As I've said elsewhere, I'm glad that the National Center for Science Education doesn't take a position on theism vs. atheism and involves many religious believers who support the promotion of good science.
Suppose U.S. demographics on belief and nonbelief were reversed, so that atheists made up 80%+ and those who explicitly believed in God were about 4-5% of the population (with the difference filled by agnostics, closeted believers, etc.). Suppose further that demographics of believers in science were reversed--with most physicists and biologists being religious believers, who commonly said things like "the Big Bang shows evidence of a beginning of time, started by a creator God," and "the intricate design of biology shows the hand of God."
Presumably Nisbet would tell those religious scientists that they shouldn't say things like that in public, even if they firmly believe them to be true, because they would cause the atheist majority to stop listening to the part that's actually science. And I think he'd have a point. To the extent that Dawkins and Myers go beyond the science into areas like philosophy and normative ethics, they are making non-scientific claims that are not entailed by the scientific evidence (though I happen to agree with them that atheistic views fit much better with the evidence than religious views). A division *can* be drawn, and if your goal is persuasion, *somebody* needs to draw the division and communicate with the audience that otherwise wouldn't listen without including the nonscientific parts that will turn them off.
But, contra Nisbet, that somebody doesn't need to be everybody, or Dawkins or Myers in particular.
Matt Nisbet and Chris Mooney have been getting support in their statements from people like intelligent design advocate William Dembski and "Expelled" co-writer Kevin Miller, but I suspect that they would not really agree with Nisbet's position if the demographics were reversed as above--they would be defenders of the religious version of P.Z. Myers. Their position strikes me as opportunistic rather than principled.
Which raises the question--if you support P.Z. Myers' approach and think that it's beneficial for the promotion of science, but you wouldn't support a religious counterpart's approach in the reversal scenario, does that show an inconsistency or lack of principle in your position? I don't think so, and my parenthetical comment is a start of the answer I'd give to why. (I think the underlying causes of the demographics are of relevance, and it's interesting that only Nisbet seems to have tackled that subject in this discussion.) But I'm interested in hearing what others have to say, either way. I suspect that John Lynch and John Wilkins would argue that it does show an inconsistency.
UPDATE (April 2, 2008): James Hrynyshyn at The Island of Doubt ScienceBlog offers a critique of Nisbettian framing. Somehow, I get the impression something's missing here, though. Claiming that scientists are completely objective and trained to be so is to miss the fact that Kuhn, Latour and Woolgar, and the sociologists of science aren't completely wrong about everything. (I'm still a big fan of Philip Kitcher's book, The Advancement of Science: Science Without Legend, Objectivity Without Illusions.)
UPDATE (April 3, 2008): John Wilkins offers a defense of "the f-word" in terms of simplification for the purposes of pedagogy.