Thinking, Fast and Slow · chapter 10 · member edition

The Law of Small Numbers

Chapter 10 warns that humans – including trained researchers – have terrible intuitions about sampling variability. The science is solid and has been vindicated by the replication crisis. The irony is that Kahneman’s own book fell victim to the same error, citing small-sample studies that later collapsed. Gigerenzer’s important critique (frequency formats reduce errors) is omitted.

So what

This chapter contains the book’s most self-aware insight and its deepest irony. The science on statistical reasoning errors is robust. But the book itself fell victim to the very bias it describes, citing small-sample studies that later failed replication. Read this chapter as both a lesson in statistics and a cautionary tale about the limits of expert judgment.

Verdict

MostlyAccurate

Claims checked in this chapter (3)

holds✓ verified
Even trained researchers have poor intuitions about sampling variability and expect small samples to be representative of the population ('belief in the law of small numbers').
needs context✓ verified
People neglect base rates when making probability judgments, relying instead on representativeness -- as shown in the lawyer-engineer problem.
holds✓ verified
Many of the psychology studies Kahneman cites in this book -- including studies on priming and social psychology -- were themselves conducted with small samples, making them vulnerable to the exact error this chapter describes.