Print Page   |   Contact Us   |   Sign In   |   Register
Complexity Matters
Blog Home All Blogs
Search all posts for:   


View all (443) posts »

Wisdom of Crowds Works Best for Easy Choices

Posted By Prucia Buscell, Friday, July 29, 2016
Small Groups May Make Wiser Choices
But Don't Abandon Need for Elections
The wisdom of crowds apparently works best when there is a pretty straight forward correct answer.   What it the weight of the ox? How many jelly beans in the a jar? Or a tougher problem, but one that still had specific right answers: the Defense Advanced Research Projects Agency (DARPA) placed 10 red weather balloons at different locations around the continental U.S. and launched a public competition to find them. Where were they?
Santa Fe Institute Professor Mirta Galesic and colleagues from the Max Planck Institute for Human Development in Berlin have investigated how many people make a wise crowd, and their research suggests when it comes to qualitative decisions, a small to moderate group may be better.    
A story in the Santa Fe Institute newsletter Parallax explains, for example, that a team of five to seven doctors is likely to do better than a much larger group of doctors at identifying a diagnosis that fits patient symptoms. Financial official forecasting unemployment, economic growth and inflation, and panels of forecasters predicting political victories are also likely to perform best in small to moderate sized groups.
While past research on wisdom of crowds has looked at decisions about how much or how many, the current research examined more difficult decisions that combined an unpredictable mix of easy and hard choices. The researchers mathematically modeled group accuracy with groups of different sizes and different combination of decision difficulties. The smaller groups did better. Galesic says the reason is a matter of probabilities. A group of experts of any size will probably get an easy decision right. For more difficult decisions, the story explains, "moderate sized groups are more noisy representations of the overall population of experts," and can by chance arrive at a correct answer even if most of the experts in the larger population wouldn't.
What about democracy? Galesic doesn't think we should abandon large scale referendums and national elections. Those choices, she says, represent preferences, with a whole spectrum of consequences, rather than a right or wrong answer.
Mark Buchanan, a physicist, author and columnist for BloombergView, contemplating this question and the new research, notes that decision making bodies around the world tend to work with small numbers-juries, parish or municipal councils, central bank boards and parliamentary committees, which usually have five to 40 participants. Buchanan suggests U.K. voters who expressed clear discontent on matters of globalization and immigration may not have reached the wisest decision on Brexit.
Arguably, he writes, the referendum didn't have a right or wrong answer, but he says it was a "crude instrument for deciding such an important and difficult issue," especially because much of the British public has adopted some inaccurate ideas, such as believing there are twice a many immigrants in the country as there actually are. He says U.K. leaders will need to examine carefully how to respect the will of the voters and determine whether that respect demands invoking Article 50 of the Lisbon Treaty, which begins the process of taking the U.K. out of the European Union.
The MIT team that got the right answers on the balloon locations had help from 4,400 volunteers quickly recruited from across the country. Click here to read how the team won the DARPA challenge.

This post has not been tagged.

Share |
Permalink | Comments (0)
Association Management Software Powered by YourMembership  ::  Legal