Wikidata:Property proposal/Antal valkretsar
Number of constituencies edit
Originally proposed at Wikidata:Property proposal/Organization
| Description | Number of constituencies related to a legislative body |
|---|---|
| Represents | electoral unit (Q192611) |
| Data type | Quantity |
| Template parameter | antal valkretsar in sv:Template:Landstingsfakta |
| Domain | councils |
| Allowed values | integer, 1-650? |
| Example | 4 -> Kronoberg Regional Council (Q3232003) |
| See also | number of seats (P1342) |
- Motivation
I am looking for a way to connect the number of constituencies/electoral districts to the political entities. I would prefer another way to solve this than with a dedicated property, but I do not know a better place to find a solution than in a "live" Property Proposal. In some cases, like in the House of Commons, you have a 1 seat/constituency, but that is far from true everywhere. It is possible to create one item for each constituency, but I doubt many such can be found on local level in Wikipedia. So that would be a tiresome way to solve it. And even if we do that work, I do not know if we have any simple way to connect constituencies with the councils. The numbers of constituencies has become important on local level in Sweden today, since it affects how high the election threshold is. And creating constituency-items when there is only one constituency for the whole legislative body does not look like a good way to do it. -- Innocent bystander (talk) 14:16, 12 September 2017 (UTC)
- Discussion
- I am not extremely familiar with the topic but I would rather expect to have links from constituency items to the item for the legislative body. The count would not be explicitly stored in Wikidata but could be derived (for instance with a SPARQL query). As I understand it, properties which count things (such as member count (P2124)) exist only for cases where the counted things would not all deserve an item on their own. Otherwise it's a bit redundant. But that's just my intuition. − Pintoch (talk) 17:20, 12 September 2017 (UTC)
- But how do you today use a SPARQL query from a template like Landstingsfakta? An option could be like calling contains the administrative territorial entity (P150) in United States of America (Q30) and simply count the number of claims in that property. But I am not aware that we have any relation like P150 between the councils and the constituencies. -- Innocent bystander (talk) 18:27, 12 September 2017 (UTC)
- One other problem with deriving the value by counting the constituencies is that in many cases we don't yet have the items for the individual constituencies. That's something that we hope to clean up a bit for national legislatures as part of Wikidata:EveryPolitician, but in general this would be easier if we could compare how many constituencies Wikidata knows about with how many there should be. (And even in the countries where this information is currently good, historic data is usually much less comprehensive. Again, it would be good to be able to record how many constituencies existed at various points in history, so we can more easily spot places where there are gaps to fill.) We have a related property with number of seats (P1342), so a parallel one for constituencies could make sense. --Oravrattas (talk) 21:03, 12 September 2017 (UTC)
- For Sweden, we have items for historical constituencies on the national level, as far back as we have elections with women's suffrage (Q205204). But on Regional and Local level, I do not think we have anything. We have good sources like val.se for at least 2014, but further back? Luckily the Swedish constituencies are not as many as in UK. Stockholm Regional Council has 12, and that is probably the highest number we have outside of national level. Most Municipalities have only 1. -- Innocent bystander (talk) 04:21, 13 September 2017 (UTC)
Support. The number of constituencies or similar administrative units under a region should be seen not just as something for queries. The number bears an importance of its own. It has bearing on how the political and administrative structure would be, the size of the collective that the region represents, etc. This is one of the first parameters to be talked about when one talks about an administrative region; it's foremost when publishing information about it. And it doesn't make sense to rely solely upon querying the sub-regions with the expectation that every sub-region must be already provisioned from before. Assuming that can lead to wikidata showing wrong numbers with nothing objective to cross-check with. These numbers are well-referenced in official sources; doesn't make sense to include everything else and exclude this just because it can be derived from querying. In Indian context, the administrative levels are : state, district, sub-district, village/city. The number of villages and cities in a sub-district and the number of sub-districts in a district are important bits of info and not easily found. One cannot rely on each and every village to have an entry; we're only just beginning to properly map their boundaries. Further, a city has electoral wards and the number is important as we cannot expect individual wards' entries to be provisioned soon, yet a lot of other data points have this number relevant to them. Nikhilsheth (talk) 14:55, 18 September 2017 (UTC)- Support ديفيد عادل وهبة خليل 2 (talk) 14:50, 29 September 2017 (UTC)
- @Innocent bystander, Pintoch, Oravrattas, Nikhilsheth, ديفيد عادل وهبة خليل 2:
Done ArthurPSmith (talk) 19:42, 3 October 2017 (UTC)
