Crowd Wisdom Adjusted to Measure Information Popularity

June 2, 2017

The article on ScienceDaily titled In Crowd Wisdom, the ‘Surprisingly Popular’ Answer Can Trump Ignorance of the Masses conveys the latest twist on crowd wisdom, or efforts to answer questions by asking many people rather than specialists. Unsurprisingly, crowd wisdom often is not very wise at all, but rather favors the most popular information. The article uses the example of asking various populations whether Philadelphia is the capital of Pennsylvania. Those who answered yes also believed that others would agree, making it a popular answer. The article goes on to explain,

Meanwhile, a certain number of respondents knew that the correct answer is “no.” But these people also anticipated that many other people would incorrectly think the capital is Philadelphia, so they also expected a very high percentage of “yes” answers. Thus, almost everyone expected other people to answer “yes,” but the actual percentage of people who did was significantly lower. “No” was the surprisingly popular answer because it exceeded expectations of what the answer would be.

By measuring the perceived popularity of a given answer, researchers saw errors reduced by over 20% compared to straightforward majority votes, and by almost 25% compared to confidence-weighted votes. As in the case of the Philadelphia question above, those who predicted that they were in the minority deserve the most attention because they had enough information to expect that many people would incorrectly vote yes. If you take away nothing else from this, let it be that Harrisburg, not Philly, is the capital of Pennsylvania.

Chelsea Kerwin, June 2, 2017


Comments are closed.

  • Archives

  • Recent Posts

  • Meta