Today was another relatively quiet day, before two days of all out, rapid fire polling. The one national poll was YouGov/Sun, which followed up days of flipping between one-point Labour and Tory leads with a tie, and the best voting intention for the Lib Dems with YouGov for exactly a month:
YouGov/Sun: CON 33 (-1) LAB 33 (=) LIB 10 (+1) UKIP 12 (=) GRN 5 (=) Fieldwork 3rd-4th Daily roundup to come #GE2015 #ELECTION2015
— NumbrCrunchrPolitics (@NCPoliticsUK) May 4, 2015
Much more dramatic was the ICM/Guardian poll of Sheffield Hallam:
ICM/Guardian (Sheffield Hallam – named candidates, chg vs 2010): CON 12 (-12) LAB 35 (+19) LIB 42 (-11) UKIP 7 (+5) GRN 3 (+1) #GE2015
— NumbrCrunchrPolitics (@NCPoliticsUK) May 4, 2015
ICM/Guardian (Sheffield Hallam – no names, chg vs 2010): CON 21 (-3) LAB 34 (+18) LIB 32 (-21) UKIP 8 (+6) GRN 4 (+2) #GE2015
— NumbrCrunchrPolitics (@NCPoliticsUK) May 4, 2015
ICM/Guardian tables out. Interestingly, the constituency Q was asked first, national Q second Dates 1st-3rd N=501 http://t.co/DiTjDsTDRz
— NumbrCrunchrPolitics (@NCPoliticsUK) May 4, 2015
As Tom Clark made clear in the Guardian's writeup, there were clear signs of a tactical swing of just about 10% from the Conservatives to the Lib Dems to support Clegg, which accounts for entire difference between the two Questions. This echos the pattern Lord Ashcroft found in this, the only non-Labour held seat in South Yorkshire. But the lead was bigger and Twitter was immediately back into the debate about whether-or-not to name candidates.
The fact is, we still don’t really have much evidence, other than that it may make a difference where a very prominent candidate is named. This isn’t to say that naming is more accurate either – the logic of that argument is that making a poll more like the ballot paper inherently increases its accuracy. However that has been shown to be untrue in some other contexts (for example, prompting for minor parties) so without any public domain data to back it up, the uncertainty remains. See also Anthony Wells’s thoughts here and Chris Hanretty’s discussion of some of the issues here.
Youth abstention is a relatively recent trend. In 1992, 63% of under 25s voted (among all ages it was 78%) https://t.co/FKqQcywtXC #GE2015
— NumbrCrunchrPolitics (@NCPoliticsUK) May 4, 2015
And finally:
Nice to see @allstarlanes getting in on the election… But which is selling the best?! HT @DamianSurvation pic.twitter.com/cM6Lo9uxzl #GE2015
— NumbrCrunchrPolitics (@NCPoliticsUK) May 4, 2015
Tomorrow we get final calls from Populus/FT online and Lord Ashcroft, plus the penultimate daily poll from YouGov for the Sun. As far as I’m aware, all other final call polls will be on Wednesday apart from MORI on Thursday morning.
