Alternatives to First-Past-The-Post Voting


by lpetrich
Tags: alternatives, firstpastthepost, voting
lpetrich
lpetrich is offline
#1
Dec24-11, 12:20 AM
P: 514
First-past-the-post or plurality voting is the dominant system in the US and in the UK's national government. You vote for one candidate, and the winner is whoever gets the largest fraction of the votes.

While that is a simple system, and while it works well when there are only two candidates, it does not work so well for more than two candidates. In fact, fear of wasting votes on unlikely-to-win candidates usually leads to a two-party system. This behavior was described in the 1950's and 1960's by a certain Maurice Duverger, and it's now called Duverger's law (Wikipedia).

However, there are several alternatives that are much more multiple-candidate-friendly, and many nations and regions and cities and organizations now use some of them.

Approval Voting: you can vote for more than one candidate. As with FPTP, whoever gets the most votes wins. It's a subset of Range Voting, where one can give each candidate a fraction of a vote from 0 to 1. It can be extended to include "disapproval voting", where you can cast a vote of -1 as well as 0 or +1. The UN Security Council has used that system to vote for a Secretary General: RangeVoting.org - Election (via approval voting) of UN secretary general

Polls created with the vBulletin software can be either FPTP (single choice) or approval voting (multiple choice).

Runoff elections. If nobody gets a majority in FPTP-style voting, then have a second election between the top candidates.

Preference voting. Rank the candidates by preference, then find an overall winner with some algorithm. Borda: for n candidates, one's kth preference gets a value (n-k+1), and these values are added for all candidates. Condorcet or Virtual Round Robin: from each ballot construct a matrix of who beats whom. Add all these matrices together, and from the total one, find who beats the others. There are various algorithms for disambiguating ambiguous cases, like the Schulze beatpath algorithm. Instant Runoff: each ballot gets a pointer that's started off at its top preference. In each round, count the pointed-to preferences, and if a candidate gets a majority, then that candidate wins. Otherwise, remove the candidate with the fewest votes, bump the pointers of that candidate's ballots down to the first candidate still in the race, and then have another round of counting.

These algorithms can be extended to multiseat elections in various ways.

Single Transferable Vote: set a quota of (total votes) / ((number of seats) + 1). Count like IRV, but first remove candidates with more than the quota as winners. Over-quota ballots get their pointers bumped down like defeated candidates' ones.

Party-List Proportional Representation: you vote for a party, and each party gets a number of seats in proportion to how many votes it had received. There are various algorithms for coming up with integer numbers of seats for each party, like the d'Hondt method. Party-list PR can be closed list, where the parties select who gets the seats, or open list, where you can vote for whomever you think ought to have priority.

Mixed-Member Proportional Representation: a mixture of single-member districts and party-list PR.

Wikipedia's Voting system has a comprehensive list.

Wikipedia's contributors have collected a Table of voting systems by country. It's rather eye-opening to see what different countries have chosen.

For presidents, runoff elections are most common, though FPTP ones are common. If the US had popular-vote presidential elections with runoffs, then several of them would have gone into runoff. Would Carter have beaten Reagan? Bush I or Dole beaten Clinton? Gore beaten Bush II?

For legislatures, party list is the most common, though mixed member and FPTP have sizable showings, followed by runoff elections. Most of the FPTP users are the UK and former British-Empire countries, including certain well-known rebellious British colonies. But even in the Anglosphere, as it may be called, some nations have moved to PR and PR-like systems, like Australia and New Zealand, and in the UK, Northern Ireland, Scotland, and Wales.

In the US, it is easier than one might think to implement PR. The US Constitution nowhere mandates single-member districts for Representatives; it states that they are to be elected state-by-state. So a state could elect is Representatives by PR if it so chose. A big plus of PR is that it enables minorities to be more easily represented; not only racial and ethnic and religious ones, but also political and ideological ones. One can see this in action in PR systems, which sometimes have several parties. This means that Libertarians, for instance, can more easily get seats under PR than under FPTP.
Phys.Org News Partner Science news on Phys.org
SensaBubble: It's a bubble, but not as we know it (w/ video)
The hemihelix: Scientists discover a new shape using rubber bands (w/ video)
Microbes provide insights into evolution of human language
ThomasT
ThomasT is offline
#2
Dec26-11, 12:48 AM
P: 1,414
What, exactly, is your point or question?
lpetrich
lpetrich is offline
#3
Dec26-11, 02:15 AM
P: 514
Quote Quote by ThomasT View Post
What, exactly, is your point or question?
It's that there are lots of alternatives to first-past-the-post, alternatives that many Americans and other inhabitants of FPTP-using countries may be unaware of. These alternatives often do better than FPTP in making possible fairness of representation and ability to challenge existing political parties.

The question of US third parties comes up every now and then, and I did not want to clutter other threads with discussions of third-party-friendly voting systems.

ThomasT
ThomasT is offline
#4
Dec26-11, 05:16 AM
P: 1,414

Alternatives to First-Past-The-Post Voting


Quote Quote by lpetrich View Post
It's that there are lots of alternatives to first-past-the-post, alternatives that many Americans and other inhabitants of FPTP-using countries may be unaware of.
Then, imo, good point and, imo, I think you're correct.

Quote Quote by lpetrich View Post
These alternatives often do better than FPTP in making possible fairness of representation and ability to challenge existing political parties.
Ok.

Quote Quote by lpetrich View Post
The question of US third parties comes up every now and then, and I did not want to clutter other threads with discussions of third-party-friendly voting systems.
Ok. I think the US system has evolved (is designed ?) to discourage real political dissent. Effectively, imo, the US has one political party which represents the interests of the rich, the status quo oligarchy.

