Advertisement
If you have a new account but are having problems posting or verifying your account, please email us on hello@boards.ie for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact hello@boards.ie
Hi all! We have been experiencing an issue on site where threads have been missing the latest postings. The platform host Vanilla are working on this issue. A workaround that has been used by some is to navigate back from 1 to 10+ pages to re-sync the thread and this will then show the latest posts. Thanks, Mike.
Hi there,
There is an issue with role permissions that is being worked on at the moment.
If you are having trouble with access or permissions on regional forums please post here to get access: https://www.boards.ie/discussion/2058365403/you-do-not-have-permission-for-that#latest

Electoral systems discussion

  • 19-07-2024 4:57pm
    #1
    Registered Users, Registered Users 2 Posts: 7,490 ✭✭✭


    Can a new thread be set up to discuss electoral systems? This has taken over a couple of threads and has been the same comments repeatedly for close to two months now. While it is related to General Irish politics, it is a specific topic which goes far beyond the scope of this thread, particularly given the depth to which it is being discussed.

    This is a fair point so I'm starting this thread as a dedicated place for these conversations



«13

Comments

  • Registered Users, Registered Users 2 Posts: 7,490 ✭✭✭Brussels Sprout


    6 or 7 seat constituencies would be geographically very large, in rural situations.

    This is true although with so many candidates there likely wouldn't be any regions that would be without at least 1 TD



  • Moderators, Science, Health & Environment Moderators Posts: 19,906 Mod ✭✭✭✭Sam Russell


    If computer counting was used, the solution is obvious.

    If I cast my vote as follows in order of preference: A, B, C, D, E.

    Now A gets elected on count 1 with 2 quotas. B gets eliminated on count 2, E gets eliminated on count 3, C gets elected on count 5, D gets elected on count 4.

    So what happens to my vote? Well, only 50% of my vote is needed to elect A. None of my vote is used for B. C gets 50% of my vote. D and E gets none of my vote, D is already elected before my preference is available.

    Once the quota is set, the vote is shared out as required at each count, and the remaining part of the vote carries on until it dies as no further candidate remains.

    So, in a 4 seater. 20% of the vote, at least, dies without a home. Think about it, if there are 5 candidates, then all the votes for the candidate that comes 5th die.

    [Edit - I assume every candidate elected gets exactly one quota.]

    So, computer counting will be able to solve the riddle of the quota.

    100 years ago, this would not have been possible.

    Post edited by Sam Russell on


  • Registered Users, Registered Users 2 Posts: 11,475 ✭✭✭✭dulpit


    Don't they have partial transfers for seanad elections? Similar to above...



  • Moderators, Science, Health & Environment Moderators Posts: 19,906 Mod ✭✭✭✭Sam Russell




  • Registered Users, Registered Users 2 Posts: 11,475 ✭✭✭✭dulpit


    But same logic could work if there was computerised counting?

    Note - I favour the retention of current system as it is, including the count.



  • Advertisement
  • Registered Users, Registered Users 2 Posts: 3,143 ✭✭✭RoyalCelt


    How about a system that takes everyone's first preference as the starting point but instead of taking the second preference of those eliminated they take the second preference from everybody for count 2? And so on. For count 3 accumulate everybody's 3rd choice etc etc.

    What are the downsides to this?



  • Registered Users, Registered Users 2 Posts: 2,820 ✭✭✭PommieBast


    The point in only carrying those eliminated is those who made it through have already got representation.

    Did wonder whether you were thinking along the lines of Condorcet voting.



  • Registered Users, Registered Users 2 Posts: 1,654 ✭✭✭rock22


    That would surely change the quota for each count.



  • Registered Users, Registered Users 2 Posts: 11,475 ✭✭✭✭dulpit


    If your first preference nearly gets elected on first count but doesn't, that vote is now gone?



  • Moderators, Science, Health & Environment Moderators Posts: 19,906 Mod ✭✭✭✭Sam Russell


    Well, to follow the logic used in the Senate count for a count like a GE or Euro one could not be done without a computer counting system. See post #3.



  • Advertisement
  • Registered Users, Registered Users 2 Posts: 24,525 ✭✭✭✭Cookie_Monster


    The count in Ireland is a complete farce. computerised counting should 100% be a thing, instead of waiting days and days for a result it could all be done on the day or the next, like the recent UK elections as a singular example, or any number of others.

    I'm not particularly here or there on electronic voting over paper, but elec counting over manual, absolutely. Especially in pr-stv, counting and allocating everything fairly instead of the current system where:

    "The first count is normally the only time that all of a candidate’s second preference votes are counted"

    Electronic counting would easily enable 2nd and subsequent counts to fully distribute the preference correctly, which is not currently the case and it becomes entirely dependent on when your vote was counted as to whether its re-used correctly or just binned/discounted and your vote is wasted through no fault of your own.

    but mainly the completely ridiculous amount of time taken to manually count.



