Heading into the next presidential election, an analysis of CNN polls shows that Republicans have reverted to the deeply negative national outlook they held prior to Donald Trump’s presidential victory in 2016. They again are convinced the nation is in decline, and more often defensive against demographic and cultural changes in US society.
In a poll conducted late in the summer of 2016, following Trump’s nomination, roughly half of Republicans and Republican-leaning independents (49%) said America’s best days lay behind us. And while most said they considered the country’s increasing diversity enriching, 37% said they felt the increasing number of people of many different races, ethnic groups and nationalities in the US was, instead, threatening American culture.
Three years later, during Trump’s presidency, only 18% of the party said the nation was past its peak days, with a similar 20% viewing diversity as a cultural threat.
Since then, the GOP has reversed course, becoming less pluralistic and even more pessimistic. In CNN’s latest polling, released this week, the share of Republican-aligned adults who said the country’s best days are over had skyrocketed to 70%, while the percentage saying that America’s culture was threatened by increasing racial and ethnic diversity rebounded to 38%. In a question not asked in 2019, a broad 78% majority of Republican-aligned Americans also say that society’s values on sexual orientation and gender identity are changing for the worse.
The party’s shift in perspective over the past four years took place across demographic lines. Between 2019 and 2023, the belief that the country’s best days are behind it rose by more than 40 percentage points across age, educational and gender lines. Additionally, the share considering diversity a threat rose by double digits in each group. That may suggest that the results sometimes represent not deep-seated beliefs so much as a reaction to the current political environment, including which party holds the presidency.
But the survey also finds Republicans and Republican-leaners are far from wholly unified in their views, with a constellation of interrelated political, demographic and socioeconomic factors dividing views.
One of the most persistent gaps appears along educational lines, with Republican-aligned college graduates less likely than those without degrees to favor a more active government, say the country’s best days have passed or to consider the country’s increased diversity threatening – though both groups share similarly negative views about changing values around gender and sexuality.
Age also plays a role, as do gender and race: Those younger than 45 are less likely than older adults to call racial diversity a threat or to say values on gender identity and sexual orientation are changing for the worse, with a similar divide between GOP women and men, and between White people and people of color aligned with the party.
Differences within the GOP are often magnified when demographics intersect. Roughly half (51%) of Republican-aligned adults ages 45 or older who don’t have a college degree say they consider the country’s increased diversity threatening, an opinion shared by a third or fewer within any other combination of age and education. And, within the GOP, 54% majority of male, White evangelical Christians find such diversity threatening, a view not shared by most of their female counterparts, or by majorities of those of other combinations of racial and religious backgrounds.
Republicans’ unease with the way that the US is changing ties into opinions of Trump’s legacy. In the latest poll, a 57% majority of Republican-aligned adults who call racial diversity threatening also say it’s essential that the next GOP nominee would restore the policies of the Trump administration. So do nearly half of those who say values on gender and sexuality are changing for the worse (49%) or who feel the nation’s best days have come and gone (46%) – in each case, a significantly higher figure than among those who don’t share those views. Belief that Trump has had a good effect on the Republican Party, meanwhile, is 14 percentage points higher among those who say the US has peaked than among those who say its best days lie ahead.
What’s less clear is whether those outlooks will drive support for Trump and his campaign, particularly with presumptive rivals like Florida Gov. Ron DeSantis also building messages around similar themes. At this early stage in the campaign, Republicans and Republican leaning-independents who say the US’s best days have passed are about equally likely to say they’d be enthusiastic about the possibility of a DeSantis nomination as they are to say the same of Trump. Relatively few currently express similar excitement about former UN ambassador Nikki Haley, former Vice President Mike Pence, or former Secretary of State Mike Pompeo.
It’s also too early to tell what next year’s GOP primary electorate will look like. That’s a key factor, given the likely demographic divides both in whom Republicans support and in how likely they are to vote at all. Older and more highly educated voters are more likely to turn out. Exit polling suggests that in past cycles, older and more highly educated voters tended to turn out disproportionately. This far from the start of voting, it’s hard to tell who’s likely to show up, but both demographics and political preference could play a role in determining initial levels of enthusiasm heading into the election season. In the latest CNN poll, Republicans and Republican-leaners over age 45 who supported Trump were far more likely to report extreme enthusiasm about participating in next year’s primaries than those over 45 with a different candidate preference, or younger Republicans and Republican-leaners regardless of the candidate they back.
The CNN poll was conducted by SSRS from March 8-12 among a random national sample of 1,045 self-identified Republicans and Republican-leaning independents drawn from a probability-based panel. Surveys were either conducted online or by telephone with a live interviewer. Results among the full sample have a margin of sampling error of plus or minus 3.8 points, it is larger for subgroups.
This story has been updated with additional information.