You just have to be careful about what conclusions you draw from the data. You have to remember that correlations do not necessarily indicate cause and effect, and they can be totally misleading.
For example, lets say that colleges become convinced that a predictor of success in college is a person's height. The colleges might then proceed to admit only those students who are 6 foot tall or more.
Eventually, you will observe that most of the people in higher paying jobs are 6 foot tall, or higher. You would see an apparent correlation between 6 foot height and earning ability. The problem with this is that the correlation between height and earning power is a false one - an invention, a self-fulfilling prophecy. We cannot accurately conclude from this correlation that 6 footers have more earning ability than shorter persons, or that shorter persons do not have as great an earning potential, although the temptation to do so is strong.
However, that said, I believe that humans are "hard wired" to be racists. Its the most natural thing in the world to categorise people according to skin color, or religeon, or nationality or what have you. That is why you have to tread carefully when discussing these issues in general terms. We are all prey to our own points of view and our human tendancy to jump to conclusions. (Jumping to conclusions is related to survival.)
I agree with Animal, that the concept of race can be misleading. I also agree with Miko 2d that as long as you treat individuals fairly, you won't go too far wrong.