The New BBC/ICM Poll

To add to a bumper polling week in Wales comes news that BBC-Wales have run a new poll with ICM. Although the main focus of the poll was not directly on matters electoral, they have included a question on voting intention for the forthcoming general election. The poll, conducted by telephone, was actually carried out slightly before the new Barometer poll that was published on Monday, but the results were only been released last night.

The voting intention numbers (with changes from the last ICM poll for BBC-Wales, carried out in September, in brackets) are:

Labour: 38% (no change)

Conservative: 21% (-2)

Liberal Democrats: 7% (no change)

Plaid Cymru: 12% (-1)

UKIP: 13% (-1)

Greens: 6% (+4)

Others: 2% (no change)


So no major changes from the previous poll, except for the significant boost in support for the Greens. The numbers are all also within the ‘margin of error’ of those produced by YouGov for the Welsh Political Barometer poll published earlier this week. Both polls showed the Greens rising by several points, raising our confidence that this is a genuine increase and not an outlier.

Using the standard Uniform National Swing assumption to project from the raw polling numbers to seat outcomes, on the figures here only two seats would change hands: Labour gaining both Cardiff Central and Cardiff North. That leaves us with the overall outcome of:

Labour: 28 seats (+2 on 2010)

Conservatives: 7 seats (-1)

Liberal Democrats: 2 seats (-1)

Plaid Cymru: 3 seats (no change)

The only difference here with the Barometer poll is that ICM’s slightly lower score for the Tories, and marginally higher one for the Lib-Dems, leads to Brecon and Radnor being projected as a marginal Lib-Dem hold rather than a narrow Conservative gain. Using the alternative Ratio Swing assumption that I have periodically discussed here on the blog, we get these seat numbers:

Labour: 30 seats (+4; gaining Cardiff Central, Cardiff North, Carmarthen West & South Pembrokeshire, and Vale of Glamorgan)

Conservatives: 6 seats (-2; losing Cardiff North, Carmarthen West & South Pembrokeshire, and Vale of Glamorgan to Labour, but gaining Brecon & Radnor from the Liberal Democrats)

Liberal Democrats: 0 seats (-3)

Plaid Cymru: 4 seats (+1; gaining Ceredigion from the Liberal Democrats)


In terms of how we interpret the findings of this poll for the individual parties, I’d say this:

Labour: Rather like the Barometer poll, I think this is a moderately encouraging one for Labour. Given the erosion of Labour support throughout 2014, to see a second poll this week were poll in which Labour are holding their ground, and at a level slightly above that which they secured in 2010, is at least modestly good news for them.

Conservatives: One rather surprising – to many people, at least – feature of Welsh politics since 2010 has been how well the Tories’ support levels have held up. Throughout 2014 they were steady at a level only slightly below the vote share won by the Conservatives in the 2010 general election. In that context, this is slightly disappointing poll for them: it is only the second poll since the last general election, and the first in over a year, in which the Tories have scored below at least 22%. On these numbers, and assuming uniform national swings, they don’t suffer major seat losses. But some of the seats that this poll projects the Tories to hold on uniform swings, like the Vale of Glamorgan and Carmarthen West & South Pembrokeshire, are on these numbers starting to get very marginal. And on ratio swings, those two seats actually fall to Labour.

Liberal Democrats: This is yet another in a very long line of disappointing polls for the Lib-Dems. There are still no signs of the party recovering much, if any, any of the substantial support they have lost since May 2010. On these numbers they have lost nearly two-thirds of their support since the last general election. (And it is perhaps worth mentioning that this is after taking into account the adjustments ICM make to the raw numbers, which are markedly favourable to the Lib-Dems. On the unadjusted numbers they are down at 5%: a loss of slightly over three-quarters of their 2010 support). At least in this poll, unlike with the Barometer one, they are still in fifth place rather than sixth…

Plaid Cymru: This is very much a so-so poll for Plaid Cymru. They are up a little on their 2010 vote share, but no more than that. While on uniform swings Plaid would hold their three current seats, it does not give them a sufficient increase in support to be projected to win any of their target seats on uniform swings, although ratio swing does project them to gain Ceredigion.

UKIP: A year ago this would have been a strikingly good performance in a Welsh poll for UKIP. But given their momentum during 2014, this may actually be slightly disappointing for them. Although changes since the last poll are small and well within the ‘margin of error’, this result is consistent with the picture from many of the GB-wide polls in January, suggesting that UKIP’s momentum may currently have stalled. But there is nothing to suggest that UKIP have yet gone decisively into reverse.

Greens: An interesting feature of the political year so far has been a relatively strong performance by the Greens in the opinion polls, coupled with an apparent surge in their party membership. This poll, as with this week’s Barometer poll, very much fits in with that picture. There is nothing to indicate that the Greens are likely to win a parliamentary seat at the general election, but they are attracting notably greater support now in Wales, just as they have begun to do in England.

The Other Barometer Results

As well as probing voting intentions for the general election, our latest Welsh Political Barometer has continued to ask about how people intend to vote in several other elections and referendums that may, or will, be facing Wales in the near future.

First, we asked people about their voting intentions for the National Assembly. With the elections for this body following on exactly one year after the general election, where do the parties stand right now?

For the constituency vote, the results of our new poll were (with changes from our previous poll, in early December in brackets):

  • Labour 34% (-1%)
  • Conservative 21% (-1%)
  • Plaid Cymru 18% (-1%)
  • UKIP 13% (+1%)
  • Liberal Democrats 7% (+1%)
  • Greens 6% (+1%)
  • Others 1% (no change)

Clearly, very little has changed since our last poll, with all the parties seeing changes in their support levels but by amounts that are well within the ‘margin of error’.

