Final Welsh Political Barometer Poll of the Election

Today’s Welsh Political Barometer poll is our final one of this general election campaign. With polling stations opening in less than twenty-four hours, how do the parties in Wales stand on the eve of this, most unpredictable general election? And how has support for the different parties changed across the course of the campaign?

Our new poll was carried by YouGov as late as possible – between Monday and this morning. Here are the figures for general election voting intention (I’ve put in brackets how our poll figures compare with the vote-share won by each party in Wales in the May 2010 election):


Labour: 39% (+3)

Conservatives: 25% (-1)

Plaid Cymru: 13% (+2)

UKIP: 12% (+9)

Liberal Democrats: 8% (-12)

Greens: 2% (+2)


This is the fifth poll that YouGov have carried out in Wales since the start of the election campaign. How much has support for the parties changed? The following table shows how support has ebbed and flowed since our late-March Barometer poll:


Party 27/03 31/03 15/04 30/04 06/05
Labour 40% 40% 40% 39% 39%
Conservative 25% 27% 26% 26% 25%
LibDems 5% 6% 6% 6% 8%
Plaid Cymru 11% 9% 12% 13% 13%
UKIP 14% 13% 13% 12% 12%
Greens 5% 5% 4% 3% 2%

* All polls conducted by YouGov. Dates listed for polls are dates when fieldwork was completed.


If we apply the swings implied by this poll from the May 2010 general election result uniformly across Wales, this produces the following outcome in terms of parliamentary seats:


Labour: 28 seats (keeping the 26 seats they won in 2010, and gaining both Cardiff Central from the Liberal Democrats and Cardiff North from the Conservatives);

Conservatives: 8 seats (losing Cardiff North to Labour, but gaining Brecon & Radnor from the Liberal Democrats);

Plaid Cymru: 3 seats (no change);

Liberal Democrats: 1 seat (losing both Cardiff Central and Brecon & Radnor, and holding on only to Ceredigion).


This final poll suggests that the campaign has had only a limited impact in terms of changing the mind of the Welsh voters. Labour’s support has been stable throughout the campaign, at a level a few points above their performance in 2010. This suggests that Labour should be in good shape to hold most, if not all, of the seats that they won in 2010; however they may gain fewer seats in Wales than they would need in order to be on course for a parliamentary majority in the Commons.

The poll also provides yet further confirmation that Conservative support in Wales remains robust at a level that should help the Tories retain the vast majority of the seats that they won in Wales in 2010.

Two parties have the polls suggest, made some progress during the campaign. One, perhaps surprisingly, is the Liberal Democrats. While their overall support remains well below half of the level they won five year ago, they are up three points on the start of the campaign. If this rise in support is concentrated in their existing seats, then the Lib-Dems may well have a chance of hanging on to all three of them.

The one party that has made some, modest progress is Plaid Cymru. Their support has risen by three points from YouGov’s previous Welsh poll. This is the second successive poll that has placed them – narrowly – ahead of UKIP in third place, and they are up four points on their low-point in a poll early in the campaign. Nonetheless, on uniform swings Plaid would still be struggling to add to its current three seats.

The poll adds further evidence that UKIP’s support is now several points below its highpoint last autumn. It also suggests that the Greens have had a poor campaign. Neither party would now appear to have realistic hopes of winning a parliamentary seat in Wales in 2015.

Still, this is only an opinion poll. As we all know the only poll that really counts is the one that is coming tomorrow. There’s not long to wait now!

Postscript: And for the die-hard Welsh psephology enthusiasts out there (yes, both of you), here are the Ratio Swing seat projections from the poll:


Labour: 28 seats (keeping the 26 seats they won in 2010, and gaining both Cardiff Central from the Liberal Democrats and Cardiff North from the Conservatives);

Conservatives: 8 seats (losing Cardiff North to Labour, but gaining Brecon & Radnor from the Liberal Democrats);

Plaid Cymru: 4 seats (keeping the three seats won in 2010 and gaining Ceredigion);

Liberal Democrats: 0 seats (losing Cardiff Central, Brecon & Radnor, and Ceredigion).


But please remember, folks – a projection is not a prediction.


  • Dave Middleton

    The main change seems to be a shift away from UKIP and toward Plaid Cymru. Can’t help but think that Leanne Wood’s performances in the TV debates has helped in this. That kind of national exposure normally takes years to cultivate and even then Wales tends to get ignored. If I was Carwyn Jones (which I’m not) I’d be be complaining, though I suppose Jim Murphy could say the same in Scotland.

    • Twm Owen

      Always thought Plaid’s percentage of the vote, and I suspect the SNP’s too, suffered from being left out of the debates in 2010.

