Opinion Polls Discussion Thread (Part 2)

1390391393395396543

Comments

  • solenoidsolenoid Posts: 15,495
    Forum Member
    ✭✭
    Haha a combined Lab/Con polling of 57% A record low perhaps?
  • MARTYM8MARTYM8 Posts: 44,710
    Forum Member
    sjp001 wrote: »
    No, its to make #Cameron look legit when not wanting to debate without the #Greens #LordAshcroft is loosing all credibility as each week passes.

    It's about as credible as the Green party's claim their membership has risen 60 per cent in one month!!
  • MattXfactorMattXfactor Posts: 3,223
    Forum Member
    ✭✭✭
    Looks like someone's been reading my threads . . . . . .*cough* thief *cough*.

    http://forums.digitalspy.co.uk/showthread.php?t=2022782&highlight=

    as soon as I read Kellners prediction it made me remember yours on here a few weeks ago haha, I think you will be very close in your vote share prediction (just to add I did agree with you at the time too not just since Kellners predicted it).
  • MattXfactorMattXfactor Posts: 3,223
    Forum Member
    ✭✭✭
    Jol44 wrote: »
    Ashcroft

    Con 29% (-5) Lab 28% (nc), Lib Dem 9% (+1), UKIP 15% (-1), Green 11%(+3)

    Ashcroft always seems to poll Lab/Con lower than other pollsters , I've not really studied how he weights his polls but it could be something to do with that. the Green surge is due to how much media coverage they'd received I'd expect it to go back below 10% within next couple of weeks.

    Although he's polled CON/LAB very low compared with other pollsters I think the margin is still pretty much in line with other polls, Con/Lab for the most part either level or either party 1 point ahead. Most poll of polls I've seen recently has lab 0.5% ahead or something like that so all polls are going to be extremely tight and a lead for either party at this point is probably just the movements between the margin of error.

    I'm more concerned by seeing the trends now as we approach the election the big questions for me are this, will the historical trend of polls moving towards the incumbent government occur this time around or will the fact that this election date has been fixed for ages mean that peoples minds are already made up.

    My own personal belief is that we'll still see a swing back to government that will just be enough for the Conservatives to have marginally more seats than Labour (I think Kellners estimate is probably around about correct) , but whether or not that will mean Cameron can hold onto power remains to be seen, a coalition may be tough to form for the Tory's & a minority government may be too unstable to survive.
  • AnnsyreAnnsyre Posts: 109,398
    Forum Member
    ✭✭✭✭
    BanglaRoad wrote: »
    That is a crazy looking poll! Never seen Con/Lab both below 30% and the Greens so high. Also with the two big boys polling so low it would be reasonable to assume that UKIP would have gained but this has not happened. No idea what to make of this

    It's the figures for the marginals.
  • Jol44Jol44 Posts: 21,048
    Forum Member
    ✭✭✭
    Annsyre wrote: »
    It's the figures for the marginals.

    It's Ashcroft's National Poll.
  • smudges dadsmudges dad Posts: 36,989
    Forum Member
    Jol44 wrote: »
    Ashcroft

    Con 29% (-5) Lab 28% (nc), Lib Dem 9% (+1), UKIP 15% (-1), Green 11%(+3)
    http://lordashcroftpolls.com/wp-content/uploads/2015/01/ANP-150119-Full-data-tables.pdf
    As always, the interesting bit is in the detailed figures. Tha Table has Labour and Conservative tied on 28%, but Ashcroft has probably done some neat rounding to get the Conservatives 1% higher.

    In Scotland it is very different:
    Con = 8%
    Lab = 24%
    LD = 4%
    UKIP =1%
    Green = 4%
    SNP = 58%
    Conservatives down to 8%! That's like the Lib Dems in the whole of the UK!

    Other snippets
    Greens hit 15% in the Midlands
    Conservatives low in Wales and SW, where UKIP and Labour both have their highest support
  • BanglaRoadBanglaRoad Posts: 57,514
    Forum Member
    http://lordashcroftpolls.com/wp-content/uploads/2015/01/ANP-150119-Full-data-tables.pdf
    As always, the interesting bit is in the detailed figures. Tha Table has Labour and Conservative tied on 28%, but Ashcroft has probably done some neat rounding to get the Conservatives 1% higher.

