NOTE: The following diary is adapted from a series of articles that were posted last week on my blog. Links to these articles are provided at the end of this diary.
* * * * * * * * * * * * *
Recently, I discussed the need for two fundamental electoral reforms that I believe are needed in order to lay the groundwork for a more open and truly representative multi-party system: ballot access reform, to make it easier and fairer for minor parties and independent candidates to both get on the ballot and stay on the ballot in successive elections; and electoral fusion, to make it easier for voters who have grown accustomed to having to choose between two monolithic parties to transition over to a multiparty system by giving them the option of supporting minor parties that cross-endorse major-party candidates.
While I believe that these two elements of reform are essential and should be given first priority for tactical reasons, I nonetheless acknowledge that they are only components within a broader reform program, and that by themselves they probably won't be enough to bring about the change we need to open up American politics. Ultimately, we will also need to address basic deficiencies in the way that votes are cast and counted.
I. Getting past first-past-the-post
If there's one thing that many pro-reform advocates seem to agree on, it's that the voting regime we're using right now isn't working. Currently in America, the vast majority of elections are decided using the first-past-the-post voting method, in which each person is given one vote and the candidate that receives the most votes is declared the winner. In some cases, when the candidate with the most votes receives less than a majority, a runoff election is held in which the top two vote-getters are subject to a second round of voting, but in most cases, a mere plurality is enough to ensure victory.
The shortcomings of the first-past-the-post system are numerous and glaring. In the case of plurality voting, an election can in some cases be won by a candidate who falls far short of a majority and of achieving the mandate that goes with it. An extreme (and for we in the Lone Star state, unfortunate) example of this can be seen in the 2006 Texas gubernatorial election, in which Republican incumbent Rick Perry was able to win despite receiving only 39% of the vote. Perhaps the most often cited disadvantage of first-past-the-post voting in general is the one expressed by Duverger's law, which asserts that plurality voting tends to reduce the number of viable parties within a political system and lead to the emergence of a two-party system, like ours here in America. This often leads to tactical voting--most voters are aware that only two candidates have a realistic chance of winning and don't want to throw their vote away on a longshot, or worse, split the votes on their end of the political continuum and thus facilitate the election of the major-party candidate from someone on the other side. So rather than voting for the candidate that best reflects their political beliefs, most voters who don't prefer one of the two major-party candidates nonetheless hold their noses and vote for "the lesser of two evils".
Another often-cited deficiency of the first-past-the-post voting system is that it tends to result in a large number of "wasted" votes, or votes that are cast for either losing candidates or for winning candidates beyond the threshold needed for victory. In the 2005 UK General Election, for example, 70% of votes were "wasted"--52% were cast for candidates who lost, and another 18% were cast for winning candidates in excess of the number needed for victory. The relatively high number of wasted votes in first-past-the-post voting tends to encourage gerrymandering due to the fact that redistricting is left up to the whims of elected partisan legislators, who often use their control of the process to attempt to marginalize voters likely to support for their opposition. They do so by concentrating those voters in relatively few districts where their candidates will win with overwhelming majorities, while creating a majority of districts in which their party has a smaller majority. For an example of this, I am again forced to plumb the dark electoral history of my own home state, this time going back to the 2003 battle over congressional redistricting.
The great Texas redistricting battle of 2003 began in the aftermath of the previous year's midterm elections, which produced Republican majorities in both houses of the Texas Legislature for the first time since Reconstruction but still left a slim 17-15 majority of congressional seats under the control of Democrats. Thanks to the machinations of former bugkiller and then-current House Majority Leader Tom DeLay, along with Governor Goodhair, the Republican-controlled legislature took the unprecedented step of attempting to redraw district lines that had already been determined by the 2001 redistricting process. Despite the best efforts of the "Killer D's", and the later efforts of the Texas Eleven, the Republicans were ultimately able to pass a bill targeting the districts of 10 white Democratic incumbents. The redistricting effort survived later legal challenges (although the state was forced by the US Supreme Court to redraw the boundaries for District 23 due to racial gerrymandering), and the GOP began reaping the fruits of their efforts in the next election cycle, taking a 21-11 majority among the state's congressional delegation for a net gain of six seats in the US House. Had it not been for the Republican chicanery in Texas, the GOP would have actually lost seats in the House during the 2004 election cycle.
