Increasingly, in the scholarly publication landscape, journals are considering and developing policies around the sharing or publication of data underlying the manuscripts they are publishing. There is a wide range of these policies, such as:
- no mention of data sharing or publication
- requiring a statement on the authors’ willingness to share the data, e.g. Annals of Internal Medicine, British Medical Journal (BMJ), International Committee of Medical Journal Editors
- a statement encouraging data sharing, e.g. Monthly Notices of the Royal Astronomical Society
- requiring all data underlying a journal article to be made available with no or minimal restrictions, e.g. PLOS Medicine, Nature, PNAS.
A journal’s policy about data sharing often suggests a location/repository for the data to be archived. For examples of this see PLOS Medicine and Scientific Data. Some journals have relationships with specific repositories, and have integrated data submission to that repository into their manuscript submission system. Some journals have integrated data submission to Dryad within their article submission process, and Wiley is currently piloting this process with figshare.
Dryad provides recommended elements and examples of journal data policies.
ANDS Guide: Research data for journal editors
This ANDS Guide is intended to provide a starting point for Editors considering developing or improving data policies for their journals.
Data journals are publications whose primary purpose is to expose datasets. They enable the author to focus on the data itself, rather than producing an extensive analysis of the data which occurs in the traditional journal model. Fundamentally, data journals seek to:
- promote scientific accreditation and re-use
- improve transparency of scientific method and results
- support good data management practices and
- provide an accessible, permanent and resolvable route to the dataset.
Publishing in a data journal may be of interest to researchers and data producers for whom data is a primary research output. In some cases, the publication cycle may be quicker than traditional journals, and where there is a requirement to deposit data in an "approved repository", long term curation and access to the data is assured.
Publishing a data paper may be regarded as best practice in data management as it:
- includes an element of peer review of the dataset
- maximises opportunities for reuse of the dataset and
- provides academic accreditation for data scientists as well as front-line researchers.
While individual publisher policies vary, it's worth noting that publishing data through a data journal does not necessarily prevent the publication of data analyses and research results in a traditional journal, along with a reference and links to the data journal paper. This provides readers with access to all relevant information about a piece of research and may result in citation of both the journal article and data paper.
Formal publication and citation of data supports the recognition of research data as a first class research output. It also enables the generation of citation metrics for research data outputs. With products such as theThomson Reuters Data Citation Index(DCI) capturing data citation metrics, the potential for formal recognition and reward mechanisms based on data publishing is enhanced.
Working in collaboration with Thomson Reuters, ANDS has established a service to feed records in Research Data Australia to the Data Citation Index. Citation counts for those records are retrieved from the DCI and displayed in Research Data Australia.
A number of data journals also support 'altmetrics' that track the number of article views, number of downloads, and social media 'likes' and recommendations. These can be early indicators of the impact of data, before the long tail of formal citation metrics can be assessed. Like traditional journals, an increasing number of data journals have Journal Impact Factor rankings reflecting the number of times articles are cited boosting the credibility of the journal.
Data journals, like traditional journals, have differing requirements for submission, review and publication. However, to give a sense of requirements for data journals and how they may differ from traditional journals, the following points are indicative:
- Depositing data: data may need to be deposited in an "approved repository" or with the journal itself. There may be restrictions or guidelines on file size and format as well as specific requirements for metadata or data description.
- Citation and identifiers: some data journals require that data be assigned a Digital Object Identifier (DOI) or other form of persistent identifier. There may also be a defined or recommended data citation format.
- Researcher profile: you may be asked to provide details of your research profile, organisation and affiliations.
- Copyright and licensing: in addition to copyright, you may also need to consider (and agree to) licensing and access conditions for the data to be published.