Conversations in the Blogosphere

Came across a study today (Herring, Kouper et al. 2005)1 which investigated the extent to which and in what patterns, blogs are interconnected, taking as its point of departure randonly-selected blogs. They used "quantitative social network analysis, visualization of link patterns, and qualitative analysis of references and comments in pairs of reciprocally-linked blogs". They carried out the study because they claimed that findings by others e.g. Marlow (2004) that conversation among blogs is "completely connected" and Bloods requirement that a blog link to other blogs in order to be characterized as a blog focussed on a particular sub-type of blog, the so-called filter blog. However Herring et al claimed that only 12.6% of active blogs at the time were filters and that 48.8% contained no link to other weblogs at all. They were keen to investigate if the blogosphere would appear to be a completely connected coversation from the perspective of a random ordinary blog. Their results showed that most blogs contain no links or comments. Of those that did, filter blogs contain more links and link preferentially to A-list blogs, which are consequently overrepresented in the network. At the same time unexpected clusters of non A-list blogs emerged that were highty reciprocally-linked, forming topic-based communities. Direct manifestations of online conversation were found in both A-list and non A-list blogs, but varied in extent and nature according to blog topic and gender of writer. They concluded that the blogosphere is partially interconnected and that blog conversations, while occasionally intense, are the exception rather than the rule.

1Herring, S. C., I. Kouper, et al. (2005). Conversations in the Blogosphere: An Analysis "From the Bottom Up". Proceedings of the 38th Hawaii International Conference on Systems Sciences, Hawaii.