From two places, first David Brooks in the NY Times:

The reality is that this was a broad victory for the president. Bush did better this year than he did in 2000 in 45 out of the 50 states. He did better in New York, Connecticut and, amazingly, Massachusetts. That's hardly the Bible Belt. Bush, on the other hand, did not gain significantly in the 11 states with gay marriage referendums.

He won because 53 percent of voters approved of his performance as president. Fifty-eight percent of them trust Bush to fight terrorism. They had roughly equal confidence in Bush and Kerry to handle the economy. Most approved of the decision to go to war in Iraq. Most see it as part of the war on terror.

The fact is that if you think we are safer now, you probably voted for Bush. If you think we are less safe, you probably voted for Kerry. That's policy, not fundamentalism.

And from Slate:

Much has been made of the fact that "moral values" topped the list of voters' concerns, mentioned by more than a fifth (22 percent) of all exit-poll respondents as the "most important issue" of the election. It's true that by four percentage points, people in states where gay marriage was on the ballot were more likely than people elsewhere to mention moral issues as a top priority (25.0 vs. 20.9 percent). But again, the causality is unclear. Did people in these states mention moral issues because gay marriage was on the ballot? Or was it on the ballot in places where people were already more likely to be concerned about morality?

More to the point, the morality gap didn't decide the election. Voters who cited moral issues as most important did give their votes overwhelmingly to Bush (80 percent to 18 percent), and states where voters saw moral issues as important were more likely to be red ones. But these differences were no greater in 2004 than in 2000. If you're trying to explain why the president's vote share in 2004 is bigger than his vote share in 2000, values don't help.

If the morality gap doesn't explain Bush's re-election, what does? A good part of the answer lies in the terrorism gap. Nationally, 49 percent of voters said they trusted Bush but not Kerry to handle terrorism; only 31 percent trusted Kerry but not Bush. This 18-point gap is particularly significant in that terrorism is strongly tied to vote choice: 99 percent of those who trusted only Kerry on the issue voted for him, and 97 percent of those who trusted only Bush voted for him. Terrorism was cited by 19 percent of voters as the most important issue, and these citizens gave their votes to the president by an even larger margin than morality voters: 86 percent for Bush, 14 percent for Kerry.

In fact:

Amusingly, one big improvement in Bush ’s performance actually came from those who never go to church. He won 36% of this group compared to 32% last time.

And then there's this:

A variety of exit polls indicate lesbian/gay voters gave George W. Bush the same support, 23 percent, as they did in the 2000 presidential election, according to the Washington Blade.

So what was that about all the evangelical homophobic morons voting Bush into a second term?