A third property we should want to find in a voting system to be evaluative. This means that a voter should simply be asked evaluate each candidate independently. This avoids coercing voters to making comparisons between candidates. Besides making the job of the voter easier, the important advantage of evaluative voting systems is that these systems reduce the need for voters to consider extraneous issues such as electability. Instead, voters can concentrate strictly on candidate suitability for office and let the election (rather than the pundits) determine which candidates are electable. Voters may choose to take extraneous considerations into account but using an evaluative system they are at least not forced to turn to such perverse rationales.
Plurality voting and ranked voting satisfy (subject to the reservations noted above) only the condition of logical completeness. Approval voting is both logically complete and evaluative but it does fail to be balanced. In this series we have investigated many systems that are balanced but not evaluative, but only two, balanced approval voting and balanced-randomized voting, that satisfy all three of these important properties. It is these types of voting systems that deserve special consideration.
In the United States, which voting system to use is a decision that each state must make. Given the great potential for improving our democracy by adopting a better voting system, states should long have been experimenting with different approaches to voting that seem promising. Maine is now experimenting with one form of ranked voting and other states should follow this lead but they should consider other promising voting systems and not just replicate the experiment that Maine is already performing. The states have often been touted as laboratories of democracy and this is an opportunity for them to play that important role.
(Note: You can view every article as one long page if you sign up as an Advocate Member, or higher).
