According to Wikipedia, the wisdom of the crowd is the collective opinion of a group of individuals rather than that of a single expert. An explanation for this phenomenon is that there is idiosyncratic noise associated with each individual judgment, and taking the average over a large number of responses will go some way toward canceling the effect of this noise.

We’ll take the results of a YouTube video as an example, to test the wisdom of the crowd. I used Majk’s Show as a dataset. It’s a young local guy that interviews random people on the streets, asking them random questions.

We will agree that any hits above 50% will validate the wisdom of the crowd. Follows a table with results:

Question	Answer	Correct answers (%)	Valid?
How do you pronounce “schedule”	sheh·jool or skeh·jool	13/34 = 38.23%	No
Who’s the man on the picture	Albert Einstein	14/32 = 43.75%	No
8/2(2+2)=?	16	11/25 = 44%	No
What’s closer, America or the Moon?	America	9/22 = 40.9%	No
Who was the first president of Macedonia?	Kiro Gligorov	22/50 = 44%	No
What is larger: -4 or -10?	-4	20/36 = 55.55%	Yes

We can further categorize the questions and see where the wisdom of the crowd fails specifically. For example, English seems to be below 40% while everything else is above 40%.

In any case, I think I might have set a bit high grading point for the citizens of my country 🙂 But at least they seem to know how to compare two numbers!