British Election Study

[clyde1998]

Wave 21 of the British Election Study has finally dropped; conducted in May 2021.

I know these numbers are very out of date now, especially for electoral matters, but can be interesting due to large sample size.

General Election VI (21,600 responses with a voting intention)
Con 45.5% (+0.8 on 2019 election)
Lab 29.8% (-3.1)
Grn 8.4% (+5.6)
LDm 7.0% (-4.9)
SNP 4.7% (+0.7)
RUK 2.2% (+0.2)
PC 0.7% (+0.2)
Oth 1.8% (+0.5)

For context, the average of YouGov's four polls in May 2021 showed: Con 44%, Lab 30%, LDm 7.5%, Grn 7.5%, SNP 5%, RUK 3%, PC 1%, Oth 2%

By area:

	North East	North West	Yorkshire	East Midlands	West Midlands	Eastern	London	South East	South West	Wales	Scotland
Con	45.4% (+7.2)	39.3% (+1.8)	45.3% (+2.2)	55.4% (+0.6)	55.0% (+1.6)	54.1% (-3.1)	34.5% (+2.5)	55.4% (+1.4)	52.5% (-0.3)	36.8% (+0.7)	24.6% (-0.5)
Lab	40.7% (-1.9)	41.5% (-5.0)	34.8% (-4.1)	27.0% (-4.7)	28.7% (-5.2)	25.5% (+1.1)	44.0% (-4.1)	20.9% (-1.2)	21.6% (-1.8)	35.9% (-5.0)	16.4% (-2.2)
Grn	6.0% (+3.6)	7.7% (+5.2)	10.3% (+8.1)	8.7% (+6.1)	8.0% (+5.0)	7.9% (+4.9)	10.0% (+6.9)	10.1% (+6.1)	10.4% (+6.6)	5.5% (+4.5)	3.7% (+2.7)
LDm	3.8% (-3.0)	6.1% (-1.8)	6.2% (-1.9)	4.4% (-3.4)	4.6% (-3.3)	7.9% (-5.5)	7.7% (-7.2)	10.7% (-7.5)	10.5% (-7.7)	4.1% (-1.9)	4.3% (-5.3)
SNP											49.0% (+4.0)
RUK	2.5% (-5.6)	3.2% (-0.7)	2.2% (-3.7)	2.4% (+0.9)	2.0% (+0.7)	2.6% (+2.2)	1.8% (+0.4)	1.8% (+1.5)	3.0% (+2.6)	2.5% (-2.9)	0.9% (+0.4)
PC										14.3% (+4.4)
Oth	1.5% (-0.4)	2.2% (+0.4)	2.1% (+0.3)	2.0% (+0.4)	1.7% (+1.2)	2.1% (+0.5)	2.1% (+1.6)	1.2% (-0.2)	2.0% (+0.5)	0.9% (+0.3)	1.2% (+0.9)
Weighted Base	881	2,268	1,890	1,669	1,829	2,034	2,692	3,070	2,151	1,062	2,054

Putting those figures in the Electoral Calculus regional seat calculator gave:

• Current boundaries: Con 420 (+55), Lab 147 (-56), SNP 59 (+11), PC 4 (nc), Grn 1 (nc), LDm 0 (-11), Brx 0 (nc), Oth 19 (+1)
• New boundaries: Con 412 (+34), Lab 161 (-34), SNP 51 (+3), LDm 5 (-3), PC 2 (nc), Grn 1 (nc), Brx 0 (nc), Oth 18 (nc)

[Pete Whitehead]

Looks like dodgy samples in the East of England

[clyde1998]

Jan 20, 2022 13:25:19 GMT Pete Whitehead said:

Looks like dodgy samples in the East of England

I'll take a look at that. I've already corrected the headline VI figures as I accidently applied analytic weights rather than importance weights, which meant the data was weighted wrong; the regional figures had the right weights applied. On first glance, it looks like it may be a combination of 2019 Conservative voters being slightly more likely to move to undecided than voters of other parties in the region and 2019 non-voters being a lot less likely to break Conservative than the 2019 election result in the area (42% of 2019 non-voters with a Conservative VI compared to 57% who voted Conservative in the East of England in 2019).