Following the Scholix framework, in order to create a link between a dataset and an article, there are 3 metadata fields that must be populated: relatedIdentifier, relatedIdentifierType and relationType (see previous section). This section looks in more detail at the use of relationType, how/when these relations become citations and how this information will be exposed for data level metrics.
The DataCite Metadata Schema defines relationType as a "Description of the relationship of the resource being registered (A) and the related resource (B)"
The full list of relationType controlled vocabulary and directionality is shown below. A definition of each relationType is included in the PDF on the DataCite Metadata Schema website.
IsCitedBy | Cites
|
|---|
A relationType should be selected to best describe the relation and direction between the subject and the object. This information will be then be used to define whether the link is a Citation, a Reference or another type of Relation.
The figure below explains how each relationType will be counted by DataCite for Data Level Metrics:
Use Case
Dataset A is cited in Article B. The repository is updating the DOI metadata for dataset A and have already included the relatedIdentifier and relatedIdentifierType of Article B. They need to select the relationType.
- Describe the relation
Following the recommendation above, the repository must choose the most appropriate of these 3 relationTypes: IsCitedBy / IsReferencedBy / IsSupplementTo. The relationType can now be included in the DOI metadata of dataset A. This will count as 1 citation for dataset A.
Only one relationType should be used to describe a link between a dataset and an article. It is not necessary to select more than one, even if multiple relationTypes could apply.
-
Select the direction
Keep in mind the direction of the relationType. In this case dataset A (the subject) IsReferencedBy/IsCitedBy/IsSupplementTo Article B (the object). -
Data level metrics
This information can now be exposed via Event Data in DataCite Search and via the Data Metrics Badge.
Updated about a year ago