Wednesday, October 14, 2009

Book Review draft 2

What do syphilis in Baltimore, Sesame Street, Hush Puppies shoes, and crime reduction in New York City all have in common? They are all epidemics. They started with a small group of people, reached some ethereal level and exploded across a community, be it local, national, or worldwide. They are also prominently featured in Malcolm Gladwell’s book, The Tipping Point. In The Tipping Point, Gladwell examines epidemics to explore the reasons why these disparate things have been able to affect so many people. He seems personally most interested in those epidemics that can make a positive difference in people's lives, such as the learning epidemics caused by Sesame Street and Blues Clues. Regardless, he traces the elements involved in the spread of many different epidemics–be they fashion statements, non-crime waves or messages such as Paul Revere's famous cry of impending battle.

Central to the book is the idea of a tipping point: An idea, trend, or germ--a contagion--may exist at some basal level within a population, infecting roughly the same number of people that overcome it in any given time period. Some stimulus can cause an imbalance in this equilibrium, though, and the contagion will start to reach more new people than it leaves behind. This is the tipping point, and an epidemic follows. Without that push to disrupt the equilibrium, it may never reach that critical mass, that threshold it needs to cross in order to infect the whole population. Gladwell's analysis of epidemics centers around this idea: How can we identify strategies to "tip" phenomena into epidemics?

Perhaps because of the discussion of this equilibrium, I began to formulate a mathematical model to track the spread of an epidemic to go along with Gladwell's commentary. The model works roughly thus: Each member of the population in question would have a state, either "infected" or "not infected." They would also contact "x" number of people in a given time step "t", and, if infected, have a "y"% chance of "infecting" each of the people they came into contact with. Each individual would remain infected for "z" time steps. Each of these variables would be different for each person in the population (except for the time step, which would necessarily be the same for all), and, when all members of the population were considered, the total spread of the contagion could be tracked. You could predict the percent of a population "infected" with some contagion after a time period by setting an initial condition (maybe only one individual is infected to start with) and extrapolating to the desired number of time steps.

Gladwell defines the transmission of epidemics by three laws. The first of these, "The Law of the Few," explores three special types of people that can help an idea become an epidemic: his “Connectors,” “Salespeople” and "Mavens." These are the people who can tip that contagion over the edge into boundless effect. In my model, these are people who have exceptional coefficients "x", "y", and "z". The first of these, the Connectors, spread contagions exceptionally well simply by the sheer magnitude of people they know. Their "x" is through the roof. If I am studying the spread of the newest pop hit, “Bulletproof” by La Roux, an average person would pass the song along to maybe two people per day. The Connectors would get ten people to listen to it: they know more people, they are in contact with them more, and their contact is meaningful. These people have huge social networks.

What if you don’t like pop? Especially dance pop from the UK, like "Bulletproof"? Gladwell’s Salespeople will convince you that you do. Their "y" is near 100%: they have the ability to transmit ideas to everyone they meet. Gladwell’s Mavens are an interesting case. These people are experts in a narrow field. As a result, they hold a lot of sway when giving advice about that field, and people they "infect" are likely to stick with whatever the Maven is spreading. Given the right contagion--something in their field--they have a high "y," and the "z" of those they infect is likely to rise. Mavens also practice selection among contagions--they will only select the best. The case of the Maven may not apply as well to the spread of a pop hit. Maybe a music reviewer would be a maven, and a favorable review would encourage others to listen.

In this way, these three types of people infect more than their fair share; they diffuse ideas farther than normal people. Whether from sheer numbers as with the Connectors, from careful selection as with the Mavens, or with efficacy of transmission as with the Salespeople, these special individuals have the ability to tip things into epidemics.

Just by looking at my simple model, I disagree that these three types of people would be the only ones who would have the power to transform a contagion into an epidemic–There are other ways to manipulate the "x," "y" and "z" values to produce extraordinary individuals. What about someone with a high "z" who harbors a contagion for a long time–a 'Collector?' These people would be able to upset the equilibrium by transmitting the contagion for a long time–they will be singing “Bulletproof” for weeks rather than just the day or two after they hear it. I also wonder if these Connectors, Salespeople, and Mavens are truly different from the average population, or if there is a broad spectrum of ability. In any case, it is useless to define these people if you can't find them to help tip your epidemic.

