Talk:List of metropolitan areas in the United Kingdom/Archive 1

Errrrr....where are these figures from?

This page is just completely odd! It does not cite a single source of these population stats.

What on earth is the "Birmingham Metropolitan area"? (or any of them for that matter) - where are we drawing the lines here? Given Greater London is just over 7.5 million people where have the other SIX MILLION people come from?

This Brum link takes us to a disambiguation between either the American city or the West Midlands county (which is quoted at 2.6 million inhabitants). Where has the West Midlands suddenly inherited an extra million people from???? has Tamworth become a boom town we didn't know about?

In contrast other "metropolitan areas" are quoted at the same pop as their Met. county or some other made up figure.

It is simply un-encyclopediac to use one criteria for one place and another for a similar one. Besides, not a single population figure is justified. This article should be massively improved or if not deleted as there are several other much more reliable lists of UK urban areas. Mapmark (talk) 00:46, 5 November 2010 (UTC)[reply]

More importantly the "NUTS 5" European areas to which this article refered to were phased out in 2006 (there are only 3 levels now: http://epp.eurostat.ec.europa.eu/portal/page/portal/nuts_nomenclature/introduction) and those in the UK much more closely rememble the old Met/shire county council areas - see this map for example:

http://epp.eurostat.ec.europa.eu/cache/GISCO/mapjobs2009/0603EN.pdf

It certainly might be of historical interest to retain elements of this page as a list of NUTS 5 areas in the UK giving the figures during that period when they were viable, and if it can be justified by EU data but it certainly should NOT be regarded as a well sourced or practical contemporary guide to UK metropolitan areas, therefore I have placed it up for discussion by fellow wikipedians.

Mapmark (talk) 01:01, 5 November 2010 (UTC)[reply]

These population figures ARE a joke. We, in Greater Manchester area, told constantly, Birmingham and districts around, are only a few thousand more than us!!! — Preceding unsigned comment added by 95.149.183.197 (talk) 13:28, 2 August 2011 (UTC)[reply]

NUTS 5 was NUTS!

No wonder it was abolished! What a mess! Have a look at the figures (http://www.espon.eu/export/sites/default/Documents/Projects/ESPON2006Projects/StudiesScientificSupportProjects/UrbanFunctions/fr-1.4.3_April2007-final.pdf#page=119) for the "Birmingham Metropolitan Area" just as an example (I've no beef with Birmingham, just using it as an example)

It's absolutely comical: "Birmingham 2,363,000 - OK fair enough, that's the entire West Midlands county, possibly minus Coventry or one of the other boroughs
Then it ADDS to that:
966,000 for Wolverhampton
433,000 for Coventry
299,000 for Warwick
78,000 for Dudley
192,000 for Cannock
96,000 for Kidderminster
and 55,000 for Tamworth

Well OK, if you wanna count people two or three times thats all fine, but it certainly is not statistical and its data has no place on Wiki except as an example of bad counting!

966,000 for example was a figure once bandied around for Birmingham City Council adult pop (NOT Wolverhampton!). It MAY refer to the entire pop of the Black Country, though they would have already been counted in the 2.363,000 for "Birmingham" so that's just farcical. (Wolverhampton Met Borough itself as you can see has a pop of around 236,000)
Never heard of Coventry having 433,000 people but let's let that one ride for a moment
But the 78,000 in Dudley were surely part of the 2.3m already counted as in "Birmingham"?
Now we come to Warwick - well we have it as 29,000 on Wiki, so Brussels has just add a nought for good measure! (by the way in case you are wondering, Warwickshire has a pop of over 500,000 - so no connection at all with either figure!

Look at Liverpool metropolitan area - given as 2,241,000 by including all of Merseyside (1.3m) PLUS Warrington, Wigan, Chester and Skelmersdale!

Or Tyneside as almost 1.6m - Tyne and Wear has just over a million granted...but outside the met area there's nothing but sheep

Or "Portsmouth/Southampton" as 1.5 million - well OK you can drive between the two and see urban development but not even Fareham can muster a million extra souls - thats the figure for the entire Hampshire county and then some

These NUTS 5 stats were a result of some really very poor research and have quite rightly been abolished.

