iDentifying your researchers: challenges and opportunities
This post was edited on 19 June 2019 to correct a factual error in the information on asserting affiliation and update information on the way information was added in the ORCID registry web API
When I first started at my current job in Jisc last year, I was asked to identify ORCID iDs connected with member institutions of the UK Jisc Consortium as a nice easy start to my role… Or indeed, an introduction to one of the ongoing challenges and most persistent issues I face in my day-to-day work of supporting UK institutions that are keen to monitor uptake and use of ORCID iD by their researchers…
You see, providing a persistent unique identifier for researchers, and owned by them is at the core of ORCID’s mission. But this creates tension for a piece of information (the ORCID iD) that is more and more critical as a central identifier in institutional workflows, for identifying researchers’ connections to their funding and research outputs (such as publications) in infrastructure such as repositories and CRIS.
There are two challenges with trying to associate a researcher to a particular institution: access to that information and the completeness/currency of it. Let’s look at those in turn:
Access to Information in ORCID records
The visibility of information in an ORCID record can be set at three different levels – all controlled by the researcher. At any time, any piece of information can be set to “Only me”, meaning it is entirely private and hidden from access (i.e. only available to the researcher); “Trusted parties”, available only to the researcher and to ORCID members they have authorized to access it or “Everyone” – openly available to all (i.e. public). By default, the email address, one of the easiest ways to check for a researcher’s association with an institution (via the domain name, for example, in the UK looking for the email which ends university.ac.uk), is set to private when added to a record. Other ways of associating an ORCID record to a particular institution, such as affiliation, or employment, are open by default, but less commonly filled in, as they are not required to set up an ORCID record, whilst an email address is.
Completeness and currency of ORCID records
Most of the current personal information on the ORCID Registry is not date stamped or the date stamps are optional. This means that, even if you find a researcher with a publicly available email address for your institutional domain, you don’t know for sure whether they are currently associated with your university. However, it is possible to find out when an ORCID iD was created and last modified, so you can have some idea of whether or not a record is being actively used, as opposed to one that was created but never used again – a “zombie” record that might have been briefly brought to life for a mandatory entry in a grant submission then abandoned or a “tombstone” record that was once active but hasn’t been updated for a significant period of time (over a year, say) and may have been abandoned. By checking when a record was last updated you can also identify candidate records that have not been maintained – these may indicate that the researcher needs to be prompted to add to their works, or connect to a new system if they have moved institution, or perhaps it is the record of a now inactive researcher (through change of career or retirement). That ORCID iD are also used by those that administer the ORCID Records within institutions (via ORCID Self Service) further confuses this picture.
The fluffy bunny confounder
Even if a record is being actively used, the researcher may not have added the information that is needed in order to make a connection between them and their institution. For example, their record may not contain any employment or affiliation information and only a personal email address, (like firstname.lastname@example.org ) but they are, in fact, a current researcher.
Implications, gotchas and first steps
So, in summary, the above implies that when you derive a list of IDs linked to your institution, you may find that:
– Some of the records returned are for people no longer associated with your university, but were past affiliations
– Some of the records returned are out of date
– Some of the records will not be returned because there is no linked email with your institutional domain, or the email is private
– You may not have found all the records because you have not used a complete list of affiliation id – some institutions have more than one OrgID
You can improve your searches for affiliation by using measures such as ensuring you have covered all versions of your institutional email address (some places may have university.ac.uk AND univ.ac.uk), and using employment records, and other institutional identifiers, such as Ringgold id or GRID
For consortium members, there’s a monthly report that gives you a number of your currently affiliated researchers, along with activity through any integrations you have registered.
Improvements in infrastructure
So, what is happening to address these issues for institutions? There are three improvements and approaches on the horizon:
Since API 2.0, start & end dates can be added on to affiliation items. There has recently been much discussion about whether or not a start date should be mandatory for employment information, but currently it is still optional . In fact, since API 3.0 launched just recently, adding a start date is possible in the web interface and in the API (and strongly recommended).
A number of systems are working on inference based reporting, where you can use the connections about iDs that are stored in another system as well as the information you hold in your own system – for example by researchers adding it to the CRIS and a report being generated from the CRIS directly.
Finally, Jisc is developing a platform to coordinate all the sources of information that might help identify those researchers that do (and do not) have an ORCID iD. The Community ORCID Dashboard project is currently under active development and you can find out more about it on its project page.
What can researchers do?
There are also several ways that researchers can be encouraged to make their ORCID records more useful to their institution — and, therefore, to themselves.
- They can add their institutional email to their ORCID record and make it public or open to trusted parties, enabling everyone — or just ORCID members — to view this information and associate them with their institution
- They can add an employment or education affiliation to their record either manually, or by using their institutional repository or CRIS or other platform to verify their association with an institution, and ensure this information is available to everyone or to trusted parties
What can institutions do?
- Communicate the above to your researchers – especially the importance of having an institutional address in your ORCID record that is accessible to all
- Start adding affiliation to researchers who have given you permission (if this is supported by your integration) or influence your vendor to improve your integration to allow you to add affiliation and dates to records
- Check all your email domains are included in the monthly reports so they are more complete