Randomization in such studies is arguably a negative in practice

Refuted papers continue to be cited more than their failed replications

Relating t-statistics and the relative width of confidence intervals

The Dean of Engineering at UofNevada wrote a paper so bad, you can't believe it

Pinker was right, I was wrong

Bad stuff going down at the American Sociological Association

What is a standard error? [pdf]

The immediate victims of a con would rather act as if the con never happened

Clarke's Law, and who's to blame for bad science reporting

Bayesians moving from defense to offense

Russell's Paradox of ghostwriters

Dorothy Bishop on the prevalence of scientific fraud

I disagree with Geoff Hinton regarding "glorified autocomplete"

A successful example of "adversarial collaboration"

Questions about synthetic control methods in causal inference

Why are functional programming languages so popular in the programming languages community? (2018)

Non-Standard Errors [pdf]

What happened with HMOs?

Slides on large language models for statisticians

Point Estimation of Covariance Matrices in Hierarchical Models (2015) [pdf]

Additional recommendations for designing experiments to learn about the world [pdf]

Should we take measurements at an intermediate design point? [pdf]

When can you know the Type I error rate for your analysis

Using predictions from arbitrary models to get tighter confidence intervals

Analog computing and hybrid computing: The view from 1962

ChatGPT4 writes Stan code so I don’t have to

Example of the general issue of the importance of measurement in statistics

They say that stocks go down during the day and up at night

The Undying Holiday-Suicide Myth

Scale transform -50 x*x + 240 x – 7

More →