All relevant contigencies have been investigated in order to establish the frequency with which a Condorcet winner is elected under the plurality and approval procedures when voters are assumed to act sophisticatedly, and given that: (1) there are three candidates of whom one must be elected; (2) voters have complete information regarding all other voters (strict) preference orderings; (3) the electorate consists, alternatively, of three, four, or five blocks of voters; 4) all contingencies are considered equally likely. The main results are: (i) the approval procedure is significantly more Condorcet-efficient than the plurality procedure; (ii) when a Condorcet winner exists, a non-Condorcet winner is elected very rarely under both voting procedures; (iii) voting under the approval procedure results more often in a determinate winner when a Condorcet winner exists, whereas voting under the plurality procedure results more often in a determinate outcome when there are cyclical majorities; (iv) sophisticated voting under the approval procedure seems to be more Condorcet-efficient than the sincere version of this procedure, whereas the Condorcet-efficiencies of sincere and sophisticated voting under the plurality procedure are not significant different.
No takes yet. Share an insight, caveat, or question.
Felsenthal et al. (1990) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: