Ultimately, our graphics should tell a story about our research, whether it’s about changing media habits or shifting social norms. Here's a few of our favorite visuals from 2025, how we made them, and what makes them successful.
In the U.S. House, the president’s party has lost seats in all but four midterms since 1862.
(@pollsandvotes.bsky.social)
More, via Opinion Today:
opiniontoday.substack.com/p/251209-top...
The reported peace plan developed by envoys for Trump and Putin sounds more like a list of Moscow's demands than a compromise between two combatants.
It's not clear what if anything Russia is willing to concede even as it insists that Ukraine give up a lot. www.ft.com/content/2353...
It is incredibly simple to acknowledge that violence in any form is awful and erodes society.
“.. Unemployment rates among tech workers between 20 and 30 years old jumped by 3 percentage points since the start of this year .. , “a .. larger increase than we’ve seen for other young workers,” he said.
@cnbc.com
www.cnbc.com/2025/08/05/a...
Given the other questions on this survey, it's reasonable the researchers would want to check for people who might be yanking their chain
A Washington Post poll found most Americans strongly support releasing all files in the Epstein case and suspect the documents contain embarrassing information about President Trump, Democrats and billionaires.
A few days ago, the new Washpost Opinion boss extolled Bezos's shift in the direction of the page and warned that the changes will be "unwelcome for some"
Which is what managers say when they're trying to get people to quit before they get fired.
While MAGA is ubiquitous, little is known about what it means to the American public. Ten years on, what do Americans think when they hear or read this phrase?
(@jesserhodesumass.bsky.social @eichen.bsky.social Douglas Rice, Gregory Wall and Tatishe Nteta)
theconversation.com/what-maga-me...
The Economist/YouGov poll in April found that 11% of Trump voters are medicaid recipients
Across 12 high-income countries, public satisfaction in democracy has declined over time. In 2017, a median of 49% were satisfied with the functioning of democracy. By 2025, just 35% say the same thing.
www.pewresearch.org/...
Interesting analysis.
“Pew found that Trump’s success with Hispanics was primarily due to changes in turnout rather than voters changing their preferences. Latino voters who voted in ‘20 but not ‘24 backed Biden by 2-1. While Trump won Latino voters who skipped ‘20 by 23 points.”
A new Pew Research Center study solidified what pre-election surveys suggested, that President Trump was able to build a far more diverse Republican coalition in the 2024 election than ever before. Here's what else the research showed.
NYT channeling a 1950s greaser
Murder remains down more than 21 percent in the 30 cities that reported the most murders to the FBI in 2023 (the most recent official release). Murder is down in 26 of the 30 cities. All but 4 cities have data through late May and all but 1 (c'mon Phoenix!) have data through April.
Smart look at what we can and can't extrapolate right now from Democrats' in-party perception weakness (and also a good aside on how changes in survey mode can have unexpected effects on trendlines)
I don't know that it's especially difficult to conceptualize the use case for surveys that gather data on a) alcohol consumption or b) internet usage among the American public
Must-read on the state of the polling landscape
Conservative and Liberal Americans flatly reject authoritarian rule.
RIP Joe Nye, the dean of American political scientists who coined the term “soft power” — the idea that America’s global influence was more than its military might. www.nytimes.com/2025/05/08/u...
Was looking thru Gallup 1975 polling data (for fall of Saigon) & found this:
Party ID (a year after Nixon resigned)
Dem 46%
Rep 22%
Ind 22%
Educational pattern was reverse of today: Dems’ narrowest edge, 35%-27% among college grads, widest (59%-20%) among ‘grade school’ grads.
This is a nicely thorough overview:
One very clear change from our October/November 2024 survey of Canadians to our March/April 2025 survey of those same (n=1,181) respondents is the 10 percentage point jump in those saying being Canadian is their single most important identity.
Screenshot of a news article reading "A recent survey by the scientific journal Nature revealed that 75% of the 1,600 scientists surveyed were considering relocating to Europe or Canada, citing actions taken by President Donald Trump." The reference is to a non-representative convenience sample.
And predictably, this "stat" - again, reflecting only readers of Nature who filled out a form - is now turning up in other stories as though it's based on an actual survey, which it is not.
(The bigger story itself is, of course, far more important and troubling: www.nbcnews.com/news/us-news...)
The @WSJ headline says it all: Dow Headed for Worst April Since 1932 as Investors Send ‘No Confidence’ Signal. @hannaherinlang.bsky.social www.wsj.com/finance/inve...
Line chart from Pew Research Center showing changing attitudes among Democrats and Republicans about various potential outcomes of the Russia-Ukraine war.
% who are extremely/very concerned about Ukraine being defeated and taken over by Russia
🔵 Democrats: 62%
🔴 Republicans: 25%
% who are extremely/very concerned about Russia invading other countries in the region
🔵 Democrats: 66%
🔴 Republicans: 29%
www.pewresearch.org/global/2025/...
Public opinion of Pope Francis over the years
A good time to read Gene Healy's The Cult of the Presidency, which traces the history of this notion, which starts with Woodrow Wilson www.cato.org/books/the-cu...
74% of respondents—including 91% of Democrats and 55% of Republicans—said it was reckless for senior leaders in the Trump administration to discuss attack plans on the Signal app, while 22% said it was a harmless oversight.
(Reuters/Ipsos)
More, via Opinion Today:
opiniontoday.substack.com/p/250403
This graphic got a lot of action last week. I was curious about the underlying data so I pulled a copy from the same source, but included all the airports being reported, not just five.
And I included the total data, not just YoY change.
The numbers are a bit less shocking.