Pig Data Longitudinal Analysis: Size Doesn't Matter
via http://ma.tt/2014/03/statistical-validity-in-big-data
For each person George Gallup’s pollsters interviewed, The Literary Digest received 800 responses. All that gave them for their pains was a very precise estimate of the wrong answer.
http://www.ft.com/cms/s/2/21a6e7d8-b479-11e3-a09a-00144feabdc0.html
More on big pigness: http://businesses.entrechat.org/2014/03/big-data-vs-pig-data/
nmw 10:43:19 on 2014/04/01 Permalink
ps: note that the “Google Translate” example glosses over the fact that the data set was probably limited to translations performed by certified translators, and did not include the stuff that relatively ordinary schmucks translated. 😉