One major statistical question Freakonomics discusses is that of relationships, otherwise stated as the question of what causes what. As the authors point out in the book's Introduction, a statistical relationship indicates that things are affected by something along with other things, but a statistical relationship indicates no direction of change or of affect. In other words, while statistics may show a relationship that indicates that--hypothetically speaking--the price of greens go down when the price of fruits go up, that relationship cannot say whether the fruits cause a change in the price of greens or whether greens cause a change in the price of fruits or whether the changes to the price of both greens and fruits are caused by some other factor, say, the changing seasons or the price of gasoline. So one major statistical question addressed by journalist Stephen J. Dubner and economist Steven D. Levitt is that of statistical relationships and the directions of relationships or the unrevealed causes that underlie the directionless statistical relationships.
A correlation simply means that a relationship exists between two factors--let's call the X and Y--but it tells you nothing about the direction of that relationship. It's possible that X causes Y; it's possible that Y causes X; and it may be that X and Y are both caused by some other factor, Z. [...] Unless you have more information, however, it's hard to say what's causing what. (Freakonomics, Introduction)