On these figures, and assuming uniform national swings across Wales, only two constituency seats would change hands from the results in the last Assembly election in May 2011: the Liberal Democrats would gain Cardiff Central from Labour, while Labour would also lose Llanelli to Plaid Cymru.

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

  • Labour 32% (+1%)
  • Conservative 20% (no change)
  • Plaid Cymru 15% (-4%)
  • UKIP 16% (+1%)
  • Liberal Democrats 8% (+2%)
  • Greens 8% (+1%)
  • Others 2% (no change)

Here there is a little more change evident. Plaid Cymru see a quite large fall in their regional list vote (after having had a significant rise in our previous poll), while several other parties edge upwards by smaller amounts.

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: 28 (-2): 26 constituency AMs, 2 list AMs
  • Conservative: 10 (-4); 6 constituency AMs, 4 list AMs
  • Plaid Cymru: 9 (-2); 6 constituency AMs, 3 list AMs
  • UKIP: 8 (+8): all list AMs
  • Greens: 3 (+3): a list AMs
  • Liberal Democrats: 2 (-3); 2 constituency AMs

So, for the second time in a row, our Barometer poll projects an outcome which would mean six different parties being represented in the National Assembly. This time, though, the relatively strong performance of the Greens in our new poll actually relegates the Lib-Dems to sixth place in terms of projected seats. The Greens are projected to win list seats in North Wales, Mid and West Wales, and South Wales Central.

We should note, though, that with so many different parties in the mix, several of the list seats are projected to be won by tiny margins. In South Wales West, the final list seat was allocated to UKIP over the Conservatives in a calculation that went to the third decimal place – equivalent to about two votes! As the polls bob up and down in the period leading up to the next Assembly election we should expect to see the projected regional list seat outcomes showing quite a lot of turbulence.

As well as asking about election voting intentions, however, our Barometer polls have continued to ask people in Wales how they would vote in the two potential referendums we may be facing in the not-too-distant future. One of these is on the UK’s membership of the EU. In our latest poll, 44% of respondents in Wales said that they would vote for Britain to remain in the EU and 36% said that they would vote for Britain to leave, with the remaining 20% indicating either they didn’t know or that they would not vote in such a referendum.

This 8% margin between those who want to stay in the EU and those who want to leave is quite narrow. It is, though, the largest gap that we have seen for some months in Wales. The table below shows the results of all polls on this question conducted by YouGov in Wales since the launch of the Welsh Political Barometer in December 2013. The ‘remain’ camp has led in all of them bar the first, but that lead is normally very slender. If we do see an EU referendum at some time in the next few years it is far from clear which side Wales would end up supporting.


Poll% Remain% Leave% Don’t Know / Wouldn’t Vote% ‘remain’ Lead
December 2013384022-2
February 201444332311
June 20144138223
July 20144136245
September 20144337206
December 20144239193
January 20154436208


Another question that has been consistently run in Barometer polls concerns a possible referendum on handing some powers over income tax to the National Assembly. Our latest poll finds 37% of respondents indicating that they would vote in favour of the Assembly gaining some powers to raise or lower income tax levels in Wales but 39% suggesting that they would vote against; the remaining 24% were either unsure or indicated that they would not vote.

The table below again shows the various polls conducted by YouGov on this question since the launch of the Barometer poll just over a year ago. We can see here that, as with the matter of Britain’s EU membership, opinion in Wales has consistently been fairly evenly divided on this question. But here also we find that one side has consistently led, if mostly by fairly small margins. Not one poll conducted by YouGov has yet shown those in favour of devolving income tax in the lead. It seems unlikely that Welsh political leaders will be eager to hold an income tax referendum any time soon, with the balance of public opinion being so marginal, and actually leaning towards opposition.


Poll% Yes% No% Don’t Know / Wouldn’t Vote% ‘No’ Lead
December 20133538263
February 201431422811
May 20143339286
June 20143441257
July 201432422610
September 20143839241
December 20143738251
January 20153739242



One Hundred Days to Go

Our new Welsh Political Barometer poll is published today – as we hit the one hundred days to go point in the general election campaign. With the election on May 7th drawing ever closer, where do each of the parties stand here in Wales?

Here is what the Barometer poll found regarding general election support for each of the main parties (with changes from our last poll, conducted in early December, in brackets):

  • Labour 37% (+1%)
  • Conservative 23% (no change)
  • UKIP 16% (-2%)
  • Plaid Cymru 10% (-1)
  • Greens 8% (+3)
  • Liberal Democrats 6% (+1%)
  • Others 1% (-1)

So what does that mean in terms of who represents us in parliament? Well, if the changes since the last general election implied by these figures were repeated uniformly across Wales, we would get the following outcome in terms of seats:

  • Labour: 28 seats (+2)
  • Conservatives: 8 seats (no change)
  • Plaid Cymru: 3 seats (no change)
  • Liberal Democrats: 1 seat (-2)

Only three seats, of the forty in Wales, would change hands: Cardiff North and Cardiff Central would both won by Labour (from the Conservatives and Liberal Democrats respectively); Brecon & Radnor would be narrowly gained by the Conservatives from the Liberal Democrats.

There’s an interesting paradox here. Party politics in the UK currently seems more uncertain and turbulent than for a long time –maybe more than it has ever been. We’ve seen big recent movements in the support levels of several parties, including the rise in Wales of UKIP and now a notable increase for the Greens. Yet, at the moment, a direct projection of poll findings produces only very small changes in terms of who wins which seats. We could be on course for an election in which lots of things change, but the basic fundamentals of which parties represent us in parliament are hardly disturbed.

