|
|
Post by manchesterman on Jun 20, 2024 19:35:08 GMT
I think if you offered the Tories that outcome [only 14% behind Labour & 11% ahead of Reform] on 4 July, they'd snatch your hand off!
|
|
Sibboleth
Labour
'Sit on my finger, sing in my ear, O littleblood.'
Posts: 16,057
|
Post by Sibboleth on Jun 20, 2024 19:43:48 GMT
This is the Brendan Cox outfit no? One of the co-founders, but he hasn't been involved in the running of it for about six years. Luke Tryl is, or at least was (who knows with anyone now), a Conservative. They did pretty well with the Mayoral polls the other month (apart from the North East) though, as we know, that is a different ball game. |
|
Sibboleth
Labour
'Sit on my finger, sing in my ear, O littleblood.'
Posts: 16,057
|
Post by Sibboleth on Jun 20, 2024 19:44:43 GMT
I think if you offered the Tories that outcome [only 14% behind Labour & 11% ahead of Reform] on 4 July, they'd snatch your hand off! Yet both would be objectively disastrous. A statement in itself. |
|
|
|
Post by hullenedge on Jun 24, 2024 16:51:46 GMT
|
|
|
|
Post by carlton43 on Jun 24, 2024 16:57:53 GMT
So, we have Polling showing Reform 10-points behind the Conservative at the same time others place then 1-point ahead!! These are charlatans and witch doctors of no consequence to anyone. |
|
|
|
Post by greenrichard on Jun 24, 2024 19:24:56 GMT
So, we have Polling showing Reform 10-points behind the Conservative at the same time others place then 1-point ahead!! These are charlatans and witch doctors of no consequence to anyone. Or, and hear me out on this one, perhaps they just ask different people. |
|
|
|
Post by carlton43 on Jun 24, 2024 21:54:05 GMT
So, we have Polling showing Reform 10-points behind the Conservative at the same time others place then 1-point ahead!! These are charlatans and witch doctors of no consequence to anyone. Or, and hear me out on this one, perhaps they just ask different people. And what would be the point of that? I ask 10 young students in Sheffield Central and then 10 OAPs on the greens at Dore Golf Club in Hallam and get two completely different answers. Well, I would wouldn't I? |
|
|
|
Post by nobodyimportant on Jun 24, 2024 21:58:56 GMT
This pollster definitely has a strong Conservative skew compared to basically any other. I doubt that's down to who they ask unless there's some dodgy sampling strategies going around.
|
|
|
|
Post by batman on Jun 24, 2024 22:44:17 GMT
Perhaps it's the methodology rather than the nature of the sample. It seems amazing to think this so close to a general election, but I do think they are slightly overestimating the Conservative vote, low as it still is in these poll results. Perhaps others here have a perspective as to why they come up with relatively good results for the Tories?
|
|
Sibboleth
Labour
'Sit on my finger, sing in my ear, O littleblood.'
Posts: 16,057
|
Post by Sibboleth on Jun 24, 2024 22:51:21 GMT
They are one of those firms who do an extra re-weighting. The unadjusted figures are: Lab 41, Con 23, Ref 16, LDem 9, Green 6.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jun 24, 2024 23:09:52 GMT
Not far off 40% Labour, 20% Reform, 20% Conservative, 10% LD, 10% other which would be a bigger landslide than Blair '97 with the Corbyn 2017 vote share! |
|
|
|
Post by greenrichard on Jun 25, 2024 8:52:39 GMT
Or, and hear me out on this one, perhaps they just ask different people. And what would be the point of that? I ask 10 young students in Sheffield Central and then 10 OAPs on the greens at Dore Golf Club in Hallam and get two completely different answers. Well, I would wouldn't I? I think you missed my point entirely. If you and I both ask two perfectly chosen cross section of the population who they vote for, we’d likely get different percentages. That doesn’t make us snake oil salespeople. It just means we’d accurately report the result and they might be different. The fact some people pay a lot of attention to polls and expect them to accurate predict how millions of people will actually vote is a problem for people who read the polls, not for those who compile them. |
|
|
|
Post by stodge on Jun 25, 2024 9:21:48 GMT
This pollster definitely has a strong Conservative skew compared to basically any other. I doubt that's down to who they ask unless there's some dodgy sampling strategies going around. Looking at the data, there's a key difference between MiC and Redfield & Wilton (R&W) on the voting intention of older people. A sub-sample admittedly (and to be treated with caution) but MiC has the Conservatives at 40% among those over 65 while R&W's "mega poll" (so a much larger sample) has 25%. With MiC, that 40% represents nearly half the total Conservative vote - the other key point is the likelihood to vote. Older people are more likely to vote (75-80% certainty) so that core Conservative vote is more likely to turn out. Context - in December 2019, the Conservatives won over 65s by 47 points (64-17), the lead now is 15 (40-25 with MiC) so that's a healthy 16% swing away from the Conservatives so even if more older people are still voting Conservative it's not in the numbers it was and will be the worst result for the Conservatives in that age group since 2010. Back then, Cameron enjoyed majorities among the 30, 40 and 50 somethings as well but it's very different now. |
|
|
|
Post by carlton43 on Jun 25, 2024 9:24:13 GMT
