DS Forums

 
 

Opinion Polls Discussion Thread (Part 3)


Reply
Thread Tools Search this Thread
Old 24-06-2016, 05:54
Mark39London
Forum Member
 
Join Date: Feb 2007
Location: London
Posts: 3,863
So the only polls that were correct were the ones we ran here on DS
Bizarre and a first

Opinion Polls in general need changing from the top down as they clearly provide incorrect results (maybe even biased).
Mark39London is offline   Reply With Quote
Please sign in or register to remove this advertisement.
Old 24-06-2016, 06:02
CarlLewis
Forum Member
 
Join Date: Nov 2005
Posts: 3,998
52-48 is close on paper, but given that people are, on the whole, risk averse and inclined to stay with the status quo, I think this is a big rejection of the EU.
CarlLewis is offline   Reply With Quote
Old 24-06-2016, 06:03
CarlLewis
Forum Member
 
Join Date: Nov 2005
Posts: 3,998
Bizarre and a first

Opinion Polls in general need changing from the top down as they clearly provide incorrect results (maybe even biased).
But will DS get important things right like tonight's Big Brother?
CarlLewis is offline   Reply With Quote
Old 24-06-2016, 06:04
Mark39London
Forum Member
 
Join Date: Feb 2007
Location: London
Posts: 3,863
52-48 is close on paper, but given that people are, on the whole, risk averse and inclined to stay with the status quo, I think this is a big rejection of the EU.
Absolutely. The likely outcome would be to keep the status quo, so a change vote is always a big vote.
Mark39London is offline   Reply With Quote
Old 24-06-2016, 06:05
Gibman
Inactive Member
 
Join Date: Nov 2011
Posts: 591
52-48 is close on paper, but given that people are, on the whole, risk averse and inclined to stay with the status quo, I think this is a big rejection of the EU.
Very true. And factor all the money, time and non-stop Project Fear that the govt chucked at this and it's even more impressive that the GBP held their nerve at the ballot box.
Gibman is offline   Reply With Quote
Old 24-06-2016, 06:28
ArtfulDodger_
Inactive Member
 
Join Date: Jun 2016
Location: Independent United Kingdom
Posts: 3,749
Clearly people on these polls were fibbing.
ArtfulDodger_ is offline   Reply With Quote
Old 24-06-2016, 07:07
marke09
Forum Member
 
Join Date: Oct 2006
Location: mid west wales
Posts: 9,623
52-48 is close on paper, but given that people are, on the whole, risk averse and inclined to stay with the status quo, I think this is a big rejection of the EU.
1.3 million votes !
marke09 is offline Follow this poster on Twitter   Reply With Quote
Old 24-06-2016, 07:11
mossy2103
Forum Member
 
Join Date: Aug 2005
Posts: 68,712
Bizarre and a first

Opinion Polls in general need changing from the top down as they clearly provide incorrect results (maybe even biased).
Yes, that final YouGov poll quoted on BBC last night was Remain 52, Leave 48. The complete opposite of the actual result.
mossy2103 is offline   Reply With Quote
Old 24-06-2016, 07:30
gs1
Forum Member
 
Join Date: Dec 2004
Posts: 5,061
Higher margin of error this time around, but I'd be confident in a remain victory.
After all the discussion about polls, the Leave.eu one- "52/48 Leave"- that Nigel Farage apparently didn't trust at the start of the night- proved to be the most accurate, with the final percentages being 51.9/48.1
gs1 is offline   Reply With Quote
Old 24-06-2016, 11:01
dodrade
Forum Member
 
Join Date: Aug 2005
Posts: 6,121
Look whos changed his tune

Peter Kellner ‏@PeterKellner1 16m16 minutes ago

Looking much closer now; Remain victory no longer certain
Kellner's powers of prediction now rival aurichie's for their certain inaccuracy!
dodrade is offline   Reply With Quote
Old 24-06-2016, 11:02
Chisato Geeste
Forum Member
 