Overall, what does this poll tell us about the prospects for each party, as we enter the final hundred days of campaigning?

For Labour, this poll is at least modestly encouraging. A persistent feature of the opinion polls in Wales during 2014 was the decline of Labour support: they finished the year well below the point that they started it. Our new poll seems to suggest that Labour have stopped, and may even have begun to reverse, this erosion in their support. This poll doesn’t put Labour on course to gain as many seats as they would need to help secure a parliamentary majority for Ed Miliband. But it does place them slightly ahead of where they were in 2010, and indicates that Labour are currently on track to make at least some ground in May.

The Conservatives have surprised many observers with the robustness of their support levels since 2010, holding steady at a level only slightly below the vote share they won in the last general election. Here is yet another poll that supports this pattern. Although the poll projects the Tories to lose the ultra-marginal Cardiff North, on these figures they ought to retain all their other Welsh seats. And it puts them in with a very good chance of taking Brecon & Radnor from the Liberal Democrats.

For the Tories’ coalition partners, the Liberal Democrats this is yet another in a long series of disappointing polls. They have at least managed a very small up-tick in their support levels. But this poll still indicates that they have lost more than two-thirds of the support that they gained in Wales in 2010, and with the rise of the Greens they are now in sixth place! There seems to be no substantial improvement in their position in sight. The best the party can do for now, it seems, is to try to hang on to the three seats they currently hold. But even that will now be very difficult.

For Plaid Cymru this poll will be at least a little disappointing, putting them as it does a little below their vote share in 2010. One piece of slightly better news for them comes from a question where YouGov asked respondents how certain they were to vote in the election: Plaid supporters were the most likely to indicate that they were absolutely certain to vote. This poll suggests that Plaid may well be able to hold their existing seats. But they are nowhere near threatening the sort of breakthrough that their sister-party is doing in Scotland.

For UKIP, this poll may also be mildly disappointing. Perhaps the big story in Welsh politics in 2014 was the UKIP breakthrough. Our latest poll indicates, as have many of the recent Britain-wide ones, that UKIP’s forward momentum may well have been checked, at least for the moment. Nonetheless, UKIP have not yet gone into a clear reverse. They are currently on course to get lots of votes in Wales in May. But the party still remain up against it to convert this significant public support into a win in any specific constituency.

Finally, what about the Greens? This poll shows them making significant ground in Wales, relegating the Liberal Democrats to sixth place (as they did in last May’s European elections). As with UKIP, however, it is currently very difficult to see the Greens converting such support into actually winning a seat anywhere. But the more proportional voting system used for devolved elections makes a Green presence in the National Assembly after 2016 look increasingly likely.

First results from the new Welsh Political Barometer Poll

Results from the latest Welsh Political Barometer poll will be released this week. Later on today the first of the voting intention figures should be released by our partners at ITV Wales. I will, of course, be making full results available, along with some analysis, at this website.

The first figures released, however, relate to a couple of questions added to the latest poll by ITV Wales. These probe attitudes to, and experiences of, the NHS in Wales. The results thus tie in very well with the Ashcroft poll that I discussed here last week. Helpfully, YouGov also asked very similar questions to those in our Barometer poll to a sample of 1,344 English respondents, enabling us to get direct comparisons of attitudes in the two nations – and a much more up-to-date comparison than was available from the Ashcroft poll, fieldwork for which was carried out in November last year.

The first question asked about the NHS was: ‘How confident, if at all, are you that the NHS in Wales will provide you with a high standard of healthcare when you need it?’. Responses in Wales (and to the equivalent question in England) were:


Very Confident1216
Fairly Confident4154
Not Very Confident2919
Not at all Confident148
Don’t Know44


As we can see, the majority of people in Wales did select one of the two ‘confident’ options. But, as we saw consistently in the Ashcroft poll, overall Wales’ NHS fares somewhat less positively than the English ones.

The second question asked in our Barometer poll inquired about people’s own experiences of the NHS: ‘How satisfied or dissatisfied would you say you are with the overall experience you have had when being treated by the NHS in Wales?’

Very Satisfied2631
Fairly Satisfied4549
Not Very Satisfied1513
Not at all Satisfied54
Don’t Know83


Here we see very nearly three-quarters of Welsh respondents express satisfaction with the NHS in Wales as they have experienced it. But we once again see ratings of the Welsh NHS lagging behind those of the English service, if only narrowly.

In short, although our data about NHS attitudes is rather more up-to-date than that released by Lord Ashcroft, it comes to a very similar conclusion (and one also supported, as I mentioned last week, by a question in the latest ICM poll as well). There is a consistent gap in attitudes to the performance of the NHS between England and Wales, with Welsh respondents consistently less satisfied (with Lord Ashcroft’s evidence having suggested that the Scots are even more satisfied than the English with their health service). The gaps are not, in the main, huge. But they appear to be pretty consistent – across multiple questions, and also across surveys conducted by different survey companies.

Lord Ashcroft’s Health Poll

At the end of last week I discussed the voting intention figures from the recently-reported (though conducted in November) Lord Ashcroft ‘mega-poll’. Interesting though those findings doubtless were to regular readers of the blog, the main purpose of the poll was not simply to test the waters regarding voting intention. Rather, the bulk of the poll was devoted to exploring public attitudes to, and experiences of, the NHS. Large samples of voters in Scotland, Wales and across the regions of England were asked about the health service through a range of different questions.

