Social Network Analysis

Have you read Thomas Kuhn's The Structure Of Scientific Revolutions? It's appeared on lists of the most influential books of the last century; it would certainly sit near - if not on - the summit of my top ten. Here comes the 'in a nutshell bit'. Pay attention. 

Kuhn said that science isn't this objective, rational pursuit that many like to think it is. He argues that scientists are subject to the same social forces that affect the rest of us and this colours their practice. Thus they work within an accepted framework (or 'paradigm') and ignore anomalous information that's occurring outside the paradigm. For example when scientists working in the framework of Newtonian physics found anomalous results (curvature of spacetime) they put it down to measurement error. He suggests that it takes a great deal of social pressure to get scientists to adopt a new paradigm (after all they've invested so much in the old one). He termed this framework transition a 'paradigm shift'.

Once I'd learnt about Kuhn and paradigm shifts and such (as an undergraduate) I took  to bugging the hell out of my lecturers. I kept challenging them to point to the paradigm shifting work that was occurring in our discipline, for it was here that I considered the exciting stuff existed. Paradigm shift as growth curve, if you like. Here's a more recent example: the internet was a paradigm shift. Good old Bill Gates was stuck in the old manner of thinking when he initially suggested that it wasn't going to amount to much. That was one huge misjudgment, but he's human and subject to the same cognitive processes as the rest of us. 

Just before we get too carried away always remember that a lot of the anomalous stuff will not quite make it to shift status: Google Wave anyone?

So without further ado here's my contribution to the soup of anomalous phenomena that's buzzing around current ITSM thinking: Social Network Analysis.

SNA doesn't relate to the internet-based social networking tools that we're all familiar with (although we can perform analyses on the data that these collect). Rather it's a method of analysing social interactions; for example people that you talk with during the course of your day. Here's what you do:

1. You capture the social interactions of all the people that you're interested in. As I'm blogging about ITSM let's use as an example the IT service support staff in your organisation. You can simply administer a short questionnaire asking them to create an ordered list of the twenty people they interact with most in a week. You could also add to this PABX telephone data which will detail who they talk to on the phone. You can go further and include email sent to and received from headers. Going even further you could use the connection information from your corporate instant messaging application?
2. Now you have the data of who interacts with whom you can then plug this into a social network analysis application such as UCInet. This will do all the complicated maths on the data and determine where you have structural holes and positions of influence and power and similarities. You can even plot a map of the social network so you can get a visual of your organisation.

What will this tell you? Well it will give you a calculated picture of where informational power lies, and perhaps how communication needs to be re-configured to ensure that knowledge is more evenly distributed. And it can do more - research has shown that individuals who occupy similar positions in different parts of a social network tend to have similar views. Crazy but true.

Here's a SNA plot of Incident data that I analysed (at the workgroup level) from a huge service support organisation. The data was simply taken from the service management tool. You will notice two huge bottlenecks. Shortly after this analysis one of those bottlenecks was removed, resulting in a speedier flow around the network. I could go on, but I'll leave you with a few other interesting SNA plots (courtesy of Visual Complexity). So you decide: is SNA a genuine anomaly or the beginnings of a paradigm shift in the methods we use to understand our ITSM organisations?