January 26, 2018


  • “Are bigoted, irrational robots hurting humans?” (Herald, from the Daily Mail).
  • From 538: ” If I had a magic wand, I’d develop an algorithm that: a) draws the shortest possible line(s) necessary to split a state into equally populous districts and b) requires that only as many pre-existing jurisdictions be split as necessary to achieve equally populous districts.” That’s the easy part. Getting a fair voting system isn’t primarily difficult because math is hard, but because power corrupts.
  • Sharon Begley at STAT on the hype around exercise and dementia “But we thought it was a reasonable thing to say, especially since [exercise] doesn’t have a lot of risk. To be honest, we’re looking for something positive to tell people.”
  • Peter Ellis writes on how to recruit data scientists in the public sector
  • There’s a famous medical paper reviewing all the randomised trials of parachute use when jumping from a plane (the point being, of course, that there aren’t any, and shouldn’t be). There’s a new medical paper looking at people using this analogy. Of 35 interventions where the parachute analogy was used explicitly to argue that randomised trials were impossible and there was no room for doubt about effectiveness , 22 had been examined in randomised trials. Six had been shown effective.

Thomas Lumley (@tslumley) is Professor of Biostatistics at the University of Auckland. His research interests include semiparametric models, survey sampling, statistical computing, foundations of statistics, and whatever methodological problems his medical collaborators come up with. He also blogs at Biased and Inefficient See all posts by Thomas Lumley »

Add a comment

First time commenting? Please use your real first name and surname and read the Comment Policy.