Template talk:Largest cities of Somalia

Population of the largest cities

The section "Largest cities" lists the number of inhabitants per city. These numbers cite sources like Jubba-airways.com. I don't believe these figures are correct for the following reasons:

• According to The World Fact Book by the CIA, the number of inhabitants of Mogadishu is 1.3 milion, not 2 million as stated in this wiki article
• If you take the sum of all the population figures in this section, about 74% of the population of Somalia lives in one of the largest cities. That is unlikely, as this article states elsewere that only 37% of the nation's population live in towns and cities. Although this percentage is increasing, it is increasing at a lower rate.
• If you look at a city like Galkayo via Google Maps, it simply does not look like a city with 545,000 inhabitants.

Does anybody know more realistic figures? Countability (talk) 07:53, 1 October 2012 (UTC)

The population figures are all estimates because of a substantial nomadic constituency in many areas, among other factors. The CIA's Mogadishu population estimate is from 2009. It's quite a bit higher than that now due to diasporans and others moving back to participate in the reconstruction process. This has also contributed to an increase in the size of other major cities. Middayexpress (talk) 12:59, 1 October 2012 (UTC)

It probably also has to do with city-definitions etc. I agree with countability that the source is not very reliable. The philipino-cupid commercial (subtitle: dating and singles) on the left of the page does nog give me the impression of a very reliable organization; let alone a local airline. I would also be interested in more more sources (and preferrably; on view of varying city definitions: a single source to use). Anyone? L.tak (talk) 13:25, 1 October 2012 (UTC)

Jubba Airways is one of the main private carriers in Somalia and has been throughout the past decade, including to its hub Mogadishu. I'm also not sure what you mean by city-definitions. Middayexpress (talk) 13:29, 1 October 2012 (UTC)

The Benadir Regional Administration which administers Mogadishu likewise indicates that the city's population is around 2 million [1]. Middayexpress (talk) 13:36, 1 October 2012 (UTC)

But this source doesn't give the impression that it has reliable information either. Their motivation that the number of inhabitants is at least 2 million contains some assumptions, a questional source (GeoNames database) and an inconsistent use of the term "significant". And then there still is the problem that with the sum of all the populations listed in this section, 74% of the population would live in cities which is much higher then the 37% stated elsewhere in the Somalia article Countability (talk) 13:58, 1 October 2012 (UTC)

The CIA explains that the population is in flux because of constant movements, so estimates will vary. The rate of urbanization is also for the country as a whole, not just for the largest cities, and it too is an estimate. GeoNames is a commonly used database for population estimates (c.f. [2], [3]). Middayexpress (talk) 15:21, 1 October 2012 (UTC)

I agree that the rate of urbanisation is for the country as a whole. That means that the figures presented in the template are even less credible: they suggest that the rate of urbanisation is at least 74%. And this makes the inconsistency with the figure of 37% even larger. In addition, I think the discrepancies are to large to blame it on the flux (the flux would be that large that you cannot speak of inhabitents, but of "travelers") . So the problem must be the data. I think these differences are to large to call it an "estimate". It's not even an educated guess, more a wild guess. Countability (talk) 08:42, 2 October 2012 (UTC)

The total urbanization estimate tells us nothing about the actual weighting of that urbanization. The cited 20 largest cities constitute a fraction of the total urban areas in the country. Individually, they could and probably do have much greater urbanization than do other urban areas, but in aggregate terms, this is not apparent. Also, the CIA explains that general population counting in Somalia is complicated by movements of nomads and IDPs. Some estimates are thus evidently only counting the 'core' populations of towns, whereas others include recent migrants from other areas as well. For this reason, the CIA extrapolates its total population estimate from the 1975 census. Middayexpress (talk) 12:33, 2 October 2012 (UTC)

City definitions: it is hard to indicate where a city starts and ends. I can only give examples for my own country where the question whether Amsterdam or Rotterdam is larger also depends if you count the municipality or the agglomeration (which would include the municipality Amsterdam Southeast... These definitions can give rise to major differences and problems with comparing sizes.

And while I believe Jubba is large, it is to my knowledge not an authority for population sizes; it's an airline and therefore not a reliable source for this type of info. As for the commercial site yyou use (this I checked with google chrome commercial), we have no means to see where they have their data from and when it was updated. So again: better sources would be helpful. L.tak (talk) 14:08, 1 October 2012 (UTC)

City boundaries in general are frequently not really clear-cut due to lack of demarcations, so the population data is often just an estimate. Please see above for GeoNames. Middayexpress (talk) 15:21, 1 October 2012 (UTC)

that's exactly my point (about the city boudneries); that could be a reason for the differences! I don't think geonames is very good as a source; can you expand its sources or authority (instead of mentioning an article with 10 edits only). Can you expaline why you think it is a reliable source (maybe some background about the organization behind it; mentioning in secondary sources etc)? L.tak (talk) 22:24, 1 October 2012 (UTC)

I think that you are right that city boundaries are an important part of the definition of "population" and that they do matter. But I think that the descrepancies between urbanisation rate and the figures presented in this template are to large, so this cannot be the only explaination. According to the template, 7447900 persons are living in one of the largest cities. According to the rest of the article on Somalia, the population is 10085638 and 37% of them live in cities. That means that at maximum, 0.37*10085638=3731686 live in cities. So what happened with the other 37% of the population? Are they only living in the countryside in the direct vicinity of the large urban agglomerations? I don't think that this is realistic, so the boundaries issue cannot explain the inconsistancies in data. Nor can any other factors do this, IMHO. Therefore, it must be the quality of the data that is the problem. So what would be a solution? Not presenting any data in the template but adding a phrase that estamations vary? Countability (talk) 08:42, 2 October 2012 (UTC)

Please see above. Middayexpress (talk) 12:33, 2 October 2012 (UTC)

L.tak: The city boundaries demarcation issue applies to most areas in the world. GeoNames also cites its sources on its 'Datasources' page [4]. Its population info comes from World Gazetteer, which culls its data mainly from official sources [5]. Middayexpress (talk) 12:33, 2 October 2012 (UTC)

I agree about the worldwide problem with city boundary demarcation. It's not a reason not to use a certain set of data; but it is a problem if we use different sets of data int he same table.

