Dean's World

Aziz also makes a good point: if you preferentially pull your samples from one end of the bell curve -- whether that's an intelligence bell curve, a suicidal impulses bell curve, a despondent and willing to die bell curve, or a willing to die for a cause bell curve -- the remainder will form either a new bell curve or a one tail curve, and either way the mean will shift away from what you're selecting for. Aziz assumes it's an intelligence bell curve, and I'm inclined to agree with him.

Aziz
Excellent points, especially the "up yours" point. ROFL.