20110614 - Wisdom of crowds

[<Normal page] [PEREZGONZALEZ Jose D (2011). Wisdom of crowds. Journal of Knowledge Advancement & Integration (ISSN 1177-4576), 2011, pages 46-47.]

The wisdom of crowds

The Wisdom of Crowds is both the title of a book and a concept introduced by Surowiecki1 with that book in 2004 . The concept refers to how groups (of people) can outperform even the brightest individuals most of the time (or, as the subtitles goes, how the many are smarter than the few).

This concept actually reflects known statistical principles, such as unbiased sampling, aggregation, and central tendency. Therefore, the wisdom of crowds is not so much a new concept as a 'translation' of those statistical principles into folk psychology. Thus, we could expect the wisdom of crowds working better with larger groups, or working well with smaller groups if those 'statistical' properties have been controlled.

Four pillars

The concept of The wisdom of crowds rests on four principles or conditions (Surowiecki, 2004, p.101):

  • Diversity of opinion: each person have some private information, even if it's just an eccentric interpretation of the known facts
  • Independence: people's opinions are not 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

Furthermore, Surowiecki's book deals with three kind of problems: cognition problems (which have definite solutions either in the form of a single correct answer to each of them or a best answer among competing ones), coordination problems (which require members of a group to coordinate their behavior knowing that each member is trying to do the same), and cooperation problems (which require people to contribute to a group knowing that others may be trying not to do so).

Supporting evidence

As said earlier, the concept of The wisdom of crowds is not so much a new concept as a translation of statistical principles into folk psychology. This, in turn, makes statistics the main supporting evidence for the concept.

  • Diversity of opinion mimics the statistical principle of diversity of measures, such as taking information from a diversity of individuals in order to prevent biases and systematic errors. The ideal (statistical) situation is that where individuals are randomly selected as this ensures that no biases are consciously or unconsciously included. But, when this ideal situation is not possible, ensuring as much diversity among participants as possible is the second-best strategy. An (statistical) idea close to the principle of diversity is the size of the group (or sample size). As the size of the group gets bigger, it 'normalizes' any measures taken by compensating for extreme opinions (pending, again, that no known biases exist in the sample). Notice that this principle is subtly reflected in the concept of The wisdom of crowds, as it makes reference to crowds (in general) and not to a group of a particular size (ie, the bigger the group, the greater its wisdom, pending, again, that all other principles are fulfilled).
  • Independence of opinions mimics the statistical principle of lack of self-correlation among the measures taken. That is, that individual opinions are not being affected by others' opinions or other variables unknown to us. When variables are not independent they mask reality, leading to biased results and invalid conclusions.
  • Decentralization is related to independence of opinions, and so mimics the statistical principle of lack of self-correlation.
  • Aggregation mimics the statistical principle of aggregating variables into categories and summarizing them as results of higher-order analysis. This can be seen in descriptive statistics (which summarize trends in the population, such as averages, standard deviations, etc), bi-variate statistics (which show relationships between variables, such as correlations), and multi-variate statistics (which shows relationships between multiple variables, such as factor analysis or regression analysis).
References
1. SUROWIECKI James (2004). The wisdom of crowds. Abacus (London, UK), 2005. ISBN 9780349116051.

Want to know more?

SUROWIECKI James (2004). The wisdom of crowds. Abacus (London, UK), 2005. ISBN 9780349116051.
This is Surowiecki's book, a amenable reading about basic 'statistical principles' made into folk psychology…without the reader even suspecting it!
PEREZGONZALEZ Jose D (2007). //Knowledge Management Edition™ guide to Surowiecky's book The Wisdom of Crowds.// Pergonomas (US), 2007. ISBN 9780473120436.
Users who apply the instructions contained in this guide will be able to underline/underscore the main ideas in Surowiecki’s book, representing some 13% of the text, in their own copies of "The Wisdom of Crowds". Furthermore, they will be able to highlight the core ideas, representing some 4% of the text. Having those main ideas so highlighted may help the user to review them at any other time without the need for reading the entire book again.

Author

Jose D PEREZGONZALEZ (2011). Massey University, Turitea Campus, Private Bag 11-222, Palmerston North 4442, New Zealand. (JDPerezgonzalezJDPerezgonzalez).


Other interesting sites
320
Journal KAI
105px-Stylised_Lithium_Atom.png
Wiki of Science
120px-Aileron_roll.gif
AviationKnowledge
Artwork-194-web.jpg
A4art
Artwork-162-web.jpg
The Balanced Nutrition Index
Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License