PVI for the UK

[Deleted]

Inspired in part by jamesdoyle's work with by-elections, I've tried to come up with a way of analysing how far to the left or right of the country a particular seat has voted in any given election. Hopefully this is more useful than "OMG Blyth voted Tory". Obviously a US-style PVI wouldn't really work in the UK but I've come up with something that I think is similar. I'd like to know what you all think.

1) Calculate votes and percentages for the are you're comparing and the area you're comparing to (usually constituency country but regions, counties, wards, nations etc. will also work)
2) Find the Conservative's percentage majority over Labour (or vice versa - some numbers will be negative)
3) Calculate the difference between the majority in the area you're comparing to and the area you're looking at (e.g. constituency majority - national PV lead)
4) Divide the difference between the two by the national PV lead (I'm not sure if this is actually necessary but the results look less hysterical than simply cutting the process off at three. Plus I think "4 time more left/right" is easier to understand)
5) Positive numbers equal right of the country, negative ones left.

So, for example, looking at the last GE in Stroud:

1) Calculate votes and percentages for the are you're comparing and the area you're comparing to
Stroud: Con 47.9% Lab 42.1%.
GB: Con 47.2% Lab 33.9%

2) Find the Conservative's percentage majority over Labour (or vice versa - some numbers will be negative)
Con majority in Stroud = 5.8%; Con majority in GB = 13.3%

3) Calculate the difference between the majority in the area you're comparing to and the area you're looking at (e.g. constituency majority - national PV lead)
5.8 (Stroud majority)-13.3 (PV lead) = -7.5


4) Divide the difference between the two by the national PV lead 
-7.5/13.3 = -0.42

5) Positive numbers equal right of the country, negative ones left
So the "PVI" is -0.42 or slightly to the left of GB as a whole. For context the numbers range from -7.33 in Liverpool, Walton to 4.3 in South Holland & the Deepings. The closest to the national average were Wolverhampton North East and Preseli Pembrokeshire, both at 0.1. Lanark & Hamilton East came up with the same figure but I think that just means it has problems in Scotland.

So what do you think? Useful? Waste of time? Sound idea but flawed execution? I think most of the results look reasonable, areas with strong LD performances look a little understated but not hugely. The SNP in Scotland have thrown a spanner in the works in several seats but I'm not sure how to get around that

[jamesdoyle]

Jul 2, 2022 15:12:49 GMT @europeanlefty said:

Inspired in part by jamesdoyle's work with by-elections, I've tried to come up with a way of analysing how far to the left or right of the country a particular seat has voted in any given election. Hopefully this is more useful than "OMG Blyth voted Tory". Obviously a US-style PVI wouldn't really work in the UK but I've come up with something that I think is similar. I'd like to know what you all think.

1) Calculate votes and percentages for the are you're comparing and the area you're comparing to (usually constituency country but regions, counties, wards, nations etc. will also work)
2) Find the Conservative's percentage majority over Labour (or vice versa - some numbers will be negative)
3) Calculate the difference between the majority in the area you're comparing to and the area you're looking at (e.g. constituency majority - national PV lead)
4) Divide the difference between the two by the national PV lead (I'm not sure if this is actually necessary but the results look less hysterical than simply cutting the process off at three. Plus I think "4 time more left/right" is easier to understand)
5) Positive numbers equal right of the country, negative ones left.

So, for example, looking at the last GE in Stroud:

1) Calculate votes and percentages for the are you're comparing and the area you're comparing to
Stroud: Con 47.9% Lab 42.1%.
GB: Con 47.2% Lab 33.9%

2) Find the Conservative's percentage majority over Labour (or vice versa - some numbers will be negative)
Con majority in Stroud = 5.8%; Con majority in GB = 13.3%

3) Calculate the difference between the majority in the area you're comparing to and the area you're looking at (e.g. constituency majority - national PV lead)
5.8 (Stroud majority)-13.3 (PV lead) = -7.5


4) Divide the difference between the two by the national PV lead 
-7.5/13.3 = -0.42

5) Positive numbers equal right of the country, negative ones left
So the "PVI" is -0.42 or slightly to the left of GB as a whole. For context the numbers range from -7.33 in Liverpool, Walton to 4.3 in South Holland & the Deepings. The closest to the national average were Wolverhampton North East and Preseli Pembrokeshire, both at 0.1. Lanark & Hamilton East came up with the same figure but I think that just means it has problems in Scotland.

