Monday, December 08, 2008

So much for the wisdom of sociologists

A tenet of pop social psychology is based in the 2004 book the Wisdom of Crowds by James Surowieck. Basically it holds that if you have the right group of people solving some puzzle, they will do better than just a single person.

Well, it's a fun idea to think about, but it's hard to see how it works. (In statistical terms, increased sample size means more efficient estimates, but not less biased.)

One can simply reflect on past presidential elections to know that sometimes even the general population gets things wrong.

This last election, we had a betting pool here in the department picking the number of electoral votes that Obama would win. There were about a dozen of us in the pool, and every single one of us were too low. Obama won 365 votes, and the highest guess was in the 350s.

Maybe the wisdom of crowds doesn't apply to sociologists?


3 comments:

Gruntled said...

The wisdom of crowds works with a group of people with different expertize. A group of sociologists, even a big group, has too much of the same kind of expertize. In the illustration that Surowieki uses as the beginning of the book, Francis Galton watched a very mixed group at a country fair estimate the dressed weight of a cow. Galton thought the butchers and cattlemen would be closest, but the crowd as a whole guessed better than the professionals.

Michael Kruse said...

Ditto on Gruntled. Then again:

"Maybe the wisdom of crowds doesn't apply to sociologists?"

Maybe its just wisdom that doesn't apply to sociologists.

:-)

Brad Wright said...

Yeah, you guys are right, so I'm changing title. :-)