How to Read Likelihoods
Reading likelihoods means asking how expected the evidence is under each live hypothesis.
Compare, Do Not Merely Confirm
The key question is not "Does this fit my view?" Many things fit many views. The better question is "Is this evidence more expected if my view is true than if an alternative is true?"
List Live Alternatives
Before interpreting evidence, list at least two alternatives. Include mundane explanations, error, noise, coincidence, and selection effects where relevant. Evidence that survives comparison deserves more weight.
Look for Discriminators
Strong evidence discriminates. It rules some possibilities down while raising others. Weak evidence is compatible with too many possibilities to move confidence much.
The Rival-Explanation Habit
Likelihood reasoning becomes powerful only when rival explanations are actually allowed into the room. A claim can fit your theory and still fail to support it if the same evidence fits ordinary alternatives just as well.
Practice with simple cases: a friend is late, a study reports an effect, a political opponent behaves badly. Ask not only whether the evidence fits your favored explanation, but whether it fits human error, incentives, coincidence, selection, or measurement noise.
Adjacent References
These are starting points rather than exhaustive citations. They connect this page to nearby work in probability, calibration, cognitive bias, and social epistemology.
- |Thomas Bayes, "An Essay towards solving a Problem in the Doctrine of Chances" (1763).
- |E. T. Jaynes, Probability Theory: The Logic of Science (2003).