Editor's note: This story is part of a series based on a survey of Duke community members conducted by The Chronicle from Oct. 16 to 18. You can read more about our methodology and limitations here, or read all of our survey coverage here.
The Chronicle surveyed Duke students, faculty and staff about their voting plans in advance of the 2024 general election.
We found that over 80% of respondents plan to vote this election cycle, and over 75% intend to vote for Democratic nominee Kamala Harris in the presidential election. The majority of respondents identified as liberal, a trend that was consistent across demographic groups.
Registration status
Over 4 in 5 Duke community members plan to vote in the U.S. presidential election, and nearly all were already registered at the time of the survey.
Roughly 88.74% of respondents said they planned to vote in the 2024 presidential election, compared to only 2.94% who said they do not plan to vote. An additional 8.33% responded that they do not plan to vote due to their citizenship status.
Respondents were asked what their current voter registration status was for the 2024 general election. About 87.37% of respondents stated that they were already registered, and 2.45% of respondents were eligible to register but had not yet done so at the time of the survey. About 8.92% of respondents were ineligible to vote due to citizenship or age. Roughly 1.27% of respondents were unsure of their registration status.
Most respondents are registered to vote in North Carolina, and registration status varies little by demographics.
A majority of those who were registered to vote — 74.38% — were registered in North Carolina, followed by Florida, at 2.07%.
Out of the respondents registered to vote in North Carolina, 42.9% were originally from the Tar Heel state. Californians made up the next-largest cohort of North Carolinian registrants at 6.09%, followed by New York at 5.17% and Texas at 4.77%.
Across all racial groups, a significant majority of respondents were registered to vote. Notably, 17.54% of Asian respondents were ineligible to vote due to citizenship. According to a Duke Facts dashboard published by the University, the top three home countries for undergraduates as of fall 2023 were China, India and South Korea.
About 90.89% of women respondents were registered to vote, compared to 83.74% of men. A higher percentage of male respondents were ineligible to vote: 11.57%. Roughly 3.69% of male respondents reported being eligible to vote but hadn’t registered at the time of the survey, compared to 1.55% of women.
You can view the disaggregation of registration status by political identity on Tableau.
Political affiliation
Respondents were asked for their political identifications, from a scale of “very liberal” to “very conservative.” Roughly 30.56% of respondents identified as “very liberal,” 41.53% identified as “somewhat liberal,” 18.41% identified as moderates or centrists, 7.44% identified as “somewhat conservative” and 2.06% identified as “very conservative.”
The majority of respondents identify as liberal; more women respondents identify as liberal than men.
When grouped by gender, 83.33% of women respondents identified as either “somewhat liberal” or “very liberal,” compared to 55.91% of men that identified as either “somewhat liberal” or “very liberal.” Roughly 13.4% of women respondents identified as moderate/centrist, compared to 25.12% of men. Only 3.26% of women respondents identified as either “somewhat conservative” or “very conservative,” compared to 18.97% of men.
There were similar breakdowns of political identification across all Duke affiliations — undergraduate and graduate students, faculty, staff and other — with faculty having a slightly larger proportion that identified as moderate/centrist and slightly less that identified as either “somewhat liberal” or “very liberal.” The majority of respondents of all Duke affiliations identified as either “somewhat liberal” or “very liberal.”
Of racial groups with at least 15 respondents, the breakdowns of political identification were generally similar, with the majority identifying as either “somewhat liberal” or “very liberal.” Black or African American respondents had the largest proportion of respondents identify as liberal, at 81.02%, and Middle Eastern/North African respondents had the smallest proportion of respondents identify as liberal at 66.66%
You can view the disaggregation of political identity by party affiliation on Tableau.
Presidential picks
Over 3 in 4 Duke community members plan to vote for Kamala Harris.
The majority of respondents — 78.06% — intend to vote for Democratic presidential and vice presidential candidates Kamala Harris and Tim Walz. Roughly 7.64% of respondents intend to vote for the Republican ticket of Donald Trump and JD Vance. About 11.26% of respondents are not voting in the presidential race, and 3.04% intend to vote for third-party candidates.
