Talk:Leicester urban area

Latest comment: 7 years ago by The Equalizer in topic 2011 Data

Towns and Villages included

edit

This section is a bit of a mess. Beaumont Leys was in the first list although it states later that it isn't included (it's an estate in Leicester, so that makes no sense). And why is there a list of towns and villages, and then another list afterwards? The main list now matches the ONS list. Are the others based on refs, or just someone's opinion?--Michig 13:52, 25 March 2007 (UTC)Reply

In the absence of any refs, I have removed the second list of areas/towns.--Michig 12:58, 24 June 2007 (UTC)Reply

Other towns and villages - "Leicester" vs "Leicester UA" (2001 Census)

edit

Just dropping some rather useless original research here. For those unaware, the ONS have two popuation figures for Leicester - "Leicester" (336,153) and "Leicester UA" (279,921). "Leicester" (336,153) is used for the Leicester Urban Area stats, while "Leicester UA" (279,921) covers the unitary authority and actual city boundaries.

My issue with the defined Leicester Urban Area is the omission of several very prominent, highly urbanized villages/towns on the edge of Leicester (while including Queniborough and Cosby...). It's clear the difference of 56,232 between the two Leicester figures encompasses these settlements that aren't included separately in the ONS's Leicester Urban Area list. I've determined that the villages below are included in the "Leicester" (336,153) figure and are therefore also part of the Leicester Urban Area:

Total = 61,646. Subtracted from "Leicester" (336,153) = 274,507...adding the ONS's "Beaumont Leys" figure of 5,579 (who knows why they separated it) then gives us 280,086, a very close difference of 165 from the "Leicester UA" (279,921) figure!

I came up with the list by looking at ONS maps[1] and Google Earth satellite images. The ONS maps usefully highlight the Leicester Urban Area in red, and it clearly also highlights settlements outside of the Leicester UA boundaries that are not already on the main article here. All the towns and villages mentioned here and on the main article are highlighted, and none are left out. The only sticking point is Leicester Forest East, which is highlighted on the map, but isn't included in my list. This is because the ONS oddly reports the population of nearby Kirby Muxloe (already part of the Leicester Urban Area) as 9,500, while Leics County Council reports a population of only 4,523...I'm assuming the LFE population is included (at least in part) in the ONS's figure for Kirby Muxloe.

Would be nice if this could be included in the article, as the above villages are all very much suburbs of Leicester and should be noted as such, but alas, no sources or references, just speculation... Outrune (talk) 12:25, 30 October 2011 (UTC)Reply

The list of those villages included in the urban area is included in the ONS documentation somewhere, so there's no need to guess at what it is. Anstey is not in the list and is not a suburb of Leicester. There appears to be little consistency in the way the ONS defined these urban areas, meaning that relatively close-lying settlements in Leicestershire are omitted while relatively far areas from Nottingham get included in the Nottingham one for instance.--Michig (talk) 12:34, 30 October 2011 (UTC)Reply
I believe the list currently in the article is the one used by the ONS.--Michig (talk) 12:39, 30 October 2011 (UTC)Reply
The list in the article is correct, but it doesn't tally with the ONS Urban Area map I linked to above, which also highlights villages such as Thurmaston, Syston and Anstey (although they aren't specifically labelled). Do you agree with my reasoning behind the ONS's two conflicting Leicester population figures (336,153 vs 274,507), or is there another reason for it that I'm not aware of? Outrune (talk) 13:42, 30 October 2011 (UTC)Reply
Thurmaston and Syston are highlighted on the map, as are Oadby and Wigston, which are not on your list and between them make a significant contribution to the figure. Anstey is labelled but not highlighted - it is on the other side of the A46 which forms part of the boundary of the highlighted area.--Michig (talk) 14:16, 30 October 2011 (UTC)Reply
You're right about Anstey, my bad. I didn't include Oadby and Wigston on my list, as they are already on the list in the article - they contribute to the greater Leicester Urban Area population figure of 441,213, rather than the Leicester figure of 336,153. The Leicester Urban Area section in this ONS spreadsheet lists Oadby, Wigston and all the other villages on the list (including Beaumont Leys...) separately to "Leicester". Outrune (talk) 15:14, 30 October 2011 (UTC)Reply
Six years later...
I've added the suburb figures for which I see there's been some lively discussion on, and wanted to explain how I totted those up - I wrote up notes for the 2011 figures I compiled in the next section, and the methods for obtaining the 2001 data were pretty similar - while using parish counts as mooted by Mr Outrune is on the right track, but using ONS census output area (OA) and enumeration district (1991 and earlier) figures and a bit of thrashing about with Excel and a calculator are really the key to cracking the numbers.
Specific OAs within a parish/district/UA are used if they contain some population of the overall urban area - The city, Glenfield and LFE don't use all their OAs to cover their parts of the urban area - so lets work out which portion of their totals do:
  • Glenfield - parish population in 2001 is 9,721 - minus the 2 same OAs in 'The Brantings' area as 2011 (see that section below) which is separated to the west:
E00130033 (pop. 304) & E00130035 (224) - total 528
Remainder of parish is 9,193.
  • Leicester Forest East as mentioned in the notes section of the article consists of 20 OAs - the parish portion east of the M1 is counted as part of the Leics subdivision urban area, it consists of five OAs:
E00130078 (272), E00130085 (267), E00130087 (316), E00130090 (344), E00130091 (320) - total 1,519.
The western remainder (15 OAs) as hinted to by Outrune above gets counted within Kirby Muxloe's subdivision figure (LFE 6,496 - 1,519 + KM 4,523 = 9,500)
For comparison the 2 parishes together are LFE 6,496 + KM 4,523 = 11,019.
  • Again as 2011, Ashton Green is subtracted from the city figure, so too is Beaumont Leys - which needs some explaining. Only the OAs (18 in total) hemmed in to the north of the A50-Glenfrith Way/Krenfeld Way/Beaumont Leys Way corridor count towards the Beaumont Leys figure of 5,579, while the full Beaumont Leys council ward itself is 13,000+ persons. This implies there was some separation between this part of B-Leys and the south of those roads at the time for the ONS to class it as a subdivision, the shopping centre, Walkers factory/industrial area, Beaumont Park and extensive forestry certainly contribute to this division.
So to clarify, the B-Leys subdivision is within the ward - but is not the whole ward which includes Bradgate Heights and Ashton Green - and is not the whole suburb, which extends south of the aforementioned corridors but doesn't include Bradgate Heights or Ashton Green. Bear in mind that Ashton Green is north of and separate to the subdivision.
The other city surrounding parishes' OAs are all completely allocated to the Leics subdivision - Thurmaston, Glen Parva, Braunstone, Thurnby and Bushby, Scraptoft, and Syston so those parish totals can be used.
Thought it better to put this detail here as there is no document on the web explaining these specific OA details and how they all come together.
Any further queries about these figures, ask away. Pictures certainly would help visualise a lot of this, and I can provide some.
Regards --The Equalizer (talk) 15:20, 25 July 2017 (UTC)Reply

2011 Data

edit

The short version

I have added suburb figures and the totals absolutely add up to the BUA subdivision amount for 2011. Expand the section below if you want to see how I obtained these numbers, as well as explaining the correcting of figures an IP user added using parish counts and my attempt to show why those aren't always the right geography to use. Main rule is that subdivision figures do not always match the numbers for political/administrative areas as these are not the same in many cases.

Not long enough: will read it all

The figures for BUAs are actually taken from a extrapolation of Census Output Area (OA) figures which 'best-fit' the built up areas, and not just the total residents in a UA or parish (especially in those cases of urban areas on the outskirts of a city/town). There are single digit, tens, hundreds and thousands of these OAs which are added up to make a total for a parish, district, or city/UA area, and indeed, regions and higher still the whole country. OA areas can go way past the edge of the BUA as they don't align with anything other than political borders, but it's the lowest statistical geography used to collate the numbers.

