Statistics

Statistics is a way to lie to people or, at the very least, confuse them enough about a topic with have the math backing us up so that it makes us sound like we know what we’re talking about. At least that’s how unscrupulous people who wants to push an agenda define statistics. Instead, statistics is the study of patterns in a set of data and how those patterns can be interpreted or displayed to help us gain an understanding of what the data really represents. Speaking of patterns in data, how about Abraham Wald’s Memo from WWII?

Seems the Allies were trying to analyze the damage to planes coming back from missions over Germany to see if they could find a pattern in the damage and therefore armor against that damage pattern. Wald’s statistical analysis of the damage pattern on planes led them to armor specific areas of the planes and greatly, like saved countless lives greatly, improved the durability of the planes. Want some statistical weirdness? There’s nearly a perfectly linear relationship between per capita cheese consumption and the number of people who die by becoming tangled in their bedsheets. What about when statistics sent an innocent women to jail?

Lucia de Berk, a nurse, was sentenced to jail for the murder and attempted murder of children in the hospital where she worked based solely on the fact that there was a 1 in 342 million chance that one nurse’s schedule could randomly have lined up to have that nurse working on the nights of all the crimes. Lucia had to have been the culprit because it was so statistically unlikely for there to be another person with the same access to the victims. There was no physical evidence nor eye-witness testimony. She was later cleared when the real culprit was identified and confessed. Just so we’re clear, no decent court should ever convict someone solely on statistical evidence.