This weekend marks perhaps the most contentious battle over American calendars: Columbus Day. While people who get a day off this Monday are unlikely to complain about it, there are significant divides among Americans about how we remember Christopher Columbus and whether he should be honored with a holiday. A YouGov survey last week asked about views of Columbus, his accomplishments, and his holiday.

How do we view Christopher Columbus as a person?

About half of Americans have a somewhat or very favorable opinion of Christopher Columbus — 52% view him favorably and 32% view him unfavorably. Columbus is more popular among older Americans and Republicans. Americans 65 and older overwhelmingly view Columbus favorably — 68% are favorable toward him and 21% are unfavorable. Republicans are even more likely to have a positive opinion of Columbus — 68% are favorable and 19% are unfavorable.

Younger adults and Democrats are more likely to view Columbus unfavorably. Adults under 45 are almost evenly divided on Columbus: 40% view him favorably and 39% view him unfavorably. And Columbus is viewed unfavorably by more Democrats than view him favorably (48% vs. 41%).

Americans’ views of Columbus also differ based on race. A majority of white Americans view Columbus favorably: 57%, compared to 28% who view him unfavorably. Americans of other racial backgrounds don’t view him quite so positively. Black Americans are about evenly divided on their attitudes about Columbus – 42% view him favorably and 39% unfavorably – while Hispanic Americans view him a little more positively: 45% favorable vs. 34% unfavorable. However, the group with the most negative views are Americans of other racial backgrounds, which includes Asian Americans, Native Americans, and people who identify with more than one racial group, among others. A majority of Americans of other racial backgrounds (56%) view Columbus unfavorably, while only 31% view him favorably.

Because Columbus hailed from Genoa in Italy, many Columbus Day observances include celebrations of Italian heritage. To see if this has any effect on Italian Americans’ attitudes towards Columbus himself, we asked respondents if they have any ancestors from Italy. Italian Americans are only slightly more likely to view Columbus favorably than are non-Italian Americans (54% vs. 51%). However, Italian Americans are also about as likely to view Columbus negatively — 33%, compared to 32% of non-Italian Americans. All these numbers are within the margin of error for this survey, meaning we cannot conclude that Italian ancestry has any significant impact on views about Columbus.

After asking about respondents’ personal view of Columbus, we asked whether he was a good or a bad person. About one-third (32%) of Americans say that Columbus was equally good and bad and 16% are not sure. Very few Americans take the absolutist stances that he was either completely good (5%) or completely bad (5%), and Americans are more likely to see him as mostly good (30%) than mostly bad (13%).

Attitudes on Columbus’ moral character are split similarly to favorability. Older Americans are much more likely than younger adults to view him as mostly or completely good (40% of those 45 and older vs. 27% of younger adults. White Americans are more likely to view him as good (40%) than are Black Americans (25%), Hispanic Americans (27%), or Americans of other racial backgrounds (17%). Republicans are much more likely to think of Columbus as a good person (52%) than are Democrats (23%) or Independents (29%).

Attitudes are similar about whether Columbus was a hero or a villain. The most common response is that Columbus was neither a hero nor a villain (36%), but Americans are more likely to view him as a hero (35%) than as a villain (20%).

Opinions vary by age, race, and political party in similar ways to other positive and negative attitudes about Columbus. Viewing Columbus as more of a hero is more common among Americans 65 and older (44%), White Americans (37%), and Republicans (54%). On the other hand, Columbus is more likely to be viewed as more of a villain than as more of a hero by Americans between 30 and 44 (30% more of a villain vs. 24% more of a hero), Americans of other racial backgrounds (33% vs. 24%) and Democrats (33% vs. 25%). Black Americans are roughly evenly split between those who view Columbus as more of a hero (31%) and those who view him as more of a villain (29%), while Hispanic Americans have about the same likelihood of seeing Columbus as more of a hero (31%), but were less likely to see him as more of a villain (23%).

What was Columbus’ impact?

Americans’ views of Columbus’ historical importance are more unified than their views of Columbus as a person. A large majority of Americans (69%) think Columbus is important or very important to the history of the United States, while only 22% view him as not very or not at all important. Columbus is more likely to be viewed as important than unimportant among all major demographic groups, showing that some Americans who don’t like Columbus believe he had an important role in history.

