Twitter Research: What To Study and How To Study It
A quick glance at the Twitter research bibliography compiled by danah boyd shows that Twitter is increasingly being seen as a serious object of study within academic circles; on the aforementioned page there are 69 Twitter-related journal publications, conference papers, MA theses and tech reports, placing the purposes micro-blogging platform in categories as diverse as ambient journalism, online activism, language learning and informal communication. When thinking about the latter, a short-term study of Twitter done by Pear Analytics, suggesting that 40% of tweets are “pointless babble” (BBC News, August 17 2009), comes to mind. The same study revealed that 37.5% are conversational while only 8.7% have “pass-along value” (BBC News, 2009).
In other words, very little of the shared content is substantive and news worthy. The most prominent example of such content are the Iran revolution tweets. According to research done by the Digital Methods Initiative, between June 10 and June 30 2009, over 650,000 tweets contained the hashtag #iranelection, with more than 612,000 of them in English, which alludes to the international impact of this event. Ari Berman of The Nation blog wrote:
Forget CNN or any of the major American “news” networks. If you want to get the latest on the opposition protests in Iran, you should be reading blogs, watching YouTube or following Twitter updates from Tehran, minute-by-minute. (Berman, May 16 2009, emphasis added)
Such updates are a prime example of citizen reporting and a valuable example of Twitter as a “broadcast medium” (ct). A study carried out at the Harvard Business School comes to this conclusion and claims the platform is few people tweet and even fewer listen consistently, thus pointing out Twitter’s broadcast character as opposed to “an intimate conversation with friends” (Heil in BBC, June 9 2009). If people on Twitter don’t listen, who does? Tools like Monniter allow everyone, including non-users, to observe how different topics and hashtags are used in real-time; Twitter becomes an important news source in times when the mainstream media’s site access to events is limited or impossible. What limits the research of the large-scale, so-called “Twitter Revolution” is precisely that: its size. To perform discourse analysis on a sample of this magnitude is a daunting task that would require immense amounts of staff, time and money. What can be done is to select, for example, the top 3 retweets per day and then filter out issues of importance in a qualitative manner, as was shown by the DMI-produced film “For the ppl of Iran – #iranelection RT”.
“Self-organizing content” is another project in the same vein, which delves deeper into what relation there is between the organizing principles – hashatgs and retweets – of the platform or between these principles and the content. It deals with “how the Ground Zero Mosque issue, which is not actually a mosque and not on Ground Zero, but an Islamic center in lower Manhattan, is organized on Twitter” (Poell et al, 2010) and aims to examine what kinds of accounts of this particular controversy are produced and what actors are involved. Some telling results show how the use of the term ‘Ground Zero Mosque’ is associated with the #tcot hashtag or how the more objective ‘New York Islamic Center’ often appears in connection to the #cnn hashtag. Hashtags and the relations between them thus grow from being mere labels to indications of political partisanship.
To acknowledge Twitter’s broadcast medium status when it come to tweets with pass-along value (such as controversial news events) means to shed light not only on citizen journalistic practices, but also on how individuals align or disassociate themselves from mainstream media accounts, offering alternative story lines. I argue for an interdisciplinary approach that makes use of the quantitative benefits of tools and then qualitatively inquires into was is actually being transmitted, by whom and to what ends.
BBC News (June 9, 2009). Twitter hype punctured by study. Retrieved October 10, 2010 from http://news.bbc.co.uk/2/hi/8089508.stm
BBC News (August 17, 2009). Twitter tweets are 40% ‘babble’. Retrieved October 10, 2010 http://news.bbc.co.uk/2/hi/8204842.stm
Cernison, M., Petkova. S and Poell, T. (2010). Self Organizing Content. Retrieved October 10, 2010 from http://wiki.digitalmethods.net/Dmi/SelfOrganizingContent
Digital Methods Initiative (2009). For the ppl of Iran – #iranelection RT. Retrieved October 10, 2010 from http://movies.digitalmethods.net/for_the_ppl_of_iran.html