Blog · corpus.blog/blogs/statmodeling.stat.columbia.edu/posts
Andrew Gelman
statmodeling.stat.columbia.edu
2023
30 Jul 2023
We want to go beyond intent-to-treat analysis here, but we can’t. Why? Because of this: “Will the data collected for your study be made available to others?” “No”; “Would you like to offer context for your decision?” “–“. Millions of taxpayer dollars spent, and we don’t get to see the data.original ↗
29 Jul 2023
Causal inference and the aggregation of micro effects into macro effects: The effects of wages on employmentoriginal ↗
28 Jul 2023
24 Jul 2023
23 Jul 2023
Problem with the University of Wisconsin’s Area Deprivation Index. And, no, face validity is not “the weakest of all possible arguments.”original ↗
22 Jul 2023
20 Jul 2023
19 Jul 2023
When your regression model has interactions, do you need to include all the corresponding main effects?original ↗
18 Jul 2023
What data to include in an analysis? Not always such an easy question. (Elliott Morris / Nate Silver / Rasmussen polls edition)original ↗
17 Jul 2023
Why is every action hero named Jack, John, James, or, occasionally, Jason, but never Bill, Bob, or David?original ↗
11 Jul 2023
10 Jul 2023
Here are the data from that cold showers study. So you haters can now do your own analyses!original ↗
10 Jul 2023
Would you allow a gun to be fired at your head contingent on a mere 16 consecutive misfires, whatever the other inconclusive evidence?original ↗
9 Jul 2023
Before reading this post, take a cold shower: A Stanford professor says it’s “great training for the mind”!original ↗
8 Jul 2023
“The Role of Doubt in Conceiving Research.” The capacity to be upset, to recognize anomalies for what they are, and to track them down and figure out what in our understanding is lacking:original ↗
6 Jul 2023
4 Jul 2023
Welcome to the grand opening of the Alexey Guzey Sleep Center at the University of California!original ↗
3 Jul 2023
Instructors are trying their best but they are busy and often don’t themselves understand statistics so well. How can they avoid using bad examples, or does this even matter?original ↗
2 Jul 2023
If the Nobel Prize winners and the bigshots at the National Cancer Institute are doing hype on the daily, then it makes sense that the loser wannabe bigshots at the Cleveland Clinic and the Scripps Translational Science Institute will imitate that behavior.original ↗
29 Jun 2023
They came in through the window: The migration of tech hype from the fringes to the media and academic mainstreamoriginal ↗
27 Jun 2023
Variation over time or across scenarios is more relevant than within-sample standard errorsoriginal ↗
26 Jun 2023
Here are some ways of making your study replicable. (No, the first steps are not preregistration or increasing the sample size!)original ↗
22 Jun 2023
Ted-talking purveyors of fake data who write books about lying and rule-breaking . . . what’s up with that?original ↗
21 Jun 2023
It’s Harvard time, baby: That thing where well-connected policy advocates can’t handle criticism and call it “Stasi” or “censorship”original ↗
17 Jun 2023
16 Jun 2023
15 Jun 2023
Explaining the horribly wasteful U.S. heath care system as a combination of rich-countries-spend-more-on-health-care and diminishing-returns-to-health-care-spendingoriginal ↗
14 Jun 2023
12 Jun 2023
11 Jun 2023
Joe Simmons, Leif Nelson, and Uri Simonsohn agree with us regarding the much publicized but implausible and unsubstantiated claims of huge effects from nudge interventionsoriginal ↗
11 Jun 2023
10 Jun 2023
9 Jun 2023
Junk science: It’s not just for Ted talks, NPR features, and airport bestsellers. It can also ruin people’s lives.original ↗
3 Jun 2023
Before data analysis: Additional recommendations for designing experiments to learn about the worldoriginal ↗
1 Jun 2023
31 May 2023
“The narrow beach between the continent of clear effects and the sea of confusion”: that sweet spot between signals that are so clear you don’t need statistics, and signals that are so weak that statistical analysis can’t helporiginal ↗
30 May 2023
Maybe Paul Samuelson and his coauthors should’ve spent less time on dominance games and “boss moves” and more time actually looking out at the world that they were purportedly describing.original ↗
27 May 2023
“If you do not know what you would have done under all possible scenarios, then you cannot know the Type I error rate for your analysis.”original ↗
20 May 2023
18 May 2023