Benjamin Egerod

How much data do you need to conduct an informative staggered diff-in-diff? In our new working paper @fhollenbach and I simulate the power of #DiD estimators, and find that you might need *a lot*, even to detect large effects. We also provide suggestions for improving power 1/

See our No-Spin report on the Baby’s First Years RCT, studying the effects of providing unconditional cash transfers to low-income, new mothers in the US. In brief: This high-quality RCT found no impact on any of the study’s primary child development outcomes at 4y follow-up.🧵

🚨 New paper in @ScienceAdvances Can changing how we argue about politics online improve the quality of replies we get? @THeideJorgensen, @a_rasmussen, and I use an LLM to manipulate counter-arguments to see how people respond to different approaches to arguments. Thread 🧵1/n

Excited that our (@Apoorva__Lal, @xuyiqing, @zu_gary) paper won the Political Analysis' 2024 Editor's Choice award! It was really a lot of work (we started this in 2018!), so nice to see we've had some impact on the field. It's also open access. cambridge.org/core/journals/…

This looks like a must-read for those who test multiple hypotheses. Beyond Bonferroni: Hierarchical Multiple Testing in Empirical Research "standard approaches [for correcting for multiple hypotheses] typically ignore hierarchical relationships among hypotheses—structures that…

#FirstView from @polanalysis - Estimating the Local Average Treatment Effect Without the Exclusion Restriction - cup.org/4fjAykx - Zachary Markovich

There’s so much interesting research coming from @VandyPoliSci. As dept chair, it is a pleasure to highlight a bunch of it. First, up, @trounstine and @SarahAnzia in the @apsrjournal on civil service adoption: cambridge.org/core/journals/…

Again: *vast* majority of university researchers are trying to cure cancer, understand inflation, see why black holes really exist, trace the Bantu migration, etc. But we're being screwed by small number of folks who think public role of the university is something else. 11/11

Our article introducing CampaignView is now officially online! nature.com/articles/s4159…

We are doing it again! Please submit your papers and come see us in Lisbon :)

💸 How should scholars measure early money in campaigns? Our new @PSRMJournal article in FirstView (w/ @colinrcase) offers a theoretical and empirical framework, as well as practical guidance for working with FEC data 🔗doi.org/10.1017/psrm.2…

🚨 New paper with Jie Ma and @keithschnak: Democratic Accountability with Citizen Coproduction (Link in next tweet) As always: read it, cite it, love it

This looks a must read for those interested in triple differences (DDD) designs. It shows that the standard way many estimate DDD is wrong. It shows why and gives you tools that work with covariates and staggered adoption. Bonus: free R package.

📢 Thrilled to share our new article introducing CampaignView—a comprehensive open-source dataset of congressional candidate campaign bios and policy platforms (2018–2022). Paper + data here: campaignview.org & doi.org/10.7910/DVN/JI… 🧵1/4

Genes predict children's cognitive abilities. But an RCT of preschool + better nutrition erases any "genetic disadvantage"

New study shows that unfettered access to ChatGPT can harm student learning. Without guardrails, students use GPT-4 as a “crutch” during practice problem sessions, and subsequently perform worse on their own. But (!) these negative learning effects are largely mitigated by…

Sektordimensioneringen vil formentlig øge negativ social arv, men ikke få flere til at søge velfærdsuddannelserne, viser analyse fra @TaenketankenDEA. Sektordimensioneringen vil helt sikkert koste erhvervslivet dyrt i form af færre ingeniører, it-specialister og økonomer

"Floods", by @dev_a_patel, is one of the more astounding Big Data econ papers I have ever seen. Machine learning can be used on satellite data to essentially do magic. How damaging are floods? How much do they affect human capital attainment? A thread on the paper 1/

Really cool paper, which was unfortunately never published due to ethical concerns. Thompson and Hanley found that creating new Wikipedia articles changed the content of scientific papers. As a way of disseminating information, it is ludicrously cost effective.

My new paper asks, "theorizing after looking at the data is bad, right?" Actually, post-hoc theorizing might be OPTIMAL in the modern era of huge datasets and mature theories. 🧵

Is economics a sandwich? No Is econometrics a sandwich?...