IMHO this page needs some very very serious revisions.

Any thoughts? Mapmark (talk) 01:33, 5 November 2010 (UTC)[reply]

I don't quite see what the problem is here.

You say that none of the population figures are cited, but every single figure is taken directly from the document cited, which is a study on exactly this subject by a project set up by the European Union that is both offical and scholarly. The source seems about as reliable as it could possibly be.

Nothing being measured here has anything to do with local authorities, so it is no surprise that the figures given bear no relation to local authority populations.

All of the population figures given in the MUA column are for "Morphological Urban Areas". MUAs are quite similar to the urban areas defined by the Office for National Statistics in the UK census, and many of the figures for these are quite similar. The Birmingham/Wolverhampton MUA population figure of 2,363,000 given is very similar to the population of the West Midlands Urban Area in the 2001 census - 2,284,093. This isn't surprising as the two are very similar ideas - continuous urbanity - but are measured slightly differently - the census defines continuous urbanity by land use, ESPON defines it by population density.

NUTS5 wasn't abolished in 2006, it was renamed LAU2 (see Local administrative unit). Although the names NUTS4 and NUTS5 are still commonly used, strictly speaking it's incorrect, you're right. This minor misuse of terminology hardly renders the entire article pointless though - neither the article nor the source is about these areas (which in the UK are just the same as Wards anyway), the study just uses these very small areas as the "building blocks" that get combined into MUAs and FUAs using the methodology it describes, to produce the metropolitan areas which are the subject of teh article.

Above all, the areas in the left hand column that are the subject of the article are metropolitan areas, which are completely different to urban areas, as the article makes clear. It doesn't matter how much countryside there is between Southampton and Salisbury, if the economic linkages between the two are close enough (which is what the study has studied) then they are in the same metropolitan area. So yes, the Southampton/Portsmouth metropolitan area covers quite a lot of Hampshire, and bits of Wiltshire and West Sussex too, while the Southampton Urban area and the Portsmouth urban area are separate from each other and much smaller. There is no contradiction here - they are completely different geographical concepts.

JimmyGuano (talk) 07:02, 5 November 2010 (UTC)[reply]

I do agree that the Birmingham metropolitan area page - if it needs to exist at all - should not link to West Midlands (county) though - this is highly misleading. Whoever did that has got confued between metropolitan areas and metropolitan counties when they are completely different. JimmyGuano (talk) 07:09, 5 November 2010 (UTC)[reply]

This is by far the best list on this topic. Regarding Greater London, it is a misnomer and outdated. "Greater" areas in other countries expand over time, just as the London Commuter Belt has done as it now stretches far beyond the boundaries of Greater London with millions of people travelling from the Home Counties to London every day. 82.5.217.254 (talk) 13:12, 6 November 2010 (UTC)[reply]

I agree with Mapmark (talk). This data is out of date and isn't used by the European Union anymore. I'm not sure it justifies having a page and think the page should be put up for discussion for deletion. Elland1 (talk) 15:36, 7 November 2010 (UTC)[reply]

Thanks Jimmy, some well made points there and yes indeed there is a distinction between the (old) Metropolitan counties and the MUA's in that years calculations, but IMHO the figures used for the NUTS5 list were just plain simply wrong in some cases!

For example compare the "Birmingham" area to "Manchester" one. The Brum one clearly takes in a HUGE hinterland to add itself up to 3.6m! This is perfectly fair enough, but my point is that whoever compiled these figures made an ERROR by double-counting (or even triple-counting) some people in the West Mids; Wolverhampton is the highest example of this. You cannot have 2.6m W Mids people then ADD almost a million around Wolverhampton who will have already been included in the first figure! It just doesn't make any mathematical or statistical sense Jimmy!
If the MUA is what you say it is them Manchester Met area should hoover up all the SE Lancs towns and most of North Cheshire (Just as the Liverpool area seems to include Warrington and Wigan)- that would clearly give the Manch figure as somewhere over 3 million but then raises the question as to how/why some areas (Wigan/Warrington etc) would be counted twice - once in Liverpool, and another in Manchester. This is why I'm questioning the figures in the NUTS5 report itself. My contention is that they may be flawed and I hypothesise that this is why I guess they were re-named AND their method of calculation was changed. If you look at the current figures and maps quoted above the old Met Counties have been followed much more closely and in the case of the Southampton/Portsmouth one, it does appear to resemble Hampshire (with no salisbury or West Sussex places added, though we could do with seeing a detailed breakdown).