The best thing that US voters can do to effect real change, imo, is to simply stop voting. This would bring to the forefront the hypocrisy of the 'system' and perhaps be a precursor to positive change in American politics.
ThomasT
ThomasT is offline
#5
Dec26-11, 02:17 PM
P: 1,414
Quote Quote by lpetrich View Post
It's that there are lots of alternatives to first-past-the-post, alternatives that many Americans and other inhabitants of FPTP-using countries may be unaware of. These alternatives often do better than FPTP in making possible fairness of representation and ability to challenge existing political parties.

The question of US third parties comes up every now and then, and I did not want to clutter other threads with discussions of third-party-friendly voting systems.
How about this?:
http://www.americanselect.org/
lpetrich
lpetrich is offline
#6
Dec26-11, 10:15 PM
P: 514
Looks like a third party to me.

I looked at the Debates section but I couldn't search it. Looking through the first few pages I found no mention of voting-system reform, however.
JDStupi
JDStupi is offline
#7
Dec31-11, 01:36 PM
P: 111
I do not have much to add to this discussion as of now because I actually have not read enough of the relevant materials. I had a class which touched on this subject towards the end and unfortunately I did not pay as much attention as I should have and so I must go back and learn it, which is why I do not have much to add. In any case, I have attatched two papers which may be of interest to the people who want to know more of what the OP is talking about. Also if you follow this link http://brian.weatherson.org/LectureNotes.shtml you will find a link to the lecture notes for the course I took and you can go to the last chapters on group decisions and voting systems. Enjoy
Attached Files
File Type: pdf Alternatives to Majority Rule.pdf (91.5 KB, 0 views)
File Type: pdf The Nonobviousness of Majority Rule I.pdf (89.9 KB, 0 views)
jaystar
#8
Feb13-12, 05:22 PM
P: n/a
FPTP has some merit as a way of electing a single winner in a single member district, but is not a good way of electing a parliament/legislature or government.
An alternative you didn't mention, but might like to consider is Direct Party and Representaive Voting (see www.dprvoting.org)
BobG
BobG is offline
#9
Feb13-12, 08:42 PM
Sci Advisor
HW Helper
BobG's Avatar
P: 2,275
I'd go with Approval Voting if I had my preference.

It comes closer to reaching a consensus on an acceptable candidate vs the system we currently have in the US. Between the primaries and the general election, a voter is often confronted with a choice between which major party candidate he dislikes the least.

In my personal experience for Presidential elections:

Carter vs Ford. I wasn't enthusiastic about either, but actually found both acceptable. In Ford's case, my lack of enthusiasm was because I thought it would have been better to start with a clean slate after Watergate, but I did wind up voting for him.

Reagan vs Carter. Was even less enthusiastic about Carter. Definitely wouldn't vote for Reagan.

Reagan vs Mondale. I was wrong about Reagan - he actually turned out to be a good President. No way I would have ever voted for Mondale.

Bush vs Dukakis. I wanted Bush to beat Reagan in the 1980 primaries and I still liked him. Did not like Dukakis.

Clinton vs Bush. I still liked Bush. Did not like Clinton.

Clinton vs Dole vs Perot. Disgusted with both Clinton and Dole after the government shutdown, so I voted for the cranky, senile guy.

Bush vs Gore. Did not like either. That was a tough choice and I eventually decided Bush was the lesser of two weevils.

Bush vs Kerry. I was wrong about Bush. He was even dumber than I ever could have imagined. Didn't like Kerry, either, but at least he wasn't Bush.

Obama vs McCain. Liked both, but I would have liked McCain better when he was young enough to be likely to complete a full term. He would have been a good choice in 2000, but not in 2008.

So I voted for the guy I liked 5 times (counting acceptable as like), and a person I didn't really like 4 times. That's just not a good percentage.
nicoleannB
nicoleannB is offline
#10
Feb18-12, 02:23 AM
P: 2
Interesting. Was that really effective?
KingNothing
KingNothing is offline
#11
Feb18-12, 09:53 PM
P: 949
I think this would discourage less intelligent or less informed voters. Maybe that's a good thing.
BobG
BobG is offline
#12
Feb19-12, 07:26 AM
Sci Advisor
HW Helper
BobG's Avatar
P: 2,275
Quote Quote by KingNothing View Post
I think this would discourage less intelligent or less informed voters. Maybe that's a good thing.
Just eliminating party affiliation from ballots would discourage less informed voters. How many voters just vote for the candidate with the right letter after his name in spite of knowing absolutely nothing about that candidate - especially for minor offices such as County Coroner, County Engineer, etc.
DoggerDan
DoggerDan is offline
#13
Feb21-12, 07:02 PM
P: 77
Quote Quote by lpetrich View Post
Borda: for n candidates, one's kth preference gets a value (n-k+1), and these values are added for all candidates.
Where n is the number of candidates?

Hypothetically, if three people were running for President:

Obama
Romney
DoggerDan

then n=3, and if I ranked them like:

DoggerDan
Romney
Obama

The values would be:

DoggerDan = 3-1+1 = 3
Romney = 3-2+1 = 2
Obama = 3-1+1 = 1

I think that's a complex way of saying "rank order them."

We used to rank order decisions where there wasn't a clear financial difference. We could have debated it endlessly, but instead had each board member rank the options from the most to lease desirable, and we simply counted up the numbers for each position.

We did it backwards, though, with the most desirable getting a 1, then so on, so the option with the lowest score was the winner, so: k-n+3


Register to reply

Related Discussions
Voting system- voting against? Current Events 20
Alternatives for oil General Physics 3
ID: and other alternatives General Discussion 10
Alternatives to LaTek? General Math 2