An approximate answer to the right question is worth a great deal more than a precise answer to the wrong question.
Super Freakonomics page 224: "as the renowned statistician John Tukey once reportedly said,"