Like I say above this article is perfectly fine as a historical reflection of one way of counting people in a snapshot few years in the mid 2000's, but it simply is not a fair reflection of the current situation and this should be made very clear in the article. I will prepare a new article soon on the current NUTS2 areas as they relate to the UK and you can do what you want this one but it just isn't very useful as a description of the way things are done today.Mapmark (talk) 12:59, 8 November 2010 (UTC)[reply]

Mark did you actually read the PDF you quoted? Because the stats and arguments you make don't tally with the information presented in the report. For instance you site 966,000 attributed to Wolverhampton (double counted as you say), when it is clearly allocated to birmingham, Wolverhampton is allocated the figure of 433,000. No double counting.

Secondly the Portsmouth-Southampton figure is accurate too, it does not represent hampshire as clearly the areas of the north east of the county are added to London metropolitan area (I take it you didn't know Aldershot and basingstoke are in hampshire). It clearly takes in much of the rest of Hampshire, the isle of wight, Western West sussex (where bognor regis is located) and south eastern Wiltshire (where Salisbury is located). —Preceding unsigned comment added by 90.204.234.182 (talk) 20:00, 9 November 2010 (UTC)[reply]

Yep, OK thanks Unsigned - I've looked again at the report (http://www.espon.eu/export/sites/default/Documents/Projects/ESPON2006Projects/StudiesScientificSupportProjects/UrbanFunctions/fr-1.4.3_April2007-final.pdf#page=119) and it's possible that you are right that the "966,000" refers to Birmingham CC NOT Wolverhapmton - I humbly extend my sincere apologies!

However, if you examine these pages (page 118 onwards) it is EXTREMELY messy and open to a great deal of misinterpretation. I apologise unreservedly if I have misread it. BUT my substantive points REMAIN!

First of all, this old NUTS5 way of understaning urban conglomerations in the UK are at best misleading and at worst, deeply erroneous (hence they have changed the way they calculate these figures)

Second, it just doesnt add up that the "Birmingham Metropolitan Area': is over 3.6 million - where the hell does this figure come from? It is NOT explained within this report (unless I have made another error - I would be humble enough to accept that is the case IF there is a full explanation of the calculation of this 3.6m figure. Can you find it?) and the same is true of other Metropolitan areas which are given as smaller than their old Met County pops. It is NOT explained in the report, and is not a generally accepted figure of a MUA or any other measure of West Midlands or Birmingham population, no matter where you arbitrarily draw the line! We have to accept - as the EU does - that they made some errors in this report. HENCE my belief that this table is not accurate or relevant anymore.

All I am arguing is that if you are going to exapand an urban area to include the TTW/MUA/wider hiterland, then these critera should be equally applied to ALL examples given. In this table, as disucssed above, they are clearly NOT. Therefore the table is statistically meaningless and its status on WP should be rightly questioned. Mapmark (talk) 00:10, 10 November 2010 (UTC)[reply]

I'm afraid I still don't see what the substance of your objection is.

You still seem to be arguing that the fact that these figures are bigger or smaller than respective counties makes them invalid, but these figures have nothing to do with administrative borders. Some metropolitan areas will be bigger than their main county, some will be smaller, some (such as Portsmouth/Southampton) will include some bits outside the county and exclude bits inside the county. They are completely different measures.

As the poster above has pointed out, there is no double-counting going on in the West Midlands. That column of the table headed "Espon 1.1.1 Population" only contains figures from the previous study for comparison anyway. Those are the figures that have been superceded by this study - which is why they're not included in this Wikipedia article.

