Research Data Management: Metadata

 Metadata is another form of documentation and is simply ‘data about data’.  

• It is related to the broader contextual information that describes your data, but is usually more structured in that it conforms to set standards and is machine readable.
• Metadata Standards provide specific data fields or elements to be used in describing data for a particular use.
• Some research fields have predefined metadata standards.
• Dublin Core is used here as an example of a metadata schema to illustrate the types of metadata you might need to capture but also the fact that metadata can be captured in a standardised way using controlled vocabularies.

2b What metadata will accompany the data?

Points to consider: 

• Indicate if you will be using a metadata  standard to help others identify and discover the data. Use disciplinary metadata standards where these are in place.

The UK Data Service explain: "metadata can describe the content, context and provenance of datasets in a standardised and structured manner,  typically describing the purpose, origin, temporal characteristics, geographic location, authorship, access conditions and terms of use of a dataset."

Rich metadata (elements which describe the data) enhance the findability, interoperability and reusability of your data. To comply with the FAIR Principles metadata should be accessible, wherever possible, even if the data aren’t.

The quality of the descriptive information (metadata and documentation) regarding the data has a profound impact on their reusability so the more documentation and metadata you can provide, the better.

Your chosen Data Repository or Archive may have a metadata template you can complete or a required standard you must use. If not you should follow relevant disciplinary standards.

Dublin Core is used here as an example of a metadata schema to illustrate the types of metadata you might need to capture but also the fact that metadata can be captured in a standardised way using controlled vocabularies.

Dublin Core is comprised of 15 “core” metadata elements. It is one of the simplest and most widely used metadata schema. The name "Dublin" is due to its origin at a 1995 invitational workshop in Dublin, Ohio, nothing to do with Dublin, Ireland unfortunately. Originally developed to describe web resources, Dublin Core has been used to describe a variety of physical and digital resources.

Built into the Dublin Core standard are definitions of each metadata element that state what kinds of information should be recorded where and how.  Associated with many of the data elements are suggested controlled vocabularies.

All elements are optional and repeatable.

The example below uses basic Dublin Core metadata elements to describe the book using Title, Creator, Date, Description and Type.

Notice how the author's name is formatted. This is a standard way of formatting names within Libraries and facilitates alphabetical sorting by author. Additional metadata elements that could be added to increase the richness of the metadata include the ISBN, information about the publisher, the location of this specific book within the library, some subject terms, the language of the book, the physical size of the book (helpful to know when shelving), the number of pages etc.

The example below uses basic Dublin Core metadata elements to describe a photograph using Title, Creator, Date, Description and Type.

Additional metadata that could be captured about the photo includes the location it was taken, the copyright owner, licence information, information about the camera used to take the photo, technical specification include image resolution ec.

The example below uses basic Dublin Core metadata elements to describe a piece of research data, in this case an interview, using Title, Creator, Date, Description and Type.

In the case of a single interview it's likely that multiple 'digital objects' for each interview will exist, for example the audio recording, a text file with the transcript and perhaps an NVivo analysis file. These can all be linked using metadata. Other metadata elements that might be useful to record include the topics covered in the interview, some demographic information about the interviewee ec.