A common fact told in my science classes to help distinguish between correlation and causation was that when ice cream sells are up there are more murders committed. It is easy to make the mistake that ice cream sales are directly related to murder. The reality is ice cream sales and murder are directly related to the temperature, and only indirectly related to each other. When it is hotter more ice cream is sold, and more murders are committed.
Still I can't help but think of some guy in his bedroom, air conditioner busted, thinking to himself, "Should I kill someone or go get ice cream?"