User:DemocracyProfessor/Comparison of instant runoff voting to other voting systems

This article is a comparison of various voting systems with "Instant-runoff voting" (IRV), also called the "Alternative Vote", "preferential voting" and "ranked choice voting."

Categories

Voting systems fall into three broad types.

• Plurality/majority systems, sometimes called majoritarian.
• Proportional representation systems, sometimes called proportional.
• Semiproportional systems.

The types of various legislatures and their voting systems are given below.

Voting system	FPTP	AV	STV	PRO	Parallel	Notes
Category	Majoritarian	Majoritarian	Proportional	Proportional	Semi-proportional
Example legislature	UK HoC	AU HoR	IR DE	NE TK	JP SH

Voting system criteria

Political scientists rate voting systems using voting system criteria. Underlying these criteria are Arrow's Theorem and the Gibbard-Satterthwaite theorem, which presume that voters rank all candidates in a strict preference order, among other assumptions. No ranked preference method can satisfy all of the criteria, because some are mutually exclusive. Some of those criteria are given below, along with a list of voting systems. Voting systems that pass the criterion are labelled with a "Y", those that fail with an "N" and those that suffer an especially intolerable failure with an "IN".

Only three of these voting systems are used to elect candidates in one-winner elections held for national, regional and local government offices: First-past-the-post voting, FPTP; Instant-runoff voting, IRV; and Two-round systems of runoffs, RO. Because both systems involve automatic elimination of trailing candidates, every criteria failed by IRV is also failed by traditional runoffs, although IRV meets certain criteria that runoffs fail to meet.

Summary Table: Compliance of Several Voting Procedures to Various Voting Criteria

