Skip to main content
Coronavirus (COVID-19): Latest updates.


The November Welsh Political Barometer Poll

27 November 2017

Labour remain well ahead of the other parties, despite some signs of an ebbing in their support. But none of their opponents are making consistent progress. These are the main findings on the state of the parties from the latest Welsh Political Barometer poll, the final systematic test of public attitudes and voting intentions to be reported in the politically extraordinary year of 2017.

As per usual, our new poll examined voting intentions for both Westminster and the National Assembly. These are the voting intention figures for a UK general election that emerged (changes on the previous Barometer poll, in September, are indicated in brackets):

Labour: 47% (-3)
Conservatives: 31% (-1)
Plaid Cymru: 11% (+3)
Liberal Democrats: 5% (+1)
UKIP: 3% (no change)
Others: 3% (+1)

There are no very major changes evident here since either the general election or our last poll. But the very strong showing for Labour support that has characterised electoral politics in Wales since mid-May does, for the first time, show some signs of abating. After a poor poll last time Plaid Cymru will be pleased to move up three point; but this maybe emphasises that their unusually weak showing in September was something of an outlier.

If we use the standard uniform national swing method of converting these poll findings into projected electoral results, then our latest poll implies the following overall result:

Labour: 28 seats
Conservatives: 8 seats
Plaid Cymru: 4 seats

Eagle-eyed readers may realise that these seat numbers are exactly the same as at the general election. The very small changes in support since the election implied by our poll mean that not a single seat is actually projected to change hands!

Our new poll also asked once again about voting intentions for the National Assembly. First, here are the figures for the constituency vote (with changes since the September Barometer poll again shown in brackets):

Labour: 43% (no change)
Conservatives: 26% (+1)
Plaid Cymru: 19% (no change)
Liberal Democrats: 5% (no change)
UKIP: 4% (no change)
Others: 4% (+1)

These could hardly be said to be dramatic findings. No party has moved more than a single point since September: the overall picture is one of almost complete stagnancy, and we do not see the three-point movements here for Labour and Plaid seen for Westminster.

As with our September poll, the constituency numbers imply quite large changes in party support from the National Assembly election in May last year. And yet, on a uniform swing assumption, no constituency seats are projected to change hands. As I explained back in September, this reflects the fact that most of the Assembly constituency seats which produced close results last time were held by Labour; with their support now much higher than it was last year such places simply move back to being safe Labour seats instead of marginals.

Now for the figures for the regional list vote (with changes in party support since September once more shown in brackets):

Labour: 38% (-2)
Conservatives: 27% (+4)
Plaid Cymru: 18% (-1)
Liberal Democrats: 5% (no change)
UKIP: 4% (-1)
Others: 8% (+2)

These numbers suggest (as with Westminster voting intention, but not as seen for the constituency vote), some ebbing of Labour support. They also seem to imply a significant rise in Conservative support – something not seen elsewhere in our poll. This is puzzling – but we should remember that September’s poll showed a four-point drop in Conservative support for the list vote; that may again have been a slight outlier, and our new poll may be showing things reverting to what is the current norm. There are no major changes for any other parties evident here.

Taking into account the projected constituency results, and once more assuming uniform national swings since May 2016, our poll provides the following projected outcome for the regional list seats:

North Wales: 3 Conservative, 1 Plaid
Mid & West Wales: 3 Labour, 1 Conservative
South Wales West: 3 Conservative, 1 Plaid
South Wales Central: 3 Conservative, 1 Plaid
South Wales East: 2 Conservative, 1 UKIP, 1 Plaid

This, in turn, gives us the following overall projected result for the National Assembly:

Labour 30 seats (27 constituency, 3 regional)
Conservative 18 seats (6 constituency, 12 regional)

Plaid Cymru 10 seats (6 constituency, 4 regional)
UKIP 1 seat (1 regional)
Liberal Democrats 1 seat (1 constituency)


The Labour party continue to be well ahead in Wales. But the upwards trajectory in their support seen throughout the last few months appears to have been stemmed, at least for the moment. There are no clear signs of any of their opponents making progress, however. Plaid have gained ground for Westminster; the Conservatives for the Assembly list vote. Yet both of these apparent moves may be more apparent than real, reflecting unusually poor showings by both parties on those respective ballots in our last poll. Perhaps the clearest thing to come out of our new poll with regards to party support is the continued failure of any other parties to make ground: UKIP once more look like a marginal player in Welsh politics, and the poll emphasises the huge task facing the new Welsh Liberal Democrat leader Jane Dodds. At the moment, only three parties seem to matter in Welsh politics; and there is no doubt that, of those three, it is Labour that is doing much the best.

The poll, for ITV-Cymru Wales and Cardiff University’s Wales Governance Centre, had a sample of 1016 Welsh adults and was carried out by YouGov from 21-24 November 2017.


1 comment
  1. Christian Schmidt

    It seems to me that at least for the last decade the Assembly poll figures are consistently closer to the Westminster results than the Assembly results, esp. when it come to Plaid and the Tories. Scottish polls showed a similar regular inconsistency (there between Labour and the Nats up to 2011).

    Any thoughts why? Would it make a difference if a poll asked for the Assembly first (i.e. more respondents try to answer consistently than actually vote consistently)? Or is it based on media coverage of Westminster and devolved government (which only swings towards the latter in the weeks before an Assembly election)?

Leave a reply

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