New Poll!

As promised last week, YouGov have produced a new Welsh opinion poll for Elections in Wales. Many thanks to them. Total sample size was 1,012 Welsh adults. Fieldwork was undertaken between 18-22 July 2013.  The survey was carried out online. The figures have been weighted and are representative of all Welsh adults (aged 18+).

Without further ado, here are the figures on voting intention (with changes since the last election in brackets):

UK General Election Voting Intention

Labour: 48% (+12%)

Conservative: 23% (-3%)

Lib-Dems: 8% (-12%)

Plaid Cymru: 9% (-2%)

UKIP: 8% (+6%)

Others: 4%


National Assembly Election Voting Intention (Constituency Vote)

Labour: 46% (+4%)

Conservative: 19% (-6%)

Lib-Dems: 8% (-3%)

Plaid Cymru: 17% (-2%)

UKIP: 6% (+6%)

Others: 3%


National Assembly Election Voting Intention (Regional List Vote)

Labour: 25% (-12%)

Conservative: 12% (-11%)

Lib-Dems: 9% (+1)

Plaid Cymru: 23% (+5%)

UKIP: 16% (+11%)

Others: 14% (+6%)


What would be the implications of these changes for seats (if we assume uniform national swings from 2010/2011)?

The next UK general election would yield the following seat outcomes (change from 2010 in brackets):


Labour: 34 (+8)

Conservative: 4 (-4)

Lib-Dems: 1 (-2)

Plaid Cymru: 1 (-2)


The seats to change hands would be:

Aberconwy (LAB gain from CON)

Arfon (LAB gain from PC)

Brecon and Radnor (CON gain from LD)

Cardiff Central (LAB gain from LD)

Cardiff North (LAB gain from CON)

Carmarthen East and Dinefwr (LAB gain from PC)

Carmarthen West and South Pembs (LAB gain from CON)

Preseli Pembrokeshire (Lab gain from CON)

Vale of Glamorgan (LAB gain from CON)


The Assembly voting intention figures suggest the following outcome:


Constituency Seats:

Labour: 31 (+3)

Conservative: 3 (-3)

Lib-Dems: 1 (no change)

Plaid Cymru: 5 (no change)


The seats to change hands would be:

Aberconwy (LAB gain from CON)

Carmarthen West and South Pembs (LAB gain from CON)

Preseli Pembrokeshire (Lab gain from CON)


Regional List Seats:

Labour: 0

Conservatives: 3 (1 each in North, Mid & West, and South Central)

Lib-Dems: 1 (South West)

Plaid Cymru: 8 (1 each in North and Mid & West; 2 in each of South West, South Central and South East)

UKIP: 8 (1 each in South West and South Central; 2 each in North, Mid & West, and South East)


This produces an overall outcome of:

Labour: 31 seats

Conservative: 6 seats

Lib-Dems: 2 seats

Plaid Cymru: 13 seats

UKIP: 8 seats


These numbers clearly show that Labour continues to be in a very strong position in Wales. The Labour resurgence in Wales that began shortly after the 2010 UK general election continues more-or-less unabated.

For the other parties, the news is not great. None of them are doing very well, although the outlook for the Liberal Democrats is particularly bleak. Plaid Cymru may be particularly disappointed that they do not yet seem to be making any progress in Westminster or Assembly-constituency voting intentions, with the exception of the Assembly list vote – figures about which I am not entirely convinced. (I’ll discuss them in a later post – for now, see my earlier post). Leanne Wood’s leadership appears to have raised spirits within the party, but there’s not much sign of her yet connecting to much of the electorate.

I’ll be back later in the week with more detailed analysis of the poll. But that’s probably enough for you all to chew on right now!


  • Dewi

    Questionable results in my opinion!

    UKIP to win EIGHT regional seats? There support is high- but achieving this would be a heavy task.

    Who are the ‘Others’ who get 24% of the reg vote

    Other seats I think are wrong:
    LAB gain Aberconwy from Con – unlikely
    LAB gain Cam E+D from Plaid – even more unlikely!

    The most interesting result in this poll in my opinion is that as a party (see reg vote). Plaid Cymru and Labour are neck and neck – this is v.interesting and I imagine many in Transport House would be quite worried.

    What’s also interesting is that the above results is very similar to the poll pre NAfW elections – in other words none of the parties have moved since this point – this is v.interesting bearing in mind the cuts made in London.

    • Roger Scully

      Diolch, Dewi.

      On the individual constituency results – this is simply an application of uniform national swing, not taking into account of local factors. You may well prove to be right about those two seats.

      I have my own doubts about the list vote numbers, for reasons I will elaborate on in a later post.

  • Roger Scully

    Apologies: error in original post. The regional list figure for Others should be 14%, not 24% as originally posted. Now corrected.

  • Dafydd Trystan

    So based on this poll, if there was a UNS, the Ynys Mon by-election would return the following result:
    PC 39; Lab 30; Con 23; LD 1; UKIP 6

    Lets see how accurate UNS is – when applied to an ‘interesting’ constituency

  • Save Pehrhos

    And yet they have just had a full slate county council election here on Anglesey and it does not reflect the shifts shown in this opinion poll . The actual swings were:-
    Plaid 32.9% (+11.7%).
    Labour 17.1% (4.4%)
    Tory 5.5% (+0.5%)
    Lib Dem 5.2% (+0.8%)
    UKIP 7.3% (+6.8%)
    Others 29.5% ( -26.7%)

    Not only that we have an Assembly Constituency by-Election and it looks like not only are Plaid going to hold it but with a significant swing to them but that Labour’s vote will be well down.

