Notes

[MUSIC]
So now we're going to get into some
very, very interesting material that
relates to what
we were discussing earlier about lead
users and
people who are important to reach out to.
We're going to examine that issue in
more detail by trying to understand how
information
spreads from one person to the next,
either
in an offline environment, or an online
environment.
And what it means for influence and
contagion to take place.
So let me just give a little bit of a
road map of where we're going to go with
this material.
So first
of all, I'm going to show you a very, very
controversial study.
And I'm going to ask you to click on the
link to play it yourself.
It's about a minute 45 seconds.
It's a study on the spread of obesity
through a network of people in Boston.
I think you'll find it interesting just in
terms of
understanding what a network really is and
what the elements are.
I'm making the summarize by definition
what things take place in networks, do
you need individuals, do you need them to
be connected and so on.
And then I'm going to talk about
neighborhoods.
So a neighborhood is just another unit of
analysis.
Instead of the individual, we're going to
think about locations.
And then we'll wrap everything together
with four examples
of research that have been done on the
following topics.
The first example is going to be some of
my own research on Internet retailing.
And we're going to look at how
a company called netgrocer.com spread
itself throughout
the United States through a process of
word of mouth, contagion, and so on.
Secondly, we're going to look at another
study done by some colleagues at UCLA
that tried to examine who's influential in
the social network, and who's not.
So think of that as being in Facebook,
or one of those other social networking
environments.
The third study is again going back now
into the off line world.
It's going to be looking at a network of
physicians in Los Angeles.
And what influences who prescribes what
drugs to their patients.
So, if I'm friends with Chris, Chris now
is the doctor, he's our all-purpose
videographer and he's prescribing a
certain kind
of drug to his patients, maybe because
he and I are friends or he's influencing
me, I'm going to prescribe the same drug.
That's a study done by some colleagues
here at
the Wharton School in conjunction with
another professor at USC.
Then the final study is another one of my
own.
This is also going to look at diffusion
for an Internet retailing company.
But this time, the company's called
Bonobos.
It's a fashion retailer.
So let's begin.
here's the controversial study.
And it's actually out of a book by some
professors at Harvard.
And I'd like you just to click on the link
and go through it.
But I'm just showing a visual here on the
screen.
Basically, what the study proports to show
is
that obesity spreads almost like a, a
virus.
Like a flu, or something else, that's why
it's controversial, from people who are in
a network.
And what that means is if I'm a friend of
somebody.
Who's perhaps struggling a little bit with
obesity
then I'm more likely to become obese than
somebody who's not in that social network.
That's why it's a little bit controversial
because it's hard for
us to get our heads around the idea of
something like
obesity, which is a physical thing, really
spreading more like a
virus which is a cold and we can imagine
that being transmitted.
So just play the video because I think
it's
a good way to understand some of the basic
concepts.
Again it's about a minute and 45 seconds.
Now that you actually watched that, let me
just sort of reiterate
what the elements of networks are.
And share what I think are some of the
most interesting research findings here.
So one of our colleagues at NYU, Sinan
Aral, who
does a lot of work in the area of
networks,
he says, really what a network is, is,
involves pathways
through which information and resources
and support flow between people.
So you've got to have a flow of something.
You got to have a connection.
And you must have people.
Or, we could also extend this to, to
neighborhoods.
Now networks can
either be physical.
So it could be a social network of people
who are
going to the same church or the same club
in Philadelphia.
Or it could be virtual, like Facebook,
LinkedIn, Twitter, and so on.
Now what's really interesting here and
what's written at the bottom
of the slide is that networks usually
exhibit something called homophily.
Now, that's a buzzword.
I don't like to be giving you guys too
many buzzwords.
But that's a good one to hang on to.
If you mention it at the next party you're
at, you'll be very popular.
Trust me.
So homophily means, birds of a feather
flock together.
So people who are using friends, people
who are your close associates.
Probably on average, are more like you,
than they are just like random people.
So people have similar cultural
backgrounds, similar
tastes, similar income levels, tend to
kind of
flock together, whether it's in a virtual
neighborhood, or whether it's in a
physical neighborhood.
And we'll say more about this later on,
because when we get to our third week,
we're going to be looking at various forms
of
advertising and communication, including
advertising over Facebook and we're going
to take advantage of that principle of
homophily to
send advertisements to people who are
friends of particular brands.
So just bear that in mind.
We'll be coming back to that later on.
So, here's some of the definitions of the
elements that you
need to know when you think about what a
network is.
So a network can be really, really simple.
It could be just two people.
Chris
and I, we're friends.
That could be a simple network.
So it could also be hundreds, or
thousands, or even millions of people.
I guess, by now, at least a billion people
are connected in one big social network.
And Facebook, I think even LinkedIn now is
over 225, 250 million people.
So in order for a network to exist you
have to have nodes.
Nodes can be either people or
neighborhoods or some other unit.
