New Framework for Measuring Collective Disagreement in Voting
Researchers have introduced a stratified framework to analyze structural disagreement among voters over alternatives. The plurality matrix generalizes pairwise comparisons by recording, for every subset S of alternatives, the probability that each alternative ranks first in S. This allows computing disagreement measures like rank-variance and divisiveness at level 3, meaning they require information about subsets of size 3. The work, published on arXiv (2605.19521v1), shows that pairwise comparisons alone cannot distinguish structural disagreement from noise.
Key facts
- arXiv paper 2605.19521v1 proposes a stratified framework for analyzing disagreement.
- The plurality matrix generalizes pairwise comparisons to subsets of alternatives.
- Disagreement measures like rank-variance and divisiveness are at level 3.
- Pairwise comparisons cannot distinguish structural disagreement from noise.
- The framework identifies minimal aggregated preference information needed for disagreement measures.
- The paper analyzes disagreement among a population of voters over a set of alternatives.
- Surveys typically ask for pairwise comparisons or full rankings.
- The work is published on arXiv.
Entities
Institutions
- arXiv