Who's smarted whose?The Wisdom of Crowds by James Surowiecki © 2004

Host: Steve M.

Date: November 1, 2007

Judges’ Score: TBA

Food & Drink: Ethiopian Food + Home-made Anna cake

Other Sources:

Host’s Review:

Overall, some tough punches thrown — was the book’s theory arbitrary? Did Surowiecki take a theory he noticed in anecdotes and force his data to conform? Sure crowds are wise if you apply such stringent rules: a crowd will ONLY be smart if this AND this AND this AND this. That is, a smart crowd must have diversity of opinion, independence of individuals, decentralization (not controlled by one person or group), and aggregation of opinions. Who fits the bill?

However, there seemed to be a general consensus that crowds could provide useful information…sometimes. Corporations are now polling employees to predict future performance, and even the belligerent Broken Spines group cobbled their collective wisdom to successfully answer impromptu history questions and guesstimations (on average).

Maybe the true value of a crowd lies in creating the right question. You might not ask a random sample to build a bridge, but they might be equipped to define the sturdiest overall design. And though innovation may still originate with the individual, tools become truly useful once put to the crowd for aggregation: even if Google’s algorithm was created by one person, its worth comes from the crowd’s aggregated opinion. Similarly, the Iowa Electronic Markets.

The communal Ethiopian food was amicably received, Anna’s cake creation arrived amidst whoops and hollers, and the BS crowd tossed back another book in the hope that a rising class of informed citizenry lifts all boats.

Let them eat cake!

deliciaaaas.jpg

10 Responses to “#10 – Wisdom of Crowds”


  1. 1 richh November 2, 2007 at 8:11 am

    Nice bout, very interesting theory, but something didn’t smell right as an all encompassing solution for decision making (or maybe it just adverse effects of Ethiopian food.)

  2. 2 tbone24oz November 2, 2007 at 8:11 am

    I little out of sinc. but here are my thoughts from before the meeting:

    The book is basically trying to prove a single argument (crowds make better decisions), but the method being used to prove this argument is flawed.
    And this is how it goes:
    Assumption: there are 3 ways you can make a decision 1) by an individual, 2) by a small group, 3) by a crowed.
    Argument: individuals suck at making decision and there is no problem finding plenty of examples for that.
    Argument: small groups also suck at making decisions and we have plenty of examples for that as well.
    Conclusion: Since “large crowds” is the only method left, the book deducts that it’s the best for making decisions. And, we got some anecdotal examples for this.
    Oh, wait a minute, there are some cases where also large crowds make bad decisions (i.e. most democracies), so let’s make rules so the conclusion will stand.
    Rule: cases in which large crowds are bad for making decisions are excluded from the discussion.

    Summary: except for the cases in which it doesn’t work well, large crowds are the best for making decisions. Simply, because the two other methods (i.e. individuals and small groups) are not effective.

    An equivalent argument: babies are brought to the world by Cranes.
    I can easily prove babies don’t grow on trees. I can also prove babies are not delivered by UPS.
    I have some anecdotal examples that babies were delivered by their mother (after 9 months of pregnancy), so let’s exclude these cases from the discussion.
    Conclusion: babies are delivered by Cranes. Simply because there is no other way they can come into the world.

  3. 3 bobbyjones November 2, 2007 at 9:01 am

    How come I’m the only one smiling in the photo? Because that Anna Cake was so good. I thought Dave set the bar with home made ice cream – but the Anna Cake was a work of art.

    And, like the Anna Cake, last night’s bout had all the right ingredients: well written book, enough conflict to stir a good debate, and spicy Ethiopian food. Well done Steve.

    In the spirit of keeping our conversations on Broken Spines rather than email, here is my opening volley back in September:

    On Stupid people: http://www.youtube.com/watch?v=Bb8kJV15Kns

    OK Steve M – you drew first blood. You want to take the gloves off and throw down some applied behavioral economics? Well, I’m ready to weigh in and get in the ring.

    I’ll save you the trouble of responding to my post, here’s my retort:

    Steve M would have us believe that Web 2.0 is the manifestation of New Age collective consciousness…what Tim O’Reilly calls the successor to the human potential movement (Esalen in the 70’s).

    Well, call me a skeptic, but I don’t buy the hype. I don’t think the crowd has ever proved itself wise – and certainly not with Web 2.0. Steve M’s selection for the next bout supports this view by setting a high bar for “wise crowds” by requiring four attributes:
    (1) Diversity of opinion
    (2) Independence of members from one another
    (3) Decentralization
    (4) A good method for aggregating opinions

    By Surowiecki’s criteria, wise crowds are rare indeed. I’m looking forward to reading the book to see an example of a wise crowd (turns out they can count jelly beans and the weight of a cow really well).

