Find me on bluesky @felixthoemmes.bsky.social Associate Prof @Cornell. Used to live in Germany, Indiana, Arizona, Texas, now settled in Ithaca.
Ithaca, NY
Joined September 2015
- Tweets2.2K
- Following136
- Followers557
- Likes7.7K
Felix Thoemmes retweeted
Somewhere along the way “correlation is not causation” morphed into “until the causal link is proven beyond a reasonable doubt, ideally with a large scale RCT with global external validity, than the correlation likely reflects the impact of some unobserved factor I can’t name.”
Felix Thoemmes retweeted
🤩Scrollytelling with @quarto_pub, closeread extension and #rstats!
Great new possibilities for data-driven stories. Check out my first project: mschnetzer.github.io/scrolly… (on a laptop)
👀 @geokaramanis @R_Graph_Gallery
Felix Thoemmes retweeted
My paper is out in @PNASNews! I replicate a paper on the impact of COVID vaccine mandates on vaccine uptake. Removing a single bad control variable sign-flips several of the paper’s headline results. The reply’s findings are also not robust. 1/x
pnas.org/doi/10.1073/pnas.24…
Felix Thoemmes retweeted
I adore this work. So many little tweaks to my mental model as I read through this thread.
When Double Descent & Benign
Overfitting became a thing, I was a masters student in statistics — and so confused. I couldn't reconcile what l had literally just learned about bias-variance&co with modern ML
Here's what I wish someone had told me then: arxiv.org/abs/2409.18842 1/n
Felix Thoemmes retweeted
Are you interested in perception, culture, & big-data analysis? Join my new lab at Cornell: COCOCO: COrnell COmputational COgnition! Multiple PhD & Postdoc opportunities available: norijacoby.com/cococo.html. Please retweet & spread the word! @CO3lab 1/6
Felix Thoemmes retweeted
When Double Descent & Benign
Overfitting became a thing, I was a masters student in statistics — and so confused. I couldn't reconcile what l had literally just learned about bias-variance&co with modern ML
Here's what I wish someone had told me then: arxiv.org/abs/2409.18842 1/n
Felix Thoemmes retweeted
No defiers is a required assumption for identification of LATE in IV analysis. If that is false, the best thing you can do is to calculate bounds for the LATE.
But what happens if limit the maximum proportion of defiers. For instance, what if I don't believe is > 10%?
Felix Thoemmes retweeted
Excited about modeling intensive longitudinal & passive sensing data? Join our dynamic research team at UW-Madison as a postdoc! 🧠📊 Flexible start date, and applications are reviewed on a rolling basis. centerhealthyminds.org/asset…
Felix Thoemmes retweeted
IPWRA is great; screenshot from Imbens (2004) review article, py and R implementations.
Don't be beholden to statacorp; write in a programming language that lets you implement the math almost verbatim
Felix Thoemmes retweeted
The Social Psychology in Legal Contexts Preconference returns to #SPSP2025! Talks from Saul Kassin, @kurtjgray Kyle Scherr, @JackieMChen and more!
Please RT!
Poster & data blitz submissions due 10/17 !
ow.ly/ZJ3L50TwsVG
@SPSPnews @SPSPGSC @APLSsc @APLS41 @margem_
Felix Thoemmes retweeted
If you preregister a wild study, and your collaborating on THAT study, then the resulting paper should focus on that wild study. You preregistered it, and you collaborated on it.
Nothing stopping you writing a paper on the discovery but that is EXPLORATOARY & not preregistered.
You have a friend who had a wild idea and, being an open minded person, you agree to collaborate on it (preregistering the wild hypothesis). Along the way, you discover something completely different that is of real scientific interest.
Which paper is better to write?
12%One focused on crazy idea
56%One focused on discovery
32%[Show me the results]
766 votes • Final results Felix Thoemmes retweeted
🚀 ANNOUNCING: The release of my free and open e-book, "Applied #MachineLearning in Python: A Hands-on Guide with Code"! 🎉
🔗 Access it here: geostatsguy.github.io/Machin…
📚 16 chapters packed with well-documented #Python workflows, open-source code, and datasets — with direct links to my YouTube lectures for deeper learning! This e-book blends theory with practical coding and effective visualizations to help you understand and apply #MachineLearning concepts.
⚙️ Try the examples, explore the code, and follow along with hands-on exercises. It’s a living document, so I’ll continue adding new chapters and improving content over time.
I'm here to help you dive into #DataScience and #ML — let’s learn together! I'm having an awesome sabbatical! Stoked!
Felix Thoemmes retweeted
The decline effect is the mechanical consequence of applying a threshold to p-values for claiming discovery when initial studies tend to be too small (which we learn with hindsight). Initial well powered studies won't show that. No recourse to woo required. That is all. #subtweet
Felix Thoemmes retweeted
Another shoe just dropped in the ongoing Dan Ariely scandal: JMR, a top 4 marketing journal, has issued a formal “Expression of Concern” for Mazar, Amir, & Ariely (2008), the infamous Ten Commandments study
🧵
Felix Thoemmes retweeted
I'm currently on sabbatical, keeping a low profile—so well-hidden that no one can find me! In reality, I’m in my office on campus, working long hours to complete my "Applied #MachineLearning in #Python" e-book.
I love the e-book format because it allows for live, interactive illustrations with repeatable code, is fully updateable, and accessible to everyone. It’s the perfect way to share the years of work I’ve put into developing educational content for my students.
The positive feedback from my recently released "Applied Geostatistics in Python" e-book (geostatsguy.github.io/Geosta…) has been incredibly motivating! Stay tuned—the new e-book will be out soon! Stoked!
Felix Thoemmes retweeted
I am excited to announce the release of the first edition of my e-book, "Applied Geostatistics in Python: A Hands-on Guide with GeostatsPy".
This e-book is designed to support students & professionals learning and applying #geostatistics / #spatial #DataAnalytics through 25 chapters with well-documented, demonstration workflows that help you understand theory and apply best-practice all in modern, flexible #Python codes. By expanding from my GitHub repositories to an e-book, I aim to reach a broader audience with a dynamic and evolving educational resource.
This living document will continuously grow in response to your feedback and needs. Also, it is integrated with my online lectures, interactive dashboards, and workflows, all based on my #opensource Python package GeostatsPy and other common open-source packages. Accessible and actionable educational content!
Check it out @ geostatsguy.github.io/Geosta… ∀. #DataScience
Felix Thoemmes retweeted
If you or your students are interested in studying measurement and quant methods I will be reviewing grad applications for the Fall 2025 admission. Some info about applying: jessicakayflake.com/join-the…
If you have any questions ask them here so I can answer them and update the FAQ!
Felix Thoemmes retweeted
Extremely excited to share that I will be joining @Cornell as an Assistant Professor of Psychology in Fall 2025!
I’m just so thrilled to be joining such a lively department, and the thriving moral psych/philosophy community there.
Felix Thoemmes retweeted
AI is gonna forever change the way we disseminate research. Here's a NotebookLM-generated podcast on my 2022 AER on racial discrimination in bail decisions
drive.google.com/file/d/1HOB…
This took <5min to generate from uploading the PDF
Pretty cool!
