Diffusion and Deviance
"Diffusion Networks" was an accessible read dealing with opinion leaders, S Curves, and idea diffusion. Rogers' peppers his chapter with examples of how diffusion networks operate through opinion leaders, for both positive (promoting safe sex in gay bars with logo-wearing ambassadors, photovoltaic enthusiasts in DR, Paul Revere, populist sleeper hits) and negative (the new math flop fad) ends. One example I found somewhat specious was his discussion of the modernization of agriculture. He notes the high cost barrier adopting mechanical farm equipment in Dutch farm communities, but cites the large land plots of opinion leaders as being discordant with small landowner's adopting their mechanical methods. Using a situation in which primogeniture and other rural community morays shape social mobility and land acquisition does not really support the overall opinion leader generalizations.
He describes the two-step flow model, which cuts the monolithic and hegemonic ideal of mass media down to size while emphasizing the importance of interpersonal communication in the transfer of information. I agree with this analysis, as many people need an intermediary translator to extract relevant information from the news or current events and frame it for them. Will the advent of new media journalism such as prestigious bloggers (evaluated through web search hits and embedded links) or sites like Digg which allow readers to vote and rank stories (seems similar to interactivity or mutual discourse) ever replace the "face-to-face contact that influences people's political decisions?" (304) Much has been made of Web 3.0 semantic advisors or information gatekeepers, but we have been retreading the "what in network tie structure or human tendencies is impeding these online replacements of opinion leaders?" question.
The author focuses on homophilious communication's ease, frequency, and use in acclerating diffusion in a tightly knit network versus hetereophily's integral role in diffusing innovation across a broad expanse of network communities. In one of his first generalization's, Rogers claims that "interpersonal diffusion networks are mostly homophilous," explaining that individuals of highest status in a system seldom interact with directly with those of the lowest status. This definitely correlates with my job experiences but seems to contradict Milgram's findings in six degrees reachability studies that reaching a high-ranking employee in a company is a good target to trickle down to the lower employees. I know I had hoped for this with my Wistar packet, but does this access to lower rung employees also translate to opinion leadership? Does just being the head boss of a big company give one enough social capital (or using simulated tactics to mimic close ties like a weekly newsletter from the boss with smiling picture) to influence lower employees' decisions on consumption patterns or is there a face-to-face element needed as in the political decisions situation?
Rogers describes the methods of measuring opinion leadership (recall, roster, key informant's identifications, self-designation, and the laborious observation) and their disadvantages. I would have liked for Rogers to expound upon how they identify key informants, especially in our urbanized, non civic participatory, ego-centric networks of today. He presents the differentiation between polymorphism (opinion leading for various topics) and monomorphism (opinion leading for one topic). He then presents unique social characteristics of opinion leaders such as high SES and high levels of social participation. One of the characteristics states that opinion leaders have greater exposure to mass media, but I think Rogers should have specified perhaps that they access non-redundant and heterophilous media, like structural holes. There are neat discussions of change agents facilitation, state opinion leaders, and the effects of various experiments in promoting social change or beneficial health policies. The concept of radial personal networks and low communication proximity seem to correlate with these opinion leaders, who reach a broad sparsely-knit population. Rogers introduces critical mass and network externalities in the diffusion of innovations, and uses examples of fax and Internet adoption. He cites a study by Shermesh and Tellis that claims the average time-to-take off for kitchen and household products is 7.5 years, compared to 2 years for information and entertainment products, an interesting discrepancy which could be due to planned obsolescence or differences in technology investment. This was a general easy to read chapter definitely reminscent of a Gladwell piece. I would have liked to also read more about opinion leader's specific development of their networks, and how one maintains this status with new ties and through life cycle changes. While much of this diffusion of innovation talk is hijacked by annoying marketers, I think the applications such as the Korean birth control villages is interesting and show the power of unconvential approaches to social change through networks.
In Deviance as a Search Process," Lorne Tepperman focuses on individual deviant behavior that pits individuals against communities. Tepperman characterizes the deviant continuum, using a description of search versus contagion that reflects our previous readings on assimilation versus homophily in teenage smokers. The discussion of intention versus deviant acts was interesting, but I would have liked to seen an application of this idea to support resource fruition in another area, with entrepreneurial idea and vc funding or microloan access. I enjoyed reading Tepperman's introduction of the beginnings of search processes through the targeting of WWII enemy sea ships, which puts the topic in perspective but also shows the breadth of applications. Tepperman makes a crucial point about search efficacy depending on the social-cognitive map a searcher has of their deviant object or social behavior. Do you think that the breadth of information available on the Internet could act as a suitable replacement for "authentic" first-hand experience with a subculture in obtaining information on marijuana smokers' location or identities?
I strongly disagree with Tepperman's assertion that the more acculturated the policeman to a deviant subculture, the "more likely his cooptation by criminals." (5) This may be the case of victimless crimes such as marijuana buying, but I highly doubt he has any proof of this for officers tracking people that have sex with children or forge checks from the elderly. He goes on to explain how homophily and our uncertainty of our own networks hinders deviant searches, and the process sort of resembles a twisted version of Milgram's Six Degrees, where we need to access an unidentified tie with a taboo social capital resource through our personal tie networks. Tepperman presents three types of neighborhoods, and his description of a "structurally interchangeable" closed network again resembles Durkheim's mechanical solidarity societies. Tepperman's presents these closed networks as having "little emotional neutrality," a hallmark of weak ties; I had never thought of that perspective before and think it is very useful for analyzing weak ties. Tepperman concludes by describing various methods of deviant search processes, including "closing in," "following a path, through breadth first or depth first," and heuristic searches. I had heard about Lee's pre-Roe v. Wade abortionist study, and was glad it and its findings of a median of 2 contacts was included at the conclusion of this piece.
In 'Social Capital of Opinion Leaders," Burt refines the opinion leader concept by emphasizing their role on the edge of social networks, as brokers of information between groups. I had never noticed that he was a professor at INSEAD, which is sort of Wharton's counterpart in Europe, and strange since he refers to GE's Jack Welch as Jack Welsh. Burt begins his discussion with a parable thats shows a micro-interaction that fueled diffusion of an idea in the business community at large and then touches upon the two-step flow model of diffusion and adoption. He then approaches the network structure of interpersonal contagion and research on the structure of social capital to formulate his argument on opinion brokers. I interpreted his discussion of equivalence and cohesion as slightly echoing our previous readings that characterized strong ties and adoption in heterophilous networks as unlikely without the facilitation of opinion brokers. Burt, like Rogers, uses an example of the adoption of new prescription drugs in a medical community. I would have liked to seen a less institutional example, maybe of the diffusion of religion amongst prison inmates, or a deviant diffusion and adoption of ectasy in suburban high schools. His finding of the role of strong ties between weakly equivalent doctors in decreasing time intervals was interesting, and I think makes sense in light of issues of trust and structural holes (but being strong here). I found his graph illustrating the findings from doctors, officers, and most interesting lobbyists to be a little hard to understand. The reading reviews Burt's concepts of competitive advantage and structural holes, which continually point out the importance of having heterophilous social networks, and brokering strong cohesion across these groups. I know we have talked about the problems with measuring weak ties, but is there a way we could identify opinion brokers by measuring their social networks? I did not really enjoy the conclusion of this article as much as the middle and beginning, as Burt discusses "marketing strategies" and "training sales people as opinion leaders."