We should not consider homophily – “the principle that we tend to be similar to our friends” (Easley and Kleinberg 2010, 86) – an assumption of how people connect; instead, let’s think of people having attributes that may or may not explain how people connect. This may sound a bit pedantic, but the issue is to first take the reported structure for what it is, look at characteristics of the network, and then see if attributes help explain connections.
In fact, I’d like us to make attributes even less defining of people, but instead think about things in terms of affiliation networks. Many anthropological network studies take this approach (perhaps for methdological reasons), and we will see this later on when we talk about incidence matrices and bipartite graphs (as seen in the example of Figure 4.8).
For class workshop today:
Gephi download (for visualization) (may need to install Java as well)