Contribution to the handbook
Supporting · Assessed 2026-10-28
The question and its evidence give a concrete operational measure: classify participants by which side of neutral their ratings fall on, fit group-statement positions to combinations of participants’ opinions, and sum estimated minority weights across levels of group division. This is useful background for evaluating representation, but a study-specific measure is not itself a governance design or proof of fairness.
Does this challenge the goal?
In-scope gap
The question seeks detail about a method for measuring minority influence, which can inform how a deliberative process is evaluated. It does not contest the exploration’s goal or beneficiary boundary.
This is an inquiry about one study’s method, not evidence that the handbook or the wider field lacks such methods.
Why this category?
- useful background
- test and measurement method
- limits of the measure
The evidence is an AI interpretation of one shared article source, not independent corroboration. The source itself cautions that the estimated weight is not a complete measure of influence or fairness.
This is an AI assessment of contribution to the goal, separate from whether the question has an answer.
Authority, accountability and resilience →
Tests, stop conditions and unresolved work →
Read the exploration goalExplain how participants were classified as minority or nonminority for the analysis.
What we can say so far
The authors classified participants as minority or nonminority according to whether their position ratings fell on one side or the other of neutral.
What remains open
No material gap is recorded within this part’s current scope.
Conditions and limits
- This is an interpretation of the readable text of the 10-page Science article; supplementary materials were not included.
- The classification is based on position ratings relative to neutral.
Evidence and related ideas
Supporting evidence, qualifications and context play different roles in this answer.
supports
Estimating minority influence The authors classified participants as minority or nonminority according to which side of neutral their position ratings fell on. To estimate minority influence, they fitted group-statement positions to combinations of the participants’ opinions, summed the estimated weights for minority opinions, and repeated this analysis for different levels of group division. They reported the average minority weight.
Provides the authors’ operational definition of minority and nonminority participants.
AI can help humans find common ground in democratic deliberation
Read full source: AI can help humans find common ground in democratic deliberation (opens in a new tab) ↗Read source discussion: AI can help humans find common ground in democratic deliberation →
Describe how the authors fitted group-statement positions to participants’ opinions, calculated minority weights, and examined them across levels of group division.
What we can say so far
The authors classified participants as minority or nonminority according to whether their position ratings fell below or above neutral. They represented opinions and group statements as positions on a question-specific axis, then fitted each group statement’s position to combinations of participants’ opinion positions. They summed the estimated weights assigned to minority members’ opinions to obtain a minority-weight estimate, and repeated the analysis at different levels of group division, reporting the average minority weight. This describes the authors’ operational estimate, not a direct measure of fairness or of every kind of influence.
What remains open
No material gap is recorded within this part’s current scope.
Conditions and limits
- This is the method reported in the article as represented by the supplied evidence; supplementary materials are not included.
- The estimate depends on the chosen text and rating measures and does not establish that it captures influence beyond those measures or whether the statement was fair.
- Alternative text models or position-axis definitions could yield different measurements.
Objections to consider
- The source notes that alternative text models and axis definitions could produce different measurements; the supplied evidence does not establish robustness across those alternatives.
- The source limits the estimate to the chosen text and rating measures and does not establish that it captures influence completely or measures fairness; no supplied evidence resolves that limitation.
Evidence and related ideas
Supporting evidence, qualifications and context play different roles in this answer.
supports
Text scores used a question-specific scale The researchers encoded each written opinion as a 768-number text vector, then rotated those vectors onto a question-specific line running from agreement to disagreement. They checked the resulting position scores against participants’ earlier ratings; the scores were correlated, but the match was not exact.
Reports the question-specific position scale and comparison with earlier ratings, contextualizing how opinion positions were measured.
AI can help humans find common ground in democratic deliberation
Read full source ↗