#	Criterion	Meaning of criterion	FPTP	IRV	Approval	Borda	Bucklin	Kemeny [-Young]	[Minimax] Condorcet /Simpson- Kramer	Range	Ranked pairs/ Tideman	Two- round/ runoff	Schulze	Coombs	Dodgson	Nanson	Copeland
1.2	Condorcet [Winner]	Pass-"if a candidate would win a head-to-head competition against every other candidate, then that candidate must win the overall election" OR Fail-"A candidate x is not elected despite the fact that it constitutes a 'Condorcet Winner', i.e., despite the fact that x is preferred by a majority of the voters over each of the other competing alternatives."	N	N	N	N	N	Y	Y	N	Y	N	Y	N	Y	Y	Y
1.3	[Absolute] Majority [Winner]	As a special case of the Condorcet Winner, to fail this means to also fail the Condorcet Winner. Pass-"if one candidate is rated first by an absolute majority of voters, then that candidate must win" OR Fail-"A candidate x is not elected despite the fact that x is preferred by a majority of the voters over each of the other competing alternatives."	Y	Y	IN	IN	Y	Y	Y	IN	Y	Y	Y	Y	Y	Y	Y
1.3j	Mutual Majority / Majority for Solid Coalitions / Generalized Majority	Similar to but more strict than the Majority Winner. Pass-"if a majority of voters prefer every member of some group of candidates to every candidate not in that group, then one of the preferred group must win"	IN	Y	IN	IN	Y	Y	IN	IN	Y	IN	Y	Y	IN	Y	Y
1.4	Condorcet Loser / Borda	Pass-"if a candidate would lose a head-to-head competition against every other candidate, then that candidate must not win the overall election" OR Fail-"A candidate x is elected despite the fact that it constitutes a 'Condorcet Loser' i.e., despite the fact that a majority of voters prefer each of the remaining candidates to x."	IN	Y	IN	Y	IN	Y	IN	IN	Y	Y	Y	Y	IN	Y	Y
1.5	Absolute Loser / Majority Loser	As a special case of the Condorcet Loser, to fail this means to also fail the Condorcet Loser. Pass-"if one candidate is rated last by an absolute majority of voters, then that candidate must not win" OR Fail-"A candidate x may be elected despite the fact that it is ranked last by a majority of voters."	IN	Y	IN	Y	Y	Y	IN	IN	Y	Y	Y	Y	Y-IN -	Y	Y
1.6	Pareto / Dominated Candidate	Pass-"ver.1:when every voter strictly prefers alternative a to alternative b, then alternative a must perform better than alternative b. ver.2:when no voter strictly prefers alternative b to alternative a, then alternative b must not perform better than alternative a." OR Fail-"A candidate x may be elected while candidate y may not be elected despite the fact that all voters prefer candidate y to x."	Y	Y	IN	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
2.1	Monotonicity / Additional Support / Mono-Raise / [Non-]Negative Responsiveness	To fail this, it is likely to also fail Participation / No-Show. Pass-"a voter cannot harm a candidate's chances of winning by voting that candidate higher, or help a candidate by voting that candidate lower, while keeping the relative order of all the other candidates equal." OR Fail-"If candidate x is elected under a given distribution of voters' preferences among the competing candidates, it is possible that, ceteris paribus, x may not be elected if some voter(s) increase their support for x by moving x to a higher position in their preference ordering."	Y	IN	Y	Y	Y	Y	Y	Y	Y	IN	Y	IN	IN	IN	Y
2.2	Consistency / Reinforcement / Multiple Districts / Separability / Convexity	Pass-"if two groups of voters both elect the same candidate, then a single group of all these voters should also elect that candidate." OR Fail-"If x is elected in each of several disjoint districts, it is possible that, ceteris paribus (all other things being equal), x will not be elected if all districts are combined into a single district."	Y	N	Y	Y	N	N	N	Y	N	N	N	N	N	N	N
2.4	Participation / No-Show	As an extreme version of the Truncation, to fail this, it is likely to also fail Truncation & Twin (neither tabulated). Pass-"the best way to help a candidate win must not be to abstain" OR Fail-"A voter may obtain a more preferable outcome if one decides not to participate in an election rather than, ceteris paribus, if one decides to participate in the election and vote sincerely for one's top preference(s)."	Y	N	Y	Y	N	N	N	Y	N	N	N	N	N	N	N
2.6	Independence of Irrelevant Alternatives (IIA/IIR) / Subset Choice (SCC)	Pass-"the election outcome remains the same even if a candidate who cannot win decides to run." OR Fail-"when candidate x is the unique winner, then x may become a loser whenever any of the original losers (irrelevant excessive alternatives) is added or removed and all other things remain the same."(e.g.vote splitting & spoiler effect)	N	N	N	N	N	Y-N -	N	Y	Y	N	Y	N	N	N	Y-N -
2.8	Strategic Voting / Tactical Voting	Pass-"a voter cannot improve the chances of a candidate by voting insincerely" OR Fail-"Ceteris paribus, a voter may obtain a preferred outcome if one votes strategically, i.e., not according to one's true preferences. All known voting procedures suffer from this paradox [to various degrees inversely proportional to complexity/runtime]."	N	N	N	N	N	N	N	N	N	N	N	N	N	N	N
3.a	Resolvability	Pass-"the probability of an exact tie must diminish [towards zero] as the number of votes cast increases [towards infinity]"	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	N
3.c	Reversal symmetry	Pass-"reversing the order of every ballot paper must alter the final winner"	N	N	Y	Y	N	Y	N	Y	Y	N	Y	N	N	Y	Y
3.e	Independence of Clones	Pass-"the election outcome remains the same even if an identical candidate who is equally-preferred decides to run."	N	Y	Y	N	N	N	N	Y	Y	N	Y	N	N	N	N
3.p	Polynomial [run]time	Cobham's thesis states that polynomial time is a synonym for "tractable", "feasible", "efficient", or "fast". Pass-"if its running time is upper bounded by a polynomial expression in the size of the input for the algorithm, i.e., T(n)=O(nk) for some constant k."	Y	Y	?	Y	Y	N	Y	?	Y	?	Y	?	N	Y	Y
9.z	Later-No-Harm	Pass-"if a voter alters the order of candidates lower in his/her preference (e.g. swapping the second and third preferences), then that does not affect the chances of the most preferred candidate being elected"	Y	Y	N	N	N	N	N	N	N	Y	N	?	N	N	N
	Total Especially Intolerable Failures	"there seems to be a wide consensus that a voting procedure which is susceptible to a 'cardinal sin' (i.e.which may elect a Pareto-dominated candidate, or elect an Absolute Loser, or display non-monotonicity, or not elect an Absolute Winner) should be disqualified as a reasonable voting procedure regardless of the probability that these paradoxes may occur."	3	1	5	2	1	0	3	4	0	2	0	1	3.5	1	0
	Total Failures		8	7	9	7	9	6.5	10	7	4	9	4	8	11.5	7	6.5
#	Criterion	"What does that mean?"	FPTP	IRV	Approval	Borda	Bucklin	Kemeny [-Young]	[Minimax] Condorcet /Simpson- Kramer	Range	Ranked pairs/ Tideman	Two- round/ runoff	Schulze	Coombs	Dodgson	Nanson	Copeland

