Tuesday, March 28, 2006

Poll Results

As of 18:00PM Tuesday March 28

Link http://www.cvk.gov.ua/vnd2006/w6p001e.html
Source Offical results as of 6:00PM 84.75% Counted
Date March 28, 2006     Poll Date: March 26, 2006 
Name of the party or electoral bloc
PR Party of Regions (PR)
BYuT Yuliya Tymoshenko Bloc (BYuT)
NS-NU Bloc 'Our Ukraine' (NS-NU)
SPU Socialist Party of Ukraine (SPU)
KPU Communist Party of Ukraine (KPU)
LPB Lytvyn's People's Bloc (LPB)
NVB Nataliya Vitrenko Bloc 'People's Opposition' (NVB)
UPBKP Ukrainian People's Bloc of Kostenko and Plyusch
VICHE Viche Party
CBP Civil Bloc PORA-PRP' (CBP)
OBNY Oppositional bloc NOT YES! (OBNY)
Party Block Poll% Vote% Seats% No of Seats
PR 30,64% 30,64% 39,14% 176
BYuT 22,43% 22,43% 27,51% 124
NS-NU 15,01% 15,01% 17,01% 77
SPU 6,00% 6,00% 4,25% 19
KPU 3,60% 3,60% 0,85% 4
LPB 2,65% 2,65%  
NVB 2,63% 2,63%  
UPBKP 2,07% 2,07%  
VICHE 1,59% 1,59%  
CBP 1,49% 1,49%  
OBNY 0,95% 0,95%  
Others 10,94% 10,94%  
Sum 100,00% 100,00% 100,00% 450


Kevin L. said...

Hello. Interesting site (and I mean that in a good way). One thing though; I think you may be using an incorrect formula for the allocation of seats in the Ukrainian parliament, or are relying on a source that is using an incorrect formula. Some guy had earlier edited Wikipedia's page on this election ( http://en.wikipedia.org/wiki/Ukrainian_parliamentary_election%2C_2006 ) to say that the requisite 3% required to make the quota would docked from each party before the proportional allocation of seats (by whatever method used, which I've been told in the post http://uselectionatlas.org/FORUM/index.php?topic=38060.msg854627#msg854627>this post in the Atlas Forum of the awesome (shameless plug) Dave Leip's Atlas of U.S. Presidential Elections is the largest remainder method with what I can tell is the Hare Quota). Your seat tallies from the election results at various stages in the count and from the two exit polls of this election that are cited in your blog seem to assume such a method of seat allocation. But that same post I referred from Dave Leip's Atlas Forum also said that the person who made that Wikipedia edit was wrong and translated what he implied were the relevant sections of Ukrainian election law. These sections seem to make it clear that seats in the Ukrainian parliament are allocated to each party meeting the 3% threshold in proportion to that parties percentage of the total vote for all parties meeting the 3% threshold. If you could let me know where you got your information about either the formula or the seat tally data in your exit poll and election tally charts I would appreciate it. I actually see you have cited something called Oblast and when I have more time I will do a Google search on that. But I thought I'd let you know that your seat data might be incorrect.

Kevin L. said...

I had left a long message earlier but I think I lost it all. Anyway, I think your using an incorrect seat allocation earlier, although someone had edited the Wikipedia page on this election describing a formula that was a lot like the one you seem to be using. Someone on a web forum later told me that guy was incorrect and translated relevant sections of Ukrainian election law that led me to believe him. Basically, I think those parties meeting the 3% threshold are allocated seats proportionally through the largest remainder method with the Hare Quota. The Communists have been in the neighborhood of 20-21 seats rather than merely 4.

Online Editor said...

Kevin. Thanks for this information.

I have been trying to ascertain how they calculate the number of seats.

The question that was not answered is do they discount the 3% threshold then allocate proportionally or does the 3% count towards the seat allocation.

I went off the formula on Wikipedi. I will review and publish both outcomes as I am still not sure. Overall I am satisfioed with teh elecrtion process but have little praise for the Altas site design and management.

Thanks again

Online Editor said...

Kevin can you send me your contact deatils. They wont be published.

