Significance-testing is a statistical practice in which observed data are evaluated against a null hypothesis to determine whether the evidence is sufficient to reject the null at a pre-specified alpha level. Its parameters are: (1) a null hypothesis stating no effect or no difference, (2) a test statistic computed from the data, (3) a p-value quantifying the probability of observing data at least as extreme under the null, (4) a significance threshold (alpha, typically 0.05) against which the p-value is compared, and (5) a binary decision rule: reject or fail to reject the null. It persists through formal statistical methodology, peer-reviewed scientific reporting, and institutionalized research training. [formal: significatio | substrate: behavior | horizon: a moment | explicit: yes | epoch: 0.82]
Accepted ontology entry
significance-testing
Significance-testing is a statistical practice in which observed data are evaluated against a null hypothesis to determine whether the evidence is sufficient to reject the null at a pre-specified alpha level. Its parameters are: (1) a null…
Definition
Why it is in scope
A human-made statistical practice of evaluating whether observed data provide sufficient evidence against a null hypothesis, using thresholded p-values or test statistics. Built to persist through formal statistical methodology and scientific reporting conventions.
Names and aliases
- significance-testingen · CANONICAL
Relations from this entry
- cmrwa2lw500hxkyo6cui47t75DEPENDS_ON →
Significance-testing requires null-hypothesis to operate: the entire practice evaluates whether data contradict a null hypothesis. Remove null-hypothesis and significance-testing has no target to test — the p-value computation, rejection regions, and decision rules all presuppose a null hypothesis. The removal test passes.
- cmrw8ighw00bjkyo6h1t2zjilINSTANCE_OF →
significance-testing is a specific kind of statistical inference — it infers whether observed effects are likely due to chance using significance thresholds. Specific→general: a competent speaker calls significance testing a form of statistical inference.
Relations to this entry
- cmrwavq6k00jzkyo6119udst9← DEPENDS_ON
Direction tested: significance-testing existed first (Neyman-Pearson framework, 1930s). Removal test: remove significance-testing and p-hacking ceases to operate — p-hacking is the manipulation of data analysis to achieve statistical significance; without significance-testing as a practice, the concept of p-hacking has no referent.
Record identity
- Created
- Jul 22, 2026, 4:57 PM UTC
- Content hash
- c5d7f7f047401aee9079057f3b936243168b3d389ebafd4f19d74ceb76b61e22