Legend/Notes:
Voting systems that pass the criterion are labelled with a "Y", those that fail with an "N" and those that suffer an especially intolerable failure with an "IN".
Column headings highlighted in lime pertain to non-ranked procedures;
Column headings highlighted in pink pertain to ranked non-Condorcet-consistent procedures;
Column headings highlighted in yellow pertain to ranked Condorcet-consistent procedures.
Row headings highlighted in turquoise pertain to simple criteria (when failed is where the relevant data leads to a 'surprising' and arguably undesirable outcome.);
Row headings highlighted in grey pertain to conditional criteria (when failed is where changing one relevant datum while holding constant all other relevant data leads to a 'surprising' and arguably undesirable outcome.).
Data held in the "#" column denotes the same sequence of each tabulated criterion as listed by the references' authors. Criteria in need of references is tabulated last. Additional references are needed to verify claims and to help resolve (3) disputes between claims from existing references.

Tabulated references:

Voting system results

Are the results proportional?

Criterion	What does that mean?	FPTP	AV	Notes
Is it proportional?	"Does the number of seats awarded match the number of votes cast?"	No	No	See below

The intention of IRV is to find one candidate acceptable to a majority of voters. It is intended as an improvement on the 'first-past-the-post' (plurality) voting system. Under 'first-past-the-post' the candidate with the most votes wins, even if that candidate has less than a majority of votes and even if considered the worst candidate by a majority of voters (the "Condorcet loser"). IRV will always result in the defeat of the Condorcet loser. When used to elect legislative bodies, however, IRV can produce results that can be unrepresentative of voter preferences across the entire jurisdiction. Like all winner-take-all voting systems, IRV tends to exaggerate the number of seats won by the larger parties. For that reason some backers of proportional representation oppose IRV for legislative elections.

How unproportional are the results?

Average deviation from proportionality 1945-2010 (see description for sources).

FPTP   AV   STV   PRO

The "deviation from proportionality" for four systems since 1945 are as follows:

Criterion	What does that mean?	FPTP	AV	STV	PRO	Notes
Deviation from proportionality	"How unproportional is the voting system?"	29.4	27.3	12.0	5.22	See below

The "deviation from proportionality" is a linear scale that measures the degree of unfairness of a given voting system from 0 to a higher number. In the case above, 0 = perfectly proportional (seats assigned match the votes cast), and 200 = perfectly unproportional (a party with no votes gets all the seats).

Deviation from proportionality (adjusted scale) (see description for sources).

FPTP   AV   STV   PRO   AMS

Another study, (this time of 24 councils in the United Kingdom) came up with the result below:

Criterion	What does that mean?	FPTP	AV	STV	PRO	AMS	Notes
Deviation from proportionality (adjusted scale)	"How unproportional is the voting system?"	22%	16%	7%	7%	3%	See below

That study adjusted the score so that 0% was perfectly proportional and 100% was the most unproportional score possible in practice.

What are the results under different systems?

How well do different political parties do under different systems? Some results are given below:

Criterion	What does that mean?	FPTP	AV	STV	PRO	Notes
UK Conservative 2010 seats	"How many seats would Conservatives have in 2010 under this system?"	306	281	246	234
UK Labour 2010 seats	"How many seats would Labour have in 2010 under this system?"	258	262	207	189
UK Liberal 2010 seats	"How many seats would Liberals have in 2010 under this system?"	57	79	162	149

A different study focusing just on FPTP and AV gave these results:

Criterion	What does that mean?	FPTP	AV	Notes
UK Conservative 1983 seats	"How many seats would Conservatives have in 1983 under this system?"	397	391
UK Labour 1983 seats	"How many seats would Labour have in 1983 under this system?"	209	190
UK Liberal 1983 seats	"How many seats would Liberals have in 1983 under this system?"	23	48
UK Conservative 1987 seats	"How many seats would Conservatives have in 1987 under this system?"	376	381
UK Labour 1987 seats	"How many seats would Labour have in 1987 under this system?"	229	202
UK Liberal 1987 seats	"How many seats would Liberals have in 1987 under this system?"	22	44
UK Conservative 1992 seats	"How many seats would Conservatives have in 1992 under this system?"	336	328
UK Labour 1992 seats	"How many seats would Labour have in 1992 under this system?"	271	268
UK Liberal 1992 seats	"How many seats would Liberals have in 1992 under this system?"	20	31
UK Conservative 1997 seats	"How many seats would Conservatives have in 1997 under this system?"	165	70
UK Labour 1997 seats	"How many seats would Labour have in 1997 under this system?"	419	445
UK Liberal 1997 seats	"How many seats would Liberals have in 1997 under this system?"	46	115
UK Conservative 2001 seats	"How many seats would Conservatives have in 2001 under this system?"	166	140
UK Labour 2001 seats	"How many seats would Labour have in 2001 under this system?"	413	423
UK Liberal 2001 seats	"How many seats would Liberals have in 2001 under this system?"	52	68
UK Conservative 2005 seats	"How many seats would Conservatives have in 2005 under this system?"	198	171
UK Labour 2005 seats	"How many seats would Labour have in 2005 under this system?"	356	377
UK Liberal 2005 seats	"How many seats would Liberals have in 2005 under this system?"	62	68

Are the results decisive?

How decisive are elections under different systems? Some results are given below:

Criterion	What does that mean?	FPTP	AV	Notes
1983 UK election outcome	"Would there have been a "hung parliament" in 1983 in the UK under this system?"	No	No
1987 UK election outcome	"Would there have been a "hung parliament" in 1987 in the UK under this system?"	No	No
1992 UK election outcome	"Would there have been a "hung parliament" in 1992 in the UK under this system?"	No	No
1997 UK election outcome	"Would there have been a "hung parliament" in 1997 in the UK under this system?"	No	No
2001 UK election outcome	"Would there have been a "hung parliament" in 2001 in the UK under this system?"	No	No
2005 UK election outcome	"Would there have been a "hung parliament" in 2005 in the UK under this system?"	No	No
2010 UK election outcome	"Would there have been a "hung parliament" in 2010 in the UK under this system?"	Yes	Yes

No overall majority

Incidence of "No Overall Majority" (see description for sources).

FPTP   AV   STV   PRO   AMS

"No overall majority" is achieved when no single party has more seats than the other parties combined. How often does this happen? Some results are below:

Criterion	What does that mean?	FPTP	AV	STV	PRO	AMS	Notes
No overall majority	"How often will no single party have more seats than the other parties combined?"	16.7% (4/24)	16.7% (4/24)	37.5% (9/24)	37.5% (9/24)	54.1% (13/24)

Ballot issues

Number of ballots per voter

The "number of ballots per voter" is the number of times a voter has to complete a single ballot.

Criterion	What does that mean?	Single-round Plurality (inc. FPTP)	Preferential (inc. AV)	Multiple-round (inc. EB and RO)	Notes
Number of ballots per voter	"How many times does a voter have to complete a single ballot?"	Once	Once	More than once

Voter choice

"Voter choice" is the number of candidates on the ballot.

Criterion	What does that mean?	Single-round Plurality (inc. FPTP)	Preferential (inc. AV)	Multiple-round (inc. EB and RO)	Notes
Voter choice	"Which system gives the higher number of candidates"	Plurality gives fewer candidates than preferential or multiple-round	Preferential gives more candidates than plurality	Multiple-round gives more candidates than plurality	See below

Like the two-round system, IRV tends to give voters a wider choice of candidates than plurality. More independent and third party candidates are likely to run because the spoiler problems are less severe. However, most jurisdictions (regardless of voting system) limit the total number of candidates by requiring deposits, a large number of nominators, or other measures.

Voter exhaustion

"Incomplete ranking" occurs when a voter fails to list a preference for all the candidates. The ballots on which incomplete ranking has occurred are called "exhausted ballots". The result is referred to as "voter exhaustion".

Criterion	What does that mean?	FPTP	AV (with OPV)	AV (without OPV)	Notes
Exhausted ballots	"Does the system prevent exhausted ballots being counted?"	No	No	Yes	FPTP fails exhaustion by definition. AV fails it if exhausted ballots are counted. See below.

There are two sources of incomplete ranking:

• Some implementations don't allow complete ranking, either due to voting machine limitations or other reasons; for example, San Francisco allows only three ranks, even with over twenty candidates.
• Some voters may simply choose not to rank all the candidates.

To avoid exhaustion, Australia generally does not count exhausted ballots. But New South Wales and Queensland do count such ballots (a variant known as optional preferential voting, or OPV). Antony Green notes that "The exhaustion rate has approached 80% in some seats....optional preferential voting almost always assists the party with the highest primary vote."

Voter confusion

"Voter confusion" takes place when a voter cannot discern who to vote for.

Criterion	What does that mean?	STV	AV	Notes
How-to-vote cards	"Are how-to-vote cards used?"	Yes	Yes	See below

To cast a valid ballot, Australian voters must rank every candidate. Australian political parties use "how to vote" cards to encourage votes for their candidates. Without compulsory rankings, voters would have the option to stop ranking when indifferent to their remaining choices.

Invalid ballots

Latest incidence of invalid ballots (see description for sources).

FPTP   AV

"Invalid ballots" are ballots that cannot be counted. The most recent rate of invalid ballots for two systems is as follows:

Criterion	What does that mean?	FPTP	AV	Notes
Invalid ballots rate	"What is the most recent percentage of votes that cannot be counted?"	1.0% (overall), 0.28% (non-postal votes), 3.8% (postal votes)	5.55%	See below

"Invalid ballots" are called "informal ballots" in Australia (where voting is compulsory) and "spoilt ballots" in the United Kingdom (where voting is optional).

Voter participation

Latest turnout (see description for sources).

FPTP   AV

"Turnout" is the number of people who do vote compared with the number of people who can. The most recent turnouts for two systems is as follows:

Criterion	What does that mean?	FPTP	AV	Notes
Turnout rate	"What is the most recent percentage of people who can vote that do vote?"	65.1%	93.21%	See below

Voting is compulsory in Australia and optional in the United Kingdom.

Risk of fraud

Risk of fraud in the United States

Fraudulent counting in the United States takes place when the count is interfered with.

Criterion	What does that mean?	FPTP	AV	Notes
Fraudulent counting in the USA	"Are elections fraudulent in the USA?"	No	No	See below for the discussion

Most counties in the United States use voting machines to register and count their votes at the polling place. Although first choices can be counted at the polling place, AV ballots usually are counted in a central location in races without a first round majority winner. Changing the voting system by counting the vote centrally may increase the risk of fraud or may not increase the probability of fraud associated with such machine counts. FPTP need not be counted centrally.

Risk of fraud in Australia

Fraudulent counting in Australia takes place when the count is interfered with.

Criterion	What does that mean?	STV	AV	Notes
Fraudulent counting in Australia	"Are elections fraudulent in Australia?"	No	No	See below for the discussion

Australia (except for ACT) uses paper and pencils to vote and counts the votes by hand in a central location after unofficial tallies at the polling place. The fear of fraud is less.

Tactical voting

Main articles: Tactical voting, Vote-splitting, and Spoiler effect

"Vote-splitting" occurs when voters split their votes between similar candidates, allowing a dissimilar candidate to win. One or more of the similar candidates may be characterised as a "spoiler" and the result as the "spoiler effect". "Tactical voting" occurs when a voter lists an "insincere preference" for a candidate due to the "spoiler effect". An "insincere preference" is when a voter deliberately gives a candidate a dishonest preference to prevent somebody winning (or another undesirable outcome).

Criterion	What does that mean?	FPTP	AV	Notes
Is tactical voting possible?	"In theory, can people vote insincerely to prevent somebody winning?"	Yes	Yes	Tactical voting is theoretically possible in both FPTP and AV elections, see below
Does tactical voting occur in practice?	"In actual elections, do people vote insincerely to prevent somebody winning?"	Yes	No	Tactical voting occurs less in AV elections, see below
Does the spoiler effect occur?	"Can vote-splitting happen?"	Yes	Yes	See below
Does the spoiler effect occur when there are two major parties?	"Does vote-splitting happen when there are two major parties?"	Yes	No	See below

Is tactical voting possible in theory?

The Gibbard–Satterthwaite theorem demonstrates that no voting system can be entirely immune from tactical voting unless it is dictatorial (there is only one person who is able to choose the winner) or incorporates an element of chance.

IRV permits tactical voting if voters have complete and reliable information about the other voters' full preferences. The 2009 mayoral election in Burlington, Vermont provides an example where tactical voting could have happened: most supporters of the candidate who came in second (and led the first-round) preferred the Condorcet winner (who IRV gave third place) over the IRV winner. If some of these voters had known that beforehand and insincerely raised their second-choice to first-choice, the Condorcet winner would have won the instant runoff, which these voters would have preferred. In a three-party election where voters for both the left and right prefer the centrist candidate to stop the "enemy" candidate winning, those voters on the left and right who care more about defeating the "enemy" than electing their own candidate may cast a tactical first preference vote for the centrist candidate.

Does tactical voting occur in practice?

Under FPTP, voters have an incentive to vote insincerely for one of the two major candidates instead of their true favorite, because a vote for their true favorite is likely to be "wasted."

