23 Research Data Things – 16: Publishers & Funders

Explore a new form of data publishing: the data journal. Data journals focus on data, rather than discuss an analysis of the data (as in traditional journals).

I think data journals are actually very similar to traditional journals – they look the same, smell the same, follow the same kind of format. The experience of accessing a paper in a database is the same. However, in a weird way they’re the opposite of traditional journals, because the intention is backwards. In a traditional journal article, the research is done, data collected, to support the writing and publishing of the article. Here, the article is like a piece of supporting documentation for the dataset. The focus is on the data, with more tables, graphs, and explicit reference and links to the dataset in places of prominence. I think that this suits researchers that are more focused on putting the data out there, rather than the analysis and discussion.



23 Research Data Things – 15: Data Management Plans

In the past I’ve had a look at the “Mantra” course from The University of Edinburgh which is really helpful.

They give an example of core components of a research data management plan (based on the DCC’s checklist). Most data management plans I have looked at contain some variation of these:

  • Administrative Information
    Name of the project and description, important dates, associated organisations, funding body, etc.
  • Data Collection
    Collection methods, types, formats, systems and structures for file management, etc.
  • Documentation and metadata
    Supporting documentation to help others read and interpret the data, metadata standards, controlled vocabularies
  • Ethics and Legal Compliance
    Consent needed for data collection, privacy and confidentiality concerns, copyright concerns.
  • Storage and Backup
    Where data will be stored, backup and recovery procedures, data security.
  • Selection and Preservation
    Contractual, legal or regulatory requirements impact on retaining, sharing, preserving data, length of time, details of repository or archive.
  • Data Sharing
    Discoverability, how it will be shared (e.g. repository), citation and persistent identifier.
  • Responsibilities and Resources
    Name of reponsible person(s) for each activity, hardware and software requirements, charges required, etc.


23 Research Data Things – 14: Identifiers and Linked Data

ORCID is great. I’ve learned about it before and I’ve helped with an info session for researchers in setting up an ORCID profile. It’s a great way to achieve author disambiguation and can be helpful for researchers in promoting their output, and in a lot of cases is mandated in order to receive research funding.

When I searched ORCID for “John David Burton” there were 3 results that potentially matched the name. Two of them were “John Burton”, but there was no publication information on either of these profiles so it’s hard to tell if they are the same person. There are situations where the same person has duplicate IDs. The third potential result is for “J. D. Burton“- this is our guy. He has listed other names: John David Burton, John Burton, John D. Burton. Filling in a list of aliases is great for both disambiguation and discovery. He’s also linked his Scopus author ID and ResearcherID to his author profile, which is a good move because ORCID lets you import your paper details from these products.

Toby Burrows has done the same thing with Scopus and ResearcherID. There are quite a few datasets in his list of publications, but the publication type for these is “other”. At first I felt a bit offended, assuming ORCID was implying that datasets aren’t important enough to warrant their own category. But then I found a list of supported work types in ORCID and data-set does appear as a work type under the sub-heading “other”. Either this subheading causes it to display in ORCID as “other” (though I doubt it – since other fields like journal-article appear as they are and not under their wider subheading – “publications”) or if it was just the way the records were added/imported.