|
|
Post by jimjam on Sept 3, 2022 21:38:56 GMT
Peter, the key section from the Link Shevii put up is the below imo:
''Our new methodology fixes that, because it weights up the 2019 Conservative voters who do give a voting intention to account for the fact that others are undecided. This is the main reason our Labour lead is smaller than it otherwise would have been.
We believe this is the more accurate way of reflecting voting intention midterm, although no method is perfect. The main reason for this is because, when we squeeze the voting intentions of those 2019 Conservative voters who are undecided, the majority say they would still vote Conservative if they were forced.
However, it will mean a shift is required in the way our polls are analysed. For the past few months people have generally considered the high number of Conservative don’t knows to be a caveat against the high Labour poll leads. In our case, the opposite is now true. The existence of the high number of 2019 Conservative undecided voters is more likely to increase the Labour lead at a later date rather than decrease it. Our tables will still release all this information so people can see for themselves the number of undecided voters and where they are coming from.''
JJ comment, There is nothing intrinsically wrong with trying to calculate likely swingback but it does make the headlines numbers a forecast based on current opinion rather than reporting current opinion.
Graham suggests 6-8% difference but that will vary and when the Lab lead is bigger the impact is less due to more direct switchers being reflected in the uprating.
I don't have the energy to take a close look at their tables to be more accurate for this poll but saying that without the above process the reported VI would be within moe of other recent polls is fair.
Kantar do something similar by the way.
Final comment, I am not buying those Peoples Poll for GB news either.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 3, 2022 21:40:59 GMT
Can someone please tell me what Opinium polls are supposed to represent. While the 17% lead a couple of days ago might be on the high side of MOE, 4% is just ludicrous. I remember reading some time back after Opinium changed methodology that its new method reflected that their figures represented what the result is likely to be at the next election. If that is true, then WHAT A LOAD OF RUBBISH. I suspect that if things continue as most experts predict, then the result is more likely to be 25% lead than 17%. Apart from anything else, many of the over 65s will unfortunately have died due to hypothermia or starvation!!! Surely if Opinium are going to publish data like this then it should be accompanied by words of explanation and then the interested public can take it with the pinch of salt it deserves. Opinium's last Labour lead of 8% was very high for them; they have reverted to their recent norm. They are not measuring current opinion but trying to forecast the result of the next GE by applying epic levels of swing back to their figures. Seems daft to me, so I now ignore them in my own calculations. I do think it is confusing for the general public who will assume it is a normal poll. Agreed, but Just For Fun, even on those figures, Electoral Calculus makes LAB comfortably the largest party, on c305 seats. The Conservative hegemony would be over. |
|
pjw1961
Member
Government, even in its best state, is but a necessary evil; in its worst state, an intolerable one.
Posts: 8,601
|
Post by pjw1961 on Sept 3, 2022 21:50:25 GMT
Opinium's last Labour lead of 8% was very high for them; they have reverted to their recent norm. They are not measuring current opinion but trying to forecast the result of the next GE by applying epic levels of swing back to their figures. Seems daft to me, so I now ignore them in my own calculations. I do think it is confusing for the general public who will assume it is a normal poll. Agreed, but Just For Fun, even on those figures, Electoral Calculus makes LAB comfortably the largest party, on c305 seats. The Conservative hegemony would be over. Agree that. The implication of Opinium's polling for some time has been a minority Labour government. I just object in principle to it being called an opinion poll, which should be a poll of current opinion. Opinium's is an election forecast. Potentially useful, but a different animal. |
|
|
|
Post by peterbell on Sept 3, 2022 22:00:56 GMT
Shevii, thanks for the link re Opinium. They have given a detailed explanation here: www.opinium.com/resource-center/uk-voting-intention-27th-january-2022-2/The info from Opinium is stating that the new methodology takes the liklihood to vote into açcount. However, my recollection of recent polls is that where the data are published, then those voting Lab are more likely to vote than Cons. If this is correct then the old method would surely be overestimating the likely Con vote rather than underestimating it. |
|
|
|
Post by jimjam on Sept 3, 2022 22:15:40 GMT
Peter, they made a few changes at the same time but all the others were to try to get more accurate figures for current opinion.
