The day after the election the British Polling Council announced it was going to have an inquiry into what went wrong with the polls, we’ve now got some more information about how the inquiry is going to proceed. Over on the National Centre for Research Methods website they have announced the membership of the inquiry team, timings and terms of reference.

The Chairman of the inquiry, Pat Sturgis, was announced earlier this month. The rest of the team include several names who regular readers will be familiar with: Steve Fisher from Oxford University who ran the ElectionEtc model and worked on the exit poll, as did Jouni Kuha of the LSE, Will Jennings of Southampton University who is part of the Polling Observatory team, Jane Green of Manchester University who is the current Director of the British Election Study and Ben Lauderdale of the LSE who did the ElectionForecast model that was on Newsnight and 538. The rest of the inquiry team are Nick Baker of Quadrangle Research, Mario Callegaro of Google and Patten Smith of Ipsos MORI.

The terms of reference for the inquiry are to assess the accuracy of the 2015 polls and investigate the cause of any inaccuracy, whether it’s connected to inaccuracy at previous elections, to look into the possibility of herding, to see if enough information was provided and communicated to people about how polls were done and what they meant and make recommendations on how polls are conducted and published in the future and on the rules and obligations of the BPC.

The inquiry are inviting written submissions via their website, and there will be a public meeting on the 19th June – it’s due to report to the BPC and MRS by the 1st March next year.

571 Responses to “Details of the polling Inquiry”

1 10 11 12
  1. Well, the Conservatives undercut Labour with £1, but it’s less important there. Tesco vs Sainsburys and Aldi wins.

  2. I rather like Kendall, Creagh. They have the advantage of being new and different.

  3. Lazlo

    BPC rules say that pollsters must respond to queries about their polls.

  4. Pete B

    They said the same thing when the Tories held their Totnes ‘primary’.
    The result was that Dr. Sarah Wollaston, my favourite MP, was selected.
    She was head and shoulders above the other two candidates, and the Totnes public voted accordingly – it almost didn’t matter which party they supported.

  5. @ OldNat

    Yes, I know that. I’m more interested in the clients’ attitude in this one. I’m just surprised by it. How would it affect the service provider (the polling company)? OK just two points, they don’t even know if the polling company surveyed registered pullers only (I hope they did, but I have my doubts). Secondly, they were also surprised about the gender distribution, yet they didn’t queried it apparently, and advised me to contact the polling company. Yet they didn’t have the problem with reciting the results.

  6. Pullers = voters

  7. Lazlo

    Good points.

    I’ve only been involved in one poll from the stance of helping the client understand the results.

    On that occasion, we went through the tables pretty thoroughly and had a number of questions for the pollster. Most of the points were answered to our satisfaction, though the tables were adjusted in a couple of areas where we discovered errors.

    In retrospect we felt that, despite discussion with the pollster as to the questions, the poll was less useful than it could have been as there had been simple acceptance of the pollster’s methodology, rather than specifying those demographic factors which were most important.

    In the case you mention, the client (unlike the one I was involved with) has undertaken frequent polling, so our naivety wouldn’t be an excuse open to them.

  8. @ OldNat

    “there had been simple acceptance of the pollster’s methodology, rather than specifying those demographic factors which were most important.”

    Indeed, and Hawthorn had some comments on this. They were very sharp, but lacked details.

    Several years ago I did some work on polling for an English football club. They (the club) just took the polling in without questions. As I normally work with clients that are much more vulnerable that this one, I gently led them to the recognition that they were committing several millions of pounds without really looking at how the data was created (it was fine by the way, even if the client abandoned the project eventually). Then I had a two hours long session on the methodology (not on the data) to verify the opportunities and constraints in the data. It was an eye opener. They had no knowledge about methodology, yet they understood what I was saying, and had the right questions, and they nodded at the right time. They understood what MoE was, yet they didn’t miss the magnitude. Oh, digressing …

    Anyway … What’s the responsibility of the client (when it goes beyond the headline figures)? What messages it gives to the polling company? Who loses credibility? How does it obstruct or encourages learning, etc?

    I went much longer than I wanted instead of just saying that I agree with you …

  9. MS has tweeted that the wording of the EU referendum will be as follows:

    “Should the United Kingdom remain a member of the European Union?”

    I’m rather surprised at this in that (as MS also observes) the last two major referenda within the UK (AV and the IndyRef) were both won by the “No” sides.

    While it could be argued that the electorates in those referenda voted No to reject change, and that in the case of the EU referendum the reverse would be true (I.e. Yes would be to maintain the status quo), I’m far from convinced that that people in the UK are more likely to register a positive Yes to remain in the EU, than that would a No to leave the EU, had the question been phrased that way.