So what do you think? Useful? Waste of time? Sound idea but flawed execution? I think most of the results look reasonable, areas with strong LD performances look a little understated but not hugely. The SNP in Scotland have thrown a spanner in the works in several seats but I'm not sure how to get around that

Interesting, and much needed. Will need to read and ponder on (not a job for a Saturday night!), but will get to it asap.

[clyde1998]

Jul 2, 2022 15:12:49 GMT @europeanlefty said:

So what do you think? Useful? Waste of time? Sound idea but flawed execution? I think most of the results look reasonable, areas with strong LD performances look a little understated but not hugely. The SNP in Scotland have thrown a spanner in the works in several seats but I'm not sure how to get around that

I'd recommend having it over the previous three elections to minimise the risk of a unique outcome affecting the results too much.

A possible solution to the third party problem could be to have weights for each party; for example: the Lib Dems being considered to the left, but less so than Labour.

As a proof of concept for for Stroud:

	2015	2017	2019	Ave	Adjust	Wght	Adj*W	Notes
Con	45.7%	45.9%	47.9%	46.5%	46.0%	1	23.0%
Lab	37.7%	47.0%	42.1%	42.3%	41.8%	-1	-20.9%
LDm	3.4%	3.2%	-	3.3%	3.3%	-0.25	-0.8%	Would need to work out how to best deal with elections were one of the major parties didn't have a candidate in any one of the elections.
UKIP/Brx	8.0%	1.6%	1.6%	3.7%	3.7%	1.5	2.8%
Grn	4.6%	2.2%	7.5%	4.8%	4.7%	-1.5	-3.5%
SNP/PC	-	-	-	-	-	-1	0.0%
Oth	0.6%	0.1%	0.9%	0.5%	0.5%	0	0.0%
Sum	100.0%	100.0%	100.0%	101.1%	100.0%		0.51%

The figure of 0.51% is then used as the majority figure in your calculation and continue as before. This gives a score of -0.58, suggesting it's slightly more left than average.

Using this method, South Holland & the Deepings scores 14.3 and Liverpool Walton received -15.9.

The weights for each party are simply guides; I could have a look at the British Election Study to work out the average left-right position of voters for each party to get a clearer picture. The figures I've used will invariably skew the figures left due to the totals of the weights being -0.25 in England or -1.25 in Scotland and Wales.

[Deleted]

Jul 2, 2022 20:42:44 GMT clyde1998 said:

Jul 2, 2022 15:12:49 GMT @europeanlefty said:

So what do you think? Useful? Waste of time? Sound idea but flawed execution? I think most of the results look reasonable, areas with strong LD performances look a little understated but not hugely. The SNP in Scotland have thrown a spanner in the works in several seats but I'm not sure how to get around that

I'd recommend having it over the previous three elections to minimise the risk of a unique outcome affecting the results too much.

A possible solution to the third party problem could be to have weights for each party; for example: the Lib Dems being considered to the left, but less so than Labour.

As a proof of concept for for Stroud:

	2015	2017	2019	Ave	Adjust	Wght	Adj*W	Notes
Con	45.7%	45.9%	47.9%	46.5%	46.0%	1	23.0%
Lab	37.7%	47.0%	42.1%	42.3%	41.8%	-1	-20.9%
LDm	3.4%	3.2%	-	3.3%	3.3%	-0.25	-0.8%	Would need to work out how to best deal with elections were one of the major parties didn't have a candidate in any one of the elections.
UKIP/Brx	8.0%	1.6%	1.6%	3.7%	3.7%	1.5	2.8%
Grn	4.6%	2.2%	7.5%	4.8%	4.7%	-1.5	-3.5%
SNP/PC	-	-	-	-	-	-1	0.0%
Oth	0.6%	0.1%	0.9%	0.5%	0.5%	0	0.0%
Sum	100.0%	100.0%	100.0%	101.1%	100.0%		0.51%

The figure of 0.51% is then used as the majority figure in your calculation and continue as before. This gives a score of -0.58, suggesting it's slightly more left than average.

