Publishing and sharing your research data increases research visibility and discoverability. Increasingly funders and publishers are requiring research data be made publicly available following the completion of research projects. For example, the Marsden Fund contract specifies (unless prohibited by ethics approvals) the establishment of “adequate and reasonable access to metadata, data and samples within twelve months of the Completion date of the Contract to (i) people carrying out research; and (ii) national and international repositories”.

Factors to consider when publishing and sharing data include:

• Format, type, size, complexity and sensitivity of data
• Ethical issues such as obtaining consent from participants and research partners to share data. See template documents at the Human ethics website. Specify in consent forms that de-identified data may be shared publicly
• Māori cultural concerns around the sharing of data considered tapu or culturally sensitive, such as whakapapa or images related to death and dying
• Ensure that data have been de-identified or anonymised if needed
• Data formats should be general or commonly used and non-proprietary, such as .txt, .csv and .jpg, .gif or .tiff for images.
• Consider the need for a data sharing agreement
• Describe the data collection methods, dataset variables etc in an accompanying Readme text file
• How will the data be licensed for reuse?
• Are there any restrictions on the reuse of third-party data?
• Will data sharing be postponed / restricted e.g. to publish or seek patents?

Research data can be published in data journals or stored in data repositories and linked from journals. Open data are research output such as software and datasets licensed for re-use. The Open Data Handbook gives advice on legal, social and technical aspects of open data.

See the Open data subject guide for more detail on platforms to share data.

Discipline-based and general repositories are available to publish/share your data during your research (search the Registry of Research data Repositories) and preserve it on a long-term basis. 

Data journals focus primarily on data, not analysis, but can include links to articles. They are often open access and offer fast peer review. Examples include:

• Geoscience Data Journal, from the Royal Meteorological Society
• Scientific Data, from Nature publishing
• Journal of Open Archaeology Data, peer reviewed data papers describe archaeology datasets with high reuse potential
• Biodiversity Data Journal, an international, interdisciplinary journal for the publication of articles on original research data
• Earth System Science Journal, an international, interdisciplinary journal for the publication of articles on original research data

Digital Object Identifiers (DOIs) are unique identifiers that provide persistent access to published articles, datasets, software versions and a range of other research inputs and outputs. If you store data in a repository it will typically be allocated a DOI to enable the data to be cited.

DataCite registers and allocates DOIs for datasets, images, software and other research material enabling the location, identification and citation of research data.

The Digital Curation Centre provide examples for how to cite data and datasets within scholarly literature, another dataset, or any other research object: How to cite datasets and link to publications