I have some questions on younger voters in recent election polls
Recent polls show Harris polling well behind Biden's 2020 performance among young voters. A few thoughts on what could be going on.
In the 2020 presidential election, Joe Biden won 18 to 29-year-old voters by 24 points and 30 to 49-year-old voters by 12. With those numbers in mind … I’ve been a little perplexed about the latest crop of national polls that have come across my inbox. Several polls show Harris with a much smaller margin of victory over Trump among younger voters.
Not all pollsters publish the demographic results of their surveys. But some do. Below is a sampling of some of the latest polling that showed close young voter results:
Morning Consult shows Harris up +8 among likely voters ages 18 to 34 (52% vs 44%)
USA Today/Suffolk has Harris leading Trump by +5 among likely voters 18 to 34 in their two-way matchup (48% vs 43%).
Fox News has Harris leading Trump by +7 among voters 18 to 34 (53% vs 46%).
A few things could be going on here…
First, we could be seeing a dramatic shift in political ideology among younger voters. Young voters could be growing more conservative. There is some evidence of a rightward shift. Consider some results from the Harvard Kenny survey of adults 18 to 29, comparing April 2020 and April 2024:
Agreement that recent immigration into this country has done more good than harm has declined from 36% to 29%, a 7-point decline.
Agreement that Government spending is an effective way to increase economic growth has declined from 34% to 28%, a 6-point decline.
Agreement that the Government should do more to curb climate change, even at the expense of economic growth, has declined from 57% to 49%, an 8-point decline.
At the same time, plenty of data makes me hesitant to declare a sizeable rightward shift among young voters. A few data points that underscore my reluctance …
Gallup found that women young women have become increasingly liberal, while young men have been mostly stable in their political ideology … with perhaps a slight uptick in conservative identification. Most polls show that the liberalization of young women is more pronounced than the conservatization of young men.
In the 2022 midterm elections, a Pew analysis found that 68% of voters aged 18-29 supported Democratic candidates, while 31% supported Republican candidates. This margin is similar to past elections.
The same Harvard survey that shows a slight rightward shift on immigration, government spending, and government action on climate change showed Biden beating Trump by 33 points among young voters. Their most recent survey has Harris beating Trump by 32 points.
If there were a substantial shift in the political allegiances of young voters, you would expect to see a larger movement in ideological beliefs. Going from about a +18 Biden margin of victory among 18 to 34-year-old voters to +5 or +8 for Harris would be such a dramatic shift that it would anticipate it showing in polling on other attitudes. That shift does not appear to be there.
Second, fewer left-leaning 18 to 34-year-olds plan to vote, leaving a more right-leaning voting population among young voters. There is some evidence to support this. The October NBC News survey found 49% of 18 to 34-year-olds say they are a “9 out of 10” or a “10 out of 10” in their interest in voting this year … a 14-point decline from the 63% who were at those levels in 2020. Voter registration for 18 to 29-year-olds across many states is below voter registration levels in 2020.
On the other hand, the latest NY/Times Sienna poll shows 54% of likely young voters saying they are “almost certain” they will vote. Compare that to the 29% who gave that response in October of 2020. It could very well be that even if enthusiasm is low … voter participation will still be high.
In addition, what evidence we have shows the enthusiasm gap among young voters favors Harris, not Trump. The September 2024 Harvard Youth poll found that 74% of young Democrats say they will "definitely" vote, compared to 60% of young Republicans. Pew found that 50% of young Harris supporters are “extremely motivated” to vote, compared to 34% of young Trump supporters.
Third, young voters may be turned off by Harris. NY/Times Seinna shows 18 to 29-year-olds with a 59% favorable rating of Harris. The same poll in October 2020 gave Biden a 63% favorable rating. Biden seems to have been slightly more favorably perceived in 2020 than Harris in 2024 among young voters. But even still, the NY Times/Sienna poll shows Harris beating Trump among young voters by 19 points.
Harris doesn’t appear to generate the same enthusiasm as Biden did in 2020. But it’s hard to see a 4-point difference in favorability between Biden in 2020 and Harris in 2024 leading to a double-digit drop in support for Harris.
Fourth, there are problems with sampling younger voters in these election polls. I think there is strong support for this. Response rates to election polling continue to plummet. Fewer people are taking surveys. The likelihood that the respondents who take surveys are meaningfully different from everyone else is high … meaning that the polls are not adequately representing young voters in their data.
Young adults are more digitally literate and better able to screen requests from pollsters to take surveys. Most of the research on this relates to young people being better able to avoid online scams, phishing, and spam … but it’s not too hard of a leap to assume those strategies extend to avoiding survey requests. Given the low response rates, it’s safe to assume people are trying to avoid taking surveys. They are well-equipped to do it.
What if there is an issue with sampling young voters on top of lower participation or a slight rightward shift among young voters in 2024? Let’s assume Harris wins 18 to 34-year-olds by 14 points instead of the 18 points Biden got in 2020. How would rebalancing the surveys to a more realistic youth vote affect the results?
The Morning Consult poll, with Harris leading Trump by +4 points among all likely voters, would have Harris up about +6 points over Trump.
The USA Today Suffolk poll, with Harris leading Trump by 1 point among all likely voters, would have Harris up about +3 points.
The Fox News poll, which has Trump leading Harris by 2 points among all likely voters, would have Harris and Trump tied.
It’s not a dramatic result … but it is enough to move the results into a clear lead for Harris in the Morning Consult poll, on the cusp of Harris having a national buffer in the USA Today Suffolk poll, and tied instead of losing in the Fox News poll.
We’ll know soon enough if there’s any validity to my concerns over the sampling of youth voters. But it’s enough for my skepticism of recent election polls to continue.