No man (or woman) is an island, as the saying goes. We’re all part of a global network comprised of many, many smaller and interlinking ones where people are connected to each other based on location, friendships, profession etc. A social network is defined by the constructs of “nodes” and “links” (or “actors” and “ties”), which are applied to a community of people, which can be formal (a working group of scientists) or informal (friends). The nodes, therefore, can be individual people, while the links are the relationships that connect them to one another. Network analysis requires a relationship around which the nodes or actors are connected. For example, a co-authorship network is based around scholarly articles; if two academics are listed as authors on the same paper they are co-authors and the paper acts as a link between them. If you want to know more about social network analysis I suggest you read Social Network Analysis – Methods and Applications by Wasserman and Faust (which I’ve only started reading myself).
Networks of scientific communities are especially interesting to those of us involved in science communication. How do scientists work together in groups? Individual scientists can be viewed as part of the network when they share commonalities with other scientists: co-authored papers, working groups, conferences they both attend, etc. This is interesting for the science communicator when you realise that deeper analysis of these networks can shed some light on public perception of trust in the scientific community as this nice conference poster by Jeanine Finn demonstrates.
Finn’s paper describes how she carried out a network analysis of the co-authorship patterns (using the ISI Web of Science database) of both a group of 16 scientists who wrote an op-ed in The Wall Street Journal entitled “No Need To Panic About Global Warming” and the group of 38 scientists who responded with an article stating that, in their opinion the group of 16 were non-experts on the subject and that there is evidence to support the fact that human-induced climate change is occurring. Interestingly there is “almost no co-authorship activity” amongst the 16 critics while the network of 38 climate science supporters is “much more densely connected”.
Finn also carried out a hyperlink analysis of the scientists by using URLs that the scientist controls (personal blog, university profile page etc.) rather than, say, a Wikipedia page about the scientist. This hyperlink analysis, she says, can possibly detect those who may important in the science communication process but don’t go through the traditional channels of scholarly publishing. Finn looks at how they interact with each other online in term of whether they have linked to each other’s websites. These 16 scientists might not have worked together but they have connected online.
To me, this seems like an intriguing way of analysing how climate science actors interact, which, one would assume, has an effect on how the science is communicated to the lay (or non-expert) public. Perhaps this could be extended to their social media activity: are they connected to each other on platforms like Twitter and what does their extended Twitter network look like? It also brings up issues of trust and credibility. Would you trust a group of scientists – who have co-authored many papers together – to present a more cohesive picture of their field of research or a group who connected more by their online interaction than traditional scholarly avenues? Which comes across as more credible when weighing up scientific evidence?