Is Poll biased against Glasgow? I should probably ask an expert

SNP on course for landslide | Scotland | The Times
JEFF J MITCHELL/GETTY IMAGES

Looking at the YouGov poll for 4th – 8th March, which suggests a 2 point lead for NO and a 3 point fall in SNP support, I was browsing the M/F and age ratios to see they made sense. They do, but when I looked at the sample in terms of region, something struck me. It may or may not have been a worthwhile strike.

One of you will know.

Anyhow, here is the distribution and the Yes/No ratio:

Region                                    Electors                  Sample                   Yes          No

North East Scotland            ?                              164/164                 44           56

Highlands & Islands             206K                    109/113                 49           51

South Scotland                     208K                    167/178                 43           57

West Scotland                      ?                              112/111                 51           49

Central                                   ?                              126/127                 47           53

Mid-Scotland & Fife            ?                              132/117                 54           46

Lothians                                 650K                     183/184                 48           52

Glasgow                                 462K                     108/106                 59           41

Why is the strong Yes region of Glasgow sample (106), much smaller than the one for No majority region of Southern Scotland (Borders, Dumfries & Galloway) (178), yet it has more than twice the electorate?

In YouGov’s November poll the Glasgow sample was 136 and the South Scotland sample was 154.

More interesting, the Ipsos MORI poll in October which recorded the highest lead for Yes of 52 to 39, had a Glasgow sample higher than the South Scotland one by 156 to 125.

Also, the Savanta poll in December which returned a lead for Yes of 52 to 38 had a Glasgow sample of 139 compared to a Southern Scotland one of 83.

Ideas?

29 thoughts on “Is Poll biased against Glasgow? I should probably ask an expert

  1. More dirty tricks, they are everywhere when the yoons are in full cry.

    Have you pointed this out to James Kelly (Scot goes Pop!) to see what he thinks?

    Liked by 1 person

    1. Stewart
      This poll must be torn apart
      E.G.Where is Dundee in the mix
      What represents Cent.Scotland
      Ditto for the Islands as Orkney/ Shetland heavily NO.whilst Western Isles heavily YES
      This must be done first then passed not to polling experts but to a reputable highly renowned Statistical modelling expert
      With no axe to grind and even from overseas
      Over all there are many many holes to explore and expose on this one and certainly
      Within the abilities of the vast YES movement
      To burst this balloon

      Liked by 1 person

  2. One would have to see a boundary map to be able to decipher properly
    None the less it is more than reasonable to assume that the balls have been weighted
    Deliberately to produce a Headline making result and bullets for the NO side to load into
    Their blunderbus barrels
    Further more this has the paw marks of the Westminster beast tramped all over it
    In biology and if you were attempting to explain
    Changes in a species spread in a eco system
    Such methods of investigative sampling would only unbalance and produce a result that can but be wrong
    I smell Rat Urine and it is pungent and strong
    Proper analisation should be carried out by a highly qualified statistician
    It appears this poll set off with a mean standard of deficiency skewed to produce the result
    They sought
    But to those who are knowledgeable of such
    Matters it is blatant cooking of the books
    If I were them I would exercise great caution from now on
    Particularly so as more of their cooks enter the kitchen and all with different recipies
    To produce the cake they seek
    Then the chances of being caught out fair and square increase exponentially as their efforts increase
    If i was on the NO side i most certainly would not pull upon the trigger of the blunderbus
    As it may just backfire full onto your face
    Overall this smacks of sheer desperation
    And from that we who are Yes should actually take much comfort
    Remember all political organisations conduct
    Secret private polling and me thinks the recent results their private polling has been yielding is at the root of what is going on now
    With this poll and S on S recent one
    2 swallow,s can make a summer
    And i see flocks upon the horizon

    Like

  3. Very interesting John, just as well we have folk like you not prepared to take things at face value . As someone living in South Scotland I can see how a focus on voters here would skew any results. Sadly we have 8 more weeks of this to go but maybe better having a slight setback now to wake folk up. In 2014 it was the slight yes lead that provoked the vow and wall to wall MSM coverage

