A new poll carried out by Survation for the Liberal Democrats has shown that Jo Swinson is in a tight race with the SNP in her East Dunbartonshire constituency.
The Kirkintilloch Herald reports:
The poll, with methodology designed by the LibDems but carried out by an independent company, shows a mere two per cent divide between Ms Swinson from SNP candidate John Nicholson.
It also predicts that Labour will take just 16 per cent of the vote and the Conservatives just 13 per cent.
The results were: Lib Dem 34.5%, SNP 32.1%, Lab 16.2%, Con 13.1%, Green 2%, Ukip 0.7%A total of 413 people took part in the telephone poll with seven questions, carried out on April 9.
We can now bring you the full analysis including all the tables which you can access here.
This fieldwork was carried out in the wake of high profile visits to the constituency by both Nick Clegg and Nicola Sturgeon.
Jo’s incredibly well known, and popular even amongst other parties’ voters. There can’t be very many MPs whose name is recognised by 96.7% of people. 60% of people overall have a favourable opinion of Jo. 56.1% of people who currently say they are voting Labour think favourably of her, as do 45.1% of those currently saying they will vote Conservative and 68% of those who have not yet decided who they will vote for.
The poll shows that 24% of people who voted Conservative in 2010, and 16% of people who voted Labour in 2010, have already switched to Jo to help stop the SNP.
It also shows that 46% of people who voted NO in the independence referendum in East Dunbartonshire currently plan on voting for Jo, while 69% of people who voted YES are currently planning on voting for the SNP candidate.
84% of people planning on voting Conservative, 77% of those planning on voting Labour and 62% of people who are currently undecided all say they voted NO. These voters need to know that a vote for Labour or the Conservatives in May make it likely they will be represented in Westminster by the SNP for the next five years
When asked how they would vote if only Jo and the SNP were standing, 35.8% of people who said they were planning on voting Labour in the voting intention question then said they were more likely to vote for Jo Swinson in those circumstances; conversely only 15.4% said they were more likely to vote for the SNP.
Similarly 44% of people planning on voting Conservative said they were more likely to vote for Jo Swinson having heard that statement and 10.1% said more likely to vote SNP
I’ve known for a long time how brutally hard Jo and her team are working, and that’s being recognised by the voters. 79% had recently received literature from Jo. Just 58% remember receiving stuff from the SNP.
Many commentators have assumed that Jo would lose her seat. This poll gives strong, recent evidence that she is very much in contention.
* Caron Lindsay is Editor of Liberal Democrat Voice and blogs at Caron's Musings
Follow @LibDemVoice on Twitter
Like us on Facebook
Subscribe to our feed
Sign-up for our daily email digest
Calling them out
The railway wars come to Market Harborough
Constructive opposition: a phrase you will hear a lot
Friends of Riverside Nature Park AGM #dundeewestend
Hugh Young
Roland
Kevin Hawkins
Steve Trevethan
54 Comments
Looking at the tabs, having the VI question take place only after questions on the candidates and 2010 vote, as well as including ‘knowing who is standing?’ at the end of the question might have skewed the results. That said, I’d be surprised if it made so large a difference as to completely warp the figures, so it looks like the race might be pretty close after all!
More importantly a poll like this means that a nice bar chart can be produced saying it’s neck and neck between Jo Swinson and the SNP – which should squeeze enough Tory/Lab voters to push her ahead on polling day.
Read the full data tables. It’s a push poll, pure and simple.
SNP landslide coming to East Dunbartonshire. Nothing to see here, move on and divert resources elsewhere, to seats we might actually hold.
As a politics lecturer, I’m concerned by the small sample size of this poll. As a Lib Dem I believe this poll shows that this is winnable for Jo provided we carry on campaigning: hard 🙂
I’m curious to know why a bespoke methodology was used?
That would be a truly sensational result if she wins.
I do not know how much credibility the poll has but to seems as though the seat will at least be close, and that in itself would be quite an achievement.
What George Potter says. Squeeeeeeze!