As with questions about personal regard for Columbus, he is most likely to be viewed as important by older Americans (74% of those 65 and older), white Americans (71%), and Republicans (84%). Democrats (29%) and Americans of other racial backgrounds (33%) were more likely than other groups to say that Columbus is not very or not at all important to U.S. history, although this perspective is still held by only a minority of members of these groups.

Two in five Americans (41%) say Columbus’ impact on the U.S. was mostly or completely positive, although 30% say his impact was both positive and negative and 15% are not sure (15%). Americans who see Columbus’ impact as mostly or completely negative are in a small minority (14%).

Among most demographic groups Columbus’ impact on the U.S. is more likely to be seen as positive than as negative. Only two groups are more likely to say Columbus’ impact was both positive and negative than that it was mostly or completely positive: Democrats (33% vs. 31%) and Americans of other racial backgrounds (31% vs. 28%). And only 22% of Democrats and 24% of Americans of other racial backgrounds say his impact was mostly or completely negative. Columbus’ impact, like the man himself, is viewed positively by more older Americans (48% of those 65 and older), white Americans (43%), and Republicans (57%).

Here, opinion diverges between Italian and non-Italian Americans. Italian Americans are more likely to view Columbus’ impact on the U.S. as mostly or completely positive (49%, compared to 41% of non-Italian Americans).

Do Americans think they’ve gotten the truth on Columbus?

Most Americans think the things they learned about Columbus in school are somewhere in between completely true and completely false. Americans’ most common viewpoint (30%) is that their education on Columbus was mostly true, and about as many say it was about equally true and false (29%). Very few Americans (8%) believe that what they learned was completely true, but an even smaller share (5%) believe it was completely false. Most Americans are skeptical about what they were taught, but a majority also do not entirely reject it.

Skepticism is highest among Americans between 30 and 44: 32% think what they learned was mostly or completely false and 24% think what they learned was mostly or completely true. All other age groups are more likely to think their education about Columbus was mostly or completely true than to think it was mostly or completely false.

Skepticism also varies by race and party identification. White and Hispanic Americans are more likely to trust what they learned about Columbus in school than Black Americans and Americans of other racial backgrounds: 40% of white Americans and 38% of Hispanic Americans believe what they learned is mostly or completely true, compared to 32% of Black Americans and 27% of Americans of other backgrounds. Democrats are more likely to believe that what they learned about Columbus was mostly or completely false — 29% do, the same as the 29% who believe what they learned was mostly or completely true. Conversely, Republicans are more likely to trust their education — 52% think what they learned was mostly or completely true, while just 14% believe what they learned was mostly or completely false.

Americans are not significantly more confident about the truth of what children in school learn today about Columbus. While many Americans understandably say they are not sure what children learn about Columbus today (28%), just 4% say they believe what children learn today is completely true. The most common opinion is that schools today provide information about Columbus that is “about equally true and false” (23%).

When it comes to age and party identification, the more likely members of a group are to believe what they learned about Columbus as children, the more likely they are to doubt today’s approach to Columbus education. Americans 65 and older are the least likely to believe that what children learn today is mostly or completely true (18%), while adults under 30 are the most likely (33%). This may reflect that what children learn today is most similar to what younger adults learned as children. Younger adults also are more likely to be parents of children under 18, who are more likely to say what is being taught today is completely or mostly true (31% vs. 22%) and slightly less likely to say it is completely or mostly false (17% vs. 29%).

Republicans are more likely to say that what children learn today is completely or mostly false (22%, compared to 18% of Independents and 14% of Democrats). Conversely, Democrats are more likely to believe that what children learn today is mostly or completely true (28%, compared to 19% of Independents and 25% of Republicans).

White and Hispanic Americans are only slightly more likely than Black Americans and Americans of other racial backgrounds to believe that what children learn today is mostly or completely true and only slightly less likely to believe it is mostly or completely false. The racial gap on this question is much smaller than the one for beliefs about the truthfulness of Americans’ own education on Columbus, suggesting that Black Americans and Americans of other backgrounds may be more likely than White and Hispanic Americans to believe that education about Columbus has become more accurate over time.

