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


The Land of Wishful Thinking?

19 November 2013

Though this site is entitled Elections in Wales, I’ve mentioned before that we’ll make occasional excursions into my other main academic interest, public attitudes to devolution. After all, how the devolution settlement in Wales is perceived by people may well influence how they vote, and what they see elections in Wales (particularly devolved elections) as being about. This post is about some research findings on attitudes to devolution which I’m currently writing up for an article.

Since the referendum in 1997 which led to the creation of the National Assembly, many surveys have explored people’s views about devolution. One common form of question in these surveys has been to ask people to choose their preferred option for how Wales should be governed from several plausible alternatives. A survey conducted immediately after the 1997 referendum, and several subsequent ones, asked people to choose between the following options:

·    Wales should become independent, separate from the UK and the European Union.

·   Wales should become independent, separate from the UK but part of the European Union.

·   Wales should remain part of the UK, with its own elected parliament which has law-making and taxation powers.

·    Wales should remain part of the UK with its own elected assembly which has limited law-making powers only.

·     Wales should remain part of the UK without an elected assembly.

 Alternative question formulations have been tried. In 2009, a study of public attitudes was conducted in fourteen ‘regions’ across Europe, including Scotland and Wales. Needing a wording that would work across multiple different political and constitutional contexts, this asked respondents to choose between the following options:

 ·    There should be no devolved government.

·     The devolved institution should have fewer powers.

·     We should leave things as they are now.

·     The devolved institution should have more powers.

·     The country should become independent, separate from the state.

 In general these options seem to work as broad measures of public attitudes to devolution. The vast majority of survey respondents are able to answer them, and using consistent measures over time allows one to trace change in attitudes. Surveys conducted from 1997 onwards using the ‘Assembly/Parliament’ question above showed a big drop in opposition to devolution after 1997.

 However, there are also limits to these questions. While No Devolution and Independence are the obvious options to place at each end of the scale, it’s not intuitively obvious what the options should be in between. Nor is it self-evident how people will understand the options they are given. How different do most people see the ‘Parliament’ option in the first list above from either the ‘Assembly’ one or from Independence? Nor do these questions tell us how people see the status quo. This raises concerns about interpreting answers to the second question: if people say they favour ‘more’ powers, what does that mean – more than what?

To address some of these concerns, we included a different way of measuring attitudes to devolution in the 2011 Welsh Referendum Study. We presented people with a 0-100 scale, anchored by the two options of No Devolution at one end (scored 0) and Independence at the other (scored 100). We then asked people to indicate on this scale: ‘Where would you place things as they are right now’, and ‘How you would like to see Wales being governed’. We also asked about the preferences of the main parties (e.g., ‘How you think the Labour Party would like to see Wales being governed?’), and about the consequences of a Yes vote in the referendum.


The first thing to check was whether this new form of question ‘worked’. Happily it did: most respondents could answer it, and their answers made sense in relation to how they answered other questions (asked in other survey waves). So, if we group people by how they responded to the More/Fewer powers question above, we get the following average response on the 0-100 scale to ‘How you would like to see Wales being governed’:

     Those who chose the No Devolution option on the More/Fewer powers question averaged 9.1 on the 0-100 scale;

    Those who chose Fewer Powers on the More/Fewer powers question averaged 12.6 on the 0-100 scale;

    Those who chose Leave as Now on the More/Fewer powers question averaged 33.7 on the 0-100 scale;

    Those who chose More Powers on the More/Fewer powers question averaged 61.8 on the 0-100 scale; and

    Those who chose Independence on the More/Fewer powers question averaged 83.2 on the 0-100 scale.

Knowing that our 0-100 scale works, we can use it to investigate some interesting aspects of attitudes towards devolution. For instance, what is the relationship between people’s perceptions of the current situation and their own preferences? How does ‘Where would you place things as they are right now’ correlate with ‘How you would like to see Wales being governed’?

My expectation was that we would see what political scientists have called a contrast effect: that is, that people would tend to see a strong contrast between the current situation and how they would like things to be. Thus, I expected that those who remain opposed to devolution would tend to perceive it as having gone much too far, and they would certainly perceive devolution has having gone much further than would those who support independence or enhanced devolution. The latter group, I expected, would tend to perceive devolution as having gone not nearly far enough, and this would make them much more likely to place ‘things as they are right now’ nearer to the No Devolution end of the spectrum.

