There are two ways to predict the outcome of elections and referendums: opinion polls and big data. Big Data is a newcomer on the scene and attempts to use algorithms to scour social media for clues as to which group is gaining or losing and make a prediction about the plebiscite results.

The first group doing some tracking is [Equimedia](https://www.equimedia.co.uk/blog/eu-referendum-will-turnout-be-decisive/), which accurately forecast the results of the Scottish referendum in 2014. 

![](https://i.imgsafe.org/d56993ecd1.jpg)

As you can see from the above graphic they believe that Remain has the advantage at the moment. In addition they show that voters are extremely interested in issues about Jobs, Business and Trade, and these favour the Remain side too.

Here is a breakdown of the social media stats for each campaign:

![](https://i.imgsafe.org/d5846e9c78.jpg)

Again the Remain campaign (which is backed by both the Prime Minister and Leader of the Opposition) has the advantage, though Leave is closing fast.

The other people doing Big Data  analysis are Professor Massimo Poesio and Professor Udo Kruschwitz of the University of Essex, who have their own results page at [Sense-Eu](http://www.sense-eu.info/).

Thus far their prediction is Stay 51% and Leave 49%. Their website correlates the most influential tweets (the ones which are being shared and retweeted the most). Interestingly the Leave side dominate.

As at the time of writing this, there are just 10 days to go to the referendum.