Friday, February 12, 2021

There's no particular reason to think opinion polls are overestimating the Yes vote

First of all, I have an analysis piece in The National today about the new Savanta ComRes poll which puts the Yes vote on 53%.  You can read it HERE.  

Elsewhere in the same paper, Shona Craven argues that the Yes lead in the polls might be misleading, and asks whether Yes voters are significantly more likely than No voters to join volunteer online polling panels.  To which the answer is: almost certainly yes, but it probably doesn't matter that much, because that's what political weighting is there to correct for.  Most polling firms start by asking respondents how they voted in the 2014 referendum, or in some cases it's already known how they voted due to responses in previous surveys.  If there are too many Yes voters in the sample, they are downweighted accordingly.  For example, in the recent Panelbase poll for the Sunday Times, the 494 Yes voters from 2014 were downweighted to count as only 471, and the 560 No voters were upweighted to count as 583.  If that adjustment hadn't been made, the Yes lead would have been higher than the 52-48 that was reported.  And in many other surveys, the correction has been a lot bigger than that.  

If anything, the greater concern ought to be that an over-correction is going on, and that Yes are actually being underestimated as a result.  That's exactly what happened to the SNP in the 2019 election - the polls underestimated them, because people who voted for them two years earlier were being downweighted by pollsters to a target figure that was too low, due to the disproportionate number of potential SNP voters who stayed at home in 2017.  

Don't get me wrong: I think there's no room at all for complacency about the current polling situation.  But that's not because I think there's any particular reason to suppose the polls are wildly overstating Yes at the moment.  The problem is more that the Yes lead is relatively modest, and public opinion can change in the blink of an eye.

36 comments:

  1. Does Shona Craven think the 20 other opinion polls putting YES ahead are misleading?

    ReplyDelete
    Replies
    1. She was querying the polls in general, not just one in particular.

      Delete
  2. Why can't we have a more detailed Poll. One in every constituency Council area. Would that be expensive I be willing to contribute towards something like that.

    ReplyDelete
  3. I am of a view that running at around 50% plus, without a genuine attempt to work that number, is likely to head North, rather than South.

    Frankly, dealing with a pandemic in a humane and sensible manner, whilst not exactly anything other than what you'd expect. We getit here. I am not at all sure that we get the same from Westminster.

    I kind of doubt that there is likely to be a significant roll-back in the Scottish numbers for independence.

    Given the initial numbers before the previous referendum, I'd imagine that we might see a similar increase - yeah beyond stratospheric - as long as we are treated as ignortant and poor fools.

    Which appears to be Westminster's strategy.

    ReplyDelete
  4. Averaged out, Scottish polling has historically been pretty accurate. We can conclude that most Scots, tentatively and by a small margin, back indy now.

    Yes does seem to be ahead, hence the massive unionist push of 'SNP civil war', tinfoil hat conspiracy theory stuff and the appearance of new faces on the Yesblogosphere supporting 'indy for Scotland' yet doing nothing for that cause apart from relentlessly attacking yes parties and their supporters.

    ReplyDelete
  5. A big shift in Yes support was amongst older voters, in part, due to the corona response. This is clearly dissipating, but shows these voters aren't completely unreachable.

    ReplyDelete
  6. The polls are a positive indicator of the change taking place in Scotland.

    The YES vote is very, very different from the SNP vote. The Unionists only view support for Independence based on the SNP result at Holyrood. The will argue on that basis. I suspect that the constituency vote will remain fairly strong but I suspect the list vote for the SNP will fall considerably. The key issue is where that list vote goes. Will it be devided? Will it firm up behind one party.

    My point is that the road to an Independence ballot has to get past the Holyrood result.
    Anyone who thinks people are going to follow SNP 1/2 now is ignoring the current reality of politics in Scotland.

    ReplyDelete
    Replies
    1. The vast majority of the voters are not political junkies like you or me. Your perception of what is current in politic may not match the vast bulk of the electorate who have a passing grasp of politics. Polls are only a snapshot of what people are thinking at a particular time. A lot will depend on the campaigns from now up to polling day. Most people don't get their news from websites like this or twitter or Facebook. A lot depends on the "political tide" on polling day. The jury is out on what will happen. We will know on the evening on polling what the jury has decided. We just don't now at the moment.

      Delete
    2. I can't see this happening until after a Yes vote. Then it would be natural for new parties to emerge and become influential.

      Delete
    3. I must have imagined the data I seen recently regarding SNP members resigning and joining the ISP.
      It's just a few political junkies....is burying your head in the sand!
      It's a very large number. It is experienced activists and they have a lot of contacts after years campaigning for Independence.

