August 9, 2016

Briefly

  • The Australian Census. Comments from the Statistical Society of Australia (“The Statistical Society of Australia is concerned that these changes, brought in with the 2011 Census and repeated in 2016, and that have many potential benefits, have not been handled well”.) and Troy Hunt, a computer security person (“There are some good reasons to question the whole thing, plus some good reasons why it’s really a non-event” ).
  • False positives — Russell Brown writes about the ‘meth contamination’ panic from Housing NZ. “But a dwelling being rendered uninhabitable and needing to be torn apart simply because meth was consumed in it? It didn’t seem possible.
  • There has been a lot of discussion about the basic zoning in the Auckland Unitary Plan. Aaron Schiff has maps of the other stuff — the ‘overlays‘ that restrict development in other ways.
  • There’s a huge difference in style between @realDonaldTrump tweets posted on the iOS and Android Twitter clients. The obvious conclusion is that the Android tweets are real Donald himself, and the iPhone tweets are his staff.
  • An interesting survey measurement issue in estimating religious support for the presidential candidates — ‘evangelical Christian’ is a reasonably well-defined group, but there are big differences in how they are measured in surveys.
avatar

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 »