We have an apparent cure for most people with pancreatic cancer—a profoundly swift, deadly cancer—and the position of the U.S. government is that it should be banned. This is real fall-of-empire stuff.
Why not review for Elsevier? (Pardon my ignorance, honest question, I don't review for MDPI!)
Jan-15 45.4% Feb-15 52.4% Mar-15 51.0% Apr-15 49.8% May-15 48.0% Jun-15 47.6% Jul-15 49.5% Aug-15 54.8% Sep-15 53.1% Oct-15 52.1% Nov-15 52.8% Dec-15 53.3% Jan-16 57.2% Feb-16 58.3% Mar-16 56.7% Apr-16 54.2% May-16 55.3% Jun-16 51.9% Jul-16 56.9% Aug-16 56.6% Sep-16 50.3% Oct-16 56.2% Nov-16 55.6% Dec-16 55.1% Jan-17 56.3% Feb-17 59.4% Mar-17 60.3% Apr-17 59.4% May-17 59.9% Jun-17 61.7% Jul-17 65.9% Aug-17 64.2% Sep-17 63.0% Oct-17 63.5% Nov-17 62.3% Dec-17 63.7% Jan-18 62.0% Feb-18 64.5% Mar-18 64.5% Apr-18 61.8% May-18 64.6% Jun-18 66.9% Jul-18 68.6% Aug-18 67.6% Sep-18 68.0% Oct-18 68.5% Nov-18 66.5% Dec-18 67.7% Jan-19 74.1% Feb-19 68.1% Mar-19 69.8% Apr-19 68.4% May-19 65.1% Jun-19 66.0% Jul-19 65.7% Aug-19 68.3% Sep-19 68.5% Oct-19 69.5% Nov-19 67.7% Dec-19 67.0% Jan-20 69.5% Feb-20 70.7% Mar-20 69.3% Apr-20 61.0% May-20 63.7% Jun-20 66.3% Jul-20 59.7% Aug-20 55.7% Sep-20 63.6% Oct-20 63.0% Nov-20 64.9% Dec-20 65.4% Jan-21 62.9% Feb-21 56.7% Mar-21 56.4% Apr-21 58.7% May-21 61.4% Jun-21 55.3% Jul-21 43.7% Aug-21 45.8% Sep-21 47.2% Oct-21 49.3% Nov-21 44.4% Dec-21 44.4% Jan-22 46.7% Feb-22 51.7% Mar-22 49.7% Apr-22 43.8% May-22 41.2% Jun-22 45.8% Jul-22 52.7% Aug-22 55.5% Sep-22 54.0% Oct-22 52.7% Nov-22 50.6% Dec-22 46.8% Jan-23 47.1% Feb-23 49.8% Mar-23 47.8% Apr-23 49.7% May-23 45.8% Jun-23 45.6% Jul-23 45.2% Aug-23 43.4% Sep-23 42.9% Oct-23 44.2% Nov-23 45.5% Dec-23 46.5% Jan-24 47.9% Feb-24 46.4% Mar-24 47.5% Apr-24 49.3% May-24 50.8% Jun-24 50.8% Jul-24 52.1% Aug-24 56.6% Sep-24 59.2% Oct-24 59.1% Nov-24 62.0% Dec-24 60.3% Jan-25 64.1% Feb-25 71.2% Mar-25 73.0% Apr-25 75.1% May-25 75.4% Jun-25 72.0% Jul-25 73.9% Aug-25 74.8% Sep-25 78.3% Oct-25 78.8% Nov-25 78.2% Dec-25 84.4% Jan-26 90.5% Feb-26 94.7%
Hey, has anyone visited TRAC Immigration and looked up asylum denial rates in US immigration courts?
October 2000-January 2025: 56% of cases denied
February-November 2025: 75% denied
December 2025: 84%
January 2026: 91%
February 2026: 95%
tracreports.org/phptools/imm...
