At the Indy X event on Sunday past I said at one point that the single biggest marker of someone who isn’t yet voting Yes is confidence. This is quite a bald statement and will benefit from some unpacking – but that will also help show where the weakness in strategic thinking about independence is.
The conversation around independence strategy usually goes the same way; there are people who don’t vote Yes. How can we learn more about them? Let’s put them into groups, find out which groups don’t vote Yes and then by working out what that group is like we can work out why they don’t vote Yes.
And when we group we group in terms of what we can see – age, sex, income, profession, geography, the basic demographic categories that we can instantly recognise by looking at someone or asking them a single question.
Then we track the Yes/No voting patterns in each group to find out how ‘indy’ they are. We then look at the characteristics of the groups and tally that across to their voting history or intentions when it comes to independence. And finally we draw conclusions about the varying likelihood to vote Yes based on the characteristics of the group.
So basically the groups that voted No tend also to have a higher preponderance of voting on the right of the political spectrum, and so we conclude that the problem with independence is that it is too left-wing.
(This has been the go-to approach for the small and unrepresentative right-wing of the independence movement which has used this analysis, flawed as it is, to persuade the independence movement that it must swing to the right via initiatives like the Growth Commission.)
And it is not wrong to do this, to look at these preponderances of why one group votes one way and another group votes another. It does indeed tell you things. What it doesn’t do is tell you what to do to win a future Yes vote in an independence referendum.
Campaign theory points to the conclusion that it is probably easier to bring a No voter who profiles identically to a Yes voter over than it is to bring over a No voter who profiles very differently
The reason can be gathered as follows: referendum on having Christmas narrowly defeated, turkeys vote heavily against – major drive to persuade turkeys of merits of Christmas must now be the priority.
Of course there are groups which are more or less likely to support any given prospectus – that is the nature of humans. And if you see the world via demographic segments it makes the world look like those least likely to support you.
But that is only one form of ‘market segmentation’ and it isn’t the most useful one for the independence movement. Again, you can see why by looking at the following, simple statement:
“Pick any demographic at all and refine until it is as narrowly-defined as possible. The people in this demographic group are now as much alike each other as it is possible to achieve. And yet some of them voted Yes and some of them voted No.”
It is not difficult to find two (say) men, same age, same income spectrum, same category of work, same housing profile, same football team, same region, both urban, same voting pattern in parliamentary elections, same religion, same vote cast in the Brexit referendum – but who voted differently in indyref and continue to tell pollsters they would vote the same way again.
No demographic category voted 100 per cent Yes (or 100 per cent No). Scotland is filled with like-for-like people who profile identically in every way possible but who express different voting intentions on independence. Standard demographic do not tell us why.
But almost all campaign theory points to the conclusion that it is probably easier to bring a No voter who profiles identically to a Yes voter over than it is to bring over a No voter who profiles very differently.
Let me phrase that another way; if you break down into tight demographic categories you might find that ‘rural farmers’ break 80/20 against independence but that ’30-something women with children working in the NHS’ break 60/40 in favour of independence. If you take the first approach, the ‘who looks like they are our biggest opponents’ approach, the solution is to target your campaign at rural farmers.
That is not the only way to look at it. The other way is to ask who are the next ten per cent of 30-something female NHS workers and why didn’t they ‘get across the line’ to support independence? And here’s the thing – the latter are much easier to win over because their demographic characteristic makes them much less likely to be hostile.
This is important; demographic groups are often most likely to work and socialise with others in the same or similar demographic groups. Rural farmers are much more likely to hang out with other rural farmers and opinions can often be self-reinforcing. Nurses are more likely to hang out with nurses who are more likely to vote Yes and so the conditions from easing them away from their current position are more favourable.
Like for like, if two people profile exactly the same in terms of demographics then the biggest single predictor about whether they voted Yes or No is their level of risk aversion, their general sense of confidence in themselves and others
Rather than looking at the demographic groups among which independence does least well and assuming they are the target of our campaign it is better to think about all demographic groups and see ‘the next ten per cent’ of each as our target. It is much more likely to work.
So what is it that is the factor that helps us best understand these voting patterns once we’ve exhausted demographics – i.e. what was the barrier for those next ten per cents?
Luckily work has been done on this by the Scottish Independence Convention (which has done by far the most far-reaching work on public attitudes to independence). There isn’t space here to explain in detail but the SIC tested demographics – but also what is known as ‘psychographics’.
If demographics tell you what part of the social strata people are in, psychographics categorise people by their attitude and emotional outlook. And when you tally these things together there is a fairly stark conclusion – the gap between ‘the last ten per cent to get over the line’ (the softest Yes voters) and ‘the next ten per cent who didn’t get over the line’ (the softest No voters) is best explained by ‘confidence’.
Like for like, if two people profile exactly the same in terms of demographics then the biggest single predictor about whether they voted Yes or No is their level of risk aversion, their general sense of confidence in themselves and others and their sense that they can trust their wider community to adapt to change.
You cannot quickly or easily change someone’s psychographic outlook but you can change your strategies to be cognisant of it, to accommodate it and seek to counterbalance it. We can’t easily make people more fundamentally confident but we can understand that to get them over the line we must make them more confident about independence.
That explains why, since 19 September 2014 (I must admit I suspected this was the case prior to having access to the public attitude research work), I have been emphasising the need to do the things that reassure people who do not have the profile of ‘inherently hostile to independence’ but do have the profile of ‘require more reassurance than average to take a decision they view as a risk’.
In my strong opinion it is that group which is key to a victory for independence and definitely not the profile of the average person in the most hostile demographic groups. Until we understand that and bring a bit more sophistication to our understanding of how to win independence we risk going round and round in circles, failing to make progress.