New data turns our opinion about the partisanship of young Americans on its head
New data turns our opinion about the partisanship of young Americans on its head

A few years ago, I spoke with Columbia University political scientist Donald Green while researching my book on generational change in the United States. My analysis showed in detail how younger Americans differed from older ones, including being more diverse and better educated. These characteristics overlapped with another distinguishing feature: They were more likely to support Democratic candidates, although they were less likely to be members of a political party.

Green provided a caveat to this differentiation.

“If you look at the opinion profile of young people, you’ll be surprised at how ideologically heterodox this otherwise largely liberal group is,” he said. “There are some specific issues on which young people have very distinctive views – for example, LGBT equality – but on other issues they don’t.”

That’s not how politics is generally framed among younger Americans, but there’s growing evidence that it’s a useful way to view it. Rather than aligning themselves with the positions of the Democratic or Republican parties, many young, independent voters approach politics like a Spotify playlist — picking out individual songs rather than listening to an entire album. Or, perhaps more importantly, they consider some aspects important and others less important.

That’s certainly too broad, both in terms of representing younger voters and older partisans. But it’s a useful way to look at Green’s observation – an observation that’s consistent with recent polls showing a shift to the right among younger Americans. An observation that’s also consistent with Donald Trump’s unexpectedly strong support among younger Americans.

Part of the support for Trump can be explained by the way the country has evolved since 2015. One 18-year-old who voted for the first time this year was 6 years old the last time Republicans nominated someone other than Trump for president. A 29-year-old was only 17. The world of Republican politics they grew up in has always been dominated by Donald Trump.

This is different from the general shift that occurs in different ways and to different degrees in different data. In April, for example, the Pew Research Center released data that showed a continuation of the familiar pattern in party identification—from more liberal to more conservative as respondents got older.

Earlier this week, however, Pew released the latest edition of its large, comprehensive National Public Opinion Reference Survey. The NPORS is a benchmark survey that uses a number of different methods to ensure responses, including telephone, online contact and direct mail. And as Lenny Bronner of the Washington Post noted after analyzing the data, a much more complicated interplay between party identification and age emerged.

The size of the sample included in the NPORS allowed Bronner to filter out not only age groups but also often overlay race and gender. The results of this analysis are presented below, with particularly interesting findings identified. They are discussed below. Reading the chart is straightforward: Dark blue indicates respondents who identified as Democrats and light blue those who were independent and more likely to vote Democrat. Red bars indicate the same relationship with the Republican Party. The labels on the left indicate groups by age, race, and gender, among other things.

Discussion of the highlighted segments of the diagram:

A. Note that while the total number of Democrats and Democratic-leaning independents is relatively constant across all age groups, the number of respondents who identify themselves as Democrats increases dramatically as the respondent gets older. Less than a quarter of those under 30 identify as Democrats or Republicans. More than a third of those over 65 identify with one of the two major parties.

It’s not highlighted, but just below the general breakdown by age is the same breakdown among registered voters. There is more partisan identification within this group of respondents, probably in large part because states generally require partisan registration.

But we shouldn’t overlook the most important aspect here: Younger respondents are about evenly split between the two parties, which hasn’t been the case in recent years. Much of the movement is among independents, a group that often views their alignment with one party as a reaction to their dislike of the other party. So perhaps some of the movement here is due to independents changing their perceptions about which party they oppose, or to what extent they oppose it.

b. Also notable is the lower level of partisan identification among younger black and Hispanic respondents. (The pool of Asian Americans was too small to break down separately.) They are more likely to identify as Republicans or Republican-leaning independents than older members of these ethnic groups, but much less likely to identify as Democrats.

C The age group most likely to identify as Democrats is older women. The pattern of younger respondents being less partisan holds when breaking down age groups by gender: Men are less Democratic overall than women. In fact, men in every age group are equally or less likely to identify as Democrats than the youngest group of women, who are least likely to identify as Democrats.

Across all age groups, men are more likely to identify as Republicans than Democrats. Across all age groups, the opposite is true for women.

D. The group most likely to identify as Republicans or Republican-leaning independents are young white men. They are more likely to do so—and less likely to identify as Democrats—than the oldest group of white men.

It is important to note that the sample size here is small and the margin of error is therefore relatively large, but this is still a break from previous surveys of party political identification by age.

Most Democratic-leaning women are non-white women. White women of all ages are more likely to identify as Republicans than Democrats. A majority of white women under 30 in this survey identify as Republicans or independent Republicans. Again, however, the sample is small.

Regardless, these data support the notion that the familiar pattern of young Americans leaning strongly to the left has been broken, at least in recent years. If so, Green has likely identified a reason why: The extent of her liberal political identity may have been overstated. It’s also likely that younger voters – perhaps in response to the candidates and issues presented by the Democratic Party – have turned their backs on the party they once leaned toward.

Lenny Bronner contributed to this report.