The deficiencies of first-past-the-post voting are widely acknowledged, and use of the system has been abandoned in many other countries in favor of other voting regimes. Only among countries that are historically connected to the British Empire is the system still widely used, and it's very rare for countries using other voting systems to switch over to first-past-the-post voting. But if we're going to scrap first-past-the-post voting, which system should we switch over to? Well, there are two other major categories of voting regimes that are used in single-winner voting systems. Under preferential voting systems, candidates are ranked by voters in order of preference, and the rankings are used in various ways to determine the winner of the election. Different methods of preferential voting include various Condorcet methods, Bucklin voting, Coombs rule voting, the Borda count, and instant runoff voting, which is the most widely known and often-cited method of preferential voting among reform advocates. The other major category of voting regimes within the single-winner system is range voting, in which voters are allowed to vote for multiple candidates, assigning a score to the candidates they choose to vote for which is tabulated to determine the winner of the election. This category includes both traditional range voting and approval voting, in which voters are allowed to cast a vote for as many candidates in an election as they wish (while candidates are not scored by voters under an approval voting system, it's still considered a form of range voting, because voters are essentially given the option of assigning a score of either 1 or 0 for each candidate).
II. Instant-Runoff Voting
Among the various preferential voting methods, the most well-known and most popular among reform advocates is instant-runoff voting. Sometimes referred to as ranked-choice voting, IRV allows voters to cast their vote by ranking the candidates in an election in order of preference. Votes are distributed to candidates based on the first preference indicated on the ballot, and if a candidate receives a majority of the first-preference rankings, that candidate is declared the winner. However, if no candidate receives a majority, then candidates receiving fewer votes are eliminated and votes are redistributed. There are two major variations on instant-runoff voting. In an alternative voting system, the candidate with the fewest number of first-preference rankings after votes are distributed is eliminated, and the counting process is repeated until a winner emerges. Contingent voting is a variation on IRV in which all but the top two vote-getters are eliminated before votes are redistributed.
Instant-runoff voting has an extensive track record and is currently practiced in a number of countries to at least some extent: it's used to elect members of the House of Representatives in Australia and Fiji, members of the parliament in Papua New Guinea, and the President of Ireland, to cite a few notable examples. It's also used by some political parties in the UK and Canada to select party leadership, and is beginning to emerge as an option in the United States as well, having been implemented at the local level in several counties and cities including San Francisco, Oakland, Minneapolis, and St. Paul. Louisiana, South Carolina and Arkansas also use IRV on a limited basis with overseas military voters.
Proponents of IRV argue that it is superior to first-past-the-post voting because in allowing voters to rank candidates, it enables them to express their true preferences without worrying about wasting their votes or splitting the vote on their side of the political divide, thus minimizing the "spoiler" effect that works against minor-party candidates and leads voters to support one of the two major-party candidates as the "lesser of two evils". IRV proponents also point out that under plurality voting, the winner of an election often ends up polling less than 50%, violating the democratic ideal that an elected officeholder should be preferred by a majority of voters. IRV provides for a runoff, and it arguably does so more cheaply than first-past-the-post elections with a runoff round of voting because it eliminates the need for a second separate election to take place. This is a particularly important consideration for overseas and military votes, which is probably why several jurisdictions that don't allow for general IRV voting nonetheless allow it on a limited basis for overseas military personnel. IRV may also discourage negative campaigning, since ranking gives candidates an incentive to avoid alienating supporters of the candidate that they might be inclined to attack.
Popular though it may be with reform advocates, instant-runoff voting also has its fair share of detractors. Supporters of first-past-the-post voting argue that by denying voters the opportunity of basing their ranking of candidates on the results of earlier ballots, it offers less freedom of choice than separate rounds of runoff voting. However, this aspect of IRV doesn't seem to limit voters' freedom of choice so much as it limits their ability to use tactical voting methods, which some would say is an argument in its favor. By allowing voters to express their preference for more than one candidate, IRV also arguably violates the democratic principle of one-man, one-vote. This argument, however, has been undermined by at least one judicial ruling in Stephenson v. Ann Arbor Board of City Canvassers (1975), in which Michigan Circuit Judge James Fleming upheld IRV as consistent with the principle of one-man, one-vote and the Equal Protection Clause of the Fourteenth Amendment. Some IRV opponents also argue that the system's increasing prevalence could lead to the increased use of electronic voting, with all its associated problems. Supporters, however, argue that the problems associated with electronic voting can be addressed by implementing audit procedures to minimize the likelihood of data loss or fraud. It should also be pointed out that electronic voting has become increasingly prevalent even in the absence of widespread IRV implementation, and will thus have to be dealt with regardless of whether IRV becomes more widely accepted.
III. Preferential voting: alternatives to IRV
In comparison to first-past-the-post voting, IRV seems clearly superior in terms of efficiency, providing for a runoff in cases where no candidate receives a majority of the vote while avoiding the expense and hassle of holding a separate second round of runoff voting, as well as in terms of providing for a more open and representative political system. However, in addition to instant-runoff voting, there are several other less common and less well-known variations on preferential voting.
The Coombs method was devised by Clyde Coombs, an American psychologist specializing in the field of psychometrics. The method he devised is similar to instant-runoff voting, with the exception that instead of eliminating candidates with the least number of first-preference votes among remaining candidates in each round of counting, candidates are eliminated for having the highest number of last-preference votes among remaining candidates. The Coombs method is essentially a variation on instant-runoff voting, and would seem to have the same advantages and disadvantages as IRV. However, at least one study has argued that the Coombs method is actually preferable to IRV because it's more likely to select the Condorcet winner (more on that in a moment) in races involving four or fewer candidates.
The Condorcet method is named for 18th-century mathematician and philosopher Nicolas de Caritat, the Marquis de Condorcet, who is credited with originally devising the concept. (The Marquis de Condorcet should not be confused with the Marquis de Sade, who advocated somewhat more diabolical but far less complicated forms of torture.) Under the Condorcet method, voters rank candidates in order of preference, as in instant-runoff voting, with the difference that tie-rankings are allowed. For each ballot, the rankings of each candidate are them compared with those of every other candidate, one pair at a time, and a "win" is counted in each of these pairings for the higher-ranked candidate. These wins are then totalled, with a separate count for each possible combination, and the candidate who wins every one of their pairings is declared the winner. If no single candidate wins all of their pairings, a winner is chosen using one of several resolution methods, which for the sake of brevity and my personal sanity I won't attempt to describe here. If the Condorcet method sounds overly convoluted, that's probably because it is, and that may have something to do with the fact that it's not currently known to be used by any government in the world. However, the idea behind the Condorcet method--that the candidate who most deserves to win an election is the one who wins every possible matchup against every other candidate in the race--is popular enough that it is frequently used as one of the benchmarks by which voting systems in general are evaluated, with a voting system considered stronger if it's more likely to select the Condorcet winner.
Proponents of the Condorcet method argue that it is superior to instant-runoff voting because it takes all candidate rankings into account simultaneously, whereas under IRV, rankings for candidates who have been eliminated are not taken into consideration. It is also said to be superior because it selects the candidate that is the most preferred overall. However, it could be argued that this is in fact a shortcoming of the Condorcet method, not a selling point. Under the Condorcet method, it's actually possible for a candidate to win an election without being the first preference of any voter, and Condorcet elections tend to produce consensus candidates that represent the "best compromise", which in the context of American politics would probably translate into more corporatist "centrists" being elected to office.
The Borda count is named after Jean-Charles de Borda, a French mathematician and political scientist who formulated the system in 1770. Under the Borda Count, as in other preferential voting systems, voters rank candidates in order of preference. Candidates are awarded points based on their preference ranking for each ballot, and the candidate who accumulates the most points is declared the winner. Use of the Borda count is not widespread among electoral systems--it is currently used on a limited basis in Slovenia to elect two ethnic minority members of the National Assembly, as well as in Nauru, where it is used for parliamentary elections, and in Kiribati, where it is used in presidential elections. Its use is more widespread among private organizations, including the NBA where it is used to select the league MVP.
Like the Condorcet method, the Borda count has a tendency to favor consensus candidates and it is possible for a candidate to have a majority of first-preference votes and still lose the election. It is also susceptible to tactical voting: voters who want to improve the chances of their most-preferred candidate winning might undervote for other candidates that they prefer in order to improve the chances of their first-preference candidate winning. And the Borda count is highly vulnerable to strategic nomination: by running more candidates in an election, a political faction can actually improve the chances that one of their candidates will win the election, rather than dividing the vote as might be the case in a first-past-the-post election.
