Back
SHARE

Guest brief: Brexit 2016: Insights Analysis

Share on linkedin
Share on facebook
Share on twitter

Guest post from John Swain using Right Relevance data to analyze the Brexit conversations on Twitter in 2016.  The report produced (linked below) visualizes and maps those conversations along with other analysis in the form of graphs and charts.

Source Link

Brexit — Remain v Leave, PR on Twitter

In a previous article we looked at the Twitter conversation about Brexit during the week ending 21 May.

Here is a map of the overall conversation with two accounts highlighted.

These two accounts are the main media accounts for the Leave and Remain campaign groups.

One of the things that is evident from the network map alone is that the Remain side has a lot more of the important people (denoted by the size of the node) compared to the Leave side. The importance is measured by Page Rank in this diagram which shows the importance of the node within the network. However, it is obvious to a person with average knowledge of the political landscape that those important users on the Remain side contain a high proportion of important global influencers from media and politics from the wider world.

We looked at the contribution of the two highlighted accounts to building connections with important Users.

To do that we extracted the ego network (network with nodes directly connected to the node of interest — a connection being a ReTweet, a Mention or a Reply on Twitter) for each user account.

If the ego networks are simply coloured pink we can see that Leave and Remain are more closely associated with their respective sides of the network as you might expect.

blank

blank

Leave (left) and Remain (right) ego networks highlighted.

It is noticeable that the Remain side as more connected overall, however there is one very significant difference which is not so obvious at this level but can be seen on closer inspection.

The section of the map with the Remain supporters has much higher density of important users — denoted by the size of the nodes.

blank

If the ego networks are analysed separately it is easier to see which media accounts are connected to the Media accounts of each side.

Vote Leave Media

This is the network map of the Twitter users directly connected to Vote Leave Media.

blank

Stronger In Press

This is the network map of the Twitter users directly connected to Stronger In Press.

There are several obvious differences:

  • More important users overall.
  • More diversity as indicated by the colour (shows different communities of users identified by machine learning) and examination of the users in the map including Nigel Farage a key proponent of the opposing view.
  • Higher prevalence of major media e.g BBC, Telegraph, Reuters, The Guardian, The Times of London.

blank

Conclusion

There are three recent sources of information which put this analysis in context.

Guardian Media Study

The Guardian suggests a media bias in favour of the Leave campaign.

FT Poll of Polls

The FT Poll of Polls currently indicates a reasonably strong overall position in favour of Remain.

blank

Betfair Odds

Betfair odds indicate a very strong position for Remain.

blank

https://www.betfair.com/exchange/plus/#/politics/market/1.118739911

If the Guardian article is correct and there is a genuine media bias for the Leave campaign then the Remain campaign are clearly doing well to counter that important element in the debate.

However, the Twitter analysis may indicate that the major stories analysed by the Guardian do not give the full picture of media communication in the Social Network era. It may be that the major media published articles and the communication from those same media in other channels are not consistent.

Either way my research shows that the Stronger In Press are running a very effective Twitter campaign.

Never miss a story about Brexit by following our feed

Sign Up

 

If you are interested in using our data under an Attribution CC-By license to write a report, please contact us.

Discover the power of Cronycle for Teams

What to read next