I have often heard mathematicians complain — most recently, last night — about their work being mangled when it gets covered in the press. Why don’t science journalists check with their sources to make sure that the science is presented accurately?
There’s a great discussion of this issue at PLOSBlogs, featuring many well-known science writers and highly-placed editors in the comments. It’s a tough issue. On one side, journalists are quite likely to make mistakes about technical subjects (not only science) even if they’re very diligent when conducting the interview. On the other hand, journalists are not public relations officers, and I tend to agree that it’s important to preserve that distinction, even when there are some costs.
As for me, I would never show copy to a source prior to publication. Then again, because I mostly write about math, I think people cut me a lot of slack — if I oversimplify somebody’s work, they know that I know that I’m oversimplifying, and respect that I’m bowing to journalistic necessity.