- Who it helps
- Survey researchers evaluating synthetic respondent datasets.
- Availability
- Public preprint describing two US surveys and six language models. Runnable artifacts were not verified.
Measure the whole response pattern
Respondent coverage gap compares the distance from a human answer profile to synthetic profiles against distances among human samples. It evaluates joint answer patterns rather than isolated question averages.
Simple generators remain important comparisons
The authors report missing human profiles across the tested models. Gaussian-copula and independent-marginal baselines expose weaknesses that aggregate agreement can hide; these findings remain bounded to the tested settings.
What you can do
- Compare the metric’s assumptions with the sampling design before using it as an acceptance criterion. Open reference ↗
Where the evidence stops
Abstract checked; no statistical replication or full-paper audit. The measure does not by itself establish whether a survey is representative.
Sources read. Sources checked October 31, 2026 UTC. No third-party software, model inference or complete workflow was executed.
Sources & further reading
Paper · paper
Respondent coverage measures missing answer profiles ↗
Provides the cited explanation or evidence for Respondent coverage measures missing answer profiles.
arxiv.org · page checked 2026-10-31
Coverage note: Independent attention and end-to-end execution remain unverified.
The record over time (1)
Dated source checks and published explanations. An observation date can differ from the original release date.
- · Added to Current
An October 22 preprint measures whether synthetic survey responses leave real human answer patterns uncovered.