This is a partial transcript of "Special Report With Brit Hume," Nov. 4, 2004, that has been edited for clarity.

Watch "Special Report With Brit Hume" weeknights at 6 p.m. ET!

BRIT HUME, HOST:And now let’s take a look at something from last night’s television.

(BEGIN VIDEO CLIP FROM "THE DAILY SHOW WITH JON STEWART")

JON STEWART, HOST: Ohio emerged as the pivotal state, too close to call due to thousands of uncounted provisional ballots. Cable news viewers were on the edge of their seats wondering would this be another Florida?

HUME: Fox News is now projecting that President Bush will win the state of Ohio.

STEWART: Oh, I guess it won’t.

(END VIDEO CLIP)

HUME: It was that call for Ohio for the presidency at about 11:30 in the evening that brought President Bush to the brink of re-election and made a Kerry win all but impossible. An outcome that earlier in the day had seemed unlikely in light of early exit polling that suggested a Kerry sweep of key states was at hand. So what, in the end, was the difference? For answers, we turn to the man who ran the winning campaign, President Bush’s campaign manager, Ken Mehlman.

Welcome.

KEN MEHLMAN, BUSH-CHENEY ‘04 CAMPAIGN MANAGER: Thank you.

HUME: First of all, let me ask you about that early exit polling (search). Exit polling has been a very reliable indicator on many an Election Day (search), primary days many, many times. We’re all accustomed to relying on it. When we all saw it, we had the same reaction. What was your reaction?

MEHLMAN: The reaction I think was, remember in 2000 and in 2002, there had been a problem with the exit data’s predictability as to the outcome of the election. I think exit data is relevant and is very reliable to talk about the attitudes of voters, what they care about, what they thought. But in this particular case, the exit data about what the result would be, the predictive value was very low…

HUME: But how did you know that when you first heard it?

MEHLMAN: Well, part of what we saw were some absurd numbers. For instance, Pennsylvania, a state that the president lost by about a point, it had us 20 points down. Then we noticed that, in fact, it was unbalanced in terms of gender. It was 58 percent female. In some states, in North Carolina, for example, was 63 percent female. So, what we found was that while it indicated the attitudes of folks, it didn’t predict the final result because of the fact that people that were interviewed were not reflective.

HUME: Well, were you entirely serene about this?

MEHLMAN: As soon as I saw the 20-point Pennsylvania, I said, these numbers don’t make sense. When you saw a 3-point victory in South Carolina, us losing Virginia, all these indicated that, in fact, this was not data that was reliable for that purpose. And then of course, remember 2000. So that was our thinking.

HUME: All right. Tell me now about what you think, if there was one factor in all this, turn out, your get out the vote efforts, the issue of moral values, which emerges at the top of the exit polling as the reason, was the key to all this.

MEHLMAN: I think that at the end of the day, the president was able to motivate more people to want to turn out and support him than John Kerry was. Throughout the election cycle, we always saw that.

HUME: But that is self-evident.

MEHLMAN: Well, but remember, there were certain things that were indicators it was going to happen. A number of strong supporters for the president were always stronger than the number of strong supporters for John Kerry.

HUME: It got pretty close though at the end, though.

MEHLMAN: It got a little closer, but it was always stronger for the Bush people. And I think at the end of the day, what finally closed the deal was that as people got in that booth and they thought about it, they recognized that during the dangerous time, they wanted to stick with someone they knew, someone they thought could keep the country safe. That was, I think, very important. Late deciders we won and did very well with. We won them overwhelmingly.

HUME: By what did you estimate?

MEHLMAN: The final number, I think we won them by like 55 percent, 56 percent late deciders.

HUME: You did not believe all along that that conventional wisdom about late deciders would hold.

MEHLMAN: Not at all.

HUME: Why?

MEHLMAN: We looked at who the late deciders were. They were overwhelmingly moderate and conservative, they were overwhelmingly white.

HUME: Now you knew this — you had a better handle on late deciders because you — was it because of the nature of the polling you were doing?

MEHLMAN: Well, we aggregated — if you aggregated who was undecided throughout the election cycle, what you saw was that the undecided voters, the people that remained to decide, were less liberal, they were more moderate and conservative.

