Exploring Social Media Networks

How network analysis can make us more innovative in the study of collective traumas?

I’ve been exploring social networks, semantic networks, networks of one form or another in different contexts. Networks are everywhere! But, how can an understanding of networks help us to become more innovative in the study of collective traumas?

In this blog post I want to explore how a deeper understanding of networks helps to identify communities online extending collective traumas. In recent years, multinational enterprises taking part in alleged violations of human rights have been brought to public attention in the media and given rise to collective traumas that have mobilized the international community. Among the most publicized cases we have the operations of Bhopal in India, Shell in Ogoniland, Coca-Cola and BP in Colombia and Chevron-Texaco in Ecuador. The latter deserves special attention since it is considered a struggle of epic proportions and worldwide consequences which involves indigenous people fighting to hold Chevron, the inheritor of Texaco's legacy, accountable for have deliberately dumped more than 18 billion gallons of toxic wastewater, spilled roughly 17 million gallons of crude oil and left hazardous waste in hundreds of open pits dug out of the forest floor in the Ecuadorian Amazon from 1964 to 1990 (Aguinda v. Texaco, Inc., 2002). Actually, it is described as the “largest oil-related environmental catastrophe ever” (New York Times, 2010).

I’ll use an example of a network generated from Twitter, showing the follow relationships between people talking about ChevronToxico, the international campaign for justice in Ecuador, and the sharing links from the site. 

A network map of 110 Twitter users talking about and sharing ChevronToxico links

In this example of ChevronToxico Twitter network, I used NodeXL to extract, analyze and visualize the relationships between users who mentioned or shared a link from ChevronToxico. Nodes are people concerned about ChevronToxico. Edges are follower relationships, mentions and retweets between them.  They show who cares about the campaign, who participates in the debate and the direction in which information flows through this group of people. Therefore, it helps us to identify the communities leading the extension of the discourse about such a crisis, which is emerging in the public sphere as a collective trauma. 

Using the Girvan-Newman algorithm, I found that users are bound together in four groups or communities. In the first community we find the 50% of the users (55). Figures such as Mia Farrow, Rafael Correa, Wall Street Journal and Amazon Watch are the most popular. This is what I call a “mediatic community”. 

The second community has 37% of the users (41). Here we find less mediatic but more leftist figures such as Pablo Iglesias, a writer and professor who leads the left-wing political party “we can” in Spain, and Cayo Lara, a politician who leads the United Left (IU) also in Spain. This can be described as a “leftist community”.

The third community with the 11% of the users (12) deserves special attention since it involves the Whitehouse, this is the President Obama and his administration account, Climate Solution and the Schumacher Center for a New Economics, whose mission is to educate the public about an economics that supports both people and the planet. Are we in front of the traditional notion of the “American protective community”? We need to go deeper in this community structure to answer that question.

Finally but not least is the fourth community which seems more like a dyad linking two actors who are not part of any group (2%). Why? What is the role of these actors in the network? Questions like this will continue capturing my attention for the next weeks!