18/6/2016

The EU Referendum debate on Twitter: Who is winning on Twitter - vote leave or vote remain?

 

This post highlights the disparity of tweeting volumes during the EU Referendum debate between different hashtags using the Visibrain Quick Trends feature, which allows users to examine the tweet frequency of keywords and hashtags.

For non-UK readers, the EU referendum is to take place on Thursday 23rd June 2016 and the UK will vote either to remain in or leave the European Union.

I recently published a blog post and tweeted various insights into the EU Referendum debate using Visibrain and NodeXL. Both the blog post and the tweets went on to receive mainstream media attention with journalists enquiring about the analysis. More specifically, the queries I received surrounded a feature offered by Visibrain (which to the best of my knowledge is current not offered on any other platform). Visibrain is a commercial tool which allows you to pull in vast amounts of data and examine features such as time series graphs, popular users, hashtags, expressions, most frequently shared URLs and so forth. They offer a free non-obligation 30 day trial for anyone looking to explore the tool more. It is a great a media monitoring tool, which specialises in crisis management and prevention for brands.

In this post, I am teaming up with Georgina Parsons from Visibrain, using the Quick Trends feature to explore and compare the various hashtags used by the Vote Remain and Vote Leave campaigns.

voteleavevoteremain.png

Figure 1: Time series graph of VoteLeave vs VoteRemain

As figure 1 above displays, the VoteLeave campaign (n= 1,024,987 tweets) significantly outnumbers tweets in comparison to the VoteRemain (n=180,370 tweets) campaign.

However, I wanted to see if the difference between the #voteleave and #voteremain campaigns was really as simple as comparing these two hashtags. Therefore, we compared all of the hashtag variations around both campaigns, as shown below in figure 2 and figure 3:

figure2.png

Figure 2: Time series graph of #voteleave, #leave, #takecontrol, #leaveeu, #voteout

figure3.png

Figure 3: #strongerin, #remain, #voteremain, #votein, #labourin

As the two figures above show, it turns out that #voteremain is not actually the most popular hashtag around the Remain Campaign: it's actually #strongerIn, which has been used over 600,000 times this past month.

What is apparent from the figures above is that the Leave Campaign is truly concentrated around the #voteleave hashtag. There are many different hashtags for the campaign, but only that one truly stands out. A possible explanation for this is that the #voteleave group is more organised on Twitter or that the hashtag is used by everyone, even those whom are neutral.

As shown below, #brexit is still the most used hashtag around the EU referendum, but the term is frequently used to refer to the referendum in general (for example, it's often referred to as the Brexit referendum). The #InOrOut hashtag is also used for this neutral purpose.

brexit.png

Figure 4: Time series graph of Brexit

We also found, as shown in figure 5 below, that Twitter has been helping to encourage people to take part and vote in the referendum. The #eueefready hashtag along with #voteregistration and #registertovote have been used over 50,000 times over the past two weeks.

figure5.png

Figure 5: #eurefready, #voteregistration, and #registertovote

The link to the "register to vote" page of the UK government website is by far the most-shared link in tweets around the EU referendum: Quick Trends shows that it was shared 141,613 times this past month on Twitter.

sharedurl.png

Figure 7: Most frequently shared URL

However, a caveat to note is that Twitter is a highly non-uniform sample of the population. Not everyone in the UK uses the Internet. Twitter also allows members of the public to hold more than one Twitter account, so in theory one user could set up several accounts to post about vote leave or vote remain. There is also the issue of bots, which are Twitter accounts that tweet in large volumes automatically or look to mimic real users.

Finally, I’d also recommend checking out the NodeXL graph gallery which contains a range of different network graphs related to the EU Referendum.

Comments