Some people seem to think that epistemic reasons and evidence come to the same thing. (Call this 'the equivalence thesis'.) I suspect that some people think that this equation suggests that some sort of evidentialist view must be the correct one.
Don't think these things!
Here's the quick and dirty argument (inspired by some things that David Owens said (or, probably said-this is from recollection) in Reason without Freedom). Suppose that you shouldn't believe p unless you have sufficient evidence to believe p. You might, if you like, think of the evidence you have as epistemic reasons that somehow help to justify believing p. According to the equivalence thesis, all the epistemic reasons will be evidence that concerns p. But that cannot be. If you don't have sufficient evidence to believe p, you oughtn't believe p. If you oughtn't believe p, you have a decisive epistemic reason not to believe p. This reason, however, is not some further bit of evidence you have. So, the equivalence thesis must be false. Some epistemic reasons must not be further evidence you have. If it were, the obligation to refrain from believing without sufficient evidence couldn't be binding on you.
There's a point here that's simple, but important, and that is that it's undeniable that some epistemic reasons will have a bearing on whether you should believe p whether or not you have those reasons in your cognitive possession. The fact that you don't have sufficient reason to believe p, for example, constitutes a decisive reason not to believe p even if it's one that you're non-culpably ignorant of.
If the undeniable point is indeed an undeniable point, it shows that lots and lots of things that people say about justification are mistaken. I've argued that McDowell misses just this point when he tries to show that we need to reject the traditional view of experience in his epistemological argument for disjunctivism. I also think that people miss this point when they criticize people for defending externalist epistemic norms. What's wrong with these norms, people often say, is that they imply that we have decisive reasons not to believe even when we're non-culpably ignorant of these reasons and it is reasonable not to refrain from believing in just the way that these (alleged) reasons tell us to. Well, that cannot be what's wrong with truth or knowledge norms, not if the undeniable point is correct, for this feature of truth and knowledge norms is a feature that all norms share in common.