Using this method, South Holland & the Deepings scores 14.3 and Liverpool Walton received -15.9.

The weights for each party are simply guides; I could have a look at the British Election Study to work out the average left-right position of voters for each party to get a clearer picture. The figures I've used will invariably skew the figures left due to the totals of the weights being -0.25 in England or -1.25 in Scotland and Wales.

That's certainly an interesting idea. My only concern is that the weighting is fundamentally arbitrary which I've tried to stay away from as far as possible. I was going to go back to previous elections eventually (I wanted to test the concept first) but I was going to work out each individual index and then average it rather than averaging the numbers. I might do that anyway just to compare with your version

[greenchristian]

If you don't incorporate third party votes you are going to end up with the occasional very odd result in some seats with a strong third party vote. Extend it back to the Blair years, when the Lib Dems were to the left of Labour, and you're going to find most seats that elected Lib Dems being counted as right of centre because the Conservatives were the second party.

That said, I agree that the precise weighting does seem somewhat arbitrary. And with the Lib Dems and their predecessors there are elections when it would seem to make more sense to put them on the right of centre rather than the left (2015 being the most recent). Which could produce some very odd-looking changes in some seats.

[clyde1998]

Using the British Election Study's May 2022 data, I've done an MRP for left-right economic position by constituency with authoritarian-libertarian position currently processing (take about five hours a pop for me to process this on my PC).

The left-right economic position is determined by the average response (Strongly disagree; Disagree; Neither agree nor disagree; Agree; Strongly agree) for the following:

How much do you agree or disagree with the following statements?
Economic position (left-right)

• Government should redistribute income from the better off to those who are less well off
• Big business takes advantage of ordinary people
• Ordinary working people do not get their fair share of the nation’s wealth
• There is one law for the rich and one for the poor
• Management will always try to get the better of employees if it gets the chance

Social position (authoritarian-libertarian)

• Young people today don’t have enough respect for traditional British values
• For some crimes, the death penalty is the most appropriate sentence
• Schools should teach children to obey authority
• Censorship of films and magazines is necessary to uphold moral standards
• People who break the law should be given stiffer sentences
  

The responses are used to put each respondent on a 0-10 scale for both economic and social position. There's a leftward skew on the economic question and an authoritarian skew on the social question; due to this, I'll show the results as standard deviations from the mean.

---

I'll put a spreadsheet up once the social scores are done, but the ten most 'extreme' constituencies on the economic axis are:

Most economically right-wing

1. Esher and Walton (2019: Con) - 2.330 standard deviations more right-wing than average
2. Chesham and Amersham (2019: Con) - 2.146
3. North East Hampshire (2019: Con) - 2.098
4. Mole Valley (2019: Con) - 2.044
5. Beaconsfield (2019: Con) - 2.026
6. Maidenhead (2019: Con) - 1.968
7. Buckingham (2019: Con) - 1.953
8. Wokingham (2019: Con) - 1.896
9. Sevenoaks (2019: Con) - 1.809
10. Ruislip, Northwood and Pinner (2019: Con) - 1.773
    

Most economically left-wing

1. Liverpool, Walton (2019: Lab) - 3.093 standard deviations more left-wing than average
2. Glasgow North East (2019: SNP) - 2.943
3. Liverpool, Riverside (2019: Lab) - 2.897
4. Glasgow East (2019: SNP) - 2.840
5. Knowsley (2019: Lab) - 2.711
6. Glasgow South West (2019: SNP) - 2.641
7. Liverpool, West Derby (2019: Lab) - 2.572
8. Bootle (2019: Lab) - 2.472
9. West Dunbartonshire (2019: SNP) - 2.367
10. Liverpool, Wavertree (2019: Lab) - 2.343
    

[Pete Whitehead]

Great stuff. This is what this forum is all about. Nicely done just ahead of the award ceremony nominations opening

[greatkingrat]

What is the most right-wing Labour seat and the most left-wing Conservative seat?