      Delete
    4. Would also like to see this data.

      Delete
    5. Smearer Skier (liar since 2014) will soon be along to call you a Unionist as he thinks people who want to see this kind of thing must be unionists.

      That's you Keaton, unknown and andy.

      Delete
  7. The crossover occurred pre-covid.

    It's this issue - i.e. right-wing English nationalism destroying Scotland - driving things.

    https://www.euronews.com/2021/01/11/post-brexit-trade-fish-prices-collapsing-in-uk-as-red-tape-hits-exports-and-supply-chain

    Post-Brexit trade 'grim' for Scottish seafood industry as red tape hits UK exports to EU

    Post-Brexit red tape is causing some UK exports to the EU to grind to a halt, industry bodies have warned, as the new rules that came into force at New Year begin to bite.

    One fish exporter said on Monday that prices were "collapsing" in the Scottish port of Peterhead, amid reports that seafood prices fell by as much as 80% due to "export blockages".

    Scottish fish exports -- particularly reliant on smooth deliveries to the continent -- seem to be among the worst hit by new paperwork demands.

    The industry body Scotland Food & Drink, quoting information from the seafood industry, reported on Tuesday that a third of the fishing fleet was "now tied up in harbour", adding that some boats were landing in Denmark rather than Scotland to avoid "Brexit bureaucracy".

    ReplyDelete
  8. Excellent news.

    I remember almost being forced into economic migration out of British Scotland in the 1990's. Economy destroyed by English nationalism pretending to be be British. Milking all the oil cash to fund a boom in Bath. Luckily, I got a job in oil and gas; one of the few industries available to graduates at the time.

    https://archive.is/pBUkA

    Migration between England, Scotland and Northern Ireland is falling

    That could weaken the union

    I cannot believe the article bigs up Scots being forced to leave Scotland in search of a better life.

    It's fine if you want that, but there's nothing 'great' about the history of desperate mass economic emigrations out of British Scotland.

    ReplyDelete
    Replies
    1. Wouldn't call it excellent news TBH. Countries becomming more insular is never a good thing to be honest. Just as Brexit will limit the number people being able to bring their skills to Scotland to help the country go, less people moving to Scotland from other parts of the UK will mean that less people will bringing their skills to Scotland from other parts of the UK

      Generally only find that its Brexiter types that think its good that people from other countries aren't migrating to their country - they don't want smelly foreigners coming at taking all the jobs etc etc

      Delete
    2. Huh? How is a brain drain / enforced economic migration in any way a good thing?

      Are you seriously arguing that Scots having to leave Scotland for England to find work is a sign of a healthy union? That could not be more insular; people leaving en masse as is happening in the UK now due to brexit (except in Scotland).

      And it is England being insular. Academic studies show Scots are being forced out due to increasing anti-Scottish sentiment.

      While in non-insular Scotland, the number of foreign workers continues to increase, bucking the UK trend.

      Delete
    3. The article even talks about how Wales is being drained of all its young graduates then told it's too stupid for independence because it doesn't have a strong economy supported by young educated talent.

      The union is isolating Wales from the world just as it has done to Scotland, forcing insularity on these. Being in the EU helped reduce this. There is nothing more insular than not being independent.

      Delete
    4. It is no surprise that the end of direct parochial insular British rule and a change to partial independence in the form of devolution is what has transformed Scotland from somewhere people left in search of a better life to an attractive place for people to live and work. Mass emigration in the middle of an oil boom is unheard of apart from in countries pillaged by empires.

      https://www.nrscotland.gov.uk/news/2016/scotlands-changing-population

      The day after the parliament opened they started arriving / not leaving. Scotland started to join the world again.

      And I'd like to see Scots choose to go and live / work / study across Europe and beyond, e.g. through Eurasmus. Focussing on one country due to jingoistic flag waving is nationalistic and insular.

      Delete
    5. I never said it was, I said less migration into a country was bad.

      Anyhow this has very little to do with independence. The very fact that less people are now leaving Scotland (as highlighted in the article) whilst it is still part of the UK shows that independence is not needed to slow/stop a countries brain drain. Looks like the main factor is the Government of that country improving things to make people want to stay. Therefore the Welsh government probably needs to look at what the SNP has done during its time in power to reduce the brain drain in Scotland and implement those ideas.

      Delete
    6. In 1706 the population of Scotland was 25% that of England, Ireland and Wales, combined. It is now approximately 8.9% that England, Wales, and Northern Ireland, combined. That is the real, so called benefit of the UNION.

