A recent post looked at a study out of Switzerland that found an exception to the belief that the wisdom of the crowd is, on average, more accurate than the wisdom of an individual. The study identified the “social influence effect”, in which individuals had access to the answers of others. The effect of questioning one’s own results as compared to others led to a decline in accuracy.
A commentary on this study posted in a Wall Street Journal article received some heated debate and discussion arguing that WSJ writer Jonah Leher, misread the results of the study.This blog post by Nicholas Carr sums up the accusations, primarily that the wisdom of crowd effect was only shown in one instance and that even in that instance, the only reason it was close to correct was because the median answer, not the mean, was used. In most cases, the average answer provided by the crowd was wrong, contradicting the wisdom of crowd effect.
While the researchers themselves do not misreport this finding, they certainly downplay it. Their introduction cites previous studies supporting the wisdom of crowd effect and focuses on their findings that the effect can be affected when a variable such as social interaction is introduced. Carr explains how the researches do so through normal, acceptable statistical manipulation and provides a concise, humorous conclusion to the debate.
Even in its most basic expression, the wisdom-of-crowds effect seems to be exaggerated. In many cases, including the ones covered by the Swiss researchers, it’s only by using a statistical trick that you can nudge a crowd’s responses toward accuracy. By looking at the geometric mean rather than the simple arithmetic mean, the researchers performed the statistical equivalent of cosmetic surgery on the crowd: they snipped away those responses that didn’t fit the theoretical wisdom-of-crowds effect that they wanted to display. As soon as you start massaging the answers of a crowd in a way that gives more weight to some answers and less weight to other answers, you’re no longer dealing with a true crowd, a real writhing mass of humanity. You’re dealing with a statistical fiction.