This blog has focused on the challenge presented by evaluation of large data sets and questions that evade scientific analysis. Adapting to the conclusion that science alone cannot answer some inquiries, the Rand Corporation pioneered the Delphi Method of consensus building. See Consensus in the Absence of Proof (January 2021).
There have also been those who criticize group dynamics. George Carlin (1937-2008) is perhaps the most memorable with "Never underestimate the power of stupid people in large groups.” A similar sentiment is expressed by Agent K: "A person is smart. People are dumb, panicky dangerous animals and you know it." Men In Black (Sony 1997).
Nonetheless, there are benefits from group analysis, collaborative or not. The Delphi example is one, but is dependent on the participants possessing expertise that is brought to bear on the challenge. What of the common man?
In the 19th century, a polymath named Francis Galton stumbled on a mathematical proof for estimating, called The Wisdom of Crowds. The legend of this method holds that Galton witnessed a contest in which people strove to guess the weight of a cow. None of them was correct, but he noted the average of their individual guesses was surprisingly close to the bovine's weight.
The method has come into repute and therefore use in estimating populations. The concept came to my attention with study of the amazing volume of theft and vandalism in a lovely city, Amsterdam. The place is famous for canals, dope, and Anne Frank. A municipality might gain fame in various paths.
Some might instead associate Amsterdam with canals. The town's fame in that regard has been reinforced with various movies. It is among the most famed canal cities, along with Venice, Italy. Few realize that neither has the “most” canals, a superlative reserved for unassuming Cape Coral, Florida. I have visited each of these three, and there are arguments for each. But if you may visit only one, Amsterdam is a good choice.
It is a city of canals, but also of bicycles. Getting around town is largely a pedal-endeavor. The tool is so popular, they believe the cycles outnumber people there: "Amsterdam has 780,559 inhabitants, who together have an estimated 881,000 bikes." Many of those end up in the canals each year.
The bikes that go swimming are not estimated. The city counts those as they are perennially dredged from the famous waterways. They call the task "Bicycle fishing," and the city claims that "Every year we fish up between 12,000 and 15,000 bicycles."
In this there is illustrated two methods of measure. First the "fished" cycles. Those are quantified based on results. The canals are dredged, cycles and more are recovered, and the cycles are counted. The number is neither estimate or guess. This is a valid method of quantification.
The second is the 881,000 volume of velocipedes. That is a great many bikes. They are not licensed as is common with automobiles and trucks. If they were, then the registration process would yield a significantly accurate quantity (a few might not be registered due to minimal use, as occurs with cars).
Thus, for the quantification, Amsterdam is said to have turned to the Wisdom of the Crowd. The description provided by Amsterdam’s Statistics Bureau, however, also suggests departure from the pure crowd, noting their conclusion is "the average derived from the guesses of a selection of experts." Thus, some suggestion of deviation toward a Delphi model and away from pure "crowd."
Some might argue with the validity of the crowd. Galton's conclusions regarding the famed bovine were, after all, subject to empirical confirmation. The Fair folks knew what the cow's actual weight was. The guesses of the crowd there could be compared to an objective, measurable, known outcome and accuracy measured.
Others might forgive this absence of objectivity. They might note the accuracy of the "guesses" perhaps implicates neither benefit or harm. In the end, what relevance is there to knowing how many cycles exist in Amsterdam (other than perhaps predicting the need for racks to which they can be locked)?
"Not everything that counts can be counted, and not everything that can be counted counts."
The Quote Investigator seems less than convinced of the provenance, but the point remains regarding what "counts." Presumably, no one would expend resources unless something "counts." On that broad assumption or conclusion, it is logical that the accuracy of the count would therefore be important also.
To what end the crowd, Delphi, or guessing? Intriguing indeed.
