All hypothesis tests have the following components:
| Component | Description |
|---|---|
| Null hypothesis (H0) | The null hypothesis is known as a hypothesis of no difference. Example: Experimental drug is no better than placebo. The null hypothesis is accepted or rejected based on a statistical test of the hypothesis. |
| Alternate hypothesis (H1) | Hypothesis accepted if null hypothesis is rejected. Example: Experimental drug is more effective than placebo. |
| Alpha (α) (Also called significance level or Type I error.) |
The Null Hypothesis is rejected if the P-value is smaller than the specified Alpha value (where Alpha is the probability of rejecting the null hypothesis when it is true). Most common α values are 0.01, 0.05, and 0.10, corresponding to 99%, 95%, and 90% confidence, respectively. Results are "statistically significant at α." |
| Test statistic | Value to which data set is reduced, used in hypothesis test. Its sampling distribution under null hypothesis must be calculable (exactly or approximately), making p_values calculable. |
| Degrees of freedom | Number of independent pieces of information needed to estimate a population parameter (for example, μ or σ2) for sample of specified size. |
| Critical value | Quantile of distribution of test statistic under null hypothesis. Used to determine rejection region. |
| p_value | Probability of test results at least as extreme as test statistic results observed under assumption that null hypothesis is true. The smaller the p_value, the stronger the evidence against the null hypothesis. |
| Hypothesis test conclusion | Acceptance or rejection of null hypothesis. |