• TV
  • MOVIES
  • MUSIC
  • SHOWBIZ
  • SOAPS
  • GAMING
  • TECH
  • FORUMS
  • Follow
    • Follow
    • facebook
    • twitter
    • google+
    • instagram
    • youtube
Hearst Corporation
  • TV
  • MOVIES
  • MUSIC
  • SHOWBIZ
  • SOAPS
  • GAMING
  • TECH
  • FORUMS
Forums
  • Register
  • Login
  • Forums
  • General Discussion Forums
  • Politics
Opinion Polls Discussion Thread (Part 3)
<<
<
10 of 378
>>
>
MattXfactor
20-04-2015
Originally Posted by batdude_uk1:
“Hi there, just a couple of quick basic questions from someone who doesn't really understand all of these different polls, who are asked to create these polls as I have never been asked, and don't know anyone who has been, and also what sort of sample size is used to generate these numbers?

How reliable are they, and why are they banded about as being important? Surely the only poll that matters is the General Election one?


Thanks for any help.”

They are very important to the parties in assessing where they currently are in the "race" for downing street (or whatever you want to call it). Polls I believe just ask random members of the electorate, and they are weighted based on various things so that they should represent a group of people that represents the country as a whole as best as possible. I think they normally poll around 1000-2000 people.

In a close election like this its probably best to assume that the constant flip flopping basically means its likely we are either tied or close to tied. They are banded about as being important as they usually give a good indicator of who will win the election although in a race as close as this its very difficult for them to do so.
Chirpy_Chicken
20-04-2015
Originally Posted by jjne:
“Bit unfair -- the last poll was a rogue/outlier and ICM said so. This is pretty much a return to their normal findings.”

I was going to say a very big drop until i read this bit.
blueisthecolour
20-04-2015
Originally Posted by MattXfactor:
“I think this is more in line with reality, phone polls (like this one) generally seem to have the Con's 2 points ahead, online polls labour 2 points ahead. Also shows what most of us suspected that the ICM poll showing a massive tory lead was an outlier (and in fairness ICM pretty much told us this when they said the pool of people were a touch too tory).”

Am i right that ICM was the crazy poll from last week where Tories were winning by 6 points? The one where they were beating Labour in Scotland?

Not surprised to see this fall back in line with the 34% give or take MOE for both parties. Just goes to show that every now and again you're going to get a poll that fails the 95% test.
MARTYM8
20-04-2015
Originally Posted by MattXfactor:
“Regarding ICM I read that the poll had Lab and Con level pegging however part of their methodology is to reassign have of "don't know's" back to where they voted in 2010, which is what gave the CON's a 2% lead. I think its probably correct to do that as historical evidence seems to show that leading into elections people tend to be more likely to go back to how they voted in the previous election.”

Yes - but voters are more volatile now and it has the effect of re confirming 2010 voting patterns which boosts the LDs and the Tories. The don't knows in the end usually follow the already decideds.
PrestonAl
20-04-2015
Originally Posted by oathy:
“Latest ICM poll (17 - 19 Apr):
CON - 34% (-5)
LAB - 32% (-1)
UKIP - 11% (+4)
LDEM - 10% (+2)
GRN - 7% (-2)

sorry out of all the polls this one I have the biggest doubts about the way the numbers fly up and down.”

LibDems Soaring!
batdude_uk1
20-04-2015
Originally Posted by MattXfactor:
“They are very important to the parties in assessing where they currently are in the "race" for downing street (or whatever you want to call it). Polls I believe just ask random members of the electorate, and they are weighted based on various things so that they should represent a group of people that represents the country as a whole as best as possible. I think they normally poll around 1000-2000 people.

In a close election like this its probably best to assume that the constant flip flopping basically means its likely we are either tied or close to tied. They are banded about as being important as they usually give a good indicator of who will win the election although in a race as close as this its very difficult for them to do so.”

I thank you very much for taking your time to reply to my questions, I know they are basic ones.

So the politicians are worried because of what 2000 people out of millions say?
If that is the case, then what is the point of the polls, as they don't represent any where near enough people to be taken seriously surely??

If the polls asked about a million or so people, then they would be far more reliable, not these small numbers.

Thanks for clearing things up, to my mind they are useless unless a bigger number of people from a wider area (i.e from all over the UK) are asked.