After consulting the ONS census tables I believe all the 2011 data to be correct and adds up to the totals as this edit so of course while I expect the article to be improved after that time, if the figures specifically are changed I would have grave concerns about how any adjusted numbers were obtained.

Lookup table for all BUA areas is here at this link. This lists all OAs and what BUA, as well as district/region they belong to. I have extracted the detail of Leicester and the immediate conurbation to put together the details for the suburbs. There are 1300+ OA's for the city alone never mind for the rest of the BUA, so I won't be listing all those here to prove the figures unless demand warrants it.

I can demonstrate how a total is obtained - using the smallest suburb Scraptoft - the 6 OA codes below make up the total for the parish (1,804) and can be found doing a query on the ONS website:

E00130973 (346 persons)
E00130974 (316)
E00165825 (389)
E00170637 (240)
E00170638 (288)
E00170639 (225)

Even though the parish is mainly rural, the lookup table states all those OAs are within the BUA, so most of the parish population will be abutting the city. The pop counts for each OA must be greater than 100 and/or 40 households to ensure confidentiality of the stat data associated with it so no one person/household can be identified. There is also a minimum size for the OAs, therefore some will be larger than others to get these minimum numbers, and indeed the average is 200-300 residents.

Having said all that, the Leics contiguous suburb figures broadly match what is given for parishes, except for these cases in 2011 (after all, we are counting only directly continuous areas as part of the Leics subdivision area)

  • Glenfield - the parish count is 9,643, however the area regards the Leics urban area in the lookup table does not include the spaced apart 'The Brantings' Glen Park Avenue/Overdale Av estate to the west that abuts Groby parish, cornered by the A46 and A50 roads.
This area has OA codes E00130033 (306 residents) and E00130035 (228) - together there are 534 persons.
Minus the overall Glenfield parish count above gives 9,109.
  • Not a parish of course, but the same for the Ashton Green/Thurcaston Park/Glebelands suburb which again is the only detached area still inside the city limits in 2011 - this has 2 OAs allocated to it:
E00068743 (308 persons)
E00068748 (248)
Total 556 which is in the table and subtracted from the UA total. This means we aren't using the whole city population.

Oadby & Wigston in 2011 make up a district, no parishes - same principle though, all OAs in O&W are within the BUA as the lookup table.

Small anomaly - Stoughton parish has only 1 OA code E00130975 and this covers what I term the Polton Hill 'leg' as it's close to that land feature. This leg is shown as being part of the Leics subdivision on the map - while it is definitely continuous to the city (the 2001 BUA map shows this leg too), the lookup table shows the OA and therefore population (351) is not allocated to the BUA in any way, only to Harborough district.

Why you may ask?

Because according to the methodology guidance doc "ONS 2011 Built-up Areas - Methodology and Guidance" (PDF). for BUA areas, it isn't 'best fit' as there is population weighting applied according to how the population is distributed within the OA. In this case, most of the residents are in the main village to the east - which is too far away from the BUA and in turn skews the 'leg' population and therefore the whole OA and hence parish from being counted in the BUA figure. This weighting is what really determines whether if any OA, particularly ones on the fringes, is close enough to be added to any nearby built up area.


Credit should be given to the IP editor who introduced the concept in the article of Leics (ONS) and Leics (Unitary) as it helps to explain the difference between the urban subdivision area and the official city area, with reference to the tables.

Lastly I need to apologise for using collapsing nested tables within the overall table, while I think it really explains the ONS definition of the subdivision in relation to the city area and outer suburbs, wiki guideline advice is not to use nested unless unavoidable as screen readers can't make sense of them as per Help:Table & MOS:DTAB. I have tested this page using a freely available reader, and the table reads properly collapsed and expanded. I have tried doing two aligned lists but it was not as simple, but will look more deeply into a conversion from table to list if lots feel it is a concern.

I'll work on 2001 figures and update this when I tabulate those. The Equalizer (talk) 08:55, 20 July 2017 (UTC)Reply

Full 2001 figures now added. --The Equalizer (talk) 15:20, 25 July 2017 (UTC)Reply