 | 'Today's findings run counter to repeated poll leads for the SNP over recent weeks, which put the Nationalists as much as 10 points ahead.' Douglas Fraser, Scottish Political Editor in the Herald, 5 th April 2007. |  |
| 'Today's findings run counter to repeated poll leads for the SNP over recent weeks, which put the Nationalists as much as 10 points ahead.' Douglas Fraser, Scottish Political Editor in the Herald, 5 th April 2007. | |
Labour has edged ahead of the Scottish National Party in the race to control Holyrood, according to an exclusive poll for The Herald.
However, with the race seeming even tighter than before, the survey has also blown it wide open by showing the unusually high proportion of people who are undecided.
Among those saying how they will vote, Labour came in ahead of the SNP by four points on the constituency vote, and two points ahead on the regional choice.
Even with that good news for Labour, the survey showed Jack McConnell trailing behind Alex Salmond by 10 points on who would make the better First Minister. Public opinion on independence remains sceptical, with opposition outweighing support by 56% to 44%.
Today's findings run counter to repeated poll leads for the SNP over recent weeks, which put the Nationalists as much as 10 points ahead. They show the race between the two front runners is getting tighter as voters focus on their options.
The telephone survey of 1000 people by mruk research, one of Britain's biggest independently-owned market research companies, found more than half of those polled throughout Scotland said they do not know how they will vote. Yet there are also strikingly high numbers of people who say they are certain or very likely to cast a ballot.
Among the undecided respondents, more than half said they were certain to vote and two-thirds are either certain or very likely to cast a ballot. With the Labour-SNP gap so tight, that large number of potential voters will probably decide the election.
On voting intentions for regional list MSPs, Labour was ahead by 37% to 35%, meaning both gaps are within the 3% margin of error for polls of 1000 people. The LibDems were on 14% and the Conservatives on 10%.
According to James Mitchell, a professor of politics at Strathclyde University, a calculation of how these might translate into seats would put Labour on 55, up five from their tally of seats at the 2003 election, the SNP would be on 45, up 18, while the LibDems would be up one on 18, the Tories would be down seven on 10 seats, and the Greens would secure only one seat, whereas they won seven in 2003. The poll, however, is national and fails to register local regional differences, so seat projections are less meaningful than voting intentions.
According to Professor Mitchell, an average of five recent surveys by different polling organisations gives the SNP a five-point lead over Labour on the constituency vote, and a three-point lead on the regional vote, with Conservatives on 12% for both votes and the LibDems on 15% and 14%. The Greens average 6% and others, including the socialist parties, are on 7%.
The average projections based on these figures show the closeness of this election campaign, with the SNP on 47 seats to Labour on 44, the LibDems on 19, Tories on 14 and the Greens on five. That would deny a majority to a continued partnership of Labour and LibDem, and an SNP-Lib Dem coalition would have a majority of only three.
Labour is aware that much of its traditional support base is reluctant to vote. The party is concentrating its attack on the SNP in order to stop those people switching. It is particularlyconcerned about the middle-class voter base which began to back Labour over the past 20 years.
While the findings suggest Labour could have closed the gap with its main rival, they also suggest Mr McConnell could be a drag on views of his party. Asked who would make the best First Minister, the Labour leader trails Mr Salmond by 47% to 37%.
However, those totals have stripped out those who do not know or think it should be none of the four main leaders. The "none of the above" option gained more support than Mr McConnell.
Among the decided, 10% chose Tory leader Annabel Goldie as best First Minister, and LibDem leader Nicol Stephen mustered 6%.
The focus on the Labour-SNP battle appears to have squeezed other parties. Setting aside half of those who remain undecided, Labour was ahead by 38% to 34% on voting intentions for the constituency vote. LibDems were on 15% and Conservatives on 10%.
The poll, carried out from March 22 to 27, asked how people would vote in an independence referendum. After discounting 17% who did not know, 46% backed independence and 54% were against, in line with recent polls.
A further question asked about voting intentions for local council elections on May 3. Under a proportional system, voters will rank candidates in order of preference. Those who have decided their first vote gave Labour 40% support, the SNP 34%, LibDems 14% and Tories 10%.
This suggests majorities for Labour in and around Glasgow, but not by much.
All the main parties believe there is everything to play for in the elections following a poll for The Herald which gives Labour a narrow lead over the SNP.
While the SNP expressed surprise at the poll, they are not unhappy with what can be seen as a wake-up call for them and a guard against complacency.
The Herald poll by mruk research gave Labour a 38-32 lead on the constituency vote and a 37-35 lead on the regional vote, but half of those polled had not yet decided how to cast their vote.
First Minister Jack McConnell said: "When Alex Salmond is subject to scrutiny on his policies, when he's put under pressure on his election campaign, he runs away. He's been hiding all week. He doesn't have the substance to back up the soundbites, whereas I have the substance, the record and the vision that will win."
SNP campaign manager Angus Robertson said: "A poll of polls' analysis for the last two months shows the SNP six points ahead on both questions, which is the strongest pre-election position the SNP have ever enjoyed.
