I don't see what this has to do with race. A Swede or Chinaman's culture is far more different from mine than an African American's is likely to be. We don't import opinionated foreigners so we can get their perspectives, even though it would be a far more efficient way of getting diversity.
If you introduce those different points of view into a college class, then the quality of classroom discussion can, and probably will, be better than if the class consisted of kids from the same racial and economic background.
Maybe. Or maybe the ideas will be trite recapitulations of what they've heard in the media.
In any case, it really depends on the field. I don't give a tinker's damn what my students' ideas are on the Second Law of Thermodynamics. They're either correct, or incorrect (or, most likely, they've never heard of the second law), but it really doesn't matter; they can either learn and learn to apply the second law, or they can fail the class. So can we dispense with 'diversity' in the sciences?(please!)
The idea that a group of immature youngsters, sitting around and expressing almost completely uninformed opinions on a subject of which they know next-to-nothing, amounts to education is a comical one, when you think about it. Sure, there are some pedagogical purposes to a Socratic method of teaching (IMNSHE, it's a good way of keeping 'em awake), but Socrates, AFAIK, never asked Phoenicians into his class to provide a 'non-Greek' perspective. I doubt Socrates would have believed ideas depend on origins.
Your comments offer very little, one way or the other. As for importing foreigners, I initially had that in there, too, but decided to leave it out: this is a discussion about American citizens.
Your second comment is useless. "It might, or it might not." Speaking from personal experience, it does offer a better discussion.
You're of course speaking as a technical guy -- which says a lot more about you than it does about the quality of, say, an ethics class.