I’m not going to attempt to summarise all the findings. The tables of results published by Lord Ashcroft come to very nearly 850 pages, so it would take a lot of summarising! Nor am I going to make any personal contribution to on-going arguments about the management of health across the UK – that would be straying a long way beyond my professional competence. Here I simply want to point out what I think are three main general points to be drawn from the morass of detail in Lord Ashcroft’s poll. But while doing so, I very much also want to encourage all of you to take some time to look through the detailed poll results for yourselves.

First point. The manner in which the poll was conducted and reported (which decent-sized samples in Wales and Scotland, plus a very large one across England) allows us to make fairly robust comparisons between attitudes in the different nations and regions of Britain. And when we do compare attitudes, the picture that emerges is not flattering to the NHS in Wales. The NHS is a bigger concern to people in Wales than elsewhere in Britain, while direct evaluations of the service offer poorer ratings here in Wales than the rest of the country:

  • When asked to nominate the ‘most important issue facing Britain’, the NHS comes fourth in Wales. But it is chosen by a higher percentage of survey respondents in Wales (14%) than in any other region/nation in Britain. When respondents are asked for the most important three issues, the NHS is again chosen by more people in Wales than anywhere else;
  • When asked about the ‘most important issue facing you and your family’, Wales again has more people nominating the NHS than anywhere else in Britain;
  • When people are asked whether they would ‘recommend’ any NHS care they had received to friends or family, respondents in Wales gave the worst ratings of any nation or region in Britain;
  • Respondents were asked to rate, on a 0-10 scale (with 10 being the best rating) the quality of their personal experiences of the NHS. Scottish respondents gave an average rating of 7.53, English ones 7.38, and Welsh respondents only 7.23;
  • When asked to make a more general rating of the NHS on a similar 0-10 scale, Scottish respondents gave an average rating of 7.01, English ones 6.40, and Welsh respondents only 6.06. Some 16% of Welsh respondents rated the NHS at between 0 and 3, compared to 8% of English ones and only 5% of Scots;
  • When asked whether the NHS had improved or declined over both the past 5 years and the past 25 years, Welsh respondents were less likely than both English and Scottish ones to detect improvements and more likely to perceive declines;
  • Similarly, when asked to look forward to the next five years, Welsh respondents were (albeit only slightly) the least likely to expect improvements, and the most likely to expect the NHS to get worse.

Quite what implications one draws from this for debates about how the NHS should be managed in Wales is for others to consider. But for all of us in Wales, these ratings must be rather disheartening. The findings also suggest why the Welsh NHS is likely to remain an issue in a UK general election where control of the NHS here is not obviously at stake.

My second point relates to another set of questions in the Ashcroft poll, where it asked respondents how important they thought the NHS was to the different parties. Again, this question used a 0-10 scale. Here are the mean average ratings given for each of the main parties by the Welsh sub-sample:

Labour: 6.6
Conservatives: 5.1
Liberal Democrats: 5.4
Plaid Cymru: 5.9
UKIP: 4.6

These results are clearly rather more heartening for the Labour party. One can readily see that, whatever difficulties might face the NHS in Wales, Labour retains a significant advantage over the other parties, and particularly the Conservatives, in being seen as caring more about the service. The NHS is, though, not absolutely and inevitably owned by the Labour party as a political issue: in Scotland, the SNP has a mean average rating of 7.0 on the NHS (with Labour there rated at 6.5).

A final point. One very disappointing aspect of the poll was the almost total lack of recognition of devolution, and the fact that the NHS is managed separately in Scotland and Wales from England. This comes through quite explicitly in a small number of questions: most obviously in one of the final questions, Q24 (see pages 701 and following in Lord Ashcroft’s published details), which talks about “the government” having “introduced a number of NHS reforms”. There is no acknowledgement that the relevant government for managing the NHS differs across Britain, nor that the reforms that have been introduced in England might not be relevant to survey respondents in Scotland and Wales. Similarly, earlier on when there is a question regarding which people survey respondents would ‘trust to tell the truth on the NHS’, Jeremy Hunt and Andy Burnham – the Secretary of State for Health and his Shadow at Westminster – are listed; no attempt is made to identity equivalent relevant figures in Holyrood or Cardiff Bay, nor to suggest that Hunt and Burnham might only be directly relevant to debates over the English NHS. I’ve greatly admired many of the contributions Lord Ashcroft has made to political polling in the UK in recent times, but I think this complete ignoring of devolution – in probably the highest-profile devolved policy area for both Scotland and Wales – was both unpardonably sloppy, and a great missed opportunity. It is amazing to me that even sophisticated political operators like Lord Ashcroft seem sometimes to ‘forget’ that devolution exists.


Postscript (21/01/15)Tables for this month’s ICM/Guardian poll have been released. This poll has a standard sample size (just over 1000), and a Welsh sub-sample of only just over 50, so it is distinctly more hazardous doing cross-regional comparisons. However, the poll includes a small number of questions that touch on the NHS, most particularly this one:

‘Which of the following do you think best characterises the current condition of the NHS?’

with the following response options:

  • It is working well, with few problems that are not being dealt with
  • It is feeling the financial squeeze, but generally functioning satisfactorily
  • It is struggling, and not delivering services it should in some places
  • It is not coping with the demands being made of it, and is in danger of ceasing to exist in the form that Britain has known it
  • Don’t Know

The pattern of responses in the Welsh sub-sample was:

Working Well: 10%

Generally Satisfactory: 12%

Struggling: 28%

Not Coping: 50%


While comparisons between such relatively small sub-samples should be made with considerable caution, it may be worthy of note that the proportion choosing the ‘Not Coping’ option in Wales was substantially higher than anywhere else in Britain. In line with the Ashcroft results it was lowest in Scotland (26%, barely half the level in Wales), and also lower throughout England. Indeed, in no other ‘region’ did even 40% choose the ‘Not Coping’ option.

