Proportional Representation Foundation

This is the first of a short series of posts on Ranked Choice Voting (RCV), inspired by the resurgence of interest in RCV in the US over the last few years. Only a small minority of recent adoptions use ranked voting for PR, but to the extent that voters become familiar with ranking candidates, it can be a stepping stone to STV-PR, hence my interest (that and a general interest in improving election methods, of course).

I will update this post from time to time, both for substance and to provide an index to the other posts in the series.

RCV-related terminology can be confusing, so I’ll begin with a kind of guide to the alphabet soup one encounters along the way.

STV-Based Ranked Voting

STV (Single Transferable Vote) is the usual term used in voting literature to refer to a set of methods that require voters to rank candidates in preference order. Each voter has exactly one vote (“single”). That single vote is initially counted for the first-ranked candidate, but can be “transferred”, all or in part (in some forms of STV-PR) to lower-ranked candidates. We use “STV-1” and “STV-PR” as shorthand references to single-seat and multiple-seat proportional representation forms of STV, respectively.

RCV (Ranked Choice Voting) is another name for elections using the STV counting method. Where the distinction matters, we’ll sometimes use the abbreviations RCV-1 and RCV-PR (as with STV-1 and STV-PR) to refer to the single- and multiple-seat variants of RCV. The name emphasizes the voter’s view of the election (ranking their choice) rather than the underlying counting method.

IRV (Instant Runoff Voting) is another name (chiefly US) for single-seat RCV (RCV-1). The name emphasizes the similarity of STV-1 to conventional runoff elections.

AV (Alternative Vote) is yet another name (chiefly British) for single-seat RCV (RCV-1), not to be confused with Approval Voting, which does not employ ranking. The name suggests that a voter lists preferred alternatives to their first choice.

Preferential Voting You’ll sometimes see STV-based voting systems referred to as “preferential” systems.

Non-STV Ranked Voting

Some voting systems are superficially similar to STV systems, but differ in the counting method. We won’t discuss them in detail, but they include for example Borda counts, Condorcet-compliant methods, and assorted scoring systems, where voters assign numeric weights to candidates (Borda is such a system, in which the score is implicit in a candidates ranking).

Non-Ranked Voting

The most familiar voting system in the US is plurality voting, where each voter votes for one candidate (or more in the case of multiple-seat elections), and the candidate(s) with the highest vote count(s) win(s). Such elections may be combined with auxiliary elections, such as primary or nominating elections, and/or runoff elections. Plurality voting is often referred to as FPTP (first past the post).

Other non-ranked systems include approval and cumulative voting (the latter being a semi-proportional multiple-seat method).

Summary

In this series of posts, I’ll look at RCV as a replacement for plurality voting, though other approaches will come up from time to time. I’ll update the above list, and the following list of posts, as we progress.

Posts

I’ll start with some common misconceptions and misunderstandings about RCV, and then move on to more substantive criticisms. I’ll add links here, as they appear.