    In Scotland it is very different:
    Con = 8%
    Lab = 24%
    LD = 4%
    UKIP =1%
    Green = 4%
    SNP = 58%
    Other snippets
    Greens hit 15% in the Midlands
    Conservatives low in Wales and SW, where UKIP and Labour both have their highest support
    Just love those figures from Scotland! Has any party ever polled as high as 58% in a GE poll before?
  • TelevisionUserTelevisionUser Posts: 41,373
    Forum Member
    solenoid wrote: »
    Haha a combined Lab/Con polling of 57% A record low perhaps?

    The latest Ashcroft National Poll shows that support for the Green Party has grown three points to 11 per cent while Ukip’s has dipped by one to 15 per cent.
    http://i100.independent.co.uk/article/green-party-closes-gap-on-ukip-as-tories-and-labour-lose-support--gyXzUkwGoe

    Yes, it's extraordinary and figures like that would almost certainly produce a hung parliament. The two big parties ought be concerned if this turns out to be a real trend and not a one-off rogue poll like the one that caused mayhem and meltdown in the Scottish Independence Referendum last year.
  • MARTYM8MARTYM8 Posts: 44,710
    Forum Member
    The more you look at that Ashcroft poll the odder it looks. Beware of small sample sizes too when looking at sub national data.

    Even on those figures you could end up with a majority govt with only 28 per cent of the vote. UKIP and the Greens might get as much combined but get few seats under FPTP. Very unfair!
  • MattNMattN Posts: 2,534
    Forum Member
    ✭✭✭
    MARTYM8 wrote: »
    The more you look at that Ashcroft poll the odder it looks. Beware of small sample sizes too when looking at sub national data.

    Even on those figures you could end up with a majority govt with only 28 per cent of the vote. UKIP and the Greens might get as much combined but get few seats under FPTP. Very unfair!

    Ashcroft has been all over the place recently.

    Latest YouGov Poll (18 - 19 Jan):
    CON - 32% (+1)
    LAB - 32% (-)
    UKIP - 15% (-3)
    LDEM - 8% (+1)
    GRN - 7% (-)

    Same as usual with yougov
  • TelevisionUserTelevisionUser Posts: 41,373
    Forum Member
    http://lordashcroftpolls.com/wp-content/uploads/2015/01/ANP-150119-Full-data-tables.pdf
    As always, the interesting bit is in the detailed figures. Tha Table has Labour and Conservative tied on 28%, but Ashcroft has probably done some neat rounding to get the Conservatives 1% higher.

    In Scotland it is very different:
    Con = 8%
    Lab = 24%
    LD = 4%
    UKIP =1%
    Green = 4%
    SNP = 58%
    Conservatives down to 8%! That's like the Lib Dems in the whole of the UK!

    Other snippets
    Greens hit 15% in the Midlands
    Conservatives low in Wales and SW, where UKIP and Labour both have their highest support

    I don't think that's being entirely fair since it would be mathematically correct to round up 28.5% to 29% and round down 28.4% to 28% to the nearest whole percentage point. Not only that, the thing to bear in mind is that all opinion polls, including this one, will have a margin of error, e.g. three percent (technical - tl;dr).
  • RadiomikeRadiomike Posts: 7,914
    Forum Member
    http://lordashcroftpolls.com/wp-content/uploads/2015/01/ANP-150119-Full-data-tables.pdf
    As always, the interesting bit is in the detailed figures. Tha Table has Labour and Conservative tied on 28%, but Ashcroft has probably done some neat rounding to get the Conservatives 1% higher.

    In Scotland it is very different:
    Con = 8%
    Lab = 24%
    LD = 4%
    UKIP =1%
    Green = 4%
    SNP = 58%
    Conservatives down to 8%! That's like the Lib Dems in the whole of the UK!

