A polling question to which the answer is yes (or at least probably)

by Atul Hatwal

Earlier today Anthony Wells posted a response at UK Polling Report to a post of mine, “Is Labour’s lead actually 3% not 10%?

Anthony makes a series of reasonable points defending the accuracy of YouGov’s 10 point Labour lead. He cites the high level of precision of YouGov’s final poll in the London election and the greater robustness of polls today compared to twenty years ago. Fair enough. But this misunderstands the purpose of my piece.

Polls measure how people feel at a point in time. There are adjustments and various psephological tricks to ensure respondents reveal their true preference, but the emphasis is on measuring how voters are actually feeling when the poll is conducted.

In mid-term, with an election some way off, many people might feel thoroughly hacked off with the government and be convinced they want rid of them. But when faced with the real choice, when it is days before polling and they engage with the arguments, a significant minority will change their minds.

By definition, this cannot be picked up by polling several weeks out from an election because voters haven’t switched yet.

The point of my piece was not to contest the accuracy of current polls, but to try measure this effect: to attempt to quantity the proportion of respondents who are relatively convinced Labour backers in mid-term but will then switch when the campaign ramps up and the polling booth beckons.

In fact, it is only because modern polling is so much more effective that we can be confident that a shortfall between mid-term polls and actual election results reflects voters who changed their mind rather than those who simply lied to the pollster in the first place.

Looking at YouGov’s poll on April 13th, when Labour voting intention was at 50%, the London election was still three weeks off. The London race hadn’t captured the public imagination and was far from the forefront of most Londoners’ thoughts.

But it was sufficiently soon after the budget to capture public reaction to the government’s escalating omnishambles.

YouGov did ask a question in this poll specifically about voting intention for the Assembly constituency section, but adjusted results based on likelihood to vote. This meant these results were not comparable to national polls which do not adjust in the same way. Hence the use of the poll’s headline voting intention figure in my piece.

This figure from this poll was the nearest approximation to the current national mid-term polls.

Comparing voting intention from this early April poll to the final result gives a sense of the numbers of voters who will change their mind and switch their vote when the choice becomes real.

It showed that almost 1 in 5 Labour supporters did not follow through and vote Labour in the Assembly elections. In comparison, roughly 1 in 33 Tory supporters from the April poll did not vote Conservative.

Anthony makes the valid point that the type of election determines whether people actually vote and this will also have impacted the London result. Turnout at the London election will likely have been lower than at the next general election.

But equally, this would also apply to the Tories so the net impact on Labour is debatable.

Understanding the extent of this soft support is critical for a party in planning its next move. It is impossible to capture in polling significantly before election day but will often be the difference between success and failure.

In his piece, Anthony closes with the observation “Labour DO have a lead of somewhere in the region of 10 points now. Whether they’ll have one in 2015 is an open question”.

I’ll make this challenge in response, Anthony, do you think if an election were called tomorrow, with just a two week campaign, that Labour would win by 10 points?


Atul Hatwal is Editor at Uncut

Tags: , , ,

6 Responses to “A polling question to which the answer is yes (or at least probably)”

  1. paul barker says:

    Wells makes the point that polls are only relevant to the election they are asking about, since all the intention polls ask about an election which will never happen they have no relevance to anything much, certainly not to the real election in 2015.
    The question we cant answer is if the introduction of fixed-term parliaments has made polls even more dtached from reality ? Are the voters beginning to realise that there will be no election for 3 years, freeing them to say what they like to pollsters ?

  2. Matthew says:

    You can’t ignore the Assembly poll as easily as you do. It gave figures of 44-35 lab to con http://d25d2506sfb94s.cloudfront.net/cumulus_uploads/document/oaxiy1v7sf/YG-Archives-EveningStandard-MayoralRace-160412bpc.pdf which compares well to the result three weeks later of 41-32. The reasons for the difference between the GE poll and the assembly poll (seems mostly lib dems, also turnout) is surely the issue, not whether people do what they say (as they did do what they said for the London-wide election)?

  3. Hi Atul,

    I took your original piece as warning that Labour’s poll lead might be soft because (a) the polls might be wrong and (b) because these may be relatively good polling times for Labour, and their lead may shrink as people consider their options, the reality of the choice looms and it becomes more about a choice of alternatives, rather than just an expression of anti-government feeling.

    I think you are right on (b), and while it would be tempting fate for me to dismiss (a) completely, I don’t think there is any reason to think it is the case. If, of course, you only meant to say (b) in the first place they we are of one opinion (I’ve written about whether Labour’s lead is big enough given the tendency for mid term leads to fade away here)

    To answer the final question (in a sort of roundabout way) if we were two weeks out from a general election and the polls were showing a 10 point Labour lead, then I suspect Labour probably would win with that sort of lead. Polls are pretty accurate these days and election campaigns don’t normally lead to big shifts in support.

    However, to give that a heavy caveat, if we actually were 2 weeks away from a general election I wouldn’t expect Labour to necessarily have a 10 point lead to begin with. Minds would have been concentrated, Labour would have set out a full range of policies, the Conservatives would presumably have put out some palatable policies rather than frontloading the nasty stuff, the Lib Dems would have tried to differentiate, etc, etc. All of those things may have changed minds.

    In the rather unrealistic circumstances of an election being called tomorrow, all those things might happen in the space of 2 weeks, and it would be perfectly possible to have great big shifts in support in two weeks (not to mention the impact of whatever event had just brought the government down!)

  4. Good discussion on both sides. For what it’s worth (and this has no scientific basis) I see the q YouGov asks ‘would you prefer a Conservative Government under David Cameron or a Labour one under Ed Miliband?’ as the key indicator of where things actually are. Last time I checked this gave Labour/Miliband a 6% lead. I think this is where things basically are. Where they will go is anyone’s guess.

  5. swatantra says:

    Polls usually get it pretty right the day before the actual election; in 2010 they predicted a Hung Parliament and they got it spot on. But I always fear that there is a ‘bandwagon effect’ of polls, and they can actually influence the way people actually vote on the day. So it would be correct to say that people actually voted for a Hung Parliament and a Coalition. The French are right to ban the publication and all talk about polls in the Media a few days before an election. And we should do the same.

  6. vin maratty says:

    it seems your search facility is not very efficient. I typed ‘use of troops during firemen’s strike’ and ‘labour closes 100s of mines’ and it came up with nothing

Leave a Reply