Better Late Than Never

Apologies for being a bit slow on this, but I was away for part of this week (supposedly visiting Dublin; in reality, spending much of my time in Dublin Airport). Anyway, as many of you will have noticed, this week Lord Ashcroft released another big chunk of polling data. This time it concerned a large sample poll he conducted in November last year that was primarily devoted to exploring attitudes towards the NHS.

The poll was based on a very large sample – approximately 20,000 respondents. (Unlike with his constituency polls, this one from Lord Ashcroft was conducted online.) This means that, although there was no over-sampling for Wales, we still have about 1000 respondents in the Welsh sub-sample. Results reported for Wales, and all the other nations/regions, appear to have been weighted for socio-demographic representativeness within those territories (although I haven’t yet been able to trace the exact details of how this has been done).

One of the questions was voting intention for the general election: as is typical with Lord Ashcroft’s polls, a number of different figures are reported. However, what seems to be the main ‘headline’ figure is one that adjusts for likelihood to vote. The Welsh figures here were:

Labour: 37%
Conservatives: 25%
Liberal Democrats: 3%
Plaid Cymru: 12%
UKIP: 17%
Greens: 4%
Others: 1%

Although these figures are now a couple of months out of date, it is still nice to have another measure of public attitudes from a different source. The poll also asked about how people had voted in the 2010 general election. This may give us some guide as to how representative the sample was in party terms. However, we must remember vote recall over such a long time period can be faulty: in particular, if a party has slipped in popularity then some people (either deliberately or not) seem to ‘forget’ having voted for them in the past. So if vote recall figures don’t match the last election this may be because the sample is skewed between the parties, or it may mean that some people are misremembering. Anyway, this is what the vote recall measure for 2010 gave (with differences from the actual 2010 result in Wales in brackets):

Labour: 38% (+1.8%)
Conservatives: 31% (+4.9%)
Liberal Democrats: 18% (-2.1%)
Plaid Cymru: 10% (-1.3%)
UKIP: 3% (+0.6%)

In short, the Ashcroft sample seems to have Labour 2010 support slightly over-stated, and even more so for the Tories, but is looking a little light on Liberal Democrats and Plaid supporters. We should perhaps bear that in mind when looking at the vote intention figures. In particular that may help explain the truly horrendous figure for the Lib-Dems, the worst they have recorded in any poll of which I am aware this century. But there may be elements of misremembering, as well as un-representativeness of the sample, at play here.

What would these vote intention figures mean in terms of seats? On the standard uniform national swing assumption, the poll projects the following seat outcomes:

Labour: 28 seats (gaining Cardiff North and Cardiff Central)
Conservatives: 8 seats (losing Cardiff North but gaining Brecon & Radnor)
Liberal Democrats: 1 seat (losing Cardiff Central and Brecon & Radnor)
Plaid Cymru: 3 seats (no change)

On the Ratio Swing assumption I have mentioned on this blog previously, the results are only slightly different:

Labour: 28 seats (gaining Cardiff North and Cardiff Central)
Conservatives: 8 seats (losing Cardiff North but gaining Brecon & Radnor)
Liberal Democrats: 0 seats (losing Cardiff Central, Brecon & Radnor and Ceredigion)
Plaid Cymru: 4 seats (gaining Ceredigion)

Finally, as this poll was conducted in 2014, even if not reported then, it seems only fair to up-date the polling averages 2014 that I discussed recently in my review of the electoral year. The following is based on eleven polls conducted during 2014, three by phone and eight on-line:

Labour: 40.5% (varying from a high of 47% in the year’s first Barometer poll to a low of 36% in the year’s final Barometer poll)
Conservatives: 23.5% (remaining between 22% and 25% in all polls)
Liberal Democrats: 6.1% (never reaching double figures in any poll)
Plaid Cymru: 12.0% (varying between 11% and 15% in all polls)
UKIP: 13.4% (varying from a low of 7% early in the year to 18% by the end)
Others: 4.7%

I’ll be back soon with some discussion of the other findings in this poll. There should also be two new Welsh polls to discuss at some time in the next week or so. Life just doesn’t get much better than this…

Is Lord Ashcroft a Closet Welsh Nationalist?

In my previous blog post, I looked at the four constituency polls conducted in Wales by Lord Ashcroft during 2014. One slightly puzzling feature of them that I pointed to was the relatively strong performance of Plaid Cymru in all four seats. Given that none of the four constituencies polled is conceivably a Plaid target seat for 2015, and given also Plaid’s rather limited progress in the national polls, it was rather surprising to observe Plaid’s vote share in all four constituency polls up by several points on Plaid’s actual 2010 election performance in those places.

To what might we attribute this slightly puzzling result? It seems very unlikely, as I said, that Plaid can have been targeting substantial resources at any of these seats. It is possible, perhaps even likely, that Plaid’s vote in these seats will get squeezed as election time approaches, and voters in those respective constituencies face up to the question of who can conceivably win in their locality. (Although, to be fair, Lord Ashcroft has sought to address this issue by asking people first a ‘standard’ voting intention question and then one related specifically to the particular constituency they are in). We don’t have equivalent constituency polls from this point in the electoral cycle prior to the May 2010 election: perhaps Plaid at that point would have been in a promising position in these seats, only to see votes drain away later.

(Of course, there are plenty of keyboard warriors/fantasists, lurking on Twitter and in the below-the-line comments sections of many publications, who will insist that most poll results are just arranged to suit the interests of those paying for them. To which all I can say is that, if Lord Ashcroft is a closet Welsh nationalist, he has been keeping it very well hidden…).

