May Welsh Political Barometer Poll Published!


The moment an entire nation has been waiting for! The detailed results of the May Welsh Political Barometer poll are available here. And below is the text of a piece that I wrote for the ITV-Wales website. (Or you can read it at the ITV website, here). Enjoy – I think you’ll agree that they are a very interesting set of results! Further detailed analysis to follow (though mostly after the European elections, which will take up plenty of time over the next week or so).

This week sees publication of the third poll conducted by the Welsh Political Barometer – a unique collaboration between ITV Cymru Wales, the Wales Governance Centre at Cardiff University, and the leading polling agency YouGov.

With the European election this Thursday, what did respondents to our poll say that they were likely to do? We asked them two important questions: how likely they were to vote, and which way they would vote if they did turn out.

Only slightly more than half (55) of our survey respondents said, on a 0-10 scale, they were 10/10 ‘definitely’ going to vote. This suggests that many voters are unlikely to participate in the election – as is normal in European elections. What is really interesting, however, is the pattern of which party’s voters are most likely to take part.

Looking at the voting preferences of all survey respondents for the European election, we got the following result (with changes on February’s Barometer poll in brackets):

  • Labour 33% (-6)
  • UKIP 23% (+5)
  • Conservative 16% (-1)
  • Plaid Cymru 15% (+3)
  • Liberal Democrats 7% (no change)
  • Others 8% (+1)

These results are interesting enough in themselves – showing UKIP continuing their rise (with their European vote intention now 10% higher than in December’s inaugural Barometer poll) and the Labour vote slipping rather alarmingly. But they get even more interesting when we look at whose voters are most likely to turn out. Plaid Cymru voters seem rather more certain to take part than supporters of the other main parties. Among those saying that they are definitely going to vote in the European election, we see the following levels of support:

  • Labour 32%
  • UKIP 22%
  • Conservative 16%
  • Plaid Cymru 17%
  • Liberal Democrats 7%
  • Others 5%

If this latter set of figures were replicated on election day, Labour, UKIP and Plaid Cymru would each win one MEP. The final one of Wales’ four seats in the European Parliament is, on these figures, literally a dead heat between Labour and the Conservatives! However, with only 600 Barometer respondents being definite voters, the poll has a margin of error of 4%: this suggests that while Labour and UKIP are both very likely to win one MEP, the final two seats are a three-cornered fight between Labour, the Tories and Plaid Cymru. Who will be left without a seat when the music stops?

What about voting intentions for the next general election and National Assembly election? Here the picture is once again of Labour seeing its support levels decline, while UKIP are the main beneficiaries.

First, Westminster: our poll got the following results for general election vote intention (with changes from the February Barometer poll in brackets):

  • Labour 43% (-4)
  • Conservative 22% (no change)
  • Plaid Cymru 11% (no change)
  • UKIP 13% (+4)
  • Liberal Democrats 7% (no change)
  • Others 4% (no change)

Although Labour is still well in the lead, it is worth noting that 43% is their lowest level of general election support in Wales found by YouGov since before the 2010 general election.

If the changes since 2010 implied by these figures were repeated uniformly across Wales, this would produce the following outcome in terms of seats:

  • Labour: 31 seats (+5)
  • Conservatives: 5 seats (-3)
  • Liberal Democrats: 2 seats (-1)
  • Plaid Cymru: 2 seats (-1)

The seats to change hands would all be won by Labour: Arfon from Plaid Cymru; Carmarthen West & South Pembrokeshire, Vale of Glamorgan, and Cardiff North from the Conservatives; and Cardiff Central from the Liberal Democrats. Labour would also retain all of the 26 seats they won in 2010.

In short, the poll indicates that Labour is still on course to make ground in Wales at the next general election. But the party must be rather concerned at the extent to which its support level has fallen – two years ago Labour was consistently polling above 50% in Wales.

What about the National Assembly? For the constituency vote, the results of our new poll were (with changes from February’s Barometer poll in brackets):

  • Labour 39% (-3)
  • Conservative 20% (-1)
  • Plaid Cymru 19% (no change)
  • Liberal Democrats 8% (-1)
  • UKIP 10% (+5)
  • Others 3% (no change)

The only constituency seat projected to change hands from 2011 on the figures from this poll is Llanelli, being won by Plaid Cymru from Labour.

For the regional list vote, we saw the following results (with changes from the February Barometer poll again indicated):

  • Labour 35% (-4)
  • Conservative 19% (no change)
  • Plaid Cymru 17% (no change)
  • UKIP 14% (+4)
  • Liberal Democrats 7% (-2)
  • Others 8% (+2)

Again, on both votes here the main change overall is Labour losing ground while UKIP advances.

Taking into account both the constituency and list results, this produces the following projected seat outcome for a National Assembly election (with aggregate changes from 2011 indicated in brackets):

  • Labour: 29 (-1); 27 constituency AMs, 2 list AMs
  • Conservative: 12 (-2); 6 constituency AMs, 6 list AMs
  • Plaid Cymru: 10 (-1); 6 constituency AMs, 4 list AMs
  • UKIP 8 (+8); all 8 would be list AMs
  • Liberal Democrats: 1 (-4); 1 constituency AM

These projections indicate the possibility, on the results implied by the current poll, of UKIP becoming a significant force within the National Assembly, and largely doing so at the expense of the Liberal Democrats. Our poll currently projects Kirsty Williams to be the only remaining Lib Dem AM – leader of a party of one.

