Social Network Analysis: An Introduction to Measuring How We are Connected

In the article ”Blaine Cook Introduces Us To Webfinger” Blaine suggested that, “Facebook is like a wedding from hell. Because it’s everybody you know and everybody you’ve ever met is just kind of hanging out. And if you were ever in a physical social space that would be like Facebook, it would be the most horrific experience you’ve ever had.” Despite having lists and groups the undifferentiated experience of using Facebook is a definite drawback for many users, myself included.

Most people’s lives are made up of involvements in many social groupings. Home, social, and work are probably the three main categories. The relationships within each of these spaces have their own associated considerations, responsibilities and obligations. While there may be a certain amount of overlap there is a distinct tendency to compartmentalize our activities in these different spaces. Having everybody you know lumped together on a social network like Facebook is not reflected in how we conduct our lives in the real world.

However much we make distinctions between the various social sets in our own lives it is still relevant and important to be able to measure how we are all socially connected to the world of people, groups and organisations that are around us and how in turn they are related to us.

In Stanley Milgram’s Small World Experiment we are shown that, for Americans anyway, only six degrees of separation – six other human contacts – separate us from anyone else.

Trying to get a handle on those connections is where Social Network Analysis (SNA) comes in. Originally, a sociological theory it is now a field of practical study with its own tools and techniques to describe our relationships, ties and interdependencies. In addition to its own descriptive terms SNA relies on heavily on graph theory. This is particularly important because once we get beyond our very immediate personal networks it becomes harder and harder to describe our interrelationships in useful ways. Families can use images of family trees to show how siblings, aunts, cousins and so on are related to each other. Businesses use organizational charts to show the deployment of responsibilities. However, beyond these tight knit internal representations can get dense very quickly.

Adding to the processing load are other factors which have to be taken into consideration. Here are just three:

  • Strength of ties: Not all ties are equal. Some are strong and some weak but weak ties can also have significant influence. A comment from a passing stranger or an unfamiliar speaker at a conference for instance can have effects disproportionate to our familiarity with that person.
  • Lots of connections in your networks doesn’t necessarily reflect influence: Bernardo Huberman from HP Labs has already shown this in his work on Twitter, ”How To Influence On Twitter: Research Results From New Algorithm Give Guidance.” What matters is where you are along the line between being at the centre of formed groups and cliques and being available to link to the otherwise connected. Being the bridge between a fully networked group and those outside that group is where the power of influence lies.
  • Vulnerability: Some networks can have a lot of connections to a single point. If that person, group or organisation is no longer available for being connected to it can result in the possible fragmentation and dispersal of the entire network.

An immediate application of SNA is in epidemiology. Being able to model and predict how diseases like HIV/Aids spread is vital for public health planning. Social networking sites can be an absolute gold mine for being able to monitor, measure and quantify people’s relationships.

The information derived from the applications of SNA techniques allows the social network operators to provide a better service and make the experience of the site more meaningful. It also provides useful data that can be used for marketing products and other commercial activities.

SNA applies to either persons or groups and the points where relationships connect are called nodes. In this article, ”Social Network Analysis, a Brief Introduction”, it is pointed out by Valdis Krebs that while all the action may seem to be occurring with those at the centre of a network there are real possibilities for change that come from those at the periphery.

Being less likely to be involved in the comfort zone of homogenous communication in a group other than their own, they can, by their position towards the edge of a group, create the opportunity to bridge one social network with another. These bridging nodes, whether they are individuals, groups, companies and so on are where the potential for the rapid expansions of networks lies.

Social Networks Analysis shows us that our connections are always dynamic in nature and alter in complexion according to how close to the centre or periphery we are in relation to others in our networks.

As social networks influence our lives more and more it will become more and more vital to be able to accurately measure and observe as precisely as possible what is going in that environment.

3 thoughts on “Social Network Analysis: An Introduction to Measuring How We are Connected

  1. Great article Tom.There’s also a study by Fowler and Christakis from 2008 which showed that happiness tends to be correlated in social networks: those nearby to a happy friend have an increased chance (25%) of being happy themselves, and this effect may even reach through three degrees of separation.


  2. Yeah I think this is an excellent overview. While we’re on studies, there’s one which found that acquaintances were better at finding you a job than close friends. “The strength of weak ties”. The rationale being that you tend to form a clique with close friends/family and you all tend to know the same people you do. wheras acquaintances exist in other cliques, and tend to know people you don’t know. Social networks are strange creatures.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s