Andres pointed out a sad corollary downstream of people’s misinterpretation of regression to the mean as indicating causality when there might be none. From Tversky & Kahneman (1982) via Andrew Gelman:
We normally reinforce others when their behavior is good and punish them when their behavior is bad. By regression alone, therefore, they are most likely to improve after being punished and most likely to deteriorate after being rewarded. Consequently, we are exposed to a lifetime schedule in which we are most often rewarded for punishing others, and punished for rewarding.
Nice post on this, with code: https://acastroaraujo.github.io/blog/posts/2022-01-01-regression-to-the-mean/index.html
Andres pointed out a sad corollary downstream of people’s misinterpretation of regression to the mean as indicating causality when there might be none. From Tversky & Kahneman (1982) via Andrew Gelman: