I only became aware of ROR at this year’s PIDapalooza! So please forgive me for this “stupid question”.
I looked up the organization where I work as Library and Publications Manager, IUCN, in the ROR Registry and found 5 entries (Research Organization Registry (ROR) Search). We are an international organization, so we have offices all over the world. It seems four of the five entries are for our country offices.
What is the ROR guidance on this? I see that “research organization” is defined as “any organization that conducts, produces, manages, or touches research.” But how should this work for an international organization with offices in various locations? Is it considered “one” organization or is each office a distinct one? Should “organization” correlate with a “legal entity”? (It could be that some of the country offices with a ROR are distinct legal entities, I’m not sure!)
Finally, I see that there is a curation form where one can request a change or update (ROR). But can those changes be initiated by anyone, or does it need to be someone (can it be anyone?) from the organization in question?
Your question is a very good one. ROR specifically aims to address what we call the “affiliation use case” - that is, how to identify which research organizations are associated with which research outputs. This is a different aim than identifying every legal entity in the world. You point out an interesting aspect of the registry, which is that in some cases there are records that correspond to distinct branches of an organization. For the purposes of affiliation, do we need distinct records for each branch? Or is a single ID sufficient? I would like to invite other community members here to share their thoughts on it as well.
ROR launched its first iteration of the registry with seed data from Digital Science’s GRID dataset, and continues to mirror GRID while we develop infrastructure and workflows to support independent curation of the registry. We are looking to community members to help advise on policies for how we maintain ROR records and metadata, so your question will be an important one to consider as part of that.
To answer your last question - yes, anyone can submit a suggestion for additions or changes to ROR.
I am interested in this question as well. I would like to see my organization incorporate RORs in the future for tracking unique ‘institution’ names. For example, among our users are academics which might be affiliated with a university, a university repository, a university lab, and any combination of the three and more. In order to locate where the data being submitted to the repository originated, the university name may not be specific enough. The lab (or specific branch) would be preferred.
I think this is a very important issue that, if possible, I would like to see on the agenda in future discussions. Hence a post to try to resurrect this thread.
The multi-national aspect is important for my own use case (tracking sponsors and funders of clinical research, as well as researcher affiliations) when dealing with pharmaceutical companies, which seem to be listed in ROR under their various national subsidiaries. Given that these probably are separate legal entities that makes sense, but it is not very useful. In general we and (we presume) our system users are not interested if the research is funded or carried out by researchers in (for example) Pfizer US, Pfizer Germany, or Pfizer France, or whatever, we just need to record that it was ‘Pfizer’. The national divisions of companies, in any case, do not normally match the more important functional divisions within those companies (e.g. Novartis Vaccines, Novartis Respiratory… etc.). The sources we use, such as trial registries, provide the data as either the national or functional division, or often with no division at all. Having a mechanism that can allow a single ROR code for a company and link the national and / or functional divisions as ‘child’ entities, albeit of different types, would make ROR much more useful to us.
Ramdeen’s point is also very relevant for our use case. The key organisation for us is often the specific trials unit or laboratory within a university, or a university hospital. It would be great to develop a mechanism for capturing this information where it was available. I don’t under-estimate the practical difficulties of tracking this data (perhaps it could only be contributed and maintained by the parent organisation itself) but I think the issue is worth discussing. A particular granularity issue I have is with government agencies, which are often in a complex hierarchy (e.g. the clinical trial registry ClinicalTrials.gov – whose staff generate research papers - is part of the National Library of Medicine, which is part of the NIH, which is part of the DHHS, which is part of the US government). Every government uses different – and often rapidly changing – structures, so this is a really difficult issue, but I would be interested in what other people thought was possible.
