[tinymollusk]

I didn't find "Everybody Lies" particularly compelling. It was an interesting piece of exploratory research, but I didn't see a lot of effort on generalizability. Several of his examples were using rare occurrence rates.

In short, the whole book felt like he got great access to data, and poked around in it until something interesting popped up. Then overarching conclusions were drawn without significant effort to describe whether or not the data was representative.