Social networks have a life of their own

[Lucia Nass]

2011 was the year in which I discovered what social media can do for me. I’m intrigued by how using social media changes me, or rather changes the way in which I work and connect professionally. For the Dutch course I am following on social media and change, I read a book on the subject. Finding books like that is not easy in Hanoi. Ever since we lost a big box of expensive Amazon books, I have stopped ordering abroad. Luckily in the Bookworm in Hanoi, which trades in secondhand books, my eye caught the title ‘Connected’, and it has done more than meet my expectations.

Connected / The amazing power of social networks and how they shape our lives, Christakis and Fowler, 2011 paperback version (internet site)

The authors explain how social networks work, with highly readable scientific ‘sex-and-drugs-and-rock-and-roll’ illustrations. There are just two chapters about social media towards the end of the book, but I highly recommend it to my social media course mates. Social media are tools, and this book shows the potential we have to use those tools to our advantage. Our capacity to be networked developed long before the internetworked society.

My first take-away from the book is: Social networks have a life of their own.

There has recently been a lot of upheaval about Facebook making people’s lives miserable. Christakis and Fowler would probably point out that people who suffer on Facebook would also suffer without it. The new insight is that the misery does not just stem from the individual, for instance their lack of self-esteem, but is highly affected by their social network. What matters in not just the people we are connected to, but also our position in the network (more central or more on the fringe), as well as the structure of the social network around us (are our friends connected or not, and how are the friends of our friends connected).

The position we have in a social network is in our genes. Our genes do not just affect how many friends we have, but also whether we are located in the center or periphery of the network. People with five friends who know each other have a different genetic make-up than people with 5 friends who do not know each other. -- The latter people often act as bridges between completely different groups. --

We cannot understand the network by studying the individuals. We have to study the whole group and the relationships between the people in it. Compare it with being stuck in a traffic jam. Your car adds to the problem, but we cannot understand the jam by just looking at your car. It has only recently become possible to study vast networks and their behavior. ‘Connected’ explains what has been learned so far about the secret life of networks. It has helped me to more deeply understand the social network analysis metrics, which I blogged about here.

Studies of social networks show some very amazing ways in which networks typically behave or move. A few examples from the book may illustrate how these new insights into the behavior of networks can help change facilitators to stop or speed up anything that flows through networks, like information and norms.

• Network studies have shown that the spread of STDs is not being driven by a high-activity core made up of sex workers and their patrons, or by other high-activity individuals transmitting disease to a low-active periphery made up of individuals with one or fewer partners. It depends on the larger patterns of contact in the overall network. Techniques to visualize and analyze networks help tremendously to adjust assumptions about network behavior, and shape more effective prevention strategies for HIV and other STDs.
• A study on smoking behavior showed that in 1971, smoking had no bearing on social position. Yet today, smokers are on the periphery of their networks. It’s not just that they became less popular, they also tend to be friends with people who are less popular. Over time smokers have formed their own cliques within networks, with progressively less interconnections by early 2000. Cliques reinforce behavior in the group. Without much connection to the non-smokers in the overall network, smokers become more and more isolated of influences that may cause them to quit. The study also shows the role of high-status individuals in the diffusion of innovation. Smoking started in the 30’s and 40’s with posters showing smiling doctors enjoying and promoting tobacco. Similarly, the study revealed that when well-educated individuals were quitting, others were more likely to follow suit. In fact the more educated, the more likely it was that people would copy the quitting behavior.
• Note that the flow of influence does not pass through everyone equally. In the smoking example, education was a factor that affects the flow, but gender can also play a role. One study showed that if women start drinking heavily, both her male and female friends are likely to follow suit. But when a man starts drinking more, he has much less effect on others. It is not clear why, but women may be the key to successful intervention against alcohol abuse in this network.

The book is full of such examples, and suggests that network analysis leads to innovative, non-obvious, more effective change strategies.

Because networking is in our genes, we naturally form different types of networks depending on the purpose the network needs to serve. So each network is fit for a specific purpose, but may not be ideal for another goal. For instance, network structures explained why in one community a latrine construction program was effective, but in another it failed time and again. The community where the program failed was self-sufficient and closely knit, with almost no relationships to strangers. The community where the program succeeded had frequent interactions with others. Such integrated networks are more likely to show pro-social behavior. In the case of latrine construction, everyone in the community must use a latrine for good health benefits. This requires people to care about others, and it appears that the ability to care about others and to adjust your behavior for the greater good of the community is weaker in closely knit networks.

This may vote well for the future, as we increasingly open up our networks to others through the use of social media. Will more connections with strangers generate more caring behaviors? The book shows that we already see this type of behavior emerging. Thousands of vigilantes voluntarily patrol Wikipedia to prevent malicious edits, and gang up to stop mean and hateful users from making further changes.

Technology clearly changes the way we connect. Although people initially felt that the telephone would lead to more shallow relationships, much like the fears with social media today, fact is that the telephone expanded and strengthened local ties, and enabled us to be more social and cooperative. The internet may do the same, as research in Canada shows. In a completely new suburb of Toronto 60% new homes were equipped with free broadband internet access. The residents with access developed deeper and broader connections to other residents in the neighborhood, and put their connections to collective use. They for instance effectively pressured the suburb developer to fix defects of their homes. Compared to non-wired home owners, the technology also helped to preserve relationships people had before their move, thus counteracting adverse impact of the move.

Connectedness also comes with a price. There have been a number of STD outbreaks related to internet use. Networks do not just carry good stuff: violence spreads in networks, as does suicide, anger, fraud, fascism and so forth. The authors call this ‘side effects’, and argue that the purpose of our social networks is to transmit positive and desirable outcomes.

And that leads to my second take-away from the book: To enable social change, we must help those at the periphery of the network to connect with those at the core.

Positional inequality occurs not because of who we are, but because of who we are connected to. The book shows how our connections often matter more that our race, class, gender or education. It suggests that to address differences in education, health or income, we must address the personal connections of the people we are trying to help. When we target the periphery of a network to help people reconnect, we help the whole fabric of society, not just the disadvantaged individuals at the fringe.

Understanding social systems is not just about the individuals, nor is it just about the whole. Interconnections between people have more effect than we ever realized. Together we form networks, superorganisms that are more intelligent than the sum of its part. I look forward to discover more about how and why we are all connected.