Sampling & Statistics
Null Set
Definition
The set of documents predicted non-responsive and set aside, also called the discard set.
Why it matters
It is the population an elusion test samples from to check what was missed.
In practice
The null set is everything the workflow decided not to review or produce: documents below the cutoff score in TAR 1.0, or the unreviewed remainder when a continuous review stops.
It is usually the largest part of the collection, which is why small error rates in it matter. An elusion sample drawn from it estimates how many responsive documents it still holds.
Worked example
Of 200,000 documents, 40,000 score above the cutoff and go to review. The other 160,000 form the null set, and the elusion sample is drawn from those 160,000.
Illustrative numbers, not from a real matter.
Compare with
- Control Set: A control set is a random sample of the whole collection coded at the start; the null set is what is left out at the end.
Work it out
- Culling Impact calculator · Did culling remove noise without losing responsive documents?
- Elusion & Recall calculator · What did we leave behind, and what recall can we defend?