fredag 22 november 2013

Theme 3: Research and Theory

Journal

Journal of Computer-Mediated Communication

The Journal of Computer-Mediated Communication (Impact factor 1.778, 2012) focuses on social science research on communicationg with computer-based media technologies. Within that general range it is broadly interdisciplinary and publishes work by scholars in communication, business, education, political science, sociology, psychology, media studies, information science as well as other disciplines. Original research articles and meta-analyses of prior research are acceptable formats for submission in JCMC.

Article

Campaigning on Twitter: Microblogging and Online Social Networking as Campaign Tools in the 2010 General Elections in the Netherlands

The study focuses on how candidates in the Dutch general elections of 2010 use Twitter. It also focuses on explaining why some candidates are more prone to adopt Twitter, have larger networks, and show more reciprocation than other candidates. Twitter usage is described as a counter-measure to deal with declining interest and participation in the political processes in the Western world. It is used to decrease the gap between the politicians and citizens, as well as an increase of visibility and interactivity between which might lead to increased political involvement. The study aims to describe how candidates use Twitter as well as to find explanations for differences by using party and candidate characteristics.

The first step in their strategy of analysis was to look at the entire set of candidates to try to explain why candidates subscribe to Twitter. To estimate these effects they used logistic regression analysis. The independent variables on these analyses (number of tweets, followers and following and reciprocal relations) represent count data, which is why Poisson regression was decided to be applicable. Some of the results included that there were no clear pattern if the adoption rate was higher or lower within parties with more seats in parliament, and that the more followers a candidate had the more tweets he/she sent out.

A problem with the whole article and its findings is that there are a relatively small number of test subjects (Twitter users) that are being evaluated and Twitter usage can differ a lot between cultures, countries and time, which makes it difficult to use the results in another scenario where parameters such as location and time are changed. It is impossible to say that "this is how politicians use Twitter which will lead to this". 

Theory

Theory aims to explain, inform or provide logical grounds and to describe, explain or help to understand a phenomenon. Theory can use and be backed up by references, data, lists of variables or constructs, diagrams and hypothesis (or predictions), but these are not theory by themselves since they have to be logically interpreted and evaluated.

Since my article is trying to predict political success based on the candidates’ observed characteristics, based on collected data, on Twitter I would say that the theory type falls into the Prediction theory. Prediction is a statement about the way things will happen in the future, often but not always based on experience or knowledge. A prediction may be a statement that some outcome is expected, and may cover a range of possible outcomes. Prediction – even if it doesn’t provide guaranteed information about the future – is necessary to allow plans to be made about possible developments. 


Limitations include that these kind studies are not very applicable on a broader scale, since they are often very specific. This was certainly the case in my article that only focused on a specific election, in a specific country with running candidates that specific year. However, this kind of research can provide information for further exploring in the specific area.

2 kommentarer:

  1. Interesting topic. As you day, there' s no way you could define the political outcome of the usage of Twitter. However, if one monitors the link between this usage and political popularity one can establish findings that the reader may interpret freely.
    Not so long ago BBC published a documentary on the link between ruling governments and the price of onions in India- and how the price of onions seemed to directly correlate to how people found their government.

    SvaraRadera
  2. Hi. I agree with your thought about the problem with the study, that it depends on location and time. Though as you mentioned, the article is supposed to only cover the Twitter usage of Dutch politicians, or no? It would be good if the study was only focused on the Dutch politicians. Further studies could then be to do the same study but in another country and then make some kind of comparison.
    As a side note, I think it's interesting in our case how Carl Bildt is very active on Twitter. I think it's a good idea for politicians because as you mentioned it decreases the gap with the citizens for example. I also remember how Carl Bildt had some arguing going on with an another politician (from an another country) on Twitter, which was talked about in media and I found quite funny..

    SvaraRadera