As we do every four years, Americans are looking again for the best man to lead the nation. Well, not quite. One of the biggest changes in what Americans want in a candidate comes from the headline above: We are no longer looking for Mr. Right, but the best man or woman for the job. In 1937, when Gallup first asked about supporting a woman for president, only a third of adults said they would vote for her. That response is virtually unanimous now. What else do surveys tell us about what Americans want in their presidential candidates?
A Look at the Trends
Gallup’s invaluable trends about willingness to vote for different groups of qualified people provide an important anchor for this discussion. Today, more than 85% would vote for a qualified woman—the same goes for someone who is Black, female, Hispanic, Catholic or Jewish. Of the groups Gallup has inquired about over time and updated recently, only one, a socialist, still faces majority opposition, but it is a bare 51%.
In a 1979 poll, Gallup used a different approach and asked whether a woman, Black person or Jewish person had a chance of being elected by the year 2000. Fewer people thought any of these groups had an excellent or good chance than said they themselves would vote for such a candidate. Only a third thought a woman would have an excellent or good chance by the year 2000, while 37% gave that response about a Black candidate, 40% about a Jewish candidate and 59% about a divorced candidate. To understand possible resistance to certain kinds of candidates and candidate behaviors, pollsters now ask about not only respondents’ personal preference, but also what they think their neighbors or fellow citizens would do. Many Americans give what they believe is the socially correct response when asked about their own views, while their “neighbors’” response may reveal more accurate sentiments.
But what about a candidate’s past or present behavior or morality? Unlike the Gallup battery discussed above, trends don’t exist for most of these kinds of questions. I found only two extensive batteries of questions that allow for comparisons, one from 1979 and the other from 1996.
The 1979 poll is now 45 years old, and we have changed our minds about many things. Back then, seven in 10 Americans, for example, said they would object strongly if a presidential candidate smoked marijuana occasionally. National support for legalizing marijuana hit 70% in a 2023 Gallup poll. A majority of Republicans and people over age 55 supported legalization. That doesn’t mean most Americans would endorse occasional use by a presidential candidate today, especially as only 17% admitted to Gallup in 2023 that they used marijuana. But I wonder if you would find 70% of the population objecting strongly.
Other behaviors were less objectionable in 1979. Thirty-eight percent said they would object strongly if the candidate was not a member of a church, 36% if he used tranquilizers occasionally, 33% if he used profane language in private, 30% if he had seen a psychiatrist, 21% if he wore jeans occasionally in the White House, 17% if the candidate were divorced and—get this—14% if he had a cocktail before dinner. It’s hard to imagine a pollster even asking most of these questions today (but I wish one would).
In 1996, in a poll sponsored by Knight Ridder, Princeton Survey Research Associates asked a question about deal-breakers but gave people a much longer list. More than half said they could not vote for a candidate who had been a member of a white supremacist group (75%), failed to pay his income tax (70%), had friends suspected of being involved in organized crime (66%), used illegal drugs within the past few years (65%), burned a flag during an anti-war demonstration (58%), favored interests of big money donors (55%), had been investigated for corrupt business practices (55%) or practiced homosexuality (51%). People were split on a presidential candidate who had told racist jokes (49% said this would be a deal-breaker). A majority, 58%, said they could overlook a presidential candidate’s past extramarital affairs, his cheating on coursework in college (61%) and, just barely, a candidate who avoided the draft (51%).
An American Enterprise Institute colleague once compared running for president to hiring a plumber. People don’t want to hear about the plumber’s problems with his wife or his kids, he said; they just want the plumbing problem fixed. So, too, with Americans’ responses to many questions about a presidential candidate’s personal life. The personal does not appear to be as political as feminists told us in the 1970s. Many Americans also appear ready to give a candidate a pass for some youthful behaviors. All of us make foolish mistakes, and we appear willing to forgive.
But this doesn’t apply across the board. Americans in 1996 were less likely to give more serious infractions a pass. Thus, most Americans said that being a member of a white supremacist group, failing to pay income taxes, having friends who were close to organized crime and using illegal drugs in the past few years were too serious to be overlooked. Two-thirds said they would not vote for a candidate who lied about what his policies would be if elected.
In 1972, Democratic candidate George McGovern selected Thomas Eagleton as his running mate. Eagleton withdrew his name after reports surfaced that he had been treated for clinical depression. In 1987, Joe Biden ended his presidential campaign after videos leaked by a rival Democratic contender surfaced comparing a speech of his to one by British Labour leader Neil Kinnock. And Bill Clinton—well, Clinton’s large and small missteps—dented but did not diminish his political success. Neither of these issues—a candidate’s mental health, charges of plagiarism—seems relevant this year. But are there other possible deal-breakers Americans might have this year?
Lessons for Today?
In many respects, the 1976 and 1996 questions are simply interesting artifacts from bygone presidential elections. Each campaign is unique. Age didn’t appear to be a significant concern in 1979 and 1980, when pollsters asked about Ronald Reagan’s age (69 in 1979). Nor was it a concern when he was 73 and ran for reelection. But the new Gallup poll asks about willingness to vote for a qualified candidate over the age of 80, and two-thirds said they could not vote for that candidate. This finding should make the Biden campaign more than a little nervous.
In recent weeks, pollsters have begun to explore attitudes about a possible Trump conviction. The questions are hypothetical, but the different approaches the pollsters are using may give us some indication of how people see this issue. Sixty-five percent of voters in the Iowa Republican caucuses said they would consider Donald Trump fit for office even if he were convicted. Fifty-four percent of those who voted in the New Hampshire primary gave this response. But self-declared independents there didn’t agree.
In Gallup’s new poll, two-thirds said they would not vote for a person charged with a felony, and, separately, 70% said that about a person convicted by a jury. The January Harvard/Harris poll found that registered voters would still vote for Donald Trump over Joe Biden if Trump were convicted of a crime related to his handling of classified documents (53% to 47%), and separately, if he were convicted of a crime relating to the Georgia case (51% to 49%). But Biden topped Trump barely (51% to 49%) if Trump were convicted of a crime related to January 6. The new late January-early February NPR/PBS NewsHour/Marist poll finds Biden and Trump closely matched among registered voters. But Biden opens up a six-point lead if Trump is convicted of a crime. Of course, attitudes may change as trials ensue and we get closer to November. Still, these new findings should make the Trump campaign nervous as well.
Polls from the past can’t tell us what will matter most to voters this fall, but they do provide some hints of what resonates and how that has changed over time. In this sense the polls are palimpsests, revealing new attitudes where traces of the old can be seen.
