Explains a pooled rate as the weighted mean of its population rates
A rate over several populations is the mean of their rates weighted by their sizes (denominators), not their plain average; a small poor population can hide in it, and a shift in the population mix moves it with no change in any population's quality.
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.
| Model | Tier | Probability of mastery | Status |
|---|---|---|---|
| gemini-4-argon | Flagship | 100% | Mastered |
| gemini-3.5-flash-lite | Low cost | 100% | Mastered |
| gpt-6-luna | Low cost | 100% | Mastered |
| claude-haiku-5-5 | Low cost | 100% | Mastered |
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 pooled rate as the size-weighted mean of the population rates, with the mix effect it can hide
How it is assessed: Computes a pooled rate from population rates and sizes, tells it from the plain mean of the rates, and names the mix effect
Related knobits
Builds on
Needed for
- Q-406 Recognises one quality ratio over differing populations
- Q-703 Recognises a quality trend read without its population change
- Q-754 Explains like-for-like trends of quality measures
All 10 knobits of Profile and assess data quality.
Source: Part B candidate of k12-dmbok for activity QO-4 (a need for school knowledge no knobit of the activity reached), accepted by Peep Küngas on 31 Oct 2026