Most “science” that denies the reality of studying complex systems is broken. The concept of studying strength of correlations, even applied in the context of r

Most “science” that denies the reality of studying complex systems is broken. The concept of studying strength of correlations, even applied in the context of randomized controlled trials is only really effective in low dimensional spaces and linear dependences (I’m being very loose here, for brevity). But complex systems break all those assumptions. To steal a quote, researchers use data/statistics like a drunkard uses a lamp post — for support, rather than illumination. This applies to nutrition, medicine, ecology, economics, etc. I wouldn’t be surprised if their experimental arms run largely on confirmation bias.