- Who it helps
- Researchers considering generated respondents for survey analysis.
- Availability
- Public preprint. Its comments say code, raw outputs and tables are available on request; public artifact access was not established.
Compare against simpler baselines
The study compares decision models, generative models and statistical approaches on 68 survey items. Statistical baselines remain competitive in the reported setting.
Group averages are insufficient
Reported individual probabilities are less reliable than aggregate frequencies. Matching a population average therefore does not establish that generated profiles represent real individuals.
What you can do
- Read the sampling and validation methods before replacing any human responses with generated profiles. Open reference ↗
Where the evidence stops
Abstract and metadata checked; results were not reproduced. One survey and population cannot establish validity across cultures or professions.
Sources read. Sources checked October 31, 2026 UTC. No third-party software, model inference or complete workflow was executed.
Sources & further reading
Paper · paper
Synthetic survey profiles can miss individual differences ↗
Provides the cited explanation or evidence for Synthetic survey profiles can miss individual differences.
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 10 preprint compares synthetic survey profiles with responses from 263 people in a Lithuanian organizational-psychology survey.