If you haven’t yet read the Sunday NY Times article on social networks (written by Clive Thompson) and their relationship to behavior, do it. Stop what you’re doing now (alright, maybe finish reading this post) and go read it. You’re going to love it.
To recap briefly…for years, social scientists have been looking for an extensive enough data set to more fully investigate the theory that certain human behaviors are “contagious”. That is, that our behavior is strongly influenced by the behavior of those around us. There are some really intriguing questions tucked into that theory…how does the shape of social network influence and affect the way people behave? How do information, gossip, influence, and opinion flow through a network and impact the people in it?
Enter the Framingham Heart Study. Starting in 1948, researchers began tracking individuals to investigate the common factors contributing to cardiovascular disease by following a large group of participants over a long period. There are a number of cohorts in the study, but overall, the researchers have followed more than 15,000 people, spanning three generations, over 50 years.
Turns out, in addition to their basic health and wellness information, the Heart Study’s researchers also collected the names of the participants’ family and friends. And it was that wonderful data pool that attracted researchers, James Fowler and Nicholas Christakis. Working with the Framingham Study data, they used the connections information to manually reconstruct the social ties of the Framingham group. Through painstaking work, they constructed a map of how 5,124 subjects were connected – friends, family, and work colleagues. Then they added in weight loss/gain data and created an animated diagram of the network, with each resident displaying as a dot that grew bigger or smaller as he or she gained or lost weight over a 32-year period. As the animation ran, they could see that obesity broke out in clusters. Check out the illustrations in this Wired magazine article about the research and you can see that groups of people would become obese, while other groups remained slender (same with the smokers). Whoa.
But they didn’t stop there. They continued to mine the data, finding more examples of contagious behavior. For instance, smoking and drinking seemed to spread socially. Even happiness and loneliness. But how does it work? Chritakis and Flower hypothesize that these behaviors spread partly through the subconscious social signals that we pick up from those around us. In other words, we decide what is “normal” from the people to whom we are most closely connected. That makes sense. But you have to wonder what comes first, the chicken or the egg….I mean, are you drinking because those around you are drinking or did you seek those people out because you are a drinker and they seemed like good company? The article points out the confounding factor of “homophily”, the tendency of people to gravitate toward others who are like them. Social relationships are so complex and it’s difficult to weed out where cause and effect begins and ends.
The scientists themselves agree that their friendship map isn’t perfect and there are many questions still to be investigated but there is clearly quite a bit here for us to chew on as we groom and cultivate our online social networking. For instance, one of their findings is that some social behaviors can skip links. That is, spread to a friend of a friend without impacting the connecting person. Hmm….And another of their findings – they discovered that behaviors appear to spread differently depending on the type of connection. They’ve identified something they refer to as “directionality” in a relationship. That is, I might identify Karen as my close friend, but Karen might not think of me in the same way. In that scenario, Karen is much more likely to influence my behavior than I am to influence hers. But if our friendship is mutual, apparently, the influence can be huge. That seems like a good one to ponder as it relates to the importance of adding value to your network connections.
It also seems natural to use this model to predict that the best way to change your behavior is to surround yourself with people who are also trying to change that behavior (or have already done it). So, if a teacher wants to change what she is doing with her students, she’d better find herself a like-minded networking group (or reshape her existing network) so as to surround herself with those reinforcing connections. Hard to do if you can only select from the people who live geographically close, but much easier to do when you have a live internet connection and tons of social networking tools at your disposal. Christakis and Fowler even point out that “you don’t need a lot of people, but you do need the right ones.” Amen.
Christakis and Fowler have published their findings in the New England Journal of Medicine and have book, Connected: The Surprising Power of Our Social Networks and How They Shape Our Lives coming out later this month.