One reflection of Americans’ skepticism about how Columbus’ history is taught is doubts about whether Columbus was the first non-Indigenous person to reach the Americas. Even though many Americans were taught this in school, a majority of Americans (56%) believe Columbus was not the first explorer in the Americas. Only 25% of Americans believe that Columbus was the first non-Indigenous person to reach the Americas, while the remaining 19% say they’re not sure.

Groups of Americans who view Columbus more positively are not more likely to believe that Columbus was the first to reach the Americas. Even though older Americans are more likely than younger adults to view Columbus favorably, older Americans are less likely to think Columbus was the first to reach the Americas. Only 18% of Americans 65 and over believe this, while 61% believe Columbus was beaten to the Americas. In contrast, 33% of adults under 30 think Columbus was first, and only 43% think he was beaten. Meanwhile, 34% of Black Americans believe Columbus was first to the Americas, compared to 24% of white Americans, 27% of Hispanic Americans, and 17% of Americans of other backgrounds — even though Black Americans are likelier than White and Hispanic Americans to hold negative views of Columbus. Some Americans may associate being first in the Americas with negative effects rather than positive ones. However, both Black Americans and younger adults are more likely to say they are not sure whether Columbus was the first non-Indigenous person to reach the Americas.

On this matter, Independents differ more from Democrats and Republicans than Democrats and Republicans do from each other. Democrats are slightly more likely than Republicans to doubt the narrative about Columbus’ arrival in the Americas (55% vs. 52%), but Independents are more likely to believe Columbus was not the first explorer in the Americas (60%). Americans who place themselves outside the two-party system are likelier to question the traditional narrative about Columbus.

Nearly half of Americans (45%) believe that Norse people or Vikings reached the Americas earlier than Columbus did. (To learn more about Americans’ views on Norse exploration of the Americas, we also surveyed Americans about Leif Erikson.)

There is robust scientific evidence for Norse settlement in eastern Canada around the year 1000. There also is scientific evidence to support contact between people in Siberia and Alaska before 1492, something 14% of Americans believe happened. Finally, there is emerging evidence of possible travel between Polynesia and the west coast of South America around the year 1200, something 10% of Americans believe happened.

However, many Americans also report believing theories of pre-Columbian exploration of the Americas with less scientific evidence. Small, but not insignificant, shares of Americans believe that the Americas were visited before 1492 by other Spanish or Portuguese explorers (12%), Celtic people (9%), Chinese people (7%), people from sub-Saharan Africa (7%), Christian missionaries (6%), ancient Egyptians (6%), or people from the Roman Empire (5%). And 20% of Americans believe some other group visited the Americas before Columbus. While many of these ideas lack scientific evidence, and some form parts of conspiracy theories, the widespread belief that Columbus was not the first explorer to reach the Americas may reflect Americans’ general skepticism about what they were taught about Columbus.

Americans also have doubts about another supposed fact many were taught in school: that Columbus proved the world is round. In reality, it was well-known that the Earth was round long before Columbus: ancient Greek mathematicians had even calculated the circumference of the Earth. A majority (56%) of Americans correctly label this claim as false, while 23% believe it and 21% are not sure.

Older Americans and Republicans are more likely to believe Columbus proved the Earth was round: 31% of Americans 65 and older and 30% of Republicans say this is true.

Should we celebrate Columbus?

Overall, Americans are more likely to say that statues of Christopher Columbus should be displayed (50%) than that they should not be displayed (22%). However, many Americans also say they don’t know whether statues should be displayed or not (27%).

Attitudes on this question are quite divided by age and race. Majorities of Americans 45 and older (60%) and white Americans (56%) are in favor of displaying statues of Columbus. In contrast, adults under 45 (39%), and Black Americans (40%), Hispanic Americans (41%), and Americans of other racial backgrounds (33%) are much less likely to favor displaying Columbus statues.
The largest demographic divide is between political parties. While a majority of Republicans (72%) favor displaying Columbus statues, Democrats are more likely to favor not displaying them than to favor displaying them (36% vs. 33%).

Italian ancestry also has a large impact on attitudes about Columbus statues. A majority (63%) of Italian Americans favor displaying Columbus statues, while only about half (49%) of non-Italian Americans say the same.

Americans are also more likely to strongly or somewhat approve of celebrating Columbus Day (43%) than to disapprove (23%). However, the sizable shares of Americans who neither approve nor disapprove (28%) or are not sure (5%) prevents either side from taking a majority.