As with the figures for a general election, these findings show that while Labour are still the party in the strongest position, that position has slipped noticeably. The last time that Labour was as low as 39% on the constituency vote in a poll in Wales was just after the 2010 UK general election, and Labour’s support level has slipped more than 10% since 2012. At the moment, and for the first time since the last Assembly election, Labour is on course slightly to lose ground at the next devolved election, rather than be challenging strongly for an absolute majority in the National Assembly.

Finally, another question asked by this month’s Barometer poll was about how people would vote in a referendum on Britain’s membership of the EU. You might well expect that, given the rising tide in UKIP support, the poll would show a growth in Welsh Euro-scepticism and support for Britain leaving the EU. Actually we found the exact opposite! 44% of respondents said they would in favour of Britain remaining within the EU (3% more than in February’s Barometer poll), while 33% said they would vote to leave the EU – fully 5% down on February’s poll. Perhaps UKIP’s rise is provoking a counter-reaction among others in Wales?


  • Jason

    Interesting as always but one small comment, this suggests that over half the Welsh electorate are certain to vote, and I suppose a very high percentage likely to do so then. That seems very high for a Euro election, no?

    • Roger Scully

      Thanks, Jason.

      Opinion polls always tend to somewhat over-state the proportion of people likely to vote; and on-line polls tend to over-state it a tad more than telephone polls. There are various reasons for this, but they include the fact that the sort of people who agree to complete political polls are also the sort of people more likely to vote.

      I don’t think this poses any fundamental problems with the polls; you simply have to ‘mentally adjust downwards the percentages likely to turn out. For these European elections, I would guess it would be c.40% turnout.

  • J Jones

    I’m speechless………but if I had to say anything, I’d say that the outcome of Thursday’s election will be somewhere between this poll and the last, in other words, very exciting. What I do believe is that Plaid will get their vote out, they have dominated local press and the blogoshere with near hysterical intensity. Labour has made little effort and, in Wales, is suffering from the earlier barage of NHS/Education bad press. Tories look solid and the poll will replicate the EU vote exactly. UKIP dominate the national, UK press and no publicity is bad publicity.

    One small thing bothers me….change in voting intention is one thing but, when attitudes to other questions changes, I’m suspicious that the sample is atypical. More people in favour of Independence for Wales? More in favour of devolution of Income tax raising powers? More people supporting continued membership of the EU? All these are consistent with catching a larger sample of Plaid supporters.

    So which of the last two polls is nearer the truth? I would say between the two…..both extreme samples.

    Is 55% turnout likely in Wales? Yougov respondents are more political than the average I think.

    • Roger Scully

      Thanks, Jon – though I’ve never known you to be speechless before…

      After the dust from the Europeans has settled, I’ll be publishing some analysis of the longer-term trends in party support for the general election and Assembly. Without pre-empting too much, this poll doesn’t look out of line in terms of the trends we see emerging over the last couple of years.

      Of course, you can get samples that are skewed one way or another. But I’d be cautious about interpretations of the findings on Welsh independence (as I’ve said on Twitter to some supporters of the idea): given the vast differences in the question wording, it really doesn’t make much sense to directly compare these findings with those of the BBC/ICM poll.

      One thing, though, that we definitely seem to agree on – turnout will be key on Thursday. Both the final seats could simply come down to who ie most effective at getting their vote out.

  • Christian Schmidt

    I know we should be careful with cross samples, but North Wales Plaid’s worst region for Assembly constituency, list and EU voting expression, and 2nd worst by just 1% for Westminster? Looks odd…

    That the majority of Conservative Westminster/Assembly supporters want out of the EU is also interesting…

    • Roger Scully

      Yes, I suppose the regional cross-breaks do look a bit surprising that way, Christian. But, of course, the poll is weighted to try to achieve representativeness across Wales as a whole, rather than between each region.

      On the EU referendum, I’m actually surprised it is that close among Tories; I would have expected a greater majority against EU membership.

  • J.Jones

    I don’t think that the North Wales poll is particularly surprising Christian, it depends on what “North Wales” encompasses; in this case (I think) Dwyfor Meirionydd, where over 50% vote Plaid, is in Mid and West Wales. Arfon is marginal at the best of times but in term time a poll may well catch students in the Bangor area making Arfon less likely to appear as Plaid. Ynys Mon is again marginal….a Labour MP and Plaid AM but a small population in comparison to Labour areas to the East. The coastal region of Conwy is dominated by Tory voting pensioners and only the inland, rural and sparsely populated areas of Conwy, Denbighshire and Wrexham have significant numbers of Plaid voters.

    It looks about right to me.

    • Roger Scully

      I’m pretty sure that they work on the electoral regions; North Wales does not include Dwyfor Meirionydd. Nonetheless, it was Plaid’s second strongest region on the list vote in 2011, so you would expect them to do a bit better. But it’s probably just some random variation in the sampling.

  • J Jones

    Harry Hayfield comes down in favour of Plaid winning a seat in Wales. Not quite clear if that’s because Labour won’t win 2 or because the Tories will miss out…I suspect the former if he’s using national polling not Wales polling.

Leave a Reply

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