Bucklin voting, also known as the Grand Junction system, is named after James Bucklin, who originally devised the concept. Like other preferential voting systems, under the Bucklin method voters rank the candidates in an election by preference--however, there are variations on how this is done, and in most elections where the Bucklin system has been used, the practice has been to allow voters to rank only their first and second preferences. Votes are tallied by first counting the first preferences listed on each ballot. If a candidate wins a majority of the first preference votes, that candidate is declared the winner; otherwise, all but the top two vote-getters are eliminated, and second preference votes are added to the tally to determine the winner of the election. Like the Condorcet method, Bucklin voting is not currently being used by any government, but it has been used in the past, most notably in Alabama where it was used in 16 primary elections between 1916 and 1930. The Bucklin method never caught on, probably because it's too easy for well-organized voters to game the system on behalf of their candidate of choice. Listing a second preference on a Bucklin ballot effectively cancels out a person's vote if the second preference they list makes it to the second round of vote-counting, thus encouraging voters to "bullet vote": either not listing a second preference at all, or picking a second preference who is perceived as having no chance of being in the top two. The peculiarities of Bucklin voting, combined with the practice of bullet voting, lead to a comparatively high probability of elections in which the winning candidate falls short of getting a majority of the vote. This was demonstrated in the Alabama primary elections where Bucklin was used: out of 16 elections, in not a single one of them did the addition of second-preference votes give the winner a majority, and in only one election did the addition of second-preference votes change the outcome of the race from the plurality winner of the first round.
With the exception of the Coombs method, which is similar enough to instant-runoff voting to be essentially thought of as a varation of IRV, each of the preferential voting alternatives to IRV has significant shortcomings that make them less attractive in comparison. The Condorcet and Borda methods in particular favor consensus candidates, in some cases even over candidates who have the support of a majority of voters, and within the context of the American political system as it currently exists both systems would probably facilitate the election of corporate accomodationists. IRV and the Coombs method would thus seem to offer the most promise among preferential voting methods as an alternative to first-past-the-post voting. But what about range voting?
IV. Range and approval voting vs. IRV
In a range voting system, voters are asked to rate each of the candidates within a specified range, such as 0 to 99 or 1 to 10, and the scores from each ballot are added up to determine the winner of the race. Range voting does not appear to have ever been widely used, and isn't currently utilized in elections by any government. However, the idea of range voting has been around for a long time; a crude form of range voting was apparently used in some elections in ancient Sparta--winning candidates would be selected based on how loudly a gathered crowd shouted for them, in much the same way that winners are picked by Showtime at the Apollo, and losing candidates would then be bumrushed off the stage by a tap dancer (okay, I made that last part up).
Like the various methods of preferential voting, range voting allows participants to register a vote on all candidates, not just one. However, range voting goes a step further by allowing voters to quantify their support for candidates, registering not just whether they prefer one candidate over another but how much they prefer that candidate. Because of this aspect, supporters of range voting argue that it offers the best hope of breaking the Republican/Democratic duopoly in allowing voters to express their support for all the candidates they approve of, without the risk of splitting the vote on their side of the political continuum which is a prominent problem with plurality voting under the first-past-the-post system. Unlike the Borda system, range voting is not vulnerable to the phenomenon of strategic nomination, or "candidate cloning", in which a political faction actually improves the chances of one of its candidates winning an election by running more candidates rather than dividing the vote as might be the case in a first-past-the-post election. In range voting, candidates are neither helped nor hurt by the addition or subtraction of other candidates of candidates with similar political views. Range voting advocates also argue that it is cheaper and easier to administer than many other voting systems because it eliminates the need for runoff elections and would also lead to fewer spoiled ballots and less risk of fraud--a hanging chad or other such malfunction or misvote would only invalidate the part of your ballot pertaining to that candidate, and thus only that part of your vote would need to be discarded. Overvotes would no longer be a concern because the system is designed to accomodate votes for multiple candidates. In plurality voting, on the other hand, overvotes or any other form of misvote could cause the entire ballot to be discarded if the voter's intentions can't be discerned. This is no small concern--the number of spoiled ballots and undervotes in a typical presidential election can number in the millions, and in 2000, an unusually high number of rejected ballots in counties with large African-American populations (due in no small part to Republican voting machine hijinks) probably cost Al Gore Florida's electoral votes and the presidency during the 2000 presidential election.
In approval voting, as with traditional range voting, participants are allowed to cast a vote for as many candidates in an election as they wish. However, in the case of approval voting they are only allowed to indicate their approval for each candidate, and do not weight their vote with a score. It's generally classified as a form of range voting because voters are essentially given the option of assigning a score of either 1 or 0 for each candidate, but it might be better thought of as a hybrid system, combining the simplicity of first-past-the-post with the flexibility of range voting. Like traditional range voting, the approval system isn't currently utilized in elections by any government. However, it is used informally by the UN Security Council to narrow the list of potential candidates for Secretary General before a formal vote is held, and it has also been used in some municipal ballot propositions, as well as for party leadership elections in Pennsylvania. Several private organizations also use range voting, including the Mathematical Association of America, the Institute for Operations Research and the Management Sciences, and the American Statistical Association. Approval voting, like traditional range voting, has a long history behind it, having been used by the Republic of Venice to elected the Doge from the 13th to 18th centuries as well as by the Catholic Church to elect the Pope between 1294 and 1621.