    Unfortunately, you don’t need to look very far for proof of mass ignorance. Let’s take an easy example: presidential elections and the issue of evolution. Here are the results from a USA Today/Gallup Poll:

    Evolution, that is, the idea that human beings developed over millions of years from less advanced forms of life.
    Definitely true: 18%
    Probably true: 35
    Probably false: 16
    Definitely false: 28
    No opinion: 3

    In other words, a slim majority (53%) of Americans believe in evolution as much as the probably believe in anything else.
    Creationism, that is, the idea that God created human beings pretty much in their present form at one time within the last 10,000 years.
    Definitely true: 39%
    Probably true: 27%
    Probably false: 16%
    Definitely false: 15%
    No opinion: 3%

    In other words, a majority (66%) of Americans believe that god created human beings as they exist today within the last 10,000 years. Interesting.

    If a presidential candidate stated that he or she DID not believe in the theory of evolution, would that make you –much more likely to vote for that candidate, a little more likely, not make a difference either way, would it make you a little less likely, (or) much less likely to vote for that candidate?

    Registered Voters:
    Much more likely 8%
    A little more likely 7%
    Not make a difference 53%
    A little less likely 14%
    Much less likely 15%
    No opinion 3%

    15% of voters would be more likely to vote for a creationist candidate!?! That means 68% of voters wouldn’t care or would be more likely to vote for a creationist candidate. Furthermore, according to Kevin O’Keefe, the average American believes in the literal truth of the bible, can name the three stooges but not the three branches of government (see my post on Ms. South Carolina). Let’s not even talk about global warming.

    What if you were forced to choose between reading blogs written by the “average American” versus subscribing to, say, the New York Times, the Financial Times, the Atlantic, and the Economist? Which would you choose?

    The lazy, slobbering masses have already dumbed down tv and they’re in the process of ruining newspapers.

    For those of you who live in St. Paul (and aren’t lazy and slobbering), be sure to vote next Tuesday:

    St. Paul City Council Primary Election on Tuesday, September 11
    Notice is hereby given that a municipal primary election will be conducted on September 11, 2007 in the City of Saint Paul. The hours of voting in all polling places will be from 7:00 AM to 8:00 PM. Voters will select nominees for the following offices:

    Council Member – Wards 2, 4, 5 and 7

  4. 4 bobbyjones November 13, 2007 at 10:48 am

    “Americans spend a 3.7 billion hours a year in congested traffic. But you will never see ants stuck in gridlock.” How do they do it?

    http://www.nytimes.com/2007/11/13/science/13traff.html?ex=1352696400&en=a667ae1a6bc726cd&ei=5124&partner=permalink&exprod=permalink

  5. 5 Matt H. November 13, 2007 at 4:10 pm

    Ok… here are a couple of data points on how my company has “tested” the wisdom of crowds.

    1) Two years ago, we did an employee survey of Total $ in November Gift Card sales… the merchants were 95% accurate in their guess… the crowd was 99.5% accurate.

    2) Last year, we did a survey at 4 different times of before and during of our $ Revenue for the holiday. The first crowd guess was very close ~1% off… each additional guess was further away… but all were closer than the “experts”. Their theory was that the crowd size (volunteers) was reduced with each survey… they lost interest and as a result fewer participants were not as good.

  6. 6 Dave K. December 12, 2007 at 3:55 pm

    I am about to conduct another ‘Wisdom of Crowds’ experiment with the upcoming college football bowl season…

    My brother-in-law is running a confidence pool for all the bowl games, and I will be submitting two entries- one from my gut, and the other based strictly on the entries for the confidence pool manager on ESPN.com. Though it doesn’t provide the number of users, my feeling is that ESPN’s user base is in the tens of thousands, so that’s a decent sized crowd to work from!

    If you’re unsure what a confidence pool is, you pick the winners for a batch of games and rank how confident you are in each of your picks- 1 being most confident and therefore awarding you the most points if you’re right, down to the least confident pick and points.

    I’ve ranked the winners, from highest to lowest confidence, based on the ESPN users overall confidence percentage levels, and applied the respective confidence level to those picks.

    My BIL instituted a catch, though… since a large majority of the people in the pool went to a school in one of the bowl games, he’s mandating that game be confidence level 1. And, oh, the blasphemy of having to pick against my alma mater on that one!

    But, it’s all in the name of science… I’ll keep you posted.

  7. 7 bobbyjones December 12, 2007 at 4:16 pm

    He’s so confident, he posted it twice. Badgers suck!

  8. 8 Dave K. December 12, 2007 at 6:05 pm

    We have WordPress to thank for the double post…

    And, way to shoot my relative anonymity out of the water… or, are you just pissed that the Gophers couldn’t even muster enough wins to be the Badgers’ water boys on Jan. 1?!


  1. 1 Senator On-line « Broken Spines Trackback on October 18, 2007 at 4:43 pm

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s




Pages

Blog Stats

  • 90,037 hits

%d bloggers like this: