How do I describe my dataset?
Queen Mary has adopted a standard for describing research datasets. The standard is called the DataCite Metadata Schema and contains minimal descriptive ‘properties’. This keeps the amount of required descriptive information (metadata) small to aid faster creation of dataset records.
When creating a record of your data, you will be asked for the following mandatory information:
- Identifier - a number designated to this particular dataset; usually a Digital Object Identifier or handle. Queen Mary Library staff will assign a DOI for you if you deposit your dataset to Queen Mary Research Online.
- Creator name - who created (authored) this dataset?
- Title - give your dataset a title describing what the data represents.
- Publisher - if you deposit your dataset to the Queen Mary Research Online repository the publisher will be Queen Mary University of London. If you deposit it somewhere else, then the publisher will be the data repository that you used.
- Publication (distribution) year/date - either the date of when a dataset was created, or when a record of the dataset was released.
Supplying the information below is also useful, though not mandatory:
- Version of the dataset - if there are different versions/releases of this dataset (earlier versions, for example).
- Resource type - for example, ‘software’, ‘moving image’, ‘still image’, ‘geo-location’.
Why describe your data using a standard?
By adopting a standard for describing your datasets, sharing information about your data and providing mechanisms for discovering these datasets is made easier. The standard introduces consistency in where and how certain types of information about your data is described, by keeping the number, types, and complexity of the metadata properties to a minimum, we ensure that creating the metadata is simple and quick.
Where do I describe my data?
Regardless of whether you intend to deposit the data files with Queen Mary or via an external data service, a metadata record of the dataset must be created in Symplectic Elements and uploaded to Queen Mary Research Online (QMRO). If you deposit the dataset to an external repository, please provide the link to the dataset in the Elements record.
Documentation describes what the dataset is (akin to an abstract) and enables others accessing your data to be able to understand how it was created, what has happened to it in the meantime, and what instruments or equipment are required to be able to view or work with the data. Whilst this takes a little time in preparing the data for deposit and sharing, it saves time in the future and also helps to avoid the dataset becoming obsolete due to lack of supporting information to make it usable.
Getting help with storing and describing datasets
For help in preparing your dataset with appropriate descriptive metadata and documentation, email email@example.com
There are several chapters in the QMPlus module that may be helpful when describing and documenting your datasets, check out the sections on:
The complete QMPlus Research Data Management course is available online.