  • Elfed Williams

    This has in many ways has been a fascinating election and I am sure there will be many more surprises by the end of tomorrow night.
    Both the Cardiff seat look definite gains for Labour, the interesting seats will be Brecon & Radnor, Ceredigion and Ynys Mon. It is so tightly balanced that the Liberals could hold both Brecon and Ceredigion (a good result for them) or loose both. The same for Plaid they look certain to retain their three seats but could gain Ceredigion and possible Ynys Mon (if their YouGov poll is accurate)

  • Jason Morgan

    What I find most interesting, given on the one hand the quite extensive coverage that smaller parties have had – such as Plaid Cymru and UKIP – and on the other hand the fact that neither Labour nor the Conservatives have truly rallied their support considering that a hung parliament is certain, is how little actual change we’ve seen in the Welsh polls – it’s margin of error stuff. It’s almost as if nothing has happened to change people’s minds!

    • Ddirpytnop

      Very true Jasper. Political parties make huge efforts during the course of election campaigns but to minimal effect.

    • Welshguy

      What I find really interesting is that the Tory vote has barely changed since 2010 – despite a huge increase in the UKIP surge. This is very different to the situation in England, where the Tory vote has declined (albeit slightly), and suggests that UKIP supporters in Wales are more likely to come from disaffected Labour voters than disaffected Conservatives (though I suspect these figures mask a fair amount of churn (LD>Lab offsetting Lab>Con offsetting Con>UKIP).

  • J. Jones

    Well….I would say that the Yougov polling wasn’t really that far out… Labour overestimated by 2%, Tories underestimated by 2%, Plaid overestimated by 1%, Ukip underestimated by 1.6%, LibDems overestimated by 1.5% and Greens under by 0.6%.

    You could easily take the view that all the Yougov polls were correct within the margin of error. Except that the errors were consistent in all the polls I suspect and the analysis (seats won) was actually quite significantly wrong.

    Because of the voting system (no Welsh MPs) UKIP’s performance will easily be forgotten but they have actually had a quite significant effect on the outcome. Ynys Mon is now on a knife edge….not particularly because of a great fall in the Labour vote or a great increase in Plaid’s ( they are still lower than their 2005 %) but because a significant enough proportion of the Labour voters turned to UKIP.

    • Roger Scully

      I agree Jon. I’ll write about that on the blog next week some time.

      More generally – the polls were pretty much dead on with Scotland, and our poll was close with Wales. It was just England that went awry…

      • J.Jones

        Close but no cigar. I would expect that Wales with a small voting population and a sample of around 1000 would be less prone to variation but the point is that the error is consistent and it is of the same nature as the error in England. In terms of Labour/Con it is 4% in Wales and 6.6% in England. To be simplistic you could take smaller parties and suggest the same error has occurred, so Plaid/UKIP is a compound error of 2.6%.

        Almost certainly the pollsters will say that the public lied to them. I would say that the pollsters fooled themselves by having roughly the same weighting system and polling methodology.

        It is the same theme that I have voiced many times on your forum Roger; polls are generally erroneous representations of opinion. The same can be said of surveys…only more so. The single biggest factor may be who isn’t asked and who doesn’t answer. If you read blogs in Wales and the limited Welsh press (as I do) then you would swear that Plaid had a significant upsurge in support but that UKIP had a negligible following. Plaid is the 4th party in Wales and its support is in limited areas. UKIP is the party with the 3rd highest following, the voter share is almost identical to England and is higher in Wales than the UK average. But we don’t want to believe it and so we will ignore it. It just doesn’t fit with the Welsh self image.

        Ignore it or not we are looking at the rise of the “right” and stagnation of the “left”. But when it comes to polls does “self image” actually skew the responses? I believe that it does, particularly in certain demographics.

        One danger of uniformity is fear of producing an “outlier”. I hear rumours now that one pollster made an 11th hour poll which replicated the relevant ballot paper for each respondent. They came up with a result very close to the actual outcome…and lost their nerve and didn’t publish.

        • Roger Scully

          The rumour you heard is right, Jon – it was Survation.

          As for our poll – I will post an evaluation piece on it tomorrow. I think we did pretty well, but there was one key mistake (which we won’t make again).

  • J.Jones

    ….and there’s more; if the Yougov polls consistently overstated the voting intentions of Labour and Plaid supporters and consistently understated the voting intentions of Tory and UKIP supporters what other mistaken analyses have we supposed? The EU referendum might well tip over into “out”. The Assembly elections may well see Labour, Plaid and Libdem losses and Tory/UKIP gains beyond what was previously considered.

    Now that the Conservatives have seen the power of English press attacks on the Welsh NHS and Education can they be persuaded to do the same hatchet job next year to help out the Tories in Wales? Would the English press waken up the people in Wales who STILL ignore Assembly elections?

Leave a Reply

Your email address will not be published. Required fields are marked *