[clyde1998]

Nov 7, 2022 12:05:56 GMT greatkingrat said:

What is the most right-wing Labour seat and the most left-wing Conservative seat?

For Labour: Ilford North (0.820σ right). A lot of the most right-leaning Labour seats are in London - Harrow West; Brent North; Ealing Central and Action - or have a large amount of students - Sheffield, Hallam; Warwick and Leamington; Reading East. I imagine most of these will be socially liberal.

For the Conservatives: Blackpool South (1.562σ left). Most of the 'lefty' Conservative seats are recent gains from Labour where Brexit was popular - a lot from the West Midlands. The reverse statement applies here, these will be socially conservative.

Interestingly, the average Lib Dem seat is to the right of the average Conservative seat - although that's down to the Lib Dems having fewer seats.

[John Chanin]

I have a simpler version which just lists all seats on a Labour-Conservative dimension, and then calculates a 0-100 score for most/least Conservative seat. I have this back to 1950, and it self adjusts for boundary changes, thus giving a very useful indication of trends over time. Obviously third party votes, particularly when very variable, create disturbance, but they don’t really make much difference.

Using the same example of Stroud, figures are (Labour high)

1950	47.9
1955	44.2
1959	44.0
1964	39.7
1970	32.5
1974	29.5
1979	29.2
1983	28.6
1987	30.7
1992	37.3
1997	36.1
2001	41.6
2005	40.6
2010	53.0
2015	49.7
2019	58.6

See how it swung towards the Conservatives until the 1980s, and there has been a sharp swing the other way since, the seat being on the Labour side of the balance for the first time in 2010. 

[bjornhattan]

Nov 7, 2022 12:28:16 GMT clyde1998 said:

Nov 7, 2022 12:05:56 GMT greatkingrat said:

What is the most right-wing Labour seat and the most left-wing Conservative seat?

For Labour: Ilford North (0.820σ right). A lot of the most right-leaning Labour seats are in London - Harrow West; Brent North; Ealing Central and Action - or have a large amount of students - Sheffield, Hallam; Warwick and Leamington; Reading East. I imagine most of these will be socially liberal.

For the Conservatives: Blackpool South (1.562σ left). Most of the 'lefty' Conservative seats are recent gains from Labour where Brexit was popular - a lot from the West Midlands. The reverse statement applies here, these will be socially conservative.

Interestingly, the average Lib Dem seat is to the right of the average Conservative seat - although that's down to the Lib Dems having fewer seats.

Also, most of the Lib Dem seats have similar characteristics to the right wing Labour seats you identify. Three of their seats are in Outer London like Ilford North or Harrow West; others like Bath or Oxford West and Abingdon have large student populations and will be socially liberal.

[jamie]

Nov 7, 2022 12:28:16 GMT clyde1998 said:

For Labour: Ilford North (0.820σ right). A lot of the most right-leaning Labour seats are in London - Harrow West; Brent North; Ealing Central and Action - or have a large amount of students - Sheffield, Hallam; Warwick and Leamington; Reading East. I imagine most of these will be socially liberal.

From the few you’ve listed I would guess a lot are going to be disproportionately non-white (at least relative to the size of the Labour majority). This would not be entirely surprising as there’s a decent chunk of the ethnic minority vote that isn’t left wing but votes Labour due to Labour seeming more pro-immigrant/the Conservatives a bit racist (a similar pattern can be found among black and formerly Hispanic Democrats in the US). The student seats would also make sense, as polling has previously found a lot of young people aren’t very economically left wing but still have no interest in voting Conservative for social reasons (the opposite is true of pensioners). 