  • Registered Users, Registered Users 2 Posts: 2,820 ✭✭✭PommieBast


    Interesting to know why it is still done manually. Only technical reason I can think of is the inputting process but handwriting recognition has come a long way even in the last few years.

    I have my suspicions…..



  • Registered Users, Registered Users 2 Posts: 2,245 ✭✭✭Glaceon


    I'd imagine that our previous shambles with e-voting machines has left a sour taste.



  • Registered Users, Registered Users 2 Posts: 2,156 ✭✭✭Lewis_Benson


    Build new e voting infrastructure, simple answer.



  • Registered Users, Registered Users 2 Posts: 11,475 ✭✭✭✭dulpit


    I reckon the key thing is any change to the current system is fixing a problem that doesn't exist.

    Allowing the count to run for a few days isn't a problem, if anything it allows the results to be digested over time.



  • Registered Users, Registered Users 2 Posts: 29,532 ✭✭✭✭AndrewJRenko




  • Registered Users, Registered Users 2 Posts: 29,532 ✭✭✭✭AndrewJRenko


    No. It stays with that candidate, while they hope to pick up transfers from other eliminated candidates. Eventually, if they are the lowest candidate, they may get eliminated and may transfer to your next preference vote that has not previously been elected or eliminated.



  • Registered Users, Registered Users 2 Posts: 29,532 ✭✭✭✭AndrewJRenko


    No. It stays with that candidate, while they hope to pick up transfers from other eliminated candidates. Eventually, if they are the lowest candidate, they may get eliminated and may transfer to your next preference vote that has not previously been elected or eliminated



  • Moderators, Science, Health & Environment Moderators Posts: 19,906 Mod ✭✭✭✭Sam Russell


    The system we use is multi seat single transferable vote.

    What that means is that in a 4 seater, the quota is 20% and at least 20% (one quota's worth of votes) are discarded (on average) of each vote.

    If computerised counting is used, less of the votes are wasted, because the transfers are more precisely accounted for because surpluses are treated precisely, and not just last in, first out.

    Currently an attempt is made to mix the votes before counting to randomise the votes, but how is that done with ballots measured in meters long?

    However, would it make much difference to the result? Certainly, when the result is down to low figures, but mostly not, I would think.



  • Registered Users, Registered Users 2 Posts: 7,490 ✭✭✭Brussels Sprout


    No it isn't. As they learned 20 years ago the devil is in the detail.

    The very concept of E-voting is political kryptonite in this country. No party will touch that for another 20 years at least.



  • Advertisement
  • Registered Users, Registered Users 2 Posts: 2,156 ✭✭✭Lewis_Benson


    They were badly implemented.

    They could be very well implemented now, no need for all the manual counting and recounting.

    Elections decided in one day.

    Time this country's government came into the 21st century.



  • Registered Users, Registered Users 2 Posts: 15,116 ✭✭✭✭loyatemu


    I think I said at the time of the eVoting farce, if we can have a Lotto that gets written on paper and scanned by a machine, why can't we do it for voting? Then there's still a paper trail for recounts if required, but if the system is proven to be accurate to a certain low percentage, then recounts would be vary rarely requested.



  • Registered Users, Registered Users 2 Posts: 20,374 ✭✭✭✭Donald Trump


    Because then one party will probably get all the seats in a given constituency.



  • Registered Users, Registered Users 2 Posts: 11,475 ✭✭✭✭dulpit


    How can you ensure the secrecy of the ballot? How can you ensure that the system is liable to hacking or tampering?

    And again, what problems are we solving with e-voting?



  • Registered Users, Registered Users 2 Posts: 29,532 ✭✭✭✭AndrewJRenko


    What problem would 'elections decided in one day' solve?

    Because the integrity of the Lotto system doesn't depend on your vote being anonymous and secret.

    The integrity of the voting system DOES depend on your vote being secret.

    If you can prove your vote, you can sell your vote. If you can prove your vote, you can be put under duress to vote.

    There are good reasons why we vote in private booths. If you don't understand the current system, you're not in a good position to design a new system.



  • Registered Users, Registered Users 2 Posts: 29,532 ✭✭✭✭AndrewJRenko


    What do you mean by votes discarded, and where did you get the 20% figure from?



  • Registered Users, Registered Users 2 Posts: 24,525 ✭✭✭✭Cookie_Monster


    Not only does it take needlessly long, its costs a fortune to pay all those people, electronic counting would save a ton of expense.



  • Registered Users, Registered Users 2 Posts: 24,525 ✭✭✭✭Cookie_Monster


    for 3rd and subsequent counts the entire ballot is not recounted for next preference, only a portion of it. So depending on where your vote is in the count the next preference could be used or it could be discarded.



