Eugene Volokh writes:
University students should generally be evaluated based on what they know, not what they believe.
The context is that he is attacking Michael Dinis policy of only providing recommendation letters for people who believe in evolution. Volokhs point, I think, is that students should know how evolution works, but not necessarily believe that it works.
As Matthew Yglesias points out, this is an odd position. Most theories of knowledge have it that for an agent to know something, she must believe it. It might be added that even more theories of knowledge require that for an agent to know that p, she must not believe not p. Even those who grant that non-belief is consistent with knowledge usually grant, as far as I can tell, that disbelief defeats knowledge. And, of course, creationists disbelieve evolution.