CALLING BULLSHIT: The Art of Skepticism in a Data-Driven World | Carl Bergstrom & Jevin West, Random House, p318. An evolutionary biologist and data scientist who teaches a college course similar to the book teamed up to pen a tour-de-force of how data misrepresentations fill our daily content intake and how to spot them. Moreover, they want you to call it “bullshit,” mainly when perpetrated in bad faith (which they encourage the reader not to assume). They take the reader through common errors: false equivalencies, confirmation bias, selection bias, visual data misrepresentation (bar and line graphs), and how the steady stream of BS obscures real understanding. They pull dozens of examples of each classification of BS from the headlines, which is mere shallow sampling. It is a book every journalist should read and required reading for every headline writer. Some of the critical tips they explore in detail include:
Question the source
Beware of unfair comparisons
If it seems too good or too bad to be true
Think in order of magnitude
Avoid confirmation bias
Consider multiple hypotheses
The two most significant shortcomings are their susceptibility to an appeal to authority - not unsurprising by authors reared in the reverential cult of expertise and their replete (although not sole) bias in the examples. Ironically, the latter stands in contrast to their spiel on “avoiding confirmation bias” as an editorial choice; it needlessly undermines their credibility. Regardless, as an introduction, it’s a start.