  • Moderators, Science, Health & Environment Moderators Posts: 19,906 Mod ✭✭✭✭Sam Russell


    Let me try to explain, as I did not make it clear.

    In a four seat constituency, the votes cast are divided by 5 to determine the quota. That means 20% of the votes cast do not find a winner.

    Now under a manual count, at least 20% of the vote does not rest with an elected candidate. However, with a computer count, it is possible to divide EVERY vote counted so that only that proportion required to elect a candidate is used, the rest continues as a small proportion until no more valid preferences remain for that vote, and the remaining proportion is discarded.

    So to in my example in post # 3, candidate A is elected on count 1 with 2 quotas, each vote for A is divided by 2, and all the votes for A continue as 0.5 votes to the next preference, and so on. When the next preference of a voter is for a candidate already elected or eliminated, that preference is ignored and the vote or fraction of a vote to the next preference. Eventually, all preferences will either end with no more preference expressed, or the preferences are for candidates either elected or eliminated. In the ultimate implementation, votes for already elected would be adjusted and the count run again to take account of this - however that would not be realistic except in very, very close runs.

    In that example, a vote with a single preference would only count for 50% of a vote, and the remaining fraction is discarded. It would be important to continue preference as far as the voter has an actual preference either for or against candidates.

    I suspect that most voters have little reason to choose one candidate over another once they are past 5 or 6 choices but certainly as they go further down the card.



  • Advertisement
  • Registered Users, Registered Users 2 Posts: 29,532 ✭✭✭✭AndrewJRenko


    Last time we tried that, the electronic system took more human resources to run than the paper system. Throwing tech at a 'once every few years' process may not produce savings.



  • Registered Users, Registered Users 2 Posts: 29,532 ✭✭✭✭AndrewJRenko


    A vote which "does not rest with an elected candidate" after Count 1 isn't discarded, unless it was a plumper, with no second preference.



  • Registered Users, Registered Users 2 Posts: 29,532 ✭✭✭✭AndrewJRenko


    Can you point to the polling rule for this please?



  • Registered Users, Registered Users 2 Posts: 24,525 ✭✭✭✭Cookie_Monster


    https://www.electoralcommission.ie/irelands-voting-system/#votecount

    A surplus is the number of votes a candidate has over the quota. Any candidate that reaches or exceeds the quota on the first count is deemed elected. The first count is normally the only time that all of a candidate’s second preference votes are counted. This surplus is then split between the remaining candidates. After all subsequent counts, it is only the remaining (surplus) ballot papers, after a candidate has reached the quota, that are redistributed (transferred) to the remaining candidates.

    My highlight above, so if your vote is in the quota part it takes no further part, but if it is in the surplus it will be used for the next count(s). So random chance as to whether your vote is counted earlier or later basically.



  • Registered Users, Registered Users 2 Posts: 29,532 ✭✭✭✭AndrewJRenko


    You're ignoring the distribution of votes for eliminated candidates,which is often way larger in numbers than the distribution of surpluses.

    The random selection applies to a fairly small proportion of votes.

    I'm still not seeing where you got your "3rd count" rule from.



  • Registered Users, Registered Users 2 Posts: 35,607 ✭✭✭✭Hotblack Desiato


    Instead we spent a fortune on machines that didn't work properly and another fortune to store them.

    Scrap the cap!



  • Advertisement
  • Registered Users, Registered Users 2 Posts: 69,612 ✭✭✭✭L1011


    in the Euros, plenty of people had damn good reasons to go quite close to the bottom; to actively vote against a few candidates.

    You cannot assume that everyone votes like you do.



  • Moderators, Science, Health & Environment Moderators Posts: 19,906 Mod ✭✭✭✭Sam Russell


    My assumption does not apply to 'everyone' but just most voters - particularly those voting for popular candidates.

    I am not suggesting that they do vote like me, and I understand that voting against as well as for candidates. That is the strength of our system.

    However, most voters vote for popular candidates - the clue is there for you in the name 'popular'. That is the nature of elections.

    If I vote for candidates who represents my favourite party - all of the candidates of that party - why would I go much beyond them by more than a few more candidates I might like before giving up. [I do not support any political parties , by the way.]



  • Registered Users, Registered Users 2 Posts: 2,681 ✭✭✭Field east


    Cookie , I assume that you do realise that the UK system - first past the post- is a vastly different animal to the Irish system - PR. Long May the Irish system last including the manual counting element. It is part of our culture and very much ingrained in it. If we are ever looking for the name of a span new festival then what about an ELECTION FESTIVAL. There may come a time when we will be the only country with such a unique approach to holding elections - once Malta ‘changes it’s tune’ !