Traditional range voting and approval voting are similar in some aspects, and because of this similarity the two have some of the same strengths. Like range voting, approval voting allows voters to register their opinion on all candidates in a race; it gives voters the option of voting for all the candidates they approve of in an election without splitting the vote on their side; it's not vulnerable to the practice of "candidate cloning"; it eliminates the need for runoff elections; and there is a reduced risk of spoiled ballots, as well as no risk of overvotes.
Range and approval voting both offer compelling alternatives to first-past-the-post voting, but how do they measure up against instant-runoff voting, and against each other? In defense of IRV, advocates point out the fact that range and approval voting, which they sometimes refer to derisively as "theoretical voting methods", have no track record in high-stakes elections to evaluate them by and are thus less likely to be implemented. IRV, on the other hand, either has been or is currently used in several jurisdictions and can be judged based on real-world examples. Range and approval voting advocates, however, have used IRV's track record against it, arguing that as with first-past-the-post voting (albeit to a lesser extent), IRV has a tendency to reinforce two-party dominance and would be a poor choice in attempting to open the American political system up to increased competition. In Australia, Ireland, and Fiji, three of the four countries where IRV has been implemented, two-party domination is the status quo despite these countries having other structural advantages that would tend to favor the emergence of a multiparty system. IRV detractors also argue that range and approval voting would be easier to implement here in America despite their relatively thin track record because, unlike IRV, they are both compatible with every voting machine currently in use without modification or reprogramming. Range and approval voting are both simpler for voters and election administrators than IRV and would thus be less susceptible to ballot disqualification due to voter error.
Detractors of range voting have argued that, despite the claims of some supporters that the system is less susceptible to tactical voting than other systems, it is in fact possible for voters to game the vote by giving maximum scores to their most-favored candidates and minimum scores to their least-favored candidates. Approval voting, in which voters have only two voting options (for or against) and thus automatically give each candidate either a maximum or minimum score, is likewise vulnerable to this criticism. This fact, however, has actually been offered by some advocates of approval voting as a feature, not a bug--they argue that in nonpartisan elections, political primaries and special elections, where candidate lists can be long and party affiliation isn't always available as a tool to aid voters in deciding who to support, approval voting gives voters who don't have strong preferences a simple way to vote for the candidates they do like without having to rank or rate them. This could be particularly useful in cases where voters don't have strong feelings one way or the other about most of the field, but strongly dislike one particular candidate for whatever reason--perhaps the candidate is an unabashed racist, as in the case of David Duke, or demonstrably unfit to hold office, as in the case of George Bush. In such cases, advocates argue, approval voting gives these voters a means of disadvantaging that candidate without having to differentiate among the rest of the field, by voting for everyone except that candidate. Voters within a range voting system can basically accomplish the same thing by simply giving maximum scores to every candidate in a race with the exception of the candidate they strongly dislike, and giving a minimum score to that candidate.
Range voting advocates assert that their system offers voters a greater range of expressivity than both IRV and approval voting, giving voters the opportunity to quantify their support for the candidates they prefer and (in comparison to IRV) to rank two or more candidates equally, if they so choose. This expressivity has the added benefit of arguably making the system less susceptible to ties and near-ties than IRV and approval voting, due to the wider range of possible voter opinion that can be expressed on a range voting ballot. However, IRV supporters have asserted that this expressivity is arguably undemocratic, and that both range and approval voting are flawed because they fail the majority criterion for judging voting systems. In other words, they can produce election results in which a candidate who represents the first preference of a majority of voters actually loses the election. This isn't the case with IRV--if a majority of voters list a candidate as their first preference, that candidate is guaranteed of winning the election. The Center for Voting and Democracy, which has been a prominent advocate of IRV, cites the following examples of how range and approval voting fail the majority criterion:
60% of voters think that candidate A is the absolute best choice, and that candidate C is horrible. They don't know much about B (B has avoided taking stands on any controversial issues), but believe almost anybody is better than C. 40% of voters like candidate C the best, and really hate candidate A. They don't know much about B, but believe almost anybody is better than A. The A supporters cast their ballots for both A and B (they include B to help assure that C won't win in case A isn't strong enough to win). The C supporters vote for both C and B (to help assure that A won't win in case C isn't strong enough to win). The outcome: A receives 60 votes, B receives 100 votes and C receives 40 votes. Candidate A loses, although an absolute majority of voters thought A was the best candidate. B wins, although not a single voter thought B was best. Because Approval Voting rewards candidates who avoid taking stands on issues that may alienate any voters, the quality of campaigns and public debate will decline in future elections as candidates seek to avoid revealing their policy intentions on controversial issues.
