主题：【英文文摘】科学决策：谁更聪明些－ 专家，人群？ -- 西风陶陶

Forget the Experts; Consider the Crowd

Sun Aug 22, 2004 07:26 AM ET

By Samuel Fromartz

WASHINGTON (Reuters) - Donald Trump gets a lot of mileage by lifting his hand, pointing his finger and uttering the words "You're fired!" on his reality TV show, "The Apprentice."

The disgraced competitor leaves the boardroom and takes an elevator to the street, no longer in the running for an executive spot in Trump's empire.

But does The Donald know best?

According to "The Wisdom of Crowds," the answer would be no.

In the Doubleday book, author James Surowiecki, a columnist for New Yorker magazine, argues that groups of people have far more smarts than any one individual, even though people tend to put their faith in leaders or prognosticators or promoters like Trump.

"Chasing the expert is a mistake, and a costly one at that," Surowiecki writes. "We should stop hunting and ask the crowd (which, of course, includes the geniuses as well as everyone else) instead. Chances are, it knows."

Surowiecki began to explore this idea by looking at markets, which through trading arrive at the "best" value for an item. They also act as uncanny predictors.

For example, he mentions that of all the companies that helped build the Challenger space shuttle, it was Morton Thiokol whose stock took the biggest fall the day of the 1986 explosion.

No trader knew that a component made by the company - O-ring seals - would later be blamed.

Yet the crowd had voted, "an unmistakable sign that investors believed that Thiokol was responsible, and that the consequences for its bottom line would be severe," Surowiecki writes.

Academics still don't understand how the market made this judgment, he notes. Continued ..

Surowiecki comes up with many examples of masses of people solving a problem better than any expert.

In one case, a naval officer was trying to locate a sunken submarine. Data about its final moments were sparse, so where it rested on the ocean floor was open to conjecture.

Rather than ask a group of experts to work out the answer, the officer assembled a team of people with diverse skills and had them wager on various scenarios for the ship's demise.

The officer whittled the theories into a collective judgment and then projected where the submarine was.

It turned out to be only 220 yards from where the group scenario found it would be.

SMART CROWDS, DUMB CROWDS

Why does group problem-solving work?

The wide extremes of opinion tend to cancel each other out, Surowiecki writes, so the group as a whole returns to the mean.

This is why a group of people will probably make a better average guess on the number of jelly beans in a jar than any single person.

Four key elements should be in place for a crowd to function wisely.

First, the group must be diverse. Those whose members have various skills and intelligence levels tend to be smarter than groups of experts.

Second, a group should be decentralized, so that each participant has a channel to say what they think. Continued ...

Third, the participants must be independent enough to avoid conforming to a supposed norm.

Finally, there needs to be a mechanism - a market, an aggregator of information, or an open-minded leader - to combine these diverse opinions into a single answer.

The best thing a wise crowd can do is to avoid the stupidity of a despot. It must also avoid the urge for a blind stampede -- the stock market bubble, for example.

To harness the wisdom of groups, businesses should take into account diverse views without influencing or controlling opinions or trying to reach a consensus.

That may mean taking honest input from customers and suppliers -- which readily arrives in the form of business -- and from employees and investors.

Start-ups might be better suited to apply this group approach, because they have much less to lose and are ideally always looking to improve. It's key to their survival.

For his part, Trump took input in his show, judging contestants by gauging their opinions of each other and listening to two of his lieutenants.

But a much more accurate gauge would have come from letting the audience decide which of the contestants should stay and who should go. Trump could have relied on the wisdom of millions rather than his own judgment, which apparently has been less than stellar.

Just witness the recent bankruptcy of his Trump Hotels & Casino Resorts Inc. (DJTC.PK: Quote, Profile, Research) and the 97 percent fall in its stock since its initial public offering nine years ago.

The crowd, alas, has spoken.

(Samuel Fromartz is a Washington, D.C.-based journalist who writes about entrepreneurs and emerging companies. He can be reached through his Web site at http://www.fromartz.com. Any opinions in the column are solely those of Mr. Fromartz.)

• 风雨，LIBERAL还有另外一个理论基础：