Metadata

Metadata is "data about data". It enables the documentation of research data, allows it to be structured internally and makes it easier to find a dataset at a later date.

Metadata is often structured according to a predefined schema. Such a schema usually includes mandatory information, such as the names of the authors of the dataset or the license under which the data is stored. In addition, many repositories use other optional metadata, such as language of a dataset, geographical information or funding agency information.

The visibility of published research results can be greatly enhanced by well-maintained metadata.

Metadata standards

There are several international, discipline-specific metadata standards for recording metadata, such as:

  • CF Metadata (Climate and Forecast Metadata) in the field of geosciences
  • DDI (Data Documentation Initiative) in the field of social and economic sciences
  • TEI (Text Encoding Initiative) Text annotation/captioning, especially in the humanities

In addition, the international consortium DataCite provides a standard for the interdisciplinary bibliographic description of research data, which is also relevant if published data are to be assigned a DOI (Digital Object Identifier). Research data published via our service datastore is also described according to the DataCite metadata standard.

Authority control and controlled vocabulary

When assigning data and metadata – whether automatically or manually recorded – so-called authority control and controlled vocabularies should be used wherever possible. These were developed, for example, for the unambiguous identification of persons, institutions, cities or research sponsors. A good overview of norm data can be found in the article "Norm data – what they can do, where you can find them".