The Internet has also caused Gladwell’s heroes, especially the Connectors, to lose some of their power. Social networking sites allow all of us to be Connectors. I can message hundreds of people with the click of a mouse on Facebook, and link them to “Bulletproof” on Youtube. This glut of information devalues the communications of the Connectors. Mavens lose their uniqueness when faced with thousands of their peers reviewing the same items on Amazon. Salespeople are perhaps immune, as good salesmanship requires face-to-face contact.

Additionally, it is short-sighted to say that certain types of people are necessary for an epidemic to transpire. Some exceptional contagions can spread on their own, like H1N1 and results of the presidential race in the USA.

The Few are not even necessary to tip epidemics. The media can take their place. Gladwell even acknowledges that the media can be an alternative vehicle to the “Few” for reaching many people at once. Indeed, as he addresses his second criteria, “The Stickiness Factor,” he talks about children's television as a vehicle for learning epidemics, a case in which messages are disseminated entirely by the media, not by the Few. Gladwell here focuses not on the population through which the epidemic spreads, or the Few who tip it; he looks at the contents of the contagion. The Few can’t do anything with an impotent virus or a worthless idea. It must be "sticky" enough to be absorbed before it can be passed on.

Gladwell highlights Sesame Street and Blue’s Clues as examples of intentionally increasing the 'stickiness' of a message, in this case increasing the stickiness of learning for children. The creators of Sesame Street studied children and what was sticky with them. They then tailored their show to meet those criteria. The sticky factor came not just from the information, but also the way it was presented, catering to children’s short attention spans. Blue’s Clues took the information and success of Sesame Street to another level by further developing the stickiness of the information, repeating it five times every week. This is a very interesting concept, that presentation can change the stickiness of a message without altering the content.

This stickiness factor is not taken into account in my simple model. It is assumed that the contagion will be sticky, that it can be transmitted, and that it will spread. This is the case for most of the things we typically consider contagions, i.e. diseases. With the example of education, Gladwell shows how non-inherently transmissible things can also become contagions. This has huge implications: We can engineer virtually anything to become an epidemic if we can figure out how to make it sticky. We can spread information, ideas and positive change just like H1N1 has spread across the world, if we only figure out how to make our information stickier.

My model is much too simplistic to describe a real-world epidemic, but so is Gladwell's commentary up to this point. He had been hinting at it, like when he described some indefinable quality that some Salespeople posses, but he truly explores the issue in the third section of the book, "The Power of Context." Here he states the very obvious point that circumstances matter. People and ideas do not exist in vacuums. There is more than just stickiness and personality type at play. We as humans have complex subconscious psychological processes that pick up on subtle environmental cues we may not even be aware of. If the song “Bulletproof” is used as a theoretical example in a class project, the context informs the reader that it is not meant as a serious suggestion. The would-be epidemic tipper needs to consider the context that the message will be transmitted in.

Within a population, up until this point, only the transmitters of epidemics (The Connectors, the Mavens, and the Salespeople) have been considered, but the recipient of the transmission is just as important. Communication is a complicated subject that demands a lot from both the sender and the receiver to be effective. The great managerial writer, Drucker, addressed this in a 1974 essay called "Functioning Communications." He delineates there the effort required by the communicator to reach the listener due to differences in mental processes. Communications can fall upon deaf ears when the communicator fails to take into account the listener's mindset. People hear what they want to hear. People's brains are hard-wired to process stimuli in certain fashions. Drucker writes how communication is almost a battle: As the communicator, you have to bring the listener around to your way of thinking. You have to bypass the natural route in the listener's mind and develop new pathways. Otherwise, they will hear what they want to hear, and not what you want them to hear.