Those may be appropriate or not but all pollsters adjust the raw data to try get a more accurate snap-shot.
The specific process I highlighted differs in that it seeks to build in forecast swingback.
As an aside, their statement that majority of Tory 2019 voters saying undecided (which includes WV and DK I think) when squeezed went Tory is true but actual published squeeze polling by R&W showed more going Lab than among those that gave a VI.
Despite requests to Chris Curtis Of Opinium) to release their squeeze data, they have not done so, hmm?
I suspect it may have been old data but may never know unless they do the right thing.
|
|
|
|
Post by mercian on Sept 3, 2022 22:49:05 GMT
ptarmigan Are you saying that these times are more desperate than the banking crisis of 2007, or the exit from the ERM in 1992(?) or the hyperinflation of the late '70s, or WWII? I'm not talking about philosophy, just practicalities. Whether tory instincts have changed or not is irrelevant. The fact is that handing out subsidies to every household is a very left-wing measure IMO. Not that I'm not grateful, but if it means my descendants will be saddled with massive government debt, maybe not such a good idea? When did we have hyperinflation in the late 1970s? Inflation - as measured by RPI - peaked at 26% - 27% inthe summer of 1975. The last full year of the Callaghan government - 1978 - saw RPI inflation of circa 8% - rising to 10.2% when Thatcher took office in May 1979. Current RPI inflation is 12.4%. Ok, mid-70s then, or doesn't 27% count as hyperinflation. I don't know what the official definition is, but happy to replace with 'huge inflation in the mid-70s' if that floats your boat. At this rate you're in the running for Quibbler of the Month. oldnat's record run of 963 consecutive wins of that particular accolade is in serious jeopardy 🤣 |
|
oldnat
Member
Extremist - Undermining the UK state and its institutions
Posts: 6,131
|
Post by oldnat on Sept 3, 2022 22:49:24 GMT
peterbell "The info from Opinium is stating that the new methodology takes the liklihood to vote into açcount"
As @jimjam says "all pollsters adjust the raw data to try get a more accurate snap-shot"
Most pollsters use "likelihood to vote" as a key weighting factor. As far as I can see they use one of 3 methods -
1. Respondents own estimate of their likelihood to vote
2. The past history of whether the respondent has previously voted regularly
3. Applying previous patterns of voting among people matching the socio-economic/age groups that the respondent is part of.
Which, if any, are the most accurate --- 
|
|
oldnat
Member
Extremist - Undermining the UK state and its institutions
Posts: 6,131
|
Post by oldnat on Sept 3, 2022 22:50:43 GMT
oldnat 's record run of 963 consecutive wins of that particular accolade is in serious jeopardy 🤣 964 |
|
oldnat
Member
Extremist - Undermining the UK state and its institutions
Posts: 6,131
|
Post by oldnat on Sept 3, 2022 22:55:22 GMT
I did like this contribution to the discussion on an indyref. 
|
|
|
|
Post by graham on Sept 3, 2022 23:33:53 GMT
When did we have hyperinflation in the late 1970s? Inflation - as measured by RPI - peaked at 26% - 27% inthe summer of 1975. The last full year of the Callaghan government - 1978 - saw RPI inflation of circa 8% - rising to 10.2% when Thatcher took office in May 1979. Current RPI inflation is 12.4%. Ok, mid-70s then, or doesn't 27% count as hyperinflation. I don't know what the official definition is, but happy to replace with 'huge inflation in the mid-70s' if that floats your boat. At this rate you're in the running for Quibbler of the Month. oldnat 's record run of 963 consecutive wins of that particular accolade is in serious jeopardy 🤣 No - I make this point simply to correct the widely-held view that Thatcher inherited rampant inflation in May 1979. She didn't - although by Spring 1980 it had more than doubled to 22%. She left office in late November 1990 with RPI inflation at 9.7% - compared with the 10.2% inherited 11.5 years earlier. Hardly evidence of a remarkable economic transformation - to say the least - given all the attendant costs incurred. |
|
oldnat
Member
Extremist - Undermining the UK state and its institutions
Posts: 6,131
|
Post by oldnat on Sept 3, 2022 23:43:15 GMT
