Histogram vs eCDF - Marc's Blog
brooker.co.za/blog/2022/09/02/ecdf.htmlIt's trivial to transform an eCDF into a histogram, by bucketing up first differences (fn+1 - fn). You can't go from the histogram to the eCDF so easily in general, because bucketing loses data.
Reporting bias makes homeopathy trials look like homeopathy works | Ars Technica
arstechnica.com/science/2022/03/reporting-bias-makes-homeopathy-trials-look-like-they-work/When clinical trials are registered in advance, homeopathy doesn't seem to work.
What is A/A Testing, and Why Should Marketers Care?
instapage.com/blog/what-is-aa-testingAre your A/B testing results entirely accurate? Learn why A/A testing can be beneficial, the pros and cons, and how to avoid false positive test results.
The Tyranny of Metrics by Jerry Z. Muller
goodreads.com/book/show/36644895-the-tyranny-of-metricsThe Tyranny of Metrics book. Read 120 reviews from the world's largest community for readers. How the obsession with quantifying human performance threat...
Violence in Blue | Patrick Ball | Granta Magazine
granta.com/violence-in-blue/‘One-third of all Americans killed by strangers are killed by police.’ Patrick Ball measures the undocumented police killings in the United States.
Scientists rise up against statistical significance
nature.com/articles/d41586-019-00857-9Valentin Amrhein, Sander Greenland, Blake McShane and more than 800 signatories call for an end to hyped claims and the dismissal of possibly crucial effects.