Join Date: Nov 2004
Posts: 19,908
Which poll got it right this time? TNS or something?
Chisato Geeste is offline   Reply With Quote
Old 24-06-2016, 11:04
Majicman
Forum Member
 
Join Date: May 2007
Posts: 22
I think when the Sunderland result comes in it will be 'game back on'
Sorry I can't help myself. That was yesterday at 2.38pm
Majicman is offline   Reply With Quote
Old 24-06-2016, 11:05
culttvfan
Forum Member
 
Join Date: Apr 2012
Posts: 1,625
Kellner's powers of prediction now rival aurichie's for their certain inaccuracy!
Kellner's useless but I don't think even he's that bad
culttvfan is offline   Reply With Quote
Old 24-06-2016, 11:07
Irritable Owl
Forum Member
 
Join Date: May 2015
Posts: 1,208
Don't know how polling companies can take the money and keep a straight face.

How many times can they get it so wrong?
Irritable Owl is offline   Reply With Quote
Old 24-06-2016, 11:11
Hacker Harrier
Forum Member
 
Join Date: Mar 2016
Posts: 811
Might as well have got Martin Kelner to carry out the polling. At least the press release would have been rammed with funny gags.
Hacker Harrier is offline   Reply With Quote
Old 24-06-2016, 11:50
inothernews
Forum Member
 
Join Date: May 2012
Posts: 3,045
Given the expectation remain will add 5 points to the final opinion polls remain will get very close to the 60-40 result I've predicted from the start.

This is why Farage has already thrown in the towel. They need to be 5-6 points up now down to have a chance.
Haven't had time to go through all the thread since I last logged on last night, but this quote from aurichie will do as well as any of his others.

Laughing my head off right now.
inothernews is online now   Reply With Quote
Old 24-06-2016, 11:57
batdude_uk1
Forum Member
 
Join Date: Apr 2005
Posts: 69,237
Surely these poll companies are now totally redundant, and will not be trusted anymore?
batdude_uk1 is offline   Reply With Quote
Old 24-06-2016, 12:11
inothernews
Forum Member
 
Join Date: May 2012
Posts: 3,045
Surely these poll companies are now totally redundant, and will not be trusted anymore?
I think it's a bit like the infamous Michael Fish weather forecast of 1987.

Computer modelling was new then, so when the computer correctly forecast the great storm human intervention downplayed what the computer model was saying as in- It can't be right. That would be the storm of the century.

I think the same thing happened here. People were telling the pollsters, but they thought they knew better, and tried factoring things in like 'When it comes to it, people will lose their nerve', only people didn't lose their nerve.

The polls that didn't try to be cute, and just gave figures, were quite often the ones that had Leave ahead (or very close) It was the so called more sophisticated polls that got it wrong (because they tried to 'interpret' findings (and thought 'When push comes to shove, people will bottle it)
inothernews is online now   Reply With Quote
Old 24-06-2016, 12:14
Rich Tea.
Forum Member
 
Join Date: Jun 2012
Location: Newport Pagnell
Posts: 21,358
After all the discussion about polls, the Leave.eu one- "52/48 Leave"- that Nigel Farage apparently didn't trust at the start of the night- proved to be the most accurate, with the final percentages being 51.9/48.1
My voting area of Milton Keynes actually voted leave with almost the exact same percentage as the national vote.
Rich Tea. is offline   Reply With Quote
Old 24-06-2016, 12:17
Nodger
Forum Member
 
Join Date: Mar 2015
Location: A bunker
Posts: 5,964
I think it's a bit like the infamous Michael Fish weather forecast of 1987.

Computer modelling was new then, so when the computer correctly forecast the great storm human intervention downplayed what the computer model was saying as in- It can't be right. That would be the storm of the century.

I think the same thing happened here. People were telling the pollsters, but they thought they knew better, and tried factoring things in like 'When it comes to it, people will lose their nerve', only people didn't lose their nerve.