I am not convincedn about the somalian source of world gazetteer. Indeed they seem to have good data from Canada etc, but I couldn't find any info on the source for Somalia (not even the year). That -in combination with not being a true authority (I have seen no articles claiming world gazetteer is always accurate- a problem; not to speak of the airline-based data... What can we do to improve that situation? L.tak (talk) 19:06, 14 October 2012 (UTC)

World Gazetter indicates that it uses official sources. Where unavailable, secondary sources such as year books, encyclopedias, atlases etc. are used. It's also a resource used by the US government (e.g. [6]). Middayexpress (talk) 02:57, 15 October 2012 (UTC)

That is not a government source; it's a scientific paper of a German group using location data from the gazetteer... I am still not convinced we have sufficient sources here if the source is a bit shaky and the source of the source is unknown (but I am repeating myself). So: what do we do? L.tak (talk) 18:47, 15 October 2012 (UTC)

World Gazetteer is a gazetteer that includes all of the population estimates on one page [7]. It is also one of the most used sources on Wikipedia for this parameter (e.g. [8]). Middayexpress (talk) 20:04, 16 October 2012 (UTC)

Gazetteer or not; it is not clear this one is notable; and hte primary source is not clear. I will place a disputed notice pending resolution of this dispute. What do you suggest for a resolution? Dispute resolution or a third opinion? L.tak (talk) 21:42, 16 October 2012 (UTC)

I've asked on rs/n if World Gazetteer is a reliable source for the city population parameter. The US government does use it, by the way [9]. Middayexpress (talk) 14:58, 17 October 2012 (UTC)

convencience link to the rsn noticeboard: Wikipedia:Reliable_sources/Noticeboard#World_Gazetteer L.tak (talk) 15:15, 17 October 2012 (UTC)

I've adjusted the table, as Tageo uses NIMA records [10]. Middayexpress (talk) 15:15, 19 October 2012 (UTC)

It does use the nima names (and population data?), but it does not specify if it uses it as a source for Somalia-based records. Would it be possible to get them from NIMA to verify?L.tak (talk) 19:00, 19 July 2013 (UTC)

NIMA is the only external source it uses [11]. Middayexpress (talk) 13:48, 20 July 2013 (UTC)

That's not what I read out of that list… Do we have an indication population data are available from NIMA so we can verify some directly? L.tak (talk) 11:13, 21 July 2013 (UTC)

It's from NIMA: "This site provides a geographic coordinate database consisting:- 5.4 million records from the NlMA GE0net Names Server - 2,7 million records of cities - 5000 records of administratives divisions - 188000 records of countries data - 7.9 gigaBytes (7.9 Go) of Land Surface, Shallow Water, and Shaded Topography". There's also World Cities, which exclusively uses government sources [12]. Middayexpress (talk) 12:34, 21 July 2013 (UTC)

I read the link on world cities and tageo, but none indicate that the population data of somalia are stemming from this source. We only know the sites use this sources in general, but not for what purpose… Do we have any sources of NIMA itself saying what type of data they provide so we can cross check? L.tak (talk) 22:12, 21 July 2013 (UTC)

The bibliographical details apply to their respective country databases in general. Tageo: "Geographic Area of Coverage of Tageo:Worldwide excluding Antarctica" [13]; World Cities: "World Cities™ provides detailed educational information for cities around the world. World city and country information is attained from government sources" [14]. Middayexpress (talk) 12:28, 22 July 2013 (UTC)

I believe we are running in circles here; I still think we need some text showing where the Somalia data come from; instead of the very generic text that may apply to some or to all of the data. Shall we discuss this at the reliable sources noticeboard to see how people there look a the NIMA-thing? L.tak (talk) 13:00, 22 July 2013 (UTC)

Both links are quite clear that the text applies to all countries in their respective databases (see underscoring above). Middayexpress (talk) 13:57, 22 July 2013 (UTC)

For me that is not clear at all. "Government sources" are very unclear and unprecize (which census, which year, which government, what type of data). As said before: the other site nowhere states NIMA is their only source (it does not even say that the population data come from NIMA… I bring it to the noticeboards tonight to get some external input… L.tak (talk) 15:56, 22 July 2013 (UTC)

I have raised the issue at Wikipedia:Reliable_sources/Noticeboard#population_data_for_Somalian_cities_sourced_by_Tageo.com_and_Worldcities.us, but I guess you already had an annoying notification in your screen because of that ;-). L.tak (talk) 19:26, 22 July 2013 (UTC)

Good to know :) Middayexpress (talk) 14:07, 23 July 2013 (UTC)

Move discussion in progress

There is a move discussion in progress on Template talk:Largest cities of Acre which affects this page. Please participate on that page and not in this talk page section. Thank you. —RMCD bot 14:33, 1 March 2015 (UTC)

baladweyne, Hiran,Hiran 197.220.90.38 (talk) 21:15, 9 May 2024 (UTC)

Baladweyne, hiiraan, hiiraan 1000000

Baladweyne,hiiraan, hiiraan Population_total:1,100,000 197.220.90.38 (talk) 21:17, 9 May 2024 (UTC)

Thanks for letting us know. Can you provide the source for the information? L.tak (talk) 15:21, 10 May 2024 (UTC)