Normally we're thinking about them as
people.
You then need to have some kind of
connection between people.
And then also some ability to
share information, share resources, and
have exchange.
So if you want to see some more background
on that, again
I've provided another YouTube link for you
to be able to do so.
Okay, so we also decide, when we go into
a network what benefit are we going to get
from that?
So that's an important thing to keep in
mind too.
Is that entering a network, whether it's
joining a local club, going to a local
church, or participating in a
social network, is a choice and presumably
other people have made the same choice.
That's why we may get the birds of a
feather flocking together.
People who have similar interests.
we also decide the networks, whether
they're
real networks or social networks, virtual
networks.
How many people we want to be connected
to.
And then thirdly, at the bottom of the
slide, an important principle
is how embedded we are within a network.
So imagine, for example, that I'm friends
with Chris and
a whole bunch of other people at the
University of Pennsylvania.
And I'm connected to almost everybody on
the campus.
Let's imagine that's true.
If that were true, then I would be a very
embedded person.
Because I'd be connected to everyone who's
then connected to each other.
So in some sense, I might be more central
than other people in th network.
Okay,
so, a little bit of interesting research
that's been done I guess over the course
of the last 40 or 50 years, about how
networks work, and how influence works.
And so I'm going to share with you the
results
of famous study that was done in New York
City,
that shows that when influence happens, it
doesn't just
have to happen through conversations,
through Chris telling me something.
I might be influenced just by seeing what
other people are doing.
So way back in 1968,
some professors at Columbia did the
following experiment.
They put a person on a street corner in
New York
City and the person was, like this,
looking up at the sky.
And what they wanted to see was, would
other people around also,
upon seeing somebody looking up at the
sky, also look up themselves?
And they found in that particular
instance, when that happened about
4%, or four people out of 100 would also
just look
up, for no other reason, than that they
saw someone doing it.
Now when the experiment is put 15 people
around, just standing looking up at the
sky,
then about 40% of the people who weren't
part of the experiment also just looked
up.
So what that tells us is, it tells us
often times, there's pressure
to do things when people around us are
engaging in some sort of behavior.
So keep that in mind, that influence can
happen either through conversation.
Or it can happen through observing things.
And this is something that we'll come back
to quite
a few times as we proceed through our
study here.
another classic study of influence that
was done in
the 1960s, really, really helps underly
the key concepts here.
It was also repeated in 2002.
So let me explain what that experiment is.
So you might have heard of the phrase Six
degrees of separation.
It's a very common phrase, and it's often
used
to indicate that there's no more than six
people
between you and anyone else in the world.
So all of us out there enjoying the
Coursera
class are connected by no more probably
than six degrees.
So what do I mean by that?
Let me go back to the original study.
So in the original study, people in
Nebraska
were asked by the researchers, the
experiment, the experimenters.
To write a letter to individuals who lived
in a different city in the United States,
in
Boston.
So again, let's imagine they were asked to
send letters to people like Chris in
Boston.
Now the people in Nebraska, were not
allowed to take
the Boston phone book and look up the
individual by name.
What they had to do, was to write a letter
and send that letter
to somebody that they thought was closer
to the person in Boston than they were.
So imagine I'm living in Nebraska in the
1960s,
and I'm scratching my head and I say, gee,
I have a friend from college who's now
living in New York City.
New York City is closer to Boston than
Nebraska is.
Let me write to that person.
And then that person repeats the same
process.
And what they found was it took about
six steps, hence the term six degrees of
separation.
Now that same experiment was repeated by
some professors who did something called
The Small
World Project, you can look that up on the
Internet, in 2002, where they got 98,000
other people to communicate random people
around the world, communicate with them.
But this time not through physical
letters, but through typing and sending
emails.
So I might be asked to send an email
to someone who is an orthodontist who
lives in Finland.
I'm not allowed to Google that person and
email them directly.
I have to email somebody who I think is
going to
be more likely to know that person than I
am.
And again, it seemed to take about six
steps.
So isn't that fascinating.
We're all connected by about six steps.
Now that's connections but connections
don't necessarily mean influence.
It turns out that for influence, there are
really just three levels or three steps
that matter.
What do I mean by that?
So again let me think about my friend
Chris in the example here.
So Chris and I are friends.
We're a first degree connection.
So if I tell Chris, hey Chris, you should
order all your detergent
from soap.com.
He might copy me.
I might be able to influence his behavior.
And then Chris might tell his younger
brother, let's say.
So now that's two steps outside.
I had some influence over Chris's younger
brother.
And maybe Chris's younger brother's
girlfriend starts doing the same thing.
But I almost have zero influence
over Chris's younger brother's
girlfriend's college roommate.
So going out four steps is just a bridge
too far.
So that's another very,
very important thing to understand in
terms of influence.
Influence spreading out from you typically
will not go more than about three steps.
[MUSIC]