Properties, Standards & Criteria for Evaluating & Comparing Multi-Alternative Single-Winner Social Choice Methods. by Mike Ossipoff


Single winner vsoting systems are for choosing 1 alternative from among several. Typically, in partisan elections, the 1-Vote Plurality (First Past The Post, or FPTP) is used. People vote for 1 alternative or candidate, and the one with most votes wins. Other methods are proposed as improvements.

Various different single-winner methods are proposed as improvements. There's little or no consensus on standards for evaluating them. Defining some of those standards & criteria, and telling how the methods do by them, with initial special attention to a few of the best methods, is the purpose of these articles.

Why are single-winner methods important? What government does, what it allows or doesn't allow, those things affect every material aspect of our world. Even a bird in a tree in the most remote forest is affected by that. So it becomes important how public wishes translate to government policy--typically, but not necessarily, via the election of candidates for office, and by voting by those candidates in their legislatures.

What's wrong with the Plurality method? The lesser-of-2-evils problem. Voters know that if they vote for their favorite, that will often prevent them from being able to help a more winnable compromise (the "lesser-evil"), who may be the only candidate who could beat someone whom they despise more. So they have to completely abandon their favorite. I believe that most agree that that is the main problem that dramatizes why we need better single-winner voting systems. Many of the criteria that I describe later are for measuring compliance with that standard.

Here are the topics covered in these articles:

Table of Contents:
1. The Approval methodApproval is one of the two best systems and very simple.

2. How voters would use Approval voting

3. The Condorcet method Condorcet is more complicated than Approval but has extra advantages

4. Strategy Criteria

5. Some Traditional Academic Criteria

Mike Ossipoff