Talk:Counties of the United Kingdom: Difference between revisions
Jump to navigation
Jump to search
question |
No edit summary |
||
(6 intermediate revisions by 3 users not shown) | |||
Line 1: | Line 1: | ||
How are the | How are the population numbers calculated? --[[User:Inops|Inops]] ([[User talk:Inops|talk]]) 15:23, 27 December 2017 (UTC) | ||
:The Office for National Statistics provide a dataset of each postcode with its population and latitude/longitude figures. The correct county is determined for each postcode by using its co-ordinates to calculate in which of the [https://wikishire.co.uk/lookup/ county border data polygons] it is situated. Once all the postcodes are allocated to a county, it is then a simple matter of summing up all the postcode populations for each county to arrive at a total. [[User:Owain|Owain]] ([[User talk:Owain|talk]]) 21:05, 27 December 2017 (UTC) | |||
::Are there not postcodes that go across county borders though? Did this take long to calculate? I am very thankful for these figures. Around 2014 I tried to do something similar but I had to split the figures up and look very closely into certain border areas because they had been split up so much when it came to comparing the traditional county and administrative area. I seem to remember Cheshire/Lancashire being particularly bad as then you had to get down to a ward level. Some of the county borders have been broken up even at that ward level which meant that I just had to estimate. I seem to remember doing most of England and Wales but gave up when I got to Scotland. | |||
::Anyway I was wondering whether it would be possible to do all the census figures like this from 1801 onward. The reason I ask is because I have been wanting to make a proper comparison of population growth in our country from the early 1800s and obviously because of constant administrative changes it is nigh on impossible to do this. <small>''-- preceding usigned comment added by'' [[User:TheBoldGrenadier|TheBoldGrenadier]]</small> | |||
:::Of course there are border postcodes, but as each postcode only covers a very small area the numbers we are talking about are small enough to be rounding errors. Given that as soon as the census is taken the figures are already out-of-date due to people constantly moving about, it is as close as one can get. The same sort of point-in-polygon allocation of postcode populations to other kinds of census geography is done by the ONS, so it is a proven methodology. The processing of the entire dataset can be done in under a day. Census figures from 1801 to 1911 should be accurate, as they are calculated with the same borders. From 1921 onwards they started using administrative counties, so the ancient counties would need to be recalculated by using civil parishes, or other subdivisions. Postcode populations only became available with the 2011 census, and indeed accurate latitude/longitude information for postcodes since the early part of this century! The [http://visionofbritain.org.uk/ Vision of Britain] web site have massaged population figures from 1801 into all sorts of geographies, which might be useful. [[User:Owain|Owain]] ([[User talk:Owain|talk]]) 23:15, 22 March 2018 (UTC) |
Latest revision as of 09:31, 24 March 2018
How are the population numbers calculated? --Inops (talk) 15:23, 27 December 2017 (UTC)
- The Office for National Statistics provide a dataset of each postcode with its population and latitude/longitude figures. The correct county is determined for each postcode by using its co-ordinates to calculate in which of the county border data polygons it is situated. Once all the postcodes are allocated to a county, it is then a simple matter of summing up all the postcode populations for each county to arrive at a total. Owain (talk) 21:05, 27 December 2017 (UTC)
- Are there not postcodes that go across county borders though? Did this take long to calculate? I am very thankful for these figures. Around 2014 I tried to do something similar but I had to split the figures up and look very closely into certain border areas because they had been split up so much when it came to comparing the traditional county and administrative area. I seem to remember Cheshire/Lancashire being particularly bad as then you had to get down to a ward level. Some of the county borders have been broken up even at that ward level which meant that I just had to estimate. I seem to remember doing most of England and Wales but gave up when I got to Scotland.
- Anyway I was wondering whether it would be possible to do all the census figures like this from 1801 onward. The reason I ask is because I have been wanting to make a proper comparison of population growth in our country from the early 1800s and obviously because of constant administrative changes it is nigh on impossible to do this. -- preceding usigned comment added by TheBoldGrenadier
- Of course there are border postcodes, but as each postcode only covers a very small area the numbers we are talking about are small enough to be rounding errors. Given that as soon as the census is taken the figures are already out-of-date due to people constantly moving about, it is as close as one can get. The same sort of point-in-polygon allocation of postcode populations to other kinds of census geography is done by the ONS, so it is a proven methodology. The processing of the entire dataset can be done in under a day. Census figures from 1801 to 1911 should be accurate, as they are calculated with the same borders. From 1921 onwards they started using administrative counties, so the ancient counties would need to be recalculated by using civil parishes, or other subdivisions. Postcode populations only became available with the 2011 census, and indeed accurate latitude/longitude information for postcodes since the early part of this century! The Vision of Britain web site have massaged population figures from 1801 into all sorts of geographies, which might be useful. Owain (talk) 23:15, 22 March 2018 (UTC)