    To properly judge those figures these are the comparisons with the General Election 2010 in Scotland

    LAB - Down 18%
    LIBDEM - Down 15%
    CON - Down 9%

    Green - Up 3%
    SNP - Up 38%
  • TelevisionUserTelevisionUser Posts: 41,373
    Forum Member
    MARTYM8 wrote: »
    The more you look at that Ashcroft poll the odder it looks. Beware of small sample sizes too when looking at sub national data.

    Even on those figures you could end up with a majority govt with only 28 per cent of the vote. UKIP and the Greens might get as much combined but get few seats under FPTP. Very unfair!

    Yes, larger sample sizes are more accurate (as would naturally be expected) and the types of polls, e.g. internet only, can induce biases and distortions all of their own which have to be taken into account. Ideally, with all these opinion polls, I'd like to see three extra criteria made very clear in addition to the results: sample size, method of polling and margin of error.
  • RadiomikeRadiomike Posts: 7,914
    Forum Member
    http://lordashcroftpolls.com/wp-content/uploads/2015/01/ANP-150119-Full-data-tables.pdf
    As always, the interesting bit is in the detailed figures. Tha Table has Labour and Conservative tied on 28%, but Ashcroft has probably done some neat rounding to get the Conservatives 1% higher.

    Based on the data the actual figures would be 28.54% CON and 28.34% LAB.

    CON rounded up to 29 as over 28.5, LAB rounded down to 28 as under 28.5.

    So standard rounding basis but effectively the figures are all but equal in real terms.
  • TelevisionUserTelevisionUser Posts: 41,373
    Forum Member
    Radiomike wrote: »
    Based on the data the actual figures would be 28.54% CON and 28.34% LAB.

    CON rounded up to 29 as over 28.5, LAB rounded down to 28 as under 28.5.

    So standard rounding basis but effectively the figures are all but equal in real terms.

    ...and no doubt both figures are subject to the poll's own margin of error too!
  • smudges dadsmudges dad Posts: 36,989
    Forum Member
    Radiomike wrote: »
    Based on the data the actual figures would be 28.54% CON and 28.34% LAB.

    CON rounded up to 29 as over 28.5, LAB rounded down to 28 as under 28.5.

    So standard rounding basis but effectively the figures are all but equal in real terms.

    One person out of 494 makes the difference. The Ashcroft poll is half the size of most pols, which could explain the variability.
  • MARTYM8MARTYM8 Posts: 44,710
    Forum Member
    One person out of 494 makes the difference. The Ashcroft poll is half the size of most pols, which could explain the variability.

    Apparently this Ashcroft poll had the lowest combined rating for the two main parties since 1981 - when the SDP Liberal Alliance was approaching 50%.

    There was apparently a Populus poll today as well which had the combined Tory/Labour share at 71% and the Greens on only 4%. (Lab 36%, Tories 35%, UKIP 13%, LD 8, Greens 4 and others 5%). On that poll UKIP had 3 times the support of 18-24 years olds than the Greens!:D

    http://www.populus.co.uk/wp-content/uploads/OmOnline_Vote_19-01-2015_BPC.pdf

    I am starting to come to the conclusion these polls are worthless - because their methodologies might have worked for a 3 party system (Lab/Com/LD) but simply don't know how to cope when you have UKIP and the Greens getting up to 25% of the vote between them - let alone the Scottish effect of the SNP. So they are all making wildly different assumptions - hence they are all over the place.

    Is UKIP on 12 or 20% - are the Greens on 4 or 11%. Because in the last week different pollsters have had exactly those ranges - and they simply can't be right.

    Maybe we will have no idea about this election until 6am on Friday 8th May - when we actually have the results!:D
  • THOMOTHOMO Posts: 7,446
    Forum Member
    MARTYM8 wrote: »
    Apparently this Ashcroft poll had the lowest combined rating for the two main parties since 1981 - when the SDP Liberal Alliance was approaching 50%.

    There was apparently a Populus poll today as well which had the combined Tory/Labour share at 71% and the Greens on only 4%. (Lab 36%, Tories 35%, UKIP 13%, LD 8, Greens 4 and others 5%). On that poll UKIP had 3 times the support of 18-24 years olds than the Greens!:D

    http://www.populus.co.uk/wp-content/uploads/OmOnline_Vote_19-01-2015_BPC.pdf

    I am starting to come to the conclusion these polls are worthless - because their methodologies might have worked for a 3 party system (Lab/Com/LD) but simply don't know how to cope when you have UKIP and the Greens getting up to 25% of the vote between them - let alone the Scottish effect of the SNP. So they are all making wildly different assumptions - hence they are all over the place.

