As Sen. Hillary Clinton's concession speech echoed through Washington last week, pundits and analysts continued to debate whether the lengthy Democratic primary fight will help or hurt the party in the fall.
Will Sen. Clinton's disappointed supporters vote for Sen. Barack Obama in November or bolt for Republican Sen. John McCain?
If history is any indication, it is unlikely that many of the 35 million Democratic primary voters will cross over to the GOP, according to political scientists who study the effects of primary campaigns on general elections.
"At this point in time, people who are for Clinton are much more negative toward Obama than they will be in November," The College of William and Mary's Ron Rapoport predicts.
[Professor Rapoport teaches Survey Methodology for the Thomas Jefferson Program in Public Policy]
Rapoport has studied how party activists and voters behave when their favored candidate loses a primary election. He's found that, in general, they come around to their party's candidate. In the case of activists, they'll even work for the winner.
"We've done surveys with activists," he said, defining activists as people who work to persuade others to vote for their candidate. "And what we find is that the more active you were for a losing candidate [in the primary], the more active you'll be for the winning candidate [in the general election]. You can recalculate... Over time people like the nominee more and more."
That makes sense, according to Lonna Atkeson, a political science professor at the University of New Mexico, who studies voter behavior in primaries and general elections. In one study of past elections, she looked back at polling data from the 1976, 1980 and 1984 presidential elections. The 1976 Republican primary was particularly long -- the party was still divided between supporters of Gerald Ford and Ronald Reagan at the beginning of the August convention.
During the '76 campaign, surveyors had asked the same group of voters how they felt about each candidate during the early nomination period, the late nomination period and before the general election.
"I looked at their attitudes," Atkeson said. "They came around -- [people who had not voted for their party's candidate originally] found their candidate more favorable during the general election than they did during the nomination."
During a primary, people divide themselves into "ingroups" and "outgroups" -- depending on which candidate they support -- creating competition and intensifying their feelings about their chosen candidate, she said.
"The interesting thing is that if you compare how a group feels about their own in-party candidates compared to out-party candidates, you'll find that [Obama supporters], for example, like Clinton and McCain about the same," she says. "But when that candidate drops out, that's irrelevant -- the question becomes 'gee, Obama's not in this race -- who do I like better, Clinton or McCain?'"
Individual voters' attitudes, however, are only one part of a larger issue of how divisive primaries affect a party's fortunes in the general election.
Political scientists have been studying that question for decades, and the answer appears to hinge on a number of factors, according to Paul-Henri Gurian, of the University of Georgia. In a primary, after all, more than individual voters' attitudes are at stake -- fundraising, organizational structures and other campaign necessities are also put in place.
Gurian, along with Atkeson and other colleagues, analyzed data from presidential elections between 1948 and 2004 and found that a divisive primary could cost the party between 2 and 5 percent of the vote in the general election. They defined a "divisive primary" by the number of delegate votes the winning candidate received compared to the losing candidate -- in a divisive primary, those numbers tend to be near equal.
But how the divisiveness question will play out this year is unclear, Gurian says. One important -- but unquantifiable -- factor is how well a party can "heal the wounds" and come together.
"In 1980, both parties were divided in their primaries. It was Governor Reagan and George Bush Sr. for the Republicans. On the Democratic side, it was President Carter versus. Senator Kennedy," Gurian said. "But on the Republican side they were able to heal the wounds and come together. On the Democratic side, Senator Kennedy was slow to endorse Carter, and even when he did it was without enthusiasm. And a lot of his supporters stayed home, or voted for [moderate independent candidate] John Anderson."
Whether the same thing happens this year, Gurian says, will depend on Clinton and Obama's actions over the next weeks and months.
But Rapoport, for one, believes that not only has the long primary process not harmed the Democratic Party this year. It may actually have helped it, by drawing in new activists and voters who will likely stick around for the general election.
"The nomination process has two functions; one is to nominate an electable candidate, but the other is party building," he said. "Nomination campaigns are an opportunity for parties to bring in new constituencies [...] and then they stay active. So I think that for the Democrats there are some very positive sides to this extended nomination struggle."