How has Social Media reflected and influenced political debates and rhetoric in the UK since 2015?

Literature Review
This project explores public perception around the influence of Social Media on politics, and its ability to predict UK political events since 2015. It will use documentary analysis to do so.

Exploring the impact of Social Media on politics is a new phenomenon, but there have been many academic articles (Jensen, 2016) and journalistic explorations on the subject. In newspapers, there are several opinion and investigative pieces, with some saying Social Media is “ruining” politics (Carr, 2015) but others remaining positive. Research has also been carried out into political participation on Social Media (Hale et. al, 2016).
For this study, the predictive aspect of Social Media are particularly relevant. Research on this is scarcer, but nevertheless investigations have been conducted. There are pieces confirming the predictive ability of social media. A study of Twitter activity in German federal elections, published in 2010, concludes that “the content of Twitter messages plausibly reflects the offline political landscape” (Tumasjan et al., 2010: 1). In the UK, retrospective analytics of the 2010 General Election (Franch, 2013) (Anstead and O’Loughlin, 2014) concluded that Social Media was a very good indicator of the result, above and beyond the power of traditional polling methods, and noted the power of Social Media in Semantic Polling.
This study is interested in how Social Media has played a predictive role, or active role in contemporary events in British politics and the public perception of this. Since 2015, two events that appear to dwarf all others are the 2015 General Election, and the European Union Referendum. There is little published work on these events, due largely to how recent they are. One paper on the 2015 General Election (MacDonald and Mao, 2016) looks at the components of this, concluding that Big Data and Social Media provide a very accurate indicator of the outcomes. Yet, it is also possible to see conflicts. In two articles analysing Social Media and the 2015 election, academics dispute the utility of Social Media analysis, declaring it does not shape the election and it cannot be guaranteed to help predict results (Shephard, 2015) (Fletcher and Schifferes, 2015).
Work has also been done around the European Union referendum. There is a huge amount of data available, particularly from Twitter (Benoit et al, 2017) but again there does not appear to be a clear thread as to whether Social Media can reflect, influence or predict UK political developments. Other studies reflect this (Polanski, 2016) (Polanski, 2016) and note limitations (Llewellyn and Cram, 2016).
This literature review shows that there is variety in opinion and evidence when it comes to social media and politics, particularly in the UK since 2015. This study therefore, aims to use qualitative analysis to explore this further, looking at five documents attempting to understand public opinion, and looks into the power of social media.

Methodology
The methodology, qualitative analysis of documentary sources, was conducted in line with good academic and ethical practice (Prior, 2003). The analysis was thematic and deductive, and a coding frame was produced (see Appendix 2) modelled on documentary analysis techniques in previous literature (Bryman, 2008). Using the coding frame, the sources were analysed individually, picking out all details relevant to a specific code (examples can be seen in Appendix 3). From this, it was possible to see the balance of emotions and opinions on Social Media and politics.
The five sources studied are as follows:
∞ Comments on the article entitled “Immigration WAS the key issue for Brexiteers in the historic EU referendum battle, according to huge analysis of social media comments” in published in the Daily Mail, 12th December 2016. (Tapsfield and Dathan, 2016)
o comments selectively sampled (Scott, 1990), ensuring they were of relevance to the research question. This was done by looking through the highest rated comments (those most approved by other users of the site) and then selecting ones containing relevant material.
∞ Opinion piece by David Fletcher entitled “Welcome to the social media election that never was” published in the Guardian, 27th April 2015. (Fletcher, 2015)
o A representation of public opinion backed by evidence
∞ Opinion piece by Rhodri Marsden entitled “General Election 2015: Twitter and Facebook have nothing on traditional electioneering” published in the Independent, 29th April 2015. (Marsden, 2015)
o A representation of public opinion backed by evidence
∞ Blog post by Rick Delgado entitled “What Social Media Will Do for Politics in the Future” on socialmediaweek.org, 23rd December 2016. (Delgado, 2016)
o A more informal discussion in which the power of social media and the author’s opinion is presented.
∞ News article by Ami Sedghi entitled “A third of young people think social media will influence their vote” published in the Guardian, 10th March 2015. (Sedghi, 2015)
o This represents an insight into the public opinion on the influence of Social Media.
∞ Comments on the article entitled “Election 2015: It wasn’t social media ‘wot won it’” in published on the BBC News Website, 11th May 2015. (Cellan-Jones, 2015)
o For the study the 20 top rated comments were selected from the site and analysed.