In fact, we find exactly the opposite. The chart below groups respondents according how they answered ‘How would you like Wales to be governed?’, and shows the average response for each group on ‘Things as they are right now’. Put simply, it shows how far people think devolution has gone – and it shows that these perceptions depend greatly on how far they would like devolution to go. Those whose own preference is for no or very limited devolution tend to see place ‘things as they are right now’ only quite a modest distance from the No Devolution end of the spectrum. But those favouring independence or much enhanced devolution see things as having actually gone quite a long way towards their desired goal!


 This result has been repeated (indeed, even more strongly) in a subsequent wave of sampling, so it is not just some fluke. Personally, I find this one of the more disturbing findings in all the research I’ve conducted on public attitudes in Wales. Those who dislike Devolution seem to be telling themselves that it doesn’t matter that much. By contrast, many of those supportive of substantial autonomy for Wales seem to have convinced themselves that they are well on the way to realising their goals! These findings may in part reflect the lack of clarity in the Welsh devolution settlement, allowing people to see in it whatever they wish.

But they also suggest that, with regards to how the nation is governed, Wales is currently the land of wishful thinking.



  1. J Jones

    Yes well…..I have long maintained that those who don’t vote in Welsh referenda are actually the people who are devo-sceptic. I think that this research rather supports that view. The devo sceptics are actually quite unaware of the powers of the Welsh Assembly and therefore see no immediate reason to concern themselves.

    It’s one of the reasons why the Silk Commission’s surveys by Beaufort and ICM were flawed; they had disproportionate non response from weak anti devolutionists. This was shown by the extraordinary level of response from Welsh speakers and the high percentage of fluent Welsh speakers….60% Welsh speakers in one case and 40% in the other.

    Beaufort did other surveys (for the BBC for instance) where they weighted the final result to Welsh speaking adult population: 17% Welsh speaking (fluent 8%).

    For the Silk commission no such adjustment took place.
    However I do acknowledge that Silk probably represents the future referendum result…unless those sleeping non voters are finally stirred from their “dream land”… this case the land is Britain and they haven’t noticed that it is now Wales.

    • Roger Scully

      Dear Jon,

      Thanks very much, once again, for your interest in the Blog and engagement with the issues discussed in it.

      However, I find it difficult to follow your interpretation of my latest post. What the findings I posted about suggest – I think – is a degree of ‘wishful thinking’ right across the spectrum – among both those hostile or cautious about Welsh devolution, and those supportive of stronger devolution or even independence.

      I am also dubious about the idea that there is a very large contingent of ‘quiet devo-sceptics’. One of the things we did in our research on the 2011 referendum was specifically look for such people: we thought that the weakness of the No campaign might have meant that there was a latent No vote that had not been effectively mobilised. We couldn’t find a substantial group of such people. Indeed, those who indicated in our pre-referendum survey that they were likely to vote, or might vote, but who in the end didn’t end up voting, were actually slightly more pro-devolution than those who did vote.

      More generally – while the Beaufort research may have produced an overly pro-devolution sample, their results were not vastly different from those that all the other major surveys over the last decade or so have produced. I think I’d want to mentally adjust downwards a little bit some of the percentages supportive of devolution in the Beaufort findings, but I don’t think they were miles away from the truth.

      But to get back to the subject matter of this post: though I’m not wholly convinced by your suggestion, I’m still a little unclear quite why we do get these results. And any further thoughts – on this, or indeed on any other matters – are always gratefully received.

      best wishes


  2. Jon Jones

    It’s this Roger:-

    ” Those who dislike Devolution seem to be telling themselves that it doesn’t matter that much.” I don’t think it’s too far fetched to say that devo sceptics who don’t think it matters very much can’t be bothered to vote in a referendum or respond to a survey. On the other hand devo enthusiasts can’t wait to talk about the subject or vote to bring it forward.

    Why am I cautious about the surveys? If ICM had 60% of their respondents who said that they speak Welsh is it likely that they also had 31% of their respondents who were “born outside Wales”? The census tells us something else about those born outside Wales……only 8% speak Welsh.

    It would have been so easy for ICM or Beaufort to weight their sample correctly or ask one extra question; “Where were you born”? Such things give such an accurate fix on the sample and in this case that was vital.

Leave a reply

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