Theme 4 - Quantitative research

My chosen research paper

For this week’s theme, I chose an article from the journal called “Journal of Computer-Mediated Communication”, which has an impact factor of 3.541. The article is called “The Mediation of Politics through Twitter: An Analysis of Messages posted during the Campaign for the German Federal Election 2013” – and as the name suggests, the article is about analyzing the political trends on twitter, in order to get an understanding if there is a correspondence between how people are tweeting about the election, and the actual vote. This was done by analyzing 1391187 tweets from 98166 users, using specific hashtags and keywords to find what is being tweeted.

The positive aspect of using this above mentioned method is the vast amount of tweets that you can analyze, which gives you enough data to statistically support different hypotheses. However, which the authors themselves are very open about, the fact is that there might be only be a small niche of users on Twitter (e.g. more students than retired people, more unemplyed than emplyed etc.) due to the fact of the technology which is used. Because of this, the quantitative data will be skewed in favor of the specific target audience of Twitter.

From the article, I learnt a lot about using open databases (in this case Twitter has an open API which you can fetch data from using specific parameters) and how this can be openly used in research papers and analysis of this kind.

One of the main methodological problems is the fact that the authors are analyzing one social medium and just one election – which might not be enough to draw general conclusions or predictions about the hypotheses at hand. However, since the authors want to analyze the question if there is a correlation if the twitter data supports the actual election numbers the methodology is a good way to disprove that statement.

Drumming in immersive virtual reality:

This article was a very interesting point of view of virtual body ownership. By changing how the avatars used in a virtual environment look, the article wanted to examine the change in behavior of the users. Different models changed the perception of the users, and with the change in perception – the behaviors did aswell. A question that I had was regarding the design of the two avatars which were used – why did they have the FL and CD avatars which were both male? I would have loved to see the behavior change when experiencing the virtual environment as the opposite sex, and how the perception would have changed from that.

One of the benefits of using quantitative methods is the data can be quantified and statistically applied to verify your hypotheses. Data in a critical mass (a certain amount of data) can be used to uncover patterns. Data is also quite easy to analyze nowadays with the high availability of analytical programs such as Google Spreadsheets or Excel. However, pure data might not take into consideration the biases of the numbers – it might not explain why the data is skewed in one direction (which could be that the data originates from a specific audience which has a certain perception of the problem being analyzed) – to clarify, the data itself does not put the numbers into a context.  It’s good for generalized results and to test hypotheses – however, this requires a high amount of data.

In contrast, qualitative methods is heavily limited by the narrow amount of data that you achieve. Due to this, it cannot be used to draw generalizations of a subject. However, when quantitative and qualitative methods are used in a combination – you get the opportunity for a much higher understanding of a phenomena at hand. By first using quantitative data by (for example) surveys, data dumps etc – and then applying a qualitative method to that data, you get a much higher understanding of the phenomena at hand. This even opens up to new perceptions which you might not previously have thought of.

To sum up – both qualitative and quantitative methods have clear benefits and downsides, but when used in a combination they can be used to gain a much higher understanding of a specific phenomenon.