With the Scottish independence vote on Thursday a number of polls are suggesting that the spread between the Yes and No vote is now in the region of 2-3%, with a slight edge for the No camp. This is a problem for pollsters when the margin of error for these polls is usually between 2-3%.
It looks like Scotland might vote no to independence — in fact that is not what the polls are necessarily saying.
Here’s YouGov’s description of the problem:
A poll of 1,000 people has a margin of error of +/- 3%, a poll of 2,000 people a margin of error of +/- 2%. The smaller the sample, the less precise it is and the wider the margin of error. Strictly speaking, these calculations are based on the assumption that polls are genuine random samples, with every member of the population having an equal chance of being selected. In many cases this isn’t true ‒ polls are carried out by quota sampling, or from panels of volunteers. Even polls done by randomly dialling phone numbers aren’t truly random, as the majority of people decline to take part.
So when we get a poll like the one done for the Sunday Times from Panelbase over the weekend shows the anti-independence No campaign ahead by 51% to 49% for Yes, we need to treat it with caution. Does it really show the pro-union camp holding on to a narrow lead? Might the trend in favour of the Yes campaign over the past month be more important that the overall standings?
The truth is this — we simply don’t know. What is giving media commentators cognitive dissonance is that close polls are telling us only that the vote is simply too close to call. A 3% margin of error means the Panelbase poll could actually be suggesting a narrow lead for the Yes vote, or the two sides neck and neck.
Within this, however, there are some indicators that could prove pivotal to the result. The share of undecided voters as a percentage of the whole appears to falling, and falling fast. Having averaged around 15% of the vote until August the share of Don’t Knows has almost halved over the past month to around 8%.
One of the often-discussed theories was that these undecideds, mostly consisting of women and the under-30s according to polling data, were actually pro-union voters who were reticent about revealing their intentions due to the aggressive campaigning of the independence camp. Others argued that status quo bias, or the tendency of people to stick with what they already have, would mean that undecideds would lean towards the No camp as the referendum approached.
Instead the polls suggest that the biggest beneficiary of the falling numbers of undecideds is the Yes vote. Professor Alan Renwick believes that the pro-independence movement has been able to confound expectations by positioning themselves as the choice for continuity against the austerity and privatization agenda of the Westminster government. Or else, as the No campaign contends, perhaps people are simply not understanding the risks.
All we can say with any confidence at this stage is that the vote is going to be much, much closer than almost anybody predicted a few months ago. With an incredible 97% of eligible voters registered, this one is going down to the wire.
Business Insider Emails & Alerts
Site highlights each day to your inbox.