In college (back when you ran computer programs with a stack of punch cards), I took a statistics class from a bona fide genius. Between teaching us the nuts and bolts of statistical analsys, he frequently reminded us of the points that melissasvh states. Every class, he regaled us with a story of some corporation or interest group hiring him to statistically "prove" their point, and then the other side hiring him later to "prove" the opposite. Being a genius, he always found a way to make his clients happy.