A model combining an individual voter’s personal preferences (a cardinal utility rating) with the preferences of the rest of the electorate (as measured by pre-election polls) is investigated to determine how a voter may maximize his expected satisfaction with the election outcome. The basic model of McKelvey and Ordeshook for a one man-one vote election procedure is generalized to one man-N votes, approval voting and committee elections. Explicit decision conditions are derived for each procedure. An approximation technique is described and some numerical results are included. The final section extends the basic deterministic model to a situation in which the optimal voting strategy may be probabilistic. The model can be used to compare voting procedures as well as the strategies available to a voter under a particular voting procedure.
[1]
Peter C. Fishburn,et al.
Approval voting, Condorcet's principle, and runoff elections
,
1981
.
[2]
Dale T. Hoffman,et al.
Relative Efficiency of Voting Systems
,
1983
.
[3]
J. Neumann,et al.
Theory of games and economic behavior
,
1945,
100 Years of Math Milestones.
[4]
William F. Lucas,et al.
Measuring Power in Weighted Voting Systems
,
1983
.
[5]
Robin Farquharson,et al.
Theory of voting
,
1969
.
[6]
Peter C. Fishburn,et al.
Utility theory for decision making
,
1970
.