Theme 3: Research and theory

Generally speaking, theory is "A set of statements or principles devised to explain a group of facts or phenomena, especially one that has been repeatedly tested or is widely accepted and can be used to make predictions about natural phenomena." (1). In simpler terms it is an explanation to a scientific problem or question that can be tested and invested.
That being said, there is a high debate on what theory exactly is. One of Gregor's types of theory is "Analysis Theory" whose key attributes are that "the theory does not extend beyond analysis and description." and that "No causal relationships among phenomena are specified and no predictions are made". Staton and Staw however, write that lack of causal relationships and the lack of predictions are exactly what theory is not. Personally, while I see the benefit of Gregors categorization and also the importance of publications that are within this analysis category, I tend to coincide with Staton and Staws view, that good theory has to connect the facts and come up with a new answer to the problem, otherwise it is just blatant repetition of facts.

I chose Studying political microblogging: Twitter users in the 2010 Swedish election campaign by Larsson & Moe (2) out of the 2012 New Media & Society Journal, edited by Steve Jones (University of Illinois at Chicago, USA). The Journal focuses on "communication, media and cultural studies" with a focus on social media as it seems. It has an Impact Factor of 3.11. It was first published in April 1999 and is being published roughly every 4 to 8 weeks.

Larsson and Moes article studies political microblogging and the participation of Twitter users within. Their research is conducted in a case study about the Swedish general election in 2010. Larsson and Moe give a brief introduction and background about blogging, especially political blogging and microblogging, in this case, Twitter.

Through an online service Larsson and Moe collected Twitter data related only to the election topic by collecting data around the most used hashtag (#val2010). According to them, previous studies always collected data from the entire "Twittersphere" and thus being less focused on the topic at hand. They analysed that data with statistical software and illustrated it using visualization graphs. They collected data starting one month prior to the election and several days after. They found that about half of the tweets about the election were on the day of the election. However some prior spikes can be seen. Larsson and Moe linked these to other medial or political events.

They furthermore analyzed the communication between individual twitter users. Using the amount of directed tweets (directly aimed at another user using the @ sign) they categorized the users into Senders (sending out @tweets), Receivers (receiving @tweets) and Sender-Receivers (using both frequently) and thus are able to determine who is among the main drivers of the political debate. They further used retweets (forwarded messages/tweets from other users) to again put them into three categories: Retweeters (users that forward many messages), Elites (users, whose messages are popular and thus being retweet frequently) and Networkers (being retweeted a lot, as well as retweeting a lot). They found that while most users belong to the group of networkers, there are also some Elites who despite not being very active with retweets seem to have a popular opinion.

They found that most of the discussion is driven by a small amount of users. These users however are mostly public figures such as politics, journalists or well-known bloggers. Nevertheless there are also some anonymous users that according to Larsson and Moe "signal the potential for outsiders and less conventional voices to speak up via Twitter". They state however, that the overall share of twitter within the election is negligible as the "estimated number of Twitter users varies between 1 per cent and 8 per cent of internet users".

They admit that there are some limitations to their approach of collecting topic related data. They say that while the hashtag #val2010 in their case was used consistently throughout the period, some data might have slipped way due to misspellings or the hashtag being removed in ongoing discussions.

It seems that Larsson and Moes article is mostly coherent with the Design and Action theory (V.), providing the reader with a method of analyzing twitter discussions, with some EP theory (IV.) content - mainly the linking of spikes in the number of tweets with media coverage of the election and the suggestion of not only analyzing the structure of the tweets, but also the content.

By combining these two theory types Larsson and Moe are able to provide solid method for further research with a case-study at hand. Some more Explanation theory would have been interesting. For example by connecting the use of directional tweets with the use of retweets. With that and also more prediction they would have been able to establish a good theory according to Staton and Staw.

(1) American Heritage® Dictionary of the English Language, Fifth Edition. S.v. "theory." Retrieved September 16 2016 from http://www.thefreedictionary.com/theory
(2) Larsson, Anders Olof, and Hallvard Moe. "Studying political microblogging: Twitter users in the 2010 Swedish election campaign." New Media & Society 14.5 (2012): 729-747.