Book summary of ‘Connected’ by Nicolas Christakis & James Fowler


The surprising power of networks and how they shape our lives – How your friends’ friends’ friends affect everything you think, feel and do 

By Nicolas Christakis & James Fowler (2011) 


Content = ****   Readability = *** Clarity & Structure = *** 

Summarised by Paul Arnold (Facilitator and Trainer –


Connectedness matters. We are subtly (and often unconsciously) influenced by people around us (and our friends’ friends’ friends) – the power of influence extends out by 3 degrees – and whilst the impact weakens, the geometric number of influence points grows. These influence many parts of our life (including our beliefs, our health, our careers and how we feel) – some for better and some for worse.

The book explores the power of connections through a number of different perspectives (such as relationships, emotions, politics, economy, health, evolution, the digital world and systems). 


How connections influence our lives (an over-view)

We live in interconnected networks. 4 key rules:

1)             We shape our networks and our network shapes us – We become like the people we spend time with. Transivity (i.e. the amount of connections we have) affects the quality of our lives (it influences our expectations, the sort of people we marry, where we live, the sort of jobs we get, our emotions, our health – even to the likelihood of committing suicide).

2)             Our friends affect us – We often copy our friends. Friends give us permission (and safety) to do things. If a friend has done something/bought something/been somewhere, then we are much more likely to also do it/buy it/go there.

3)             Our friends’ friends’ friends affect us – We are influenced by what our friends do – but also by our friends’ friends – and surprisingly by our friends’ friends’ friends. Likewise what we do echoes out through three levels of friends before it loses its energy and impact.

4)             Networks have a life of their own – No one controls or owns the network. It is complex, dynamic and constantly evolving. Cf how a flock of geese has no leader but it self organises. It has no central control point but rather a ‘shared intelligence’

Some people are on the edge of networks, others at the very heart of them.

Some people have lots of connections with others networks, other are more insular.

How connections influence our emotions

Emotions are contagious

Emotions are a genetically inspired way of quickly spreading information that people pay attention to.

Tanzania 1962 – in a mission boarding school near Lake Victoria, there was an epidemic of laughter, which affected over 1000 people. It began on January 30th and lasted for 9 months.

Certain people are more susceptible than others and likewise, certain people are more influential than others.

Likewise in teams, emotions quickly spread – And when a team is happy, its been shown that performance improves.

In a study of happiness, unhappy people cluster with other unhappy people (and vice versa).

Furthermore, unhappy people seem more peripheral in networks.

It’s not just that happy people prefer the company of happy people, it’s that happy people make other people happy (and vv).

Analysis suggests that a person is 15% more likely to be happy if directly connected to a happy person. At 2nd degree of influence (i.e. a friend’s friend) its 10% and at 3rd degree its 6% (but no effect at 4 degrees).

People with friends who have lots of friends are also more likely to be happy.

Also the closeness of happy people affects us.  When a friend lives less than 1 mile away becomes happy, it can increase your chance of becoming happy. Consequently, the people one spends the most time with heavily influence ones mood.

Likewise loneliness begets loneliness. People who feel lonely will tend to lose about 8% of their connections a year. Also proximity to other lonely people also increases ones tendency towards loneliness.

How connections influence the quality of our relationships

68% of people met their spouses after being introduced by someone they knew. Over time, spouses become increasingly similar to each other.

How we look affects how we are treated. But it’s also true that the way people react to us either builds or weakens our self-valuation, confidence and esteem. Thus networks can heavily affect our personal identity (we often live up to – or down to – the expectations of others).

Social pressure and compliance – we are heavily influenced by others

The power of social compliance is often underestimated. Muhammad Yunnis of the Grameen Bank saw that people would never lend to poor Bangladeshi’s, as they had no collateral to hold against the loan – but by lending to small communities, the power of peer group pressure ensured the loans were repaid.

Wikipedia is an open system that anyone can edit. What is interesting is it has no centralised control and like many self-organising teams with no formal authority, manages to resist abuse by applying self-policing, peer group pressure.

Comparisons to others are a key factor in determining contentment. Economist Galbraith once commented, many consumer demands stem not from innate need but more from social pressure.

People would rather works for a company that paid them $33,000, where everyone else earned $30,000 than to work at a place where they were paid $35,00 whilst the average was $38,000

Robert Merton (social scientist) suggested we use reference groups to help make decisions.

Our best friends influence how we perceive prospective partners attractiveness.

Experiment: People will cheat in an exam if they see others doing it – especially if they identify with them (‘they are like me’, so its okay to do it – cf the UK summer riots).

Social pledging – a successful way of increasing compliance to the stated behaviour.

In 80% of cases the reason we give to charities is because people we know asked us. Katie Camen studied charity giving by 75000 employees and found that employees gave more when they worked next to someone who was very generous.

Merely observing another person’s behaviour (especially someone we admire) can be as influential as words (cf Gandhi: ‘Be the change you want to see in the world”).