A selection of sources is available in Appendix 1.
The sources were selected with the aim of reflecting public opinion as they are from different viewpoints, whilst also including evidence. Both comment sources were selected to represent public opinion into the use of Social Media as a predictive tool. These allowed an exploration into the mind-set of the social media users who are being analysed. The comments on the Daily Mail article also allowed testing of the hypothesis from the research paper quoted within the article. The eclectic mix of sources were selected so that after analysis they would provide a deeper insight in to the power of social media in influencing and predicting social media, and the public opinion on this.
Very few ethical issues were raised by this documentary research. All secondary data comes from recognised archives, or appears in the documents analysed. All sources are publically available and there will be no need to access any private collections. Additionally, these sources were analysed and referenced with appropriate academic process and rigor.

Findings
The overall findings show that the public are unsure about the power of social media to influence and/or predict. There is a balance across almost all codes and evidence on each side.
The first two codes, corroboration of a claim and challenge of a claim made by Social Media analysis appeared most apparent in the Daily Mail article comments. Some comments supported the hypothesis that immigration was the main driver behind Brexit (11) and others gave different reasons (5). Although some of those giving other reasons were negative about the power of Social Media as a predictor, some did not give a reason. There is one further statement in the Independent article that also challenges Social Media analysis; the claim that “Yes” would be victorious in the Scottish referendum.
From the codes, negativity about the power of Social Media as a predictive force was one of the most prevalent, with 10 comments on the Daily Mail article alone. People stated their belief that the analysis was unrepresentative due to sample size, that more platforms other than Twitter should have been looked at and that Social Media did not include the older generation. This sentiment of negativity towards the predictive power of Social Media also appeared in comments on the BBC news article, with many comments referring to abuse users might receive if they are clear about their political beliefs on Social Media. These comments suggest that due to fear of expressing political beliefs online, the ones that are online cannot be representative. The opinion piece in the Independent also states social media samples are rarely representative.
There is also expression of positive feeling towards the predictive power of Social Media. The corroborating comments to the hypothesis in the Daily Mail article back up this power, reinforcing its accuracy. The blog post claims Social Media is used most by young people, a group who go to the polls more frequently, asserting its predictive power by extension. The Independent article refers to the power of prediction when done through semantic analysis, using the Conservatives and their victory as evidence.
Negativity around the power of Social Media to influence individuals was far more prevalent in the comments on the BBC article than the Daily Mail one, with 12 negative comments referring to the poor nature of Social Media for political discussion. The BBC article comments referenced the complex term “group think”, whereas the sample of comments from the Daily Mail referred to this but did not name it. Continuing with negative sentiment about the power of Social Media to influence, we can see it was also very prevalent in the opinion pieces from both the Independent and the Guardian. Fletcher in the Guardian, proposed that politics misses the core aspect of what social media means for its users, and that content is banal. He evidenced this with tweets from David Cameron. He also looked at group-think, as did the Independent piece, both again without naming the phenomenon explicitly. He also proposed that there is too much risk associated with a more-forward thinking social media policy in politics, citing the “white vans and St George’s flag” (Graham, 2014).