Effective communication from the communicator's standpoint demands messages that speak to the audience. Effective communication from the listener's standpoint demands a receptive attitude and a commitment to the consideration of new ideas. This is a hard thing to accomplish when so much of the information processing that goes on in our brains occurs below our consciousness. It's almost like holding you breath: you can only do it for so long before your natural processes take over. For Gladwell, this is addressed as Context. Some messages can be groomed for mass audiences (like Sesame Street), but there will still be barriers to transmission to everybody.

Gladwell devotes considerable space toward discussing nuanced studies of various epidemics, and how his three laws played into them. One example I found particularly interesting told of suicide in the nation of Micronesia: Unheard of before a teenager took his life in the 1960s over a minor disagreement with his parents, a suicide epidemic soon swept the nation. The power of context played a powerful role here: that first suicide gave permission for the rest. Stickiness was important too: suicide, unfortunately, is absolutely sticky, and there is no way to make it less sticky. Try it once, and you are dead. One thing this example lacked, though, was any tie-in to the Few. No Connectors advertised suicide to their many acquaintances, no Mavens told the best methods, and no Salesmen advocated their friends try it. As far as I can tell, the first suicide was reported in the newspapers; the media again was the culprit for tipping this epidemic.

This is indicative of the book as a whole: Gladwell presents a wealth of provocative ideas, a wealth of interesting examples, and a wealth of succinct conclusions. His writing is expertly woven and watertight. His book is chock full of marvelous anecdotes; however, nowhere does he give an example that effectively demonstrates the importance of all of his three laws in causing epidemics to tip. This does not mean his ideas are worthless; rather, he has presented tools that can be used to create change in society, in our thinking patterns, and in the way we spread our messages. If we want to spread things that really matter, such as anti-crime waves and learning, we need to identify ways that will help these things, which are not inherently transmitted, to spread to epidemic proportions.

6 comments:

  1. Here' a comment about your model. I'll post something else about the rest of the review in a while, perhaps tomorrow morning.

    First, it needs to be generalized. Not a model of one specific contagion, but a model of lots of similar contagions, some which become epidemic, others which don't.

    I wear SAS shoes. Those haven't become an epidemic. Hush Puppies have. What explains both of those facts, not just the one? The roles Gladwell has, maven in particular, cuts across all the candidates. A film critic like Roger Ebert watches lots and lots of movies. It doesn't matter whether you watch one that you absolutely love every time you watch it. That will have essentially no impact on anyone else's viewing. Ebert's judgment of a film matter.

    So in your discussion of this potential other category, a collector, is that supposed to be specific to a particular contagion or is it general across many like contagions? The specific kind won't cut it.

    If it is the general kind, to anybody observing the situation from afar, will the person look different than the other three connectors? If there is no observational distinction, what is the theoretical benefit?

    Occam's Razor says (my spin of it, anyway) add complexity to the model only if the simple one fails to explain something important. Otherwise, simple is better.

    So I'm not getting why the theoretical possibility of another category is a point of interest. It would be if some epidemics can't be explained by the categories Gladwell does come up with.

    ReplyDelete
  2. I thought it might be useful for you to see my criticism of the Tipping Point. The criticism is that he offers up what appears to be a post hoc explanation for why certain ideas diffuse and treats that as if he is giving a theory of causality. My belief is that there are many chance factors (meaning we don't know enough) that explain which ideas diffuse. There is not much prediction power in what Gladwell writes.

    If there were the ability to predict, movie studios wouldn't spend millions of dollars on making films that turn out to be bombs. Indeed, we'd have ability to discriminate between duds and epidemics in advance. Mostly, I don't think we can do that. At best, up front efforts can improve the odds.

    Gladwell's book doesn't talk much about interesting ideas that don't go anywhere. As a result, the reader has no sense of whether there are an ocean of such ideas with the one that tips the proverbial needle in the haystack or if the relative populations are more comparable. That question is not considered.

    I don't think that is a flaw of the book if we look at the book for what it is. It's only a flaw if we inflate what we believe to be the lesson from the book.

