Documentation and metadata
Metadata and documentation are crucial in making data findable and reusable in the future, by both the creators and others.
Metadata is structured information about an item of data or dataset in a controlled vocabulary; it supports the discovery, understanding and management of other data and information. Data documentation, any documents which describe the data used in a research project, can take a variety of forms. These include electronic laboratory notebooks, data dictionaries, codebooks, vocabularies and readme-files.
Data documentation explains what data is held by the project, how it has been collected, what any abbreviations mean and how the data has been modified. Documentation will be most effective when accompanied by well-organised data; this includes using a file naming convention, directory structure, and version control. The University of Helsinki provides further information on documentation and metadata including a documentation checklist.*
The University of Manchester and many research funders, including the EPSRC, require researchers to provide metadata which allows others to understand how the associated research data was created and how it can be discovered. Where possible it is best to use metadata schemas and standards which are recognised within your discipline and you can find examples of these in the Research Data Alliance Metadata Directory or at FAIRsharing.org.
Key elements of metadata
Metadata for online data catalogues or discovery tools are often structured to international standards or schemes designed for specific purposes or subjects that may include the following:
- Subject descriptors
- Creator(s) (Creator of the dataset; main researchers involved)
- File format
- Storage location of the data (including identifier information)
- Origin of the data (creation/acquisition of the data)
- Time references for the data (key dates associated with the data: start, end, release, etc)
- Access conditions