Posted by tpc at June 3rd, 2008

Most people tend to confuse randomness with uniformly distributed. They believed that if you toss a coin 100 times you would get 50 heads and 50 tails. Lots of people have discussed this, gave examples of how they detected pseudo randomness. I would like to add my very own example.

I have 1600 exam scripts ordered by student matriculation numbers, divided into 16 piles. A total of 11 students asked to checked their scripts. One would think the 11 would come from different piles right? Here’s the actual stats: number of scripts (from pile number)

2 (1), 1 (4), 1 (6), 2 (7), 2 (9), 3 (14).