Contribution to the handbook
Supporting · Assessed 2026-10-28
The question and evidence provide useful background on how one study tested preference-pattern stability: win rates were compared using R². This may inform test design, but it is a narrow study measure and does not itself specify a governance system or establish wider representativeness. The answers share a source root and are not independent evidence.
Does this challenge the goal?
In-scope gap
The question concerns a validation measure for preference estimates, which can inform testing citizen representation without contesting the exploration’s goal.
The evidence describes matching a study sample, not whether that sample represents people outside the study.
Why this category?
- testing
- useful background
- evidence limitations
The packet reports the measure and its limited target, but does not establish how useful this metric would be for evaluating representation in a real governance process.
This is an AI assessment of contribution to the goal, separate from whether the question has an answer.
Tests, stop conditions and unresolved work →
Authority, accountability and resilience →
Read the exploration goalIdentify the measure used to compare sample and full-sample preference patterns and explain how the authors interpreted its value.
What we can say so far
The authors compared each proposal’s win rate in the sample with its win rate in the full study sample, using the squared Pearson correlation (R²). They interpreted 100% as identical win-rate patterns; the measure indicates how closely the aggregate patterns match.
What remains open
No material gap is recorded within this part’s current scope.
Conditions and limits
- This measures similarity in aggregate proposal win-rate patterns, not whether every individual’s preferences match.
- The comparison is to the full study sample and does not establish representativeness of a wider public.
- The answer reflects the supplied interpretation of the authors’ method; the evidence is not independent new verification.
Objections to consider
- A close R² match concerns aggregate proposal win rates and does not show that individual preferences are reproduced or that the sample represents people outside the study. The answer states this scope limitation and does not overclaim.
Evidence and related ideas
Supporting evidence, qualifications and context play different roles in this answer.
context
Group preferences were consistent on retest The authors estimate 95.38% test-retest reliability for the group’s preferences in the full participant sample. They say this is consistent with earlier paired-comparison work, where rankings tended to converge after about 30 to 40 choices per participant.
Reports full-sample test-retest reliability, a related but distinct check from comparing a sample’s win-rate pattern with the full sample.
Large Language Models (LLMs) as Agents for Augmented Democracy
Read full source: Large Language Models (LLMs) as Agents for Augmented Democracy (opens in a new tab) ↗Read source discussion: Large Language Models (LLMs) as Agents for Augmented Democracy →
context
Testing group-level preferences The researchers also compare whether ordinary random samples and model-augmented samples can reproduce the full group’s combined preferences.