Statistical significance
A measure of whether a selection-rate gap is large enough, given the sample size, to be unlikely to occur by chance. Courts have used a two-to-three standard-deviation benchmark (Castaneda v. Partida, 1977; Hazelwood School District v. United States, 1977). It complements the four-fifths rule: a small gap in a large pool can be significant, and a large ratio in a tiny pool may not be.
Defined in 2 GAGE programs, which carry 2 distinct definitions of it. The wording above is taught in AI Governance: Applied Mastery.
How each discipline defines it
The same term does different work depending on who is using it. These are the definitions as each program teaches them, unedited.
A measure of whether a selection-rate gap is large enough, given the sample size, to be unlikely to occur by chance. Courts have used a two-to-three standard-deviation benchmark (Castaneda v. Partida, 1977; Hazelwood School District v. United States, 1977). It complements the four-fifths rule: a small gap in a large pool can be significant, and a large ratio in a tiny pool may not be.
A measure of whether an observed difference between conditions is larger than what random variation alone would plausibly produce, without regard to whether that difference is large enough to matter in practice.
Where it is taught
The exact lessons this term appears in. The first 7 topics of every program are free with a free account.
- The hiring-AI problem: bias audits, notices, and the law already watching hiring AI · The EU AI Act: The Executive Map, AI Governance: Applied Mastery
Terms it appears with
Not an alphabetical neighbourhood: these are the terms taught in the same lessons, ranked by how often they appear together.