Substantially more rigorous than the papers that are scaring everyone
One of my favorites paper got published 🤓 It covers a lot of ground and it’s the best summary of my views on misinformation and what to do about it. Give it a read :)
🔓 osf.io/preprints/ps...
👉 doi.org/10.1177/1461...
New preprint out today (osf.io/preprints/ps...). We tested whether AI agents are actually infiltrating online surveys.
Spoiler alert: they aren't
Thread 🧵
[1/9]
In his piece for Foreign Policy published Feb. 24, Swanson wrote that Iran would not capitulate after a bombing campaign, but rather escalate and “target global oil flows and international shipping, sending energy prices up and creating a serious political liability for Trump.” And indeed, Iran has made scattershot attacks on energy targets and others across the region, as well as throttling passage through the Strait of Hormuz by threatening attacks on ships.
Loomer post noting that a government expert on Iran was part of the Iran negotiations.
This guy sounds smart. Laid out Iran's moves with incredible foresight. Too bad he no longer works as the National Security Council’s director for Iran b/c some online nutjob does not like him and that is enough to get experts shitcanned from government these days.
www.politico.com/newsletters/...
But there's so much still to understand. What are the conditions under which voters will view politicians (especially of their own party) as corrupt, and under which they will vote against them? An incomplete list of categories might include
- Seniority of office
- Pivotality of seat
The Trump administration treats all of its adversaries, foreign and domestic, as non-playable characters (NPCs), and when those adversaries react in ways they did not predict (i.e., they don't fold), the Trump administration is hopelessly befuddled.
People who thought credibility and individual leaders didn’t matter much in international relations were taking years of cultivated and maintained credibility for granted, and didn’t sufficiently consider that a major power would voluntarily throw so much of it away by empowering a leader with none.
NEWS: Six months before the Trump admin began bombing Iran, the Department of State fired its oil and gas experts.
State’s energy division got completely DOGE’d. And with it went the people who knew how to plan for a global energy crisis.
FWIW I don't buy this argument. Nobody thinks p-hacking, non-replicable findings, or data errors are okay. Avoiding those things is a standard that individual researchers & fields as a whole apply to themselves constantly
Income ≠ wealth
The wealth distribution is surely much less equal
Conditionally accepted at the APSR (w/ @scottclifford.bsky.social & @patrickpliu.bsky.social):
Why does political information so often change beliefs but NOT attitudes? We highlight the role of belief relevance, or the extent to which beliefs bear on attitudes.
"...estimated last year that a $200-per-month Claude Code subscription could use up to $2,000 in compute, suggesting significant subsidization by Anthropic. Today... that $200 plan able to consume about $5,000 in compute" www.forbes.com/sites/annato...
FT comments section this morning - saying what everyone else is thinking, right?
"In particular, while expressive responding clearly causes surveys to exaggerate the extent to which partisanship acts as a perceptual screen, the same forces that produce expressive responding in surveys may also affect the political judgments people make in real life."
I appreciate your post, bit of a luddite but you & others have convinced me to try Claude for a few things on my to do list
Haha dude I was just coming here to make this same comment
I think about this Tony Benn speech much more than I used to
My take on the partisan expressive responding literature is now in print. Open access: doi.org/10.1017/S000...
Have examples of social science papers that use DAGs to justify their controls? I find these very hard to come by & would like to use for teaching.
Hell of a chart, this, from @financialtimes.com. Spot the JCPOA.
on.ft.com/4kSuQYO
And let me refine "better": Talarico makes the case in a way that I think is especially likely to resonate with an audience that really needs to hear it
Open to it. Til then, anything you'd recommend I read / watch / listen to?
Not sure I agree. Rs oppose Jesus on every issue. To the extent I've seen Talarico makes that case better than anyone in politics
I am the reason there are so many em dashes in the training data
This recent RCT of an "AI stethoscope" claims the technology "shows promise" for diagnosing cardiovascular conditions.
It does not.
It is a textbook example of the risks of conducting unprincipled 'per protocol analyses'. Once again, peer review at a major medical journal has failed.
🧵 1/