How connections influence our health

Every month 11m people play World of Warcraft – a number greater than the population of Belgium. In 2005 the game designers thought it would be interesting to build in an infectious disease  (called corrupted blood) that players could spread to their enemies.  Unfortunately, rather than operating in one isolated area of the game it rapidly spread and infected the whole game. Eventually they had to switch the game off, clean out the infection and then reboot the system. The spread mimicked how real diseases could spread and many health organizations have studied what happened in this virtual world.

Germs are not the only things that can spread – behaviours can as well. We are influenced by what other people do as they give us ‘permission’ for that activity.

The biggest predictors of grades in US Universities are the grades of the other people in their dorm.

When people sit next to a person who is over-eating they will also tend to eat more.

An obese person has more friends, friends of friends and friends of friends’ friends who are obese than would be expected by chance.  If a mutual friend becomes obese, it nearly triples a person’s risk of becoming obese.

Anorexic role models like Calista Flockhart, Kate Moss and Victoria Beckham may influence but our immediate friends are more influential.

Groups such as weight watchers and alcoholics anonymous form powerful social networks of influence – it’s the people not the techniques that drive success. An individual’s success story echoes throughout the network. This provides a positive reference experience of success, so building their own belief that success is also possible for them.

Network science can identify those influential people at the heart of networks, so they can be targeted. Education increases influence (both through knowledge and credibility/ authority).

We are heavily influenced by the norms of our society/group. – We unconsciously copy others to be part of that group. The more we relate to a group or want to be a part of it, the more influential their norms will be upon us.

We are more influenced by people of the same sex. And people at the centre of a network have more influence than those on the periphery.

Even some illnesses can be influenced by networks. Back pain is a culture-bound syndrome (e.g. before the Berlin wall fell, East Germany had much lower rates of back pain – but after reunification the rates converged).

Likewise we see contagiousness in suicides. Sociologist, David Phillips showed how suicide rates increased in the month after a front-page article in the NY Times – and the more descriptive and celebratory of the person’s life, the greater the likelihood of increased rates of suicide (as greater potential for identification).

The research suggests that younger people are more susceptible to influence (the rates drops away by age 24). There is also greater likelihood of influence if the suicide was within ones’ three degrees of connection.

How connections influence our economy and personal prosperity

The banking crisis – people get swept along by the wisdom of the crowd (where we assume that because everyone else is doing something it must be okay).

The value of something is defined by what other people think its worth – e.g. eBay. If you put in a bid for an item, and then someone else bids more for it, then that new bid has confirmed that someone else thinks that item is worth that amount of money – which then encourages you to put in a new bid and so on….

For networks to be effective they also need ties into other networks – e.g. those people who have a number of different circles of friends/acquaintances are the critical connectors that allow information to flow between networks (cf Non-Execs).

Uzzi studied Broadway successes and flops – the best mix was a core that had worked together before, but with some new people who could bring in fresh ideas.

Networks that are more insular are less equipped to solve novel problems than those networks with lots of interconnections with other networks.

The circles we live and work in are fundamental to the opportunities and quality of lives we get to lead (so its less about absolute ability as it is about the connections one has around you as they create the initial opportunities and levels of expectations) – the old adage “its not what you know, its who you know” is still true.

How connections influences politics

Part of Obama’s success was down to money. Through use of social networks he raised over $600m – from more than 3million people – just $20 each – because he managed to ‘connect’ to the average American. Furthermore, he used the Internet to communicate directly with people (rather than have his views expressed through the mouthpiece of journalists) – hence he was able to successfully distance himself from the controversial Pastor, Reverend Jeremiah Wright.

Whilst no one vote has ever swung any election, a decision to vote increases the likelihood of others to vote. Coming up to an election we will hold c20 discussions about it. If one of our ‘discussions partners’ votes, we are 15% more likely to vote ourselves. This also follows the rule of 3 degrees of influence as the impact ripples out. One person’s decision to vote will influence three others to vote.

It appears the less innocuous and less pressurised the flow of information, the more open and hence susceptible to the message one is (so the hard-sell door stepper is less influential than the casual chat over the garden fence).

Research by Lazersfeld and Berelson has shown that online social networks are homophilic (i.e. they cluster around shared likes/values/beliefs) and strongly polarized. Information is used more to reinforce existing opinions than it is to exchange differing points of view. A study of interconnections between blogs followed clearly illustrates this political divide:

The more powerful the connection one has with a person, the more ‘influential’ one can be. Research has shown that highly ‘connected’ politicians get through three times more pieces of legislature than do lesser connected politicians.

Furthermore its useful to have connections with people who themselves are well connected – thus one person may only be connected with say 10 people, but if those 10 people have 100 contacts each and each of those have 100 contacts each then that creates a potential network of 10,000 people.

How connections and part of our evolutionary DNA

Human social network behaviours are hard wired – its genetically conditioned.

