Getting the message across in the twittersphere: How Remain and Leave camps use #hashtags

Working Towards Sentiment Analysis


In our work investigating how people discuss the EU within Twitter one of our aims is to determine sentiment, both pro and anti the EU, and in relation to the referendum on UK membership of the EU – pro-remain or pro-leave.

The approach we have taken initially is very straightforward. If tweeters use hashtags associated with the leave camp (including; #brexit, #no2eu , #notoeu, #betteroffout, #voteout #eureform, #britainout, #leaveeu, #voteleave, #beleave, #loveeuropeleaveeu, #leaveeu) then we judge their sentiment to be pro-leave. If they use hashtags associated with the remain camp ( #yes2eu, #yestoeu, #betteroffin, #votein, #ukineu, #bremain, #strongerin, #leadnotleave, #voteremain) then their sentiment was pro-remain.

If we do this we get the following results:


And we can see that these results are fairly consistent day by day:


But they are not consistent with polling data – such as the ICM tracker below which shows remain with the higher score. So why is that?


Firstly – we need to remember that research using Twitter data can only ever tell us what people who use Twitter think and does not necessarily reflect the population at large. Interestingly this has been a problem for pollsters trying to include this data in their election prediction polls (Metaxas, Mustafaraj & Gayo-Avello).

Secondly, we also need to remember that, as we discussed in a previous post, people tend to tweet against things rather than for them.

Thirdly, we are currently restricting our analysis to sentiment associated with hashtags. This means that if the tweet doesn’t have a hashtag we are not analysing it’s sentiment.

If we take a look at the data we can see some differences. There is a difference in style between the way that tweets and in particular hashtags are used between both camps. The leave camps tend to use many hashtags – see below for examples from both LEAVE.EU and Vote Leave – especially LEAVE.EU.



LEAVE.EU use hashtags in their twitter bio – this may encourage followers to use them.

The remain camp do not tend to use as many hashtags in their posts – for example:


Tweets that would be considered pro-remain often also include hashtags we have classified as pro-leave. This could be for several factors such as positioning themselves within the debate by using a popular hashtag or trying to talk to those which hold opposite views.


We can see this by looking at the data. When we look at hashtags that are used in conjunction with #strongerin and #brexit (graphs below). We find, overall, a much lower use of #strongerin and we find that it is used with leave hashtags especially #brexit, #leaveEU, #voteLeave. Where as #brexit is used not used with remain hashtags at all but with other leave hashtags.



In the future we aim to do a more sophisticated form of sentiment analysis where we analyse the text with the tweets. This leads on to a final problem that is often discussed in association with sentiment analysis of text, and one that we also see here, which is identifying the target of the sentiment. The target is the item that the sentiment is expressed towards. In the tweet by Richard Corbett MEP above the ‘leave’ sentiment is expressed towards the ‘U think being part of EU holds back trade with rest of world? ‘ and the ‘remain’ sentiment is expressed within ‘Think again’. Automatically identifying the text that the target is associated with is not always easy. This is not an easy issue to tackle and we have’t even begun to discuss jokes and sarcasm yet!

So you might be tempted to think that if this data is not representative of the general public why are we looking at it? What we can look at is how this data changes. If we can identify the differences and track the relationships between Twitter data and a more general public opinion we can start to hypothesise about how changes in the Twitter data equate to public opinion more widely. We’ll talk about this a lot more in the future.


Our project is part of the Economic and Social Research Council’s UK in a Changing Europe programme. Look out for our regular updates as the project tracks developments in the debate on the UK’s continued membership of the EU and follow us @myimageoftheEU.

This article was originally published on the imagineEurope Storify.

Imagining the EU: Words associated with EU and Europe

We are continuing to investigate how people discuss the EU on Twitter. Our aim is to discover common themes which may emerge leading up to the referendum on the UK’s inclusion in the EU. One of the things that we are looking at is hashtags associated with the terms EU and Europe. The following graph shows us hashtags that are used in conjuction with either #EU or #Europe.



We have been collecting data since 9th August 2015 so we are beginning to build up quite a big dataset – as of the 27th October we had 3,109,130 tweets associated with the EU (gathered using a variety of EU based terms). See our previous article for details.

https://storify.com/ImagineEurope/building-a-twitter-dataset-to-find-out-views-on-th

We can see that the majority of the discussion over our collection period has been on refugees and migrants. We discussed these terms in some detail in another post.

https://storify.com/ImagineEurope/what-is-associated-with-europe

Does this suggest that refugees and migrants will be hot topics that people will consider when voting? Or are they just a reflection of what is currently happening? We’ll have to do more research to find out. Keep following us for updates on this question.
Looking at the data in more detail there are also some things in here that we might not expect. For example ‘opkillingbay’ shows up with a high frequency as does ‘yearinspace’. This is because using terms like EU means we have a broad collection strategy. We’ll need to do some post collection clean up before we present a more fine-grained analysis.


#OpKillingBay press release – Anonymous to save dolphins



We can see immediately that the frequency of the hashtags supporting the UK leaving the EU (brexit, no2eu, betteroffout) is much higher than that of those supporting the UK staying in the EU (yes2eu, betteroffin, votein, Bremain). The frequency of the remain-in hashtags is so low that they don’t even appear in this chart. This is not surprising when we look at our basic sentiment analysis over the same period.