There were examples where positivity about the power of social media to influence was clear. This came through most strongly in the Guardian news article, and in the blog post. The blog post, whilst light on evidence, made grand statements on how “social media will change politics forever”, as people can reach out to politicians easier, seeing their personality. The Guardian news article contained more evidence, recognising the power of social media for young people. Statements and evidence about the power of social media to influence were also present in the Guardian opinion piece, which picked out isolated examples of social media campaigns that were successful.
The Guardian news article as contained the most evidence, distantly followed by Guardian opinion piece. The comments contained very little evidence, which could be used to draw conclusions about the reliability of the sources.
Conclusions
In conclusion, the appearance of all codes within the data is balanced, although veers slightly more towards the negative codes. Interestingly, in the very same document where there are negative points around Social Media’s power to influence and/or predict, positive points appear too. The Daily Mail article comments are an example of this; many diminish the power of social media, an almost equal number back up the findings of the study. In the Guardian opinion piece, the author declares social media has been far from influential in the UK general election, yet in America in 2012 it had been incredibly so.
I would say, from the literature review, this study does fit with the narrative around social media in politics. Although the documents focus only on the 2015 General Election and the EU referendum, they do follow the pattern of social media being able to predict occasionally, whilst having limitations. Its ability to influence is also nebulous at this stage. An interesting point about the power of social media to influence comes from awareness of “group think” in the four articles which are most negative about it. This questions the power of “group think”, because if individuals are aware of this phenomenon, then it surely must lessen its power?
It is important to acknowledge limitations to this study. It cannot be assumed to be representative of social media in politics, as it focuses on the UK only in a comparatively short time frame, and on only two events. Although the sources to attempt to be representative, the comments are from mainstream media sites.
In conclusion, this study and the wider literature present ambiguous conclusions about the power of social media to influence or predict in the political sphere. This therefore suggests that deeper studies with wider samples, both quantitative and qualitative, should be conducted, and that more time may be needed before it possible to understand and analyse the impact social media has on politics.
Reference List
Anstead, N. and O’Loughlin, B. (2014). Social Media Analysis and Public Opinion: The 2010 UK General Election. Journal of Computer-Mediated Communication, 20(2), pp.204-220.
Benoit, P., Hobolt, P., Preece, D. and Plantin, D. (2017). The Brexit Debate through Social Media: deliberative discussion, or deliberate dysfunction?.
Bryman, A. (2008). Social research methods. 1st ed. Oxford: Oxford University Press.
Burnap, P., Gibson, R., Sloan, L., Southern, R. and Williams, M. (2015). 140 Characters to Victory?: Using Twitter to Predict the UK 2015 General Election. [online] Available at: https://orca.cf.ac.uk/73127/1/BurnapElections2015.pdf [Accessed 11 Mar. 2017].
Carr, N. (2015). How Social Media Is Ruining Politics. [online] POLITICO Magazine. Available at: http://www.politico.com/magazine/story/2015/09/2016-election-social-media-ruining-politics-213104?o=0 [Accessed 11 Mar. 2017].