  • Registered Users, Registered Users 2 Posts: 1,882 ✭✭✭pureza


    If the surplus is distributed pro rata to the percentages in the elected candidates full bundle and that candidate is subsequently elected or is elected without reaching the quota,theres still a relationship there between your vote in the main bundle and the end result,tiny and tenuous but there nonetheless I suppose

    If a voters 1st preference was eliminated early in the counts,the voter probably expected that anyway and should have been thinking about where it went next

    Our PR system is one of the best in my opinion



  • Registered Users, Registered Users 2 Posts: 2,820 ✭✭✭PommieBast


    Look like the sort of project that was doomed as soon as the consultants got their hands on it.



  • Advertisement
  • Registered Users, Registered Users 2 Posts: 2,822 ✭✭✭MacDanger


    the recent UK elections as a singular example

    That's a first past the post system I think? Comparing those counts to PR counts is complete nonsense, right?



  • Registered Users, Registered Users 2 Posts: 29,532 ✭✭✭✭AndrewJRenko




  • Moderators, Science, Health & Environment Moderators Posts: 19,906 Mod ✭✭✭✭Sam Russell


    The solution was to go for obsolete hardware and to a Dutch software company and to allow them to own that software.

    This was for an election system that exist nowhere else, and would use this hardware about once or twice every four years. Now hardware goes obsolete in that time, and starting with obsolete stuff was not a good idea.

    So wrong decisions on all counts.

    Remember, the Lotto used optical sense cards to enter its draw and pay out millions at the time. Any good consultant would have started with them.

    Also, the lack of any kind of audit trail or verification of the system post hoc suggests this was a transition year project.



  • Registered Users, Registered Users 2 Posts: 2,820 ✭✭✭PommieBast


    The Oireachtas docs I've read just state outside/independent consultants without naming them. From some digging Nathean Technologies came up but their involvement was limited to writing the report on the system's insecurity.



  • Registered Users, Registered Users 2 Posts: 29,532 ✭✭✭✭AndrewJRenko


    There were no consultants involved in the design and procurement. What documents suggest otherwise?

    Starting with a lottery system that has no requirement for anonymity and the secret ballot would have been very silly.



  • Moderators, Science, Health & Environment Moderators Posts: 19,906 Mod ✭✭✭✭Sam Russell


    @AndrewJRenko I merely said I think any design should look at the Lotto system that uses a system based on optical sense to read cards would be worth a bit of research. There is no reason to think this has any implications for anonymity. That is a separate requirement.#



  • Registered Users, Registered Users 2 Posts: 279 ✭✭EnPassant


    Malta uses the same PR-STV system as Ireland.

    For the Maltese Euro-election there were
    - 260,258 valid votes (about 2/3 as many as the Dublin constituency)
    - 6 seats
    - 39 candidates (23 in Dublin)
    - 37 counts (19 in Dublin)

    They held their Euro-elections on Saturday in conjunction with their local elections.

    They counted the votes by first scanning the ballot-papers and then counting by computer.

    Above is a picture of the scanning process - the count staff hold up the ballot paper before scanning so that the Maltese tally people can see them. The results were known by Sunday evening.

    See https://newsbook.com.mt/en/live-updates-on-european-and-local-election-outcomes/ for more details.

    What works in Malta may not necessarily be suitable for Ireland.

    There is also a specific problem with anonymity if STV elections are counted by computer. Any such count means that there is a file or database showing the details of each vote. I could easily prove to you that I voted in a particular way by giving you a sealed envelope before the election with my list of preferences. After the election, you can check the file to see that these preferences are present. There are so many possible combinations of preferences that the chances of anyone else voting in the same way are negligible.



  • Registered Users, Registered Users 2 Posts: 7,668 ✭✭✭corkie




  • Registered Users, Registered Users 2 Posts: 11,475 ✭✭✭✭dulpit


    But what is the benefit of doing counts like that? What problem are ye trying to solve?



  • Moderators, Science, Health & Environment Moderators Posts: 19,906 Mod ✭✭✭✭Sam Russell


    Well, the problems one could look to solve could be:

    1. The speed of count.
    2. The accuracy of count so that every vote counted equally without requiring random choice.
    3. The data could be checked against the paper ballot - each ballot has a unique reference number that carries over to the data set, but is not related to the voter.
    4. The data file (anonymised) could be passed to university depts of politics or other interested parties for research into voting statistics and choices. Tallying would not be needed. Of course, the data could be down to ballot box, but probably not.
    5. Comparison of results vs opinion polls. This would allow polling entities to hone their models. However, Polls should be banned from publishing results for a defined time (say a week or 10 days) prior to the election.

    Of course there could be a significant saving in cost as well as the counting only requires the ballots to be scanned to a data file and then whatever validation that is deemed necessary. Results are released as gives drama and authority to the process..



  • Advertisement
Advertisement