Null tests, significance and p-values
A p-value is calibrated under a specified null model.
Key knowledge
A level-α test controls the null probability of rejection at α or below. For a simple null and a statistic rejected for large values, the p-value is P_H₀(T≥T_observed).
Conditions and scope
- The null distribution, test statistic and rejection direction must be specified.
- A p-value is not the probability that H₀ is true; failure to reject does not establish H₀.
Sources · 1
References
Sources retain their recorded scope. A source link alone does not establish verification of every statement.
Evidence and review
Physical pages 27–31: type-I error, significance level, tail p-value and failure to reject.
2026-10-10 · Simple-null type-I control and one-sided tail definition.
Entry details
Permanent link to this entry ↗- Record identity
- math-hypothesis-testing
- Revision
- university-reference.1
- Structured reference record
- University reference