2% of the tweets we have collected are in favour of remaining in the EU compared with 98% associated with a desire to leave the EU. Our regular followers will remember however that twitter is not necessarily representative of public opinion. The existence of a term or hashtag does not always indicate support – it can be more likely that people tweet against a subject (Barber and Rivero 2014). We’ll be keeping a careful eye on this and will return to the question of sentiment in twitter in a later post.


Our project is part of the Economic and Social Research Council’s The UK in a Changing Europe programme. Look out for our regular updates as the project tracks developments in the debate on the UK’s continued membership of the EU and follow us on Twitter @myimageoftheEU.


Neuropolitics Research Lab – People – Politics and International Relations (PIR)

Neuropolitics research politics experiments using fMRI brain scanning.
WWW.POL.ED.AC.UK


Laura Cram is Senior Fellow, The UK in a Changing Europe, investigating The European Union in the Public Imagination: Maximising the Impact of Transdisciplinary Insights (ESRC/ES/N003985/1).

#ImagineEurope at the Turing Summit

This week the Imagine Europe project took part in the Data Science for Media Summit which was held in Edinburgh and hosted by the Alan Turing Institute and the University of Edinburgh.

The aim of the summit was to encourage members of the media to discuss problems that they have with big data and data analytics. The hope is that, from this information, the Alan Turing Institute can identify appropriate areas for research.


The Alan Turing Institute


To start off, Howard Covington, Chair of the Turing Institute, explained the aims of the Institute and gave an overview of the big data landscape.

The day featured a mix of researchers from academia, industrial data scientists and members of the media. The programme included talks by Mike Dewar from the New York Times, Steve Plunkett from Ericsson and Michael Satterthwaite from the BBC.

There were panels on data journalism, audience engagement and the value of data, alongside some great networking.

We had a booth next to the Edinburgh School of Informatics Language Technology group on geo-tagging and data analytics. We gave a demo of the beta version of the tools we are using to analyse the EU-related twitter datasets we are collecting. We are looking, amongst other things, at hashtag frequency, sentiment towards the EU and the location of tweets.

For more information, Nicola Osborne provided a great live blog of the event:


Data Science for Media Summit LiveBlog

Today I am at the ‘Data Science for Media Summit’ hosted by the Alan Turing Institute and the University of Edinburgh and taking place at the Informatics Forum in Edinburgh. This promises to be an event exploring data science opportunities within the media sector and the attendees are already proving to be a diverse mix of media, researchers and others interesting in media collaborations.
NICOLA OSBORNE


Our project is part of the Economic and Social Research Council’s The UK in a Changing Europe programme. Look out for our regular updates as the project tracks developments in the debate on the UK’s continued membership of the EU and follow us on Twitter @myimageoftheEU.


Neuropolitics Research Lab – People – Politics and International Relations (PIR)

Neuropolitics research politics experiments using fMRI brain scanning.
WWW.POL.ED.AC.UK


Laura Cram is Senior Fellow, The UK in a Changing Europe, investigating The European Union in the Public Imagination: Maximising the Impact of Transdisciplinary Insights(ESRC/ES/N003985/1).

This article was originally published on the imagineEurope Storify and the blog version was taken from the ImagineFutures blog.

The Use of #Migrant and #Refugee on Twitter

The Use of #Migrant and #Refugee on Twitter
P-GRC0233, IFRCRCS, CC-BY-NC-ND-2.0

The movement of people across borders and the implications of that movement are at the heart of current debates in the European Union. This issue is likely to play a central role in the forthcoming UK referendum on continued membership of the EU as discussed in this blog post.

What Would UK Immigration Policy Look Like After Brexit?


How the debate is framed in public discussions may have important implications for the outcome of the referendum.

There has been discussion both online and in the media about the use of the terms ‘migrant’ and ‘refugee’ and the difference between the terms.


Migrant v Refugee: What’s the Difference?

Not all migrants then are refugees, but refugees can fall under the migrant umbrella. One of the major differences between the two designations is that while migrants may seek to escape harsh conditions of their own, refugees could face imprisonment …
CNN


With the BBC using both in one headline:


Migrant Crisis: UK Public ‘Split’ over Taking Refugees

Some 57% of people in the UK are in favour of the status quo, or the government taking fewer refugees from Syria and Libya, a poll suggests. Forty per cent said the UK should take in more. One thousand people were interviewed by telephone between Friday and Sunday in a ComRes poll for BBC Newsnight.
BBC News


There was even a petition asking the BBC to use the term ‘refugee’:


Petition · Request BBC use the correct term Refugee Crisis instead …

We kindly request that the BBC use the term Refugee Crisis instead of Migrant … One word can make all the difference. … Zinon Zygkostiotis started this petition …
WWW.CHANGE.ORG


We looked to our dataset to see if there was a difference in how these terms are used on Twitter. The data was collected between 7 August 2015 and 11 September 2015 (it is ongoing but these snapshots are taken from then). Initially we looked at the frequency of the two terms:

We can see that the term ‘migrant’ is used much less frequently than ‘refugee’ – 2549 times as opposed to 7637, respectively. We also see that the use of both terms has increased dramatically after 20 August 2015. The trend continues up towards a peak around 3 September 2015, when the body of a young refugee was found on a beach in Greece.

