Gabe Lenz

@gabelenz.bsky.social

Professor of Political Science at UC Berkeley

Codex has taken the pain out of finishing a data-heavy book. Here’s a paragraph about it. Being able to talk to your book and data files feels Star Trek-like. Link to my agent instructions below. ...

Bild

I just added up the number of years Americans have spent in prison 1979-2021: 48.5 million. In the figure below, each square represents a million years spent in prison, an unfathomable number. 1/4

Bild

I wonder what tends to happen when the leader of your constitutional republic, facing no invasion, decides to bring a large number of soldiers into the capital city. Lets look at the historical examples and oh good Lord...

Jonathan Ladd@jonathanmladd.com · 2mo ago

Since taking office again, Trump keeps bringing more and more troops into DC. The only possible legitimate need for them is to assist in a second coup attempt. Maybe he'll try it, maybe he won't, but the possibility is why the troops are here.

I understand the fear and anger. But also: 1) It is connected to the belief that AI is fake and going to vanish, which means that critics who should be helping shape AI use through policy and collective work are sitting it out 2) Lumping all AI criticism/talk into us vs. "tech bros" doesn't help

Christopher Mims@mims.bsky.social · 9mo ago

I say this as someone whose entire job is finding the narrow use cases in which AI is actually useful: All the fear about AI, all the reluctance to adopt it, is entirely legitimate, and a product of the fact that it's been foisted on the public with no consideration of its negative externalities.

New paper in PNAS: Your homicide risk follows you around the country. If you’re born in the historically violent Wild West, Appalachian Highlands, or Deep South, a higher risk of violent death trails you wherever you go. 🧵1/14

BildBild

Research: Believing that the mass public in the opposing party supports undemocratic tactics leads people to be more supportive of undemocratic tactics themselves. Being assured that a majority of your opponents oppose such radical tactics increases support for democratic norms osf.io/my987/download

osf.io

Don’t trust an observational model with a bunch of arbitrary control variables @gabelenz.bsky.social @alexandersahn.bsky.social Lenz, Gabriel S., and Alexander Sahn. "Achieving statistical significance with control variables and without transparency." Political Analysis doi.org/10.1017/pan....

Achieving Statistical Significance with Control Variables and Without Transparency | Political Analysis | Cambridge Core

Achieving Statistical Significance with Control Variables and Without Transparency - Volume 29 Issue 3

doi.org

New paper. What happens when a presidential candidate steps way out of line, not only with the public, but even with his own voters? And on a highly salient issue? Candidates typically don’t do that—they’re too strategic. But Trump isn’t. He’s good for social science. 🧵

Bild

One way to think about what's happening with Trump getting elected and then waging war on the Constitution and laws is that: It's taken less than 9 months for Trump v. U.S. to almost break the American republic. /1

An underappreciated reason why Harris lost was flat real disposable income growth in the election year. RDI was above trend, but voters care about growth in the election year itself. Between Jan. and Nov. 2024, nothing happened.

Bild

New 2024 data: Over 50% of GOP fundraising now comes from just 100 mega-donors. Trump’s embrace of the billionaire class—and their embrace of him—will only supercharge this trend.

Bild

Imagine if employees of the IRS could give out cards to their friends and family and those with cards didn't have to pay taxes. I could see an argument that this is even worse.

Post nicht verfügbar.

Since their rollout, I have made heavy use of the custom GPT tools in ChatGPT. Here is a thread of some of the tools that I use regularly (all of them require ChatGPT 4). I think they are most useful for people doing quantitative social science, but some are more general.

It's added to the fun of writing papers in markdown that copilot tries to write not only the R code but also the paper, suggesting what I should say next. The spooky part is that once in a while it guesses correctly.

Across 93 countries, the preferences of the lowest socioeconomic group, rather than those of the median or upper strata, are most predictive of realized redistribution. This finding contradicts the expectations of both leading experts and regular citizens. www.nber.org/papers/w31974

Whose Preferences Matter for Redistribution: Cross-country Evidence

Founded in 1920, the NBER is a private, non-profit, non-partisan organization dedicated to conducting economic research and to disseminating research findings among academics, public policy makers, an...

nber.org

In the aftermath of the financial crisis austerity advocates claimed that contractionary policy was actually expansionary bc of confidence — which I mocked as belief in the Confidence Fairy. It seems to me now that we're seeing a somewhat parallel myth: the Credibility Fairy 1/

The real threat to humanity is humans, not AI. Every decade technology makes it easier for a small group to wipe us out. Will AI improve fast enough to protect us from ourselves? Sure hope so.