More seriously, there is one other factor that may at least partially explain why Plaid are showing up better in Lord Ashcroft’s polls than one would expect from the overall picture of the national polls. Below is a table based on dividing the ten publicly-reported national polls on general election voting intention in Wales in 2014 into two groups: those seven conducted on-line (all by YouGov) and the three conducted by telephone. The table then lists the average general election vote intention figure for each party in the two types of poll:

Online Polls4023.
Phone Polls41.323.77.314.011.3

As we can see, for most parties here the differences between the two types of poll are small. They are also quite explicable. Labour does a bit better in the phone polls than the online ones because Labour’s support ebbed during 2014 and two of the three phone polls were conducted in the first five months of the year, when Labour was doing a bit better. Very similar factors can account for the small differences between the two types of poll for Lib-Dem voting intentions. Meanwhile, the Tories – whose poll rating was pretty stable throughout 2014 – see little difference between the two types of poll. UKIP were doing better by the end of the year than at the beginning, and thus score better in the online poll average.

The puzzle remains Plaid Cymru, for whom there is the largest gap between phone and online. There was little or no obvious trend in their general election voting intention throughout 2014 within YouGov’s online polls. Moreover, all three of the telephone polls put Plaid at a higher level of support than did any of the online ones. In short, there seems to be a genuine difference between the two methods in the support levels they are showing for Plaid Cymru. (Albeit this is a difference of relatively modest size: it’s not like we are talking about phone polls showing Plaid’s support levels double those of the online polls).

These differences may well, in turn, go at least some way to explaining why Plaid have been performing quite well in Lord Ashcroft’s constituency polls. These polls are all conducted by phone – they have to be, as no online company (not even one as well-established as YouGov) have sufficient respondents in their panel to produce 1000-sample polls in individual constituencies. If phone polls are, in general, producing systematically better figures for Plaid than on-line ones, then we shouldn’t perhaps be quite so surprised at Lord Ashcroft’s constituency results for the party.

At least two further questions remain, though I’m not sure I have adequate answers to either of them. First, why do we see these differences in the two types of poll? There may be some social factors behind it: one colleague with whom I raised this issue pointed to the often poor broadband connections in much of rural west Wales as being potentially a factor leading online polls to potentially under-state Plaid support slightly. Or there may be some more directly political factors at play (though I’m not quite sure what there would be); in this regard, it is perhaps worthy of note that the best poll result for the SNP since the September independence referendum (that by Ipsos MORI in October) was also a phone poll.

Second, the other obvious question is which one is right?! It is very difficult to say. ICM and YouGov are both highly (and quite deservedly) respected companies internationally with an impressive record for measuring voting preferences accurately in elections and other events. Perhaps we will see these slight differences erode as we get nearer the election – just as the pollsters tended to converge as the Scottish independence referendum approached. But the possibility that one polling method is tending to systematically under- or over-state the support of one significant party in Wales is something that is at least worth keeping an eye on, I think.

2014 End of Year Review: A Postscript

It has been helpfully pointed out to me that, as well as the national polls which I discussed in my previous post, we have also had four individual constituency polls in Wales from Lord Ashcroft during 2014. I was aware of that, of course, and commented on each of the four polls at the time. (See here, here, here and here).

But, while we are in a retrospective mood on the blog, it is perhaps worth putting the four results together. So, please see the table below: this presents the figures from all four constituency polls for the constituency-specific question used by Lord Ashcroft. Changes from last the 2010 general election result are in brackets.

Constituency (month of fieldwork)LabourConsLibDemsPlaidUKIP
Cardiff North (April)40 (+3)33 (-5)10 (-8)7 (+4)8 (+6)
Cardiff Central (September)36 (+7)17 (-5)24 (-17)9 (+6)9 (+7)
Brecon & Radnor (November)15 (+5)27 (-9)31 (-15)8 (+6)17 (+15)
Carmarthen West & South Pembs (December)29 (-4)33 (-8)4 (-12)16 +6)14 (+11)

We probably shouldn’t expect to see too many similar patterns across these four polls – after all, we are hearing much in the media about the very different patterns that are expected to occur across different constituencies. And the four constituencies polled by Lord Ashcroft in Wales are very different: one ultra-marginal urban Conservative-Labour seat; one Liberal Democrat-held urban seat where they are under strong pressure from Labour; a Liberal Democrat-held rural seat where they face a potential threat from the Conservatives; and a mainly rural Conservative-held seat that Labour need to regain if they are to be on course for an overall majority.

So, do we see any obvious or interesting patterns in the results here? Let’s review things party-by-party:

Labour’s vote share is up, though not by an enormous extent, in three of the four seats. However, their improvement in Brecon and Radnor is of little conceivable use to them in a seat they are most unlikely to win; it must be of some concern that the most recent poll shows them actually losing ground on 2010 in Carmarthen West & South Pembrokeshire.

The Conservatives do somewhat worse than we might expect: their performance in all four polls is down several points on 2010. This is despite national opinion polls that have consistently shown them performing robustly at a support level only a little way behind the vote share won in 2010. This must be a matter of some concern to the Welsh Tories.

The Liberal Democrats continue to perform weakly across both local and national polls. The poll in Cardiff Central put them on course to lose by a fair margin in a seat they have held since 2005. However, that in Brecon and Radnor did suggest that they have at least a chance of holding onto the seat there – a rare piece of good news for the party these days.

Unsurprisingly, the four polls all showed UKIP advancing: this was wholly to be expected given their strong showing in the national polls during 2014. But there are no signs, at least in any of these seats, of them making a sufficiently major breakthrough as to render them likely to actually win a seat.

