|
Noshir Contractor
University of Illinois at Urbana-Champaign
Friday, May 11, 2007
Presenter
Noshir Contractor is a Professor in the Department of Speech Communication, Department of Psychology, and the Coordinated Science Laboratory at the University of Illinois at Urbana-Champaign, where he teaches courses and doctoral seminars on organizational communication processes, communication network analysis, dynamic systems analysis, and collaboration technologies.
Contractor is investigating factors that lead to the formation, maintenance, and dissolution of dynamically linked social and knowledge networks in 21st century organizational forms. Specifically, his research team is developing and testing theories and methods of network science to map, understand, and enable more effective (i) disaster response networks, (ii) public health networks, (iii) transnational immigrant networks, (iv) massively multiplayer online role-playing game (MMORPG) networks and (v) environmental engineering networks. His research program has been funded continuously for the past decade by major grants from the U.S. National Science Foundation with additional funding from the U.S. National Institutes of Health (NIH), U.S. National Aeronautics and Space Administration (NASA), and the Rockefeller Foundation.
Contractor's book titled "Theories of Communication Networks" (co-authored with Professor Peter Monge and published by Oxford University Press) received the 2003 Book of the Year award from the Organizational Communication Division of the National Communication Association. He is the lead developer of IKNOW (Inquiring Knowledge Networks On the Web), a community-ware web-based software and Blanche, a computational modeling environment to simulate the dynamics of social networks.
Contractor is also a Research Affiliate of the Beckman Institute for Advanced Science and Technology, Director of the Science of Networks in Communities (SONIC) Group at the National Center for Supercomputing Applications, and Co-Director of the Age of Networks Initiative at the Center for Advanced Study at the University of Illinois at Urbana-Champaign.
Theme
In this lecture, Professor Contractor discussed his work on developing a meta-theory to describe the drivers behind social networking. He based his meta-theory on several key models used to describe the social drivers behind social networking. He presented a method based on his meta-theory to determine what types of models are prevalent in certain situations. He discussed his sources of data in depth, including the World of Warcraft online gaming community.
Contractor then described the framework he used to model Social Network Dynamics, which starts from theory. He employed two methods to validate his theories: build simulation models and collect real empirical data. He then combined these predictions with the empirical data and used these to test his theories and models. He used a feedback loop to further refine his theories.
Dr. Contractor then presented several key models used to describe social drivers for networking. These models included self interest, social resource and change, mutual interest and collective action, contagion, balance, homophily, proximity, and coevolution. He provided examples of each type of social driver. Based on these theories Professor Contractor and his collaborators constructed various "structural signatures" for each theory. These signatures are similar to and x-ray or MRI and provide a visual representation of social networks. By constructing similar signatures for different networks they can determine which theory or theories are at work at any given time in that network.
These same ideas were then presented in a table listing various characteristics and goals of networks. These characteristics include exploring, exploiting, mobilizing, bonding, and swarming. It is important to note that these characteristics are not mutually exclusive. In fact, many networks exhibit multiple characteristics. Professor Contractor has applied this meta-theory to several different areas including science applications, business applications, societal justice applications, and entertainment applications. Contractor provided examples of studies in each area.
One fascinating example involved studying the network of responders to the threat and aftermath of Hurricane Katrina. Professor Contractor mapped mentions of places, people, and organizations from SitRep documents obtained after the event. He and his team graphed these items over time. From the graph they were able to follow how organizations responded over time and how they influenced the responses of others. For example, data shows that oil producers off the coast of Louisiana had an early response to Hurricane Katrina when the threat of the hurricane was still anticipated in Florida. Over time researchers were able to see the point where the American Red Cross became more important than FEMA in terms of crisis management. Professor Contractor pointed out how valuable it would be to monitor such events in real time, allowing officials to identify key players in a crisis as well as areas where networks might be failing.
Professor Contractor then demonstrated how he is applying this research to entertainment applications, most notably the multiplayer online platform game known as World of Warcraft. The main goals of this game are exploiting, bonding, and swarming. This indicated that there were three theories that might describe actions in WoW: Social Exchange Theory, Balance Theory, and Cognitive Theories. Professor Contractor surveyed participants in the game to determine who they identify as experts in certain fields and how likely they were to contact these experts for advice. From the results, Contractor determined that: "Guild members tend NOT to ask for advice from other guild members over time, guild members tend to reciprocate advice ties with other members over time, and guild members tend to get advice from the person who gives advice to the person they ask from advice over time."
Audience
The audience consisted of approximately 45 people. Audience members came from multiple academic units across campus including Electrical Engineering, Computer Science, Communications, Public Affairs, the College of Urban Planning, the Information School, and Economics. Attendees included undergraduate and graduate students as well as a good representation of faculty.
Questions
Q: Can you diagonalize this network?
A: That is why I use the word meta-theory. That is exactly what I want to do. One of the criticisms of this work is that it is just a combination of existing theories. I want to see if there are new theories that we haven't seen yet.
Q: What data did you use to develop this model? I want to know why it is that the package community of data was the only one that had both collective action and bonding. Does bonding help mobilization if it only overlaps in this case?
A: That is a good question. I don't have an answer but the framework allows us to ask these questions. The model is deductive rather than the result of empirical analysis.
Q: How does the theory deal with granularity?
A: It's not a theory yet but an ensemble of theories. The data for this is self-reported so it is based on whatever granularity the respondents provide.
Q: How did you get users?
A: We only took guilds as a whole who agreed to participate.
Q: Would this restrict the type of data? Larger guilds would have a harder time getting everyone to agree.
A: We had a systematic bias towards smaller guilds. Our goal here was to conduct a pilot study. Now we are getting data from Everquest and Second Life. Interestingly our results in WoW were very similar to what we see in real-world data so the two are very consistent.
Q: I'm excited that this research allows us a chance to develop software to facilitate network building.
A: Yes! I have another presentation entirely on that. It would be great to have something like "If you are interested in this concept you might also be interested in..." and then provide a list of 10 concepts, 10 people in that field, 10 useful documents, etc.
Q: What I really like about this framework is that it gives us a chance to look at structural dynamics as well as personal motivations. You can begin to theorize about why people form networks and look at situations where people don't follow the normal patterns.
A: Yes. You can reflect all the motivations of a particular network. You can predict what will happen if there is a change. One example would be in mobile phone betting. When it became possible people were worried that it would ruin betting at racehorse tracks. Despite these fears nothing changed. Looking at the reasons people bet explains this. People go to a track to bet not just to gamble but also for social reasons. Adding cell phone technology does not change this so people still go to the track.
Implications for National Security
This research provides an interesting framework for studying the motivations behind different social networks. By using the meta-theory developed it is possible to determine what the motivations of networks are, who the major players are, and how to improve weak points in any network. The example of Hurricane Katrina hints at a great possibility for improved government response to complex security threats. In this case, data demonstrated that the American Red Cross was far more influential in crisis recovery than FEMA. By viewing such data in real-time it would be possible to assess this potential failure, determine where potential faults might exist, and improve the exchange of information within and across networks.
Report by Tyan Hynes, UW INSER GA
|
