Social Network Blog - y11

In this week’s readings, the various authors discussed issues surrounding network size and homophily. In “Birds of a Feather”, the authors discussed the implications surrounding befriending people similar to oneself. A fact observed since the time of Plato and Aristotle, it extends further than just making friends with those your own age and gender, but also provides greater implications when it comes to race and social economic status. In reviewing prior literature, the author finds that “patterns of homophily tend to get stronger as more types of relationships exist between two people”. This is an interesting statement, since it might reveal that the more functions a tie provides (per last week’s reading), the more homophilous they are. This could indicate a causal relationship, where the more similar two people are, they more likely they are to provide multiple services for one another. This could support the finding that kin and close friends were able to provide the most support services. In the discussion on race and ethnicity, one key element in explaining the homophily was the language barrier. One example that struck me was the growing Hispanic population in the US, where the language barrier is a significant divide that stands between extending outside of race-homophilous ties. I also found his conclusions on minorities in work organizations a little misleading. The authors state, “Almost one in four business establishments employ no minorities, while slightly more than one quarter employ fewer than 10%”. Firstly, he does not mention what kind of businesses these are, and whether there are a significant number of minorities who apply, but don’t get accepted. Also, this raises questions about policies like Affirmative Action, which may favor minorities, even if they sometimes aren’t as qualified as another non-minority candidate. Lastly, I also found the mentions of gender very interesting, since it raised the issue of men as bridges and gatekeepers, since he states that “both men and women use men as network routes to accomplish tasks and to connect to information in more distant domains”.

In Pearson’s study on sport-active adolescent substance users, he brought up many interesting issues that have a direct impact on the way to promote substance-free lifestyles. He begins the paper by stating that assimilation is an important factor in the development of adolescent friendships, as teens are adapting to “riskier” behavior to solidify ties. I found the results in this study to be relatively alarming; I find it hard to believe that 12.5% of 13 year-olds smoke cannabis. I also found it interesting that young girls are more likely to smoke cigarettes than young boys. One reason not discussed was the possible beginning of preoccupation with weight, which leads them to smoke in order to control their appetite. All in all, I found the author’s finding very interesting when considering what sort of message to employ in PSAs and substance-free campaigns. Since assimilation is more prevalent with alcohol and cannabis, a campaign should be focused more at building teens up to resist peer pressure.

Before reading “Social Network Size in Humans”, I never considered that one’s network size could be measured through the exchange of Christmas cards. Adding onto this, I found it hard to believe that researchers would test this unusual method with a theory originating from the observation of primates’ social networks. Well, this was going to be one interesting paper. The author remarks that “the exchange of Christmas cards represent the one time of the year when individuals make an effort to contact all those individuals within their social network whose relationships they value”. However, I disagree with this statement; not only does this disregard the celebration of non-Christian holidays, but I’ve noticed a trend that college-aged individuals are not big card-givers, while this is not true for birthdays. Additionally, the proliferation of minimal effort “e-cards” would add some interesting data to this study. I am surprised this fact was not elaborated upon, since the study was published in 2002. Also, I found the results to be a little troubling; a mean network size of 153.5 +/- 84.5 people provides huge shifts in the data and conclusions that can be drawn from it. One interesting finding was that individuals were more likely to contact those overseas than those living further from this within the US. This could possibly be explained by an impression that those living overseas “need” their support and various services from them more, since they are in an unknown environment.

When studying cultural/geographic differences of network sizes in “Estimating the Size of Personal Networks”, I found it hard to replicate their study 16 years later. Since the researchers relied on gathering information from a phonebook, there would be problems with people having their number unlisted, or simply not having a landline due to the growth of mobiles phones and VoIP technology. While reading the paper, I was impressed by the author’s thorough examination of the pitfalls of the study, which included the lack of randomization, similar names, and the size of the phonebook. Even though this is an extensive study, the stark difference in personal network sizes makes it incredibly difficult to determine any sort of “norm” in global network size.

Discussion Questions:
1. Is homophily a concept that we can ever prevent or change? What structural differences need to be implemented in order to prevent homophily on the basis of gender, race, and ethnicity?
2. Can a movement towards “positive peer-pressure” ever be successful? (Note: As the research stated, friends assimilated into their friend’s riskier behavior, not visa-versa.)
3. Do you agree that Christmas cards are a reliable and informative way to examine network size? What other measures would be more revealing?
4. What methods would be more exhaustive than studying personal networks than through the phone book? For college-age students, could online social networking websites (facebook, myspace) offer a better pool of data to study?