Introducing the Open Data Format

The Open Data Format (ODF) is a new, non-proprietary, multilingual, metadata enriched, and zip-compressed data format. The ODF meets the FAIR Guiding Principles for scientific data management and stewardship.

A data file in the Open Data Format has the “.zip”-extension. It is a zip-compressed folder containing the raw dataset in a comma-seperated file (CSV) and the metadata in an XML-file structured in the DDI Codebook metadata standard. For further information on the specification of the Open Data Format you can read the Specification.

To work with data files in the Open Data Format the R Package opendataformat, the Stata package opendf, and the Python package opendf are available at the moment.

You can download an example data file in the Open Data Format (example_dataset.zip). Manuals for generating data files in ODF can be found on GitHub for Stata, R and Python.


About the Project

Researchers in the social sciences use various software for statistical analysis of rectangular, structured data (e.g., Stata, R). Each software has a specific data format that is only partially compatible, if at all, with other software solutions. The non-interoperability of data formats is an obstacle to replication studies and data-reuse. It undermines the FAIR principles and is not in line with the idea of open science. To meet the needs of data users, data producers offer a variety of data formats by doing a lot of redundant work, which is error-prone and leads to increasing costs. Furthermore, there is not only a demand for different data formats but also for material that describes the data; this includes, for example, study descriptions, method reports, codebooks, or questionnaires. As of now, it is common practice for scientists to have to leave their statistical software environment to search for supplementary material while performing the analysis of their data, as this material is often accessible through external data portals. This practice is inconvenient and prone to errors.

The open, metadata-enriched, non-proprietary data dissemination format (ODF) is a project of KonsortSWD, the NFDI consortium for the social, behavioural, educational and economic sciences to develop a data format that adheres to the FAIR Guiding Principles for scientific data management and stewardship.The project includes three main work aspects:

  1. The specification of a new open data format and the documentation of the specifications’ development is the core work that runs through the entire project. We start with a minimal but scalable specification and aim to evolve it into a specification that is suitable for a wide range of use cases.
  2. For a new data format to be usable with existing statistical software, the project develops statistical packages with import and export filters for a selection of software programs (Stata, R) to use the ODF within the software.
  3. Promote the new data format to bring it in widespread use by various research data centers and other data providers.

The result of the project is the Open Data Format (ODF).