Out now open access at
@ajpseditor.bsky.social. 194 potential exclusion-restriction violations for studies using weather as an instrumental variable onlinelibrary.wiley.com/doi/full/10....
It was great to see so much innovative work at the @ucbitss.bsky.social Annual Meeting today. Thanks to Abhishek Nagaraj for the mention, I’m looking forward to continuing the discussion on using AI to advance open science!
x.com/abhishekn/st...
Nothing says “partial American victory” like “all ships passing through the Strait of Hormuz must coordinate with the Iranian military” because the Strait is “under Iranian control”
It's just an utterly relentless pattern: Every time the President de-escalates in the Middle East, the stockmarket in the U.S. rejoices. It's like they think war is bad for business.
Excited to share the agenda for the next BITSS Annual Meeting! Join us April 16 in Berkeley for presentations on AI in research, open science, and data quality.
www.bitss.org/events/bitss...
With the most robust rail network in the American Midwest, Chicago has the 2nd largest number of car-free households of any metro area (behind only NYC)
Most of those households live around the L, especially the Red Line and western parts of the Blue & Green Lines
Okaun Eye of Chaos, which at the beginning of combat you flip until you lose a flip, and double its power and toughness for each time you win.
Saw @gavinverhey.bsky.social in drag on Spell Slayers which was v fun and also learned about this card which is a great demonstration of the St. Petersburg Paradox! This creature has an average power equal to infinity, and yet is realistically much smaller than measly 10k-power Jumbo Cactuar
Trump finally listening to these guys
With some help from Claude Code, I have the app I've always wanted:
elicitcausal lets you design a causal graph with your theoretical priors & preregister it. Then after you complete a study, you can upload your graph and get estimates of causal learning.
Link: causal.wilddata.solutions
#rstats
Congratulations! Assuming the PhD plans move forward as expected I am excited to apply.
Good summary for the lack of a loneliness epidemic here ourworldindata.org/loneliness-e...
I have a forthcoming paper in which this issue is explored. Domain specific metadata extraction tasks appear to be a major hurdle for current LLMs, especially with regard to specification and variable selection.
Sounds about right. We followed a similar process in developing ground truth metadata (around 600 papers) minus the RAvLLM grading. Cross-checks revealed tons of discrepancies in line with your findings.
Impressive, I have a similar ongoing project and continue to be hit with the “correct” given the definitions hurdle. Would love to set up a chat to hear more about your process!
Really nice paper! On the topic of metadata extraction, what was your retrieval method (RAG, static, etc.)? What prompting strategy did you use?
It must be very hard to publish null results Publication practices in the social sciences act as a filter that favors statistically significant results over null findings. While the problem of selection on significance (SoS) is well-known in theory, it has been difficult to measure its scope empirically, and it has been challenging to determine how selection varies across contexts. In this article, we use large language models to extract granular and validated data on about 100,000 articles published in over 150 political science journals from 2010 to 2024. We show that fewer than 2% of articles that rely on statistical methods report null-only findings in their abstracts, while over 90% of papers highlight significant results. To put these findings in perspective, we develop and calibrate a simple model of publication bias. Across a range of plausible assumptions, we find that statistically significant results are estimated to be one to two orders of magnitude more likely to enter the published record than null results. Leveraging metadata extracted from individual articles, we show that the pattern of strong SoS holds across subfields, journals, methods, and time periods. However, a few factors such as pre-registration and randomized experiments correlate with greater acceptance of null results. We conclude by discussing implications for the field and the potential of our new dataset for investigating other questions about political science.
I have a new paper. We look at ~all stats articles in political science post-2010 & show that 94% have abstracts that claim to reject a null. Only 2% present only null results. This is hard to explain unless the research process has a filter that only lets rejections through.
New data on ICE arrests (2015–2025) show that as arrests surge, the share of arrested people with criminal convictions fall sharply, from Chloe N. East, Caitlin Patler, and Elizabeth Cox www.nber.org/papers/w34794
core part of the ostensible economic justification for mass deportation was that Americans would rush in to take all the jobs immigrants left, but now unemployment rates for native born Americans are rising and the admin is sending people out to do damage control
New report out by yours truly @clasp.org !!
ICE Activity in Trump 2.0 is Increasing Employment and Decreasing Full-Time School Enrollment for Teens in Mixed-Status Families
www.clasp.org/publications...
MINNEAPOLIS WILL WIN
DACA work authorization shifted undocumented youth into safer, higher-paying jobs, showing legal barriers—not skills—limit occupational opportunity, from Aimee Chin, Kalena Cortes, and Camila Morales www.nber.org/papers/w34685
Examining whether GLP-1 medications pay for themselves finds no spending offsets through 5 years; savings, if any, may arise later or outside healthcare, from Coady Wing, Sih-Ting Cai, Daniel W. Sacks, and Kosali I. Simon www.nber.org/papers/w34678
I know people on this website are likely to be more amenable to this than I am, but I want to stress that institutional investor purchases of single family homes are a microscopic share of the market and so you’re basically not gonna see material changes in home prices from this
On DOGE’s watch, federal spending did not go down at all. It went up. But is still led to cuts that closed offices, canceled programs and deprived people of food, medicine and other aid.
www.nytimes.com/2025/12/23/u...
Making immigration easier increases international scientific collaboration.
Using China’s staggered rollout of visa-free access in 2023, this paper shows a causal increase in Sino-foreign co-authored research.