Cellan-Jones, R. (2015). Election 2015: It wasn’t social media ‘wot won it’ – BBC News. [online] BBC News. Available at: http://www.bbc.co.uk/news/technology-32689145 [Accessed 11 Mar. 2017].
Delgado, R. (2016). What Social Media Will Do for Politics in the Future. [Blog] Social Media Week. Available at: https://socialmediaweek.org/blog/2016/12/social-media-politics-future/ [Accessed 11 Mar. 2017].
Fletcher, D. (2015). Welcome to the social media election that never was. Guardian. [online] Available at: https://www.theguardian.com/media-network/2015/apr/27/social-media-general-election-political-parties [Accessed 11 Mar. 2017].
Fletcher, D. and Schifferes, P. (2015). Can trends in social media explain why the opinion polls got it wrong?. [online] electionanalysis.uk. Available at: http://www.electionanalysis.uk/uk-election-analysis-2015/section-2-voters/can-trends-in-social-media-explain-why-the-opinion-polls-got-it-wrong/ [Accessed 11 Mar. 2017].
Graham, G. (2014). Labour shadow minister accused of ‘contempt for working classes’ over white van tweet. [online] Telegraph.co.uk. Available at: http://www.telegraph.co.uk/news/politics/labour/11243726/Labour-shadow-minister-accused-of-contempt-for-working-classes-over-white-van-tweet.html [Accessed 13 Mar. 2017].
Franch, F. (2013). (Wisdom of the Crowds)2: 2010 UK Election Prediction with Social Media. Journey of Information Technology and Politics, [online] pp.57-71. Available at: http://www.tandfonline.com/doi/abs/10.1080/19331681.2012.705080 [Accessed 11 Mar. 2017].
Hale, S., John, P., Margetts, H. and Yasseri, T. (2016). Political turbulence. 1st ed. London: Princeton University Press.
Jensen, M. (2016). Social Media and Political Campaigning. The International Journal of Press/Politics, [online] 22(1), pp.23-42. Available at: http://journals.sagepub.com/doi/abs/10.1177/1940161216673196 [Accessed 11 Mar. 2017].
Llewellyn, C. and Cram, P. (2016). The results are in and the UK will #Brexit: What did social media tell us about the UK’s EU referendum?. [online] EU Referendum Analysis 2016. Available at: http://www.referendumanalysis.eu/eu-referendum-analysis-2016/section-7-social-media/the-results-are-in-and-the-uk-will-brexit-what-did-social-media-tell-us-about-the-uks-eu-referendum/ [Accessed 11 Mar. 2017].
MacDonald, R. and Mao, X. (2016). Forecasting the 2015 General Election with Internet Big Data: An Application of the TRUST Framework. Business School – Economics, University of Glasgow. Working Papers. [online] Available at: http://www.gla.ac.uk/media/media_440128_en.pdf [Accessed 11 Mar. 2017].
Marsden, R. (2015). General Election 2015: Twitter and Facebook have nothing on traditional electioneering. Independent. [online] Available at: http://www.independent.co.uk/life-style/gadgets-and-tech/features/general-election-2015-twitter-and-facebook-have-nothing-on-traditional-electioneering-10213731.html [Accessed 11 Mar. 2017].
Polonski, V. (2016). #Nofilter debate: Brexit campaigners dominate on Instagram. [online] The Conversation. Available at: https://theconversation.com/nofilter-debate-brexit-campaigners-dominate-on-instagram-59933 [Accessed 11 Mar. 2017].
Polonski, V. (2016). Impact of social media on the outcome of the EU referendum. [online] referendumanalysis.eu. Available at: http://www.referendumanalysis.eu/eu-referendum-analysis-2016/section-7-social-media/impact-of-social-media-on-the-outcome-of-the-eu-referendum/ [Accessed 11 Mar. 2017].
Prior, L. (2003). Using Documents in Social Research. 1st ed. London: Sage.
Scott, J. (1990). A matter of record: Documentary Sources in Social Research. 1st ed. Cambridge: Polity.
Sedghi, A. (2015). A Third of Young People Think Social Media will influence their vote. The Guardian. [online] Available at: https://www.theguardian.com/news/datablog/2015/mar/10/a-third-of-young-people-think-social-media-will-influence-their-vote [Accessed 11 Mar. 2017].
Shephard, D. (2015). Using social media to predict the General Election?. [online] Election Analysis. Available at: http://www.electionanalysis.uk/uk-election-analysis-2015/section-2-voters/using-social-media-to-predict-the-general-election/ [Accessed 11 Mar. 2017].
Tapsfield, J. and Dathan, M. (2016). Immigration WAS the key issue for Brexiteers in the historic EU referendum battle, according to huge analysis of social media comments. Daily Mail. [online] Available at: http://www.dailymail.co.uk/news/article-4022308/Immigration-key-issue-Brexiteers-historic-EU-referendum-battle-according-huge-analysis-social-media-comments.html [Accessed 11 Mar. 2017].
Tumasjan, A., O. Sprenger, T., Sandner, P. and Welpe, I. (2010). Predicting elections with Twitter: What 140 characters reveal about political sentiment. Proceedings of the Fourth International AAAI Conference on Weblogs and Social Media. [online] Available at: https://www.aaai.org/ocs/index.php/ICWSM/ICWSM10/paper/viewFile/1441/1852 [Accessed 11 Mar. 2017].

Leave a Reply