Travis brought up some good stuff in the comment in the post below, so much so that I feel justified noting a few things in a brand-spankin'-new post.
The following is a bit of a simplification, so please, no hate mail.
Statistics is taking individual data points and combining them to get information about groups. This is the direction of statistics. It does not work the other way. This is an important point, so know it, breathe it, live it.
As an example: If you were to survey the height of men and women in a sample of a given sample of a population, let's say University of Wyoming undergraduate students, for example, you would probably find that the average height of men is larger than that of women. There's a chance you won't, which we won't get into right now, but let's say that is indeed what you find.
Assuming you can generalize to the general population of UW undergraduate students (not much of a stretch), there are two ways that you can use this information. One is working up, as in "male UW undergrads are taller than women UW undergrads." This would be a (roughly) correct generalization. Another way to use it would be to say that a given man is taller than a given woman (without measuring). Now, you will be right more often than you are wrong, in aggregate, but the statement itself is poor because you really have no basis for the decision. Statistics cannot be used to take the general and apply it to the individual. It just doesn't work.
So, to say that "men are more likely (whether for genetic or other reasons) to be violent criminals, rapists, child molesters, sexists" may be right more often than wrong, but it puts the focus in the wrong place, which should be that those activities are most often performed by men. Which is very true, and we should do some things about that. But that's a subject for another post.
It's a subtle point, but one that we all need to keep in mind. Statistics are squirrily. They have a tendency to take on a life of their own, and they seem to have an instinct for self-preservation. Let's keep them leashed and working for us, shall we?
