Statistical Modeling (Andrew Gelman) statmodeling.stat.columbia.edu

Genre fiction: Some genres are cumulative and some are not.

statmodeling.stat.columbia.edu·69w

The theory crisis in physics compared to the replication crisis in social science: Two different opinion-field inversions that differ in some important ways

statmodeling.stat.columbia.edu·69w·Hacker News

String theory wars: An opinion-field inversion.

statmodeling.stat.columbia.edu·69w

Muckraking at the University of Oregon

statmodeling.stat.columbia.edu·69w

“The king, sir, is much better!”

statmodeling.stat.columbia.edu·69w

Postdoc, doctoral student, and summer intern positions, Bayesian methods, Aalto

statmodeling.stat.columbia.edu·69w

I would add three words to this statement by Uri Simonsohn on preregistration

statmodeling.stat.columbia.edu·70w

Junk science becomes more professionalized. Meanwhile, conspiracy theories are being more associated with the center-right and right, politically. How does al...

statmodeling.stat.columbia.edu·70w

Treasure trove of forensic details in arXiv’s LaTeX source code

statmodeling.stat.columbia.edu·70w

Truth is more realistic than fiction, and what this tells us about odious thought experiments

statmodeling.stat.columbia.edu·70w

Progress in 2024 (Aki)

statmodeling.stat.columbia.edu·70w

This looks like an excellent new business line for Wolfram Research!

statmodeling.stat.columbia.edu·70w

“I would have had a simple piece of advice: Say nothing.”

statmodeling.stat.columbia.edu·70w

What are my goals? What are their goals? (How to prepare for that meeting.)

statmodeling.stat.columbia.edu·70w

Echoing Eco: From the logic of stories to posterior predictive simulation

statmodeling.stat.columbia.edu·70w

Progress in 2024 (Jessica)

statmodeling.stat.columbia.edu·71w

Advice for weighting the results of conjoint analyses/experiments

statmodeling.stat.columbia.edu·71w

Newly published in 2024

statmodeling.stat.columbia.edu·71w

Calibration “resolves” epistemic uncertainty by giving predictions that are indistinguishable from the true probabilities. Why is this still unsatisfying?

statmodeling.stat.columbia.edu·71w

What is the minimum bloggable contribution?

statmodeling.stat.columbia.edu·71w

Log in to enable infinite scrolling