Presumably the reason HQ is selectively releasing some of these polls is to give us a good squeeze message?
“Presumably the reason HQ is selectively releasing some of these polls is to give us a good squeeze message?”
From which it is easy to infer that in Lib Dem seats where no such poll is published, the party is in trouble.
Sample size of 400 means the margin of error is something like 5%, and I think asking the favourability question before the voting intention one isn’t common practice in polling. Not quite a push poll, but I’m not sure how reliable it is.
Looking at the figures in the tables for this poll, the effect of “weighting” is surprising. Based upon 2010 Vote, 84 previous Lib Dem voters are weighted up to 101, and 57 previous SNP voters are weighted down to 34. Consequently, for the 2015 election, 111 SNP voters is rounded down to 101 while 104 Lib Dem voters is rounded up to 107, reversing the result before weighting is applied.
@Duncan Stott “Presumably the reason HQ is selectively releasing some of these polls is to give us a good squeeze message?”
One also wonders how many polls were commissioned in East Dunbartonshire before one gave the “right” result and was published!!
Anyone who knows about polling and methodology would treat this one with a wide level of skepticism.
Sad of the Lib Dems to resort to this but all parties do it I suppose. Having small samples, big lead in questions and then only releasing those that are favourable.
Current betting:
SNP 8/11
LibDem 3/1
Labour 5/1
Not sure that a poll using methodology designed by the LibDems is going to fool anyone. Looks like a last desperate act to try and hold the seat.
There is, apparently, an Ashcroft poll being done in this constituency. Or it may be an SNP internal poll. But somebody is doing a Populous constituency poll in a lot of Scottish seats. James Kelly, of Scotgoespop, thinks it might be Ashcroft III for Scotland.
I wouldn’t trust push-polls conveniently released like this.
Interesting,
I suspect it will be bad but not as bad as predicted. There is a very strong tradition of voting Liberal in Scotland. I also think Labour might do better than expected. I accept that SNP will be the biggest party, but not quite as big as the forecasts.
@Jim,
“Sad of the Lib Dems to resort to this but all parties do it I suppose. Having small samples, big lead in questions and then only releasing those that are favourable.”
What do they have left? If, as I suspect, Sir Jeremy Heywood will release the investigation about the French Memo leak before the election, and it shows that it was a LibDem leak, then even Alastair Carmichael’s seat in Orkney is probably going to go. Otherwise, there’s no conceivable way to avoid that all the Westminster parties will be reduced to single seats in Scotland. We may start to prepare for the panda jokes already.
I don’t know if I’d say it’s a push poll, but it does appear to be slightly heavy with No voters. The poll has 32% Yes voters – 59% No Voters but the actual result in the council area which admittedly is slightly bigger than the seat was 39% – 61%.
Dodgy methodology, small sample size, problems in weighting past voting recall but apart from that ray of hope.
However, it does show how incompetent the Lib Dem election ‘strategy’ is. Here we are working our socks off and pouring every resource into a seat just to remain in contention against a party which is barely trying and if we do hang on it will be due to the hardwork of the MP and her team, not because people are supporting the Liberal Democrats. When incumbency goes, so goes the chance of holding a seat.
Another new poll predicts Lib Dems losing every seat in the West Country. A more measured report is given on ukpolling report.
http://ukpollingreport.co.uk/blog/archives/9355
Fieldwork ‘9 April 2015’,
‘…..Q3. The next general election is now about 3 months away….’
Presumably the election after a hung parliament/no agreements and we have gone back to the county 🙂
The methodology, while not necessarily wrong, probably should have the voting intention question first (before asking about the candidates). The weighting seems reasonable enough (actually it’s slightly too SNP leaning, but a reasonable precaution against false recall). The biggest issue, which other’s have stated, is the relatively low sample size and the fact it was up to party whether to release it (or any other poll showing a worse result) or not. However even taking possible sampling error into account, those are good favourability numbers and given the poll numbers in Scotland as whole, it is impressive that she is still in with any chance at all.