HUME: And you knew this because of polling, right? Because of your own polling or other people’s polling?

MEHLMAN: Public and private. We looked at both.

HUME: But I know you were doing very large samples, but you weren’t doing any national polls. So it was battleground states?

MEHLMAN: It was battleground states (search). But of course, remember, the battleground states are usually reflective of where the nation is generally.

HUME: On balance, right?

MEHLMAN: On balance. If you look at the public polling, you found the same thing.

HUME: Because I mean — I remember, we looked at public polling, too, we’re not as good at it as you guys are, we’re not professionals at it in the same sense.

MEHLMAN: But if you looked at who had yet to decide, it was clear that those folks were not likely — from the perspective of being moderate and conservative, attending church more often, overwhelmingly being white, generally being a little bit older, they were not the kind of voter that Kerry was likely to do well with. The other thing we have to remember is this, look, the fact is if you look historically, late deciders in every time a president has been running for re-election, 1984, 1992, 1996, split their votes, half for the challenger, half for the incumbent. So, this conventional wisdom, in fact, turned out to be inaccurate.

HUME: Yes, it really didn’t break heavily for the challenger since 1980, did it?

MEHLMAN: Right. It did not, no.

HUME: All right. I want to ask you something about your get-out-the-vote effort. The people who were involved on the Democratic side were some very able people. Michael Whouley, a very noted organizer.

MEHLMAN: Very effective.

HUME: Very good. Steve Rosenthal, the former AFL-CIO guy, still is, I guess.

MEHLMAN: Yes.

HUME: Very able guy. These were good people, you know they had plenty of money. Their organization was good in your view?

MEHLMAN: They did a very good job. Here’s the fundamental difference, here’s the question. You’re sitting at home, you’re not sure how you’re going to vote. Who is more compelling to you, a paid worker you’ve never met who shows up at your door with a video for you to watch, or a fellow member of you PTA, a volunteer who tells you why you and she, who share a passion for education, votes to support the president? Fundamentally, the difference is we took a gamble. Our gamble, we said that we think millions of grassroots volunteers, who strongly support this president, are going to be compelling to people that they’re peers with as compared to just some paid worker who shows up at their door. And so we had I think a much larger number of people involved. We had 1.2 million volunteers that were out there working the weekend before the election. We had 7.5 million e-activists.

HUME: E-activists?

MEHLMAN: People who have and forward e-mails to their friends with directions to the polls. We were — the campaign was sending that out. They were forwarding it to people to encourage them to vote. So if you are looking at a target audience and you’re saying, in the target states, these undecided people, about who they’re going to vote for or whether they’re going to vote, if they hear it from someone they know and they trust, that’s going to be a lot more compelling than a stranger coming up to them, who’s being paid to come up to them and knocking on their door.

HUME: In Ohio, a state that had had serious economic troubles, more serious than many other battleground states, you still won that state, would you say that was organization that did it in Ohio?

MEHLMAN: I think it was organization in part. I also think — remember, Ohio is a state that does not — there is no question there is a northern industrial part of the state. There is also a southern part of the state that is much more culturally conservative. Not the kind of a place that a John Kerry is likely to do well. But organization is a huge part to do it with it.

HUME: All right. Ken Mehlman, always a pleasure to talk to you.

MEHLMAN: Thanks a lot.

HUME: Thanks very much for your analysis.

MEHLMAN: Thank you.

Content and Programming Copyright 2004 Fox News Network, L.L.C. ALL RIGHTS RESERVED. Transcription Copyright 2004 eMediaMillWorks, Inc. (f/k/a Federal Document Clearing House, Inc.), which takes sole responsibility for the accuracy of the transcription. ALL RIGHTS RESERVED. No license is granted to the user of this material except for the user's personal or internal use and, in such case, only one copy may be printed, nor shall user use any material for commercial purposes or in any fashion that may infringe upon Fox News Network, L.L.C.'s and eMediaMillWorks, Inc.'s copyrights or other proprietary rights or interests in the material. This is not a legal transcript for purposes of litigation.