Last week I accompanied Dr. Linda Garcia and her son, Steve -- a business consultant who specializes in social network analysis -- to present our research at the Harvard Political Networks Conference.

The conference was hosted by the Kennedy School of Government and brought together a wide-range of academics. This included political scientists, MBAs, and computer scientists studying how social ties affect the behavior of government officials and citizens. Using social network analysis as a methodology, these researchers examined how the strength and structure of social connections can influence outcomes of particular events, e.g., voter turnout during an election, aid following a natural disaster, or the approval of legislation in congress.

Our poster titled, "Has Barack Obama Read Ron Burt?" analyzed the social network structure of the Obama Administration and whether, as Ron Burt argues, this structure allowed for the spread of good ideas. This topic (a brainchild of Linda and Steve) arose after all the media hullabaloo surrounding Obama's cabinet nominees and the resulting speculation over whether these people would bring change to the established culture of Washington politics.

For us, the short answer to that question was yes; the structure of the Obama Administration is in fact conducive to the spread of new ideas.

We arrived at this conclusion following a fascinating research process (fascinating for me, at least). And all the more surprising to me -- a former Communication major who had a long-distance relationship with math -- this analysis was gleaned through a series of mathematical algorithms, also known as social network analysis.

How does social network analysis work? Essentially it begins with the selection of key events in order to bound your network. We knew that there were many people who could be considered '"linked" to the Obama Administration, so we narrowed our list down to civilian-appointed, Senate confirmed nominees. Then we defined the events that we thought might connect these people together, events like education, past work experience, awards, social clubs, hometown, and current department. Once this data was gathered, we logged it in Excel matrices by event. Any time that two people were found to be connected, e.g., they had worked for the same company or attended the same university, they were considered to have a relationship and thus we added a "1" to the intersecting cell. Any time two people did not share a relationship within a particular event, they received a "0". Here's a quick glimpse of this method in all its nerdy detail:

Using these spreadsheets, Steve and his co-worker, Ben were able to feed the data into a system that combined each matrix, totaled each actor's sum, and presented a number that would fall within the spectrum of 0-9. This allowed us to mathematically calculate the strength of a particular relationship between any two people, which eventually evolved into this visual representation:

Pretty cool, huh? And we had some rather interesting findings, beyond just this pretty looking graph. Our most interesting one: Obama and two other members of his administration serve as cut points in the network, meaning they act as the glue that binds disparate groups of people together, people who might not otherwise be connected. This kind of structure reduces the connections between different parts of the network and naturally minimizes the spread of redundant information. People who serve as cut points are thus positioned to receive new information from various nodes in the network more efficiently.

Mathematical equations and matrices can help calculate the strength of social ties. Who knew? Well, apparently everyone at the conference. But for me, the academic newbie, this discovery was all too enthralling, and I can't wait to start using social network analysis to investigate other cultural phenomena. Perhaps I will start with that ever-elusive contemporary subculture that has served as fodder for many of my blog posts: HIPSTERS.