(Technically a push poll, isn’t a poll at all but method of delivering a message/slur, usually with just a couple of questions and send out to thousands of people. Something like, ‘if you found out that candidate X had been taken to court over tax evasion would you be more likely or not to vote for them.’ This really is not a push poll)
Survation ‘disowns’ this poll. It was simply a tally-thing according to them, and was not conducted under proper BPC rules.
http://survation.com/in-reference-to-recent-liberal-democrat-polling-shared-privately-with-the-media/
“Survation were not responsible for drafting the questionnaires used, sampling design discussions or analysis of the results. These polls should therefore not properly be described as “Survation polls”. As a consequence, Survation is not responsible for the publication of these polls under BPC rules. Any member of the public with queries regarding the detail or further information about the mentioned polling work should be directed to the Liberal Democrats.”
Sounds like Survation would prefer it if you didn’t mention their involvement with this poll…
http://survation.com/in-reference-to-recent-liberal-democrat-polling-shared-privately-with-the-media/
“Jo’s incredibly well known, and popular even amongst other parties’ voters.”
Very much mixed feelings from me. Her work on allergies – great. Her misjudged body confidence campaign, which has helped encourage and normalise “skinny shaming” – not so great. (I have personal reasons for feeling very strongly about both those issues.) A lot of her feminist stuff is brilliant, but a lot of it wrong-headed and doesn’t help women in my view. So I can’t make my mind up about her, but on balance I hope she is returned as an MP again. The Lib Dems certainly need all the women of her calibre they can get.
George Potter 16th Apr ’15 – 1:11pm
More importantly a poll like this means that a nice bar chart can be produced saying it’s neck and neck between Jo Swinson and the SNP – which should squeeze enough Tory/Lab voters to push her ahead on polling day.
George,
I don’t know if things work quite like that in Scotland now, if they ever have in the last forty years.
Liberals and Liberal Democrats were not considered to be a Unionist party until the referendum. So to expect a Unionist squeeze to benefit a Liberal Democrat is perhaps expecting too much.
I have no wish to see any Liberal Democrat MP lose their seats but things are not as simple as putting out a bar chart and waiting for all the non-SNP voters to line up behind you.
If they were that simple, we might have won seats from Labour in the central belt in the last few years.
“…The poll, with methodology designed by the LibDems but carried out by an independent company,…”
Seriously?
A poll “designed by the LibDems” finds that all the other polls are at variance to one “designed by the LibDems”.
Does anyone else see a flaw in this methodology?
That the party’s methodology is more favourable to us is a given. But the thing is, many incumbent Lib Dems have, for many years, relied on a personal vote beyond the wider party’s voteshare in their constituency – voters who don’t think of themselves as “Lib Dems” vote for our candidates because they see them as good local MPs.
I think the party is releasing these polls at least in part to try to counter the narrative created by Ashcroft’s polls in these seats. Ashcroft’s constituency polling does not use candidates’ names at all, prompting only by party. In most LD held seats, I suspect that isn’t likely to produce accurate predictions either.
Personally, I would like to see polls done in which no favourability questions are asked first, but candidates are named. This strikes me as likely to give the best prediction, since it most closely resembles the situation of being in a voting booth with a ballot paper in front of you.
I suspect this is by the same heavily flawed methodology that showed Featherstone competitive in Hornsey and Wood Green. I suspect in reality that Swinson is significantly behind the SNP and this is a desperate attempt by Lib Dem HQ to convince unionist Tory and Labour supporters to back the Lib Dems.
I’m now seriously concerned though because I didn’t think the Lib Dems actually beleived their flawed polls… Oh well when Featherstone, Swinson, Burt ect. are all swept away in the national tide don’t say the polls didn’t warn you. If it wasn’t for Lisa Smart (Hazel Grove) the Lib Dems would be in serious danger of having no female MPs. I don’t think the Lib Dems will lose virtually everything but from analysing every seat the low to mid 30s seems most likely.
Please don’t get the impression that I am anti-Lib Dem I’m not I respect a lot of the work they have done in the centre ground of politics and I will actually vote Lib Dem in Bristol West. It’s just self delusion is one of the things I can’t stand.
