TNS BMRB’s latest monthly poll is published this morning. It is in fact relatively old. Interviewing began almost a month ago, and finished as long ago as the 9th of this month. However, that does not mean it should be ignored.
The poll puts Yes on 28%, No on 42%. That represents a one point drop in the Yes vote compared with the company’s previous poll, conducted in early February before the Chancellor’s announcement that an independent Scotland would not be allowed to share the pound. Once the very substantial number of Don’t Knows (30%, up one point) are excluded, that equates to a Yes vote of 40%, down one.
Two implications appear to follow. First, the poll is further confirmation (if any were needed) that the currency intervention has not had a fundamental impact on the referendum race. The one point drop in the Yes vote is statistically insignificant, while the 40% Yes vote is still higher than TNS BMRB recorded in any of its polls last year.
Moreover, although TNS BMRB report that the proportion of people who think that ‘currency’ is the most important issue to them in deciding how to vote has increased from 2% in January to 5% now, it is still only no. 8 in the poll’s list of most popular answers. Undecided voters show no greater interest in the issue than anyone else.
On the other hand, as I suggested when discussing ICM’s latest poll on Sunday, there is as yet insufficient evidence to be able to argue with any reasonable certainty that the Yes side has actually made a further advance in recent weeks. Perhaps the age of this poll means that it has failed to pick up any such movement. But unless and until one or two other polls register a notable increase, we can at most only presume that the Yes side have staved off the currency row and interventions from business, and may not necessarily have positively profited from (or despite) them. Meanwhile, given that the Yes side are clearly still behind – and according to some pollsters such as TNS BMRB still well behind – the Yes side do need to make a lot more progress yet.
The high level of Don’t Knows has been a consistent feature of TNS BMRB’s polls since in August of last year when they began to ask people how they intended to vote in response to the referendum question next September rather than how they would vote if the referendum were tomorrow. Even though September is drawing nearer (and even though the proportion of people who say they are certain to vote has increased to a record high of 71%) the proportion of Don’t Knows in TNS BMRB’s polls shows no sign of diminishing.
Unsurprisingly, those who say they did not vote in 2011 or cannot remember whether they did or not are more likely to say they do not know how they will vote in September. Thus the proportion of Don’t Knows in the company’s polls is bolstered slightly by its unique practice of weighting these two groups so that they match the official abstention rate of 50% – but even the unweighted data in today’s poll show that 28% of TNS BMRB’s respondents did actually say they do not know which way they will vote.
Curiously a not dissimilar change of question wording made by Panelbase in their two most recent polls has not had a similar impact on the level of Don’t Knows they obtain. Their polls are, of course, done over the internet, whereas TNS BMRB’s are (uniquely) conducted face to face. Perhaps the contrast is an indication that Panelbase’s samples contain more people who are interested in politics and thus have firm views, or perhaps it is easier for respondents to say Don’t Know to an interviewer than to choose that option deliberately on a computer screen. Either way, it is a reminder how the differences between the polls are adding to the uncertainty of the referendum race.