Wikidata:Property proposal/Per capita income

From Wikidata
Jump to navigation Jump to search

Per capita income[edit]

Originally proposed at Wikidata:Property proposal/Place

Descriptionaverage income earned per person in a given area (city, region, country, etc.) in a specified year.
Representsper capita income (Q45918)
Data typeQuantity
Domainitem
Allowed values>= 0
Allowed unitscurrency units (usually US-Dollars or local currency)
Example 1United States of America (Q30) → $35.384
Example 2Autauga County (Q156168) → $29.804
Example 3Washington, D.C. (Q61) → $58.659
Sourceen:Per capita income
See also

Motivation[edit]

I believe this is a rather straight-forward property request. For the United States, the data is available for every city, county, state and the full country, i.e. >50.000 statements only for the US. Other countries accordingly, based on the data quality that is published from official side. Yellowcard (talk) 11:29, 31 March 2022 (UTC)[reply]

Discussion[edit]

  •  Strong support The income is one of the most important factors to characterize the social structure of a community. As we have the data from the United States Census Bureau in a siutable form to add it here, we should definitely create a new property. Regards, Dionysos1988 (talk) 13:30, 31 March 2022 (UTC)[reply]
  •  Strong oppose This data changes from year to year. Numeric changeable statistics such as this do not belong to Wikidata. Keep them in Commons datafiles please.Vojtěch Dostál (talk) 16:06, 31 March 2022 (UTC)[reply]
What do you mean with "year to year"? The data is raised just every ten years in the course of the decennial United States Census. This is a very long time period, so WD is not to become overloaded with data at all. Where does WD say that there must not be stored "changeable statistics"? As far as I can see, it is very common to store numbers of inhabitants, for example, which is changeable the same way. We have the property P585, exactly to define these changeable things. Regards, Dionysos1988 (talk) 18:45, 31 March 2022 (UTC)[reply]
@Dionysos1988 There are actually other countries than USA and some provide year-to-year statistics as far as I can see. Also, it can be stated in any of the world currencies - each year, you may have sources to state the per capita income of various places in a local currency, in Euro, in Dollars, etc. So multiply the number of year data points by three. Method of determination, either nominal or PPP, various types of estimates, ..., will multiply the data points by a factor of 2-4 at least. Then you can have various sources with slighly different numbers, which further increases the number of data points per year. This data inflation is inevitable and will clog Wikidata, so I am not in favor of creating any new properties which will contribute to that. Vojtěch Dostál (talk) 19:47, 31 March 2022 (UTC)[reply]
Excuse me, but this is not a serious way to argue for me. If you argue that you are generally not able to control the mass of data to be uploaded, you could not even start a project like Wikidata. The question of implementing a particular new property or not, does not depend on this issue. I asked you about where WD makes a statement about the statistic topic. As long as there is both nowhere a statement about this and nowhere any objectives available or even commonly desired in order to limit the sources, the statement you made is just your personal opinion. Us in de-WP decided to limit the data to the decennial Census. This may - if limitation of data was a real issue - also be a strategy to be implemented in WD in general, regardless of where a Census takes place. But this issue is not to be discussed at this point. This here is only about the property itself. Regards Stefan (talk) 20:43, 31 March 2022 (UTC)[reply]
@Vojtěch Dostál: I believe you are the one missing the scope here. With this argument, there is plenty of properties that should not be used. Just in the context of places, there is a lot of (some heavily used) properties that contain dynamic data. How about population (P1082) (with its 2.3 million usages, even though all of your arguments could be applied to the population number in exactly the same way) and number of households (P1538), very basic and old onces such as head of government (P6) (not numeric, but highly dynamic), but also more similar properties such as nominal GDP per capita (P2132) and median income (P3529)? They all have the same potentional to "clog" Wikidata.
Regarding the currency and different sources: We can clearly define that only official sources (from Census bureaus and similar) shall be used. They have one currency (mostly US-dollars) and are published only once in X years, depending on the individual country.
Therefore, I agree with Stefan's impression that you are bringing up a personal opinion here that does not really comply with the idea of Wikidata. How about you naming the guideline applicable here that suggests to not have this kind of properties?
I am more than surprised about someone "strongly opposing" such a straight-forward property proposal. Yellowcard (talk) 06:27, 1 April 2022 (UTC)[reply]
female population (P1539) and male population (P1540) are just two more examples that really contradict your opinion of not creating properties for dynamic numerical data, and having a look over the existing and also the recently created properties, there seem to be thousands of properties for numeric changeable statistics, and I do not see any consensus that this would "not belong to Wikidata". Yellowcard (talk) 06:44, 1 April 2022 (UTC)[reply]
+1. As far as I got to know Wikidata, WD is a tool to provide data in Wikimedia projects, regardless of what kind the data is, as long as the data is relevant and useful to an individual project, especially Wikipedia. If there is any restriction to this idea - according to rules of WD - it should not be a problem at all to refer to it. If there isn't, then we should put the property to practice. Regards Stefan (talk) 09:53, 1 April 2022 (UTC)[reply]
  •  Strong support As the examples show, data needs to be maintained and updated in various places in WD. The amount of data plays a subordinate role.
But this is about a new property that needs to be decided.
It should only be decided whether it is a meaningful addition to the existing data or not.
The income is meaningful both as an individual value and as a value for an upcoming project and therefore it should be included in WD.
...and also: It shouldn't be a mandatory entry and therefore the data volume will probably be limited. Wikisaar (talk) 14:06, 1 April 2022 (UTC)[reply]
  •  Support Not convinced by the arguments against. It's a basic data point, not very fancy. I don't need three currencies, I need one, maybe two, also, I can ask a bot be programmed for the conversion if I choose so. As to the various methods of data aquisition, that's why those ought to be given, as well as how the data's being processed. Far as I can tell, we don't decide what's right and wrong, we just say what is. Yes, it is a hell of a lot of data on a global level, but creating the property gives one an option one might use (and do the maintenance) or not. For now, --G-41614 (talk) 20:52, 2 April 2022 (UTC)[reply]
Retrieved from "https://www.wikidata.org/w/index.php?title=Wikidata:Property_proposal/Per_capita_income&oldid=1618107748"