JJ
Don’t be too confident about Hazel Grove. I think the last poll had the LibDems about 8% ahead, but UKIP were polling in the high teens. If the UKIP vote falls the Tories could still take it.
As is usual on these matter Sir Humphrey Appleby knows the score…
https://www.youtube.com/watch?v=G0ZZJXw4MTA
It;s not called ‘comfort polling’ for nothing.
https://youtu.be/G0ZZJXw4MTA
We actually have people here who appear genuinely to believe that the Liberal Democrats would spend £250,000 on private polls that are designed to give the party false information. Come on! Do remember that most of these polls have not been published. Their purpose is to show the party organisation where to put its precious resources. Who is being delusional?
Jo Swinson’s seat is unlikely to be fertile ground for the SNP. It is stuffed full of academics and houses a fair chunk of the Glasgow middle-class. The higher one’s IQ, the less likely one is to vote SNP. East Dunbartonshire is following a similar pattern to other suburban seats where our vote is holding up more strongly than in our more rural seats. I would expect to find a similar effect in Edinburgh West.
Malc, who previously foretold doom for Paul Burstow (despite the Ashcroft poll predicting the precise opposite), is now telling us that we are going to lose Hazel Grove. We needn’t bother with the inconvenience of an election when we have people like Malc to do the job for the electorate.
And as for the Com Res poll in the West Country, which was based on a sample less than a 10th the size of the ones used by Lord Ashcroft, those who take this poll seriously are simply letting their hatred of the Liberal Democrats cloud their judgment. They would love Com Res to be right, but the evidence tells us that it is not right.
It is clear which seats the Liberal Democrats are likely to win. Follow Nick Clegg. He goes to the areas where the internal polls show that the party is either ahead or in contention.
I’m sorry to disappoint the hyenas queuing up here to eat us alive. Yes, we will lose seats, but the result will be far from carnage.
Sesenco
I think the LibDems will struggle to hang on to Hazel Grove and Sutton and Cheam. You think the LibDems will hold on to East Dunbartonshire and that the GE won’t be carnage for the LibDems. We will have to wait and see who’s right, not long to go now.
Sesenco I think the Lib Dems have every motivation for releasing dodgy polls to the public because a) it creates a positive media story and b) it gives them a greater opportunity to squeeze the votes of uncompetitive in a given seat. I remember reading that the methodology of the Hornsey and Wood Green Lib Dem poll was deeply statistically flawed so if this poll used the same methodology then it is not credible either. It is important also to note that there is no obligation for the Lib Dems to release polls to the public and therefore they will only release good ones so who knows how many other polls of this seat they have done which show disaster for Swinson.
It is true that East Dunbartonshire is not great territory for the SNP but with Scottish polling as it is they will probably get around 35% and I can’t see Swinson losing less than 4% of her vote given the Lib Dems dire Scottish polling. She will probably keep the margin of defeat respectable though. I hope I am wrong but I don’t think she has much of a chance.
My projections of Lib Dem seat losses:
To Lab: Norwich S, Bradford E, Brent C, Withington, Burnley, Redcar, Hornsey, Cardiff C
To Con: Solihull, Mid Dorset, Wells, St Austell, Somerton, Chippenham, Taunton, Berwick, Devon N, Portsmouth S
To SNP: Dunbartonshire E, Argyll, Edinburgh W, Gordon, Caithness, Inverness, Fife NE
The Lib Dems could well hold on to a few on this list but equally they could lose some that aren’t on it: I think Ross is very vulnerable to the SNP and Berwickshire to the Tories. Watford and perhaps Oxford W are potential prospects for gains. I know from first hand experience that the Lib Dems have almost completely abandoned St Albans to presumably throw everything at helping Thornhill in Watford which is a bit of a shame since Sandy Walkington is a good candidate. I also think Clegg is more vulnerable to Labour in Hallam than many would like to believe, in the end I think he will probably hold but it will be no walk in the park.
