Open ChLaaboudi opened 3 years ago
(flagging for version 2) How would we call this? coos:RegionalStatisticalOffice?
I believe that Regional Statistical Office is good, but maybe the term is context dependant, at INEGI call on that way the geographically dispersed offices that are part of the Institution. I don't know if other NSO or International Organisation call their dispersed offices in the same way.
Might (or might not) be worth glancing at these categories in Wikipedia https://en.wikipedia.org/wiki/List_of_national_and_international_statistical_services. Interesting that they have avoided using statistical office/institute/institution/organisation/organization at the top level preferring to use statistical services.
Another question related to organization: Why "organization", then "institute" and "agency", and now "office"? Perhaps we need to harmonize
Besides National Statistical Institutes, we have also national (governmental) agencies/bodies .
I would call it "Official Statistics Producer", and for sure that tehre are more than just NSI and International Agencies under it. There are Central Banks, ministries, Planning departments, etc., that produce key official statistics, so this level needs to be wider. The sub-types coos:InternationalAgency and coos:NationalStatisticalInstitute are not enough.
I'm not sure whether we should consider further down-hierarchies, since this could be very country-specific. Eventually we can identify a "coos:Sub-nationalOrganization" and/or "coos:DelegateOrganization" or "coos:SubordinateOrganization" (these last two thinking of the case of ASTAT e Istat, for example)
May be there is no need to add new classes for down-hierarchies (sub-organisations). Couldn't we reuse org:OrganizationalUnit with its properties
I would call it "Official Statistics Producer", and for sure that tehre are more than just NSI and International Agencies under it. There are Central Banks, ministries, Planning departments, etc., that produce key official statistics, so this level needs to be wider. The sub-types coos:InternationalAgency and coos:NationalStatisticalInstitute are not enough.
@egreising Are you proposing to rename StatisticalOrganization
or to add a new class?
I'm not sure whether we should consider further down-hierarchies, since this could be very country-specific. Eventually we can identify a "coos:Sub-nationalOrganization" and/or "coos:DelegateOrganization" or "coos:SubordinateOrganization" (these last two thinking of the case of ASTAT e Istat, for example)
Regarding hierarchies, we can indeed use the properties defined in ORG.
@FranckCo I propose to use "OfficialStatisticsProducer" instead of "NationalStatisticalInstitute" Sub-type. "InternationalAgency" can remain, I think.
For the down-hierarchies it's ok to use the properties defined in ORG as you suggests.
If the idea is to completely "align" both sub-types we may have something like "NationalStatisticalProducer" and "InternationalStatisticalCompiler". These would cover not only official statistics but also private sector producers, and for the international field, the term would be more precise than just "InternationalAgency", and would include producers of global statistics and also those agencies collecting national data.
@egreising is your final proposal this?
Re-"compiler": wouldn't it be enough to just say "international organization", it is under coos namespace, so we understand it as an international organization that plays some role in the context of official statistics. Not sure if we want to specify it as "compiler". Besides, I think for some surveys, role of international organizations can be argued to be more than "compiler" (e.g., UNICEF MICS).
I know this is an old argument, but can we use "Other National Organization"?
coos:StatisticalOrganization is now split into two sub-types: coos:InternationalAgency and coos:NationalStatisticalInstitute.
A third type for representing regional statistical institutes may be needed. Below few examples illustrating the use case: Germany: