After our success harnessing big data to accurately predict the outcome of the
Scottish Referendum, we decided to dust off our ‘swingometers’ and crunch the numbers over the upcoming European Union vote.
We are tracking information across multiple data sources and combining our analysis with social signals to create an equimedia take on where people are planning to place their cross on the ballot paper on Thursday 23
rd of June.
Now that the local elections and London mayoral race are out of the way, we are already seeing the media focus shift quickly to the upcoming vote.
With the majority of political heavyweights siding with the
‘Stronger In’ campaign, ‘Vote Leave’ may have an increasingly hard task to win around the large majority of undecided voters. If we continue to see the Scottish, Welsh and Northern Irish voters remain staunchly pro-Stay, only a large swing across England towards ‘Leave’ will deliver anything other than a ‘Stay’ result. Then again, as Donald Trump has proven nothing should be taken for granted in politics.
You may also remember that the outcome of the General Election did not line up with predictions. We discovered that the flaw in both the pollsters’ data collection methodologies and our own results predictor was that the data was heavily biased as the vast majority of opinion that is shared comes from men voicing their support for their party, whereas women as a group were not expressing their political opinions via social channels. Read
In reality, each outcome will have advantages and disadvantages for serious consideration by the voting public; however as with the Scottish Referendum no side can really produce solid reliable data to back up their arguments as the implications of leaving are truly unknown.
So let’s have a look at our starting point.
Currently our early model suggests an 8 percentage point win for the ‘Stay’ campaign, however this is likely to change as we gather more data to improve our model.
England is likely to become the key battle ground, holding the majority of voters with the highest propensity to ‘Vote Leave’. It will be interesting in the next few weeks to see how each campaign speaks to the various UK regions, as perceived campaign neglect could swing undecided voters to a differing position.
The EU Referendum is by any measure an extremely complex question with many key issues likely to drive opinion. We will be tracking online and social topics throughout the next few weeks. Initial data does however suggest no one key issue is likely to be a main decision maker, with each campaign likely to deliver a series of more refined audience strategies, rather than a ‘one size fits all’ message to drive success.
Alongside our regular reporting we will also be looking a little deeper into the social data. Check back with us each week as we expand our insights on top of our weekly regular reporting updated every Wednesday until voting day.
Can we repeat our referendum prediction success, we will find out in a few weeks’ time!
EU Referendum Week 1 - What does the data say?
EU Referendum Week 2 - Can hashtags provide insights?
EU Referendum Week 3 - When will the touch paper be lit?
EU Referendum Week 4 - Driving social followers
EU Referendum Week 5 - Will turnout be decisive?
EU Referendum Week 6 - Is social data underrepresenting female opinion?
EU Referendum Week 7 - Time for the undecided voters to make their decision
EU Referendum Reflection - Age the deciding factor