Not too sure that Cardiff Central will be lost, the LDs lost the corresponding assembly seat by a handful of votes but at a time when voters, and especially students, were most furious with the party. It was bad luck for the candidate then, but Jenny Willott is known, liked and could keep the seat.
We ought to be able to add Montgomery to the list of seats to be recovered, but not perhaps this time ?
It’s a different methodology but that doesn’t mean it’s a flawed methodology. AIUI the adjustment for don’t knows isn’t Survation’s usual (30% of past recall) but is on the basis of 100% of past recall. You can argue as to the validity of that but I think there are other polling companies that do this.
400 is the usual size for the party’s internal constituency polls – it gives a 5% MOE
Oh – and if people say this is a push-poll they really need to understand what that is!
Actually having read the blurb at the front this does seem a similar past vote weighting to Survation’s normal polling.
the Tory and Labour vote is collapsing in East Dun with most Tories I have spoken to voting for Jo. Labour are nowhere to be seen having no doubt given up to concentrate their resources in SNP threatened Cumbernauld Kilsyth and Kirkintilloch East which is next door. My money is on Jo getting an increased majority as Unionist leaning voters rally behind her.
I thought Jo Swinson performed very well on last nights Question Time. (I haven’t watched it for ages and watching last night reminds me why I don’t normally watch it – Piers Morgan was bullying and just seemed to want to attack all politicians and why does David Dimbleby know the wording of the motion on the balancing of the budget so he could stop Yvette Cooper and Grant Shapps arguing over the wording. It seems that the SNP are talking about the Charter for Budget Responsibility which states that a cyclically-adjusted budget surplus has to be included in the rolling five year plan, that by 2015-16 the deficit is reducing as a percentage of GDP and that welfare spending is within the welfare cap as set in the most recent budget.)
The other point I forgot to mention in earlier posting is that East Dun is Labour’s number one target seat in Scotland. (they had a big map with an arrow pointing to the constituency at their Manchester conference saying so). Astonishing that they are now polling at 16 per cent here in this poll which would reinforce the theory that they have given up and are working to hold the next door seat. Their private polls will already have told them the same grim story. Graphically illustrates the point that they are under such pressure from the SNP. Only one Labour leaflet has ever arrived at my house. What does that tell you? This poll will do the trick and cause a stampede of Tory and Labour voters to Jo. Remember also that East Dunbartonshire voted 61/ 39 to save the Union.
For an election which is the hardest to predict in maybe a century, an awful lot of people seem awfully sure about the result in a number of seats 🙂 Discussing this with some friends a few nights ago I summarised their projections as making 5 gains but losing the seats of the current and 2 previous leaders. We then agreed that that was an unlikely overall outcome!
FWIW I think the LIb Dems will do poorly by an objective standard, but people will claim that it is better than/as expected and there will be a few seats which massively outpeform the average and are either unexpected holds or gains. There will also be some seats where the results will show that people were making massively optimistic projections or suffering from rampant candidatitis! Working out which is which will be the hard task!
Hywel, the problem is that it is hard to work out what an ‘objective standard’ for failure is now in this election.
Is it not getting into government in coalition?
Is it not being even asked to the table for negotiations?
Is it a significant drop below the 56 seats we have now (and could still reasonably expect if we had a PR system and polled 8% nationally)?
Is it a significant drop below the 40 seats the leadership were predicting for most of the parliament?
Is it something in the lower part of the 10 to 25 seat range many ‘uniform swing’ pollsters are predicting
Is it a fall in our national vote share (in which case we have already failed)?
Is it a fall in our vote share / predicted seats since the start of the campaign (in which case we can claim a ‘success’)?
Is it a failure to stop or stem the SNP surge in Scotland or any UKIP re-surge in the south (but we can’t be held solely responsible for that if it does happen)?
Take your pick.
I’d say (not as an insider) that:
– failure (and immediate automatic de-Cleggification) is now below 20 seats,
– mediocrity (and a period of horrible infighting) is 20 to 35 seats,
– and success (and a longer period of barely suppressed infighting, plus the possible poisoned chalice of another coalition) is over 35.
