From colin at netech.ie Thu Feb 11 09:15:44 2010 From: colin at netech.ie (Colin Whittaker) Date: Thu, 11 Feb 2010 09:15:44 +0000 Subject: [E-voting] need help finding a reference Message-ID: <20100211091544.GA5678@infiltrator.stdlib.net> I ended up in a debate yesterday with some colleagues about the impact of plumping or bullet voting - this is voting for only a single candidate in a multi seat election. They were contending that this is bad and has an undue impact on the result and voters should be required to express as many preferences as there are seats or the vote will count as spoiled. I "know" this is wrong but I need to find a nice reference that I can point to that explains this. Anyone got any suggestions, google/wikipedia not helping so far. This is not a completely abstract debate as we intend to run a prstv election next month to fill a bunch of committee positions and it turns out this is the first time we have had more candidates than there were seats. Colin -- Colin Whittaker +353 (0)86 8211 965 http://colin.netech.ie colin at netech.ie From Michael.McMahon at Sun.COM Thu Feb 11 10:17:57 2010 From: Michael.McMahon at Sun.COM (Michael McMahon) Date: Thu, 11 Feb 2010 10:17:57 +0000 Subject: [E-voting] need help finding a reference In-Reply-To: <20100211091544.GA5678@infiltrator.stdlib.net> References: <20100211091544.GA5678@infiltrator.stdlib.net> Message-ID: <4B73D955.7080607@sun.com> Colin Whittaker wrote: > I ended up in a debate yesterday with some colleagues about the impact > of plumping or bullet voting - this is voting for only a single > candidate in a multi seat election. > > They were contending that this is bad and has an undue impact on the > result and voters should be required to express as many preferences as > there are seats or the vote will count as spoiled. > > I "know" this is wrong but I need to find a nice reference that I can > point to that explains this. Anyone got any suggestions, > google/wikipedia not helping so far. > > This is not a completely abstract debate as we intend to run a prstv > election next month to fill a bunch of committee positions and it turns > out this is the first time we have had more candidates than there were > seats. > > Colin > Generally, it's not in a voter's best interest to only vote for one candidate because if their candidate is eliminated, then they have no further influence on the outcome. If their candidate is elected, then while they have helped to elect them, they also have no further influence on the outcome of further seats. But, I think it might be a bit strong to say that voters should be required to express all preferences, or as many preferences as seats. First, because someone might genuinely have only a first preference and really doesn't care about the others. In that situation, it's reasonable in my opinion, to only express one preference. Second, it's hard to know exactly how many preferences are actually counted. It depends on the contents of the votes themselves, and obviously varies from one ballot paper to another. But, from my limited observations, it would be typically less than the number of seats. - Michael From fergal at esatclear.ie Thu Feb 11 10:51:42 2010 From: fergal at esatclear.ie (Fergal Daly) Date: Thu, 11 Feb 2010 10:51:42 +0000 Subject: [E-voting] need help finding a reference In-Reply-To: <4B73D955.7080607@sun.com> References: <20100211091544.GA5678@infiltrator.stdlib.net> <4B73D955.7080607@sun.com> Message-ID: <875029961002110251k44bdb645l86ff633bd3d96347@mail.gmail.com> On 11 February 2010 10:17, Michael McMahon wrote: > Colin Whittaker wrote: >> I ended up in a debate yesterday with some colleagues about the impact >> of plumping or bullet voting - this is voting for only a single >> candidate in a multi seat election. >> >> They were contending that this is bad and has an undue impact on the >> result and voters should be required to express as many preferences as >> there are seats or the vote will count as spoiled. >> >> I "know" this is wrong but I need to find a nice reference that I can >> point to that explains this. Anyone got any suggestions, >> google/wikipedia not helping so far. >> >> This is not a completely abstract debate as we intend to run a prstv >> election next month to fill a bunch of committee positions and it turns >> out this is the first time we have had more candidates than there were >> seats. >> >> Colin >> > Generally, it's not in a voter's best interest to only vote for one > candidate > because if their candidate is eliminated, then they have no further > influence on the outcome. If their candidate is elected, then while > they have helped to elect them, they also have no further influence > on the outcome of further seats. > > But, I think it might be a bit strong to say that voters should be required > to express all preferences, or as many preferences as seats. > > First, because someone might genuinely have only a first preference > and really doesn't care about the others. In that situation, it's reasonable > in my opinion, to only express one preference. Indeed. Colin, ask your colleagues if it's OK to abstain or spoil a vote. It seems to me that if they follow their principal every should express N preferences (N= number of candidates). Saying that 0 preferences is OK but 1, 2, ..., N -1 are not makes no sense. I can put that on a web page for you if you like :) F > Second, it's hard to know exactly how many preferences are actually counted. > It depends on the contents of the votes themselves, and obviously varies > from > one ballot paper to another. But, from my limited observations, it would > be typically > less than the number of seats. > > - Michael > > _______________________________________________ > E-voting mailing list > E-voting at lists.stdlib.net > http://lists.stdlib.net/mailman/listinfo/e-voting > http://evoting.cs.may.ie/ > From cansbro at eircom.net Thu Feb 11 14:29:50 2010 From: cansbro at eircom.net (Catherine Ansbro) Date: Thu, 11 Feb 2010 14:29:50 +0000 Subject: [E-voting] need help finding a reference In-Reply-To: <875029961002110251k44bdb645l86ff633bd3d96347@mail.gmail.com> References: <20100211091544.GA5678@infiltrator.stdlib.net> <4B73D955.7080607@sun.com> <875029961002110251k44bdb645l86ff633bd3d96347@mail.gmail.com> Message-ID: <4B74145E.4080904@eircom.net> What Michael said. 100% agreement. If you'd like data showing how different systems affect outcomes (including the impact of specific voting tactics), see Peter Emerson's slim books discussing various voting methodologies. One of his books come with a CD with sample data and spreadsheets so you can count the votes using various systems. Peter's based in N. Ireland and he's done a lot of work in really strife-torn and polarized locations, demonstrating how various kinds of ranked-preference systems can help to reveal consensus. I've learned a lot from him. You can also contact him directly and ask him your question. He might have data or insight to share, or point you to a source of further info. Cheers, Catherine