ICM Poll Shows Biggest Swing Yet

One of the most remarkable features of the referendum campaign polls so far has been their stability from one reading to the next.  Almost every company’s poll has reported more or less the same proportion of people saying they will vote Yes (and No) as did  the last poll conducted by that same company. Now, however, that pattern has finally been broken.

In a poll conducted for Scotland on Sunday and the first nationwide poll to be conducted since Christmas, ICM report that 37% think they will vote Yes this coming September, up five points on its previous poll back in September last year. Conversely, support for No is put at 44%, down five points. Once the Don’t Knows are excluded from the calculation, those figures translate into 46% Yes, 54% No, the highest Yes proportion yet in any poll, other than in a much criticised poll conducted by Panelbase for the SNP in August last year.

This swing comes on the back of four polls conducted between the publication of the Scottish Government’s independence White Paper at the end of November and Christmas that on average pointed to a two point swing in favour of Yes. Now we learn that perhaps that swing may have become rather bigger.

Today’s ICM poll also contains two further sets of statistics to alarm the unionist camp. First, as in last September’s poll, the 19% who said they did not know what they would do in September were asked what they were ‘most likely’ to do. Somewhat less than half (45%) were able to indicate how they thought they would vote, and slightly more said they were more likely to vote Yes (24%) than No (21%). If those two sets of voters are added to the Yes and No tallies, the Yes vote creeps up to 47%.

Second, of the 44% of No voters (up three points on September) who think that in the event of a No vote the Scottish Parliament should become responsible for taxation and welfare benefits, 10% think they would vote Yes if they were to become convinced that such a change would not happen. If they were all to make that switch, the Yes tally would rise to 49%, rendering the referendum too close to call.

There is, however, a caveat to be noted. The swing is entirely confined to those voters aged less than 44. Amongst those 25-44 support for independence (once the Don’t Knows are excluded) is up by six points from 43% to 49%. Amongst those aged 16-24 it has jumped from 24% to 57%. In contrast, the figures for those aged 45 and over are almost exactly the same as last time. Here we should note that the younger the voter the more difficult they are to get to participate in a poll (even when, as in this case, the poll is conducted online).  In this instance ICM were only able to interview half as many 16-24 year olds as they wanted to. Although this deficit has been overcome by upweighting every 16-24 year old in the poll so that they count as two persons rather than one, it means the poll’s estimate of how this group will behave is based on a particularly small sample and thus can be very volatile – as appears to be true in this instance.

Indeed, we might note that if we compare the raw unweighted data in this poll with that in ICM’s previous poll we find that, at 34%, the proportion of Yes voters is exactly the same. Although the weighting procedure used in the earlier poll was much the same as in this one, it reduced rather than, as on this occasion, increased the Yes vote. That is not to say that ICM’s weighting procedure is wrong, but merely that it should be appreciated that the reality of the swing this poll records is dependent on the validity of that procedure.

In any event, if we look under the bonnet of this poll, we find yet again a pattern that was highlighted only last week by findings from the Scottish Social Attitudes survey. Nothing looks set to be more important in determining the outcome of the referendum than who wins and who loses the debate about the economic consequences of independence.

ICM’s poll included three questions about what people thought the consequences of independence would be that also appeared in its previous poll. These were: whether independence would be good or bad for Scotland’s economy, would mean there would bemore or less inequality in Scotland, and whether old age pensions would be higher or lower.

Two key patterns emerge. First, as in September the item that most sharply differentiates Yes voters from their No counterparts is whether they think independence would be good or bad for Scotland.  Second, the item on which opinion has swung most in favour of the arguments put forward by the Yes camp is, yes, you have guessed it, the economy.

The proportion who think that independence would be good for Scotland’s economy is up four points from 31% to 35%, while the proportion that feel it would be bad is down six points from 48% to 42%. That represents a ‘swing’ of five points towards independence having a beneficial effect.

In the case of pensions, the proportion who think they would be higher has increased by four points to 20% while the proportion who think they would lower has fallen by two points to 23%, a swing of three points.

Meanwhile, although the proportion who think there would be less inequality under independence is up four points to 31%, the proportion who believe there would be more has also edged up a point to 21%. That represents a swing of just one and a half points.

That swing on the economy clearly matters. No less than 88% of those who think independence would be good for the economy anticipate voting Yes, while 87% of those who reckon it would be bad say they will vote No.

The equivalent figures for pensions and for inequality are both lower. Only 58% of those who reckon pensions would be higher think they will vote Yes – although 75% of those who think they would be lower anticipate voting No. Meanwhile, 63% who believe independence would herald less inequality say they will vote Yes, while 63% who reckon there would be more inequality state they will vote No.

In responding to the publication of the latest Scottish Social Attitudes results, the SNP Deputy First Minister, Nicola Sturgeon, stated the survey indicated that, ‘when we win the economic argument, we will win the referendum’. Indeed so – if the Yes side can win that argument victory could well be theirs. Presumably, we can now expect much more effort from it on that front.

Avatar photo

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.