Re:Decoding the N.F.L. Database to Find 100 Missing Concussions
by danso on Thursday March 24, 2016 @02:24PM
(#11355115)
Just for anyone who was confused as I was...the submitted URL is a sidebar for the massive article markrote links to. That said, I like the submitted URL because it's another example of how bad data (in this case, the database of concussions sloppily put together by the NFL) can be combined with old-fashioned research and reporting (looking up old football game news articles) to find damning evidence.
Or, more broadly speaking, never trust a dataset without corroborating it with your own research.
Or, more broadly speaking, never trust a dataset without corroborating it with your own research.