Lightnews — Scholar-powered news

Reposted by Scott Free

Catherine Rampell

@crampell.bsky.social

Actual 2025 tariff rates are, to date, far less than statutory rates, but the costs of the tariffs are mostly borne by the US – estimated tariff pass-through to US prices in 2025 is 94%. Tariffs on imports act like a 1 to 2pp tax on US manufacturing. brentneiman.com/research/GN2...

December 27, 2025 at 2:04 PM

Reposted by Scott Free

Aaron Sojourner

@aaronsojourner.org

AI radically lowers the cost of polluting our information ecosystem.

Our ecosystem was already stressed & under attack across many fronts.

This unleashes new floods of slop. Slop swamps high-quality info, forcing us each to pay more for costly discernment or accept losing our way in the flood.

December 15, 2025 at 11:26 PM

Reposted by Scott Free

NBER

@nber.org

Surveying econometric innovations related to differences-in-differences estimators and event-study models with time-varying treatment effects, from Alberto Abadie, Joshua Angrist, Brigham Frandsen, and Jörn-Steffen Pischke www.nber.org/papers/w34550

December 15, 2025 at 8:01 PM

Reposted by Scott Free

Paul Goldsmith-Pinkham

@paulgp.com

I've now updated this with a few updates:

1. I've added the JFE, JF, and RFS
2. I've added QJE and (most of) JPE
3. I've fixed some bugs
4. I added a way to adjust for length of papers over time
5. I've added citation counts (per OpenAlex)

paulgp.com/econlit-pipe...

December 8, 2025 at 8:15 PM

Reposted by Scott Free

Paul Goldsmith-Pinkham

@paulgp.com

Great post by Saloni on the value of visualization #linkoftheday

www.scientificdiscovery.dev/p/salonis-gu...

December 10, 2025 at 12:23 PM

Reposted by Scott Free

Better Things Are Possible

@internethippo.bsky.social

One of the many reasons AI can't produce good writing is it can't hate its own writing. It can't think to itself "Maybe I'm illiterate" during the writing process. And that's essential

December 9, 2025 at 8:45 PM

Reposted by Scott Free

Andrew Heiss

@andrew.heiss.phd

Some closing thoughts for my students this semester on LLMs and learning #rstats datavizf25.classes.andrewheiss.com/news/2025-12...

Will you incorporate LLMs and AI prompting into the course in the future?
No.

Why won’t you incorporate LLMs and AI prompting into the course?
These tools are useful for coding (see this for my personal take on this).

However, they’re only useful if you know what you’re doing first. If you skip the learning-the-process-of-writing-code step and just copy/paste output from ChatGPT, you will not learn. You cannot learn. You cannot improve. You will not understand the code.

In that post, it warns that you cannot use it as a beginner:

…to use Databot effectively and safely, you still need the skills of a data scientist: background and domain knowledge, data analysis expertise, and coding ability.

There is no LLM-based shortcut to those skills. You cannot LLM your way into domain knowledge, data analysis expertise, or coding ability.

The only way to gain domain knowledge, data analysis expertise, and coding ability is to struggle. To get errors. To google those errors. To look over the documentation. To copy/paste your own code and adapt it for different purposes. To explore messy datasets. To struggle to clean those datasets. To spend an hour looking for a missing comma.

This isn’t a form of programming hazing, like “I had to walk to school uphill both ways in the snow and now you must too.” It’s the actual process of learning and growing and developing and improving. You’ve gotta struggle.

This Tumblr post puts it well (it’s about art specifically, but it applies to coding and data analysis too):

Contrary to popular belief the biggest beginner’s roadblock to art isn’t even technical skill it’s frustration tolerance, especially in the age of social media. It hurts and the frustration is endless but you must build the frustration tolerance equivalent to a roach’s capacity to survive a nuclear explosion. That’s how you build on the technical skill. Throw that “won’t even start because I’m afraid it won’t be perfect” shit out the window. Just do it. Just start. Good luck. (The original post has disappeared, but here’s a reblog.)

It’s hard, but struggling is the only way to learn anything.

You might not enjoy code as much as Williams does (or I do), but there’s still value in maintaining codings skills as you improve and learn more. You don’t want your skills to atrophy.

As I discuss here, when I do use LLMs for coding-related tasks, I purposely throw as much friction into the process as possible:

To avoid falling into over-reliance on LLM-assisted code help, I add as much friction into my workflow as possible. I only use GitHub Copilot and Claude in the browser, not through the chat sidebar in Positron or Visual Studio Code. I treat the code it generates like random answers from StackOverflow or blog posts and generally rewrite it completely. I disable the inline LLM-based auto complete in text editors. For routine tasks like generating {roxygen2} documentation scaffolding for functions, I use the {chores} package, which requires a bunch of pointing and clicking to use.