      Delete
    7. Thats simply due to a small thing called the industrial revolution. Raw materials then became the driver were people worked. People moved to to were there are jobs. So you people move to areas were the raw materials were mined. As transporting these raw materials are time-consuming and expensive factories that use these raw materials open close by to the mining areas and also transport links are improved in these areas, increasing the size of these population centres - its simple human geography. Bugger all to being in the union.

      Delete
    8. I don't remember an industrial revolution in the 1980s+. I remember an offshore one generating hundreds of billions in oil revenues which left Scotland to fund an English boom while Scots like myself were forced there to find work. 'Get on your bike and we'll maybe give you a taste of your own cash down here' they told us.

      Graduates from top unis in Demark... Norway...Sweden don't have to move to other countries to find work. If they were ruled by another country they would; it's what made Norwegians seek independence from Sweden.

      Doesn't sound like you studied geography. I did at university level in addition to geology, so I knew all about the wealth of the north sea being plundered. Also about colonialism, resource plunder and its effects on native populations. These are forced to leave their countries while their wealth is stolen.

      When I was forced to go to a foreign country (England) to attend job interviews I vowed to find work in Scotland as soon as I could and vote for independence. This is unlike some British 'Scots' who moved to the south of England (e.g. Bath) and love it there, preferring England to Scotland. Luckily I got a job in oil and gas so was not forced to leave British Scotland in search of a better life in the end. The parliament opened the same year.

      'Raw materials'. LOL. Scotland has none of them.

      Having control over your raw materials is key or they will be plundered.

      Delete
    9. Erm I was clearly referring to Williams post. North Sea oil was only discovered just over 50 years ago, so covers a tiny fraction of Scotlands history in human geographical terms. During the industrial revolution a overwhelming large proportion of the population moved to the central belt as that was the area that the population hubs grew in for the reasons above. Initally there are plenty of jobs, but then jobs outstrip the population hence causing people to leave Scotland to go to other population Hubs in search of work. Like I said its basic human geography and bugger all to do with being part of the Union.

      As you have said since devolution and under SNP rule people have stopped leaving Scotland in search of work now getting to the stage were people are bucking the trend of other UK coutnties and moving to Scotland. Therefore independence is not needed -all that has been needed is devolution and good Governance by the Devolved government to reverse the decline of the 80s.

      Delete
    10. 'Since devolution as part of the European Union' you mean.

      I assume you haven't been to Scotland and hard the accents of new Scots.

      Delete
    11. Scotland is now facing a future as a region of a parochial, insular, English nationalist UK, cut off from the world again. The mass xodus of skilled workers leaving the UK will hit Scotland too; we are just more friendly, so are holding onto our new Scots more successfully for now.

      The economic disaster of brexit already unfolding, ruining key industries. We may not have a shellfish industry soon at this rate.

      If Scots move to England in search of work, because of their lack of support for insular nationalism, they face increasing racist abuse according to academic studies. So even that option will not be open to them like it was in the 80s/90s.

      Delete
    12. 'Since devolution as part of the European Union'

      No since devolution, Scotland as part of the UK was part of the European Union (or its previous names) previous to devolution. It was devolution the key as you say above as soon as the Scottish Parliament opened people stopped moving away. Can you show these academic studies (plural) please.

      Delete
    13. Smearer Skier (liar since 2014) - just like the liars that turn up at the Sturgeon Inquiry you cannae remember your previous lies.

      Delete
  9. The March 2021 budget could have a big impact on support for YES.
    Predictions are for higher Corporation Tax, possible Annual Property Tax to replace Council Tax, higher Capital Gains and Inheritance Tax and NIC
    for self employed to made manadatory.
    All this combined with reductions in Public Sector Spending and no stimulus package to help economic recovery post Covid will have the Scottish Middle Class running for the doors. And they comprise most of the soft 20% NI vote.

    ReplyDelete
  10. Shona Craven makes a very good point in her column in today’s National. If Yes voters are significantly more likely to respond to polling than No voters then the polls must be skewed in favour of Yes as a consequence. It is a rather obvious point… once it has been made.

    But there are many other reasons why the independence movement should treat these polling results with extreme caution. For example, there is in polling the same problem as can beset a poorly framed referendum. Are people responding to the question asked? Or are they actually answering a different question altogether. By way of illustration, it has been suggested that a significant factor in the 2014 referendum was a dislike of Alex Salmond. Although there is, to my knowledge, no reliable evidence to support the notion that Salmond was particularly unpopular, the theory implies that respondents were not answering the question on the ballot paper about whether Scotland should be an independent country but were, instead responding to the question: Do you like Alex Salmond?

