An empirical study of some election systems
| Autor: | John R. Chamberlin, Clyde H. Coombs, Jerry L. Cohen |
|---|---|
| Rok vydání: | 1984 |
| Předmět: | |
| Zdroj: | American Psychologist. 39:140-157 |
| ISSN: | 1935-990X 0003-066X |
| DOI: | 10.1037/0003-066x.39.2.140 |
| Popis: | Presidential election data of the Amer- ican Psychological Association for five years are used to compare a number of election systems in frequency of selecting Condorcet winners, consistency over sub- slates, and the effect of relevant and irrelevant can- didates. The results from each system could be scaled in one dimension for each of the five elections, using the Kemeny metric. These scales were closely related to the weighting given to the variance of the candidates' rankings. An interpretatio n is given in terms of the polarizing effect of a candidate on an electorate. Elec- tion systems are discussed as strategies for the ac- cumulation of subelectorate s to reach a decisive set of voters. The underlying objective of an election system in a democratic society is that it should reflect the will of the electorate. When only two options are presented, the mechanism of majority choice is generally ac- cepted for this purpose; it weights all voters equally and treats both candidates the same, and misrepre- senting one's preference is self-defeatin g. But when three or more options are available, majority choice (and every other mechanism) is subject to criticism for one or another reason. The difficulty with majority choice is that the pairwise choices may not be transitive; that is, if the proportion of voters choosing A over B is at least one half, and the proportion choosing B over C is at least one half, the proportion choosing A over C may be less than one half. Such an event is called a cyclical majority. A related concept familiar to psychologists is known as weak stochastic transitivity, which applies to replication within an individual (as has been dem- onstrated in much experimental work in choice be- havior and in psychophysics). Violations of weak sto- chastic transitivity are usually interpreted as mo- mentary lapses from rationality, but cyclical majorities cannot be so interpreted. They may arise from gen- uinely rational individual preferences, and so cyclical majorities must be interpreted differently from in- transitivities in individual choice; we will sometimes use the term intransitivity instead of cyclical majority because in this article they are synonymous. Condorcet (1785) showed that if a cyclical ma- jority exists, then reducing the options to two by a |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|