Social Network Analysis: Our Social Macroscope

[John McCreery]

I've been hoping that someone else would step up with suggestions for new bodies of theory that anthropologists should know about. Can it actually be true that there is nothing out there worth considering as we go about trying to understand the situations in which human beings find themselves and the paths our lives follow?

 

To me, the answer is a firm and resounding "No." That is because for the last four years I have been getting deeper and deeper involved in a project using social network analysis to guide ethnographic and historical research on the Japanese advertising industry, where I have spent a large part of my working life.

 

Why social network analysis? First, because it is intellectually exciting. As a subset of network analysis, social network analysis shares mathematical foundations with the study of any sort of system that can be described as composed of nodes and lines connecting them. Examples include not only the Internet and the Worldwide Web but also power grids, transportation systems, and protein cascades in cell biology. Could be the whole universe, with galaxies the nodes and the lines standing for gravitational or other fields connecting them. And to a kid now in his sixties with an interest in philosophy of science, networks look like the next big thing after taxonomies, standard causal explanations modeled on classical mechanics, and statistical inference based on random samples, as a new but equally rigorous way to organize knowledge. 

 

Second, because what social network analysis brings to my current project has led me to call it my social macroscope, a playful labeling that takes off from thinking about the biologists' microscope. What a microscope does in reveal structure in finer and finer detail to the limit of the microscope's resolution. It doesn't explain why those structures are there, but it brings them to our attention and poses questions that drive research. That is exactly what social network analysis does for me. 

 

I have a data set in which I have information on over 4,000 award winning ads, the more than 8,000 individual creators who were members of the teams that created them, and more than 30,000 roles connecting the ads and creators. I've worked in and around this industry for nearly three decades; but even drawing on that experience, trying to detect the structures embedded in this big, hairy knot of data takes more than my unaided eyes. Social network analysis software (I use a program called Pajek) helps me visualize what is going on. It allows me to use  node color and size and line color and width to highlight interesting structures and to extract subnetworks of particular interest for closer examination. 

 

It might turn out that structural properties of the networks I am working with explain part of what I see. What's more important to me just now is that doing the network analysis points me in directions that my other research needs to pursue. I can, for example, identify the most central creators in my networks and track their careers over time. Then I can read their books or trade press interviews with them and, at least in some cases, get to talk with them in person about what I think I've learned, both from the social network analysis itself and from the subsequent research to which that analysis has pointed me. I find this all terribly exciting. What about you?

 

What excites you, if not about social network analysis, when you think of ideas that might help you do better anthropology?