Hillary Clinton and Donald Trump were two successful US presidential contenders in 2016. Trump is emotional and often rude, Clinton is more restrained, but also optimistic and engaged. Each nominee had his own army of admirers who wanted their candidates to win more than anything else. There were also skirmishes and clashes between the allies of Trump and Clinton in the battle for the President’s position. The purpose of this paper is to analyze the press statements made by the supporters of Donald Trump and Hillary Clinton on Twitter for filter bubbles.
A lot of coverage has been paying in the newspapers and on the Internet to outlets that have aimed to draw the public. In the press and on the Internet, a lot of attention was paid to sites that tried to attract the audience for the purpose of public relations of presidential candidates (Benkler et al., 4). In 2016, there were very few center-right sites that attracted followers of Trump, and also followers of Clinton. The Wall Street Journal, for example, worked to attract supporters of Clinton and Trump in equal parts. There was another press, partisan, which tried to attract Trump supporters in a ratio of 4: 1. Unlike such Internet resources, starting with The Wall Street Journal and moving to the left, the attention spreads more evenly across a number of sites.
After reviewing and analyzing of the collections of tweets about Clinton and Trump, it can be concluded that some of them are made using sites linkis.com and twimg.com. Formally, these sites provide an opportunity to brand a link on the Internet. For example, a user before the publication gets on one of them, gets a personalized version and publishes it on Twitter. Linkis is needed by users who want to loosen before their subscribers more often than usual. To do this, this site copies the content of the page to which the user shares, to his site, adds a dice with information about the author and the text of the tweet and provides a new link. Thus, it becomes clear that those tweets, the source of which are these two sites only imply someone’s attempt to popularize a certain candidate by constantly popping such tweets on the pages of a potential or existing audience of Trump or Clinton (Pariser, 10). For instance, on 18 October the user “wpmoneymaker” wrote about Ms. Clinton: “We can open the door to every person in this country again”, and left hashtags #billclinton and #quoteoftheday. Posts on Twitter concerning Donald Trump also included such ones, placed with linkis.com and twimg.com.
However, there are also posts on Twitter about either of the two candidates citing publications, such as New York Times, for example. Nonetheless, there were real reposts from online sources. For instance, on September, 26 2016 the user nbenitez1977 left hashtags #StrongerTogether and #PDMFNB and cited the article “Why Donald Trump Should Not Be President” from The New York Times. The link of the article was http://www.nytimes.com/2016/09/26/opinion/why-donald-trump-should-not-be-president.html?_r=1. That tweet was a real one and made by a real person, so its author did not mean to use filter bubbles. On the other hand, one of Trump’s followers left a fake link, saying that it was taken from the online source www.express.co.uk, but not mentioning the exact article. It is obvious that this Twitter user did not think anyone would want to check the information.
It can be concluded that both Trump’s and Clinton’s supporters tended to use online sources, such as articles of famous magazines or newspapers, for instance. However, both of them also used various links from Twitter: both their own ones and other politicians’. The supporters of Clinton, for example, also took and used data from YouTube. Trump’s supporters mostly preferred Twitter. Additionally, both groups of people used linkis.com and twimg.com for gaining the number of supporters of their favorite candidates. The point is that if any American that is interested in politics or is a supporter of one of the two candidates, Trump or Clinton, searches for some information about the issues he was interested in on the Internet, and after all his or her online requests are accumulated, analyzed and structured, they appear on the screen Such user gets only the selected information (Baer, 2). It happens in all social networks and Twitter is no exception.
In this particular case both articles in newspapers and magazines, as well as quotations of the most candidates were used for both candidates’ promotion. The presence of resources that assign information to themselves indicates that Twitter collects information about what one likes and reposts in order to determine the preferences of a certain person concerning a particular candidate. Additionally, this is distributed to the subscribers of this person then. Thus, in this case, filter bubble are used.
The most important goal of this approach is to convince a person that uses Twitter or any other social network that the information he or she is interested in implies the main issues in the world at the moment. Thus, one can incline a person to certain thoughts and choices that are beneficial to a candidate (Sandvig et al, 3). In fact, in addition to president elections, there are many other issues in the world, but the user of Twitter would not know about it.
Baer, Drake. “The Filter Bubble Explains Why Trump Won and You Didn’t See It Coming”. Science of Us, New York Media LLC, 2016.
Benkler, Y., Faris, R., Roberts, H., Zuckerman. E. “Study: Breitbart-Led Right-Wing Media Ecosystem Altered Broader Media Agenda”. Columbia Journalism Review, 2017. www.cjr.org/analysis/breitbart-media-trump-harvard-study.php. Accessed on 21 November, 2017.
Pariser, Eli. “The Filter Bubble. What the Internet is Hiding from You”. The Penguin Press, New York, 2011.
Sandvig, C., Hamilton, K., Karahalios, K., Langbort, C. “Auditing Algorithms: Research Methods for Detecting Discrimination on Internet Platforms”. The 64th Annual Meeting of the International Communication Association, May 22, 2014; Seattle, WA, USA. www-personal.umich.edu/~csandvig/research/Auditing%20Algorithms%20–%20Sandvig%20–%20ICA%202014%20Data%20and%20Discrimination%20Preconfere. Accessed on 21 November, 2017.