When reading popular books in a field I'm often uncertain how to treat the evidence. I'm reading Haidt's "The Righteous Mind". Fascinating, and much agrees with my prejudices. But I worry about caveats & counter-evidence that may not have made it into the book

Not a criticism of the book - this is a genre convention, without which many paragraphs would need to be expanded into chapters.

Still, it leaves me uncomfortable. I usually deal with this by (attempting) to treat such books merely as useful sources of heuristic ideas and models.

It's a little frustrating. Other strategies?

(In general I spend much more time reading primary literature. And maybe that's the right approach.)

It's amusing to imagine an alternate one-paragraph version of all books: "Read the following 5 papers to obtain the actual evidence and references behind this book".

I'd love a site which collected these.

"Oh, I haven't read 'The Righteous Mind', but I read its evidence base."

"read its evidence base" isn't really right. More like: "digested its evidence base". Sometimes, of course, there's no nutritional content....

Also, something I've noticed is almost universally believed: everyone believes everyone else should up their standards of evidence, but they're just fine. Makes for amusing conversations.

I exaggerate. Slightly.

It's much like dinner parties of scientists at conferences, everyone complaining about the terrible (anonymous) referee reports that were, just as likely, written by other people at the table.

The most striking exception on the standards question: the few logicians & cryptographers (not crypto-currency ppl) I've known often seem panicked that they're wrong. Unsurprisingly, they also seem to have the highest standards...

TBC: I am very guilty of the last few things.