You keep criticising "the old NUTS5 way of understanding conglomerations". I'm not quite sure what this is. NUTS5 in the UK is just another word for "Wards", which this study has used as the basic building blocks for conurbations, by including the contiguous wards with a population density above a certain threshold. As previously remarked, most of the results are extremely similar to the Urban Areas defined by the ONS (using a slightly different measure) so I don't see how this is hugely controversial.

The article makes it quite clear that the figures, like most population figures, refer to the year of the last census - 2001. If this makes them too out of date to include in Wikipedia then we'd need to delete most of the demography articles across the whole site.

The bottom line is this, though: whether you or I as wikipedians agree with this report is slightly beside the point - our task is to judge its reliability as a source. This is a scholarly paper by an international group of academic geographers, with a full explnation of its methodology and rationale (which has been summarised in the wikipedia article) and with extensive peer review in chapter 7. As the report points out, while there are several different measures of labour pools around (TTWAs, GEMECA Functional Urban Regions from the mid-90s, LUZs from the Urban Audit) but none of these follow the classic definition of a metropolitan area of being a conglomeration and its commuter hinterland. This report was commissioned by the EU to do precisely this. It is therefore about as reliable a source as we could hope for, and a source whose relevance and authority is makes it the most appropriate basis for this wikipedia article. Obviously if the report is superceded by another equally authoritative one relevant to the specific subject of metropolitan areas then the article would need to be updated to reflect this, but no evidence of this has been produced as part of this discussion.

JimmyGuano (talk) 08:37, 10 November 2010 (UTC)[reply]

Original research

I've deleted the large amount of new content added by User:Eopsid, as it there is nothing to indicate that it is anything other than original research. In particular, there seems to be absolutely no consistent reliable source for what is or isn't included in each metropolitan area: some seem to be based on the source used for the main table (in which case it's not obvious what the sections are adding), some seem to be based on completely different measures (eg the Urban Audit's Larger Urban Zones) and some seem to be based on nothing more than the whim of the editor (eg "this estimate excludes Harrogate which is considered by some to be part of this metropolitan area."). The components of the supposed "Manchester Liverpool Polynuclear Metropolitan Area" seems to be a complete invention - there's certainly nothing I can find in the cited sources to define it, let alone relate it to the ONS's urban area measures, or even give significant evidence of its existence.

The ESPON survey isn't necessarily definitive and if another relatively recent study can be found, from a reliable source, that consistently measures metropolitan areas across the UK using the normal measures (conurbation + hinterland with a given level of commuting into the conurbation), the of course that new data should be added. This content doesn't give any evidence of this.

JimmyGuano (talk) 16:41, 26 November 2011 (UTC)[reply]

Talk:Liverpool–Manchester rivalry indicates that much of this content has also recently been deleted from the Liverpool–Manchester rivalry article on the same basis. This userpage [1] also suggests that this is a personal hobby-horse of the editor concerned. JimmyGuano (talk) 16:47, 26 November 2011 (UTC)[reply]

Some of what I added was just adding up urban areas over 20,000 within the metropolitan areas. What urban areas that were in the metropolitan areas where from the article's main source and the rest where from urban areas inbetween ones considered to be in the metropolitan area. The maps here decided what I included [2].

I tried to also do some in relation to the larger urban zones instead of what is used in the source (split Birmingham into Birmingham and Coventry). Maybe that should have been two seperate parts. The urban areas in the Belfast Metropolitan Area one contained no original research as did some others. Also this urban areas which are part of Metropolitan Areas using the article's main source ended up giving me what I would call contradictions with Leicester and Birmingham Metropolitn Areas in particular with the Nuneaton Urban Area being considered to be in both areas.

Would it be considered Original Research if I were to a similiar table but list all Urban Areas within the Metropolitan Areas according to the main source. Adding one's which aren't listed if they are small within the metropolitan area, near continuous and too small to be mentioned? So this would mean splitting up the Manchester Liverpool Metro Area table, combining the Birmingham and Coventry one and putting in the urban areas considered part of the Metropolitan Area by the main source but which I excluded?

