Using factions to estimate preference intensity: Improving upon one person/one vote

In many elections, individuals are given two alternatives and are asked to indicate which of the two they prefer. These votes are then tabulated using one person/one vote and the alternative with the most votes becomes the public choice. Can we improve upon this system? If we change the election system and get more information on voter preference intensities, there are a number of better systems: demand revelation schemes, logrolling, Borda Voting etc. But suppose we can't change the way in which voters cast their ballots; suppose we are only allowed to change the way in which votes are tabulated. Is there a better system? Generally there is one extra bit of information — usually considered incidental — which comes with an individual's vote: his voting address. But voting address tells us something about the interest groups to which the individual may belong. As this paper shows, we can use this information to devise a better way of tabulating votes. This new system generally gives individuals from moderately cohesive voting regions more voting weight than individuals from either mildly cohesive or highly cohesive voting regions. Copyright Martinus Nijhoff Publishers 1985