  10. @ProfHoward

    Me too… I’m afraid the SNP put the CONs and LAB to shame with their showing in the chamber last night. It probably won’t last once they get involved in committees and so forth but so far looks like they’re very keen to contribute and a talented bunch. I hope they don’t regress back into clapping though!


    I think £3 is probably enough of a deterrent for most outsiders to resist. However, the leadership is a slightly different matter from a local MP – if I lived in a safe Labour seat even though I’m not a Labour supporter I’d want the best candidate available to represent my constituency whilst with the leadership as a non-Labour supporter I’d rather they picked the most useless leader available.

  11. @Raf

    I think you’ve answered that one for yourself in what you say about the status quo. Phrasing it the other way also legitimises leaving far more than doing it this way. Personally think it will be quite a clear win for ‘Yes’, probably on a similar scale to 1975 – unless renegotiation blows up and Cameron calls for Brexit.

  12. Jack Sheldon

    Agreed, but what happens when the most useless Labour leader, in your opinion, ends up as Prime Minister?

    Why are we not all urging Lord Falconer to stand…

  13. As the Scotland bill has now been published -which of the powers will NS choose to use ?Income tax ,benefits etc

    If I were a candidate for Labour leader I would be pledging right now to match the shrewdly targetted snp policies of recent years in england.Free travel over 60(londoners have got it thanks to boris),prescriptions ,care,hospital car parking(maybe not tuition fees) -any others ? English voters would surely welcome these rather than the whatever incomprehensible tory EVEL proposals.

    If the tories have got their money tree labour needs to grow its own.

    Better to do it now before the snp start winding everyone up about it.

  14. 07052015

    I’ll try to avoid policy discussions, but shouldn’t government be trying to make life as uncomplicated as possible for everyone? So get rid of TV licences, get rid of tolls on roads, get rid of the winter fuel allowance, merge NI with income tax, etc., etc.

    Simplify everything!

    The first hundred days of policy drafted by Lord Falconer had Labour won, was a dispiriting collection of interventionist proposals, but by far the worst factor was their astonishing complexity. It made my head hurt.

    Love bus passes for the elderly – all sorts of benefits and an extremely simple system.

  15. @07052015

    One might argue a lack of universal policies is exactly what divided the country into devolved parliaments to decide who gets what!

    Labour can quite nicely say they are declining tax cuts (Tories will promise big tax cuts next time) in favour or universal policies that give middle income earners access to key services. Only taxes (such as national insurance) where the government promises to give something back for every pound taken are truly popular.

  16. I do wish some company, somewhere, had the courage to do a poll or two, using the same methodology as before the election. It would be interesting to see if the polls still produced a neck on neck result, or if they now reflected a more favourable picture for the Tories.

    Don’t know what those results might mean, but it would be interesting. If you’re me, but maybe not anyone else.

  17. I am still puzzled why their is need to publish the referendum bill before both the date has been finalised and negotiations are finished. Should have just waited till both are finalised then publish the bill. It will need another piece of legalisation probably to authorise the exact date and pass some of the reforms.

  18. And it wont be in the Summer has Hammond as said. Nothing electorally happens in the summer especially one where there is Euro 2016 and the Olympics to contend with.

  19. Lurgee

    Fully agree, have thought the same the last 3 weeks. Although it’s quite nice for the media to focus on policy rather than opinion polls for a change as well.

  20. Lurgee – it’s not necessarily that simple! Apart from Ipsos MORI all the other companies use some form of political weighting and in most cases that’s weighting back to the results of the previous election.

    Hence even “keeping the same method as before the election” would in most cases mean switching from weighting using recalled 2010 vote to weighting using 2015 vote.

    Of course, in terms of working out what went wrong asking people how they voted using the old weighting is a useful tool, but there’s probably a difference between polls being done & published to measure current views and polls being done & published to try and diagnose and work out the errors in the election polls. A lot of the latter will be going on behind the scenes… though ultimately, the best data to work with is that from before the election and there’s a lot to play with.

    BM11 – it won’t necessarily need any extra primary legislation. We’ll know when we see the full text, but its likely that it’s enabling legislation setting up all the rules of the referendum now except the date, and the date itself can be set by statutory instrument later.

    The AV referendum legislation and the Scottish referendum legislation both set a date for the referendum in the Act itself, but gave the Secretary of State the power to vary the date by a few months using a statutory instrument.

  21. Anthony

    Small point, but the Scottish Independence Referendum Act 2013 allowed the Scottish Ministers to delay the date of the Referendum up to the last date permitted by the Section 30 Order, but not bring it forward.

1 10 11 12