Enhancing peer review data with industry datasets

One of the key benefits of DataSalon’s new PaperStack service is that the raw data pulled from manuscript submission systems is automatically enhanced with trusted industry resources, placing it in a wider context in order to give full insight into the editorial process.

The additional data we pull in comes from:

the GRID database of research organizations (or the Ringgold database where the relevant licence is in place)
the Funder Registry dataset of funding bodies
the ORCID registry of researchers

Even when the relevant IDs already exist in the source data, linking up these various pieces of information is quite complex. For example, the ORCID ID of an author or researcher is used to retrieve full affiliation data from the ORCID API, and this data may in turn contain IDs referencing GRID/Ringgold and/or Funder Registry data.

But we also go further: our set of ‘automatching’ tools (which include fuzzy searching, lists of abbreviations and synonyms, and the use of email and web domains) allow us to identify affiliations and link to the standard datasets even in many cases where IDs are missing.

All this enables us to add some useful features to PaperStack:

Full linking of articles, individuals and organizations: Relationships between organizations, authors, reviewers and funders can be explored, and editorial activity can be viewed at an institutional level. Reports can be broken down by author/reviewer institution or by funder, supporting author retention, tracking of reviewer diversity, and analysis of funding sources.

Standardised organization profiles: Author/reviewer affiliation data and article funder data may have issues such as inconsistent naming, varying levels of details, or data entry errors. These can be overcome by linking them to a single clean dataset drawn from GRID/Ringgold and the Funder Registry, which in combination provide excellent coverage of scholarly institutions.