    ReplyDelete
  3. Time is short, but I wonder if you'd take one more pass at this focusing on this paragraph you wrote:

    "The Internet has also caused Gladwell’s heroes, especially the Connectors, to lose some of their power. Social networking sites allow all of us to be Connectors. I can message hundreds of people with the click of a mouse on Facebook, and link them to “Bulletproof” on Youtube. This glut of information devalues the communications of the Connectors. Mavens lose their uniqueness when faced with thousands of their peers reviewing the same items on Amazon. Salespeople are perhaps immune, as good salesmanship requires face-to-face contact."

    There is potentially a full essay in the issues you touch on in this paragraph. Let me illustrate some of these that you might explore:

    1. Lags. The Internet clearly has sped things up. What's the consequence of that?
    2. Trusting information. You mentioned reviews in amazon.com. Do those reviews sway a reader of them? Why or why not?
    3. Over abundance of information. There is way too much. What should we pay attention to? How does that problem get solved?
    4. Do the few entirely vanish (you suggest that only the sales people survive) or do their roles get merged? In the limited domain known as edu blogging, where I have some direct knowledge of what is going on, there are a few uber bloggers who seem to play the role as both maven and connector. I know that when one of them references a particular post on my blog, I get a lot of hits on that post. I conjecture that generalizes. It would be good for you to talk about that in areas (music perhaps) where you have some direct knowledge. Elsewhere in the piece you talk about the media as connectors as if that means Gladwell is wrong. Why can't the media be connectors? Some elaboration on that would help.
    5. There is a potential for yin-yang with the Drucker part on communication that you brought in (the listener is an individual who can block communication if he so chooses) and the alternative possibility of group-think with an idea that has tipped (Obama during the campaign might be an example). That could be explored at some length.

    Taking on any one of these points or a couple of them would make the piece richer.

    Finally, let me make one larger observation. Your piece reads now like it is making the following argument: Gladwell reads like science fiction, but is being passed off as if it is actual science. So much of the piece is in showing ways where it is not science. If you looked at Gladwell instead as if is history - looking back at ideas that did tip and seeing if there are some common elements to those - then maybe you could spend more time on those common elements in your review. I think that would make your piece a better read.

    ReplyDelete
  4. 1) Thanks for pointing out the confusion that the model (and my current discussion of it) presents. I realize that I am again failing as a communicator by making assumptions that are not explicitly stated. I think I will be able to address most of these issues in the next draft.

    2)Reading your criticism definitely helped me see your point of view. One reason my piece reads like I am comparing Gladwell to sci-fi (I would say psuedo-science) is that I am. I think that he presents his information in a way that conveys the predictive powers of science rather than the discriptive powers of history, and I don't think his work has a sound scientific basis. I tend to overlook the value of histroy (always one of my least favorite subjects in school), so I will have to try to consider this point of view.

    3)I agree that there are a lot of things to write about in that paragraph, and I will try to expand on some of them. I think the issue of the media (who I don't think are connectors because they don't have a personal relationship [I know, another arguable point] with their listeners) and the glut of information highlight one of the big problems with Gladwell's book: he doesn't consider the individual listener's receptiveness to new things. [sorry this point is a little unclear, but I think it will be better in the next draft]

    Thanks again for all your comments.

    ReplyDelete
  5. Ok - I look forward to the next version. Here's one beyond the course request and one nit to pick for this paper.

    At your leisure, perhaps after you've had a job for a while and done other things, do give history another look. Perhaps read Barbara Tuchman. Do it as recreational reading. I think you'd react to it quite differently than has been your experience with history in courses and you might grow to like it a lot.

    The nitpicking is on the use of the word trend. It is an after the fact notion. You can't see a trend prospectively, because it doesn't exist yet, though you can hypothesize it might.

    ReplyDelete
  6. Hey Dave,

    Maybe it is just me not having as keen of an eye as Professor Arvan, but I can not see weaving of much in this piece. Just to clarify, is the main thing that you are weaving it with your mathematical theory idea? Then that idea kind of fades away a little bit, so I am confused.

    I thought it would help to get the perspective of a "normal" reader... ;)

    See you soon.

    -Al

    ReplyDelete