We also looked at which hashtags are used in association, and used in the same tweet, with migrant and refugee. We also included any compound terms including refugee such as refugeesWelcome (the biggest tag in our set by far) or refugeecrisis. Considering that we built the set using the search terms ‘EU’ and ‘Europe’, it is no surprise to see them in the set. However, since they are used in the same tweets as ‘crisis’ and ‘migrant’, the Twittersphere obviously sees this as an EU-related issue. There are some specific locations mentioned in association with ‘refugee’ – namely, Syria, Kos, Munich, Germany and Hungary.

When people use the term ‘migrant’, they also often mention ‘asylum’ and ‘ISIS’. These terms don’t appear with ‘refugee’.

Our project is part of the Economic and Social Research Council’s The UK in a Changing Europe programme. Look out for our regular updates as the project tracks developments in the debate on the UK’s continued membership of the EU and follow us on Twitter @myimageoftheEU.


Neuropolitics Research Lab – People – Politics and International Relations (PIR)

Neuropolitics research politics experiments using fMRI brain scanning.
WWW.POL.ED.AC.UK


Laura Cram is Senior Fellow, The UK in a Changing Europe, investigating The European Union in the Public Imagination: Maximising the Impact of Transdisciplinary Insights (ESRC/ES/N003985/1).

This article was originally published on the imagineEurope Storify. The blog version was taken from EuropeanFutures blog.

Building a Twitter Dataset to Find out How People View the EU

We are building a dataset to explore the relationship between the UK and the EU and how people talk about this relationship.

David Cameron EU

We are using social media – in particular Twitter – to find out what people are saying and to investigate how this changes leading up to a referendum on the UK’s membership.

We have started by building a Twitter database using the Twitter API.


Documentation | Twitter Developers


In fact, we are going to end up building three databases: one on what is in the Twitter sample stream, one gathered using specific hashtags and one gathered by following what we are calling influencers – basically those that talk specifically about the European Union and are listened to. This ranges from MPs to think tanks, to academics and to pressure groups.

Of course, Twitter data is not representative of the whole population. It is used predominantly by young (74 per cent are between 15 and 25 – see Beevolve) and politically active or by those with a particular axe to grind (‘Twitter is dominated by individuals with strong political views’ – see Barberá 2015). Nevertheless, social media is increasingly the space in which major events and political decisions are debated by large numbers of people. This became very clear during the recent referendum on Scottish independence. We will explore how the debate on the UK’s relationship with the EU is framed and reframed within Twitter and how it relates to the wider offline political conversation.


An Exhaustive Study of Twitter Users Across the World

Our Social Media Monitoring platform mines through millions of conversations every day for our customers. Out of these millions of conversations, quite a few happen to come from Twitter. We realized that we could do a large-scale in-depth study of Twitter users to better understand what is the average profile of a person using Twitter and hence this mammoth study on 36 million Twitter user profiles.
Beevolve


Thumbnail for Birds of the Same Feather Tweet Together: Bayesian Ideal Point Estimation Using Twitter DataBirds of the Same Feather Tweet Together: Bayesian Ideal Point Estimation Using Twitter Data

Wilf Family Department of Politics, New York University, 19 W 4th Street, 2nd Floor, New York, NY 10012. Politicians and citizens increasingly engage in political conversations on social media outlets such as Twitter. In this article, I show that the structure of the social networks in which they are embedded can be a source of information about their ideological positions.
Oxfordjournals


We have started off by gathering data on specific hashtags. We started on 7 August 2015 using the terms: #eureferendum, #euref, #brexit, #no2eu, #yes2eu, #notoeu, #yestoeu, #betteroffout, #betteroffin, #voteout, #votein, #eureform, #ukineu, #Bremain, #EUpoll, #UKreferendum, #UKandEU, #EUpol, #ImagineEurope, #EdEUref, #MyImageOfTheEU, #UKRef, #ref.

The wordle shows the frequency (bigger = more) of hashtags in the dataset on 7 August 2015.

We added #referendum #eu, and #europe on 25 August 2015:

In response to events in Europe, we added #refugeesWelcome on 3 September 2015:

Across the whole time period:

Obviously there is a lot more to do and this is just a taste of what we are looking at. Look out for our regular updates as the project tracks developments in the debate on the UK’s continued membership of the EU and follow us @myimageoftheEU.


Neuropolitics Research Lab – People – Politics and International Relations (PIR)

Neuropolitics research politics experiments using fMRI brain scanning.
Politics at the University of Edinburgh.


Laura Cram is Senior Fellow, The UK in a Changing Europe, investigating The European Union in the Public Imagination: Maximising the Impact of Transdisciplinary Insights (ESRC/ES/N003985/1).

This article was originally published on the imagineEurope Storify and the blog version was taken from the ImagineFutures blog.