Days before the election, faculty members of the Duke Department of Political Science gathered in the Holsti-Anderson Family Assembly Room to share their insights into the election. The event included former congressman David Price’s recounting of how Democratic leaders rallied to persuade President Joe Biden to step down only months earlier. The discussion circled around the theme of political trust: how it is secured — or lost — through public opinion and election integrity.
John Aldrich, a scholar of political parties and Congress, underscored the declining trust in government institutions since the high marks seen in 1964, and noted the long-standing implications of the Civil Rights Act on Southern politics.
Sunshine Hillygus, whose research is in survey design methodology and public opinion, explained challenges to polling predictions, particularly when describing fringe voting blocs, while also raising concerns about voting access in crisis-affected areas.
Lastly, foreign policy expert Peter Feaver pointed toward differences in international strategy between Democrats and Republicans, which may be consequential in this election when polarizing crises abroad affect domestic voter turnout.
Together, their insights provide context for understanding what influences today's electorate.
John Aldrich, Pfizer, Inc./Edmund T. Pratt, Jr. University Distinguished Professor Emeritus of Political Science:
“In the American National Election Studies survey in 1964, citizen evaluation positivity peaked for government trust and confidence in institutional roles. That has been in slow decline ever since and continues today,” he said
“Also, the implications of the 1964 the Civil Rights Act and Great Society social services programming affected Southern politics at a time when incumbency was becoming an advantage.”
“[In 1964] The political parties coalitions were increasingly sorting. The Solid South and a New Deal Democratic Majority in the North lead to a maximally unsorted Democratic majority. You also had what was referred to as Rockefeller Republicans who were socially liberal but economically conservative, and then you had Goldwater types who would vote against the Civil Rights Bill and be economically conservative.
“Yet there’s movement in who supports which party. But they’re swinging: Every time the Democrats go after college-educated voters, they lose rural working-class whites, and vice-versa. This is now a consequence of having two parties that are logically and materially able to compete in any given location in the country.”
David Price professor Emeritus of Political Science:
“Joe Biden wasn't anybody's first choice in 2020. He was a second choice or a third choice. But as the primaries got underway, I was one of those who declared for Biden at a critical point and saw how Super Tuesday could really do the job for him. But it wasn't so much about Biden. It was about what it would take to prevail against Donald Trump, what it would take to get a majority of the electorate on our side.”
“I'm suggesting that maybe something analogous to that has just happened in the Democratic Party. I certainly got involved, as did hundreds of people like me, contacting key leaders, after his debate performance made clear the extent of Joe Biden's decline.
“It was important to engineer a smooth transition to Vice-president Harris as the nominee. The notion that Democrats should have an open convention and have at it — what a disastrous idea!”
Sunshine Hillygus, professor of Political Science:
“Even if we had perfect methods, the very use of polling as a prediction tool is a bit flawed. The scientific value of a survey is that if you take a random sample from a population, you can make inferences about that population from the sample with a margin of error. But we don't know the voting population until after the election.
“And unfortunately, the groups that are hardest to reach in polls — young people, minorities and Trump voters — are also the ones whose turnout is less predictable.”
Sunshine Hillygus:
“Journalists ought to focus on barriers to participation in this election cycle. For example, what are the implications for flooded areas in North Carolina? The absentee ballots might not reach people. Will dislocated voters know what is needed to cast a ballot?”
Peter Feaver, professor of Political Science:
“Foreign policy is rarely decisive in presidential elections. It matters when there's a big policy difference between the two presidential candidates and when those issues are salient. And that has occasionally happened in US history.”
“Between Biden and Trump in 2024, there were some big and salient differences, but on foreign policy the differences were more on tone than on substance. In a close election, some foreign policy issues might matter, and one that is likely to matter is the war in the Middle East, because that divides Democrats but unites Republicans.”
“But a Trump 2.0 administration would be markedly different, with a different cast of advisors, very few —- if any —willing to do what his first-term advisors did, which was to warn the president about the risks of some of his crazier ideas and thus help him make better decisions.
“The makeup of the Republican caucus in Congress will also be different in 2025 than it was in 2017. That will be another check missing to restrain a Trump 2.0 presidency. The foreign policy problems in 2025 are more daunting than those in 2017 and the world is more dangerous. For all those reasons, I think the gap between Trump 2.0 and Harris 1.0 could be fairly large.”
Political Science