March 20, 2017

A wider partisan and ideological gap between younger, older generations

The generation gap in American politics is dividing two younger age groups, Millennials and Generation X, from the two older groups, Baby Boomers and the Silent Generation.

In 2016, as in recent years, Millennials and Gen Xers were the most Democratic generations. And both groups had relatively large – and growing – shares of liberal Democrats: 27% of Millennials and 21% of Gen Xers identified as liberal Democrats or Democratic-leaning independents.

By contrast, Boomers and Silents were the most Republican groups – largely because of the higher shares of conservative Republicans in these generations. Nearly a third of Boomers (31%) and 36% of Silents described themselves as conservative Republicans or Republican leaners, which also is higher than in the past. 

The differences in partisan identification across generations are most apparent in the shares of liberal Democrats and conservative Republicans in each. All four groups have comparable numbers when it comes to groups in the middle: conservative and moderate Democrats and moderate and liberal Republicans.

The public’s overall partisan and ideological balance changes little from year to year. But there have been some long-term shifts among the public and within generational groups, according to a new analysis based on more than 15,000 interviews conducted in 2016 as well as earlier survey data. (For a look at trends among registered voters, see “The Parties on the Eve of the 2016 Election: Two Coalitions, Moving Further Apart.”)

The share of liberal Democrats in the public has grown. In 2016, 21% of Americans identified as Democrats or Democratic leaners and also as liberal. While that is only somewhat higher than a year earlier (19%), it is the highest share dating back to 2000. At that time, just 12% of the public described themselves as both liberal and as Democrats or Democratic leaners.

There has been little change since 2000 in other partisan and ideological categories, although the share saying they do not lean to either party is smaller today (11%) than in the early 2000s (17% in 2000).

Millennials were most likely to identify as liberal Democrats. In 2016, a majority of Millennials (55%) identified as Democrats or leaned Democratic; 27% described themselves as liberal Democrats – the highest share of any generation. For much of the past decade, conservative and moderate Democrats outnumbered liberal Democrats among Millennials. But in the past few years, there were about as many liberal Democrats as conservative and moderate Democrats among this youngest adult age group (those who were 18 to 35 in 2016).

The share of liberal Democrats in Gen X ticked up. As in recent years, conservative and moderate Democrats made up the largest partisan and ideological group among Gen Xers (27%), who were ages 36 to 51 last year. However, the share of Gen Xers who identified as liberal Democrats (21%) stood at its highest point since 2000. About as many Gen Xers (23%) described themselves as conservative Republicans, a figure that has not changed much in recent years.

Boomers have turned more conservative. In both 2015 and 2016, about three-in-ten Boomers (30% in 2015, 31% in 2016) identified as conservative Republicans – the highest percentages dating back to 2000. In both years, conservative Republicans made up the largest single partisan and ideological group among Boomers.

Members of the Silent Generation continued to be most likely to identify as conservative Republicans. The GOP has made large gains among Silents in recent years. Eight years ago, Democrats and Democratic leaners outnumbered Republicans and GOP leaners among Silents by 48% to 40%; in 2016, 49% of Silents identified as Republicans or leaned Republican, while 43% leaned Democratic. The share of Silents who describe themselves as conservative Republicans increased by 9 percentage points over this period, from 27% to 36%.

Topics: Baby Boomers, Generations and Age, Political Party Affiliation, Older Adults, Millennials, Political Polarization

  1. Photo of Shiva Maniam

    is a research assistant focusing on U.S. politics and policy at Pew Research Center.

  2. Photo of Samantha Smith

    is a research assistant focusing on U.S. politics and policy at Pew Research Center.


  1. Anonymous5 months ago

    What’s the margin of error? How was the survey conducted? If by phone, did you call cell phones?

    Does this match demographics, women, minorities, Hispanics?

    It annoys beyond belief when news stories, let along the survey’s own release, does not include pertinent data.

    1. David Kent5 months ago

      Thanks for your interest. This is based on multiple surveys and interviews, with sample sizes and margins of error differing for each year’s full sample. See the methodology section of the main report:…

  2. Doug Pearson5 months ago

    I’m a long-time fan of Pew Research. Keep up the good work. I’ve been a registered Republican since I was 21 and I voted for Eisenhower that year ( 1956). A year or three ago, I took your liberal-conservative test and learned that I am very liberal.

    I, too, think people tend to become more conservative as they age, but clearly there are exceptions!

    Your charts showing how the liberal-conservative split varies for each of the four groups since 2000, implies that, yes, there does seem to be a tendency to become more conservative as we age, but I don’t think such a short period tells a real story. Do you have comparable data since, say, 1980 or 1950?

    1. John Franson5 months ago

      I don’t think it implies that there’s a tendency to become more conservative as we age. People’s partisanship and ideology can be influenced by events or conditions. I’d interpret the chart to show fluctuations in partisanship/ideology in response to things like 9/11 and perceptions of different presidential administrations. Also, there is research showing that people’s partisanship/ideology get set in their political formative years, and basically stay that way their whole lives, barring political realignments.

  3. Anonymous5 months ago

    1 question and a statement:
    1. What was the definition or defining requirements to be deemed (liberal/conservative..middle)? Curious about the criteria.
    I would gather that over the years that someone ages they may get more conservative vs liberal. This is a generalization with no data behind this but would be interesting to see if this was false or true and then normalize with this data?

    1. Anonymous5 months ago

      Pretty sure its “partisan identification”: meaning that individuals are asked where they fall on the political spectrum and then they self-identify. So in theory, two people could have different definitions of what it means to be “liberal” or “conservative.” But for each respondent, we know that they are defining themselves in regards to what these categories mean “to them.”

      As for your comment, look for a “longitudinal study” on political orientation. In order to test your generalization, you would have to follow the same people over time. A longitudinal study would do that. I’m sure someone has, but I can’t think of it off the top of my head. Keep questioning!

  4. Kendall Scudder5 months ago

    Good for the Millennials. Being a liberal democratic Baby Boomer, it’s good to see strong growth in that area. That provides hope for the future despite the insanity of the current government.