Networks have been fundamental to the advancement of the human species. People who worked together were able to kill more prey and were able to protect each other against predators (human and animal). Thus the ‘connectors’ survived better than the loners.

Research has shown that selfless giving is genetic but with cultural influence.

Our brains have developed in a way that deals with relationships. The larger the group, the greater the complexity. Neuro scientists have found we use a large part of our brain to monitor social interactions. 2/3rds of our brain’s capacity is tuned to detect differences in skin colour – a critical sign to people’s emotions.

The maximum size of social groups is 150 – it’s the number where every member can know everyone and can successfully manage to maintain a relationship with each group member (and that allows the unit to operate in a co-ordinated manner).

How connections are evolving with the Internet.

The Internet is creating new ways to connect and share (e.g. Foursquare uses digital technology to tell people where their friends are so they can meet up physically). We are now hyper connected, sharing large chunks of our daily lives with a wide group of friends – thus we know more about more people.

Will these social network relationships replace our deeper personal connections? Research suggests that like the advent of the phone, these technologies supplement the development of relationships rather than supplant them.

The average number of ‘friends’ a person has on Facebook is 110. Research (by looking at mutual tagging of pictures) suggests the real level of close friends is 6.6 (thus the internet does not significantly expand our circle of close friends).

The Internet has allowed the development of new types of connections that were not previously possible.

Sociologists Keith Hampton & Barry Wellman studied a housing estate of 109 new homes in the suburbs of Toronto when it was first connected to broadband. 60% were connected and 40% chose not to be – thus allowing a natural comparison. Residents who used the Internet knew more of their neighbours by name, had made more phone calls to them, and had visited their houses more often. It also facilitated the mobilization of community activities, events along with the development of a pressure group against the builders.

The Internet has created the opportunity to connect up with people around the world who are interested in a very specific activities/interests (such as people with specific health issues but it can also be used for some less palatable groups such as self-harmers, anorexics, suicides and bomb makers). The connections with each other reinforce their belief system and legitimise their actions.

On January 2008, Oscar Morales, a 30-year-old engineer from Colombia mobilized millions on his social network. He started a group on Facebook (called ‘No More’) with just five friends protesting against the holding of hostages by the military group, FARC. Its numbers swelled to 272,578 within 4 weeks, and on Feb. 4th 2008, 4.8m people in different places in the world (inc Sweden, Spain, Argentina, France and USA) marched against the hostage taking.

The spread of information and the collision of many different perspectives are key to the process of innovation. The Internet has allowed the rapid spread and co-creation of ideas (such as Linux and Firefox and Wikipedia).

Digital connections are increasing both the number of connections we have but also the speed that unedited information can pass around (cf Wiki-leaks).

Online provides new avenues for influence and social contagion, but the spread of emotions needs face-to-face communication. Overall the evidence suggests that online networks are less influential than real life relationships.

The circles we move-in are highly influential on the quality of life and the opportunities we have.  People with more ties will get even more and those with few ties will get left behind – thus the Internet merely amplifies their current situation.

The power of connections as a super system

We achieve more in groups than we can ever achieve by ourselves. And co-operative interactions are the hallmark of all major evolutionary leaps.

Social networks have greater intelligence than the individual. They can capture and contain information that is transmitted across people and can perform computations that aggregate millions of decisions (such as setting a market price or deciding who will be elected).

As a connected system, we act more like a super organism (cf ants). And like a super organism our networks are self-replicating and self-annealing (they self-repair).

Whilst one end of science is focusing on deconstruction of the whole – reducing it down from life to organs to subatomic particles, there is another branch of science which is taking a macro view and looking more at total systems approach – seeing earthquakes and market crashes as bursts of activity in a larger system.  The recent economic problems were part of a globally interconnected economy.

The beauty about social networks is that more of us are more connected than we have ever been before – and the power of the unified super organism allows us to achieve major breakthroughs much larger than ever seen before in the history of man.

Thus, the key thing we need to do is … connect.


The book is well researched and is full of examples to support their thesis.  The thesis itself is hardly revolutionary but its potency does help us to think of both how we are influenced and how we might influence.

Clearly there are many other forms of influence that operate on us (as well as counter behavioural activity such as losing weight, sad people etc etc).


About slooowdown

Consultant in the fields of Relationships and Change
This entry was posted in Advertising, Behaviour change, Leadership, Management, Team building, The power of great relationships, Transformational teams, Uncategorized. Bookmark the permalink.

4 Responses to Book summary of ‘Connected’ by Nicolas Christakis & James Fowler

  1. davidmerzel says:

    Reblogged this on David Merzel's BLOG.

  2. Pingback: Three Factors That Encourage Healthy Eating Habits | WGRC

  3. John Smith says:

    Are the part written in italics quotes taken straight from the book itself?

  4. slooowdown says:

    I use italics to separate out stories/examples which are interesting but not central to the thesis

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 )

Google+ photo

You are commenting using your Google+ 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