Theoretical Risks and Tabular Asterisks: Sir Karl, Sir Ronald, and the Slow Progress of Soft Psychology

A foundational critique of significance testing that anticipates the replication crisis by three decades. Directly relevant to the library's meta-debate controversy: the same nose-counting of p-values that Meehl decries is central to how both psi proponents and skeptics evaluate the evidence. His distinction between statistical and substantive hypotheses illuminates why ganzfeld batting averages and Bem replication tallies are contested.

Arguing that theories in soft psychology (clinical, social, personality) neither cumulate nor get clearly refuted but merely fade away, Meehl identifies 20 intrinsic difficulties of the subject matter and then turns to an extrinsic one: the near-universal reliance on null hypothesis significance testing. Drawing on Popper's falsificationism and Bayesian reasoning, he shows that since the null hypothesis is quasi-always false, refuting it depends on statistical power, not theoretical verisimilitude, making significance tests a feeble method of theory corroboration. As an alternative, he demonstrates 'consistency tests' within his MAXCOV-HITMAX taxometric framework, achieving 94% accuracy with zero false negatives across 600 Monte Carlo samples. He prescribes multiple non-redundant estimates of theoretical quantities over significance tests.