The most puzzling aspect of these four polls, to my mind at least, is the performance of Plaid Cymru. None of the four constituencies polled is remotely a Plaid target seat for 2015 (although one might imagine that Carmarthen West & South Pembrokeshire will be a target for Plaid in the 2016 National Assembly election.) It is very difficult to believe that Plaid can have been pushing significant resources at any of these seats. Yet we see Plaid’s support levels up by several points in all four seats – both on the constituency-specific question shown here, and on the ‘standard’ voting intention question which was also asked in all four polls. This is puzzling because, unlike UKIP, Plaid was making at best only very limited progress in the national polls at the time that all these constituency polls were being conducted. I’ll return in a later Blog post to try to work out why we might be seeing such puzzling results for Plaid.

Elections in Wales, 2014: End of Year Review, Part 2

For this second part of my review of the electoral year in Wales, I turn to the state of the parties as revealed by the opinion polls. With the establishment late in 2013 of the Welsh Political Barometer, we now have more regular opinion polls conducted in Wales than has typically been the case in the past. Along with a few other polls reported during 2014, they have given us a fairly detailed picture on the state of the parties.

The three tables below summarise the results of those polls during 2014. They give, respectively, voting intentions for the UK general election, the Assembly constituency vote, and the Assembly list vote:

UK General Election Vote Intention

Lord Aschroft, Jan 2014 (published March)4024615133
ITV-Wales/WGC/YouGov, Feb 2014472271194
BBC Wales/ICM, Feb 2014422491474
FoES/YouGov, April 20144524711103
ITV-Wales/WGC/YouGov, May 20144322711135
WGC/YouGov, June 20143924512156
ITV-Wales/WGC/YouGov, June-July 20144125511145
ITV-Wales/WGC/YouGov, September 20143823611176
BBC-Wales/ICM, September 20143823713144
ITV-Wales/WGC/YouGov, December 20143623511187
2014 AVERAGE40.923.46.412.013.04.7

NAW Constituency Vote

ITV-Wales/WGC/YouGov, Feb 2014422191953
FoES/YouGov, April 2014412182072
ITV-Wales/WGC/YouGov, May 20143920819103
BBC-Wales/ICM, May 20143619524134
WGC/YouGov, June 20143622519124
ITV-Wales/WGC/YouGov, June-July 20143721520134
ITV-Wales/WGC/YouGov, September 20143621619126
ITV-Wales/WGC/YouGov, December 20143522619126
2014 AVERAGE37.820.96.519.910.54.0

NAW Regional List Vote

ITV-Wales/WGC/YouGov, Feb 20143919917106
FoES/YouGov, April 20143721719106
ITV-Wales/WGC/YouGov, May 20143519717148
BBC-Wales/ICM, May 20143821422104
WGC/YouGov, June 20143220517179
ITV-Wales/WGC/YouGov, June-July 20143421518167
ITV-Wales/WGC/YouGov, September 201431215161710
ITV-Wales/WGC/YouGov, December 20143120619159
2014 AVERAGE34.620.

Perhaps the best way to review these poll findings is to consider them in relation to a series of questions for each of the parties that I posed at the end of my review of 2013:

  • Would the apparent slippage seen during 2013 in Labour’s poll ratings for Westminster and the Assembly constituency vote continue?

Yes – this clearly has happened to some degree. Over the last two years there have been no huge falls in Labour support from one poll to the next (certainly when one looks at polls by YouGov, the only company to have reported figures regularly). But perhaps because of this, many people seem to have failed to notice quite the scale of Labour’s cumulative polling decline in Wales. The small number of Welsh polls conducted in 2012 all had Labour above 50% for the general election, and at or very close to that level for the Assembly constituency ballot. (The regional list vote is more complex, for reasons discussed here). By the end of 2014, Labour was polling in the mid-30s for Westminster, and even lower for the Assembly list ballot. This is a quite startling decline, and goes far further than the party’s fall in support in England over the same period. Of course in one sense this simply emphasises just how strong Labour’s position was in 2012. Even now any of the other parties would love to have Labour’s poll ratings. But, relative both to its recent and its longer-term history, Labour in Wales no longer appears like the all-conquering hegemon of the past.


  • Would the Conservatives continue to hold their ground in Wales?

Yes – despite the rise of UKIP, the Conservatives’ poll ratings in Wales have remained impressively robust. The Tories are showing no signs of making any significant advance in the polls. But their support has remained steady at what is, by historic standards, quite a decent level for the party in Wales. That the Conservatives have been able to achieve this despite having been in government at the UK level for nearly five years, and despite also the rise of UKIP as a competitor on the right, is rather impressive. They now look in a strong position to defend most of the seats they won in 2010, and maybe even to challenge for one or two seat gains.


  • Would the Liberal Democrats start to turn around the dire poll ratings they have experienced in Wales almost since the ink dried on the UK coalition agreement?

No – there has been no sign at all of the Liberal Democrats’ support levels recovering. Indeed, the opposite has been the case. Things have continued to get even worse for the party. Unless they are able to turn the polls around soon, or are able to hold onto support much more strongly in the seats they currently hold than across the nation as a whole, all three Liberal Democrat MPs in Wales must regard their seats as being vulnerable. It would still be something of a surprise to see them completely wiped out in the general election. But it is no longer at all inconceivable.


  • Would Plaid Cymru build on the apparent slight lift in their poll ratings seen at the end of 2013?

Yes – just about. Plaid did not advance substantially in the polls during 2014, but it maintained the modest progress made in 2013. Slightly more positively for Plaid, Leanne Wood’s personal ratings appeared to make some significant progress. Plaid ended 2014 a very long way short of the position of their sister party in Scotland. But they are now favourites to hold on to all three of their current parliamentary seats, and would appear to have at least some chance of making one or possibly two gains in the general election.