[sirbenjamin]

Nov 7, 2022 11:57:25 GMT clyde1998 said:

• Censorship of films and magazines is necessary to uphold moral standards

Nov 7, 2022 11:57:25 GMT clyde1998 said:

Nov 7, 2022 11:57:25 GMT clyde1998 said:

[greenhert]

Nov 7, 2022 11:57:25 GMT clyde1998 said:

Using the British Election Study's May 2022 data, I've done an MRP for left-right economic position by constituency with authoritarian-libertarian position currently processing (take about five hours a pop for me to process this on my PC).

The left-right economic position is determined by the average response (Strongly disagree; Disagree; Neither agree nor disagree; Agree; Strongly agree) for the following:

How much do you agree or disagree with the following statements?
Economic position (left-right)

• Government should redistribute income from the better off to those who are less well off
• Big business takes advantage of ordinary people
• Ordinary working people do not get their fair share of the nation’s wealth
• There is one law for the rich and one for the poor
• Management will always try to get the better of employees if it gets the chance

Social position (authoritarian-libertarian)

• Young people today don’t have enough respect for traditional British values
• For some crimes, the death penalty is the most appropriate sentence
• Schools should teach children to obey authority
• Censorship of films and magazines is necessary to uphold moral standards
• People who break the law should be given stiffer sentences
  

The responses are used to put each respondent on a 0-10 scale for both economic and social position. There's a leftward skew on the economic question and an authoritarian skew on the social question; due to this, I'll show the results as standard deviations from the mean.

---

I'll put a spreadsheet up once the social scores are done, but the ten most 'extreme' constituencies on the economic axis are:

Most economically right-wing

1. Esher and Walton (2019: Con) - 2.330 standard deviations more right-wing than average
2. Chesham and Amersham (2019: Con) - 2.146
3. North East Hampshire (2019: Con) - 2.098
4. Mole Valley (2019: Con) - 2.044
5. Beaconsfield (2019: Con) - 2.026
6. Maidenhead (2019: Con) - 1.968
7. Buckingham (2019: Con) - 1.953
8. Wokingham (2019: Con) - 1.896
9. Sevenoaks (2019: Con) - 1.809
10. Ruislip, Northwood and Pinner (2019: Con) - 1.773
    

Most economically left-wing

1. Liverpool, Walton (2019: Lab) - 3.093 standard deviations more left-wing than average
2. Glasgow North East (2019: SNP) - 2.943
3. Liverpool, Riverside (2019: Lab) - 2.897
4. Glasgow East (2019: SNP) - 2.840
5. Knowsley (2019: Lab) - 2.711
6. Glasgow South West (2019: SNP) - 2.641
7. Liverpool, West Derby (2019: Lab) - 2.572
8. Bootle (2019: Lab) - 2.472
9. West Dunbartonshire (2019: SNP) - 2.367
10. Liverpool, Wavertree (2019: Lab) - 2.343
    

Most of these had noticeably above average swings to the Liberal Democrats, discounting Beaconsfield where the Lib Dems backed Dominic Grieve and Buckingham where there was no Lib Dem candidate in 2017. As for the most economically left-wing seats, only two of these have recorded a saved Green deposit at any time (Liverpool Riverside did so in 2005, 2015 and 2019; Liverpool Wavertree in 2015). Then again, two of the most economically right-wing seats have recorded a saved Green deposit in the recent past (Chesham & Amersham in 2015 and 2019; Mole Valley in 2015).

[jamesdoyle]

Nov 7, 2022 11:57:25 GMT clyde1998 said:

Using the British Election Study's May 2022 data, I've done an MRP for left-right economic position by constituency with authoritarian-libertarian position currently processing (take about five hours a pop for me to process this on my PC).