  • Syd Morgan

    Thank you for this and your interpretation. One election where both national and local variations come together is next May’s for the European parliament. From this new data, can you say how that is trending, please?

  • Christian Schmidt

    Firstly thanks for the blog on Welsh elections, I hope it will be continue to interesting.

    However, on the poll, it very much doubt that outside their own campaign periods opinion polls for the Scottish and Welsh devolved legislatures are of any worth.

    The reason is that Westminster so much dominates the news (esp. here in Wales) that people rarely think about the devolved administrations even when asked about it. Not only do most devolved issues lack salience, I believe in Wales at least Westminster dominance is true even for devolved issues that are important to people – I bet more people in Wales recognised the UK health minister than the Welsh health minister.

    So Welsh (and Scottish) polls outside campaign periods are simply copies of regional Westminster polls with a skew towards the national parties (because the word devolution makes people think about the national interest that will cause some Nats waverers to change view) and an additional skew towards parties other than the three big Westminster ones on the regional vote because many people still look at it as their second (preference) vote.

    And when it comes to the next election, there will be a shock when suddenly in the few months before the election the Holyrood/Senedd polls move (and not in line with Westminster polls).

    Personally I would be much more interested in how people rate the devolved governments (and opposition leader, and how many people know who they are), and how people currently rate the performance of the devolved government against salient issues that are devolved (which I guess would he health and education), and above all how these ratings have changed since just before the last election.

    That I believe would give a much better indicators of how the parties are likely to do at the next devolved election. (At least in Wales, in Scotland there’s the little issue of a referendum to come first which should throw things about a bit.)

    • Gareth Thomas

      Wouldn’t Clwyd West be a Lab gain on these figs very close any way?

        • Gareth Thomas

          Yes you are right but Lab would be in serious contention and much would depend on the local campaign. CW is a rather complex seat ( it certainly isnt a typical valleys or west wales seat) and I have tended to think it follows voting patterns in N W England more than Wales is that fair?

          • Roger Scully

            Quite possibly, Gareth.

            The ‘seat projections’ I did from the poll were simply arithmetical applications of uniform swings. Of course in practice you never get uniform swings, but at the aggregate they are normally a good starting point for estimating the seat implications of polling numbers.

            Detailed predictions for what you think WILL happen in particular seats are, I think, something best left much closer to an election.

  • BoiCymraeg

    Is there something about the poll was worded which encouraged people to put a different party for the list and constituentcy? From a tactical perspective there’s a lot of logic in some voters doing just that – principally, it’s largely pointless to vote Labour on the list in the south. You’d have thought that non-Labour parties (particularly Plaid & the Libs) would be trying to encourage Labour voters to consider them as a “second choice” on the list – and the poll seems to indicate that this is what people are doing, but there’s no evidence that people have realised this in large numbers in the past:

    Year Lab% cons/list
    1999 37 – 35
    2003 40 – 36
    2007 32 – 29
    2011 42 – 36

    So Labour get a small boost in the constituency over the list, but there are still tens of thousands of “wasted” labour list votes. All of a sudden we have polls showing 46 – 26 and 46 – 25, which would seem to indicate a huge movment away from Labour on the lists (with the beneficiaries being exactly who you’d expect: Plaid and UKIP).

    • Roger Scully

      Diolch, Boi. I’ll – I hope – be blogging about this issue next week. But the short answer is, No, YouGov haven’t changed their question wording on this in any way that ought to encourage people to ‘split their ticket’.

      • twm

        My impression has always been that the problem is that you are asking people what they perceive to be the same question 3 times.

        Which way will you vote (westminster)
        Which way will you vote (constituency)
        Which way will you vote (assembly)

        I think there is quite a strong urge with most people who are not diehard supporters of a particular party to put something else down the third time. I suspect if yougov asked the question in a different order with Assembly (constituency) last then you would see Labour get a good regional vote but poorer constituency. All you can take from this is that a large part of the Labour vote is not particularly committed, but if they’re polling 46% what can you expect.

  • Roger Mexico

    I notice from the interim tables that the Other of the Assembly regional VI shows Green at 7%. I haven’t seen the regional breakdowns yet but could this be enough to entitle them to a Regional List AM, depending on how their vote is spread across the regions and the vagaries of d’Hondt? Mid and West Wales might be possible where the additional seats might be less likely to be taken by the big parties or South Central where they scored 5.2% last time.

    • Roger Scully

      I did look at this, Roger. They came very close in a couple of regions, but didn’t quite make it on the projections from this poll. In general, smaller parties like the Greens will be more likely to win list seats when the distribution of the constituency seats is more even: that lowers the effective ‘threshold ‘ that they face to win a list seat. It would therefore help them a lot if the other main parties could challenge Labour more effectively in the constituency races.

  • Jon Jones

    Christian Schmidt could look at the National survey of Wales 2012-2013:


    This survey repeated the Assembly recognisability question of the 2008 Living in Wales survey. Both are very large surveys…14,000 in the 2012 survey. The young and the better educated as well as Welsh speakers know most about the Assembly Government and those who know the most about it are also most likely to approve.

  • Matt Lloyd

    Swings do not always work in that way. It will come down to candidates as the Ynys Mon by election has shown. Taking this into account and if Plaid gets the positive message out I can other gains and certainly a hold in Carms with Adam Price. He’s one of the brightest stars in Welsh politics and people like him will improve the Assembly and Plaid in the next elections.

Leave a Reply

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