The Tories are allegedly making up rules to rig a referendum which they are not going to "allow": Given the British nationalist view that laws must apply equally throughout the UK and the precious union, they will apply this to a border poll in Nothern Ireland. Or perhaps not. Pleased that by your own definition you have finally recognised that while the Labour and Lib Dem parties are unionist parties they are not "British nationalists". Nice to get agreement on that fairly obvious point. I have a friend who is definitely not a "nationalist" of any description. To paraphrase his views - "I don't give a f*** about nations or which state I'm a citizen of - Scotland, UK, France, Luxembourg - it doesn't matter a bugger as those in charge will always shit on me." I must seek his opinion about your somewhat nuanced view as to whether a preference for a particular geographical configuration forming a state constitutes being a "nationalist", or not. |
|
|
|
Post by mercian on Sept 3, 2022 23:51:29 GMT
lefthanging "p.p.s. I'm sad you've lost the 'left' monicker so soon, it suited you. " Nice one. It did seem to reduce the ferocity and frequency of personal attacks, but perhaps the change in name just made me a nicer person? ------------------ c-a-r-f-r-e-w " Tony Benn went to Westminster I think. " Does that make him a class traitor? -------------------- davem "Looking at the tables on this poll there are some very sobering numbers, Tories still have a big lead with the over 65’s, but that is before they feel the hit of energy prices. The under 24’s show only 9% support for the Tories. " Differential turnout dear boy. Us geriatrics will hobble to the polling station on our zimmer frames while the young are at work paying for our pensions and will be too knackered to actually vote when they get home. -------------------------------- EmCat "Imagine if MPs had to be elected by more than 50% of the electorate for that constituency. According to that very question being asked on Quora, there would be just nine such MPs." I carried out a similar exercise in the late '60s or early '70s using Whitaker's Almanac which gave detailed results for every seat in whatever was the previous GE. I seem to remember that there were only 2 or 3 then, Ebbw Vale being one. Does that mean that we are becoming more polarised?
|
|
oldnat
Member
Extremist - Undermining the UK state and its institutions
Posts: 6,131
|
Post by oldnat on Sept 3, 2022 23:59:03 GMT
EmCat
"Imagine if MPs had to be elected by more than 50% of the electorate for that constituency."
To be fair, that was quite normal pre 1832 in English pocket or rotten boroughs or Scots burghs and counties (and after 1832 for that matter). On that measure, they were all much more "democratic". It does rather depend on how much you limit (or control) whatever electorate is judged to be appropriate.
|
|
|
|
Post by EmCat on Sept 4, 2022 0:40:20 GMT
When did we have hyperinflation in the late 1970s? Inflation - as measured by RPI - peaked at 26% - 27% inthe summer of 1975. The last full year of the Callaghan government - 1978 - saw RPI inflation of circa 8% - rising to 10.2% when Thatcher took office in May 1979. Current RPI inflation is 12.4%. Ok, mid-70s then, or doesn't 27% count as hyperinflation. Wiki says no en.wikipedia.org/wiki/Hyperinflation"Economists usually follow Cagan's description that hyperinflation occurs when the monthly inflation rate exceeds 50% (this is equivalent to a yearly rate of 12874.63%)" The article gives examples of cpuntries which have experienced Hyperinflation, which does not include UK. It also gives examples of high inflation, which also does not include UK |
|
steve
Member
Posts: 12,716
|
Post by steve on Sept 4, 2022 4:41:27 GMT
@olnat
Some constituencies had no voters at all, primarily because no one actually lived there.
Remarkably the selected candidate had both 0% and conceivably 100% of the electorate and 100% or conceivably 0% of the available vote all at the same time.
|
|
neilj
Member
Posts: 6,498
|
Post by neilj on Sept 4, 2022 5:54:20 GMT
Truss's new cabinet?
|
|
|
|
Post by alec on Sept 4, 2022 6:05:27 GMT
Did Boris Johnson just call on his successor to unite the Conservative Party and govern for the national interest?
