During the period 20th February to 6th March 2016 we collected over 780k Tweets relating to the, so called “Brexit” referendum in the UK.
The objective is to discover who is influential in the debate on Twitter.
To achieve this objective we use a methodology which analyses the communications between users (Retweets, Mentions, Replies) and the way in which these communications form communities of Users. From this structure we can discover:
- The most Influential Twitter Users overall.
- The Users who are Connectors between different groups in the overall conversation.
- Groups of Users who potentially suffer from an echo chamber effect.
This is a link for further technical information about the methodology used for this analysis.
Most Influential Users
The results presented here are the analysis of the first full week of data collection 29 Feb — 06 Mar 2016.
Top overall influence is measured by combining the quantity of connections (Retweets, mentions, replies), the quality of the connections (measured by Page Rank) and the reach of the Users tweets. These are adjusted to discount the skew towards Users with very large number of followers.
The value of a Users connectedness is measured using an algorithm called Betweenness Centrality. This measures how well a User is connected to all other Users compared with everyone else.
Most ‘Interesting’ Users
The Interesting metric finds smaller Users who made a relatively high impact. It compares how well a User does in the overall ranking compared to how well they would be expected to do given the number of followers they have.
It is useful for finding niche or local stories in a large network where they are difficult to find.
The Twitter conversation map helps identify some of the key groups and communities of Users and how the interact on Twitter.
For example the large green area of the map is made up of Users supporting the Leave campaign.
One of the key Users is @LeaveEUOfficial
LEAVE.EU (@LeaveEUOfficial) | Twitter
The latest Tweets from LEAVE.EU (@LeaveEUOfficial). A cross-party campaign advocating a vote to leave in the…
Here is a close up view of the main group with the sub groups highlighted in different colours to show the sub groups. These sub groups are called Tribes. The Tribe with @LeaveEUOfficial as the most important member is show in purple.
Here is a list of the members of this Tribe. The highlighting in green shows that most of the members are supporters of the Leave campaign.
Echo Chamber Effect
Looking at the overall map it appears that there is more activity from the Leave campaign.
However, the tables of the Top Performing Users Overall and Top Connectors shows a fairly even split between the two camps.
Also visible in the network map is the fact that the leave campaign community seem to be highly connected between themselves. In this image of the network this effect can be seen more clearly.
In this map the Users have been filtered to only include those who have a high Retweet to Original Tweet ratio. It appears that there is a much more organised structure of Retweeting each others Tweets in the leave campaign.
The fact that the leave campaign has a strong representation in the Top Connectors table would indicate that there is effective communication in the wider network from some Users on the Leave side.
However, the fact that the leave campaign appears to be much more active overall may significantly skew the Connector measurement. Therefore, the evidence of an echo chamber effect is worth further investigation and monitoring.
Twitter is clearly an important part of the Brexit campaign with both sides engaging in a high volume of Twitter activity.
Our analysis has started to identify the most influential Users and the structure of the various communities who are Tweeting on the issue.
There are indications that there is a danger of creating an echo chamber effect where the message is very strong within a community but not being communicated effectively to the wider world. This looks most likely for the leave campaign.
This preliminary work will enable us to identify which campaign the Users and individual Tweets are supporting and undertake further analysis as the campaigns progress to evaluate the effectiveness of the campaigns and the media coverage.