The youngest voters are more conservative

A deeper looking at the growing ideological divide within Gen Z

There’s been a fair bit of discussion about our most recent poll—specifically, the generic ballot question, which found that 18-21-year-olds said they favored the Republican candidate by a 12-point margin in the 2026 midterms, compared to 22-29-year-olds favoring Democrats by a 6-point margin.

We decided to launch this Substack so that we have a place to dive deeper into the data and contextualize our findings. Today, we’re looking at the growing divide within Gen Z.

Our data indicates that the conservative shift among the youngest voters can’t simply be chalked up to a high number of undecided midterm voters or a lack of enthusiasm for Democrats. Across the board, voters aged 18-21 have substantially more right-wing views and policy preferences.

Start with President Trump’s net favorability rating. Among voters overall, Trump’s net favorability was -6, according to our data.¹ But among young voters, Trump is only underwater with women aged 22-29 (-10). Men aged 22-29 are evenly split, while both men 18-21 (+5) and women 18-21 (+1) see the president more favorably than not.

Views on Trump are the least hardened among the youngest men, where less than half feel either strongly favorable or strongly unfavorable towards the president. Among women aged 22-29, over 80% of those who dislike the president feel strongly unfavorable towards him. The Republican Party brand seems to be more popular with the youngest women than Trump himself: women aged 18-21 were R+4 in our 2026 generic ballot question.

The youngest voters have more conservative policy preferences than voters aged 22-29 when it comes to specific issues.

When asked whether legal immigration levels should be increased, decreased, or kept at the same level, 20% of voters overall say decreased, 30% say increased, and 43% say kept at the current level.

• Among voters 18-21, 21% said decreased, 37% said increased, and 31% said left at the current level.

• Among voters 22-29, 11% said decreased, 37% said increased, and 43% said left at the current level.

When asked whether transgender women should be allowed to compete in college sports, 23% of voters overall say they should be allowed, and 65% say they should not be allowed.

• Among voters 18-21, 22% said they should be allowed to compete, and 67% said they should not be allowed.

• Among voters 22-29, 33% said they should be allowed to compete and 57% said they should not be allowed.

When presented with a pro-Ukraine aid and an anti-Ukraine aid message and asked which came closer to their views, 50% of voters overall favored ending American aid to Ukraine, while 40% favored continuing to provide aid.

• Among voters 18-21, 54% preferred the anti-aid message, while 34% preferred the pro-aid message.

• Among voters 22-29, 46% preferred the anti-aid message, while 42% preferred the pro-aid message.

Of course, our poll is only one data point. It should be taken with a grain of salt. But as Zachary has pointed out on X, this matches what you see when you look at precinct-level election results around college campuses. It also lines up with some of the data that David Shor shared in his recent interview with Ezra Klein, from his firm’s polling of 8 million voters.

That is to say, the best evidence we have indicates that the youngest voters are substantially more conservative than the electorate as a whole. Policymakers, elected officials, and journalists should adjust their priors accordingly.

1. Note that this matches pretty closely to Nate Silver’s polling average of Trump’s job approval, which was -5 with all voters at the time of writing.