The polls that didn't try to be cute, and just gave figures, were quite often the ones that had Leave ahead (or very close) It was the so called more sophisticated polls that got it wrong (because they tried to 'interpret' findings (and thought 'When push comes to shove, people will bottle it)
Indeed. Stop 'fiddling' around the edges, stop pressurising the calls to respondants, ask a simple question or two, stop weighting things on assumptions, publish the raw data (so to speak). Otherwise, f-off.
Nodger is offline   Reply With Quote
Old 24-06-2016, 12:56
batdude_uk1
Forum Member
 
Join Date: Apr 2005
Posts: 69,237
I think it's a bit like the infamous Michael Fish weather forecast of 1987.

Computer modelling was new then, so when the computer correctly forecast the great storm human intervention downplayed what the computer model was saying as in- It can't be right. That would be the storm of the century.

I think the same thing happened here. People were telling the pollsters, but they thought they knew better, and tried factoring things in like 'When it comes to it, people will lose their nerve', only people didn't lose their nerve.

The polls that didn't try to be cute, and just gave figures, were quite often the ones that had Leave ahead (or very close) It was the so called more sophisticated polls that got it wrong (because they tried to 'interpret' findings (and thought 'When push comes to shove, people will bottle it)
I never quite understood all that talk about "ohh the rain might have an effect" or that bit you touch on about people loosing their nerve.

People can and do often lie to polling companies, so what is the point in them??

I don't know why anyone should pay these companies any heed in the future, as they are practically just making up numbers.
batdude_uk1 is offline   Reply With Quote
Old 24-06-2016, 13:40
xeo
Forum Member
 
Join Date: Oct 2011
Posts: 6,373
TNS and Opinium both had Leave winning in their final polls - TNS by 2.4 points and Opinium by 1.3 points. ICM had Leave winning by 5 points, but that was a week ago so maybe factor in a slight swing to Remain since then as witnessed in the other polls and perhaps they'd have had it too if they ran a poll.

Some pollsters weighted their final polls based on an expected late swing to the status quo, but it didn't really materialise. TNS reckon it's because the status quo was so unusually unpopular.

Needless to say I think online polling has been vindicated and we'll be seeing a lot less phone polling from now on...
xeo is online now   Reply With Quote
Old 24-06-2016, 13:45
batdude_uk1
Forum Member
 
Join Date: Apr 2005
Posts: 69,237
TNS and Opinium both had Leave winning in their final polls - TNS by 2.4 points and Opinium by 1.3 points. ICM had Leave winning by 5 points, but that was a week ago so maybe factor in a slight swing to Remain since then as witnessed in the other polls and perhaps they'd have had it too if they ran a poll.

Some pollsters weighted their final polls based on an expected late swing to the status quo, but it didn't really materialise. TNS reckon it's because the status quo was so unusually unpopular.

Needless to say I think online polling has been vindicated and we'll be seeing a lot less phone polling from now on...
Why weight polls??

Especially for the status quo?? It has been proven to be completely the wrong thing to do
batdude_uk1 is offline   Reply With Quote
Old 24-06-2016, 14:04
Tassium
Forum Member
 
Join Date: Mar 2009
Location: It's Grim
Posts: 24,412
In fact haven't the pollsters done very well?

If they had been predicting 60% for Remain and the final result was 64% Remain everyone would be saying well done.

It's only because the referendum was on a knife-edge that such a small percentage variation was seen as being "very wrong".
Tassium is offline   Reply With Quote
Old 25-06-2016, 13:01
xeo
Forum Member
 
Join Date: Oct 2011
Posts: 6,373
On the next Conservative Party leader calling an early General Election:
Support: 50%
Oppose: 22%
(via TNS)
xeo is online now   Reply With Quote
 
Reply




 
Forum Jump


All times are GMT. The time now is 11:40.