Blog · corpus.blog/blogs/statmodeling.stat.columbia.edu/posts
Andrew Gelman
statmodeling.stat.columbia.edu
2018
19 Mar 2018
17 Mar 2018
You need 16 times the sample size to estimate an interaction than to estimate a main effectoriginal ↗
15 Mar 2018
Reasons for an optimistic take on science: there are not “growing problems with research and publication practices.” Rather, there have been, and continue to be, huge problems with research and publication practices, but we’ve made progress in recognizing these problems.original ↗
14 Mar 2018
I fear that many people are drawing the wrong lessons from the Wansink saga, focusing on procedural issues such as “p-hacking” rather than scientifically more important concerns about empty theory and hopelessly noisy data. If your theory is weak and your data are noisy, all the preregistration in the world won’t save you.original ↗
13 Mar 2018
12 Mar 2018
11 Mar 2018
Incorporating Bayes factor into my understanding of scientific information and the replication crisisoriginal ↗
10 Mar 2018
9 Mar 2018
Important statistical theory research project! Perfect for the stat grad students (or ambitious undergrads) out there.original ↗
7 Mar 2018
No, I don’t believe that “Reduction in Firearm Injuries during NRA Annual Conventions” storyoriginal ↗
5 Mar 2018
3 Mar 2018
2 Mar 2018
I fear that many people are drawing the wrong lessons from the Wansink saga, focusing on procedural issues such as “p-hacking” rather than scientifically more important concerns about empty theory and hopelessly noisy data. If your theory is weak and your data are noisy, all the preregistration in the world won’t save you.original ↗
28 Feb 2018
Pizzagate: The problem’s not with the multiple analyses, it’s with the selective reporting of results (and with low-quality measurements and lack of quality control all over, but that’s not the key part of the story)original ↗
27 Feb 2018
I’ll use this line in my talk this Wednesday at the Society for Research on Educational Effectivnessoriginal ↗
27 Feb 2018
The p-curve, p-uniform, and Hedges (1984) methods for meta-analysis under selection bias: An exchange with Blake McShane, Uri Simonsohn, and Marcel van Assenoriginal ↗
26 Feb 2018
25 Feb 2018
Rasmussen and Williams never said that Gaussian processes resolve the problem of overfittingoriginal ↗
24 Feb 2018
“Have Smartphones Destroyed a Generation?” and “The Narcissism Epidemic”: How can we think about the evidence?original ↗
22 Feb 2018
21 Feb 2018
The graphs tell the story. Now I want to fit this story into a bigger framework so it all makes sense again.original ↗
20 Feb 2018
18 Feb 2018
17 Feb 2018
17 Feb 2018
Research project in London and Chicago to develop and fit hierarchical models for development economics in Stan!original ↗
15 Feb 2018
Use multilevel modeling to correct for the “winner’s curse” arising from selection of successful experimental resultsoriginal ↗
14 Feb 2018
13 Feb 2018
Hysteresis corner: “These mistakes and omissions do not change the general conclusion of the paper . . .”original ↗
12 Feb 2018
10 Feb 2018
“No System is Perfect: Understanding How Registration-Based Editorial Processes Affect Reproducibility and Investment in Research Quality”original ↗
9 Feb 2018
What’s Wrong with “Evidence-Based Medicine” and How Can We Do Better? (My talk at the University of Michigan Friday 2pm)original ↗
8 Feb 2018
Methodological terrorism. For reals. (How to deal with “what we don’t know” in missing-data imputation.)original ↗
7 Feb 2018
5 Feb 2018
Snappy Titles: Deterministic claims increase the probability of getting a paper published in a psychology journaloriginal ↗
4 Feb 2018
2 Feb 2018
When to add a feature to Stan? The recurring issue of the compound declare-distribute statementoriginal ↗
1 Feb 2018
Postdoc opening on subgroup analysis and risk-benefit analysis at Merck pharmaceuticals research laboriginal ↗
30 Jan 2018
My suggested project for the MIT Better Science Ideathon: assessing the reasonableness of missing-data imputations.original ↗
29 Jan 2018
Looking at all possible comparisons at once: It’s not “overfitting” if you put it in a multilevel modeloriginal ↗
28 Jan 2018
27 Jan 2018
27 Jan 2018
24 Jan 2018
To better enable others to avoid being misled when trying to learn from observations, I promise not be transparent, open, sincere nor honest?original ↗
23 Jan 2018
22 Jan 2018
How to get a sense of Type M and type S errors in neonatology, where trials are often very small? Try fake-data simulation!original ↗
20 Jan 2018
19 Jan 2018
A lesson from the Charles Armstrong plagiarism scandal: Separation of the judicial and the executive functionsoriginal ↗
19 Jan 2018