5

# I am smarter than the crowd (and so are you).

fishin

1
Wed 20 Aug, 2008 02:00 pm
@roger,
That probably holds true in a pure guessing scenario and there is a fixed answer to the question.

What would have happened if half of the students worked out the problems independently and all came up with the same answers and announced their answers to the class before the voting took place though?

The odds are that the other half of the crowd would have been swayed to select the answers the the first half came up with and, if they were honestly trying to answer the problems correctly, you would have had 100% of the class respond with the correct answers even though half of them never worked it out themselves.

And what happens when the answer is entirely opinion based? When someone asks "what is the best color for a car?" there is no fixed answer. People can offer their opinions and the rationale for them but it's entirely a question of asthetics. Since there is no "correct" answer there is no expert that can prove one is better than another. Yet the car makers only have a small pallette that they offer cars in. If you survey a crowd of 100,000 people with that sort of question and look at the responses it should tell you that if I make and paint cars in the top 10 colors the crowd responded with, I'll increase my chances of selling cars to them over my competitor who paints their cars in colors that only 3% of the crowd selected. Hence, those colors are the "right answer" for their purposes.
0 Replies

ebrown p

0
Wed 20 Aug, 2008 03:17 pm
@Robert Gentel,
Robert...

1) This example constitutes one data point. You can't make any conclusions on one datapoint (this is a basic point of statistics). The experiment was to test if the crowd could determine the weight of the bull, and it tried one time. For a valid result, the experiment would need to be run many times in a controlled setting.

So yes... one result from one run of the experiment is very unscientific (the number of people used to get this one result is irrelevant).

2) The key point I am making in this thread is that the crowd can be (and often is) quite wrong.

The average in this case represents the public opinion of the crowd

In my opinion (based on experience) ... public opinion can be quite far from the truth.
ebrown p

2
Wed 20 Aug, 2008 03:44 pm
@fishin,
Roger, you make a good point.

I think the example that make a crowd very powerful involve the ability to select expertise from within the crowd (as opposed to just averaging high and low votes and assuming the average will be the real answer).

This is why the "Who Wants to Be a Millionaire" is believable iff people who don't have a clue can choose not to vote (in this case the knowledgeable self select). If this is not the case, there is no reason to believe that "ask the audience" would be any more correct than "phone a friend".

Some of the examples we discussed have problems where the solutions are hard to come by, but easy to check. The SETI program is a case in point.

It is the idea that polling a crowd with limited expertise (e.g. if we have a thousand people of which 5 have met fishin's dog) will do better than an informed individual that has me skeptical.

Thomas

2
Wed 20 Aug, 2008 05:08 pm
@littlek,
littlek wrote:
So, when does the crowd start guessing? Can I guess if I've met fishin's dog in person?

You can (as far as the wisdom of crowds effect is concerned, I suppose you will end up being one of the few individuals whose guess is better than the average.

But if I were you, I'd make better use of my insider knowledge. I would trade it with ebrown p for a bribe. (Did they teach you nothing about shady business practices? Tsk tsk tsk.)
old europe

2
Wed 20 Aug, 2008 05:13 pm
@ebrown p,
That's kinda what the "Wisdom of Crowds" author says, though. According to Surowiecki, the four elements required to form a wise crowd are

Quote:
Diversity of opinion
Each person should have private information even if it's just an eccentric interpretation of the known facts.

Independence
People's opinions aren't determined by the opinions of those around them.

Decentralization
People are able to specialize and draw on local knowledge.

Aggregation
Some mechanism exists for turning private judgments into a collective decision.
0 Replies

Robert Gentel

1
Wed 20 Aug, 2008 05:15 pm
@ebrown p,
Quote:
So yes... one result from one run of the experiment is very unscientific .... The key point I am making in this thread is that the crowd can be (and often is) quite wrong...

And you purport to get "scientific" about that by doing smaller tests?

And yes, test sample size matters in these concepts. One of the keys to the wisdom of the crowd is having a diverse enough crowd.
0 Replies

littlek

1
Wed 20 Aug, 2008 05:21 pm
@Thomas,

40lbs and 7:30 pm.
0 Replies

Finn dAbuzz

1
Mon 25 Aug, 2008 08:38 pm
@ebrown p,
The research has been done; the scientific studies concluded, and your contention has proven to be wrong.

Of course there are individuals who are "smarter" than the crowd, but that isn't what you are contending.

Your contention is born from an elitist dismissal of the masses that is too typical of Liberals.

The Proletariat is the essence of humanity but they cannot be trusted to plot their own destinies. Better that the hard choices be left to the elite few who know what is best for them.

Your suggested experiment will prove nothing other than

a) You are more knowledgeable than 20 other A2Kers

b) You are not more knowledgeable that 20 other A2Kers.

Amazing how you Libs just can't stand the notion that "The Crowd" might be right.

If they were, then when voters don't comply with your wishes, you might be wrong.

Heaven forbid!

It might even rob you of your assertion that America is hopelessly racist if Obama is not elected president.

0 Replies