In his book Collective Decisions and Voting Nicolaus Tideman uses real-world voting data to analyze all proposed election methods in terms of resistance to tactical voting, and states on page 194 that "alternative vote [IRV] is quite resistant to strategy."

Does the spoiler effect occur?

IRV removes the spoiler effect when there are two major candidates and one or more minor candidates. In Australia, (a nation that uses IRV for its House of Representatives election) a smaller third party co-exists with its coalition partner, and functions without losing seats to other parties due to vote splitting. In Australia's national elections in November 2007, at least four candidates ran in every constituency, with an average of seven, and every constituency was won with an absolute majority of votes including several where results would have been different under plurality voting.

In the United States of America (a nation where states use FPTP for allocating Electoral College votes in Presidential election), third party and independent candidates are often characterised as "spoilers" - such as in the 2000 Bush/Gore Presidential race where Ralph Nader was described as drawing support away from Al Gore, allowing George W. Bush to win.

Voter power

One person, one vote

"One person, one vote" (sometimes called "One man, one vote") is the principle that no voter be given greater weight in his or her vote over the vote of another voter.

Criterion	What does that mean?	FPTP	AV	Notes
One man, one vote	"Does every voter get the same number of votes and are they counted equally?"	Yes	Yes	See below

In Ann Arbor, Michigan arguments over IRV in letters to newspapers included the belief that IRV "gives minority candidate voters two votes," because some voters' ballots may count for their first choice in the first round and a lesser choice in a later round. This argument was addressed and rejected by a Michigan court in 1975. In Stephenson v. the Ann Arbor Board of City Canvassers, the court held that "majority preferential voting" (as IRV was then known) was not a violation of one-man, one-vote.

“ Under the "MPV System", however, no one person or voter has more than one effective vote for one office. No voter's vote can be counted more than once for the same candidate. In the final analysis, no voter is given greater weight in his or her vote over the vote of another voter, although to understand this does require a conceptual understanding of how the effect of a "MPV System", is like that of a run-off election. The form of majority preferential voting employed in the City of Ann Arbor's election of its Mayor does not violate the one-man, one-vote mandate nor does it deprive anyone of equal protection rights under the Michigan or United States Constitutions. ” — Judge James Fleming, Stephenson v. the Ann Arbor Board of City Canvassers, 1975

Power of an individual vote

Voter Power Index (see description for sources).

FPTP   AV

The "Voter Power Index" (VPI) for two systems is as follows:

Criterion	What does that mean?	FPTP	AV	Notes
Voter Power Index	"What is the power of an individual vote?"	0.285	0.352	See below

The "Voter Power Index" (VPI) is a linear scale that measures how powerful an individual vote is. In the case above, 0 = perfectly powerless (a vote in a very safe constituency of very large size), and 1 = perfectly powerful (a vote in a very marginal constituency of average size).

Voter equality

Voter equality (see description for sources).

FPTP   AV

"Voter equality" compares the least powerful fifth to the most powerful fifth of voters. Some results are below:

Criterion	What does that mean?	FPTP	AV	Notes
Voter equality	"How equal are voters in terms of their power?"	4.8% (1/21)	5.6% (1/18)	See below

The "Voter equality" is the VPI for the least powerful fifth of voters divided by the VPI for the most powerful fifth of voters. In the case above, 0% = perfectly unequal (the least powerful fifth have no power) and 100% = perfectly equal (the least powerful fifth have the same power as the most powerful fifth).

Marginality

A constituency is "marginal" if there is a 1 in 5 chance or more that it will change hands in an election. A constituency that is not "marginal" is called "safe". How many constituencies would be safe in the United Kingdom (excluding Northern Ireland) in the 2010 election under different systems? Some results are below:

Criterion	What does that mean?	FPTP	AV	Notes
Very Marginal (GB)	What would the number of Very Marginal constituencies be under this system?	81	125	More than 1 in 3 chance of changing hands
Marginal (GB)	What would the number of Marginal constituencies be under this system?	85	82	Between 1 in 5 and 1 in 3 chance of changing hands
Fairly Safe (GB)	What would the number of Fairly Safe constituencies be under this system?	135	154	Between 1 in 10 and 1 in 5 chance of changing hands
Very Safe (GB)	What would the number of Very Safe constituencies be under this system?	175	158	Between 1 in 25 and 1 in 10 chance of changing hands
Ultra Safe (GB)	What would the number of Ultra Safe constituencies be under this system?	156	113	Less than 1 in 25 chance of changing hands

