Is this the ‘smoking gun’ of YouGov’s anti-Corbyn bias?

It’s long been contended by supporters of Labour leader Jeremy Corbyn that polls are biased against him – and in particular, polls by YouGov. Those supporters would agree with Peter Hitchens’ famous analysis of polls in general:

hitchens polls

YouGov was founded by Tory Minister Nadhim Zahawi and former ConservativeHome owner Stephan Shakespeare, so suspicions among left-wingers run high.

Labour and especially Corbyn supporters have long criticised YouGov’s polling, but the company tends to keep its methodology cards close to its chest. However, a poll published today is so weak, such a stretch that it’s hard to read it as anything other than an attempt to further damage Corbyn.

The poll is titled ‘A tale of two parties – what we learned from our Labour membership survey‘ and claims to contrast the attitudes of supporters of Corbyn and those of his challenger in the Labour leadership contest, Owen Smith – which of course immediately raises the question ‘Why now?‘, given that the leadership contest finished almost 6 months ago.

The most suspicious item in the whole thing is a ‘pie chart’ supposedly showing differences in where the loyalty of each set of supporters:

yougov loyalty

At first glance, the difference looks stark – especially with the chosen strong red colour used to denote loyalty to the Labour Party rather than to its leader or a mixed position. But a closer look shows the methodology to be full of holes:

  1. The chart is clearly intended to portray Corbyn supporters as loyal to him rather than the party – in other words, as ‘entryists
  2. The poll completely ‘forgets‘ to ask Smith supporters about their loyalty to Smith – of course, most Smith supporters are not going to be loyal to Corbyn
  3. The poll looks at percentages among the two supporter-groups and, as the article’s title suggests, uses this to portray Labour as a divided party – but completely neglects to mention that Labour has almost twice as many Corbyn supporters as Smith supporters 
  4. Another major issue is that there are no ‘Smith supporters’ now. He is a defunct figure, an obscure back-bencher. Those who voted for him don’t have anything left to be loyal to except the party – and they’re not offered any other option, at least as far as the poll reveals

By its (deliberately?) poor construction and presentation, the poll portrays the party’s right-wingers who voted for Smith (or more realistically, ‘anyone but Corbyn’) as loyal to Labour while implying that those who support Labour’s direction under Corbyn are depicted as more of a ‘personality cult’.

A poll so catastrophically badly-constructed makes it hard to see beyond two possibilities. Either YouGov is monumentally incompetent at designing and interpreting polls – or this is a ‘smoking gun‘ of the fact that Peter Hitchens’ reading of the situation is accurate and it’s not meant to measure anything, but rather to influence readers of the article (or more likely, of articles in newspapers opportunistically exploiting the pie chart) into a negative opinion toward Corbyn.

In a poll of YouGov employees and owners, I wonder which one they’d prefer to be seen as.

The SKWAWKBOX is provided free of charge but depends on the generosity of its readers to be viable. If you found this article useful and can afford to, please click here to arrange a one-off or modest monthly donation via PayPal. Thanks for your support so this blog can keep bringing you information the Establishment would prefer you not to know about.


  1. No hope for any Labour candidate or any Labour leader (I wasted my money on JC) as long as they don’t outright oppose Brexshit. That simple. The LibDems, the SNP and the Greens will destroy you. Honestly.

    1. 1) You assume that the 48% are all Labour voters – they are not.
      2) You then assume that all 48% are hanging on Brexit – they aren’t.
      3) Truth is that Brexit will m ake precious little difference to how people place their votes, if people were overly worried about the EU more would have been bothered to vote!

  2. Wonder where these muppets get their Labour members from? Who gave them access to membership lists or has anyone reading this ever been approached by them anyway?

    Makes you think when you stop for a moment and look at it.

    1. They get their information from people registered with YouGov and YouGov must be the only polling company that actually keeps a record on how its members vote politically. It’s how they can manipulate to get the results they want.

      1. I didn’t know that. Keeping polling data on registered members must be the most unethical practice in the dubious world of pollsters. For statistics to be meaningful, the respondent should be chosen at random from the general population. The should not be on a pollsters list of registered volunteers, waiting to be asked what the polling company already know (or suspect) as a result of what’s
        in their their data base. If this is happening, it is time for the polling organisations to be regulated, or perhaps come under the umbrella of the Electoral Commission. Only putting Peter Lilley in charge of electoral services would be harder to believe – oh wait.

    2. I am a Labour party member and a Corbyn supporter. I also, occasionally, complete YouGov surveys. I have never been approached by YouGov for my political opinions. There are sometimes, however, questions at the end of surveys asking about political leanings. If they use this information it would be easy to choose the respondents who will give them the responses they are looking for. I had not realised, until reading this article, how YouGov operates. I won’t be doing any more of their surveys.

  3. Is that really the best that you can do? – as you say Owen Smith is now defunct – therefore point 2 -3 and 4 are also defunct.

Leave a Reply

%d bloggers like this: