|
|
Post by distingflyer on Feb 22, 2020 1:39:23 GMT
Over the last few years, have built up a datafile of my own encompassing elections from 1885 to the present: drive.google.com/open?id=1dYxOhd1afsae8DNN-tSN77Uuk6ZUKi0OA couple of items to note: Firstly, the italicized constituencies (1885 to 1970) indicate constituencies that roughly fall within modern-day Greater London; Secondly, the Irish figures from 1885 to 1918 are taken from Walker's Parliamentary Election Results in Ireland, 1801-1922, which lack tables totalling the vote by party at each election; therefore, there may be some errors in my figures as I wasn't able to cross-check my totals with anything definite (though I've gone over them several times, I can't promise there are no mistakes!). Thirdly, there are some hidden columns indicating swing figures for elections from 1959 to the present (doing it pre-1945 didn't make as much sense due to parties not always contesting a seat two elections in a row - Labour, then Liberal, then both, then just one again, and so forth, for instance). Fourthly, the vote percentages from 1885 to 1945 have been adjusted for dual-member constituencies rather than simply being taken from the raw totals (this is why the 1945 percentages, for instance, are 48.0%-39.6% as opposed to 47.7%-39.7%). (Additionally, my thanks to Georg Ebner, who alerted me to this site from uselectionatlas. Looking forward to some stimulating discussions . . . ) Just noticed a very minor error. The Tory majority in SE Staffs in 1987 was 10,885 not 11,241 since the Alliance officially overtook Labour that year (although a lot of us think there was a counting error on the night). Well spotted! Have fixed it, and uploaded the corrected file. The 1918-2017 spreadsheet from the Parliament website, which has been referenced above, also contained a 1918 turnout estimate for Cork - not sure on what it's based, as it produces a voting figure of exactly 30800, but have used it in my own spreadsheet anyway. Unfortunately it did not provide an indicator of where the extra thousand votes in my Irish total seem to be coming from (again, I think it's probably a Meath typo, since both seats are listed as having exactly the same number of voters). (Should also note that, in checking turnout figures, I was able to spot & fix a rather embarrassing transposition error for the 1955 results, where the some of the Liberal votes in Essex were in the wrong constituencies.) |
|
|
|
Post by Davıd Boothroyd on Feb 22, 2020 1:42:39 GMT
How are you dealing with the Central Wandsworth problem? Not quite sure to what you're referring - the figures for that constituency (from 1918 through 1945) seemed fine. Not in 1923 they aren't. |
|
|
|
Post by distingflyer on Feb 22, 2020 1:58:19 GMT
|
|
|
|
Post by distingflyer on Feb 22, 2020 2:02:39 GMT
Not quite sure to what you're referring - the figures for that constituency (from 1918 through 1945) seemed fine. Not in 1923 they aren't. I have the Tories on 8774, Labour on 5294 & Liberals on 4357, producing a total of 18425 votes (and a 62.0% turnout on an electorate of 29733) and a Tory majority of 3480. This matches the figures in the Craig book, so I'm not sure what's wrong. |
|
|
|
Post by distingflyer on Feb 22, 2020 2:32:46 GMT
I don't know if any regional electorate figures have been assembled anywhere else, but here's what I've got for the 1987, 1992 & 1997 elections:
1987
North - 10951043
Midlands - 6935949
London - 5111379
South - 12989409
1992
North - 10939753
Midlands - 7030516
London - 4890697
South - 13210011
1997
North - 10905143
Midlands - 7120515
London - 4910637
South - 13580208
|
|
|
|
Post by Davıd Boothroyd on Feb 22, 2020 9:39:14 GMT
I have the Tories on 8774, Labour on 5294 & Liberals on 4357, producing a total of 18425 votes (and a 62.0% turnout on an electorate of 29733) and a Tory majority of 3480. This matches the figures in the Craig book, so I'm not sure what's wrong. When the Returning Officer was conducting the count, he did it in the usual way of separating the votes and putting them in bundles of 100, then putting the 100s together to make 1,000, and finally adding up the number of bundles together with the counted incomplete bundle to make the final total. Unfortunately when he did this for Sir John Norton-Griffiths's votes, the RO misread his own writing and a bundle of 1,000 was misread as 100. Accordingly the figure of 8,774 for Norton-Griffiths is in error and he actually had 9,624 votes. The error was not picked up at the count; shortly after the Labour agent detected that the total number of votes in the result as declared was 900 down on the number of ballot papers counted in the verification stage, and wrote to the Lord Chancellor's Office to tell them and ask if it could be investigated. (As the majority was well over the error, there was no question of an election petition) It's all explained in the files of the National Archives LCO 2/2584. So the question is whether you use the erroneous number as declared, or the correct number as was determined afterwards. |
|
|
|
Post by distingflyer on Feb 22, 2020 13:46:55 GMT
I have the Tories on 8774, Labour on 5294 & Liberals on 4357, producing a total of 18425 votes (and a 62.0% turnout on an electorate of 29733) and a Tory majority of 3480. This matches the figures in the Craig book, so I'm not sure what's wrong. When the Returning Officer was conducting the count, he did it in the usual way of separating the votes and putting them in bundles of 100, then putting the 100s together to make 1,000, and finally adding up the number of bundles together with the counted incomplete bundle to make the final total. Unfortunately when he did this for Sir John Norton-Griffiths's votes, the RO misread his own writing and a bundle of 1,000 was misread as 100. Accordingly the figure of 8,774 for Norton-Griffiths is in error and he actually had 9,624 votes. The error was not picked up at the count; shortly after the Labour agent detected that the total number of votes in the result as declared was 900 down on the number of ballot papers counted in the verification stage, and wrote to the Lord Chancellor's Office to tell them and ask if it could be investigated. (As the majority was well over the error, there was no question of an election petition) It's all explained in the files of the National Archives LCO 2/2584. So the question is whether you use the erroneous number as declared, or the correct number as was determined afterwards. Had not come across this before; many thanks. Have amended the Tory & turnout figures. |
|
|
|
Post by Andrew_S on Feb 22, 2020 14:06:52 GMT
We had a similar problem a few weeks ago in Hull West where the returning officer declared an incorrect number of votes for the Green Party, 50 instead of 560.
