Written by Brian Westra, Data Services Manager
This year’s International Open Access Week is emphasizing equitable foundations for open knowledge. Across the sciences, social sciences, and humanities there is a growing recognition of the importance of open access to research data. The FAIR (Findable, Accessible, Interoperable, and Reusable) data principles encompass a set of technical/ computational characteristics that enable data discovery and reuse. Data sharing can advance the rate of research discovery, provide the basis for new forms of computational and multi-disciplinary research, and on a personal level, may lead to increased citations of the related articles.
If you’re interested in the details, these principles are outlined in the NIH Strategic Plan for Data Science, but are just as applicable to NSF’s funding of Social, Behavioral and Economic Sciences, the digital humanities, and other domains. Others who are working on pathways for implementation include publishing coalitions and the US Federal Government’s Project Open Data.
To bring this to a more personal level of application, here are just a few ways in which you can start supporting open access to your own research data.
1. Use open, non-proprietary file formats whenever they are available and appropriate for your research. By using these formats you enable others to view and use your data without the need for specialty software, and provide some assurance that when software changes, or disappears, your data will still be usable by others (and yourself).
Data repositories, research funding organizations, and some publishers provide guidelines and suggested formats for data. The UK Data Services recommended formats cover more common file types. Your research domain may have specific standards, such as those for earth sciences, medical imaging, and chemistry.
Sometimes data will need to be converted to an open format, and it’s important to be aware of the considerations about what might happen to the data, or embedded information, during this process. The UK Data Services website and other sources outline these issues, and you can also contact us for assistance.
2. Deposit in repositories that will provide long-term, open access to and preservation of the data. Do not rely on supplemental files with articles that may be locked behind a paywall. Instead, seek out repositories that are sustainably funded, and provide open access to at least the metadata via a Digital Object Identifier (DOI) or other persistent identifier. Some funding organizations list their preferred data repositories with their data management planning guidelines, and you can also contact us for assistance.
In addition to external data repositories, the UI provides a data repository in which you can deposit your research data, and we can assist you with this as well.
3. When you deposit your data, insure that it is well-documented so that others can find it, understand it and know about any restrictions on its use. Most repositories provide standardized fields for you to submit this information. Document the collection/generation process, analysis, and any other computational workflow that would be necessary to understand or reproduce the data. You may also want to deposit a readme file for additional information, if the repository allows.
These are a few basic steps that you can take when you are ready to share your research data. With some time and effort, your data will be accessible and enable you to have a broader impact through sharing these products of your research.