    Is UKIP on 12 or 20% - are the Greens on 4 or 11%. Because in the last week different pollsters have had exactly those ranges - and they simply can't be right.

    Maybe we will have no idea about this election until 6am on Friday 8th May - when we actually have the results!:D

    And that's why more people than usual might well stay up to the early hours on election results night, as no one is absolutely sure which way the result will go.
    Ian.
  • jjnejjne Posts: 6,580
    Forum Member
    BanglaRoad wrote: »
    Just love those figures from Scotland! Has any party ever polled as high as 58% in a GE poll before?

    A legitimate majority for a change.
  • MattXfactorMattXfactor Posts: 3,223
    Forum Member
    ✭✭✭
    MARTYM8 wrote: »
    Apparently this Ashcroft poll had the lowest combined rating for the two main parties since 1981 - when the SDP Liberal Alliance was approaching 50%.

    There was apparently a Populus poll today as well which had the combined Tory/Labour share at 71% and the Greens on only 4%. (Lab 36%, Tories 35%, UKIP 13%, LD 8, Greens 4 and others 5%). On that poll UKIP had 3 times the support of 18-24 years olds than the Greens!:D

    http://www.populus.co.uk/wp-content/uploads/OmOnline_Vote_19-01-2015_BPC.pdf

    I am starting to come to the conclusion these polls are worthless - because their methodologies might have worked for a 3 party system (Lab/Com/LD) but simply don't know how to cope when you have UKIP and the Greens getting up to 25% of the vote between them - let alone the Scottish effect of the SNP. So they are all making wildly different assumptions - hence they are all over the place.

    Is UKIP on 12 or 20% - are the Greens on 4 or 11%. Because in the last week different pollsters have had exactly those ranges - and they simply can't be right.

    Maybe we will have no idea about this election until 6am on Friday 8th May - when we actually have the results!:D

    That is for sure, I am convinced we wont for definite know who will have most seats until the election. There's too many "wibbly wobbly" factors that could effect the outcome. The list is endless ,

    1.) How will the UKIP/Green/SNP Surge effect things?
    2.) How big is the Lib Dem collapse going to be?
    3.) are the polls underestimating the Tory's as they have done at times in the past?
    4.) is Labours FPTP advantage still going to be as big with their seat collapse in scotland?
    5.) Will people swing back to the government in the last few days as they did in 1992 and many previous elections?
    6.) even if you can answer all of these questions, presuming its a hung parliament you still wont know for sure who the PM is going to be
  • IanPIanP Posts: 3,661
    Forum Member
    ✭✭✭
    There are so many other factors too. Such as the weather across the country on the run up to and on polling day and it's effect on turn-out and mood. Any gaffs such as Gordon Brown's bigoted woman remarks in 2010. International events such as terrorist attacks. Problems with queues at polling stations with a large turn-out (have these been fixed since last time or become worse under austerity?). With such a tight and unpredictable election their are so many factors that could tip the balance between any number of outcomes.
  • solenoidsolenoid Posts: 15,495
    Forum Member
    ✭✭
    We may be seeing the death of politics in Britain.
  • MattXfactorMattXfactor Posts: 3,223
    Forum Member
    ✭✭✭
    IanP wrote: »
    There are so many other factors too. Such as the weather across the country on the run up to and on polling day and it's effect on turn-out and mood. Any gaffs such as Gordon Brown's bigoted woman remarks in 2010. International events such as terrorist attacks. Problems with queues at polling stations with a large turn-out (have these been fixed since last time or become worse under austerity?). With such a tight and unpredictable election their are so many factors that could tip the balance between any number of outcomes.

    Yeah that's true. Very unpredictable in terms of seat numbers at this stage although I do think it will be a hung parliament with Conservatives just largest party but obviously I'm no way near being certain on that.
This discussion has been closed.