SYSTEMA CONSTRUCTUM

Accepted ontology entry

multiple-comparisons

Multiple-comparisons is a statistical situation that arises when hypothesis tests are conducted on more than one hypothesis simultaneously, causing the probability of at least one Type I error to exceed the nominal per-test significance le…

ACCEPTED THINGcmrwavehv00jskyo6t73uv4rk

Definition

Multiple-comparisons is a statistical situation that arises when hypothesis tests are conducted on more than one hypothesis simultaneously, causing the probability of at least one Type I error to exceed the nominal per-test significance level. Its parameters are: (1) the number of independent or dependent tests performed, (2) the per-test significance level alpha, and (3) the dependency structure among the test statistics. The persistence mechanism is through established correction procedures (Bonferroni, Holm-Bonferroni, Benjamini-Hochberg), statistical methodology textbooks, and built-in software implementations in statistical packages. [formal: multiplex comparatio | substrate: behavior | horizon: a moment | explicit: yes | epoch: 0.01]

Why it is in scope

A human-made statistical situation where hypothesis tests are conducted on more than one hypothesis simultaneously, inflating the family-wise error rate. Built to persist through statistical methodology, correction procedures (Bonferroni, Holm, FDR), and standard software implementations.

Names and aliases

Relations from this entry

  • cmrwa2lw500hxkyo6cui47t75DEPENDS_ON →

    multiple-comparisons operates by inflating the Type I error rate of null-hypothesis testing: when you conduct multiple null-hypothesis tests simultaneously, the probability that at least one null hypothesis is incorrectly rejected increases. The removal test: remove null-hypothesis and multiple-comparisons has nothing to operate on — its entire mechanism is the interaction of multiple null tests. The practice of multiple-comparisons correction exists solely because of null-hypothesis testing.

  • cmrwa9lop00iekyo6t6jw7bqlDEPENDS_ON →

    Multiple comparisons is the problem of inflated false-positive rates when performing many hypothesis tests simultaneously. Remove hypothesis-testing now and multiple-comparisons stops operating — it has no independent existence outside the framework of testing hypotheses. This is object-level, not meta-level: the concept's operative mechanism IS statistical inference within hypothesis testing.

Relations to this entry

No accepted relations in this direction.

Record identity

Created
Jul 22, 2026, 4:31 PM UTC
Content hash
8cd4151ad0264bd0d24eb22300cef49357c83a2e6deca3edae7c38928220784f

Open a related act record