Pinned
magdalena bennett
magdalena bennett
@maibennett.com
Research Scientist working on causal inference. The Anna Wintour of slide decks. If you have something to say, say it with a nice plot #rstats (Disclaimer: All opinions are my own)
2.7K followers327 following333 posts
magdalena bennett reposted
Our department at Vanderbilt is seeking an assistant/associate professor of higher education policy. The position is intended for a quantitative researcher with a PhD in a social science, public policy, or education. Please circulate and let me know if you have questions! apply.interfolio.com/193891
magdalena bennett reposted
I’m excited to share a new EdWorkingPaper with @andydebarros.bsky.social. Evaluations are an increasingly common way school systems globally are trying to help teachers improve, but is more always better? A 🧵 doi.org/10.26300/ms3...
magdalena bennett reposted
magdalena bennett reposted
New paper alert! “Heritability and social interaction in labor market outcomes: Evidence from the population of twins in the United States,” with David Card and Moises Yi, out today in PNAS. www.pnas.org/doi/10.1073/...
magdalena bennett reposted
I didn’t even know this was a thing until I started teaching, and yes, apparently they “remember the name” (or so they told me). I just think that (a) they never go back to the file or (b) redownload it 1000 times.
ok so you know how we all get like that overwhelming horror when we consider how many people in applied quant classes do not know what a file directory is or what a filepath is or etc. my question is, how do these people ordinarily retrieve saved files. search? like they remember the filenames?
magdalena bennett reposted
Forthcoming in the AER: "Difference-in-Differences with a Continuous Treatment" by Brantly Callaway, Andrew Goodman-Bacon, and Pedro H. C. Sant'Anna.
magdalena bennett reposted
Brown just opened a cluster hire for two tenure-track positions in Education Policy. Come join our vibrant community of scholars at @brownedudept.bsky.social & @annenberginstitute.bsky.social! apply.interfolio.com/190527 Please help spread the word.
magdalena bennett reposted
magdalena bennett reposted
Validated against contemporaneous responses, a survey of 916 trainees and their career intentions finds a –22 percentage point change in academia intentions and a US-departure shift concentrated among foreign-born scientists, from Azoulay, Sadun, and Scur www.nber.org/papers/w35330
magdalena bennett reposted
New NBER working paper. We meta-analyze 82 RCTs of low-cost parent programs in 20+ countries. We use data on unwritten RCTs from funder records, RCT registries, author queries, etc. to estimate a model that adjusts for publication bias and characterizes the effect distribution for unwritten studies.
magdalena bennett reposted
New research from Card, Rothstein, and Yi finds no evidence that Black workers live farther from “good jobs” than White workers. Proximity doesn’t explain the racial earnings gap. What other factors may be at play? Learn more in @lisacamnermckay.bsky.social 's latest piece: https://bit.ly/4aDA5YZ
magdalena bennett reposted
magdalena bennett reposted
wow — the preprint host, arxiv, is banning authors for a year if they submit papers with hallucinated citations 🤖
I find this unsettling: A new study in #Lancet shows that the number of faux citations in scientific articles is on the rise, and #AI is likely to blame. @aniloza.bsky.social reports. www.statnews.com/2026/05/07/l...
magdalena bennett reposted
Quarto 1.9 brings Typst even closer to LaTeX parity with 📖 Typst Books, ✍️ margins for figures, tables, and citations via the Marginalia package, and 🛜 new typst-gather tool for reliable, offline rendering. Details here: quarto.org/docs/blog/po...
magdalena bennett reposted
Simon Johnson and I are hiring a postdoc at MIT to research the history, social implications, and future of technology. The job description and application link are available here: shapingwork.mit.edu/careers/jame...
We're hiring a postdoc at MIT! The Postdoctoral Associate will work closely with @dacemoglumit.bsky.social and @simonhrjohnson.bsky.social to conduct research on the history, social implications and future of technology. Learn more and apply by June 15: shapingwork.mit.edu/careers/jame...
magdalena bennett reposted
I'd love to learn more about how y'all are using AI for data science, so I've set up some times where you can chat with me: scheduler.zoom.us/hadley-wickh.... I'm interested in it all, whether it's your struggles, your successes, or what you're still trying to get your head around. #rstats
Weezer is one of those bands that is all bangers, no skips. Yes, even the Teal Album, you haters. youtu.be/okthJIVbi6g?...
magdalena bennett reposted
magdalena bennett reposted
Four small Quarto extensions to sharpen your Reveal.js slides: animated list fragments, step-by-step code annotations, auto-repeating headings, and keyboard-driven tabsets. Each one fixes a single friction point. 🎯 mickael.canouil.fr/posts/2026-0... #Quarto #QuartoPub #SlideCrafting #RevealJS
As someone who has been on both sides, there are clear tradeoffs, but industry might be a much better deal for most people’s preferences (though ymmv). I miss some things of academia, but I’ve learned a ton in the past year, get (almost) unlimited access to a lot of new tools, 1/2
Getting a PhD and then working outside academia is dope. My advisees who’ve done that are more or less uniformly happy and feel that they’re using the skills they trained on. (I talked to one today who just landed a great gig). Just wanted to convey that to anyone it might benefit.
magdalena bennett reposted
1/ "Silicon samples" are becoming more and more common in research and polling. One problem: depending on the analytic decisions made, you can basically get these samples to show any effect you want. The updated version of this preprint is now online! THREAD🧵 arxiv.org/abs/2509.13397
Can large language models stand in for human participants? Many social scientists seem to think so, and are already using "silicon samples" in research. One problem: depending on the analytic decisions made, you can basically get these samples to show any effect you want. THREAD 🧵