The challenges and successes of YouGov's political polling in Wales
Over the past 30 years, there has been very little in the way of political polling in Wales. There are a number of reasons for this but, mainly, it appears be a lack of interest from a commercial perspective. That is not to say there was no interest in polling at all, just that there was little scope to conduct small pieces of cost-effective political research. Without clients, any potential for large scale political research in Wales was lost, depriving us of the kind of richness of data available in Britain as a whole.
With this in mind, it was perhaps unsurprising a company such as YouGov took the plunge and developed, marketed and continues to supply a Welsh Omnibus survey. With this product, clients can pay for research on a per question basis, reaching 1,000 representative Welsh Adults within 48 hours. On the back of this service, there has been more political opinion research in Wales in the past 12 months than there was for three decades. This has allowed us to start building a rich stream of data in Wales which will hopefully continue into the future.
This does not however tell the whole story. In our experience, the attitude towards polling in Wales is very different to that of the rest of Britain and one that perhaps needs to change to give political polling the best chance to thrive. In one conversation with a potential client we were informed that they would not be interested in polling on a certain topic as it could 'upset the parties'. The best national opinion polling is not about pleasing or upsetting political parties, though it regularly does, it is about giving the public a voice to show their (dis)satisfaction with a government, policy or programme.
At YouGov, our aim is to know 'what the world thinks' – to provide a continuous stream of data so that companies, governments and institutions can better serve the people that sustain them. This is exactly why we passionately believe in the value of tracking public opinion and reporting it as and when it changes. Vital to a democracy is the ability of its citizens to tell a government when they are pleased or angry, when they want changes or when they happy for their government to carry on. At YouGov, we go to great lengths to perfect our methods to make sure that our samples represent the right proportions of men and women, young and old, Labour and Tory etc. By doing this, YouGov is able to accurately gauge public opinion in Wales and provide a valuable voice in Welsh political discourse.
Following our success in the 2010 General Election, YouGov successfully predicted the outcome in Wales with an average error of just 1.4%, ITV Wales commission YouGov for monthly polls leading up to referendum and Assembly elections in March and May of this year. With the 1997 referendum ending in an almost 50/50 split, with Yes winning just 50.3% of the vote, many expected the 2011 referendum to be a similar, tight contest. YouGov’s polling in the months leading up to referendum however showed something different. Even as far back as July 2010, when we were showing a 60/40 split to Yes, some believed that the No vote from 1997, whilst concerned with different issues, would not drop this low. This 40% dropped even further over the coming months and, exactly as we had been claiming for almost a year, the Yes vote recorded a crushing victory over No.
While our polls in the run up to the referendum remained consistent, the Assembly elections were proving to be a much more interesting story. For the best part of a year prior to the elections, our assembly polling had been showing that if Labour did not get a majority, they would be very close. There was also the continuing story of the decline of the Liberal Democrats. In the aftermath of the General Election and the broken tuition fee pledge, Kirsty Williams and the Welsh Lib Dems, like their colleagues in Westminster, saw their support fall off the edge of a cliff. In August 2010, they were polling at just 10%, (half what they polled in May) and this was a position from which they did not recover, regularly dropping down into single figures for the regional vote.
In our final poll for ITV Wales, it was clear that Labour were going to be right on the edge of a majority. We knew it would be close but were still not sure whether Carwyn Jones would be able to lead a Labour Assembly without the need of a coalition partner, a position that was still unclear even as results were announced on the Friday following the election. Traditionally, national polling has had a tendency to overstate Labour support and underestimate the Conservative vote. Whilst we got the story spot on with Labour on the cusp of a majority, our final Assembly figures appear to contain some of this bias and we are actively looking to put this right for future elections. There was however some anecdotal evidence suggesting a further reason which could have contributed to the Conservative’s excellent performance.
Some claimed that opposition to AV amongst Conservative voters played a major role in motivating them to get out to the polls. We tested this theory in our first post-election poll, asking respondents whether they went to vote mainly in the AV referendum or the Welsh Assembly election. Whilst the majority of voters for the four main parties all stated that they were motivated to vote in both elections, three times as many Conservative voters (16%) went mainly for the AV election as those (5%) who went for the Assembly vote. Only Liberal Democrat supporters had as few people stating they were there mainly for the Assembly elections.
Going forward, YouGov is hoping that it can continue in the same vein. Whilst the past 18 months have been an exciting time for Welsh politics, with four elections, it does not mean that polling should simply cease until the next set of elections. With Labour now looking to run a minority government, Welsh politics shows no signs of losing its excitement. How the Labour party copes with compromise and conflict will be crucial for their electoral success and this can all be tracked, analysed and built upon as the Assembly faces its changing circumstances.
There is also a real interest in the changing nature of devolution in Welsh politics. Whilst there appears to be no real appetite for independence amongst Welsh adults, just one in ten actually wants full independence right now; there are two important factors that could influence the devolution debate in Wales. Firstly, the additional powers gained in the referendum have given the Welsh Assembly a greater mandate to rule without Westminster interference and it will be vital for Carwyn Jones to prove that the faith the Welsh population showed in March was not unfounded. It will be important to track what impact, if any, the additional powers have over ordinary people in Wales as well as indicating how well, or badly, the Assembly is working in light of these changes.
Secondly, there is the issue of Scottish independence which could potentially have an impact for Welsh feelings on devolution. Whilst there is much greater demand for independence in Scotland it would be foolish to assume that there is little appetite for constitutional changes in Wales. One in three Welsh adults want the Welsh Assembly to have more powers than it was granted in March this year and, should Scotland break from the rest of Britain, it would focus many Welsh minds on exactly how they saw the country going forwards. So, whilst an appetite for independence looks extremely unlikely, there will be important questions of devolution that polling can help to answer.
2010 and 2011 will go down as historic years in Welsh politics but there are plenty more twists and turns to come and YouGov hopes to accurately track them all.
For more information please contact YouGov’s political polling team atpolitical@yougov.com