  • Would UKIP continue to establish themselves as a significant force in Welsh electoral politics?

Yes. Perhaps the most significant change in 2014 in Welsh politics – even more important than the fall in Labour support – was the rise of UKIP, who ended the year as unquestionably a significant part of the electoral scene. UKIP is currently on course to gain a substantial share of the Welsh vote at the general election. At present it still seems unlikely that UKIP will actually win a parliamentary seat in Wales. But it is no longer completely mad to suggest such a thing. And even if UKIP does not itself win anywhere its support level is sufficient that it could, in some marginal seats, make the difference as to which party does win. Moreover, the semi-proportional system used for devolved elections makes the prospect of UKIP members of the National Assembly in 2016 currently appear not merely possible but probable.

Because of the changes discussed here, we now face a very different general election in Wales than seemed plausible merely twelve months ago. Then, the main question appeared to be the scale of the gains that Labour would make in Wales on the 26 seats they won in 2010: whether their opponents might limit Labour only to modest gains (1 or 2 seats) or whether, as then seemed much more likely, Labour would make more substantial ground (possibly even ending up with something like the 34 seats won in 1997 and 2001). Now, with the most recent poll actually putting Labour’s support level no higher than the level they won in 2010, we face a rather different and more complex environment. But that’s for the future, and something I’ll explore in much more depth in blog posts over the next few weeks and months.

So How Was it for You? Elections in Wales, 2014, Part 1

As we begin 2015 I thought it would be appropriate to start this year’s blogging in the same way that I began 2014: with some reflections on the electoral year in Wales that has just passed.

By far the most important single electoral event to occur in Wales during 2014 was the European Parliament election in late May. In the months leading up to the election, the Welsh Political Barometer conducted several polls on voting intention; these were supplemented in April by publication of the findings from an academic study:

ITV-Wales/WGC/YouGov, Dec 20134120913135
ITV-Wales/WGC/YouGov, Feb 20133917712187
FoES/YouGov, April 20143918711206
ITV-Wales/WGC/YouGov, May 2014

(Certain to Vote)













As the table shows, the first poll conducted in the lead up to the election – that in December 2013 – put Labour in a very strong position. Had the election produced this outcome, Labour would have won three of the four Welsh seats in the European Parliament. Even in April, and as UKIP’s ratings were steadily rising, Labour still looked in a very strong position to win two of the four Welsh seats. Meanwhile, every poll except for the final one suggested that Plaid Cymru were odds-on to lose their representation in Europe.

As with previous European elections, the final turnout was poor – albeit marginally less abysmal than in 2009. This was pretty much as expected. Far less expected was the result produced by those who did vote. The table below gives the full European election result in Wales:

PartyVotesShare (change from 2009)MEPs
Labour206,33228.15 (+7.86)1
UKIP201,98327.55 (+14.76)1
Conservatives127,74217.43 (-3.79)1
Plaid Cymru111,86415.26 (-3.25)1
Greens32,2754.52 (-1.04)0
Liberal Democrats28,9303.95 (-6.73)0
BNP7,6551.04 (-4.38)0
Britain First6,6330.90 (n/a)0
Socialist Labour4,4590.61 (-1.20)0
No2EU2,8030.38 (-0.87)0
Socialist Party of GB1,3840.19 (n/a)0
Turnout = 32.1% (+ 1.7% on 2009)


Two things were particularly surprising about this result. The first was the allocation of the four seats. The gap between the party coming first and that coming fourth was sufficiently small that, on the d’Hondt formula used to allocate the seats, the seats were split four ways. Given the prior expectations generated by the polls, this was a major relief for Plaid Cymru, in particular, and a serious disappointment for Labour.

Possibly an even bigger surprise, however, was the strength of the performance by UKIP. Prior to 2014, Wales had always been UKIP’s second or third weakest ‘region’ in European elections in Britain since the regional list system was introduced in 1999. Now, UKIP made once of its largest advances anywhere right here in Wales, and came within 0.6 percentage points of topping the poll. This was a remarkable, and unexpected, performance from them.

The European election was not, however, the only electoral contest in Wales during 2014. There was a steady stream of local government by-elections for seats across the 22 Welsh local authorities. My friend Harry Hayfield has very kindly prepared for us a detailed list of the results. The overall patterns are summarised in the following table:

PartyTotal VotesN of candidatesSeats WonNet Gain/LossAverage Swing#
Plaid Cymru2,44862+1+12.0%

(#Mean average swing from the previous election, for all by-elections where a party stood candidates in both the by-election and the previous election. This measure therefore does not include cases where a party failed to stand a candidate either in a by-election or the previous election. It has only been possible to calculate this measure for the four main parties.)

Observing detailed patterns in these local results is hampered by the fact that parties tend to be somewhat selective in where they stand. No party stood candidates for every by-election; and in the Bowydd and Rhiw ward in Gwynedd, there was no by-election because Plaid Cymru nominated the only candidate for a vacancy! However, the results for Labour, who stood in most of the seats, do seem in line with those of the opinion polls during 2014, in showing declining Labour support. The Welsh Conservatives continued their strange under-performance in local elections – strange in the context of a robust opinion poll showing. Less strange was the continued weak showing from the Liberal Democrats, who saw their vote share slide even in the relatively few places where they chose to stand. UKIP performed creditably where they stood. But the only party making significant ground during local by-elections in 2014 was Plaid Cymru.

In the second part of my review of the year, I’ll turn from actual elections to the opinion polls in Wales, and consider what they suggest about the parties’ prospects for forthcoming elections.