The online political arena is, contrary to popular assumptions, not a digital “Wild Wild West” where blogposts are irrational, emotional and one-sided, according to a new study by the Institute of Policy Studies (IPS) released on Wednesday (Feb 11).
One of the findings looking at the correlations between blog type and identity, objectivity, emotionality and partisanship showed that the more political a blogpost is, the more objective it is. This is “counter-intuitive”, as it goes against conventional assumptions that highly-political content tends to be irrational, said Dr Carol Soon, a research fellow at IPS and one of the authors of the study.
Additionally, blogposts with “low” political content tend to be more pro-Government or pro-Opposition. “High” political content blogposts, on the other hand, tend to be more anti-Government and also more anti-Opposition, Dr Soon noted.
Blogs with low political content are also more likely to post identifying information about their writers.
The study also looked into identity and its impact on how objective, emotional or partisan a blogpost was. It found that there is no relationship between a blogger’s identity and objectivity.
Dr Soon said: “This challenges the assumption that people who blog anonymously tend to be less objective.”
As for emotionality of the posts, the study found there was no relationship between emotionality and the author’s partisanship to the Government or the Opposition.
Of the 195 blogs studied, only 6.7 per cent were labelled as “5” – indicating the use of expletives. On the other end of the scale, 20.5 per cent were found to be “1”, or “very calm”.
- About 70 per cent of the blogs were not completely one-sided in their commentaries. A third scored between “3” and “5” in journalistic objectivity that offers alternative viewpoints in their posts
- Political blogs tend to be two-sided in their approach – perhaps an indication that balanced arguments are more persuasive than one-sided blogposts, said Dr Soon
- Journalistic objectivity and being unemotional go hand-in-hand
Mr Tan Tarn How, a senior research fellow at IPS and one of the study’s co-authors, provided caveats to the study. He said the length of time – June and July 2014 – was short, and that the sample size of blogs and posts studied were small. A total of 197 blogs and more than 1,000 posts were analysed in the study.
Facebook data was not factored in because access to the information was not readily available, the study’s authors said. Twitter data was analysed but the 140-character nature of the microblog meant meaningful analysis of a standalone tweet is not possible, they added.
Dr Soon said: “(The study) has enabled us to empirically establish that the internet or the online space is not as much of a ‘Wild Wild West’ that we have feared it to be. Moving on, if we can develop big data tools to help us up-scale the effort, we can use it to look at Twitter and public Facebook pages for instance. From there, we can hopefully get a more representative indication of the online space.”