March 15, 2016
Briefly
- How many people are killed by police in the USA? Comes with a description of the statistical technique of mark-recapture estimation, from Granta magazine.
- Another example of cheating being exposed by surveillance and data collection — this time in contract bridge. From the New Yorker.
- If you want views on the `Kiwimeter’ survey, I suggest Tze Ming Mok’s piece at Public Address.
- The Ombudsman has released guidelines on Official Information Act requests through social media (PDF). Summary: it’s still a question, so it still gets answered.
- From NiemanLabs, how some news publishers are doing interactive graphics for mobile devices
And from XKCD: how much of various fluids does the US consume, using pipeline diameters to illustrate
(Update: Yes, I realise this is the sort of bubble plot we usually say mean things about. Not the point, here).
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 »
I regularly blog on how many people are killed by police in the USA.
The Guardian, and to a lesser extent the Washington Post databases include in the same total a man shot dead attempting to kidnap two children while stealing a police cruiser and a man run over by a police cruiser that was looking for him at night.
I do not really think you are adding up the same variables when you include people shooting at police with unarmed people
8 years ago
I’d agree with that, but there are also quite a lot of cases where ‘unarmed’ vs ‘shooting at police’ (or at least, plausibly threatening to do so) is exactly the point of controversy.
8 years ago