I mean, really?
|
|
|
|
Post by catfuzz on Sept 4, 2022 6:40:33 GMT
@olnat Some constituencies had no voters at all, primarily because no one actually lived there. Remarkably the selected candidate had both 0% and conceivably 100% of the electorate and 100% or conceivably 0% of the available vote all at the same time. Schrodinger’s Mandate? |
|
alurqa
Member
Freiburg im Breisgau's flag
Posts: 781
|
Post by alurqa on Sept 4, 2022 7:10:34 GMT
EmCat "Imagine if MPs had to be elected by more than 50% of the electorate for that constituency. According to that very question being asked on Quora, there would be just nine such MPs." I carried out a similar exercise in the late '60s or early '70s using Whitaker's Almanac which gave detailed results for every seat in whatever was the previous GE. I seem to remember that there were only 2 or 3 then, Ebbw Vale being one. Does that mean that we are becoming more polarised? What amazed me about the list was that six of the nine were constituencies in Liverpool. Liverpool is, therefore according to that list, the most democratically represented area in the UK. |
|
|
|
Post by steamdrivenandy on Sept 4, 2022 7:17:33 GMT
Rather than 50% of an electorate, how many achieve 50% of those who voted in the last GE?
|
|
|
|
Post by alec on Sept 4, 2022 7:40:34 GMT
hireton - thinking further on the Times story of Truss' alleged desire for a bonfire of employment rights, if true, she might have to have a quick read of her 'brilliant' trade deal with Australia. Specifically, Chapter 21, 'Labour', where she promised to "..provide for and encourage high levels of labour protection and shall strive to continue to improve such laws and policies with the goal of providing high levels of labour protection" and agreed that "it is inappropriate to encourage trade or investment by weakening or reducing the protections afforded in each Party’s labour laws". She was also remarkably woke, when she acknowledged "... the importance of gender equality and non-discrimination in employment and income opportunities for sustainable, equitable, and inclusive economic growth. Accordingly, each Party affirms its commitments to non-discrimination in employment, occupations, and places of work, and to take measures to advance anti-discrimination practices and address discriminatory practices, including those related to workplace sexual harassment, gender-based violence, gender pay gaps, and flexible working arrangements, as well as improve women’s access to decent work." So tearing up UK labour laws will likely break the terms of both her fantastic trade deal with Australia as well as all those photo-copied former EU deals. But it makes for a good headline in the Times. Link to Australian trade deal, Chapter 21, for those interested - assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1040570/uk-australia-free-trade-agreement-fta-chapter-21-labour.pdf
|
|
|
|
Post by pete on Sept 4, 2022 7:49:39 GMT
Meanwhile, the Resolution Foundation warned:
🔺 The number of people living in absolute poverty is set to rise to 14 million people in 2023-24, unless policy or economic forecasts change
🔺 Relative child poverty is projected to reach 33% in 2026-27.
|
|
|
|
Post by davwel on Sept 4, 2022 8:08:12 GMT
I am unhappy at how much prime air-time Radio 4 has given to Trump in its 8am and 9am News today, to preach poison against Biden.
To me, it is an unchecked way of helping farRight Conservatives. It`s like giving a criminal air-time to explain why he did a robbery, but victims little or no time to tell how they have suffered.
Probably the same producers who gave Lawson unopposed time to attack facts on global warming, have been at work.
|
|
steve
Member
Posts: 12,716
|
Post by steve on Sept 4, 2022 8:17:35 GMT
When you think things couldn't actually get much worse along comes our next probable health Secretary Therese Coffey promoted for blind loyalty to cosplay Maggie of course a competence vacuum.
She will move from her post at the DUP where she's forced disabled people into jobs they clearly can't do ,sometimes resulting in suicide and where she has shown a callous disregard for both staff and service users.
|
|
|
|
Post by crossbat11 on Sept 4, 2022 8:22:01 GMT
Isn't the flaw in Opinium's methodology that it fast tracks us to an event likely to still be two years away, totally ignoring any opinion shifting events that will take place in the interim? It also relies unduly on voting behaviour orthodoxy. Voters behaving in age old unchanging ways.
It's second guessing imponderables. Dangerous.
|
|
|
|
Post by crossbat11 on Sept 4, 2022 8:37:02 GMT
I've almost given up on the BBC's current affairs coverage, but I'm watching the execrable Kuennsberg's debut on the old Marr programme. First impressions are not good, although I await Joe Lycett's views on the cost of living crisis with great anticipation. We've already been informed that "Rishi" has a sweet tooth.
