Further analysis of the YouGov poll

The latest YouGov poll commissioned by ITV Wales was, as others have remarked, rather slow to get noticed (it was on Tom Bodden's blog first, I think) and much of what others have said doesn't need repeating by me. So I thought I'd try to say some things that haven't been said about it instead.

I found the regional percentages interesting, particularly when contrasted with the 2005 figures. However we should bear in mind that the sample sizes for the six regions (YouGov splits Cardiff from SWC) are very low, and that this means that they shouldn't be considered as anything more than a rough indicator of trends. If anyone wants to check my maths, the 2005 spreadsheet is here. Please let me know if I've made any mistakes.

North Wales

Labour 34% ... down 6.8% from 40.8%
Plaid 17% ... up 2.8% from 14.2%
Conservative 33% ... up 8.7% from 24.3%
LibDem 11% ... down 4.5% from 15.5%

Mid & West Wales

Labour 28% ... up 1.7% from 26.3%
Plaid 20% ... down 3.4% from 23.4%
Conservative 28% ... up 4.1% from 23.9%
LibDem 14% ... down 9.8% from 24.2%

South Wales West

Labour 36% ... down 14.4% from 50.4%
Plaid 11% ... up 1.4% from 9.6%
Conservative 32% ... up 15.2% from 16.8%
LibDem 16% ... down 2.9% from 18.9%

South Wales Central (excluding Cardiff)

Labour 43% ... down 11.4% from 54.4%
Plaid 15% ... up 4.2% from 10.8%
Conservative 25% ... up 6.5% from 18.5%
LibDem 8% ... down 5.9% from 13.9%


Labour 39% ... down 2.4% from 41.4%
Plaid 12% ... up 5.8% from 6.2%
Conservative 32% ... up 8.7% from 23.3%
LibDem 13% ... down 13.2% from 26.2%

South Wales East

Labour 36% ... down 12.9% from 48.9%
Plaid 4% ... down 3.1% from 7.1%
Conservative 37% ... up 16.7% from 20.3%
LibDem 13% ... down 0.7% from 13.7%

All Wales

Labour 35% ... down 7.7% from 42.7%
Plaid 13% ... up 0.4% from 12.6%
Conservative 32% ... up 10.6% from 21.4%
LibDem 13% ... down 5.4% from 18.4%

From Plaid's point of view it's good to see that in overall terms we're marginally ahead of the percentages we got in 2005. The fact that the LibDems have fallen generally, and especially sharply in MWW, will help us in Ceredigion, and Labour's fall should see us through comfortably in Ynys Môn. I don't think many people doubt that Plaid will win both.

In Llanelli, the percentages last time were Labour 46.94% to Plaid 26.48%. That's a wide margin, and the key is whether any fall in Labour's vote will transfer equally to Plaid. We'd be looking at a Labour fall of just over 10%, matched by a Plaid increase of just over 10%. The figures for MWW show that Labour's vote is holding up, but everything will depend on local variations. In particular Llanelli shares an industrial heritage with the Valleys to the east, where the drop in Labour support is consistently greater than 10%.

Our fourth target seat is Aberconwy, which will be a fight between us and the Tories. In predicting this will go to the Tories, most commentators have not really understood the effects of the boundary changes and I am quietly confident that Plaid will win this.


It is also worth looking at page two of the YouGov survey to see what people think are the most important issues ... and especially those issues which supporters of different parties see differently.

There are no prizes for guessing that the Economy is by far the most important issue identified, but supporters of all parties rate it roughly equally. Although lower down the list, supporters of all parties have roughly equal concern for issues such as Unemployment, Tax and Crime.

It is no surprise that Plaid voters rate Devolution as much more important to them than it is to supporters of other parties. But Health is another issue that rates much more highly to Plaid voters (51%) than it does to the supporters of other parties. It is comparatively low on the Tories' list (25%) ... and that might seem to suggest that Cameron's first mistake of making a big issue in Wales about not cutting NHS funding in England was compounded by the fact that Welsh Tory supporters don't really rate it as an issue anyway. The Tories in London seem to understand neither Wales, nor what Tories in Wales think ... as anyone can tell just by comparing their relatively-enlightened AMs with the rabid and irrelevant utterings of their MPs.

The thing that disproportionately matters to Tory supporters in Wales is Immigration (51%) while supporters of other parties put this issue much further down their lists. Once again this is an issue where what matters to Middle England doesn't matter very much to most people in Wales.

Tory supporters also attach very little importance to the Environment (6%) compared with supporters of other parties.


These figures might give us some ideas about how we campaign. Take for example the issue of Housing which I mentioned in my post yesterday. I suggest that the long, drawn out delays and refusals by, in the main, Labour MPs to let the Assembly legislate on Housing should not be presented as a Devolution issue (even though it is) because that won't resonate with the supporters of other parties that we want to win over. It would be much better to present it as a Housing issue (which is also is) because that will resonate with them, and therefore win them over. That approach should also highlight the fact that we want primary lawmaking powers for the Assembly not for their own sake ... but because having them will enable us to do practical, concrete things to make Wales better.

We should certainly exploit the fact that this is a Westminster election, and that there will therefore be a good number of Labour supporters who are relatively happy with the performance of their AMs in the One Wales Government ... but disgusted at the way Labour MPs have continually delayed and watered-down the LCOs that would enable their AMs to make a difference. Labour are hopelessly split on this issue ... and we should do all we can to expose that split in the election campaign.

