The AEA trial registry is basically uniformly used in economics and has a fixed set of fields. The analysis plans people write are a free for all.

Did you learn to speak French when you lived there?

In development econ we pre-specify the sample size in the trial registration form

Mostly obstinacy :)

I can now understand French fairly well and this helps to maintain that skill. It’s also how I learned French; my only other investments in the language have been marrying a native speaker (who speaks it sometimes) and brief trips to Haiti and France.

My Duolingo streak is now a decade old!

Do people file pre-analysis plans for this sort of study? Where do they post them? The AEA trial registry only accepts randomized experiments.

This is an important issue that I’ve never seen addressed in an actual PAP, even the very long-winded ones

I’m clear on the conceptual issue here, but have you seen this happen? If you aren’t comfortable posting publicly about a study with this issue, can you DM one to me?

That all sounds like content for a paper rather than a document about the plan for analyzing the data

What do you mean by conceptualization? And how would that lead to incorrect p-values?

Can you give an example of an RCT where this was an issue?

I think it's all costly signaling, but your view may be right

This is great lol

Did you have AI do it?

I agree but to be clear nobody is doing that right now.

Is any of the rest of that necessary for the analysis plan? It needs to be in the paper, but the point of the PAP is to tie your hands on the analysis.

Link to post: jasonkerwin.com/nonparibus/2...

Dear development economists: stop writing a ton of extra stuff in your PAP, and just tell me what analyses you are going to run

Six Questions with Monica Lambon-Quayefio

Link to the full post: blogs.worldbank.org/en/impacteva...

Development Impact is featuring my fantastic coauthor Monica Lambon-Quayefio! Amazing to see all the cool stuff she is working on.

Team "Elsevier Highlights Subversion" strikes again! Best one yet!
@deankarlan.bsky.social, Monica Lambon-Quayefio, Utsav Manjeer, @christopher-udry.bsky.social

Happy birthday to @audreydorelien.bsky.social, the most beautiful & amazing woman I’ve ever met.

I don’t totally believe that but they’re clearly better than the alternatives

Yes, a lower bar for significance is common in A/B tests. I was surprised too! But it makes some sense if the goal is to decide quickly what to invest more in

Also over on Twitter ("X") we are in day two of a debate about the value of RCTs in education research. Lots of value over at the bad place! That's where the people who disagree with me are.

The effects on quality are already about 1/4 as big as those of the main intervention we are studying (results coming pretty soon!). So the potential for improving learning is quite high.

In an A/B test we found that training principals to coach teachers in their schools improved teaching quality by 24 percent! These results are still tentative—but good enough for our implementing partner to keep running this intervention.

Re-upping my advice about how to write a good title and abstract for an academic paper, appropriately called:

"How to Write a Title and Abstract"

Feel free to share this thread, which will focus on titles.

#EconSky #AcademicSky

A fun side note: the seminar is named in honor of the late Paul Theodore Heyne, who was a long-time lecturer at UW. He was one of the undergraduate advisors for Jeff Smith, who was one of my own Ph.D. advisors. Jeff now holds the Paul T. Heyne Distinguished Chair in Economics at (the other) UW.

Why USAID is Great for America and the World - Jason Kerwin Last Thursday I was invited by UW’s Economics Undergraduate Board to give this quarter’s Paul Heyne Seminar—a public lecture about economics aimed at a general audience. My talk was about why USAID is...

Here's a brief blog post about the talk: jasonkerwin.com/nonparibus/2...

And here are the slides: jasonkerwin.com/Slides/Paul_...

Comments welcome!

Last Thursday I gave this quarter's Paul Heyne Seminar. My talk was about why USAID is great for America and also the world, why we should keep it, and how we can make it even better.

A key point is that it is the best brand in the industry, by far.