Deutsch Intern
Research Data Management

Data publication and archiving

Research can be published in different ways

  • as supplementary material of a text publication
  • with a descriptive text publication in a data journal
  • as an independent publication in a data repository

Data supplement

Publishing research data as supplementary material of scientific articles is a common way of data publication and is offered by most journals. In this case, the data are either stored directly with the article or it is outsourced to a data repository.

When data is attached directly to an article, it has no own citable ID (e.g., DOI or handle) by which it can be found independently of the article. Therefore, it is not a data publication in the classical sense. If the journal is not available as Open Access publication, the access to the data may also be restricted by the terms of use of the journal.

The text publication can also be stored separately from the text publication. In this case, the research data is published as an independent research output in an external data repository while the article is published in a journal. In some cases, there are already cooperations between publishers or journals and data repositories (e.g. Elsevier and PANGEA or Zeitschrift für Soziologie and GESIS) or there are at least recommendations where the data can be stored. The data is assigned their own persistent ID in the data repository, so that data can be referenced permanently.

Data journals

When published in a data journal, the data are mainly described in terms of their underlying research design and their potential for reuse, without reporting empirical analyses or substantive findings about the data. The quality assurance of the publications is usually ensured by peer review. The research data itself is stored in an external data repository and can be accessed separately from the data article via a permanent identifier.

Data journals are particularly useful for presenting a dataset to the public, facilitating its subsequent use by providing further information, and rewarding one's scholarly efforts during the research process with academic output.

Data repository

Data repositories are storage locations for archiving digital research data over a longer term and, in many cases, for publishing it. They can be divided into three types based on their focus on specific subjects and who provides the service.

A subject-specific repository specializes in a scientific discipline or a specific subject area. Several scientific disciplines already have established subject-specific repositories. The subject-specific focus offers some advantages:

Subject-specific repositories are known and respected in the research community, so the data are more easily found and more frequently used by the target group. This increases the scientific visibility. They are strongly oriented towards subject-specific quality standards. Therefore, they usually offer additional services, such as quality checks by data curators with subject-specific expertise. In addition, they typically use subject-specific metadata standards and enable the storage of extensive material for documentation, making it easier to find and reuse the data.

Examples of subject-specific data repositories:

Cultural studies and humanities

Social, behavioral and economic sciences

Life sciences

Natural sciences and computer science

A multidisciplinary data repository publishes and/or archives a wide range of data from very different scientific disciplines. It is a particularly useful option if no suitable subject-specific repository is available. Due to the multidisciplinary approach, these types of data repositories place lower demands on the accompanying documentation and data preparation and mostly use general metadata standards rather than subject-specific ones.

Examples of multidisciplinary data repositories:

Another option for data publication is an institutional research data repository. It is operated at universities or other research institutions and enables members of the institution to publish or archive their research data digitally.

WueData: The institutional data repository of the JMU

Scientists at the JMU who want to publish their research data according to the FAIR principles can use the institutional research data repository WueData. WueData offers the following advantages, among others:

  • Data can be described with metadata to be found more easily.
  • A Digital Object Identifier (DOI) is assigned thus the published data can be found permanently.
  • Documents related to the study can be attached to the data for better reuse.
  • Almost all types of data can be published. 
  • An (open) license is specified to make copyrights more transparent.
  • It is possible to set an embargo to open the data package for third parties at a later date.
  • The data including backup is stored locally on servers of the JMU.
  • Currently, a minimum retention period of 10 years is guaranteed. In the future, the aim is to transfer the data to a long-term archive.

Link to WueData

There are resources on the internet that can help you to find a suitable subject-specific or generic repository. Some directories and portals are listed below.