Undergraduate and graduate students indicated very similar opinions on their votes for president. Roughly 76.92% of graduate respondents intend to vote for Harris, and 7.69% intend to vote for Trump; while 75.69% of undergraduate respondents intend to vote for Harris, and 7.85% intend to vote for Trump.
Faculty and staff were even more likely to vote for the Democratic presidential ticket: 89.66% of faculty reported their intention to vote for Harris, and 88.89% of staff did the same. While staff respondents were significantly less likely to be unregistered, they were also less likely to sit out the election, as none of them indicated that they would not vote despite being eligible.
Presidential candidate preference largely aligns with political identification.
When grouped by political identification, none of the respondents who identified as “very conservative” indicated that they intend to vote for Harris, while none of those who identified as “very liberal” indicated that they intend to vote for Trump. Meanwhile, 14.86% of respondents who identified as “somewhat conservative” intend to vote for Harris, while only 0.24% of respondents who identified as “somewhat liberal” intend to vote for Trump.
Respondents who identified as moderate/centrist were the most likely to not vote despite being eligible, at 7.53%, followed by respondents who identified as “somewhat conservative” and “very conservative,” at 5.41% and 4.55%, respectively.
Preferred presidential candidates
Most liberal respondents prefer Harris, while conservative respondents are more divided.
Respondents were asked to identify their preferred presidential candidate — “even if they aren’t on the ballot” — from a list of nine options, with a 10th option of “other.” The nine options offered were Kamala Harris, Donald Trump, Joe Biden, Ron DeSantis, Nikki Haley, Robert F. Kennedy Jr., Jill Stein, Chase Oliver and Cornel West.
Respondents who identified as “very liberal” or “somewhat liberal” heavily preferred Harris as a presidential candidate, at 81.15% and 84.05%, respectively. Harris was also the preferred candidate for a plurality of moderate/centrist respondents, at 44.62%, followed by Haley, at 18.28%.
Conservative respondents were more split. Trump received support from the majority of “very conservative” respondents, at 54.55%. He was closely followed by Haley, at 22.73%, and DeSantis, at 18.18%. A plurality of “somewhat conservative” respondents preferred Haley — 32.43% — compared to 28.38% who preferred Trump and 10.81% who preferred DeSantis.
More Harris voters are satisfied with their choice than Trump voters.
Harris was the preferred presidential candidate for 80.78% of respondents voting for her, while Trump was the preferred presidential candidate for 42.31% of respondents voting for him.
The top three preferred candidates — other than Harris — for Harris voters were Haley (4.27%), Stein (3.14%) and Biden (2.14%). The top three preferred candidates — other than Trump — for Trump voters were Haley (24.36%), DeSantis (14.1%) and Kennedy (7.69%).
Respondents who answered “other” were given an option to name an alternative candidate of their choosing.
The majority of respondents — 11 — who chose “other” reported not having a preferred alternative candidate. Ten respondents named Transportation Secretary Pete Buttigieg, eight named Rep. Bernie Sanders, six were unsure what their preferred alternative candidate would be, six named Republican candidate Vivek Ramaswamy, three named Michigan Gov. Gretchen Whitmer and two named former First Lady Michelle Obama.
Gubernatorial picks
Over 9 in 10 Duke community members plan to vote for Stein for N.C. governor.
An overwhelming majority of respondents — 91.14% — plan to vote for the Democratic nominee for N.C. governor, Attorney General Josh Stein. Roughly 5.23% of respondents plan to vote for the Republican nominee, Lt. Gov. Mark Robinson and 3.62% plan to vote for a third-party candidate.
No respondents who identified as “very” or “somewhat” liberal indicated an intent to vote for Robinson. While a majority of “somewhat conservative” respondents backed Trump, they were more divided on the gubernatorial race, with 44.68% backing Robinson and 34.04% backing Stein. The majority of “very conservative” respondents — 64.71% — indicated their intent to vote for Robinson, compared to only 5.88% who intended to vote for Stein and 29.41% who intended to vote for another candidate.
You can view additional disaggregations of gubernatorial preference by race and gender, race and political affiliation, Duke affiliation, political affiliation and religion on Tableau.
Get The Chronicle straight to your inbox
Signup for our weekly newsletter. Cancel at any time.