Hopefully with modern technology making that easier and easier to achieve, in the future polls will be just like that.
blueisthecolour
20-04-2015
Originally Posted by PrestonAl:
“LibDems Soaring!”

Obviously another dud poll
blueisthecolour
20-04-2015
Originally Posted by batdude_uk1:
“I thank you very much for taking your time to reply to my questions, I know they are basic ones.

So the politicians are worried because of what 2000 people out of millions say?
If that is the case, then what is the point of the polls, as they don't represent any where near enough people to be taken seriously surely??

If the polls asked about a million or so people, then they would be far more reliable, not these small numbers.

Thanks for clearing things up, to my mind they are useless unless a bigger number of people from a wider area (i.e from all over the UK) are asked.

Hopefully with modern technology making that easier and easier to achieve, in the future polls will be just like that.”

Actually batdude you can get a surprisingly accurate result from only 1,000 people. I won't bore you with the statistics/mathematics but basically from that many people they are able to predict the actual level of support in the country to within 3% either way, 95% of the time.

To be fair to your point, I actually agree that politicians and the press focus far too much on individual polls when the actual figures will fall within a 6% window - meaning that a poll that shows the parties tied could mean that Labour are winning by 6% or losing by 6%. Both results are just as likely as them being tied (from one poll). It's only when you look at a large number that you can aggregate and start to see a real figure.
carnoch04
20-04-2015
Originally Posted by batdude_uk1:
“I thank you very much for taking your time to reply to my questions, I know they are basic ones.

So the politicians are worried because of what 2000 people out of millions say?
If that is the case, then what is the point of the polls, as they don't represent any where near enough people to be taken seriously surely??

.”

That is how the world works. TV ratings are based on boxes in a tiny number of homes but are still considered accurate by broadcasters and advertisers.
There is a science to polling and they have been pretty much on the money over the years (when you take in to account the 3% margin of error which they all have).
blueisthecolour
20-04-2015
Originally Posted by Radiomike:
“This survey from You Gov is worth a look:-

https://yougov.co.uk/news/2015/04/20...ory-defenders/

as is this piece from the Political Betting site:-

http://www2.politicalbetting.com/ind...-simply-wrong/

There is still plenty of scope for things to change in many directions.”

Sorry, I know this is my third post in quick succession, but just wanted to agree with the general points in the second article. The pollsters are only as good as their models and when the political realities shift their predictions become less accurate.

You only have to look at Israel last month to see how wrong they can get the result in a multi-party system where the main right wing party is socially unpopular. I work with Israelis and I heard a lot of stories about how young and urban people absolutely hate Netanyahu, however many naturally conservative people in more rural areas feel that he is their only option. It's not surprising that they may have keep quiet about their support to polling companies when the general atmosphere at the time was very anti-netanyahu.

The general atmosphere in many areas of UK society is arguably anti-UKIP and anti-Tory; so I wouldn't be if some of their supporters are also 'keeping quiet'.
batdude_uk1
20-04-2015
Originally Posted by blueisthecolour:
“Actually batdude you can get a surprisingly accurate result from only 1,000 people. I won't bore you with the statistics/mathematics but basically from that many people they are able to predict the actual level of support in the country to within 3% either way, 95% of the time.

To be fair to your point, I actually agree that politicians and the press focus far too much on individual polls when the actual figures will fall within a 6% window - meaning that a poll that shows the parties tied could mean that Labour are winning by 6% or losing by 6%. Both results are just as likely as them being tied (from one poll). It's only when you look at a large number that you can aggregate and start to see a real figure.”

Okay if you say so, I will take your more knowledgeable word as guidence then, thanks for updating me, I will bear all this in mind when I see some more polls.


Originally Posted by carnoch04:
“That is how the world works. TV ratings are based on boxes in a tiny number of homes but are still considered accurate by broadcasters and advertisers.
There is a science to polling and they have been pretty much on the money over the years (when you take in to account the 3% margin of error which they all have).”

It is one thing to base how successful a TV show is or not, on a few people, but what our future rulers, or at least our MP's are is something entirely different, surely?
carnoch04
20-04-2015
Originally Posted by batdude_uk1:
“It is one thing to base how successful a TV show is or not, on a few people, but what our future rulers, or at least our MP's are is something entirely different, surely?”

They are not basing anything on polls other than, possibly, how they run their campaigns. Opinion polls don't decide anything.
jjne
20-04-2015
Originally Posted by blueisthecolour:
“Sorry, I know this is my third post in quick succession, but just wanted to agree with the general points in the second article. The pollsters are only as good as their models and when the political realities shift their predictions become less accurate.”