    Liked by 3 people

    1. Brobb
      Remember that in military terms if you mount a surprised attack then it only ever works if your foe remains unaware in order for it to succeed
      We were woken up by the Scot.on Sunday headline
      So the game of surprise is already defeated
      If i was overall in command of this overt operation I would be scurrying for cover
      Dreading the forthcoming counter attack

      Like

  4. THE ENGLISH NATIONALISTS ARE SCREWING ALL POLLS

    PITY THEY DONT DO HONESTY
    TBATS WHY BREXIT WAS FOISTED ON SCOTLAND
    LIES FROM SOUTHEAST ENGLAND

    Like

  5. Statistical analysis was never my strong point, I was always more a theory and law sort. Though there does appear to be a rather irrational arbitrariness to the sampling.

    On the Difference or Equality of
    Information, Misinformation, and Disinformation:
    A Critical Research Perspective
    inform.nu/Articles/Vol9/v9p083-096Stahl65(dot)pdf

    Like

  6. The Pollsters manipulating. To influence the vote. In tight margins.

    1 million in outer Glasgow and 1 million in Edinburgh. 3 million in the central belt. Smaller electorate than 3 million, Less under 16’s.

    Approx 1/2 million in NE Scotland. 250,000 City 220,000 shire. Or the other way round.

    Borders less population. Rural. Tory support.
    .

    Like

  7. OT

    I note that opinion on Wings is turning against intersectionality theory and practice. Just because the SNP haven’t a clue about how to implement just law, does not mean that acknowledging intersectionality is harmful. Intersectionality theory offers a critical perspective that originated in “women’s studies”, and is kind of crucial to overcoming misogyny, and the social oppression of other marginalised and excluded identities.

    N.B. An individual’s psychology is produced through the interaction between their biology and the social/built/natural environments. So those who wish to identify as the opposite gender due to their feelings, simply can’t escape the influence biology has on their identity.

    A Best Practices Guide to Intersectional Approaches in Psychological Research
    https://link.springer.com/article/10.1007/s11199-008-9504-5

    Like

  8. Got the electorates (probably not 100% accurate, but are a mishmash of 2019 electorates at that point):

    Electorates
    Highlands & Islands: 335,216
    North East: 548,238
    Mid & Fife: 495,506
    Lothian: 515,528
    Central: 486,263
    Glasgow: 511,994
    West: 532,097
    South: 525,784

    Total: 3,950,626

    Broadly speaking they should all be similar, apart from the Highlands & Islands, however, there will be weighting for different turnouts (e.g. Glasgow turnout being approx 75% of South). The polling companies ‘should’ weight the results based on VI though. It is interesting though. Maybe a polling company with some balls can do a 1K per region (8k total) to get a really good idea of things.

    Liked by 1 person

    1. So going back to John’s original point, even allowing for turnout differences, the south does appear to be over represented compared with Glasgow.

      Like

      1. both Glasgow weighted and unweighted samples seem low. South Scotland and lothian seem high. Same in November poll but not so pronounced.

        Like

      1. Feet on the ground is the strong point of tge Yes movement / SNP, obviously that can’t happen at this point. . . . .
        Large amounts of money funding internet campaigning is the strong point of Westminster, that goes on unhindered.

        The full disaster of Brexit will kick in after May.

        The committee that decides whether the election is to go ahead is multi party plus presiding officer. (Unionist dominated)

        Liked by 2 people

    2. High electorate. Only 1/2 million+ under 16’s? 5.6million pop.

      Turnout? More for referendums. Less for Holyrood elections, (16+) Westminster. FPTP. (18+) All different voting systems. Referendums one/two questions. Higher turnout. Westminster Election FPTP over 18s. Holyrood election D’Hond’t. 16+.Extended franchise. (migrants included) Lower turnout?

