Keepin' It Real: Performing Authenticity On Twitter Disinformation Accounts

My research will focus on the representation of migrants on social media, specifically, on Twitter. The research would investigate how users 'talk' about migrants, and how other users interact with such talk: whether they negotiate, contest or co-construct views. The form of interaction would also be investigated in terms of the profile of the users, whether they are big news corporations, or laypeople, and whether such statuses affect the negotiation of representations. The influence of Twitter on modern society is huge, uncontrollable, and somewhat unpredictable. Its scope can be appreciated by acknowledging that on the day of the 2016 US election, it was the biggest source of breaking news: 40 million tweets were sent before 10pm (Isaac and Ember, 2016), emphasising the prominence of Twitter in modern society, and why it is important to investigate the attitudes reflected by users on the site. This research may also make valuable links with the topic of conformity. Conformity can sometimes be hard to measure, due to societal norms, and desirability bias. Often, people will not discuss their true feelings, opinions, or political leanings, due to not wanting to appear negatively to their peers. This is consistently reflected in political polls. Frequently, prediction polls will not accurately predict voter turnout, often because people say they will vote, as to appear socially desirable, but will then not actually cast a vote once the election takes place (Blais et al. 2004). This research would therefore make it possible to investigate elements of conformity, based on laypeople's reactions to big news corporations, compared with reactions to other laypeople: are laypeople more likely to agree with accounts that appear more powerful? Conformity studies offline have found that people feel uncomfortable, and even encounter physical pain (Eisenberger et al 2003) at the prospect of being out of step with their peers. No research has shown whether this is the case online, and although my proposed research would be based on language, not brain activity, it may illuminate areas for future research and development in different fields. This research project will take the form of critical discourse analysis, examining tweets that were written in recent times, regarding migration. Critical discourse analysis is a framework that enables analysis of the relationship between language and wider social practices. Its focus is largely on uncovering power imbalances, common attitudes, and cultural values (Fairclough, 2001:123). This research would therefore take place in a context of critical discourse analysis, as it aims to investigate wider social attitudes towards migration, and power relationships between different users online. To perform critical discourse analysis, it is necessary to collect a large amount of data. Several previous studies have analysed tweets, and thus, there are numerous methods that could be used to collect the data that I would need. The 'HERMES' corpus, and 'Python' are two systems that have been used in previous data collection of tweets. Twitter data is accessible to anyone that has a Twitter account, due to the website's API. The API is the 'application programming interface', and can be navigated using systems such as the 'HERMES' corpus, and 'Python'. Michele Zappavigna outlined how the HERMES corpus can be used, in her 2012 Study of Twitter discourse. The corpus contained over 100 million words, and up to 7 million tweets. Python is also recognised as a valuable tool in seeking out data from Twitter's API. Matthew Russell documents his use of Python in his 2013 work, 'Mining the Social Web', outlining how that too can be used to submit specific requests to Twitter's API, to retrieve data such as Tweets containing specific words. The Masters training inform data collection methods and improve my skills in corpus linguistics and critical discourse analysis.