Statistics and Lies

Ben Goldacre, over at the Bad Science blog, makes a book recommendation. It’s Darrell Huff’s 1954 classic: How to Lie with Statistics. He is prompted to do so by some sloppy journalism in The Daily Telegraph. Goldacre uses the techniques outlined in Huff’s book to show that the conclusions reached in the Telegraph story bear absolutely zero correspondance with truth. There’s a surprise.

 

I’ve already got another of Huff’s books (How to take a chance: The laws of Probability); I’ll definitely be adding How to Lie with Statistics to the library. It will sit between my other Huff book and Mark Monmonier’s How to Lie with Maps as the perfect bridge.