In this lecture we'll drop into a little more detail on network structure. Let's start with asking you a question. What do you think the best position is to hold in a network structure, what's your intuition for it? And, you might answer that by thinking about what do you believe the benefits of being in a network are. What benefits are you looking for from your informal networks? So people come up with all kinds of things on this, but the obvious ones are let's be humans. Affiliation, warmth, companionship, entertainment, that kind of thing. Professionally we think about opportunity, resources, information. And it's really that last one that we're going to focus on. We're going to move toward some models here and a necessary part of working with a model is that you simplify the world. And one of the simplifications that people make in this research area is we think about the benefits, first and foremost, as being information. What about costs? What do you consider the cost of being in a network? And this is a little more ambiguous but one that most people seem to get and resonates, especially with our over subscribed students are at the time involved with maintaining your network. Most of us have difficulty spending as much time with their friends and professional contacts that we have. And it's pretty obvious to us that there's a scarce resource there. It's our time. And so, we can come up with other costs, but for this exercise, to understand the basic ideas in this field, let's think about the benefits being information and the costs being time. So, you might think, all right given that setup, what does it mean about where we want to be? Do we want to have a strong ally for example? That's one way of thinking about it. I want to be wherever in the network the strongest person is. It gives you an advocate in hiring and promotion, provides training, maybe a little mentorship, maybe a little support. It's a reasonable thing to say. Or you might say, well I don't really care where I am. I just want to be in a cohesive network. I want to work with people who know each other, they understand each other. Maybe they all coordinate very easily. There are lots of benefits from that. And so that's understandable as well. Others say, I want a large network. As many ties as possible. This is probably the most common answer, the most intuitive answer. If we're going to talk about networks, we're impressed with people who can maintain big networks. Others say well you know, size is one thing but I'll take centrality. I want to be at the center of things, I'd have to be at the heart. Everyone has to come through me. That's reasonable as well. There are benefits, there are some cost to all of these things. I think these are all relatively intuitive. And because of that I really want to emphasize another one, a different one and that is this notion of a structural hole. And structural holes we'll unpack it in more detail momentarily. But these are basically centrally placed people with ties to those who are not cohesive. So the idea is that. In networks everyone is not connected to everybody else. And if you can sit between disconnected groups you're spanning the structural hole. And maybe there are some real value in doing that. So let's look at a picture and make this more plain. Here is, let's say, a network of six people. And consider especially the person in the red dot there. This is what is considered a dense network. They are all connected to each other. This is what might considered a disconnected network. A larger number of people. But the big feature here that differs across the two is that the one on the left is dense and the other on the right is disconnected. What happens when you remove the red dot? The person in the position of the red dot. You remove that person from the dense structure and everybody still talks to everybody. Information flows the same way, not a whole lot changes. On the other hand in the disconnected network if you remove the person in the position of the red dot all of the sudden these groups don't talk to each other at all. The organization's in a very different place, information doesn't flow, and something's going to function very differently. So clearly a person who is in the role of the red dot will play a very different role in the organization if they're in that dense structure than when they're in the disconnected structure. That gap, the gap that the red dot is spanning there in the disconnected structure is what we consider a structural hole or a disconnect. And we really want to talk about the benefits of that. It's going to drive pretty much everything else that follows here. So a dense structure, relatively few structural holes. People are highly connected. They have mutual influence on each other. This is often a cohesive, similar group. They work closely together, might think alike. They probably have what the economist would call or the sociologist would call redundant information. They're often tapping a restricted set of resources. Contrast that with the disconnected structure, where there are many structural holes. This more often reflects a non-cohesive, heterogeneous group. They provide opportunities in terms of differentiated, non-redundant Information, and they access a wider array of these resources. This idea comes from research originally by Mark Granovetter on weak-ties and then really built out by the sociologist Ron Burt. And we want to dig in to it in a little more detail. So here's another picture to talk about how you might think as your grow a network From 4, to 8, to 16 individuals. What it would look like if you did that in a very disconnected way versus a very dense way. We're going to use the terms redundant versus efficient. So these terms, again, there's very much an economics base to the work on social networks. He comes out of sociology, but this part of sociology is highly connected to economics, these guys think about utility maximization. The actors here are highly rational and they're going to, very much like economics, talk about cost and benefits. And they're just going to find the optimal solution given the cost and benefits. So they're going to use terms like redundant and efficiency. These are modeling terms but let's explain to you what they mean. The top there, it's Network A, B, and C. As you move, for example, from Network A where you have 4 connections and Network B, you have 8 connections, and Network C you have 16 connections. This case over time you add someone to the network your adding a new relationship. So you're going from 4 to 8 to 16 relationships. Great you've got more people, but we know there are costs associated with that. It is what the economists and sociologists call redundant ties. Contrast that with three on the bottom: A prime, B prime, and C prime. This person goes from 4 to 8 to 16, always with four relationships. In each case, they have perfectly efficient networks. They're spanning these structural holes, this is a disconnected network. And with the same four relationships they're able to tap into 16 people now because of the way they built it. So the sociologists would consider the bottom networks efficient and the top networks redundant. Because at much less cost, they're tapping into the same information. So the assumption here is that the resources needed to maintain relationships, time and emotion, are scarce. I think that's a very reasonable assumption. But given that assumption, these guys going to take it to its extreme and ask well if you guys got scarce resources maybe you want to think about building a more efficient network. So with these ideas out there, let's start talking about now how networks might differ if you took them seriously. Consider diversifying your network. In this picture we begin with what is a very dense network A person has five relationships tapping into something like 10 or 12 individuals most of whom are connected to other people in the same network. And then after diversification, after taking this theory really seriously you've got the same five relationships but now it taps into much broader, maybe 20. 