Social Network Blog - r10

I was a bit surprised that Granovetter’s famous article seemed more like a relevant literature review and opinioned conclusion than any type of monumental survey or experiment. The “strength of weak ties” has been an integral part of courses I have taken in departments such as Communication, Marketing, and Psychology, and I was surprised that this primary article did not appear to be as earth-shattering as the actual idea of the strength of weak ties.

One thing I noticed is that Granovetter defined the “strength” of a tie as “a (probably linear) combination of the amount of time, the emotional intensity, the intimacy (mutual confiding), and the reciprocal services which characterize the tie” (p.1361). However, Granovetter measured tie strength by asking respondents “how often they saw the contact around the time that he passed on job information to them” (p.1371). This measure of strength refers to amount of time two individuals spend together but not any of the other markers of relationship strength mentioned above.

What about contacts that you feel strong intensity towards (i.e. family, old college friends who live in other states) but do not see more than once a year? Granted, how often two people see each other is the most quantitative measurement, but I do believe that he could have defined a quantitative standard to measure the other variables of tie strength. Thus, I ask: Do you think that if Granovetter had measured tie strength based on some other quantitative measure/standard, he would have come to the same conclusion about the strength of weak ties?

A picture of 2 strong networks, linked by a weak tie, as graphically imagined by me:

I disagree with elements of Burt’s structural-hole argument. Burt believes that people (who he calls “players”) who have well-structured networks will obtain a higher rate of return on their social capital (as measured by the amount of opportunities and information to which they are introduced). Burt defines this well-structured network as consisting of an individual who is connected to many clusters of people (who are not connected to each other).

In my opinion, this reasoning relies on the assumption that people only live in cliques/clusters throughout which finite information circulates. This presumes a type of circular representation of social relationships. For example, one may divide a group of sports fans into baseball fans, basketball fans, and football fans. These three groups can be presumed to be mutually exclusive. However, the structural hole argument would go further, saying that a person getting information from a friend who is a baseball fan would receive no additional information on this subject from another friend who is a baseball fan.

This idea ignores the fact that the baseball fan does not identify himself solely by one characteristic but rather that he may, as Gladwell demonstrated, belong to many different circles. Thus, he would have a variety of information to offer; and a different baseball fan would have different information to offer. People belong to many different groups, and messages can be passed linearly. If this is true, then bodies of knowledge do not get actually trapped in clusters, circulating among the same people and not being shared with additional people, as Burt seems to indicate.

Thus, it would be untrue to say that a person increases his or her rate of return on social capital by maintaining relationships with a variety of different clusters. Indeed, it is almost impossible for this to be true, for Burt explains that social capital is based on networks of relationships (and not the individual); but yet his theory presumes that only the individual can maintain relationships with various clusters. He fails to understand that the clusters can have their own clusters. People don’t just have one role.

I ask: How would Burt defend his theory against the following scenario? Bob, Martha, Samuel, and Marie are all friends with each other and avid Democrats. There is redundancy in their network of contacts. However, Bob is a businessman, Martha is a lawyer, Samuel is a nurse, and Marie is an engineer. Does that mean that the people in the cluster of Bob, Martha, Samuel, and Marie all know what each other knows (and only what each other knows)? Or do people wear multiple “hats”? What would Gladwell say?

The most notable aspect of McPherson’s article was its consistent emphasis on being thorough. Whether it was covering all aspects of the research methodology, pointing out possible extraneous variables in the questionnaire, or indicating potential causes of the results, the authors of this article appeared determined to fully explain both the benefits and limitations of their research.

Although the decrease from 1985 to 2004 in the average number of people with whom respondents discuss important matters seems apparent, further analysis makes it seem a bit confusing. Granted, the data indicate that people are discussing important matters more with their families. However, the idea of people’s networks as “smaller, more tightly interconnected, more focused on the very strong bonds of the nuclear family” (p.371) seem contrary to a priori assumptions. People seem to talk more to co-workers and friends and spend less quality time with their families. So the idea that one’s close network is dominated by family members seems a bit obscure, when viewed this way.

Perhaps the resolution to this apparent conflict is that although people are spending much of their days talking about work and other non-family issues, they choose to discuss important matters with the people who are in their families. Thus, it is not a question of the amount of time spent with either group but rather solely a measure of the quality of what is being discussed that has caused this growth in discussions with kin. This view conforms to general lay knowledge about the structure of today’s society.

I ask: Do you think people are talking to their family members more (in terms of time) than before, or do you think that they are just choosing to discuss certain issues more exclusively/extensively with family members?