Social Network Blog - g44

The Freeman and Wasserman articles were very useful in giving detailed theoretical models of popularity, while the Krebs, Moupatta and Valente articles provided unique instances where popularity and prestige play a role in antisocial behaviors. The star model is agreed to have the most central and prestigious placement, yet there is heated debate concerning the measurement prestige and centrality. While degree, closeness and betweenness are all keys to the idea of centrality, what in fact are we measuring when we examine central nodes? Are we looking at actor control, independence or level of activity in a given network? Are we more trying to hint at visibility and importance of actors? Since high levels of indegrees are associated with popularity it is expected that those who are popular are likely very visible, with high levels of activity and much control. Is it sufficient that our ideas of centrality be this broad?

Question:

What in your opinion should be the agreed upon measures of centrality? Should prestige be a part of this definition? Why or Why not?

The Valente article was interesting in that actors who were consistently referenced as popular were not the same actors who were seen as opinion leaders. This is in direct contrast with the Freeman article which connects centrality (or popularity in this case) to leadership. By focusing on how an actor’s social network affects her behavior, I believe we discount the basics of homophily. People who are similar in opinions, beliefs and behaviors tend to be attracted to one another. I believe that homophily is more active here than peer pressure. Those with similar predispositions to smoking and risky behaviors tend to gravitate toward one another. The Valente article also contradicts last weeks reading on smoking where smokers were measured to have more outdegrees than indegrees. In essence the group examined in the Valente article as popular (by definition having many indegrees) in the Pearson article, is not.

Question:

What could account for the differences reported between indegrees and outdegreesof smokers in the two studies?

The Moupatta article’s findings on bullying and victimizing behaviors support social cognitive theory which asserts that behaviors are learned and reinforced within peer groups. Group members are believed to actively live up to the norms associated with their prospective peer groups and this rings true not only for bullying behaviors but for risky sex, smoking and syringe sharing. I found it quite interesting that “model” males and bad boys or rather those supporting prosocial and those engaging in antisocial behaviors similarly have high levels of peer-nominated leadership, centrality, and sociability.

Question:

Why are those supporting prosocial and antisocial behaviors equally regarded? How could this be?

The Krebs article explores how terrorist networks mobilize their efforts. Through snowballing up from one contact, sparse networks are revealed where strong ties appear to be weak due to very low frequency of contact. I was very worried by these assertions since it appears as though the only time investigators would be able to track these groups and map these networks would be at times of high activity, which translates to high risk. Since these networks are isolated, incomplete and very sparse how can we combat them?