Kevin Lamoreau said...

Sure, that contact's forum member summary is at http://uselectionatlas.org/FORUM/index.php?action=profile;u=1456 . If you become a forum member you will be able to contact him through the forum's free private messaging service (which is meant for forum issues, but the Ukrainian Parliamentary Election is being discussed on the forum, in fact that's where I first learned that such an election was being held there. I think the text ag (my "contact") translated would have mentioned somewhere in the sections ag included if 3% of the total vote was deducted from each party or bloc meeting the quota before the descibed method of seat allocation was utilised. The electoral quota is also described as the "sum total of the votes given for the parties (blocs) that obtained at least 3% of the vote" divided by the number of seats (450) in the Verkhovna Rada (Ukraine's parliamant), and it would be weird for the entire vote for those parties and blocs receiving 35 of the total vote or more to be used in determining the quota (the "ideal" number of votes per seat) if only part of the vote for each of those parties and blocs were used in determining how many whole quotas, and what fraction of an additional quota, that party or bloc had. With a largest remainder method, unlike with a highest average method like the d'Hondt or Sainte-Laguë methods, the ideal number of votes per seat or "quota" you set is critical because the number of seats each party or bloc (not counting those parties and blocs that fail to meet the threshold) receives will always be within 1 of the determined real (and usually non-integer) number of quotas. If the entire vote for those parties and blocs meeting the 3% threshold is what is devided by the total number of seats in parliament to determine the quota, and less than the entire vote of each party or block meeting the threshold is used to determine how many quotas each party or bloc has, than it might not be possibly to allot every seat in Parliament to each party or bloc.

Also, even before I "reversed" (see http://en.wikipedia.org/w/index.php?title=Ukrainian_parliamentary_election%2C_2006&diff=45753558&oldid=45749028 ) the February 18 edit http://en.wikipedia.org/w/index.php?title=Ukrainian_parliamentary_election%2C_2006&diff=40143150&oldid=39870153 that added the mention of a 3% exclusion, so during the time that mention was in the article, first a seats collumn that seemed to assume a proportional representation of the total vote among those parties meeting the 3% threshold was added to an existing results table in http://en.wikipedia.org/w/index.php?title=Ukrainian_parliamentary_election%2C_2006&diff=45749028&oldid=45748238 and then that table was replaced by the template table http://en.wikipedia.org/wiki/Template:Ukrainian_parliamentary_election%2C_2006 that has consistently had the number of seats for the five parties or blocs above the 3% threshold that seem to be proportional to those parties' relative percentages of the vote rather than their relative percentages of the vote over or beyond 3%. I suppose you know that anyone can edit Wikipedia, so information there is not necessarily correct. But the history of editions to Wikipedia's page on this year's Ukrainian Parliamentary elections are further evidence in my mind that the person who made that February 18 edit was wrong.

I've never heard of a method such as that one and the one you had assumed until yesterday (understandably so, because Wikipedia is fairly well-monitored overall and incorrect information on popular pages is not likely to last long) having been used. France gave, last I checked, a quarter it's seats in regional parliaments or whatever they're called to the highest polling slate in the second round of voting (or in the first in the unlikely event that one slate going into that first round, generally before the two major parties on at least the left and often the non-Le Pen right combine slates, gets a majority; you may have heard of France's two-round runoff system in Presidential elections, well their regional elections use a two-round runoff/PR hybrid) which has a slightly similar effect on the allocation of seats as the "old Wikipedia method," to give it a name. I can see some logic to such a method, as there isn't the big sudden jump or drop-off in the number of seats won by a party when the party goes from just being on one side of the threshold to being just on the other side, but results in an even less proportional allocation of seats than simple PR with quota, which you mentioned the dis-proportionality of in your "Results and Analysis of Seat Allocation" post of this section of your blog.

I hope this information is helpful to you. Thanks for your prompt mention of the possible error I pointed out on your blog and for pointing out where you got your info from. That made me more confident that ag the Atlas Forum guy was right and the Wikipedia guy was wrong; if you had gotten your info about the seat allocation method from another source it might have been a different story.