Seven weeks to go until the first results start to come in! Tonight we have a couple of new polls:
YouGov/Sun: CON 35 (+2) LAB 33 (-1) LIB 8 (=) UKIP 13 (-1) GRN 6 (=) Fieldwork 18th-19th Daily roundup to follow shortly #GE2015 #bbcqt
— NumbrCrunchrPolitics (@NCPoliticsUK) March 19, 2015
CORRECTED Survation/Record (Scot Westminster): SNP 47 (+2) CON 16 (+1) LAB 26 (-2) LIB 4 (-1) UKIP 3 (=) GRN 2 (-1) #scotland2015 #bbcqt
— NumbrCrunchrPolitics (@NCPoliticsUK) March 19, 2015
As explained yesterday, tonight’s YouGov/Sun poll will have the impact of the evening news, but it won’t be until their poll for the Sunday Times (out Saturday) that we get the full effect of this morning’s press coverage. Nevertheless, the last few have shown small Labour leads, so the two-point Tory lead looks a lot like it might be budget-related. Whether it’s real and sustained is the next question.
We also have a Survation/Daily Record Scottish poll. Some of the numbers floating around online seem to have been extracted from a Scots poll they did for Unison, rather than the actual Record actual poll – numbers above are the correct ones. So the SNP’s lead stretched to 21 points over Labour, with a more modest gain for the Scots Tories than the 3pts YouGov showed last week. I don’t yet know the fieldwork dates although I believe most of it would have been before the budget.
One point that bears repeating, since it’s frequently overlooked by some that should know better is that opinion polls cover Great Britain only, not the whole UK. Therefore, the only apples-to-apples comparison is with the GB popular vote figures from 2010, which were as follows:
As a reminder the actual 2010 mainland vote shares (comparable to opinion polls) were: CON 36.9 LAB 29.7 LIB 23.6 UKIP 3.2 GRN 1.0 #GE2015
— NumbrCrunchrPolitics (@NCPoliticsUK) March 12, 2015
Some sources mistakenly report the Conservative share as 37.0, rather than 36.9, reportedly because one broadcaster’s website included John Bercow’s votes in the Tory tally. As the Speaker seeking re-election, this is incorrect.
Another thing to be way of is the margin of error, which is often quoted as a single figure, but the reality is far more complication (and, generally speaking, larger) than commonly thought. Oxford’s Tom Lubbock explains here.
Today I also published a piece analysis the latest batch of Ashcroft polls, which look good for Labour on the surface, but suggest we need more evidence that their relative underperformance in English marginals has come to an end. As things now stand, my best guess is that the Tories need about a 1 point lead in the popular vote to be the largest party.
Also worth keeping an eye on if you’re interested in individual seats, a new site called Democratic Dashboard, which looks interesting:
If people could follow @DemocraticDash that would be most pleasing. *Finally* launching next week. Exciting. pic.twitter.com/rEcXoFuhjS
— Carl Cullinane (@CullinaneCarl) March 19, 2015
Tomorrow the only poll we’re expecting is Populus late in the morning. The fieldwork would have straddled the budget, and its not clear how much was before and after.