"However, the SNP are taking nothing for granted. We will be working extremely hard to achieve success for Scotland on May 3."
LibDem campaign director Tavish Scott said: "The high level of undecided voters in this poll demonstrates that people are fed up with Labour but don't want independence."
A Tory spokesman said: "We know on the doorstep that our support is growing and in the poll of polls we are in a much stronger position than at the same time four years ago."
Labour is four points ahead of the SNP in the constituency vote for the Holyrood elections, a poll for The Herald newspaper has suggested.
Market research company MRUK questioned 1,000 people across Scotland by telephone between 22 and 27 March.
It put Labour is on 38%, the SNP on 34%, the Liberal Democrats on 15%, the Conservatives on 10% and others on 3%.
This bucks the trend of recent polls which have suggested a substantial SNP lead over Labour in constituency votes.
The Herald poll suggested that about 50% of those questioned had not yet decided who would get their vote.
Almost two thirds of the undecided said they were either very likely or certain to cast a ballot.
The poll also put Labour 2% ahead of the SNP on the regional vote, by 37% to 35%, with the Lib Dems on 14%, the Tories on 10%, the Greens on 3% and others on 1%.
Professor James Mitchell of Strathclyde University said the poll findings would translate into 55 seats for Labour, 45 for the SNP, 18 for the LibDems, 10 for the Tories and one for the Greens.
The Herald's Scottish political editor, Douglas Fraser, told BBC Scotland that it was only the second poll this year which had put Labour ahead.
"It possibly is a helpful reality check to the SNP about how close this race is," he said.
The Nationalists' campaign manager, Angus Robertson, said the poll was "well past its sell-by date".
"Three polls conducted at the same time and two after it all show the SNP with a clear lead," he said.
"And a new poll-of-polls analysis for the last two months shows the SNP six points ahead on both questions, which is the strongest pre-election position the SNP have ever enjoyed."
However, a spokesman for Scottish Labour leader Jack McConnell said: "This poll proves that there is still everything to play for.
"We will be using every minute of every day between now and 3 May to contrast our plans to grow the economy by investing in education and young people with the SNP plans to wreck the economy and cost families more by separating Scotland from the rest of the UK."
The poll suggested that 54% would not back independence in a referendum, while 46% would support it.
SNP leader Alex Salmond was six points ahead of Mr McConnell when people were asked who they thought would make the best first minister.
IF THERE is one group of people (apart from political opponents) that must be taking great pleasure from the current unpopularity of the Labour government it is opinion pollsters.
They lost some business during the extraordinarily long honeymoon that Labour enjoyed with the electorate from 1997 onwards. Newspapers and other organisations became unwilling to continue with the relatively expensive business of commissioning regular polls when, especially in Scotland, they told the same story month after month: relentless Labour leads. Some pollsters - notably the venerable Gallup organisation - withdrew from political polling altogether, although others continued as a sort of gesture of charity to "politics junkies".
How things have changed. With Labour apparently on the ropes and the SNP poised to make advances, polls of voting intentions in the Scottish elections are now all the rage. Six have been reported in the last couple of weeks and even the London-based media have started to pay attention to the results. We can expect the flow to continue until polling day on May 3.
Inevitably, party spokespeople put the best possible spin on poll results. "The only poll that matters is on May 3," say those clutching at straws. "This shows our message is getting through," say those who seem to be doing well.
In fact, everyone should treat Scottish polls at this stage of the game with a healthy dose of scepticism.
Firstly, pollsters have a rather poor record as far as Scottish parliament elections are concerned. In 2003, there were four major companies reporting during the campaign - but in their final pre-election polls only one, YouGov, gave a reasonably accurate prediction of the vote distribution in constituency contests. The other three all significantly overestimated Labour and SNP support and underestimated the vote for the Conservatives. The predictions for the list voting were even worse. If even final polls can be so inaccurate, then the pinch of salt to be taken with current polls has to be large.
Secondly, there is still a long way to go. According to British Election Study data, at the 2005 general election, more than a quarter of Scottish voters (27%) said they finally decided which party to vote for during the election campaign itself. If the figure is anything like that this time, then clearly the election is not "all over bar the shouting", as some would have us believe.
Thirdly, the translation from poll figures to seats likely to be won in the Scottish parliament is fiendishly complicated. It has to be assumed that the changes in party support implied by the polls will apply uniformly throughout Scotland. But that is highly unlikely. In the 2003 elections, for example, compared with 1999, the swing between Labour and the SNP ranged from 10% to Labour and 9% to the SNP. In 42 constituencies the Liberal Democrat share of votes increased, while in 31 it fell.
Part of the problem - and the fascination, of course - is the highly complex nature of party competition in Scotland. Currently Labour and the SNP fill the top two spots in 40 constituencies, but there are six different combinations in the other 33 seats.
The other big problem in estimating how the parties will fare in terms of seats relates to list voting. Not only are the polls less than reliable in this respect but, before calculating list winners, assumptions have to be made about constituency winners in the regions concerned. This further increases uncertainty.
Return to home page