A recent YouGov survey for the Fabian Society has made a few headlines this week, and justfiably so because it’s rather more interesting than the usual ones we get. It covers a wide range of topics, with a particular focus on Labour, resulting in an entertaining but ultimately not very useful headline in the Scotsman. (Though if the poll had asked respondents to select characteristics for the other parties too, our guess is that the SNP would still have come out on top.)
Other places have chosen instead to highlight the outcome of a curiously-worded question about independence, showing a 54-30 lead for the No campaign with 16% still undecided, while Lallands Peat Worrier breaks down some of the demographics to his usual fascinating effect. But it’s a derivative of one of those breakdowns that produces an intriguing result.
The survey shows a significant closing of the gap in Westminster voting intentions between Labour and the SNP, largely at the expense of the Lib Dems. While the difference between the big two at the 2010 UK General Election was a hefty 22%, it’s now down to 14%. The poll also broke down support for independence according to those voting intentions.
But we suspect the referendum, which will be run by the Scottish Government, is an issue likely to be considered by the electorate as (like Holyrood elections) essentially a Scottish issue rather than a UK one. So just for fun, we took the figures by voting intention and applied them instead to the votes cast at the last Holyrood election.
For simplicity, we’ve assumed that when push comes to shove, the supporters of each party who are currently Don’t Knows will follow the party line, which makes almost no difference to the final tallies. They come out as follows:
2011 votes: 890,000
89% YES: 792,000
11% NO: 98,000
2011 votes: 577,000
20% YES: 115,000
80% NO: 462,000
2011 votes: 261,000
7% YES: 18,000
93% NO: 243,000
2011 votes: 131,000
22% YES: 29,000
78% NO: 102,000
(We’ve discounted the 87,000 votes of the Greens, because there’s no data in the poll about which way they’ll split and as far as we can tell it’s pretty even, with a narrow Yes majority.)
That’s a wafer-thin 51-49 win for independence, fairly dramatically at odds with the thumping 40-60 defeat (discounting Don’t Knows) actually suggested in the poll. And given the wildly unscientific premise of our calculations it’s not a prediction.
But with trends showing a substantial closing of the Westminster gap between the SNP and Labour, and the surprising revelation that more Labour supporters back independence than SNP voters oppose it (something at odds with every previous survey we’ve seen on the subject), and factoring in a bit of wiggle room as compensation for YouGov’s very odd formulation of the question, we have a feeling they’re a bit closer to reality than the headline figures.
(Remember, these numbers are based on the current Westminster voting intentions, so they already include most of the effect of disgruntled Lib Dems moving to other parties, which the 2010 General Election ones don’t.)
We’re not calling it psephology. It’s just something to think about.