March 28, 2014

Briefly

By

Reader request edition

  • Margin of error. The Herald has a reasonable story on public opinion about Labour’s baby-bonus plan. It would have been good to say what the margin of error was for the difference between men and women, since that was the headline. If the gender split was about 50:50 the margin of error for that difference is going to be just over 7%, and the observed difference was 8%. The headline is ok, but that’s the sort of calculation someone should have done, and having done, should have reported. We already have doubts about this particular poll, though. (via @danylmc)

 

  • The New Republic thinks sunglasses make you less moral and advises against them, based on the fact that masks are used for anonymity and that undergraduates in a psych experiment gave away an average of 90c less when they were wearing sunglasses.  The data on benefits of sunglasses are somewhat better founded. (via @juha_saarinen)
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 »