And what would be the point of that? I ask 10 young students in Sheffield Central and then 10 OAPs on the greens at Dore Golf Club in Hallam and get two completely different answers. Well, I would wouldn't I? I think you missed my point entirely. If you and I both ask two perfectly chosen cross section of the population who they vote for, we’d likely get different percentages. That doesn’t make us snake oil salespeople. It just means we’d accurately report the result and they might be different. The fact some people pay a lot of attention to polls and expect them to accurate predict how millions of people will actually vote is a problem for people who read the polls, not for those who compile them. And you are missing my point. We both know a lot about polling structure and method. We know it is a snapshot and not predictive. We understand modelling. If two companies have two 'perfectly chosen cross section' samples then I would expect broadly similar results with a trend that is very similar indeed. Where we have an 11-point difference (more than half the value of the higher result) then one or both are grossly wrong and I suspect the elusive presence of snake oil! |
|
Raddy
Non-Aligned
Posts: 70 
|
Post by Raddy on Jun 25, 2024 12:02:39 GMT
Surely the data analysis will always be much of a muchness, it is the make up of the panel that determines the difference.
Only anecdotal, but a few years ago during an election campaign, I asked a group of 8 friends if they were on polling panels or would they carry out a street interview if asked, they were all 50 to 64.
Three from across the political spectrum who were politically engaged were enthusiastic to take part when asked, but never volunteered, another two had little interest in politics but might take part if it suited them on any given occasion. The other three would rather have their finger nails pulled out.
Has any pollster ever included a question to gauge how politically engaged people are, because the days of ascribing peoples voting preference to a socio economic group with historical voting traditions can surely no longer be considered even remotely reliable.
|
|
The Bishop
Labour
Down With Factionalism!
Posts: 39,067
|
Post by The Bishop on Jun 25, 2024 12:12:10 GMT
We will see how "reliable" it is in this instance soon enough, I suppose.
There are many problems with VI polling, but being able to measure it against something concrete and indisputable has always been one of its strengths.
|
|
|
|
Post by John Chanin on Jun 25, 2024 13:17:36 GMT
Surely the data analysis will always be much of a muchness, it is the make up of the panel that determines the difference. Only anecdotal, but a few years ago during an election campaign, I asked a group of 8 friends if they were on polling panels or would they carry out a street interview if asked, they were all 50 to 64. Three from across the political spectrum who were politically engaged were enthusiastic to take part when asked, but never volunteered, another two had little interest in politics but might take part if it suited them on any given occasion. The other three would rather have their finger nails pulled out. Has any pollster ever included a question to gauge how politically engaged people are, because the days of ascribing peoples voting preference to a socio economic group with historical voting traditions can surely no longer be considered even remotely reliable. YouGov are well aware of the need to get the non engaged on their panel and have discussed this at length. The standard method, used I think by all pollsters now, is to pay people for participating. |
|
Raddy
Non-Aligned
Posts: 70
|
Post by Raddy on Jun 25, 2024 14:17:14 GMT
Surely the data analysis will always be much of a muchness, it is the make up of the panel that determines the difference. Only anecdotal, but a few years ago during an election campaign, I asked a group of 8 friends if they were on polling panels or would they carry out a street interview if asked, they were all 50 to 64. Three from across the political spectrum who were politically engaged were enthusiastic to take part when asked, but never volunteered, another two had little interest in politics but might take part if it suited them on any given occasion. The other three would rather have their finger nails pulled out. Has any pollster ever included a question to gauge how politically engaged people are, because the days of ascribing peoples voting preference to a socio economic group with historical voting traditions can surely no longer be considered even remotely reliable. YouGov are well aware of the need to get the non engaged on their panel and have discussed this at length. The standard method, used I think by all pollsters now, is to pay people for participating. Do they actually pay them cash individually? I seem to remember market research generally used to give you the opportunity to go in a draw for something or other. |
|
|
|
Post by John Chanin on Jun 25, 2024 14:22:24 GMT
YouGov are well aware of the need to get the non engaged on their panel and have discussed this at length. The standard method, used I think by all pollsters now, is to pay people for participating. Do they actually pay them cash individually? I seem to remember market research generally used to give you the opportunity to go in a draw for something or other. Yep. I'm on YouGov's panel and get £50 periodically. |
|
|
|
Post by hullenedge on Jun 26, 2024 16:19:27 GMT
|
|