Also in the Liverpool Manchester rivalry thing i found numerous sources which identified the possible existence of a Manchester Liverpool Metropolitan Area. Eopsid (talk) 19:43, 26 November 2011 (UTC)[reply]

Where the information is already in the article and cited to the ESPON study - for example the fact that the Newark urban area is within the Nottingham-Derby metropolitan area - then there is nothing to be gained except confusion and ambiguity from repeating it a second time in a different bit of the article. Where the cited source does not make it explicit that a given urban area is part of a given metropolitan area, then trying to guess by looking at a map is very clearly original research, and quite poor original research at that.

A metropolitan area is a precisely-defined concept in economic geography, determined not by how close urban areas are together (which you can tell from a map) but by how great a proportion of the working population of one urban area commutes daily into the other (which you can't tell from a map - you need workplace origin-destination matrices). No delineation or comparison of metropolitan areas can take place without a consistent and systematic analysis of such data, and to be included in Wikipedia such analysis needs to have been published by a reliable source (for example ESPON, but not you or me). If you can find another such study then please do include it, but if you can't please don't try to make up your own. JimmyGuano (talk) 20:54, 26 November 2011 (UTC)[reply]

There really isnt much data on UK Metro areas and the ones that do exist do not go into much detail as to what areas constitute that metropolitan area and even the source used here even says that the UK bit is pretty shoddy compared to the rest because they couldnt find data for 'Polynuclear metropolitan areas'. There also seems to be numerous anomalies and stuff (like certain areas much closer or continuous with one metropolitan area being part of another). But I dont know how the methodology works and probably different organisations use different methodologies. Eopsid (talk) 21:31, 26 November 2011 (UTC)[reply]

World Gazetteer

User:Eopsid's personal campaign to establish a "Liverpool-Manchester Polynuclear Metropolitan Area" has now led to the inclusion of lots of data from World Gazetteer in this article. The reliability of this as a source has been widely scorned in other wikipedia articles (for example here [3], here [4] and here [5]). Its description of its data sources [6] is written in the first person singular, strongly suggesting it is a self-published source written by a single individual without editorial oversight, and one that is happy to include "data from other stats lovers". It reckons here [7] that Liverpool and Manchester were considered a single agglomeration in the official census data of 1991 and 2001, which is blatantly untrue (see the official breakdown from the official source here [8]). It additionally reckons [9] that Liverpool and Manchester between them have added almost a million people to their populations in the last seven years, and gives no source for the current population figure except to state that it's a "calculation", with no explanation of what the calculation was or what data it was performed upon.

This being an extremely weak source, especially as compared to the official and scholarly ESPON work, I've removed it.

JimmyGuano (talk) 21:13, 5 February 2012 (UTC)[reply]

Personally I think both sources are rubbish as well as European Union's Larger Urban Zone. The ONS's travel to work areas are I think the best thing which shows metropolitan areas in the UK but then that splits some conurbations eg West Midlands into a number of areas which if my interpretation of what a metropolitan area is, is correct puts that source into doubt as well (I currently think the metropolitan area is a conurbation and the around which has commuting ties to this conurbation). Also World Gazetteer is used as a source on many other articles see List of most populous metropolitan areas in Pakistan, List of metropolitan areas in Asia, List of metropolitan areas in the Americas, Metropolitan areas of Mexico. The ESPON source even says that it isnt very good at identifying polynuclear metropolitan areas. Eopsid (talk) 22:42, 5 February 2012 (UTC)[reply]

I dont see where World Gazetteer says that the census refers to it as one urban area. The site refers to all metropolitan areas as the geographic entity of agglomeration. I too see it mysterious where the large population growth has come from I think they've decided more places are part of the metropolitan area. What I prefer about World Gazetteer though is if you click cities it breaks down what places are included within a certain metropolitan area which EPSON does but not to the same extent. Eopsid (talk) 22:48, 5 February 2012 (UTC)[reply]