The same study covered the United Kingdom (including Northern Ireland). The results are below:

Criterion	What does that mean?	FPTP	AV	Notes
Very Marginal (UK)	What would the number of Very Marginal constituencies be under this system?	82	126
Marginal (UK)	What would the number of Marginal constituencies be under this system?	86	86
Fairly Safe (UK)	What would the number of Fairly Safe constituencies be under this system?	142	159
Very Safe (UK)	What would the number of Very Safe constituencies be under this system?	180	162
Ultra Safe (UK)	What would the number of Ultra Safe constituencies be under this system?	160	117

"Kicking the rascals out"

Average churn rate 1945-2010 (see description for sources).

FPTP   STV   AV   PRO

The "average churn rate" (characterised in the UK as "kicking the rascals out") for four systems since 1945 are as follows:

Criterion	What does that mean?	FPTP	AV	STV	PRO	Notes
Average churn rate	"How easy is it to kick the rascals out?"	17.3%	19.0%	17.4%	21.0%	See below

The "average churn rate" is a percentage that measures how easy it is to remove an elected member - the higher the number, the easier to "kick the rascals out". In the case above, 0% = impossible to remove any member in any election, 100% = every member loses every seat in every election.

Cost

Operating cost

On 2010-08-21 the 2010 Australian federal elections were held. The cost has not yet been released. On 2007-11-24 the 2007 Australian federal elections were held. The cost have been released. It does not give a quote for the cost of the House of Representatives election (held under AV) separately from the cost of the Senate election (held under STV) held on the same day. The costs are below.

On 2010-05-06 the 2010 UK general election was held. The cost has not yet been released but an estimate of the cost has. It does give a quote for the cost of the House of Commons election (held under FPTP) separately from the cost of the local elections (also held under FPTP) held on the same day. The costs are below.

Voting system	FPTP	AV/STV	Notes
Date of election	2010-05-06	2007-11-24
Cost of election (excluding candidate funding) in local currency	£92,100,000	$114,073,467
Local currency	GBP	AUD
Exchange rate on the day of election (local->GBP)	1	0.4251038342
Cost of election (excluding candidate funding) on the day of election in GBP	£92,100,000	£48,493,068.20
Number of people who voted validly	29,687,604	12,419,992
Cost of election (excluding candidate funding) on the day of election in GBP per person who voted validly	£3.10	£3.90

Because it does not require two separate votes, AV costs less than two-round primary/general or general/runoff election systems.

Conversion cost in the USA

The "conversion cost in the USA" is the cost to convert from one voting system to another in the USA.

Criterion	What does that mean?	FPTP->AV	Notes
Conversion cost in the USA	"How much would it cost to convert to AV in the USA?"	See below	See below

Writing in the Canadian Journal of Political Science, Harold J. Jansen studied the Alternative Vote in Canada, concluding that "On balance, it differed little from the single member plurality system."

The United States uses machines to register their votes and machines to count their votes. When voting takes place on machines relying on software, changing the voting system can lead to new costs. Pierce County, Washington election officials outlined costs of $3,291,340 to implement IRV for its elections in 2008, covering software and equipment, voter education, testing, staff time, consultants and ballot printing and postage costs. In 2009 the auditor [chief elections director of Washington counties] reported ongoing costs that were not necessarily balanced by the costs of eliminating runoffs for most county offices, as those elections may be needed for other offices not elected by IRV. Other jurisdictions have reported immediate cost savings.

Conversion cost in Australia

The "conversion cost in Australia" is the cost to convert from one voting system to another in Australia.

Criterion	What does that mean?	FPTP->AV	Notes
Conversion cost in Australia (AV)	"How much did it cost to convert to AV in Australia?"	See below	See below

Australia (except for ACT) uses paper and pencils to vote and counts the votes by hand in a central location. The cost of running an election per capita is greater Australia, however, as AV involves more than one round.

