends on them being right.

Pollsters don’t suffer this as much, they just have to write a report on what they’ve learned from their mistakes.

Brandwatch’s data shows Corbyn is dominating the chatter.

This reflects, chiefly, the novelty of his position. He is talked about because he is newsworthy, a story that stands out against the current narrative.

This measure doesn’t necessarily translate into votes though, especially as the leadership will be decided by a rather small sample (approx 600,000 party members), compared to the massive sampling of social.

 

Above: Volume of mentions over the last 14 days, for Brandwatch React

But this volume does give an indication of the trend, the momentum of interest in the candidate.

It is also interesting that the sentiment analysis, for what it’s worth, errs on the side of positive. Although, as evidenced with Farage, volume of mentions, be they good or bad, is what matters more.

As Oscar Wilde said, “There is only one thing in life worse than being talked about…”

---