---

Alpha: disambiguation

• Probability of making a Type I error: within a Neyman-Pearson approach to hypotheses testing, alpha represents a conventional significance level used for rejecting a hypothesis (called either the null hypothesis or hypothesis 1) in favor of a second one (called the alternative hypothesis or hypothesis 2). Alpha (α) represents the probability of rejecting the null hypothesis when it is true, that is, the probability of making a Type I error. This statistical meaning is also necessairly associated to a probability of making a Type II error (or beta, β), the error of failing to reject the null hypothesis when it is false.
• Significance level: an "erroneous" use of the concept to refer to a conventional level of significance within Fisher's approach of estimating the probability of data against a null hypothesis. This erroneous use derives from confusing Neyman-Pearson's significance levels of error (eg, α = 0.05, or α = 0.01) with Fisher's conventional levels of probability (eg, p < 0.05, or p < 0.01), a confusion originated with Neyman-Pearson themselves.
• Cronbach's alpha: coefficient of reliability or internal consistency in a test.

References

1. NOYMER Andrew (undated). Alpha, significance level of test. In, Paul J LAVRAKAS (undated). Encuclopedia of survey research methods. ISBN 9781412918084.

---

Contributors to this page

Authors / Editors

JDPerezgonzalez

---