Rules of Order

Main article: Rules of order

"Rules of Order" are rulebooks that recommend how parliamentary assemblies should be run. Their comparative assessment of voting systems are given below

Criterion	What does that mean?	Plurality (including FPTP)	Preferential (including AV)	Multiple-round (including Exhaustive ballot and Two-round system)	Notes
Robert's Rules of Order	"Which system does that rulebook recommend?"	That rulebook ranks plurality below preferential and multiple-round systems	That rulebook ranks preferential above plurality but below multiple-round systems	That rulebook ranks multiple-round systems above preferential and plurality	See below
The Standard Code of Parliamentary Procedure	"Which system does that rulebook recommend?"	That rulebook ranks plurality below preferential and multiple-round systems	That rulebook ranks preferential above plurality but below multiple-round systems	That rulebook ranks multiple-round systems above preferential and plurality	See below
Riddick's Rules of Procedure	"Which system does that rulebook recommend?"	That rulebook ranks plurality below preferential and multiple-round systems	That rulebook ranks preferential above plurality but below multiple-round systems	That rulebook ranks multiple-round systems above preferential and plurality	See below

The sequential elimination method used by IRV is described in Robert's Rules of Order Newly Revised, 10th edition. as an example of "preferential voting," a term covering "any of a number of voting methods by which, on a single ballot when there are more than two possible choices, the second or less-preferred choices of voters can be taken into account if no candidate or proposition attains a majority. While it is more complicated than other methods of voting in common use and is not a substitute for the normal procedure of repeated balloting until a majority is obtained, preferential voting is especially useful and fair in an election by mail if it is impractical to take more than one ballot. In such cases it makes possible a more representative result than under a rule that a plurality shall elect...."Preferential voting has many variations. One method is described ... by way of illustration." And then the instant runoff voting method is detailed.

Robert's Rules continues: "The system of preferential voting just described should not be used in cases where it is possible to follow the normal procedure of repeated balloting until one candidate or proposition attains a majority. Although this type of preferential ballot is preferable to an election by plurality, it affords less freedom of choice than repeated balloting, because it denies voters the opportunity of basing their second or lesser choices on the results of earlier ballots, and because the candidate or proposition in last place is automatically eliminated and may thus be prevented from becoming a compromise choice."

Two other books on parliamentary procedure take a similar stance, disapproving of plurality voting and describing preferential voting as an option, if authorized in the bylaws, when repeated balloting is impractical: The Standard Code of Parliamentary Procedure and Riddick's Rules of Procedure.

Political parties

One-party dominance

"One-party dominance" is achieved when the opposition is so small that it cannot act as an opposition. How often does this happen? Some results are below:

Criterion	What does that mean?	FPTP	AV	STV	PRO	AMS	Notes
One-party dominance (90% plus)	"How often will the ruling party have 90% or more of the seats?"	16.7% (4/24)	25.0% (6/24)	0% (0/24)	0% (0/24)	0% (0/24)	See below
One-party dominance (70%-90%)	"How often will the ruling party have between 70%-90% of the seats?"	20.8% (5/24)	12.5% (3/24)	0% (0/24)	8.3% (2/24)	0% (0/24)	See below
One-party dominance (70% plus)	"How often will the ruling party have 70% or more of the seats?"	37.5% (9/24)	37.5% (9/24)	0% (0/24)	8.3% (2/24)	0% (0/24)	See below

If 90 per cent of council seats are held by a single party, opposition is effectively neutered. If 70 per cent of council seats are held by the majority party there will be some effective opposition but it can always be outvoted.

Conduct (ethnic campaigning)

Main article: Ethnic separatism

"Ethnic campaigning" occurs when a candidate attacks another candidate's ethnicity to encourage the voter to vote against that candidate.

Criterion	What does that mean?	FPTP	AV	Notes
Ethnic conflict	"Does this system decrease ethnic conflict?"	No	Yes	See below

Benjamin Reilly suggests instant runoff voting eases ethnic conflict in divided societies. This was a leading reason why Papua New Guinea adopted instant runoff voting. The result in Fiji which adopted it for the same reason were not encouraging.

Conduct (negative campaigning)

Main article: Negative campaigning

"Negative campaigning" occurs when a candidate attacks another candidate to encourage the voter to vote against that candidate.

Criterion	What does that mean?	FPTP	AV	Notes
Negative campaigning	"Does this system decrease negative campaigning?"	No	See below	See below

John Russo, Oakland City Attorney, argued in the Oakland Tribune on July 24, 2006 that "Instant runoff voting is an antidote to the disease of negative campaigning. IRV led to San Francisco candidates campaigning more cooperatively. Under the system, their candidates were less likely to engage in negative campaigning because such tactics would risk alienating the voters who support 'attacked' candidates", reducing the chance that they would support the attacker as a second or third choice. Others state that there is a lack of evidence that such an effect occurs as often as suggested. Indeed, Lord Alexander's objections to the conclusions of the British Independent Commission on the Voting System's report cites the example of Australia saying "their politicians tend to be, if anything, more blunt and outspoken than our own."