Metadata

Metadata is data that describes data. For instance, it can provide information about

• the authorship, and the processes of data sampling or data generation
• publications describing the data and/or using them for investigations or experiments
• the procedures used for processing and analysis of the raw data
• access authorization and copyright notice
• the formats and coding of digital data
• the variables, parameters and categories used in tables, vectors and annotated data
• the application software that was used for generating or processing the data
• the application software for visualizing, observing, or searching the data
• the quality (validity, reliability) of the data

Metadata should comprise an informative headline, an abstract and some keywords. 

Most research projects produce thousands of data units. Metadata are the foundation for

• the administration and organization of data
• data retrieval and browsing within large amounts of data
• the  re-use and further processing of the data

Metadata should be comparable and linkable within and across different research projects.

It is therefore necessary to make sure that the notions used to describe the data are consistent and the terminology is clearly defined.  For instance, the language of a historical document could be labelled with the terms deutsch or German or deu. According to the ISO 639-3 norm, the only correct option is the last one (see https://en.wikipedia.org/wiki/List_of_ISO_639-1_codesExternal link).

Metadata standards offer conventions for the generation and description of research data. They specify and define the structure of metadata.

Search engines for metadata (e.g. http://openarchives.org/pmh/External link of the Open Archive Initiative)  can find such standardized metadata in repositories of metadata and make the corresponding data available for the research community.