Scottish independence: Are referendum polls on the move?
- 27 January 2014
- From the section Scotland politics
Until now, one of the most remarkable features of the opinion polls on how people propose to vote in the referendum on Scottish independence has been their stability.
Almost every company has reported much the same result from one poll to the next.
True, the different companies have often produced results that were different from each other, thereby creating some uncertainty about the exact level of support for "yes" and "no". However, neither camp appeared to have acquired any kind of momentum.
There are now though consistent signs of movement in favour of a "yes" vote, movement that follows on from the publication of the Scottish government's White Paper on independence at the end of November.
Since that unveiling, five nationwide polls of people's inclination to vote "yes" or "no" have been published, four before Christmas and one last weekend. In each case the same company had also previously conducted a poll in September or October, well before the White Paper was unveiled.
Six point swing
Every single one of those polls has reported an increase in the proportion saying "yes" rather than "no" (that is after leaving aside those who say they do not know how they will vote). The increase has not always been a large one - in one instance, for example, the rise was no more than a statistically insignificant one point.
However, in the most recent poll, conducted by ICM, there was no less than a six point swing to "yes". At 46% that poll's estimate of the "yes" vote among those voters that have made up their mind was the highest yet to be recorded by any independently commissioned poll - and put it not that far from the 50% winning post.
On average the five post-White Paper polls have put the "yes" vote at 39%, up three points on the equivalent figure recorded by those same polls in the early autumn.
Inevitably, the most recent poll from ICM has particularly caught the attention of those campaigning for a "yes" vote.
They are hoping it shows that the trickle of support in its favour uncovered by the four polls conducted shortly before Christmas is turning into a steadily flowing stream now that the year of the referendum is upon us.
There is though perhaps reason to exercise a degree of caution in looking at the size if not necessarily the direction of the swing in ICM's poll. The movement was entirely confined to those aged less than 44. The level of support for "yes" among those aged 45 and over was almost exactly the same as it was last time.
The "yes" vote was up six points among those aged 25 to 44. But the really big increase - one of no less than 33 points - was registered amongst those aged 16 to 24. Such an increase might be thought improbably big.
The trouble is all pollsters struggle to get young people to respond to their surveys. As a result they often end up interviewing fewer younger people than they would like, and to try to overcome the deficit have to give more weight to the response of those who they do manage to contact.
That can result in a lot of weight (in ICM's case twice as much) being given to the interviews given by a relatively small number of young people.
Yet because relatively few young people have been interviewed in the first place, a poll's estimate of how young people will vote can easily bump up and down from one survey to the next simply as a result of chance.
Only time - and more polls - will eventually tell us whether ICM's poll has exaggerated the swing to yes or not. In the meantime a campaign that was seemingly at risk of being a little dull has finally generated a degree of excitement.
John Curtice is professor of politics at Strathclyde University, and chief commentator at whatscotlandthinks.org, where details of all of the referendum polls can be found.