We finished training the 15 (!) interns from the program, with hosting support by @wellesley.edu , participants from 7 other institutions (!), and in collaboration this year with the World Bank! It was an intense weekend (the interns are still recovering, I think), but they will be great!
The session at #EALE2026 in Barcelona 🇪🇸 is confirmed, with Marie Connolly ( @canadianeconoassoc.bsky.social data editor) and hosted by Joan Llull ( @ecmaeditors.bsky.social data editor). For more details, aeadataeditor.github.io/talks/2026-0....
Sorry, there's no Catalan flag emoji..
Dates and times for #weai2026 are confirmed, June 30 4:30PM and July 1, 8:15 AM! See aeadataeditor.github.io/talks/2026-0... for more details!
I love the pop-ups over the data availability tag! (And thanks to all those RAs of mine who have proof-read the input text that makes it work!)
@paulgp.com you are missing JEL and JEP... they, too, have repositories!
BTW, a very nice repository, and for those looking for a small class project that also teaches "how to get data" - perfectly doable! (Just not "data included")
RFS has some very active data editors, but very recent.You should see some improved contributions soon! review-of-financial-studies.github.io @christophscheuch.bsky.social might have some interest in this.
Believe me, it's hard. I know.
("Lisez-moi" is not...)
Well, I am at fault, for allowing READMEs that are NOT called ... "README". But often, it's the ONLY document in there, and generically has "readme" somewhere in the title. But "READ_me.txt", "Read ME.pdf", "!!!Read This.md" are all accepted (and observed).
My (=AEA) goal would be to make the LLM prompt good enough to allow authors to self-check whether their README is detailed enough. Then my team has less work to do.... and everybody is better off.
@paulgp.com I will be very interested in seeing what your LLM makes of other READMEs (from journals without data editors). And more than happy to try this out (privately) on the first submission by authors (you don't get to see those).
This adds important context to cases where some data are not included, because the first two are usually relatively straightforward to obtain. Categories 3 and 4 of access are the FSRDC and private company datasets of this world.
Graph of access categories and by whether data was shared privately with the data editor.
I plan to release soon (...) cleaned version of internal data, that classifies replication packages by the information on this form www.aeaweb.org/journals/for... which authors provide to us, corrected by my assessment (sometimes). This can generate 5 categories, like this github.com/larsvilhuber...
Data availability: thanks for trying this out on the READMEs. We try to ensure that READMEs contains enough information, but we don't go overboard on editorial suggestions... The IDEAL README has a table like this: github.com/social-scien... which would make it very clear. Try searching for that.
Example: your code has flagged doi.org/10.3886/E239... as not having a README, but it is here: www.openicpsr.org/openicpsr/pr... That's only 1 directory deep, so your algorithm should have found it.
- there are never repositories without README - or there should not be. Any such repository probably slipped through some crack, give me the list, and we'll follow up (please not all at once, we're behind as it is).
Two things:
- there are never papers with analysis that do not have repositories (but there are lagging repositories... working on getting those out - usually missing some sort of click-through thing by the authors, or us)
As the guy whose work is being highlighted here: nice!
Last but not least, I will be training the next AEA Data Editor Summer Interns in April aeadataeditor.github.io/talks/2026-0... If you are an undergraduate at participating colleges, apply! If not... convince somebody at your college to sign-up (contact me!) If you are too old: tough luck!
For these and other talks, AMAs, or workshops, see aeadataeditor.github.io/talks/. There may be a few more coming up.
And not yet fully confirmed, but I plan to be at #EALE2026 in Barcelona 🇪🇸 , together with multiple other data editors aeadataeditor.github.io/talks/2026-0... (stay tuned)
In late June, see you in Denver, Colorado for the @weai.bsky.social #WEAI2026 meetings aeadataeditor.github.io/talks/2026-0... (precise dates are still to be confirmed, and we might mix things up a bit this year, with some new topics)
If French is not your thing, but 🇨🇦 is, come to Vancouver at the #CEA2026 for THAT summer school: aeadataeditor.github.io/talks/2026-0... @ubcvse.bsky.social @sfuecon.bsky.social (again with Marie Connolly, Data Editor for #CJE)
I start off in 🇫🇷🇨🇦 for the "summer" school at @ciqss-qicss.bsky.social (May is NOT summer in Montréal), soyez les bienvenus! aeadataeditor.github.io/talks/2026-0... conjointement avec Marie Connolly (Data Editor for #CJE)
I am firming up spring and summer conference planning, and if you are too, here are a few where I will be leading or contributing to training on reproducibility (for journals and for your own research) 🧵
FWIW, I added my (= AEA Data Editor) weight ⚖️ to argue for removal of unnecessary data collections. Thanks to all those who already reported this. The 🛞 are in motion.
I will continue working with ICPSR to improve the user experience, but you can help by describing your problems via the ICPSR helpdesk ("Contact Us" link at the bottom of the page).
To be clear: you need a REAL email address. ICPSR asks for registration (and I agree with that) so they can record that you agreed to the terms of use. "Free to download" does not mean "free to do anything with it", though we will always ensure that at a minimum, you can reproduce with it!