As encouragement, it is interesting to note that voters are already beginning to see this, and that the Valley heartlands where Labour have in the past weighed rather than counted votes all show double-digit falls in Labour support, as mentioned above.

Bookmark and Share


Anonymous said...

This is an excellent analysis, but we should still be making more of an impact. Why in your opinion are we getting squeezed?

Paul Williams said...

The Ynys Môn result may not yet be in the bag for Plaid - take a look at my full analysis of the latest YouGov poll and how it predicts the Ynys Môn result here:


James Dowden said...

1) The sample size is pathetic. They talked to 157 voters in Cardiff, and 115 in South Wales Central; the other regions are scarcely any better. With the four Cardiff constituencies being so different from each other, the wild swings suggested look like an oversampling of Cardiff West and an undersampling of Cardiff Central.

2) Their region definitions are extremely unhelpful. Dwyfor Meirionydd drags the whole Llais Gwynedd (even if they don't stand, people will mention them in a poll) issue into Mid & West Wales, clouding how Plaid are doing in constituencies that are genuinely in the West.

3) "Immigration" is a rubbish word to include in a poll in Wales. It can mean three different issues:
a) The Daily Mail Little/Middle Englander issue;
b) Non-Welsh-speaking immigration to Welsh-speaking areas; and
c) Stopping the English Home Office's racist campaign against Welsh Patagonians (i.e. the polar opposite of (a)).
We can safely say that most of the Tories didn't mean (b) or (c), but what about the Pleidwyr?

4) Overall, this poll shows a failure to connect to the electorate by both Plaid and the Lib Dems. Neither party is providing the critique of the other two or the distinctive forward-looking agenda for Wales that is so needed.

MH said...

I'll start with Druid. Welcome to the Welsh blogosphere. It's nice to see people doing some number crunching, and it is interesting that your figures make it a much closer contest between Plaid and Labour than the bookies' odds suggest. That said, I—and probably the whole party—would be very disappointed if Plaid didn't win.


James, I wouldn't be too critical about the sample sizes. Before YouGov set up their Welsh panel we had very little polling information specific to Wales, so I'd be more inclined to praise them (and ITV Wales for forking out the money) than be critical. 1000 is regarded as a good sample size, and it's interesting to see that YouGov have increased the panel size to get 1137 for this poll.

As for regional breakdowns, these simply reflect the five regions used for electing additional members to the Assembly (except for the fact that Cardiff is split from SWC). So the division makes sense. I'd much rather have this degree of information than not have it ... but we must be careful how we use it because it is not as reliable as a larger sample would be. It's not a precise science. In a sense YouGov are still steering by the seat of their pants and we (and they) will get a far greater degree of trust in the figures only when they have been tested out against a real election. I just hope ITV Wales are planning two or three more before the election. We should also have the BBC St Davids Day poll by ICM.


Now the big question, raised by both James and Anon, is why Plaid (I don't care too much about the LibDems) is not making greater headway.

I'm sure there are many reasons. But there is a big difference between Plaid's performance in Westminster and Assembly elections. If we get five MPs, that will be the highest number we have ever got. Six or seven will be an excellent performance.

We're not actually getting "squeezed" because in the constituencies in which we are strong we are one of the two horses in the local race. But we have different opposition in different seats. However we are likely to be squeezed in constituencies in which we are also-rans in a tight contest between other parties ... usually Labour v Tories. I think that's why our vote has gone down in SWE, and the same phenomenon would probably mean that our vote will go down in NE Wales (if that was measured separately). If this analysis is correct, it would mean that Plaid is doing better in NW Wales than the figures for North Wales as a whole suggest ... and that would in turn suggest that I have good grounds to expect Plaid to win both Ynys Mon and Aberconwy.

Welsh Left said...

The regional votes broadly reflect what should be happening, so that suggests to me they make sense to an extent.

South Wales Central seems to be showing progress for Plaid in a way the other regions don't- mainly in my experience because of the 2 Plaid AMs there Leanne & Chris who work their communities week in week out. We need to copy that approach in Newport, Swansea, Bridgend, Cwmbran, Merthyr- all big places where Plaid does sod all usually.

Plaid Whitegate said...

N E Wales will be interesting for Plaid for a number of reasons, not least because they have succeeded in improving on the quality of their candidates and campaigns in all 6 seats.
That perhaps showed in the Euros last year, where Llyr Huws Griffiths came second in Clwyd West, Plaid got 15% in Wxm and was within 100 votes of beating Labour in Clwyd S.
Because things are so fluid now - with the Tories doing well in the Euros on a low turnout - I don't see the clear "two-horse race" scenario developing as it might do in Aberconwy or Ceredigion.
Of course, other issues will also affect the vote - changing demography, growing identity with Wales and the amazing decline in the Labour Party's organisation (it doesn't control a single council in the area and has struggled to come to terms with opposition).
I think the digital switchover is quite significant here because it means that the default channel for BBC1 is now BBC1 Wales, whereas before many areas close to the border couldn't receive a good signal or traditionally tuned their aerials into Granadaland. Will this, over time, help the Welsh political process up here? Who knows!

Anonymous said...

I consider immigration very important to me. What's worrying to me is the level of movement from large English conurbations to (mainly) rural Welsh speaking Wales. I link this to the immigration into the English cities massively pushing up house prices there and allowing those with properties to sell up and buy a propery (even properties) in west Wales with money in the bank to spare. It's also fuelling the buy-to-let phenomenom. I'm not totally against immigration but the levels seen in recent years are without a doubt having a knock on effect on my community. I would like it to stop.

Post a Comment