The left-right economic position is determined by the average response (Strongly disagree; Disagree; Neither agree nor disagree; Agree; Strongly agree) for the following:

How much do you agree or disagree with the following statements?
Economic position (left-right)

• Government should redistribute income from the better off to those who are less well off
• Big business takes advantage of ordinary people
• Ordinary working people do not get their fair share of the nation’s wealth
• There is one law for the rich and one for the poor
• Management will always try to get the better of employees if it gets the chance

Social position (authoritarian-libertarian)

• Young people today don’t have enough respect for traditional British values
• For some crimes, the death penalty is the most appropriate sentence
• Schools should teach children to obey authority
• Censorship of films and magazines is necessary to uphold moral standards
• People who break the law should be given stiffer sentences
  

The responses are used to put each respondent on a 0-10 scale for both economic and social position. There's a leftward skew on the economic question and an authoritarian skew on the social question; due to this, I'll show the results as standard deviations from the mean.

---

I'll put a spreadsheet up once the social scores are done, but the ten most 'extreme' constituencies on the economic axis are:

Most economically right-wing

1. Esher and Walton (2019: Con) - 2.330 standard deviations more right-wing than average
2. Chesham and Amersham (2019: Con) - 2.146
3. North East Hampshire (2019: Con) - 2.098
4. Mole Valley (2019: Con) - 2.044
5. Beaconsfield (2019: Con) - 2.026
6. Maidenhead (2019: Con) - 1.968
7. Buckingham (2019: Con) - 1.953
8. Wokingham (2019: Con) - 1.896
9. Sevenoaks (2019: Con) - 1.809
10. Ruislip, Northwood and Pinner (2019: Con) - 1.773
    

Most economically left-wing

1. Liverpool, Walton (2019: Lab) - 3.093 standard deviations more left-wing than average
2. Glasgow North East (2019: SNP) - 2.943
3. Liverpool, Riverside (2019: Lab) - 2.897
4. Glasgow East (2019: SNP) - 2.840
5. Knowsley (2019: Lab) - 2.711
6. Glasgow South West (2019: SNP) - 2.641
7. Liverpool, West Derby (2019: Lab) - 2.572
8. Bootle (2019: Lab) - 2.472
9. West Dunbartonshire (2019: SNP) - 2.367
10. Liverpool, Wavertree (2019: Lab) - 2.343
    

Superb.

[clyde1998]

On the social axis:

Most socially conservative

1. Clacton (2019: Con) - 1.694 standard deviations more conservative than average
2. Castle Point (2019: Con) - 1.659
3. Wolverhampton South East (2019: Lab) - 1.619
4. Doncaster North (2019: Lab) - 1.615
5. Ashfield (2019: Con) - 1.605
6. Hartlepool (2019: Lab) - 1.575
7. Barnsley East (2019: Lab) - 1.531
8. Walsall North (2019: Con) - 1.477
9. South Holland and The Deepings (2019: Con) - 1.475
10. Boston and Skegness (2019: Con) - 1.475
    

Most socially liberal

1. Bristol West (2019: Lab) - 4.051 standard deviations more liberal than average
2. Brighton, Pavilion (2019: Grn) - 3.930
3. Islington North (2019: Lab) - 3.150
4. Vauxhall (2019: Lab) - 3.118
5. Sheffield Central (2019: Lab) - 3.062
6. Holborn and St Pancras (2019: Lab) - 3.011
7. Cambridge (2019: Lab) - 3.004
8. Battersea (2019: Lab) - 2.997
9. Islington South and Finsbury (2019: Lab) - 2.974
10. Liverpool, Riverside (2019: Lab) - 2.895
    

Extremes by party

• Con: Clacton (1.694σ conservative) & Cities of London and Westminster (2.713σ liberal).
• Lab: Wolverhampton South East (1.619σ conservative) & Bristol West (4.051σ liberal).
• LDm: Caithness, Sutherland and Easter Ross (0.261σ conservative) & Richmond Park (2.165σ liberal).
• SNP: Glenrothes (0.351σ conservative) & Glasgow North (2.871σ liberal).
  

