0:01
Okay, so we're still talking about our structure of networks and so forth, and
now I want to say a little bit about dynamics and things like tie strength,
so, we've been looking pretty much at, at networks as static objects with zero one
kinds of relationships So I want to just mention that, you know, obviously in, in
looking at real data, networks are going to be changing over time, there's
going to be relationships which come and go relationships can be stronger or
weaker, and so let's just say a few things about that.
And in particular, we're still sort of within the part of our course which has
to deal with background and fundamentals. And, you know, talking a little bit about
the, the characteristics of networks and so forth.
And, what I'll do here is just give you a couple of links.
these are links to, James Moody's website where he has, pictures.
In particular these, the first one that's highlighted.
as pictures of this romance relationship that we, looked at in the high school
data before. What's important about that is, is it
will show you these changing over time. So it's not as if all these relationships
were present in one point in time. people were having relationships with
different people at different points in time.
And understanding the dynamics of this. The frequency with which things happen
and so forth, this is sort of an on going of area of research where there's
increasingly rich data becoming available.
People are struggling with ways to actually code it and take care of it.
Because it has real consequences for things like contagents.
So, the frequency of, of interactions that the individuals will have, in ,
important implications for fuse, what kinds of travel patterns, uh,uh, exists,
communication patterns. When we talk about diffusion of products
and things like that, dynamics will play n important role.
So I, I just want to sort of point that out at this point.
I don't have much more to say about it yet, we'll see a little bit more of it
as, as we look at different examples in the course.
Now the other thing that we have touched on briefly is that fact that ties aren't
all zero one. It's not as if I have the same strength
of friendship with all of my friends. Some of them I talk to on a very regular
basis, some I don't talk to very frequently, some can be very strong
emotionally, even though they're less frequent, and so forth.
and sort of, a very important study of this and, and, one that sort of made
clear the strength of weak ties Was a study by Granovetter and uh,[COUGH] , in
the early 1970s, and what Granovetter did was interview a series of people and
asked about how they found their jobs. And the interesting thing that
Granovetter found was that out of the 54 people that he interviewed, one sixth or
the more than 16 percent had found their, had, had, had found their jobs through
strong ties. And what does a strong tie mean?
In Granovetter's definition was that they had at least two interactions per week.
in contrast there were 50, roughly 56% of the people that have found their jobs
through contact with which whom they had a medium tie, at least one interaction
per year but not two interactions per week so not a strong tie.
And then weak ties still accounted for 27.6% of the job contacts and they were
less than one interaction per year. And so what grand order was pointing out
was that even though these are very weak ties in the sense that you don't have
much interaction with htese individuals at all they still acounted for a very
large, more than a quater. Of the information about jobs that people
were getting in terms of, of how they, managed to find their, their jobs.
No, this has led to a lot of follow up research and so forth, and understanding
weak ties, and understanding when they're important, and when they're not, and, and
how important they are is a, you know, is a, a, loud subject.
Let me say a couple of things here. one is if you begin to think about the
number of people with whom you have strong ties, then you interact with them
regularly, its, you know maybe dozens, maybe even fewer.
if you think about the number of people that you've known in your life and that
you could actually talk to if you actually saw them.
that might be on the order of thousands. Right?
So you have many, many weak ties and only a few strong ties.
And so, even though weak ties are, are people that you might not see very
frequently, or in rec with very frequently, it could also be that there's
just many more of them. And so, you know we have to put these
things in context. Well, what one, one part of, of
Granovetter's, discussion was about the role of weak ties, and one point that he
was making was that people with whom you have less frequent interaction or, or
less strong interaction, tend to be people that might connect you to another
part of the network. And we already saw a little bit of this
when we looked at our add health data. when we look at, weak ties in the sense
that people just had to name each other as friends we saw some connectedness
between, different groups. And if we put in strong ties those the,
the ties that tended to go across, the different groups had now disappeared,
largely, and so there's different patterns and so that more of the ties
across the races tended to be ones that disappeared once we put in the stronger
notion, and the idea is that, you know, bridging, meaning connecting different
components of a rgraph. might be more frequently done by weak
ties, so even though somebody is somebody you don't interact with much, and you
might have less in common with, they might connect to you to a part of the
world that you don't normally have access to, and so they can still be very
important in accessing information that might not be redundant with people that
you often interact with. And so, weak ties can still play a very
important role. Okay, so what we've talked about is a
series of different ways of, of representing networks and, and
characterizing them. We've talked a little teeny bit about
dynamics and, and strength of ties. the next thing we're going to look at is
now beginning to zoom in on particular nodes and to, to sort of talk about what
their position in the network is. Which is also going to be important in
understanding things like diffusion, contagion and how people behave.
Matthew O. JacksonProfessor
