Suppose you flipped a normal looking penny 5 times and got "heads" each time. What do you predict is the probability of the next flip being a head? Pick one:
If you fall prey to the "gambler's fallacy" you will believe that chance/nature is self-correcting, and you will predict that the coin is due for a tail (to counter all those heads). You would guess a p value < .50.
A scientist would say that:
- the coin has no memory,
- the coin is unbiased,
- flipping pennies is a quasi-random process
So the probability of a head on the 6th flip is .50-- the same as it would be on the first flip. This is commonly assumed to be the correct answer, but it is only correct if the assumptions are true (level 3 examines some of these assumptions) and if the question is interpreted in the normal manner (level 2 examines a common misunderstanding).
go to level 2 and 3 explications