Why don't university rankings incorporate the relative share of tenured vs. non-tenured faculty?
This seems to be a major factor in explaining educational quality for students; i.e., good teaching on hard topics requires support from unis vs. a constant threat of being dismissed.
In 1777, George Washington mandated that all members of the Continental Army be inoculated for smallpox.
In 2026, Pete Hegseth—citing the Biden administration's COVID vaccine mandate—ended the requirement that service members receive a flu vaccine.
Four small Quarto extensions to sharpen your Reveal.js slides: animated list fragments, step-by-step code annotations, auto-repeating headings, and keyboard-driven tabsets.
Each one fixes a single friction point. 🎯
mickael.canouil.fr/posts/2026-0...
#Quarto #QuartoPub #SlideCrafting #RevealJS
BJPolS abstract of a research paper discussing the relationship between immigration and crime rates. The text outlines the study's focus on examining crime rates among immigrants compared to local averages and highlights issues with large group comparisons.
NEW -
Null by Design: Statistical Dilution in Immigration-Crime Research - cup.org/4vDBgkh
- @riazsascha.bsky.social
#OpenAccess
Just 26% of Americans say they approve of how the president is handling prices/inflation, 35% on jobs and the economy. His net rating is way underwater on both
www.gelliottmorris.com/p/2026-04-21...
I have several advisees who have chosen this route, most of whom could have landed entry level academic jobs (postdocs or better) and simply wanted something different. All of them seem extremely happy.
sure, depends on the size of the codebase & whether you ask it to do something like read an article. The challenge is that it can consume a ton of tokens in a short amount of time when it goes through some recursive loop (try out new code, try again, etc).
So hard to predict token usage.
As Claude Code gets more expensive (⬆️ in pricing this past week 😬), I am wondering whether we will transition to a cheap & slow GPU option for many ordinary users.
I don't really need to process 20k tokens in 10 seconds. I could get away with slower hardware if it keeps costs down.
cheapAI 😁
Holy cow.
Ten months detention for simply being out of status with your visa.
How much time did Trump spend in jail for paying an adult porn star with hush money and falsifying the records?
Mismatch in the Egyptian labor market Caroline Krafft1 and Carmen Armas Montalvo2 Abstract High rates of unemployment and non-participation in the labor market are often attributed to labor market mismatch. This paper examines the mismatch between job seekers’ expectations and Egypt’s labor market realities. For the non-employed, analyses examine reservation wages and what occupations the unemployed would accept. For the employed, analyses explore skills and educational attainment relative to the skill and educational requirements of their jobs. The results demonstrate that the non-employed generally have reasonable wage expectations and are willing to accept public sector jobs, but these jobs, and to some degree formal private sector jobs, are not readily available. The non-employed, and especially women, are less likely to accept more readily available informal private sector employment. As a result of women being more selective in what employment they accept, among the employed, women’s skills and qualifications better match their job requirements than men’s, although overeducation and over-skill are substantial issues for both men and women. There are not differences in overeducation by vocational secondary specialization or among most higher education specializations, suggesting a pervasive problem of mismatch throughout the education system. Keywords: Mismatch, labor markets, education, skills, unemployment, reservation wages, Egypt JEL codes: J21, J23, J24, J31, J64, I21
🧵Excited to share our new working paper: "Mismatch in the Egyptian labor market" with Carmen Armas Montalvo. We use the 2023 Egypt Labor Market Panel Survey to investigate why high unemployment and low participation persist despite rapid educational expansion. hdl.handle.net/10419/339993 1/ #EconSky
The data warehouse includes information on household wealth levels and balance sheets, distributional statistics, and wealth transfer taxation. It brings together data on tax revenues, rates, exemptions, and schedules. Explore: wealthproject.gc.cuny.edu
What drives the growth of private wealth? Has inequality increased? How does tax policy shape wealth across generations? Now published in Nature-Scientific Data: the data descriptor of the GC Wealth Project Data Warehouse by @stone-lis.bsky.social & Roma Tre University
www.nature.com/articles/s41...
'Texas A&M philosophy professor Martin Peterson is leaving the university after administrators told him in January that he couldn’t teach Plato’s Symposium in his philosophy class; they said the ancient Greek philosopher’s work violated the system’s restrictions on gender and sexuality content.' 1/3
Worse than that, a some of that money is tied up in Trump assets, so it's effectively a bailout of Trump firms indirectly.
All complete datasets are alike; each partially complete data set is missing data in its own way.
I think too that part of the problem is not just "bad research methods" but also disciplinary incentives that require scholars to have big findings to get published.
This is a bad *philosophy of science.* We all know we can learn from nulls, but we in practice we ignore them.
To our detriment.
I have now examined two different highly-cited papers built on panel data sets derived from historical research.
I like both of the papers. The underlying research is sound & well-thought out.
But the panel data specifications are very weak. Both fell apart with obvious tests.
This is not great 🤷
🫡
CfP for our 3rd Annual HPE Conference @uscprice.bsky.social @usc.edu sponsored by @usccis.bsky.social and PIPE. Submit by May 30. If you work on MENA, I am running a smaller pre-conference for HPE papers focused on the region. @polisky.bsky.social docs.google.com/forms/d/e/1F...
Mamdani made more money from rapping than I did from my recent hit book* in an academic press
*according to my mom
That's what you get with saturated fixed effects models (or what economists call high-dimensional fixed effects).
Not commenting on the research, I do like the theory.
While approval is a polarized metric that changes slowly among Trump's supporters, “soft” measures of Trump’s popularity are eroding among Republicans.
New at 50+1:
blog.fiftyplusone.news/p/soft-measu...
email from NSF ACCESS
Likely to fly under the radar, so I wanted to flag it--
I received this email about a new requirement for the NSF to collect "country of citizenship" for all users of ACCESS, a program to share computational resources across universities.
Unclear why they suddenly want citizenship info...
😬
PaCSS (Politics and Computational Social Science) is back! APSA preconference at BU joint with Political Communication, deadline for submission is May 8. Should be great!
cssi.umass.edu/pacss2026
quote from conversation article
Smart piece in @theconversation.com about real-world effects of AI on jobs. I like the quote below 👇.
AI doesn't necessarily replace jobs because models struggle when they lack human supervision. The models can create *new* jobs & opportunities when industries are dynamic enough to take advantage.
poo, I liked the idea that they could live in that apartment on an academic salary even if I know it's not true
It's their lucrative royalties from books in top-end academic presses. Mamdani will be able to live comfortably in retirement when he inherits those copyrights.
I am looking to do some consulting this summer!
I've published research in AI/ML/causal inference/Bayes & have experience managing data science projects in political campaigns, marketing, epidemiology, economic development, surveys/experiments--just to name a few.
Pls reach out via DM 🙂
