Abstract

Chow ably defends classical significance testing by relating this method to venerable principles for inductive reasoning. Chow's success does not preclude the use of other approaches to statistical reasoning, which is fortunate not only for Bayesian rivals, but even for some fellow classicists.