Blog · corpus.blog/blogs/statmodeling.stat.columbia.edu/posts
Andrew Gelman
statmodeling.stat.columbia.edu
2023
Debate over effect of reduced prosecutions on urban homicides; also larger questions about synthetic control methods in causal inference.original ↗
12 Oct 2023
Rubric-based holistic word salad represents a change from traditional approaches to publishing papers in physicsoriginal ↗
10 Oct 2023
“What role will hydrological science play in the age of machine learning?” I think it will depend on the problem you’re trying to solve.original ↗
7 Oct 2023
No, this paper on strip clubs and sex crimes was never gonna get retracted. Also, a reminder of the importance of data quality, and a reflection on why researchers often think it’s just fine to publish papers using bad data under the mistaken belief that these analyses are “conservative” or “attenuated” or something like that.original ↗
5 Oct 2023
You can learn a lot from reanalyzing data (“Signing at the beginning . . .” example, but also many others)original ↗
4 Oct 2023
The connection between the psychological concept of “generic language” and the problem of overgeneralization from research studiesoriginal ↗
3 Oct 2023
3 Oct 2023
Beneath every application of causal inference to ML lies a ridiculously hard social science problemoriginal ↗
2 Oct 2023
How did some of this goofy psychology research become so popular? I think it’s a form of transubstantiation.original ↗
2 Oct 2023
29 Sept 2023
28 Sept 2023
“We live in a time of mad kings, megalomaniacal sociopaths granted dangerous power through wealth and/or political position, prone to wild schemes of empire and grandeur”original ↗
27 Sept 2023
How to teach certain tricky things such as the difference between standard deviation and standard errororiginal ↗
26 Sept 2023
26 Sept 2023
(again) why I don’t like to talk so much about “p-hacking.” But sometimes the term is appropriate!original ↗
23 Sept 2023
West Point, like major league baseball, was purely based on merit and achievement before Jackie Robinson and that Tuskegee Airman guy came along and messed everything up.original ↗
22 Sept 2023
My SciML Webinar next week (28 Sep): Multiscale generalized Hamiltonian Monte Carlo with delayed rejectionoriginal ↗
21 Sept 2023
A message to Parkinson’s Disease researchers: Design a study to distinguish between these two competing explanations of the fact that the incidence of Parkinson’s is lower among smokersoriginal ↗
20 Sept 2023
How to quit smoking, and a challenge to currently-standard individualistic theories in social scienceoriginal ↗
19 Sept 2023
18 Sept 2023
“Are there clear examples of the opposite idea, where four visually similar visualizations can have vastly different numerical stats?”original ↗
18 Sept 2023
Harvard law prof sez: “I believe that if [universities] are going to accept blood money . . . the should only ever accept that money anonymously.”original ↗
17 Sept 2023
Evaluating Visualizations for Inference and Decision-Making (Jessica Hullman’s talk in the Columbia statistics seminar next Monday)original ↗
14 Sept 2023
Better Than Difference in Differences (my talk for the Online Causal Inference Seminar Tues 19 Sept)original ↗
14 Sept 2023
13 Sept 2023
The authors of research papers have no obligation to share their data and code, and I have no obligation to believe anything they write.original ↗
10 Sept 2023
9 Sept 2023
When I said, “judge this post on its merits, not based on my qualifications,” was this anti-Bayesian? Also a story about lost urine.original ↗
8 Sept 2023
5 Sept 2023
“Latest observational study shows moderate drinking associated with a very slightly lower mortality rate”original ↗
2 Sept 2023
The variation-ignoring junk science that’s promoted by the Association for Psychological Science and related academic celebrities. It’s like a poker player thinking: “okay, if push all in from the button I’ll win 3.6 big blinds each and every time.”original ↗
31 Aug 2023
My two courses this fall: “Applied Regression and Causal Inference” and “Communicating Data and Statistics”original ↗
30 Aug 2023
Some Open Questions in Statistics (my talk this Fri, 1 Sep 2023, at the University of Michigan)original ↗
29 Aug 2023
Differences between boy and girl dinosaurs: Going beyond p-values and dichotomous thinkingoriginal ↗
27 Aug 2023
24 Aug 2023
Editorial processes and ethics approval — Case study of 248 studies with the same ethics approval numberoriginal ↗
23 Aug 2023
22 Aug 2023
21 Aug 2023
17 Aug 2023
16 Aug 2023
15 Aug 2023
NYT does some rewriting without attribution. I guess this is standard in journalism but it seems unethical to me.original ↗
12 Aug 2023
Freakonomics and global warming: What happens to a team of “rogues” when there is no longer a stable center to push against? (a general problem with edgelords)original ↗
8 Aug 2023
7 Aug 2023
Unifying Design-Based and Model-Based Sampling Inference (my talk this Wednesday morning at the Joint Statistical Meetings in Toronto)original ↗
6 Aug 2023
5 Aug 2023
Fully Bayesian computing: Don’t collapse the wavefunction until it’s absolutely necessary.original ↗
4 Aug 2023
Statistical mistake in published paper is fixed. Correction is issued. Original paper still sitting online with the wrong findings. Press release still sitting online with the wrong findings. Elsevier!original ↗
3 Aug 2023
“They got a result they liked, and didn’t want to think about the data.” (A fish story related to Cannery Row)original ↗
2 Aug 2023
Studying average associations between income and survey responses on happiness: Be careful about deterministic and causal interpretations that are not supported by these data.original ↗
1 Aug 2023
In the real world people have goals and beliefs. In a controlled experiment, you have to endow themoriginal ↗
31 Jul 2023