20 plus individuals many of whom do not connect with other people and so you're tapping into more highly varied resources. This is what the sociologists would claim are benefits of a more disconnected structure. They would advocate even pursuing a more efficient network and assuing redundancy that you see in the denser networks. So there many benefits of course the people in the disconnected networks will be the first to know information outside of each pocket. They can arrange partnerships so they want to they can play rival factions against each other. And they might be sought out as a broker to introduce, make contacts, make introductions from one part of their organization to another. There are many benefits that fall right out of it. We don't want to ignore the costs. What do you think the costs might be of this? And we'll spend a little more time on that but I think it's fair to say that while extreme, these guys are mostly just making plain something that many of us do intuitively. And in fact, I would push you to reflect a little bit on how is it that you're deciding where to spend your time? Who to invest in? What relationships to cultivate, what relationships to let ride. It is something that many people want to avoid is a little bit uncomfortable to think about. But the economic model and the sociological model takes us a starting place that we've got scarce resources. We drive utility from things like relationships and were essentially maximizers were going to try to maximize how much utility you can get from our scarce resources. You bring that model to relationships and this is the kind of thing that we end up with. I'm not going to say you should all think about it this way, that you should act exactly like this, that you should make as efficient a network as possible. But I like the conversation and I like having it on the table, not least because I think we're all doing something like this. Anyway, and lets be a little bit more explicit about it. Let's talk about the tradeoffs and then lets see how we can exercise these ideas. How we can use these tools in a way that isn't offensive to us. We don't want to act in a way that isn't in the spirit of how we want to live. We don't want to act in a way because the sociologists tell us this is more efficient just for that reason. But can we borrow those ideas, can we borrow those tools and do it in a way that actually fits with the way we want to live? That's where we're coming from. So a little more motivation for considering it. There is work here by Ron Burton, this was empirical work. So far we've just talked theory, right? Burt went out and ran a big study in a large organization that looked at how employees rose through their ranks there. What rate of promotion they got over time, and he compared that to the networks they had when they began. What he found was strong relationship, a social science skills this is a very strong relationship between those who had many disconnects. So when a relatively disconnected structure, compared to his who had few disconnects and a more dense and formal structure. Those with more disconnects showed a higher rate of promotion in the organization over time. So this was kind of the beginning of what has turned into a huge industry in academic work. On what are the empirical consequences of these various network structures and a little bit of motivation for maybe that's not just theory. Maybe there are things that make a difference. This idea that by tapping into a more diverse network, we get more information and opportunities that we wouldn't if we tapped into a more homogenous, more dense more dense network. So, are there limitations? Of course there are limitations, we've named a few. But, we always want to mind these limitations and in other parts of the course we unpack them more explicitly. But, if you think about, I'm going to maximize the efficiency of my organization, my network. What's the risk here? If you're sitting there on the right-hand side with 4 relationships into 16 people, what's the risk? Obviously one of the biggest risk is what happens if that one contact, that you have in a cluster of 4, goes away? What happens if that person moves or something goes wrong in that relationship? Then you're only bridge into that group is gone. So this is fine with a sociologist. They'll just say look we can still deal with that. If you have an especially important cluster of people then you need to have more than one relationship in there. You need some redundancy, just in case something goes wrong. They can drop that right into their model, run the utility maximizing thing again and it's going to give that kind of answer. The more important the group, the more important that you have some redundant ties into it. Let me give you one small example from politics in the United States that ends with a quote that I love that captures this pretty well. So Jim Wright was a long-time Congressman from the Dallas-Fort Worth area in Texas. He was in Congress for 34 years and finally, near the end of his career there, he took over as Speaker of the House. He succeeded Tip O'Neill, who was a long time and famous Speaker of the House. So Jim Wright goes in and he only lasts two years. This poor guy was in Congress for 34 years and he gets indicted as Speaker of the House and has to resign. Why did that happen? This was the beginning of Newt Gingrich, Gingrich took it on himself to run Wright out, basically. He wanted more power, he wanted to undermine the democratic party. And so he found ways to run him out, Gingrich, there's a lot of detail on this, and that doesn't matter as much for the moment. But what Gingrich said about Wright is very relevant to us today. Gingrich says that when he began initiating formal investigations into Wright, this is in late 1987, he said, I've watched my colleagues' body language toward me. It's actually improved. If I had gone after Tip O'Neill this hard, I would have been a social outcast. Wright should get a solid A for performance. As a technician of power, he's done a great job. The downside is he's a loner still. There's no deep body of affection for him. Being a loner eliminates a safety net of both information and good will. That line says a lot. This is why we tell the story. This line from Gingrich who, whatever you feel about Gingrich he's got some inside into power and politics. He says it about right, being a loner eliminates the safety net of both information and good will. This is something that applies to all of us. This is not about speakers of the house. And he puts in just this perfect way; information and good will. It's both of these things. Of course even that simplifies things too much, but that's a nice two bucket summary of what we're looking for in informal networks. Okay, a few caveats before we leave it. Burt's approach characterized very much by cost benefit calculations, a lot of intentionality and strategy here. There are trade offs. This is inherent to the traditional socialized group model of networks. We've presented very quickly kind of extreme version of this. We want to be provocative. We want to push you and we also want to stay kind of simple. We want to give you a simple framework for thinking about this and we want you to reflect on how much of you are doing a lot of this tacitly already. This is more or less just a theoretical benchmark. It's going to push us to think about the opportunity costs of the way we build our relationships and the way we move through organizations.
