Explains the population, measure and threshold of a check

Q-353 · Knowledge knobit · Activity: Derive data quality rules · Competence: Data quality

The rows a check examines, what it counts, and the pass rate it requires.

Can AI do this?

AI does this well Most of the tested AI models master this. 4 of 4 models in the current benchmark master it.

* From the earlier benchmark of 2026-10: 10 data-modelling test tasks, 3 runs per task, graded by claude-opus-5-5. These models have not taken the current 123 tasks yet; the two methods differ, so compare their figures with care. They are left out of the “most models” counts and the best-model picks.

Mastered means a probability of at least 90% that the model has this knobit, estimated from its graded answers. “Not shown” means the answers do not settle it either way. How the models are tested.

What it looks like in practice

Expected result: The rows a check examines, what it counts, and the pass rate it requires

How it is assessed: States the population and threshold sample rules imply

Related knobits

Builds on

• Q-1804 Explains the concept of a percentage

Needed for

• Q-303 Recognises a constraint deferred without a check
• Q-313 Recognises a check measuring the wrong population
• Q-320 Recognises a threshold at odds with its rule
• Q-326 Recognises an empty population reported as a pass
• Q-355 Explains severity, threshold and failure action
• Q-357 Explains segmented data quality reporting

All 30 knobits of Derive data quality rules.

Source: Data-management practice used by the stage 3 reference and review checklist; formerly data-modelling K-753