Under Range Voting it is possible for a candidate that 80% of voters think is the best choice to be defeated by a candidate that only a minority favor. A fundamental precept of Range Voting is that the intensity of a minority of voters' opinions about candidates should be able to trump a majority of voters with a different, but less vehement opinion. Advocates refer to this concept as "Bayesian regret," a sort of averaging of overall society satisfaction with an election outcome. There are many scenarios in which Range can defeat the candidate that the majority of voters prefer. Take as an example the election of Lyndon Johnson over Barry Goldwater in 1964 by 61% to 39%. If Johnson supporters saw him as a flawed candidate and some gave him a 2 and Goldwater a zero, and all Johnson voters together averaged a lukewarm score of 6 out of 10 (while giving Goldwater a zero), but Goldwater supporters gave their candidate an enthusiastic 10, the 39% minority would triumph.
Range and approval voting advocates, however, point out that while IRV does not fail the majority criterion, it can produce erratic results: for example, it's possible for a candidate in an instant-runoff election to be favored by a wide plurality of voters in a particular election and win every head-to-head Condorcet matchup by a 2-to-1 margin and still lose the election. Like range and approval voting, IRV is vulnerable to tactical voting and can encourage voters to vote for the "lesser of two evils". And there is some mathematical support for the idea that range and approval voting both have lower Bayesian Regret values than IRV, meaning that voters as a whole are less likely to be unhappy with the results of an election.
Range and approval voting both seem to offer better potential for making the American political system more competitive for minor parties than instant-runoff voting, but some range voting advocates argue that their system is more conducive to the formation of viable minor parties than approval voting because of a phenomenon known as the "nursery effect". According to this theory, voters participating in a range vote who might be inclined to give a minimum rating to major-party candidates that they dislike and a maximum rating to major-party candidates that they like will be more inclined to see no strategic threat in giving an honest rating above the minimum to minor-party candidates that they like, whereas voters participating in an approval vote, who don't have the same range of expressivity available to them, will be less likely to vote for minor-party candidates with all else being equal. Thus, range voting would have a gentler effect on minor parties and give them more time to grow into viability. This theory is supported by at least one study, which involved voters participating in the 2004 presidential election; however, it does not have a substantial body of evidence to support it. And approval voting advocates counter that their system is better positioned as a reform option because it is easier for voters to understand and simpler to administer than IRV and range voting.
V. Conclusion
So which is the superior voting system? IRV, range voting, and approval voting, while all demonstrably better than first-past-the-post voting, are all potentially susceptible to being undermined by tactical voting and the possibility of producing election results that to some degree don't reflect the wishes of voters. This possibility seems more pronounced in the case of range voting when taking the majority criterion into account, because it asks voters to make subjective quantifications of their support for individual candidates and thus opens the system up to the criticism of being incompatible with the concept of one-man, one-vote polling. However, range and approval voting seem to have an advantage over IRV in being more likely to offer the prospect of an open political system with viable minor parties. IRV has the advantage of real-world results from other political systems that aid in its evaluation and has a broader base of familiarity working in its favor, but from a logistical standpoint, range and approval voting would be easier to implement and administer. This is especially true in the case of approval voting, which would also be much more straightforward and easier to explain to most voters because of its similarity to first-past-the-post voting--rather than having to rank candidates or come up with a subjective score for them, voters can simply check off the candidates they like and skip the candidates they dislike.
Ultimately, I think that any of the three voting systems discussed in depth here is better than first-past-the-post voting and would reflect a significant improvement over the status quo. But taking into consideration how easy it would be to implement these systems, to administer them once they are in place, and what kind of results they would produce, approval voting appears to be the best option available for reform advocates to get behind in attempting to promote change within the context of the single-winner system that predominates in this country.
* * * * * * * * * * * * *
Adapted from the following articles:
Getting past first-past-the-post
A closer look at instant-runoff voting
Preferential voting: alternatives to IRV
Range and approval voting vs. IRV