The spreadsheet with the data is here: link

[Pete Whitehead]

Nov 8, 2022 3:44:45 GMT clyde1998 said:

On the social axis:

Most socially conservative

1. Clacton (2019: Con) - 1.694 standard deviations more conservative than average
2. Castle Point (2019: Con) - 1.659
3. Wolverhampton South East (2019: Lab) - 1.619
4. Doncaster North (2019: Lab) - 1.615
5. Ashfield (2019: Con) - 1.605
6. Hartlepool (2019: Lab) - 1.575
7. Barnsley East (2019: Lab) - 1.531
8. Walsall North (2019: Con) - 1.477
9. South Holland and The Deepings (2019: Con) - 1.475
10. Boston and Skegness (2019: Con) - 1.475
    

Most socially liberal

1. Bristol West (2019: Lab) - 4.051 standard deviations more liberal than average
2. Brighton, Pavilion (2019: Grn) - 3.930
3. Islington North (2019: Lab) - 3.150
4. Vauxhall (2019: Lab) - 3.118
5. Sheffield Central (2019: Lab) - 3.062
6. Holborn and St Pancras (2019: Lab) - 3.011
7. Cambridge (2019: Lab) - 3.004
8. Battersea (2019: Lab) - 2.997
9. Islington South and Finsbury (2019: Lab) - 2.974
10. Liverpool, Riverside (2019: Lab) - 2.895
    

Extremes by party

• Con: Clacton (1.694σ conservative) & Cities of London and Westminster (2.713σ liberal).
• Lab: Wolverhampton South East (1.619σ conservative) & Bristol West (4.051σ liberal).
• LDm: Caithness, Sutherland and Easter Ross (0.261σ conservative) & Richmond Park (2.165σ liberal).
• SNP: Glenrothes (0.351σ conservative) & Glasgow North (2.871σ liberal).
  

The spreadsheet with the data is here: link

I'd like to do some maps of these if you weren't planning to do them yourself?

[clyde1998]

Nov 8, 2022 7:05:17 GMT Pete Whitehead said:

Nov 8, 2022 3:44:45 GMT clyde1998 said:

On the social axis:

Most socially conservative

1. Clacton (2019: Con) - 1.694 standard deviations more conservative than average
2. Castle Point (2019: Con) - 1.659
3. Wolverhampton South East (2019: Lab) - 1.619
4. Doncaster North (2019: Lab) - 1.615
5. Ashfield (2019: Con) - 1.605
6. Hartlepool (2019: Lab) - 1.575
7. Barnsley East (2019: Lab) - 1.531
8. Walsall North (2019: Con) - 1.477
9. South Holland and The Deepings (2019: Con) - 1.475
10. Boston and Skegness (2019: Con) - 1.475
    

Most socially liberal

1. Bristol West (2019: Lab) - 4.051 standard deviations more liberal than average
2. Brighton, Pavilion (2019: Grn) - 3.930
3. Islington North (2019: Lab) - 3.150
4. Vauxhall (2019: Lab) - 3.118
5. Sheffield Central (2019: Lab) - 3.062
6. Holborn and St Pancras (2019: Lab) - 3.011
7. Cambridge (2019: Lab) - 3.004
8. Battersea (2019: Lab) - 2.997
9. Islington South and Finsbury (2019: Lab) - 2.974
10. Liverpool, Riverside (2019: Lab) - 2.895
    

Extremes by party

• Con: Clacton (1.694σ conservative) & Cities of London and Westminster (2.713σ liberal).
• Lab: Wolverhampton South East (1.619σ conservative) & Bristol West (4.051σ liberal).
• LDm: Caithness, Sutherland and Easter Ross (0.261σ conservative) & Richmond Park (2.165σ liberal).
• SNP: Glenrothes (0.351σ conservative) & Glasgow North (2.871σ liberal).
  

The spreadsheet with the data is here: link

I'd like to do some maps of these if you weren't planning to do them yourself?

You can do some maps if you like 👍. The constituencies should already be ordered by ONS ID if that's helpful.

[Pete Whitehead]

Here is the Economic left-right map. This has been one of the most enjoyable and interesting maps I've done for a long while



Much to comment on here but one obviously striking feature is how the 'myth' that Scotland is to the left of the rest of the UK (or especially England) is not a myth afterall, at least on this metric

[greatkingrat]

So Na h-Eileanan an Iar is actually more socially liberal than the average constituency!