|
|
|
|
Post by Davıd Boothroyd on Feb 22, 2020 15:06:18 GMT
It's pretty much acknowledged that the 1 vote recorded for Catherine Taylor-Dawson in Cardiff North in 2005 was an error. Possibly the real figure was 51 or 101 but the bundle was missed.
Another known error which hasn't been explained is in the 1997 Scottish referendum in Fife. On the first question there were 166,025 ballot papers recorded (125,668 agree, 39,517 do not agree, 840 spoilt). On the second the number of ballot papers was nearly 2,000 higher at 167,999 (108,021 agree, 58,987 do not agree, 991 spoilt).
|
|
|
|
Post by 🏴☠️ Neath West 🏴☠️ on Feb 22, 2020 15:11:40 GMT
It's pretty much acknowledged that the 1 vote recorded for Catherine Taylor-Dawson in Cardiff North in 2005 was an error. Possibly the real figure was 51 or 101 but the bundle was missed. She should have threatened to sing to them unless they gave her a recount! |
|
|
|
Post by Davıd Boothroyd on Feb 22, 2020 15:22:08 GMT
Incidentally the total 1918 electorate in Ireland I have is Boroughs 386,667, Counties 1,539,607, Universities 10,399, making 1,936,673. This is the total in the Parliamentary Return of 18 November 1918 (HC 138 1918). It is just a coincidence that the Northern and Southern divisions of Meath have the same number of voters in them.
|
|
|
|
Post by distingflyer on Feb 22, 2020 19:10:58 GMT
Those numbers fit with the Craig/Rallings/Thrasher electorate figures; the Walker book must just have a difference of a thousand somewhere. Since that book seems to have some disagreements on vote totals (as well as party totals) from the former books, this perhaps shouldn't be too surprising.
|
|
|
|
Post by Peter Wilkinson on Feb 28, 2020 19:53:23 GMT
|
|
|
|
Post by distingflyer on Dec 11, 2021 21:31:03 GMT
The earlier spreadsheet was a somewhat 'polished-up' version of the one in this folder (https://drive.google.com/drive/folders/1jRJEDGxmGXYJGRyVyLzMlFUPDuqMyYFl) which has more statistical breakdown pages & so forth.
Have also included similar files for Scotland, Canada & all the Canadian provinces.
|
|
Georg Ebner
Non-Aligned
Roman romantic reactionary Catholic
Posts: 9,796 
|
Post by Georg Ebner on Dec 11, 2021 22:44:08 GMT
The earlier spreadsheet was a somewhat 'polished-up' version of the one in this folder (https://drive.google.com/drive/folders/1jRJEDGxmGXYJGRyVyLzMlFUPDuqMyYFl) which has more statistical breakdown pages & so forth. Have also included similar files for Scotland, Canada & all the Canadian provinces. Great! Fantastic! |
|
|
|
Post by LDCaerdydd on Dec 11, 2021 22:54:23 GMT
It's pretty much acknowledged that the 1 vote recorded for Catherine Taylor-Dawson in Cardiff North in 2005 was an error. Possibly the real figure was 51 or 101 but the bundle was missed. Another known error which hasn't been explained is in the 1997 Scottish referendum in Fife. On the first question there were 166,025 ballot papers recorded (125,668 agree, 39,517 do not agree, 840 spoilt). On the second the number of ballot papers was nearly 2,000 higher at 167,999 (108,021 agree, 58,987 do not agree, 991 spoilt). The figure of 1 was definitely orally announced on the night. If there were 51 or 101 etc someone at the count must surely have seen 2 or more votes or at some point in the count have seen the bundle of 50/100 and queried it. I heard from a Lib Dem in Cardiff who was active in 2005 that they think there were 0 votes for her and the RO ‘moved one ballot across’ - equally comical, not sure if it’s plausibility though. In terms of incorrectly recorded/annouced results, I’ve noticed looking at the 2011 regional figures for the National Assembly that UKIP got just 73 votes in Cynon Valley - when 730 (for example would be much more realistic) en.wikipedia.org/wiki/Cynon_Valley_(Senedd_constituency)?wprov=sfti1 |
|
|
|
Post by WestCountryRadical on Dec 11, 2021 23:33:57 GMT
The figure of 1 was definitely orally announced on the night. If there were 51 or 101 etc someone at the count must surely have seen 2 or more votes or at some point in the count have seen the bundle of 50/100 and queried it. I heard from a Lib Dem in Cardiff who was active in 2005 that they think there were 0 votes for her and the RO ‘moved one ballot across’ - equally comical, not sure if it’s plausibility though. From The Sunday Times of 8th May 2005: Sad news for amateur singer Catherine Taylor-Dawson, who last week gloried in the dubious achievement of getting one single vote in Cardiff North. Now it seems that another 200 votes for her Vote For Yourself party were mixed into the Liberal Democrat pile by mistake. |
|
