It’s a provocative question, Why we do not know what we do not know? You’ve effectively described problems with the empirical side. Is there anything else to be said about the explanatory side, what you referred to as research “in pursuit of establishing a causal relationship”? Would it be fair to say that too often we don’t know why because we don’t ask why?

Similarly, you highlighted Romer’s “mathiness”, criticism presumably directed to lack of guidance about “the connections between its theoretical and empirical statements.” The theoretical connections may not only be imprecise but lacking entirely!

And to your question about big data as a solution, I think you answered your own question with your criticisms of data mining.

Thanks for writing this. It’s a great summary of the problems with statistical inquiries.