You are here: Home / Dissemination / Quality Indicator

Quality Indicator

The generation of research data is both work extensive and costly. Data users are searching for standardized high-quality data. Funding agencies wish to encourage the re-use of data whose production they have funded. To foster the reuse of research data, the FAIR data principles were established: findable, accessible, interoperable and reusable. Assigning DataCite DOIs and metadata to datasets are the first steps to meet the FAIR principles. However, DOIs and metadata do not guarantee that data are actually reusable if quality information on the data and metadata are not provided and data are saved in proprietary or undocumented file formats. The AtMoDat project (Atmospheric Model Data) aims on establishing a generic quality indicator to bridge these gaps and foster reuse of data.

NOTE: The Quality Indicator concept was redesigned as Data Maturity Indicator. Please consider the information on this page as outdated. We leave it here for some weeks as some kind of provenance information.

Aim

The AtMoDat project (Atmospheric Model Data) aims on establishing a generic quality indicator in the DataCite DOI metadata schema which

  • supports data users to identify high quality research data in and beyond their field of expertise,
  • incentivizes data producers to publish high quality data, and
  • honors data repositories which perform extensive data curation.

Instead of a purely discipline-specific implementation, a generic implementation within the DataCite schema makes sense, as we want to facilitate the reuse of research data by an interdisciplinary audience.

Approach

The core of this approach would be that published data records would receive a quality indicator - a new property either named ‘Quality Assessment’ (Response: ‘yes/no/n.a.’) or ‘Quality level’ (increasing level with increasing number of fulfilled quality aspects ). The Quality levels following the FAIR principles would have to be discussed, developed, communicated, and implemented by the different scientific communities. We are aware that this is a long-term process and must be driven by active community discussions and that the outcome will differ amongst different disciplines in their ‘level of practicality’. However, given the high volumes of digital scientific data, such a quality indicator will help data producers, consumers, data centers, and even funders in the long run. The quality indicator should refer to the dataset, rather than to the repository or data center itself. That means that such a Quality indicator will differ from repository certificates such as Core Trust Seal or DINI-Certificate. These certificates usually include a time-period during which the certificate is valid. In contrast, the quality status proposed here is permanently assigned to the published dataset.

Aspects described by this quality indicator should be

  • the use of all applicable recommended and optional DataCite metadata,
  • documentation by means of discipline-specific metadata,
  • the use of open file formats which follows discipline-specific standards, and
  • checks of metadata, file formats and discipline-specific standards for correctness and completeness by the publishing data repository

The approach described here would improve the labelling of quality-checked data and thus significantly contribute to their re-use.

Adding new properties to the DataCite metadata schema takes a considerable time span. Therefore, a purely discipline-specific qualtiy standard was developed that can be quickly applied  to atmospheric model data and, later, be converted into a discipline-specific implementation of the generalized quality indicator described above.

Example "Quality Level"

An exemplary classification for a ‘Quality level’ metadata property could be:

  • Level 0 (or “not filled”): mandatory DataCite metadata filled
  • Level 1: recommended and optional DataCite metadata filled if possible
  • Levels m (m>1): DataCite metadata schema extended by discipline-specific metadata; metadata checked for completeness by the data repository; open file format used and discipline-specific standard applied for file contents, if appropriate

Further material

Please find brief one-page-long descriptions of these two quality standards below for download:

 

NOTE: The Quality Indicator concept was redesigned as Data Maturity Indicator. Please consider the information on this page as outdated. We leave it here for some weeks as some kind of provenance information.

 

We invite you to get in touch with us via [Email protection active, please enable JavaScript.] .

Filed under: , , , ,
Contact / Remarks

General questions on AtMoDat? Suggestions for improvements of the webside? Contact us:

Funded by

BMBF Gross Crop

within the framework of "Forschungsvorhaben zur Entwicklung und Erpro-bung von Kurationskrite-rien und Qualitätsstan-dards von Forschungs-daten".

Details zur Förderrichtlinie