Political polling

[gosub]

Revealed: Newly registered voters piling up outside Labour's 'red wall'

 

[The39thStep]

Revealed: Newly registered voters piling up outside Labour's 'red wall'

Anyone else hate the term red wall? Those places are mainly municipal labour and any one who has had to endure a Labour Council would struggle to call those councils or their politics as red.

 

[treelover]

I think that is one reason it may be crumbling, anyone who has to deal with a long term labour council in the North will know how disheartening it can be, how arrogant and disinterested it can be.

 

[Smokeandsteam]

Anyone else hate the term red wall? Those places are mainly municipal labour and any one who has had to endure a Labour Council would struggle to call those councils or their politics as red.

It's not a phrase I've ever heard before this election.

It's also misleading. The cleavage between Labour and the working class has been a long run one.

 

[gosub]

14m UK voters live in areas held by same party since second world war

 

[Proper Tidy]

Yeah this red wall stuff is new to me, never heard it called that before, am probably wrong and it's a long used term but it does seem designed to be a new stick to beat labour with if it loses trad seats - the red wall has crumbled etc. Never heard of a blue wall or whatever

 

[danny la rouge]

I was polled this morning for this (I think - the question was a little different to the standard VI question)

I think I was too.

 

[killer b]

Yeah this red wall stuff is new to me, never heard it called that before, am probably wrong and it's a long used term but it does seem designed to be a new stick to beat labour with if it loses trad seats - the red wall has crumbled etc. Never heard of a blue wall or whatever

It's an import from America where Clinton's 'Blue Wall' of reliably democrat-won states collapsed in places in 2016 - presumably the expectation is for a similar collapse here.

 

[Dogsauce]

Presume it’s also a Game of Thrones reference.

 

[cupid_stunt]

ICM has the lead down 1% to just 6%, which according to Curtice is hung parliament territory, but ICM has constantly been coming up with a lower lead than the rest.

 

[littlebabyjesus]

Well it was just one company that got anywhere near last time. So if there's just one company out on a limb this time, it must be right.

Science, that is.

 

[brogdale]

ICM has the lead down 1% to just 6%, which according to Curtice is hung parliament territory, but ICM has constantly been coming up with a lower lead than the rest.

That still has the LDs 4.6% above their national share of popular vote in 2017. More work needed destroying the yellow vermin.

 

[cupid_stunt]

Well it was just one company that got anywhere near last time.

Two, and both of those also got it bang on in 2015 too.

 

[littlebabyjesus]

Two, and both of those also got it bang on in 2015 too.

Survation was one. Who was the other?

 

[MrSki]

Just heard on the news that there are 30%ish undecided up from 15% last week. If this is true then there is still all to play for.

 

[cupid_stunt]

Survation was one. Who was the other?

SurveyMonkey, IIRC they are US based, so don't tend to get commissioned by to do polls by the UK media, in both 2015 & 2017 they published just one poll, the day before the GE.

 

[brogdale]

Survation was one. Who was the other?

e2a: source

 

[littlebabyjesus]

SurveyMonkey, IIRC they are US based, so don't tend to get commissioned by the to do polls by the UK media, in both 2015 & 2017 they published just one poll, the day by the GE.

Ah ok. So the pattern is safe then.

I'm clutching at straws a bit here, I know, and we're not quite seeing the changes we were beginning to see at this stage in 2017, but then Labour came back from even further behind in 2017.

 

[cupid_stunt]

View attachment 192418

e2a: source

That doesn't include SurveyMonkey, who do use a very large sample compared to the UK based companies.

2017 - both were in the so-called margin of error +/-2%



And, about bang on in 2015

 

[littlebabyjesus]

View attachment 192418

e2a: source

Interesting that Survation were also the only ones to get the UKIP vote anywhere near right. I'm still suspicious of the weighting process they all use and the way they downgrade the responses from certain demographics. I'm confident the result will also be closer this time than the poll average, just depends how much.

 

[littlebabyjesus]

I'd clean forgotten that the polls got it wrong the other way in 2015. The hope that kills you can take many different forms.

 

[brogdale]

That doesn't include SurveyMonkey, who do use a very large sample compared to the UK based companies.

2017 - both were in the so-called margin of error +/-2%

View attachment 192420

And, about bang on in 2015

View attachment 192421

May have got this all arse about face, but isn't SurveyMonkey some sort of self-selecting internet only type of affair? Not sure of their 'methodology'.

 

[killer b]

May have got this all arse about face, but isn't SurveyMonkey some sort of self-selecting internet only type of affair? Not sure of their 'methodology'.

They're a site people (including market research companies) use to create online surveys - iirc they just added a political question to a load of surveys created by their users - probably a really good way of getting a decent sample tbh.

 

[cupid_stunt]

May have got this all arse about face, but isn't SurveyMonkey some sort of self-selecting internet only type of affair? Not sure of their 'methodology'.

No idea, TBH, but they can't be ignored considering their record.

Ironically both them & Survation are the new boys on the block, both having only started doing GE polling in 2015, and ended-up making a complete mockery of the others.

 

[Knotted]

Interesting that Survation were also the only ones to get the UKIP vote anywhere near right. I'm still suspicious of the weighting process they all use and the way they downgrade the responses from certain demographics. I'm confident the result will also be closer this time than the poll average, just depends how much.

I think I'm right in saying that only Kantar still downgrade responses from certain demographics, and they do it in such a way that barely makes any difference. The big factor that downgraded Labour's polling in 2017 has been removed this time and the factor that exaggerated Labour's polling in 2015 (ie. sampling bias) is probably still present to one degree or another.

 

[littlebabyjesus]

One thing about the '+/- error'. If these statements of possible error really were true, then nearly all the polls would fall within twice the error margins of all the others. Since they don't, the statements of possible error margin by the pollsters are demonstrably false.

 

[littlebabyjesus]

I think I'm right in saying that only Kantar still downgrade responses from certain demographics, and they do it in such a way that barely makes any difference. The big factor that downgraded Labour's polling in 2017 has been removed this time and the factor that exaggerated Labour's polling in 2015 (ie. sampling bias) is probably still present to one degree or another.

Upthread there's one poll that weights for age in a way that could lead to a very large error. Can't remember which it is, but whichever it is, its weighting could make a massive difference.

 

[Fez909]

 

[Dogsauce]

^ better than I would have thought. Any previous polling to compare with?

 

[Smangus]

icm

Britain Elects@britainelects
https://twitter.com/britainelects/status/1204053905602166785

Westminster voting intention:

CON: 42% (-)
LAB: 36% (+1)
LDEM: 12% (-1)
BREX: 3% (-)

via @ICMResearch, 06 - 09 Dec
Chgs. w/ 02 Dec

See more polls:https://britainelects.newstatesman.com/who-leads-in-our-poll-tracker/ …


Who leads in our poll tracker? - Britain Elects
The Britain Elects poll tracker is a linear moving average tracker with additional weightings so as to reduce volatility and provide the most accurate representation of public opinion on key politi...

britainelects.newstatesman.co