How should a poll be reported when it finds that around half agree with something while half do not? Does it mean the glass is half full or half empty? Does the answer, perhaps, simply depend on your political perspective?
Last week marked the twentieth anniversary of the 1997 devolution referendum in which voters in Scotland voted by almost two to one in favour of creating the Scottish Parliament that now sits at the bottom of the Royal Mile, opposite the Palace of Holyroodhouse. The Sunday Times decided to mark the occasion by commissioning a poll from Panelbase (the first Scottish poll of any kind since June) that, inter alia, asked a set of questions (many of them not dissimilar to ones that have previously been asked on the Scottish Social Attitudes survey) about the impact the creation of the Scottish Parliament had had on public services, the economy and politics in Scotland.
The poll found that 45% thought that devolution had resulted in a better health service, and 49% that it had given Scotland a stronger voice in the UK, while the same proportion (49%) reckoned it had given ordinary people more say in how Scotland is governed – the latter two are, indeed, criteria where SSA has also previously found people are inclined to view the devolved institutions relatively favourably. True, rather fewer (37%) reckoned that devolution had helped strengthen Scotland’s economy, while only 35% reckoned that schools have got better – further evidence, if any were needed, that perceptions of the performance of the country’s education system is proving to be the Scottish Government’s Achilles’ heel. However, even in these instances the proportion who reckoned the Scottish Parliament had had a beneficial impact still outnumbered the proportion who felt that it had resulted in things actually getting worse – many simply felt it had not made much difference either way.
The Sunday Times decided these figures not good enough, and served up the headline, ‘Devolution has been a dud, say most Scots’. Equally in reporting the poll the following day, The Express, posted the headline, ‘Devolution has failed to improve key services, say over half of Scots’. In contrast, at the other end of the political spectrum, The National reported the findings as ‘Survey finds most Scots believe devolution has improved the nation’ while The Scotsman too reckoned, ‘Majority think Holyrood has improved NHS and education, new poll finds’. In short, everyone seemed able to find support for their existing perspective from a poll in which on balance the Scottish Parliament is more commonly viewed in a positive light than a negative one – especially by SNP supporters, though mostly not so by Conservatives – but perhaps by not as much as the advocates of devolution might have hoped twenty years ago.
A few days after the initial findings from the Panelbase poll were reported, a second poll appeared, conducted by Survation for the Scottish Daily Mail. The Mail was not particularly concerned about the twentieth anniversary of the devolution referendum, or indeed the third anniversary of the independence referendum which occurred on Monday, but rather was focused on the possible consequences of the devolution of income tax, the major new devolved power that the Scottish Parliament has acquired (in two phases) during the last eighteen months. The Scottish First Minister, Nicola Sturgeon, signalled in her Programme for Government statement earlier this month that the Scottish Government might now use this power to increase income tax in Scotland, with the implication that the tax might therefore be higher than in England.
The poll suggests that while increasing the basic rate of income tax would be none too popular – only 13% back the idea – raising the higher 40p rate might not be especially politically risky. As many as 44% reckoned that those earning more than £43,000 a year should pay more tax, slightly more than the combined total who reckoned they should pay less (12%) or the same as now (29%). In particular, as many as 57% of those who voted for the SNP in June said they were in favour. That said, when the poll went on to ask a rather different question about increasing the top rate of tax, that is, whether increasing it to 50p would be good or bad for the Scottish economy, only 32% reckoned it would be good and as many as 27% that it would be bad. Much evidently depends on question wording, suggesting that politicians on both sides of the argument potentially have popular buttons they could press to persuade voters of their case, though this did not stop the Scottish Daily Mail coming up with the unambiguous headline, ‘So keep your hands off our pay packets’.
But perhaps what was most interesting about these two polls was not so much what they found in respect of their commissioners’ original preoccupations, but rather what they told us about the current standing of the parties. There has until now been an important unanswered question about the outcome of June’s UK general election in Scotland in which support for the SNP fell from 50% (in May 2015) to 37% (in June). That question is whether, rather than simply representing a decline in the popularity of the SNP, the decline in the SNP’s fortunes represented a renewal of the relative reluctance of voters to vote for the SNP in Westminster as opposed to Holyrood elections, a reluctance that had been evidence in every UK general election until two years ago. Whether or not this was the case was unclear because none of the polls conducted in the run-up to June’s general election asked people how they would vote in an imminent Scottish Parliament election, and focused entirely on the then forthcoming Westminster ballot.
However, both the Panelbase and the Survation polls asked people how they would vote now in both a Scottish Parliament election and a Westminster one. In both cases, the proportion saying they would vote for the SNP in a Westminster ballot was much the same as the proportion that indicated they would do so on the constituency ballot of a Holyrood contest. Panelbase put the SNP on 41% for Westminster and 42% for Holyrood, while Survation credited the party with 39% and 42% respectively. Between them, these figures suggest the SNP’s decline in June cannot be put down to any marked renewed relative reluctance to vote for the party in a UK general election. Moreover, they suggest too that support for the SNP in any imminent Scottish Parliament election is rather lower than the 46% the party secured on the constituency vote in May last year, underlining the impression that the party has indeed lost electoral ground and raising (albeit perhaps rather premature) questions about the prospects for a continued pro-independence majority at Holyrood after the next Scottish Parliament election in May 2021. That said, there is no sign in these polls that SNP support has fallen further below the level it secured in June this year, while they suggest that none of the nationalists’ opponents are as yet even close to displacing it as the most popular party north of the border.
Meanwhile, the decline in electoral support for the SNP should not be presumed to be indicative of a similar decline in support for independence. True, at 43%, the proportion who told Panelbase that they would vote Yes (after excluding Don’t Knows) is not just the lowest the company has recorded in any poll since the EU referendum but also the lowest since the independence referendum three years ago. That said, it only represents a one point drop on the company’s previous reading, a drop that could simply have arisen as a result of the chance variation to which all polls are subject. Meanwhile, Survation put support for Yes at 46%, the same as in its last online Scottish poll conducted shortly before June’s election, though again down a point on polls conducted before that. At most these two polls might be thought to be supportive of the suggestion that there has been a slight easing in support for independence during the course of this year.
(There were, by the way, reports of a third poll on independence in some papers earlier this month. These were based on an ICM poll for the think tank, British Future, which in fact was conducted last June – immediately after the general election – across Britain as a whole, though with a boosted 1,052 sample in Scotland. It found that just 41% support for Yes (after Don’t Knows were excluded) but it should be noted that respondents were not asked how they would vote in response to the 2014 independence referendum question, ‘Should Scotland be an independent country?’, but, rather, in a referendum on whether Scotland ‘becomes independent from the UK’.)
What, though, is quite clear is that, even though as many as 65% say (according to Panelbase) that they would vote Remain in another EU referendum (slightly up on the 62% that did so in June 2016), Brexit is continuing to fail to become a recruiting sergeant for independence as the SNP might have hoped and Nicola Sturgeon seemed once to have anticipated. Indeed, support for the idea of holding a second independence referendum on the back of Brexit seems to have faded yet further. Just 17% now think there should be one before the Brexit negotiations are concluded, down from 22% in June and 32% in April. It is perhaps little wonder then that, according to an interview published yesterday, even the First Minister is now no longer sure that an early indyref2 would necessarily be a good idea. Doubtless she will be keeping her eyes on the polls.
About the author
John Curtice is Professor of Politics at Strathclyde University, Senior Research Fellow at ScotCen and at 'UK in a Changing Europe', and Chief Commentator on the What Scotland Thinks website.