The World Gazetteer describes its Liverpool-Manchester concoction as an "Agglomeration" here [10] as if this was the same thing as a metropolitan area, which is how it describes it here [11]. As you and I both know, an agglomeration is not the same as a metropolitan area, so that makes the source look suspect from the start. In the same place [12] the site suggests that the existence of this "agglomeration" is officially sourced from the census (which is untrue, as can be seen from the official, authoritative ONS source) and even then it gives the latest census figures as a "calculation" with no source whatsoever! So even if it was felt that world gazetteer in general could be used as a tertiary source when no better sources were available, in this case it is failing to even demonstrate an ability to do that. JimmyGuano (talk) 21:53, 7 February 2012 (UTC)[reply]

More talk about it's reliability as a source here and here Eopsid (talk) 08:42, 6 February 2012 (UTC)[reply]

And even more here [13] - the one thing both parties in this dispute agreed on was that world gazetteer was untrustworthy as a source! JimmyGuano (talk) 21:41, 7 February 2012 (UTC)[reply]

Under the criticism of the UK information section of the ESPON source it even says that it is innacurate. To quote from page 118 "However, ESPON 1.1.1 data seems to be very innacurate". Eopsid (talk) 14:22, 6 February 2012 (UTC)[reply]

Which is why this article doesn't use that data but uses the ESPON 1.4.3 data that supersedes it. Though even the earlier data (had it not been superseded) would have been more reliable than World Gazetteer as a source, as it is scholarly, from an official source and gives a clear methodology. JimmyGuano (talk) 21:44, 7 February 2012 (UTC)[reply]

My quote was from ESPON 1.4.3 link on this article that pdf also if you read it gives the ESPON 1.1.1 populations (it says so on the header columns)

I think we should put the reliabilty of World Gazetteer up for debate on Wikipedia:Reliable sources/Noticeboard as you question it's reliabilty (and so do I to a lesser extent) and as I pointed out earlier it's used as a source in a number of articles. It's already beeen up for date before as my links show but that never stopped it from being used and those debates dont seem to have got very far. This link from Princeton seems to support World Gazetteer as a reliable source (this was mentioned in one of my links to an earlier discussion on Reliable Sources/Noticebeard from 2008 which i will link to again for convenience)[14]. Eopsid (talk) 22:12, 7 February 2012 (UTC)[reply]

That's quite funny - the Princeton page is clearly talking about a completely different website - "The World Gazetteer's Web site is owned by the National Imagery and Mapping Agency which maintains a database of foreign geographic feature names." This one [15] I guess, which doesn't seem to exist any more. JimmyGuano (talk) 08:48, 9 February 2012 (UTC)[reply]

Look's like your might be right about that as that Princeton page was last updated in 1998. Eopsid (talk) 12:41, 9 February 2012 (UTC)[reply]

Just browsing Wikipedia and found yet another article which uses World Gazetteer as a source. List of largest cities and second largest cities by country Eopsid (talk) 15:12, 10 February 2012 (UTC)[reply]

I have now put a new section up about the reliability of World Gazetteer up on Wikipedia:Reliable sources/Noticeboard. Eopsid (talk) 13:18, 12 February 2012 (UTC)[reply]

Putting this up on noticeboard unsurprisingly got nowhere (See archive 114 it only got one reply and it doesnt look that relevant) But Citypopulation.de seems to like World Gazetteer and it's used as a source in lot's of places. [16] It refers to it as an international web resource but there are a number of user editable ones on his list so it's not much of a good showing for its reliability.Eopsid (talk) 09:44, 13 March 2012 (UTC)[reply]

It didn't get nowhere. You asked a simple question and got a simple answer - it's not a very reliable source, it has no obvious authority and there's little point using it when there are so many better and more reliable sources available. It didn't turn into a huge convoluted debate because it doesn't need to - the situation is quite clear from looking at the site and reading WP:RS. Can we leave it at that now? This is beginning to look like continuing to ask the question until you get the answer you want. JimmyGuano (talk) 20:11, 13 March 2012 (UTC)[reply]