ORCID Pilot Progress Meeting 17th September in London
At the meeting the pilot HEIs and other institutions involved in ORCID related work came together to discuss common issues and challenges and solutions when implementing ORCID identifiers in HEIs. We’ve summarised the discussions at the meeting below. For further information see also the agenda and presentation slides from the meeting.
Summary of Discussion
Many thanks to Helen Henderson, Hazel Woodward and Rob Johnson for summarising the group discussions and to Laure Haak from ORCID for providing further information and comments on the questions.
Things that worked well so far in your ORCID pilot project. What would you would like to build on?
- Where there has been buy in from the top in the institution it is comparatively easy to start a project, especially where it is part of an OA initiative
- University support for ORCID via a publications policy requiring that researchers include an ORCID identifier when they record their publications on systems such as repositories aids advocacy efforts, as does drawing on Postgraduate students and Early Career researchers.
- It’s encouraging to see the new (optional) HESA fields to capture ORCID identifiers https://www.hesa.ac.uk/article/Student_Record_2014_15___ORCID
- The support received from ORCID has been very helpful, with good documentation on the website
- The new functionality to ‘round trip’ information back to an ORCID record that was made public with an ORCID identifier as part of the metadata, is very promising. For example, for research papers a publisher captures the ORCID identifier during the manuscript submission process and includes the identifier in metadata when minting a DOI. Through a process facilitated by the publisher or CrossRef, researchers can then give permission for information on published papers to be automatically added to their ORCID record. See also:
ORCID is working on similar workflows for affiliations (including the ability to update affiliation date), datasets, and funding.
- Some institutions have found it useful to look at the approach taken by Chalmers University of Technology in Sweden and University of Oxford. Both institutions have decided to facilitate record creation (rather than creating records on behalf of their researchers) and to provide linking tools to local systems. See also:
-a recent presentation by Jonas Gilbert (Chalmers University)
-a technical document created by ORCID detailing this “Click-and-connect” approach http://support.orcid.org/knowledgebase/articles/369982-create-on-demand-workflow
-a supplementary document describing how such an approach can be integrated into an email workflow
What hasn’t worked so well and suggestions for institutions implementing ORCID
- Internal technical delays have hampered a number of pilot sites. All institutions should ensure that their CRIS and HR and other relevant systems are ‘ORCID ready’.
- For some institutions there is confusion about ‘what is an ORCID’. Is it an identifier or an ORCID profile?
- Some institutions have found it difficult to identify a specific group of researchers to focus the pilot project on. Some have selected those who have signed a specific contract, others those who have publications in the institutional repository
- There appears to be an STM/ AHSS discipline divide in the take up of ORCID (with more take up from STM academics) which other project participants should note.
This also came up during the ORCID implementation at the University of Colorado. They took the opportunity to broaden the conversation about reporting with their HSS departments. CU presented on this at the May ORCID Outreach meeting. http://www.slideshare.net/ORCIDSlides/johnson-cu-boulderorcidoutreachmeetingslides (See slide 7)
See also discussion at OR 14 ORCID workshop https://orcidpilot.jiscinvolve.org/wp/blog
Suggestions and Questions for ORCID
- It isn’t possible to easily see who from your institution has already got an ORCID identifier.Could ORCID develop a ‘push mechanism’ to provide member institutions with updates on the number of ORCID identifiers claimed and used by the institution’s researchers?
Laure Haak: We are currently looking into ways to support longitudinal reporting on records by email domain. We also encourage universities to (a) use the click and connect workflow so that no orphan records are created, (b) use the autofill functionality to add in the name and email address during the click and connect workflow so that it is possible to track registrations by email domain, and (c) include affiliation in the click and connect workflow as one of the types of information that can be linked to the person’s record once it is created.
- How can unintended duplication of ORCID identifiers be prevented?
On ORCID’s duplicate detection process please see: http://repository.jisc.ac.uk/5378/1/ORCID_Functionality_and_Plans_20130820.pdf and
ORCID is also working on adding affiliation as another field to be considered in the detection algorithm.
- Duplicates of publications and other research outputs in ORCID records (this happens e.g. when importing publications from Scopus or ResearcherIDs) are problematic
Laure Haak: This will be addressed in a release in about 1 month. The back end and user interface have been updated to collate ‘works’ with the same persistent identifier (for example the same DOI/PubMed ID/ISBN). These ‘works’ will be grouped (not merged) in the interface. For a preview, log into your ORCID record, and add a 3 after the URL and click refresh. (https://orcid.org/my-orcid3). The issue with ResearcherID has also been addressed.
- In some cases it isn’t easy to link an ORCID identifier to an institution as the list of institutions doesn’t have the correct structure. Could there be an option to just link to the top level of an institution?
Laure Haak: We encourage users and integrators to use the top-level organisation name. Any “bulk” creation process or outreach effort should include guidance to the researchers as to which organisation name to select (or to create a tool that pushes this info into the record). If there is an issue with the way the organisation is specified in the list, the ORCID member organisation should contact either ORCID or Ringgold to make an adjustment. See also
- If HEIs create ORCID records on behalf of researchers these are made public (name and ID and any data marked as public) 10 days after creation whether they are claimed or not. There are concerns around the publication of unclaimed records. Is this going to be changed?
Laure Haak: For any organisation concerned about privacy with the bulk create process, we encourage you to use the click and create process. In this workflow, the record is not created until the researcher clicks on the button in a user interface or email. See above.
- There is a large difference between basic and premium ORCID membership. It’s not entirely clear what the extra benefits of premium membership are.
Laure Haak: Premium membership provides access to a call-back API for push notifications when a record of interest to you is updated. As the round-trip functionality is put into place, this means your local repository can auto-synch with information being posted to an ORCID record. Premium membership provides 5 tokens, which supports integration of ORCID services into multiple systems, in particular those with distinct security or privacy requirements. Premium membership comes with monthly usage reporting so you can track traffic using your integration. And, premium membership provides more hands-on technical support
Suggestions for the Jisc ARMA ORCID pilot project
- Some institutions had concerns that it would be legally complex for universities to create ORCID records on behalf of their researchers. Can we examine this in more detail as part of the pilot also to see if, and in which scenario, the above mentioned ‘click-and-connect’ workflow is the preferable route to take.
- We need to consider the range of institutional scenarios, and recognise that best practice in ORCID adoption will vary according to institutional size and culture
- We need to get advocates from within the local community. Could the voluntary ORCID ambassadors recruited by ORCID be encouraged to attend the next pilot meeting so that we can learn from their experience?
- The benefits of ORCID are easily articulated and in general widely accepted, but they will be in the future rather than immediate. We need to establish concrete benefits of ORCID for researchers and HEIs. We should also look at the results of the Sloan funded adoption and integration program in the US for evidence of benefits.
Laure Haak: The ability to validate information in Scopus is an immediate benefit, as is the ability to push ORCID iD information into Scopus, WoS, and EuropePMC. With these tools, discoverability is already improving. Also, with the launch of several third-party metrics tools, the ability to render personal usage statistics is easier and more accurate.