Woo.
Hywel – “FWIW I think the LIb Dems will do poorly by an objective standard, but people will claim that it is better than/as expected and there will be a few seats which massively outpeform the average and are either unexpected holds or gains. There will also be some seats where the results will show that people were making massively optimistic projections or suffering from rampant candidatitis! Working out which is which will be the hard task!”
That’s what betting markets are for.
We support candidates who are defending seats and those working to win new LD targets. However, there are many LDs who live in constituencies, wards etc with no chance of a winning LD candidate. Tactical voting is available – to help elect the next best party to work for our principles. We would do this if we had STV anyway. Vote for a second party. There is a useful graphic in the Guardian which reminds voters of the state of the parties in each constituency.
matt (Bristol) – “I’d say (not as an insider) that:
– failure (and immediate automatic de-Cleggification) is now below 20 seats,
– mediocrity (and a period of horrible infighting) is 20 to 35 seats,
– and success (and a longer period of barely suppressed infighting, plus the possible poisoned chalice of another coalition) is over 35.
Woo.”
All political careers end in failure. Otherwise we’d have one-party government in perpetuity.
After last night’s performance by Jo on Question Time when she took the SNP to task, she should pull further ahead and win the seat.
My aged memory suggests that even back to the 80s public opinion polls for constituencies aimed for about 400. Willing to be contradicted.
Swinson’s performance was alright though it was nothing special. The best performing panellist was undoubtedly Piers Morgan (although I don’t like him). Her major blunders were when she said the Tories banned Clegg from going to the debate (a very silly thing to say) and when she asked Piers Morgan why he wasn’t a solid Lib Dem voter which reopened the tuition fees can of worms. A game changer for Swinson it most certainly was not (not least because the vast majority of her constituents don’t watch Question Time).
I am currently predicting 33 Lib Dem seats but it could in reality be a fair bit worse or a fair bit better, it depends how this targeting strategy is going. There is absolutely no point throwing resources at several Labour facing seats such as Brent C, Withington, Norwich S ect. where bloodbaths for the Lib Dems are likely. Also it would be wise to cut Some Tory facing seats like Somerton adrift because in this case you have Heath retiring, very small majority, dreadful local results, not selecting a local (David Rendel is a very poor choice) and lots of more salvageable seats nearby. If the Lib Dems ‘ruthless’ targeting is properly executed they could end up closer to 40 seats.
Those numbers are interesting when compared with the Ashcroft polls for Scotland released a few minutes ago. On his constituency question East Dunbartonshire is 40% SNP and 29% LD, but with a substantial Lab (16%) and Con (12%) support still remaining. This feels like a classic “2 horse race” squeeze campaign. And the squueze message should lead on Ashcroft and ignore the dodgy internal polling.
Disappointing Ashcroft poll for this seat – though with the squeeze effect we might still have a fighting chance.
On a more positive note, at least we’re within the margin of error of holding Berwickshire, Roxburgh and Selkirk.
https://pbs.twimg.com/media/CCzXejlWEAAyr3p.png:large
I said earlier I wanted Jo to keep her seat, but I’d forgotten that she was one of the MPs who broke the NUS pledge. This tips the balance back the other way so now I think it’s essential she loses.
The latest Ashcroft poll makes grim reading:
http://www2.politicalbetting.com/index.php/archives/2015/04/17/new-ashcroft-scottish-seat-polling-finds-snp-gains-from-lab-ld-con/
The amount of LibDems who refuse to accept what is happening is amazing. The party is polling about 4% in Scotland – how on earth is Jo Swinson going to hold her seat.
PS. Surely LibDems knew the Ashcroft poll results were due out, who’s clever idea was it to release the results of their own “poll” the day before. It just makes them look desperate and silly.
malc and assorted other pessimists – this is for you: https://www.youtube.com/watch?v=w7RIgs3eygo
There are nae pubs in Kirkintilloch (The Corries) but there is a local newspaper?