
I’ve taken this data from the excellent http://scotgoespop.blogspot.co.uk. You might wonder why I don’t just share it but I want to make a slightly different view of what it means. Like Scot goes Pops’ sharp-as-a-tack James Kelly, I recognise it might be a rogue poll we’re about to look at but I’m more optimistic or naïve than him. We’ll soon find out.
Here’s what the Survation online poll on the 3rd June has for the UK:
Conservatives 40% (-6)
Labour 39% (+5)
Liberal Democrats 8% (n/c)
UKIP 5% (+2)
Wouldn’t that be luverly?
It’s different from the other pollsters but as I’ve already pointed out their methods are often too Tory-friendly. See:
Despite methodology biased against shy, young and less-well-off, Ipsos MORI / STV poll shows ‘Yes’ vote holding up again
After the surprise Tory win in 2015, they’ve been compensating for ‘shyness’ and supposed higher turnout amongst Tories. They’ve also been using the less-reliable landline telephone data collection method which neglects the younger and poorer voters.
I do agree a 1% gap is a bit unlikely but I do believe it’s get uncomfortably close for Theresa.
Now for the Scottish subset:
SNP: 44%,
Labour 28%,
Conservatives 21%,
Liberal Democrats 6%,
UKIP 2%
James wrote ‘Obviously an individual Scottish subsample can’t be regarded as remotely reliable…’
I’m not so sure. The subset will have been scientifically based just like the overall sample. Here’s what I wrote about the Kantar TNS Omnibus poll:
‘It’s only a subset of 108 Scots from a survey of 1200 UK voters but like the overall sample it is considered representative of the electorate so it’s still pretty interesting. Remember the 1200 represent 0.0026% of the roughly 46 million-strong electorate and the Scottish subset of 108 represents 0.0027% of the roughly 4 million-strong electorate. Let’s not quibble over the 0.0001% difference.’
Either way, 44% and a wee Labour recovery taking votes back from the Tories and/or splitting the two Unionist challengers can only help the SNP hold on to nearly all their seats even with a reduced majority.