The BBC/ICM Poll: Brexit

The recently-published ICM poll for BBC Wales also included a short set of questions about Brexit. Usefully, these questions were repeated from the same poll conducted last year, so it gives us some opportunity to see if there has been any change in attitudes over the last twelve months.

The question format first of all asked respondents the following:


“Thinking about Brexit, the decision to leave the European Union, do you think it will have a positive or negative impact or make no difference?”


Respondents were then asked about three specific matters:


  • To the Welsh economy
  • To your own personal finances
  • On the way of life in Wales today in general


What did ICM find? Here are the results for each item – with the changes since the poll in 2017 indicated in brackets.


  Welsh Economy Personal Finances Way of Life
Positive impact 24% (-9) 14% (-3) 23% (-3)
Negative impact 49% (+5) 36% (+6) 46% (+9)
Make no difference 17% (-1) 41% (-4) 22% (-9)
Don’t know 10% (+4) 9% (+1) 9% (+3)


It is clear that the balance of attitudes was negative on all the questions that ICM asked. This is not particularly new – views on the consequences of Brexit have tended towards the negative for pretty much the entire period since June 2016. Nonetheless, things have become more negative since last year. On all the questions asked, positive and neutral expectations have declined and negative ones have increased.

When one examines the details of the poll, we see that Conservative supporters have more positive expectations of Brexit than do supporters of Labour, Plaid and the Liberal Democrats. This is unsurprising, although even among Conservatives the overall balance is some way from wholly optimistic. For instance, on the ‘Welsh economy’ question, some 38 percent of Conservative (Westminster) supporters say that they expect a positive impact, while 28 percent expect it to be negative (with 22 percent opting for ‘no difference’ and 12 percent saying don’t know).

One thing that we can’t know from the BBC/ICM poll is if these increasingly negative expectations about Brexit are feeding through into whether people still support it, or would do so in the event of another referendum. The poll did not ask about that. However, the Welsh Political Barometer polls have been continuing to ask about that, and the results from this questions in the next such poll will therefore be of particular interest.


  • Bobbybob

    An important part would be the level of significant error in the results. What is the varience and is the change above considered significant.
    The second important part is that the question is so badly worded by BBC/ICM that it is loaded for a negative response. It would be more useful to have a better set of questions that can really show what people think. Sadly you wont get such from the BBC.

  • J.Jones

    And once again…look at the correlation between Welsh speaking ability and attitude to leaving the European Union. Non Welsh speakers most likely to see Brexit as positive, least likely to see Brexit as negative.
    Fluent Welsh speakers most likely to see Brexit as negative, least likely to see Brexit as positive.

    Having an unlikely percentage of Welsh speakers in these ICM/BBC polls is something that seriously needs to be investigated. If nothing else it is an indication of either a biased pool or faulty sampling technique.

  • Paul Jerome

    J.Jones? Sounds like just the sort of non de plume that would be adopted by someone with an illogical dislike of the Welsh language.

    Someone who would make an irrelevant, random comment about Welsh speakers and cast unfounded aspersions at the methodology.

  • oldnat

    As always, I post here somewhat tentatively, looking for understanding rather than to make a point.

    I note J Jones comments about the difference between Welsh speakers and English monoglots, but I noticed recently that in the 2011 census around 14% of those in Wales described their identity as “English”. I presume (though I may be wrong) that few of these people speak Welsh.

    So, though the language issue is obviously much stronger in your country than mine, is it the best factor to use? Wouldn’t it be useful to have a question on the respondent’s national identity?

  • oldnat

    Where I will make a point, however, is to suggest that the state broadcaster has a charter obligation to promote British unity, so that’s probably the very last question the BBC would want to ask!

  • J.Jones

    AHA Oldnat….exactly the point that I made to ICM and the BBC in a complaint I made about the 2017 St David’s day poll.
    In fact 21% of adults in Wales were born in England and only 8% said that they were able to speak Welsh. On top of that there are people from mainland Europe (largely Poland and the East) Asia and Africa. Of these people, ability to speak Welsh is very low according to the census; much lower than those born in England.

    There is more evidence however on this matter of self selection bias; The National survey for Wales had to make corrections, as other ONS surveys have in the past, because Welsh speakers, once identified, were more likely to respond to in depth questionnaires than non-Welsh speakers.
    You see the problem; if Welsh speakers sign up in disproportionate numbers for polling companies, and those companies only sort by demographic, socio economic, and geographical weightings then they will have a preponderance of the “willing” participants….disproportionately Welsh speakers. It makes little difference on many questions but a big difference on Welsh Constitutional questions and attitudes to the Welsh language…also Brexit .

    I am saying that having 66% of the sample able to speak some Welsh is a “tell tale” that something else is wrong. These polls are used by Nationalists to “Drive” opinion on Welsh politics. The BBC has a duty not to display bias; others (Cardiff University) have not.

  • Penddu

    Jeff, one point you never address is what is meant by a knowledge of Welsh. I am not fluent, conversant or have any meaningful ability in Welsh- but I know a lot more than just a few words- as I expect does anyone who has attended school in Wales since the 70s onwards.

    Depending on how the question is asked I would respond as “Can not speak Welsh” or “I speak some Welsh”. The two are not mutually exclusive.

  • Penddu

    Interesting Jeff but doesn’t really adress the point I am making.

    The designation of being able to “speak Welsh but more than a few words” is nonsense. There is a standard EU system used to grade degrees of fluency in different languages which makes more sense.

  • Penddu

    EU system grades language ability on a sliding scale of 1-6, with
    A1 – Entry
    A2 – Elementary
    B1 – Intermediate
    B2 – Upper Intermediate
    C1 etc
    With definitions on how to categorise etc, and WJEC has an equivalent grading system for Welsh.

    Personally I am A1 – may be A2 on a good day – but never would I call myself a Welsh speaker.

Leave a Reply

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