Neil O'Brian

This is interesting. Why the proportional increase in PAC spending in early 2010s? Citizens United?

Some # of people on Medicaid don't know they are on Medicaid (www.pbs.org/newshour/hea...). Whether there is a partisan tilt to that, IDK. But a 2025 survey shows that just 4% of Rs say they are on Medicaid, which is less than Ds and Is. This could also be due to social desirability in polling, etc.

Another year of marriage in the books. And what a year it has been! Looking forward to many more :)

Nothing beats Oregon on a sunny day. (Clementine checking out South Sister on the left; Broken Top on the right.)

The percent of Americans who report they are in "excellent" health has been declining for some time, but has accelerated since 2012. Source: GSS

Conservative are more likely today to perceive medical doctors, and medicine, as filled with liberals. Among conservative respondents, their perception that most people in medicine/med doctors:

In 2025:
36% Liberal
45% Mod
19% Conservative

In 2019 :
21% Liberal
54% Mod
25% Conservative

Image description Partisanship and Trust in Personal Doctors: Causes and Consequences Abstract In the first decades of the twentieth century, the gap in age-adjusted mortality rates between people living in Republican and Democratic counties expanded; people in Democratic counties started living longer. This paper argues that political partisanship poses a direct problem for ameliorating these trends: trust and adherence in one’s personal doctor (including on non-COVID-19 related care) – once a non-partisan issue – now divides Democrats (more trustful) and Republicans (less trustful). We argue that this divide is largely a consequence of partisan conflict surrounding COVID-19 that spilled over and created a partisan cleavage in people’s trust in their own personal doctor. We then present experimental evidence that sharing a political background with your medical provider increases willingness to seek care. The doctor-patient relationship is essential for combating some of society’s most pressing proble

Figure 1. Line graph with three panels showing trends in attitudes toward Education, Medicine, and the Scientific Community from 1988 to 2020. Each panel displays two lines: one for Democrats (light blue) and one for Republicans (red), with vertical error bars. Y-axis values range from 1.75 to 2.5. In all panels, Republican ratings decline more steeply over time, especially after 2010. Democrat ratings remain relatively stable or increase slightly, particularly in the Scientific Community panel after 2010.

Figure 2. Dot-and-whisker plot showing treatment effects (Treatment – Control) on three outcomes: “Trust Own Doctor,” “Adhere Doc Advice,” and “Conf. in Medicine.” Three groups are plotted: Vote Biden (light blue), Vote Trump (red), and Biden–Trump difference (gray). The y-axis ranges from -0.5 to 1.0. For “Trust Own Doctor” and “Conf. in Medicine,” the Biden group shows positive treatment effects, while the Trump group shows negative effects. The Biden–Trump difference is positive for all outcomes, with error bars indicating uncertainty. A horizontal dashed line at 0.0 marks no treatment effect.

Image description Figure 3. Dot-and-whisker plot with four panels showing Average Marginal Component Effects (AMCE) for Democratic (light blue) and Republican (red) respondents across different attributes: Male, Ivy League, Far Away, Democrat, High Rating, Medium Rating, Black, and Hispanic. Panels display results for All Respondents, Female Respondents, Black Respondents, and Latinx Respondents. The x-axis ranges from -0.25 to 0.50 with a vertical dashed line at 0.0 indicating no effect. Each dot represents the AMCE estimate with horizontal error bars indicating uncertainty. Some estimates differ by respondent group, and not all attributes have data points for both political affiliations in all panels.

Despite trust in personal doctors becoming a partisan issue, experimental evidence suggests that sharing a political background with one's medical provider increases willingness to seek care, finds @obrian.bsky.social & Bradley Kent in @bjpols.bsky.social doi.org/10.1017/S000...

4th most read story in local newspaper is about the cat lounge in town. The fourth estate is healthy!

The #1 read story in Eugene’s newspaper is about a dog who climbed Mt Hood. Local journalism is alive and well!

Want to learn more about how women are, and are not, role models for adolescents in American politics? Our new book, See Jane Run: How Women Politicians Matter for Young People, will be out this Spring. press.uchicago.edu/ucp/books/bo...

E-pub day is tomorrow (8/27 – hard copy still 9/25)! Roots of Polarization argues that contemporary polarization on “culture war” issues like abortion and gun control in the 1970s-1980s was molded by the racial realignment of the 1960s. #polisky

The Roots of Polarization A deeply researched account of how battles over civil rights in the 1960s shaped today’s partisan culture wars. In the late twentieth century, gay rights, immigration, gun control, and abortion debate...

"Roots of Polarization" has a 9/25 release date! In the mean time, plan on coming to the author-meets-critics panel at APSA (Thurs @ 8am). Really great panel: @cwolbrecht.bsky.social, @chriswarshaw.bsky.social and Vince Hutchings discussing, & chaired by Eric Schickler. #polisky
tinyurl.com/ycypu4nm

Interesting free-response comments from American Soldier surveys of Black GIs in WWII (part of @edgitre.bsky.social 's fantastic mass transcription project) about the surveys themselves. Attitudes range from deep distrust to support.

do you know of any randomize controlled trials on effects of phones/social media on learning outcomes? from clicking through links can't tell if ambiguity on sm effects is because there is not good evidence, or there is good evidence that shows null effects.

Is the argument that if Haidt didn't publish book/have a platform that then there wouldn't be a push to limit phones and social media?

thanks! and a great quote!

True. But rank-and-file white evangelicals (including in SBC) already conservative on abortion by 1960s. Leadership moved/replaced to represent conservative insurgency. More in forthcoming book "Roots of Polarization": tinyurl.com/45evxyby

congrats!

please do. looking forward to it.

Thank you, Shana!

Neil A. O'Brian | Carnegie Corporation of New York

Thank you to the Carnegie Corporation for this opportunity to pursue research on the role politics plays in shaping health outcomes in our polarized society. #CarnegieFellows
t.co/DJtcc2zyWt

where did the independents go? even after asking lean there are more than ~5% Ind I would think? And it looks like higher educated african-americans are leaving Dems more than lower?

New attention check just dropped (and yikes) www.pewresearch.org/short-reads/...

My thought is while older generations, or even myself, were socialized that democracy was a core value, people who came of age in the Trump era see it as a contested ideal or something that isn't the default. That said, support is still high, and these things are still common, but it's fraying.

Messages about Trump as a threat to democracy might be relatively less effective in bringing young voters to Biden camp. We primed Dems with messages about J6/democracy and find that young Ds not moved by that message but older Ds are. Article in The Conversation: tinyurl.com/387k7keu polisky

other polls have asked about that. this was about J6 as threat to democracy. i've pasted one, below (of Dems only(, that paints a broader picture. philip bump in wapo did an interesting piece on this a while ago among all young voters.

that means 158 observations (data are weighted). percentages are the second number in each cell. poll is fine.

Another example: when asked if entering capitol on J6 was a legitimate exercise of 1st amendment rts 71% of Ds 18-29 said "not at all legitimate" compared to 89% of Ds 65+

Young people, including young Dems, are relatively less committed to democracy as an institution than older ppl and perhaps see J6 as less of a threat to democracy. Below is q about perception of J6 as a crisis/problem (asked in 2022). Responses are D only. polisky

Colleague Chandler James & I wondered if we gave young Democrats information about Biden (e.g., economy is doing well) whether it would increase support/enthusiasm. It didn't move the needle much. Older Dems, though, became quite a bit more supportive. polisky

theconversation.com/young-people...