      Changes. Now extended franchise. Increased population? Referendum residential qualifications?

      2021. Not 2016.

      Liked by 1 person

  9. There are 1million under 18s in Scotland. 50,000 die each year, on average. More in the winter than the summer.

    Different electorate nos now? Extended franchise. More supportive?

    2021. Not 2016. More support for SNP/Holyrood Gov. Changing circumstances. More important to vote SNP/SNP. Vote for Independence. To improve the economy and stop people from starving and dying prematurely.

    Liked by 1 person

    1. The leadership would not listen when I was in the Party. Perhaps they will listen when I don’t vote SNP. Perhaps I will get their attention when I vote ISP on the list and spoil my constituency vote. Perhaps they will listen when others do the same or simply stay at home.

      I want Independence for Scotland, not a Wokerati Empire Business Unit.
      I want Women’s Rights recognised and at the moment I have no confidence that will happen under a SNP Government.

      Liked by 1 person

  10. It doesn’t matter anymore! We are divided

    A large proportion of the YES movement no longer trust Independence in the hands of NS, PM, etc.
    It does not matter what the loyal SNP scream at them they will no longer vote for SNP.
    One group feels let down by the SNP leadership and the other group blames the first group for not following the wheesht for Indy call.

    It doesn’t matter who is right or wrong anymore. The movement is divided and with weeks to go I doubt that the situation can be saved.

    We can throw rocks at each other all day. We can argue how big the respective camps are all day. However it cannot be disputed that we have two camps.

    We already have the Hate Bill and the SNP/Greens will probably still have a Wokerati majority (a carefully constructed list selection was designed for this) to push through GRA. One group will get something from failure.

    It will not be the same for Women’s Rights unfortunately.

    Liked by 1 person

    1. Vote for the front runner. Split the vote and lose the independence majority in the parliament. If there are really two camps we can say goodbye to independence

      Like

      1. Alex
        If there are 2 camps
        Then 1 of them is Traitorous
        So ignore them
        Let them turn and flee
        And fill a traitors grave

        Liked by 1 person

    2. To thesnpleftme I hope you don’t mind me adding this on the matters you raised – just for information, for balance, for perspective.

      Source: https://www.gov.scot/groups/misogyny-and-criminal-justice-in-scotland-working-group/

      ‘The Working Group on Misogyny and Criminal Justice in Scotland has been set up to independently consider how the Scottish criminal justice system deals with misogyny. This includes looking at whether there are gaps in the law that could be addressed by a specific criminal offence to tackle such behaviour.”

      “The group will also consider whether a statutory aggravation and/or a stirring up of hatred offence in relation to the characteristic of sex should be added to the Hate Crime and Public Order (Scotland) legislation by regulation at a future date.”

      The group is chaired by Baroness Helena Kennedy QC. She has appointed a panel of six experts with specialisms in Scots law, human rights, women’s equality and perpetrator behaviours relating to gender based violence.”

      The Remit of the Group includes but is not limited to:

      “Identify whether there are gaps in the existing law and/or where there is a failure to implement existing legislation in a way that protects women and girls. This will include whether the characteristics of sex should be included within the hate crime framework as a statutory aggravation and/or if sex should be added to stirring up hatred offences (subject to the Bill as passed).

      “Consider the legal practice in Scotland and internationally that offers the best protection for women and girls and examine how misogyny may be best tackled through a legal lens. “

      And on remit it adds: “ … work in partnership to develop a specific definition of misogyny within a Scottish legal context, taking account of behaviours that already fall within criminal law and actions that can be taken out with the criminal law to address women’s experiences relating to misogynistic behaviour or inequality, challenge men’s behaviour and wider societal attitudes.”