Even though I use Positron, I purposely do not use either Positron Assistant or Databot. I have them disabled.

So in the end, for pedagogical reasons, I don’t foresee me incorporating LLMs into this class. I’m pedagogically opposed to it. I’m facing all sorts of external pressure to do it, but I’m resisting.

You’ve got to learn first.

December 9, 2025 at 8:17 PM

Reposted by Scott Free

Matt Darling

@besttrousers.bsky.social

Before castigating a book for not using a term, you should check to see if the book uses a term.

October 26, 2025 at 8:29 PM

Reposted by Scott Free

JD Long

@jdlong.cerebralmastication.com

New code review response just dropped:

September 20, 2025 at 12:26 PM

Reposted by Scott Free

Andrew Heiss

@andrew.heiss.phd

look what you made me do

(#rstats code here gist.github.com/andrewheiss/... )

Original world map sliced into 8 areas, using the Mercator projection

The same 8 slices, but with the Robinson projection. Slice A is tiny and curvy now

The same 8 slices, but with the Robinson projection. Slice A is even tinier and curvier now

September 17, 2025 at 4:18 AM

Reposted by Scott Free

The UnPopulist

@theunpopulist.net

"So I actually like the deep state. I think that any functioning political system has to have a civil service that meets these basic Weberian requirements of being non-partisan, expert, professional, and the like."

- ‪‪‪@frankfukuyama.bsky.social‬ at #LibCon2025

August 14, 2025 at 9:25 PM

Reposted by Scott Free

Paul Hünermund

@p-hunermund.com

😯 I recently came across a sobering paper on here: among 16,649 hypothesis tests reported in political science, the median test has only about 10% power to detect the consensus effect size reported in the literature. Fewer than one in ten tests reach the conventional 80% power threshold.

August 13, 2025 at 11:56 AM

Reposted by Scott Free

Katie Mack

@astrokatie.com

"PhD-level experts in your back pocket" is a completely nonsensical description of AI but a pretty good description of social media if you follow the right people

August 9, 2025 at 11:07 PM

Reposted by Scott Free

Justin Wolfers

@justinwolfers.bsky.social

Stephen Moore never fails to make careless errors, cherry pick numbers, and misrepresent the data. My friend Christooher Clarke picks apart Moore's latest dishonesty, exposing mistakes and misrepresentations that are big, embarrassing, and disqualifying.

Worth watching: www.tiktok.com/t/ZT6aPQUJt/

Economist Stephen Moore cherry picks data to present in White House. He claims BLS overestimated Biden job numbers by 1.5M. However, he commits numerous errors. Over Biden's 4 yrs, the BLS initially u...

TikTok video by Christopher Clarke

www.tiktok.com

August 10, 2025 at 1:36 AM

Reposted by Scott Free

Julia M. Rohrer

@dingdingpeng.the100.ci

We’ve discovered a magic trick where we tell the kids that they absolutely *have to* help us clean up, and suddenly they’re perfectly able to entertain themselves for extended periods of time while maintaining a low profile.

August 10, 2025 at 4:04 PM

Reposted by Scott Free

Simon P. Couch

@simonpcouch.com

After a month or so of not using Claude Code much to write R code, I recently returned to using the tool most days. Giving coding assistants the ability to peruse #rstats package documentation via MCP is so, so helpful: www.simonpcouch.com/blog/2025-07...

How I’m using Claude Code to write R code | Simon P. Couch – Simon P. Couch

www.simonpcouch.com

July 17, 2025 at 7:07 PM

Reposted by Scott Free

Rachel Leah Childers

@donskerclass.bsky.social

"Useless estimator" story: I recently consulted with someone using Maximum Score for a real well-motivated empirical project! I was excited because I'd known it mostly as a theoretical curiosity, w/ the ugliest asymptotic distribution known to man that also breaks the bootstrap.
🧵

Paul Goldsmith-Pinkham @paulgp.com · Jul 16

There’s an analogy here about econometrics theory and applied economics work.

faculty.wcas.northwestern.edu/cfm754/et_in...

July 19, 2025 at 1:39 AM

Reposted by Scott Free

Julia M. Rohrer

@dingdingpeng.the100.ci

👀

June 26, 2025 at 7:00 AM

Reposted by Scott Free

Julia M. Rohrer

@dingdingpeng.the100.ci

New blog post! Let's say you've measured two variables repeatedly and want to investigate how one affects the other over time. Here are some recommendations for how to do that well.

www.the100.ci/2025/06/25/r...