Proportional representation

[Emsworth]

I disapprove of proportional representation, but if it must be used, STV would, in my opinion, be best. Specifically, the Hare-Clark method should be used

I particularly dislike party lists because the individuals elected are not truly accountable to the voters, but only to their parties.

[A18]

Whichever one sucks the most, so that people will get tired of it and switch back to a sane system later.

Okay, seriously, what's STV?

[John Dibble]

Yeah, I didn't know there were different types, really.

[Emsworth]

Sorry for not posting an explanation.

STV: Each voter ranks all candidates in order of preference. In order to be elected, a candidate must receive a required quota (usually = total votes/ total seats). If any candidate receives the required quota, then the votes in excess of the quota are redistributed. If no candidate receives the required quota, then the candidate with the lowest number of first preferences is eliminated, and his votes are redistributed. This goes on until the required number of seats are filled.

Essentially, STV is preferential voting for a multi-member constituency; it does, however, result in proportional representation.

Closed list: First, each party prepares a ranked list of candidates. At the election, each voter votes for a party. The parties are allocated seats in proportion to the number of votes received. The seats are filled based on the ranking in the list. For example, if the Democrats win 10 seats, the first 10 people shown on their list are elected.

Open list: This is just like the party list, except that the party does not get to rank the list. Instead, the voter must rank all candidates on the appropriate list.

[I spent the winter writing songs about getting better]

Tossup between Closed and Open list, depending on the situation. Closed list is better if there are loads of candidates, open is better for smaller elections.

I like the D'hondt method. I prefer proportional representation since it benefits extremist parties.

[A18]

Dictatorship is far preferable to proportional representation or any form of IRV.

[Jake]

I agree, but STV is the only sane choice between these. It forces candidates to run as candidates and allows voters to have a choice.

[Hatman 🍁]

STV and PR should never be mixed, it just makes things more confusing. Closed list is just undemocratic.

BRTD, I don't know what d'hont is (pardon the pun) but I assume there is no threshold? I agree, thresholds in PR are undemocratic.

[Emsworth]

BRTD, I don't know what d'hont is (pardon the pun) but I assume there is no threshold?

For each party, a quotient is calculated: Votes/(seats + 1), where "seats" is the number of seats that the party has received so far (originally 0, of course). The party with the highest quotient receives the next seat, and then the quotients are recalculated.

It can be used with or without a threshold.

[I spent the winter writing songs about getting better]

Yep. it's used in Israel, where I think they have a threshold of 1%.

[Hatman 🍁]

I like basic PR where it's not complicated, and without threasholds (except of course the minimum % you need to get one seat). If there are extra seats left over, than give them to the parties who came closes to getting an extra seat. Not very difficult.

[Fmr. Gov. NickG]

I like the system used by the Netherlands, where the parties rank the candidates to use as a default, but where voters can vote for individual candidates from anywhere on the list.  If a candidate gets enough individual votes, they can be elected over someone further up on the default rankings.

STV is a good system for non-partisan elections, where there are not too many candidates.  But if you have a ton of different parties, each of which have their own list of candidates, this system can be very confusing for voters.

[I spent the winter writing songs about getting better]

I like basic PR where it's not complicated, and without threasholds (except of course the minimum % you need to get one seat). If there are extra seats left over, than give them to the parties who came closes to getting an extra seat. Not very difficult.

so like the system that was used in Iraq?

[minionofmidas]

Yeah, known as Hare-Niemeyer.

Me, I'm a big fan of open lists, especially STV (really just a complicated open list). Closed lists are almost 1% as bad as fptp.

[Hatman 🍁]

I like basic PR where it's not complicated, and without threasholds (except of course the minimum % you need to get one seat). If there are extra seats left over, than give them to the parties who came closes to getting an extra seat. Not very difficult.

so like the system that was used in Iraq?

Something like that, but I would use a number like 200 or 400 for simplicity sake. The more simple and democratic a system is, the better. Plus, Iraq had a threshold as well, but it was very small.  In addition, a country should be divided into regions of equal number of seats so the ballots are smaller.  Also, Iraq has a system of every third candidate on the list should be female, when it should be every second.

[Platypus]

We use a version of STV in our federal senate, and whilst there are some blips (eg Steven Fielding getting less then 2% of the vote last year but still being elected because of preference deals-more info if you want it later-and reaching the ~14% quota) it generally works quite well.

[Emsworth]

We use a version of STV in our federal senate, and whilst there are some blips (eg Steven Fielding getting less then 2% of the vote last year but still being elected because of preference deals-more info if you want it later-and reaching the ~14% quota) it generally works quite well.

My only qualm about the Hare-Clark system is that the counting of votes involves some random behavior. After a candidate's quota has been reached, the determination of which ballots are surplus ballots and which ballots count toward the quota is made randomly.

[Fmr. Gov. NickG]

We use a version of STV in our federal senate, and whilst there are some blips (eg Steven Fielding getting less then 2% of the vote last year but still being elected because of preference deals-more info if you want it later-and reaching the ~14% quota) it generally works quite well.

My only qualm about the Hare-Clark system is that the counting of votes involves some random behavior. After a candidate's quota has been reached, the determination of which ballots are surplus ballots and which ballots count toward the quota is made randomly.

I thought that a fraction of each ballot would be counted as surplus, but maybe this is a different system than "Hare-Clark".  So if someone needs 100 votes to get elected and gets 150, each person who voted for them first counts 2/3 of their vote for that candidate and 1/3 of their vote for their second choice.

[Emsworth]

We use a version of STV in our federal senate, and whilst there are some blips (eg Steven Fielding getting less then 2% of the vote last year but still being elected because of preference deals-more info if you want it later-and reaching the ~14% quota) it generally works quite well.

My only qualm about the Hare-Clark system is that the counting of votes involves some random behavior. After a candidate's quota has been reached, the determination of which ballots are surplus ballots and which ballots count toward the quota is made randomly.

I thought that a fraction of each ballot would be counted as surplus, but maybe this is a different system than "Hare-Clark".  So if someone needs 100 votes to get elected and gets 150, each person who voted for them first counts 2/3 of their vote for that candidate and 1/3 of their vote for their second choice.

For the first redistribution (that is, the transfer of votes to second preferences), yes, Hare-Clark is not random. However, beyond the second preference, it is.