To get a flavour of how well the polls are doing you can look at the last major national poll; that of the European elections.

Many polls were predicting 35% for UKIP; in the end they got around 3/4 of that figure.

The pollsters know Labour and Conservative voting patterns very well now. Just as with the weather, polling data is much more scientific now than it was in 1992.

The big unknown is of course UKIP. But it has already been shown to be the case that the online pollsters inflated UKIP's support; I think they're still wrong and the higher values of UKIP support are too high by around 20-25%. YouGov and ICM had it about right last year, Survation and others were too high.

So the ICM/YG polls are likely to be pretty much spot-on at the moment also, and we're looking at 11-13% for UKIP.

My personal view is that this will drop a couple of percentage points in a last day or two so 10-11% is my estimate of their final count.

Tory/Labour/LD won't be far wrong.
InMyArms
20-04-2015
Edinburgh South constituency poll (Ashcroft):
SNP - 37% (+29)
LAB - 34% (-1)
CON - 16% (-6)
LDEM - 8% (-26)
Change from 2010.

LibDems Plummeting!
oathy
20-04-2015
Latest Ashcroft poll (17 - 19 Apr):
CON - 34% (+1)
LAB - 30% (-3)
UKIP - 13% (-)
LDEM - 10% (+1)
GRN - 4% (-2)
Inspiration
20-04-2015
Don't think I've seen LAB on 30% in any other polls so far.
PrestonAl
20-04-2015
Originally Posted by oathy:
“Latest Ashcroft poll (17 - 19 Apr):
CON - 34% (+1)
LAB - 30% (-3)
UKIP - 13% (-)
LDEM - 10% (+1)
GRN - 4% (-2)”

Like a roller coaster with the LibDems! Soaring within seconds.
Chirpy_Chicken
20-04-2015
Originally Posted by Inspiration:
“Don't think I've seen LAB on 30% in any other polls so far.”

Me neither
Jason C
20-04-2015
Originally Posted by oathy:
“Latest Ashcroft poll (17 - 19 Apr):
CON - 34% (+1)
LAB - 30% (-3)
UKIP - 13% (-)
LDEM - 10% (+1)
GRN - 4% (-2)”

So Labour and Green have collectively lost five points and three of those have gone to the un-shown Others

Let's see what YouGov says later.
Boyard
20-04-2015
Originally Posted by Jason C:
“So Labour and Green have collectively lost five points and three of those have gone to the un-shown Others ”

BNP surge!
jjne
20-04-2015
Originally Posted by Jason C:
“So Labour and Green have collectively lost five points and three of those have gone to the un-shown Others

Let's see what YouGov says later.”

All MoE.

Lab on 30 looks a bit outlier-ish though.
marke09
20-04-2015
had to laugh with this gem from the Lord Aschcroft polling focus group in Plymouth


“I work for Royal Mail, so I got privatised. Now I’ve got to work for a living and it’s a killer.
Boyard
20-04-2015
Keep in mind Lord Ashcroft is a phone poll which Tories always lead in anyway usually. Plus he has his own agenda being a Tory. Yougov are the most trusted polling company.
tiger2000
20-04-2015
Leaders in the last 10 National Opinion Polls (Most Recent First)

Lab
Con
Lab
Con
Lab
Con
Tied
Con
Lab
Tied

Good luck to anyone trying to predict the outcome

http://en.wikipedia.org/wiki/Opinion...neral_election
heiker
20-04-2015
Originally Posted by Boyard:
“Keep in mind Lord Ashcroft is a phone poll which Tories always lead in anyway usually. Plus he has his own agenda being a Tory. Yougov are the most trusted polling company.”

Was a Tory, not anymore!
<<
<
10 of 378
>>
>
VIEW DESKTOP SITE TOP

JOIN US HERE

  • Facebook
  • Twitter

Hearst Corporation

Hearst Corporation

DIGITAL SPY, PART OF THE HEARST UK ENTERTAINMENT NETWORK

© 2015 Hearst Magazines UK is the trading name of the National Magazine Company Ltd, 72 Broadwick Street, London, W1F 9EP. Registered in England 112955. All rights reserved.

  • Terms & Conditions
  • Privacy Policy
  • Cookie Policy
  • Complaints
  • Site Map