21 Comments
User's avatar
Cayce Jones's avatar

The Pew numbers on gender have 53% to 55% of women voting for the Democratic candidate in the last three presidential elections. Then there are the recent polls on Trump's immigration actions, with Quinnipiac's listing men as evenly divided, while 65% of women oppose.

This division is pretty much the same across all issues polled. It appears that cultural identity is more important than policy, and that is especially the case for the least informed, least likely voters.

Expand full comment
Deborah La Torre's avatar

I’m wondering how much the numbers for women have to do with higher percentages of women getting college degrees.

Expand full comment
ANTHONY MAHERN's avatar

On the matter of having a pod cast, I say skip it. Voracious consumers of political data, like myself, typically READ a pod cast transcript if available. Why? Because you can read way faster than a pod cast audio takes.

Expand full comment
Martha Ture's avatar

I want to see data and analysis on the effect of rightwing propaganda media - Fox, e.g., et al - on voter turnout and on voter move rightward. I think this is important. If you live in an information world that believes X and other people live in an information world that believes Not-X, how much does X world penetrate Not-X world? How much does Not-X world penetrate X world? What's the impact on voting? Further, the companies that fund X world and Not-X world with ad buys - have their C-level execs been polled or quoted on their awareness of their effects on elections?

Expand full comment
Christie Manussier's avatar

This would be a fantastic rebbit hole to understand. My biggest concern over the last decade, under the heading of How Do We Fix This, is combating disinformation in the MAGA-curious and those completely disengaged from information (the true cultists are a different problem that, if it can be solved, requires a different process).

Expand full comment
Samantha Millar's avatar

Agree that understanding what gets non-voters into the pool is tough. What does the Mamdani data show with regards to that? I read that new voter turnout was big in the election but not sure if that was across the board or just for Mamdani.

Expand full comment
Sam's avatar

It seems like asking nonvoters who they would have voted for shortly after the election would inherently bias responses in favor of the winner.

Expand full comment
G. Elliott Morris's avatar

I believe the data is pre-election

Expand full comment
Sam's avatar

The questionnaire is dated November 12-17 (relevant question on p. 3): https://www.pewresearch.org/wp-content/uploads/sites/20/2025/06/PP-2025.6.26_validated-voters_questionnaire.pdf

Expand full comment
G. Elliott Morris's avatar

Apologies then. I may reach out and ask if they are concerned about this. But note Nate Cohn’s story: This is the latest analysis confirming the same finding https://www.pewresearch.org/politics/2025/06/26/validated-voters-2024-methodology/

Expand full comment
Jon Saxton's avatar

I appreciate this data report as far as it goes, but it would be helpful to have demographic information, especially those who swung from the prior election in one way or another.

Expand full comment
axlee's avatar

TBH, if you just look at the 7 battleground states, had we had a compulsory voting, the only one Harris might have done better, or even been able to carry narrowly, is Georgia. A 5.3m turnout from about 7.5m-ish eligible population, the remaining 2million plus pool of non voters are not just majority minority, there are literally more Black nonvoters than White.

All the other 6, would have had larger gaps.

Expand full comment
Laura Belin's avatar

I recently dug into Iowa data on voter registrations and midterm election turnout. The Pew findings deepen my concern about how hard it will be for Democrats to dig out of a hole here. Many Democrats assume that most non-voters largely agree with them and oppose Republican policies, but the data don't seem to bear that out.

I don't know how feasible it is for Iowa Democrats to find and register a bunch of disengaged people who would actually support the party's agenda.

https://laurabelin.substack.com/p/real-talk-on-the-long-odds-facing

Expand full comment
Malcolm Kottler's avatar

Toy write: "but that roughly one out of every four voters behaved differently in 2024 than they did in 2024."

Correct this typo. You mean:

"but that roughly one out of every four voters behaved differently in 2024 than they did in 2020."

Expand full comment
Malcolm Kottler's avatar

Now I correct my typo! You write: "but that roughly one out of every four voters behaved differently in 2024 than they did in 2024."

You mean: "but that roughly one out of every four voters behaved differently in 2024 than they did in 2020."

Expand full comment
G. Elliott Morris's avatar

Yes I do. Corrected. Thank you.

Expand full comment
DJAnyReason's avatar

The ultimate conclusion for the Pew section - 'campaigns matter!' - is, I think, stronger than the data justifies. 25℅ of voters changing from 2020 to 2024 is also consistent with everyone having made their minds in 2022&2023 (a hypothesis which has generally been argued in this very substack!). It's more like 'events matter!', and a campaign is a type of event

Being pendantic because I'm not willing to comceed that "They're eating the dogs!" was a winning campaign message

Expand full comment
G. Elliott Morris's avatar

Pedantry accepted

Expand full comment
Jay Morris's avatar

I'd like to know what effect dark money had since Citizens United, but I guess its dark for a reason!

Expand full comment
Terry P's avatar

Voters hate inflation more than they hate Trump.

Expand full comment
Laura Belin's avatar

I also think that the pandemic sucked for most people and they associate Trump with "good times before the pandemic" (even though it started on his watch and his actions made it worse than it needed to be).

Expand full comment