This article conflates all research in the "social sciences", when in fact methodological practices vary widely within disciplines and sub-disciplines. Within each field, there is a "qualitative" literature, much less successful and popular than it once was, and probably not very useful, although I like some ethnographic and anthropoligical studies. Mainstream economics and political science is incredibly mathematically sophisticated, in fact people are coming around to the idea that there may have been too much emphasis on quantitative gymnastics above things like formulating simpler hypotheses or more descriptive work. Sociology has a bit of both. Psychology was the main culprit in the replication crisis, but even the behavioural psychology work cited approvingly in the article (Kahneman) is, I think, somewhat speculative in relating its hypotheses to experiment.
Without doubt, in economics and political science at least the problem is that the research questions are infinitely more complex than in 19C medicine; not that the methods are not quantitatively rigorous. The questions are obviously also of a more normative and moral nature than in medicine.
Without doubt, in economics and political science at least the problem is that the research questions are infinitely more complex than in 19C medicine; not that the methods are not quantitatively rigorous. The questions are obviously also of a more normative and moral nature than in medicine.