Skip to content


Voodoo psephology

With such a catchy title this post has to be about the Guardian and indeed, one of their reporters has managed to ease himself from behind his comfy London desk this morning so as to consult a map to refresh his memory as to where exactly Wales is.

Any indication that he actually crossed the Severn Bridge to meet and talk to voters and get a feel for the political landscape before setting fingers on keyboard is missing from the article. The fact that he ended with the line: ‘Land of my fathers no more’ suggests that he accidentally picked up a book on clichés rather than one on politics.

Martin Kettle writes today that Wales has been lost to Labour and that it is the Conservatives that are the ascendant party. I know of many people who might disagree with him though in my view there is no doubt that Labour will lose ground in Wales at the next General Election and the Tories will gain seats.

To be fair Mr. Kettle’s conclusion is that ‘Any claim that Wales is a Conservative nation now – especially based on the support of fewer than one voter in three – is ridiculous. But the idea that it is still a Labour nation is increasingly ridiculous too.’ That is a reasonable assessment. However, his reliance on an out-of-date and statistically unreliable YouGov poll together with the Electoral Calculus website makes this piece more voodoo psephology than informed comment.

Mr. Kettle’s claim that ‘the distinctive pattern of Welsh politics has many resemblances to England. If England is indeed about to turn to the Tories, it is not unlikely that Wales will do so’ is unbelieveable. Perhaps he might like to elaborate. The fact that England has often secured a Tory majority of votes (and seats) is important here. Wales has not been in that position since the mid 19th Century, before the franchise was widened, with the sole exception of the last European Election on a very low turnout.

If I recall correctly (and one of the weaknesses of the article is that it does not source the poll it is based on nor give figures) the YouGov poll was based on marginal seats across the UK. It did not contain a sufficiently large sample of the Welsh electorate to accurately make any prediction as to how people would vote, it was sampled some time ago and certainly before the European Elections, and did not take account of specifically Welsh circumstances, such as the four party system for example.

As for the Electoral Calculus website, well don’t get me started. It is based on uniform swings when all the evidence is that we now experience major regional variations in party support. It does not take any account of individual circumstances nor does it recalculate predictions to take account of contests subsequent to the last General Election. Mr. Kettle might as well have stuck a wet finger in the air.

In my view the electoral map of Wales after the next General Election is still very much up for grabs. There are seats such as Aberconwy that could go to either Plaid Cymru or the Tories. We do not yet know whether the nationalists can carry over their support on a UK level as they so significantly failed to do in 2005. And nobody should write the Liberal Democrats off yet.

Evidence I have seen and my own campaigning experience suggest that we have a very strong chance of holding onto our most marginal seat in Ceredigion and that although there will be a strong Tory challenge in Brecon and Radnor, we should hold on there as well. I am confident of holding Montgomery and Cardiff Central and I fully expect to take one, if not both of Swansea West and Newport East.

Wales is no longer a Labour stronghold, recent election results indicate that, but in the absence of regular and reliable polling it is impossible to accurately predict what will happen at the next General Election. Perhaps next time Mr. Kettle tries he might come to Wales first and actually sample the air.

Related posts:

  1. New Welsh poll puts Lib Dems on course for gains
  2. European Elections – a retrospective
  3. Poll puts Welsh Liberal Democrats on course for outstanding result

Posted in Uncategorized.

Tagged with , , , .


4 Responses

Stay in touch with the conversation, subscribe to the RSS feed for comments on this post.

  1. Jack says

    This is too Welsh for me. Can anyone explain what poll the author refers to, why it is outdated and unreliable, when YouGov polls on elections have in the past turned out to be the most reliable, predicted actual results to within one percentage point?

  2. Alwyn ap Huw says

    No matter how reliable YouGov might be on an UK basis, their polls can not be used to get any indication of Welsh voting patterns because Wales is polled as part of the “Midlands” by YouGov rather than as a separate “region”

  3. Jack says

    I now see that the Guardian story referred to “a detailed YouGov regional opinion poll analysis by the Electoral Calculus”. If YouGov is reliable but has not polled Wales separately but as part of the Midlands, are you then assuming that the analysis by the Electoral Calculus is unreliable?

  4. Peter Black says

    I outlined in the post precisely why Electoral Calculus is unreliable.