Thinking, Fast and Slow

by Daniel Kahneman

Regression effects are a common source of trouble in research, and experienced scientists develop a healthy fear of the trap of unwarranted causal inference.