      For more information, membership of the working group includes but is not limited to:

      Baroness Helena Kennedy QC – Chair – Labour Party member of the House of Lords where she is active on a range of human rights advocacy matters. domestically and internationally. She is the author of the book ‘Eve was Framed: Women and British Justice’: a critique of the treatment of women in the British courts

      Emma Ritch – Executive Director at Engender, the Scotland-based feminist policy and advocacy organisation. Its aim is “to make women’s inequality visible, and persuade those with power to make positive changes to services, policy, regulation, practices, and laws that negatively affect women.”

      And finally, you wrote: “Wokerati Empire Business Unit”. Is that really the quality of discourse we’ve reached to in Scotland? Early days of a better nation? I despair!

      Liked by 2 people

      1. More on the : “Wokerati Empire Business Unit”.

        I’m a regular reader of Richard Murphy’s Tax Research blog especially for his insights into Modern Monetary Theory and into arguments in favour of an independent Scotland having its own currency.

        Professor Murphy’s blog also discusses social policy issues from time to time. He did so on 11 March following a Channel 4 News programme discussing race and claims made by Ms. Meghan Markle. He noted a representative of the right wing think tank Policy Exchange used the term ‘woke mob’.

        Source: https://www.taxresearch.org.uk/Blog/2021/03/11/i-am-not-a-member-of-a-woke-mob-i-believe-in-the-most-basic-of-human-rights/

        Murphy notes that the Cambridge Dictionary defines ‘woke’ when used as an adjective as: aware, especially of social problems such as racism and inequality. He adds: “By this definition I am most certainly woke. The government and every employer has a legal obligation to be so. Every major wisdom tradition is, of course, also woke by this definition. Quite explicitly, it’s simply not possible to claim to be Christian and not be so. Loving your neighbour as yourself, the most basic rule upon which it is based, requires it.”

        Murphy asks: “And what should happen to this supposed mob, for having the temerity to stand up for rights enshrined in the UN Declaration on Human Rights, which we signed up to long ago? What is going on here? And why is such language unchallenged, as if to imply that to support human rights places you outside society, as it would seem Policy Exchange might wish? I refuse to be described as part of a mob for defending very basic human rights. I condemn those who describe the protection of such rights as the work of a mob.”

        Murphy concludes: “I fear that if this continues it will not end well. This was the deliberate use of language to create ‘otherness’ to which blame is then attached.”

        As one of the btl contributors noted: “I’ve decided to adopt a personal motto: Never knowingly unwoke.”!

        Liked by 1 person

  11. From Panelbase:

    The first thing to note is that the South Scotland electoral region actually has a very similar population size to Glasgow, so the disparity there is not quite as extreme as you might think. The South Scotland electoral region includes places like East Lothian that you might not intuitively expect.

    I can really only speak for Panelbase but my hunch would be that any company who does what we do, weighting their results to match the 2014 referendum result and/ or 2019 Westminster result would probably not end up with wildly different results by changing the regional balance a bit, even though you’d logically think that they might. An anecdote to illustrate this: in the aftermath of the relatively poor performance of (GB-wide) polls in 2015 (across the whole polling industry), we decided to start breaking down our oldest age group, 65+, into a more detailed 65-74 and 75+ as we knew that the latter would be – on balance – more likely to vote Conservative than the former. Our theory was that this could be part of the reason why we had understated the Conservative vote share in the election – that perhaps we had too many people aged 65-74 within that broad 65+ group, and not enough of the older people.

    The impact of this was pretty much as we expected – our raw data had some more Conservatives in it. However, here’s the rub – when we then applied our standard weighting based on previous election votes, the weights had less effect because the raw sample was already more naturally in balance. The upshot was that in terms of predicted future vote, our change appeared to make no real difference. I think that regions would probably be the same – if a pollster increased the number of Glaswegian residents, the raw data would be more Yes-friendly (based on 2014 recalled vote) and as a result when weighting was applied using the 2014 result, the same overall result would probably be found.

    This is not to say that there is no merit in making the raw data as naturally representative as possible of the target population, it’s just that it probably wouldn’t have as significant an impact as one might think.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.