The Wisdom of the Crowd: A Two-Part Odyssey

The wisdom of the crowd is a recent phenomenon, toppling traditional paradigms of expertise to scientifically produce better information when making decisions. Expertise is still relevant within the wisdom of the crowd concept; however statistically speaking, individual answers can suffer from random errors of judgment. Thus, it is frequently better to rely on mathematical combination of people's opinions in order to make the best decision. This approach, which is referred to as the wisdom of the crowd, lowers random error and delivers an accurate answer that is more reliable than any response within the cohort. Already used in retail sales and internal company improvement policies, as well as in prediction markets, there currently exists no market solutions for end users - both big and small - to harness this concept in decentralized opinion markets.
Concepts, 2018.08.09
In this two-part blog series, we take a deeper dive into the history of the wisdom of the crowd, its components, what the opposition would say, and how the Finnoq protocol creates an environment by which crowd wisdom can be extracted. In part one, we take a look at its recent history and discovery, as well as the landmark breakthrough of James Surowiecki and some common objections brought to the floor on the matter.
 

History

Sir Francis Galton published “Vox Populi” or the voice of the people in 1907. In this article, Galton posited “The material about to be discussed refers to a small matter, but is much to the point.” In hindsight, how wrong he was. Galton observed how close to the actual weight 787 participants could be regarding a “fat ox”. In doing so, he realized the crowd’s average response was closer to the actual weight than the responses participants who were mathematically “smarter” (averaged among the 25th and 75th percentiles, respectively).

After this initial experiment, sociologist Hazel Knight of Colombia performed many studies in the 1920s. In one of her studies, students estimated classroom temperature. The actual room temperature was 72 degrees Fahrenheit, and the group’s average guess was 72.4 degrees Fahrenheit. The striking accuracy of the group was noteworthy.

Many years later, Jack Treynor concluded in his 1987 study that, “attributes are abstractions. Many of us are diffident about our ability to evaluate abstractions, deferring instead to experts.” His inference referred to the classic jelly bean experiment, where 56 people averaged a collective response only 3% off of the total jelly beans in a jar. These disparate instances were at once encapsulated by James Surowiecki and his landmark 2004 book “The Wisdom of Crowds”.

 

Surowiecki’s Breakthrough

These phenomena (and more) were analyzed at length by Surowiecki. He describes why and when the community is smarter than one single participant in that community, building off of the studies presented over more than 100 years ago. According to Surowiecki, a wise crowd has at least three key characteristics:  

 

Aggregation methods can take many forms, but what is most important is having a consistent methodology for collecting responses. Whether it be mathematic aggregation, deliberation groups, or prediction markets as examples (and complexities therein), being able to effectively tally the results is step one in establishing the potential for crowd wisdom.

Diversity is proven scientifically to tap into crowd wisdom, whether broadly speaking or within specific groups containing particular knowledge sets. The same 2004 study by Scott Page and Lu Hong demonstrated that a diverse group of participants guessed better collectively than the highest-performing participants in the group. Even experts cannot always respond perfectly to open questions when compared to a diverse group. Further, as “opinion distance” increased in one study (representing diversity), so did crowd performance.

Independence and its impact on crowd wisdom, was highlighted with an experiment at ETH Zurich. A control group responded to questions, and were informed after of the experimental group’s responses to the same questions. However, an experimental group was informed of the control group’s responses before rendering theirs. The resulting conclusion was that the more actors know about the answers of others, influence and bias will morph answers. In the end, collective results become unreliable when such circumstances are met. 

Overconfidence, Systematic Extremes, and Crowd Size

With that in mind, many raise objections to the wisdom of the crowd concept. Specifically, Surowiecki looked at situations where the wisdom of the crowd fails (e.g. bubbles within financial markets). He argued that members of the crowd can be overconfident about the opinions of others, adjust and imitate accordingly, and ultimately make irrational choices. As examples, Surowiecki mentions the “dotcom bubble” as well as the “US subprime crisis of 2007”. To second such claims, dually-examined cases of overconfidence and systematic extremes have also been observed as a precondition to bias for opinion formation. Thus, when it is easy to point the finger at the crowd itself, one must take a look at crowd dynamics to determine if the crowd is actually capable of being wise.

Also, for those who consider that crowds too small cannot generate wisdom, such was debunked even within a group of less than 10. Psychologists even observed that individuals can create crowd-like circumstances internally; among 428 participants, the average of two guesses from a single individual was more accurate than either guess alone. Even considering some understandable and necessary objections, there is evidence refuting that a particular obstacle voids the wisdom of the crowd approach. Greater observation is required when crowds are unwise.

Conclusion

As you can see, Surowiecki really moved the ball forward for wisdom of the crowd. While his foundational book gained traction, there have been many interested in testing his ideas. Further, others have added to the list of essential elements in harnessing the wisdom of the crowd. This, as well as how the Finnoq protocol keeps all of these factors in mind, will be the focus of part two. Stay hungry for knowledge in the meantime, and get active in the discussion through the myriad of channels below.

Haven’t had enough? Join our global community today:

  1. Knowledge with one click: Sign up for our newsletter
  2. Join the finnoq community on
You are unable to comment and share this blog post, as your Browsers' "Do Not Track" settings are currently enabled.
×

finnoq logo

Finnoq Newsletter

Stay informed on the latest news from Finnoq and its Blockchain protocol.

Get the inside scoop on the FNQ token model and upcoming ICO
15k