    I have long suspected that many of those participating in the 21 polls (so far) indicating a lead for Yes are responding to unasked questions about Brexit and handling of the public health emergency. They are not answering a question about Scotland’s constitutional status. Although even they might think they are.

    The point of this is that it makes the polls extra unreliable as indicators of firm support for independence. Even if the effect described above and that described in Shona Craven’s article ar small, they are cumulative.

    I don’t know whether polling on the Scottish constitutional issue is especially susceptible to these effects or to what extent they are effects which can be compensated for with some statistical algorithm. But not knowing this only gives me more cause to be cautious about relying on polls.

    We must always ask ourselves if what these polls are indicating is solid support for the restoration of Scotland’s independence. We must never simply assume that they do. Or is it at best the development of a statistically significant cohort of ‘soft Yes” voters? If it is not the former then should we really be celebrating them?

    My own thoughts are that support for independence hasn’t shifted to any significant degree since 2014. That the present level is fully explained by factors other than support for independence. That Yes should be polling nearer to 60% or perhaps more given the near-ideal circumstances for persuading the ‘soft Noes’. And that we should therefore be worried rather than elated.

    Allow me in closing to preempt the inevitable whining about my ‘negativity’ which totally fails to address any of the points made either in the article or in my comment. Being positive is fine. So long as you have good reason to be positive. We must constantly question our reasons for being positive just as we must constantly question all our prejudices and preconceptions. That is all I’m doing here. If we don’t persistently interrogate our attitudes and beliefs then how can we be sure they are valid even in terms of our own values far less in terms of their utility in assessing situations?

    People want to believe that the polls are reason enough to be confident of success in the fight to restore Scotland’s independence. They may not be. This may be an unfounded belief. Or a belief lack a solid foundation. What this implies is not negativity but a positive incentive to do better. If we even suspect that the current approach to the constitutional issue is not sufficiently effective in persuading people away from the Union and towards independence then it is incumbent on us to demand a rethink of that approach.

    ReplyDelete
    Replies
    1. If Yes voters are significantly more likely to respond to polling than No voters then the polls must be skewed in favour of Yes as a consequence.

      No, the polls are weighted to past votes, which ensures this isn't a problem.

      This is why averages tend to predict something really close to the result. Pollsters have decent record in Scotland within MoE.

      You would need huge number of people to consistently tell pollsters they voted a different way in past referendums / elections for current intentions to be skewed. There is some evidence here for people saying they voted Yes when in fact they voted No (regret?). This would result in Yes being somewhat underestimated.

      Delete
    2. The polls will get another test in May anyway, before iref2. My guess is those in the few weeks before the vote will again call the result pretty well.

      Delete
  11. Like comments above, I am very wary about fluctuations in independence polling. I don’t know anyone who supported independence but now does not. I do know a few people who were ‘prepared to give it a go’ and some who were voting yes because ‘well why not’ back in 2014.
    I also think that many many people would be delighted if the decision was made for them.
    But I don’t believe actual support for independence fluctuates. I do think that the younger people today tend to assume ‘independence is normal’ in a way that those of us with parents who grew up through the war and the creation of the health service etc do not; we have to make that decision and you only need to make it once; then you don’t change your mind.

    ReplyDelete
    Replies
    1. Semi independence in the form of devolution within the EU has given young people a future they didn't have when I left school / uni.

      As a graduate, the way to get a job in my day was to 'get on your bike' and move to the south of England where you could at least get a look at Scottish oil wealth.

      Devolution has partly addressed that, putting the control of many trade laws in the hands of all Europeans who, unlike England, were not out to milk Scotland for every penny. This has helped Scotland go from a backwater you left, to quite an attractive country to live and work in.

      Brexit is already undoing 24 years of progress, rolling back devolution and closing Scotland off from the world. Destroying the hopes and dreams of young Scots.

      If Scotland stays in the UK, the 1980's will return. Young Scots won't even have the chance to move to England as there they now face increasing racist abuse fueled by a PM that calls them 'vermin who should not be able to vote for 40 years' with 'Scottish taxes best spent in England'.

      Young Scots have never had the chance to vote for indy and English racists would deny them that. They will never forgive that and they are the future.

      Union's fucked.

      Delete
  12. Thanks Scottish Skier.

    Always find your posts as useful as James’s.

    I think the influx of voices into this blog seeking to argue that Scotland has done well out of the union, or arguing that Yes is doomed, indicates just how far ahead of Project Fear we now are.

    ReplyDelete