While approval of Columbus Day is more common than disapproval among Americans, among Democrats disapproval is more common. Democrats are more likely to disapprove than approve of celebrating Columbus Day (31% vs. 36%). Among Republicans, 60% approve and 11% disapprove. Adults under 30 are only narrowly more likely to approve of celebrating Columbus Day than to disapprove (33% vs. 30%), while Americans 65 and over are much more likely to approve than to disapprove (51% vs. 13%).

While we did not find significant evidence that Italian Americans view Columbus more favorably than non-Italian Americans do, our data does show that Columbus Day is more popular among Italian Americans. About half (52%) of Italian Americans approve of celebrating Columbus Day, compared to 43% of non-Italian Americans. While only 24% of non-Italian Americans disapprove of celebrating Columbus Day (24%), this share is even lower among Italian Americans (20%). Given that views on Columbus himself do not significantly differ between the two groups, this is likely evidence that Italian Americans’ greater approval is driven more by the aspects of Columbus Day that celebrate Italian heritage, and less by the celebration of Columbus as a person.

We asked respondents if they think the federal government should recognize Columbus Day, offering four options: “yes, a federal holiday for which government workers get a day off (similar to Labor Day or Martin Luther King Day),” “yes, a national observance for which government workers do not get a day off (similar to Mother’s Day or Flag Day),” “no,” and “not sure.” We found that Americans are almost evenly divided between these options: 28% support a federal holiday, 21% favor a national observance, 31% believe Columbus Day should not be celebrated, and 20% are not sure. This means that Americans are more likely to favor some recognition of Columbus Day (49%) than to prefer none (31%). However, the relatively even split between multiple options means that if Columbus Day were put to a direct vote of American citizens, the results would be quite unpredictable. If the election didn’t use a system such as ranked-choice voting, those who favor downgrading Columbus Day to a national observance may be a spoiler, perhaps swinging the vote like how voters for Ralph Nader did in 2000.

Demographic differences on this question are similar to the splits on Columbus Day approval. Older Americans and Republicans are more likely to believe Columbus Day should be recognized, while younger Americans and Democrats are more likely to favor removing it from the calendar. However, even among each of these groups there is more support for recognizing Columbus Day in some way than there is for ending recognition of Columbus altogether. White, Black, and Hispanic Americans are also more likely to support recognizing Columbus Day than removing it, but this is not true of Americans of other racial backgrounds, 50% of whom favor no federal recognition for Columbus Day.

Consistent with the strong approval of Columbus Day within the Italian American community, we find that a majority of Italian Americans (64%) believe Columbus Day should be recognized by the federal government. This is much higher than the 48% of non-Italian Americans who believe Columbus deserves federal recognition. Only 25% of Italian Americans believe Columbus Day should not be recognized at all, compared to 31% of those without Italian ancestry. However, Italian Americans are split on how Columbus Day should be recognized much like non-Italian Americans are. While 35% of Italian Americans believe Columbus Day should remain a federal holiday, 29% believe it should only be a national observance.

A day does not need federal recognition to be important. Holidays such as Halloween and Valentine’s Day are very popular despite receiving no official federal recognition. And even if Columbus Day loses federal recognition, it could remain a celebration of Italian heritage, similar to how St. Patrick’s Day celebrates Irish heritage despite receiving no federal recognition.

If the federal government ever does decide to remove recognition of Columbus Day, two other early October holidays could take its place. Our survey found that Indigenous People’s Day would make a popular replacement, while considerably fewer people are clamoring for Leif Erikson Day to be a federal holiday.

Related:

See the results for this YouGov poll

Methodology: This YouGov poll was conducted online on October 4 – 6, 2024 among 1,137 U.S. adult citizens. Respondents were selected from YouGov’s opt-in panel to be representative of adult U.S. citizens. The sample was weighted according to gender, age, race, education, 2020 election turnout and presidential vote, baseline party identification, and current voter registration status. Demographic weighting targets come from the 2019 American Community Survey. Baseline party identification is the respondent’s most recent answer given prior to November 1, 2022, and is weighted to the estimated distribution at that time (33% Democratic, 31% Republican). The margin of error for the overall sample is approximately 4%.

Image: Getty