I must admit that I'm baffled as to why we have set piece interviews with both Tory leadership candidates, taking up half of the programme. No opposition politicians being interviewed at all. Thornberry is on some rather ridiculous panel a la Peston, but that's it. Baffling. The leadership election is over. An interview with the winner next week, maybe, but pseudo hustings giving platforms to both candidates, one of whom has already won?
Very odd, but no doubt Tim Davie and Mr Gibb will explain.
By the way Lycett is taking the piss out of the whole thing and is proving to be hilarious. The licence payers voice.
|
|
|
|
Post by jimjam on Sept 4, 2022 8:42:58 GMT
CB, Totally agree.
As I and others, including you, have said many times poll watchers such as ourselves will mentally reduce leads to take account of swingback but our tariffs will vary.
In this case, for example, at the last GE we know many Tory leaners held their noses (especially former remainers) and voted for them as the idea of a Corbyn Premiership was an anathema.
That Starmer won't push as many back to the Tory fold from not voting, or even LD, is not a novel notion and as such applying swingback levels from pre-2019 or even early in this parliament is 'brave' to say the least.
Other than contributing to a few 'Brilloesque' selective 'poll' narratives it matters little and relative experts (sadly we are on here) will look at movement by polling company.
So that Opinium 8% a fortnight ago was a significant shift and the 4% means some but not all has been clawed back which seems in line with most others with the new entrant Peoples Polling being an outlier at present.
NB) I say 'sad' not as a self critical remark but because most political journalist either don't have a clue or choose to report misreading of polls to suit their narratives; which is worse not sure?
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 4, 2022 9:00:20 GMT
Colin: "Do you think that the reality and consequences of Putin's subsequent actions in Germany and Ukraine could have been imagined then ?"No. But I was talking about the clearly evident, but ignored in UK discourse, surge in gas prices in Autumn '21. And the looming crisis in energy bills when the lagged UK system caught up with market prices. The great flood of failures amongst energy suppliers had nothing to do with events that would transpire in Ukraine 6 months later. My other point was that by focussing so much on intra-UK navel-gazing, and dismissing developments in our neighbours as irrelevant or uninteresting, we (as in media, but also the collective running of the country) risk missing a trick in spotting hazards in timely fashion. The combination of narrow, self-absorbed media focus and a government that acts largely on the basis of reacting to media pressure is a dangerous one. That was my wider point back then. Well no doubt all those criticisms of the UK are valid. They would carry more credibility from you , however ,if they were accompanied by similar observations about the myopic , self serving foreign and energy policy , over many years, of the EU's largest and most powerful member, which facilitated this economic war by Putin on Europe's Industrial base and families. You might also have observed, back in 2021, as others did, that Germany's reliance on renewables and abandonment of nuclear was another strategic error. And even if you had , I doubt whether you would have foreseen the shutting down of chunks of Germany's heavy Industry and a threat to the viability of the Mittelstand which even Robert Habeck is now voicing. .....which was the focus of my post ,to which you responded. |
|
neilj
Member
Posts: 6,498
|
Post by neilj on Sept 4, 2022 9:06:51 GMT
The biggest problem with Opinium is it uses the baseline of the 2019 election. This gave an unusual result because of two major issues.
One the unpopularity of Corbyn and two Brexit
Come the next election Corbyn as an issue is effectively finished and Brexit is unlikely to play as big an issue for Brexit voters next time.
As a result I think Opiniums decision to use the result of the 2019 election for don't knows to revert to is flawed
|
|
|
|
Post by crossbat11 on Sept 4, 2022 9:35:11 GMT
steve
Now I'm on the Road to Wembley, I'm taking the victors of the first round game yesterday (Redditch United) into the 2nd Qualifying Round. This could, if the draw tomorrow determines so, involve a game at the Metropolitan Police. Your old boys drew with Corinthian yesterday and could progress if they win the replay.
If this tie does transpire, I trust you will help me get the best tickets in the house and, if need be, some access to the good natured hospitality associated with the capital's boys in blue.
Grace and favours too. Exemption from a body search perhaps, and being issued with some anti-tasering PPE; useful in the event of things not going well for the Met on the field of play. I hear they have a large hooligan firm.
|
|
