Public Release: 

Lessons from 2004 point the way in 2008 election

University at Buffalo

Democrats and Republicans today are divided more deeply in their views than they have been historically, says election forecaster James Campbell.
Click here for a high resolution photograph.

BUFFALO, N.Y. -- In a scholarly assessment of the 2004 presidential election, University at Buffalo political science professor and election forecaster James E. Campbell, Ph.D., makes several observations about what trends may influence the 2008 contest.

"Perhaps the most interesting thing about 2004 was the impact of voter turnout," says Campbell. "Turnout was increased by the parties working hard to mobilize their potential voters, the competitiveness of this election -- and the memory of how close 2000 was -- and the severe polarization of views between Democrats and Republicans.

"Democrats and Republicans today are divided more deeply in their views than they have been historically," says Campbell, author of "The American Campaign: U.S. Presidential Campaigns and the National Vote."

According to Campbell's research, nearly 60 percent of the voting eligible population voted in 2004, about a 6 percentage point increase over 2000. Increased turnout, he notes, was greatest in states that were competitive and which favored President Bush. Bush carried 12 of 15 states registering the largest gains in turnout, Campbell points out.

"The 2004 election contradicted the conventional wisdom that high turnout routinely favors Democratic Party candidates," he says. "If high turnout helped the Republicans in 2004, high turnout may also be a good sign for Republicans in 2008."

Campbell makes these and other observations in the study "Why Bush Won the Presidential Election of 2004: Incumbency, Ideology, Terrorism and Turnout," published in the current issue of Political Science Quarterly, a non-partisan research journal, available at

Though not as close as the 2000 election in terms of the popular vote, the 2004 election was the ninth closest presidential election since the Civil War and the fourth closest in terms of electoral votes, Campbell notes.

"American politics nationally is now quite equally balanced between the Democrats and Republicans, and likely will remain that way," he says. "So the 2008 election, like 2000 and 2004, should be a close one."

According to Campbell, the net result of the 2004 campaign was quite small, shifting no more than one percentage point of the vote in Bush's favor. The election was as close as it was, and there was so little change during the campaign, because of the extent of party polarization in the electorate, he says. "With the public polarized, the campaign in 2008 is unlikely to shift many voters one way or the other."

Other observations from Campbell:

  • Presidents can survive sub-50 percent approval ratings. "The actual neutral point for presidential approval appears to be in the mid-40 percent range," Campbell says. "Some disapproving voters may still vote for the president rather than his opponent. "

  • Candidates ahead in the polls at Labor Day seldom lose elections. In the 15 presidential elections since 1948, only two candidates ahead on Labor Day lost the popular vote. Bush slightly lead John Kerry in the Gallop Poll of registered voters around Labor Day.

  • The 2004 election demonstrated again that an incumbent whose party has held the White House for just one term is very difficult to defeat. Of the 13 incumbents running since 1868 in this situation, only two have failed to win a popular-vote majority in their re-election bids (Benjamin Harrison in 1892 and Jimmy Carter in 1980).

    "With the 2008 election lacking an incumbent candidate, and with one party seeking a third term, we should expect a close race," Campbell says. "More than half of the open-seat elections since 1868 have been near dead-heat elections."

  • Opinions on the war of terrorism favored Bush in 2004 by a margin of at least 10 points, but opinions about the war in Iraq were nearly evenly divided.

  • Presidential debates can make a difference, but their effects are generally muted and largely dissipate by Election Day. Although the Gallup Poll indicated the public thought Kerry outperformed Bush in the first and third debates and that the second debate was a draw, Kerry only gained about three percentage points from before the first debate to after the third, Campbell notes.

  • For the first time in the history of the American National Election Study (beginning in 1952), voters who indicated that they were strong Republicans outnumbered those who said that they were strong Democrats. (About 40 percent of all voters indicated that they strongly identified with one of the major political parties.)

    "The fact that strong Republicans outnumbered strong Democrats in 2004 for the first time in several generations, may mean that this is more of a 52-48 nation than a 50-50 nation," Campbell says.

  • The 2004 election once again demonstrated that northern liberal Democrats face an uphill battle in post-1968 presidential elections. A majority of the electorate consistently regarded Bush as the candidate who shared their values. The polls consistently found more voters saying that Kerry was too liberal than that Bush was too conservative.

    "With the newly acquired Republican advantage among strong partisans, and unless conditions in the economy or internationally dictate otherwise, a northern liberal Democratic presidential nominee may be even more difficult to sell nationally in 2008 than in past years," Campbell says.


    The University at Buffalo is a premier research-intensive public university, the largest and most comprehensive campus in the State University of New York. UB's more than 27,000 students pursue their academic interests through more than 300 undergraduate, graduate and professional degree programs. The university offers the only degrees in law, pharmacy and architecture in the SUNY system, and is the home of the only comprehensive public school of engineering and only school of informatics in New York State.

  • Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.