The D’s of Data: Data Availability Statements
The final guest post from Mel Oakley, and part of our Ds of Data series, this post is an output of the InFrame project ‘Driving good practice in data sharing and use of data availability statements to support funder requirements.’ All other outputs can be found at https://doi.org/10.17605/OSF.IO/ZP4TG.
This work is licensed under CC BY 4.0.
Introduction
Data* is amazingly flexible in that it is reusable in its own right and allows research to be reproduced. It can be used by the researcher who collected/generated it or by someone else for a number of reasons:
- to replicate work you have done and get the same results, confirming your work
- to take the work further, uncovering more possibilities and providing opportunities to build on the initial research
- to answer new research questions different from those the data was collected for and open up a completely new zone of research.
In order for others to use the data to confirm the results analysed in your publication, build on your work or create something new, they first have to find the data. There are many ways this can happen. Sometimes data is found serendipitously or through personal connections, or through other routes. It’s easy to assume that readers of an article will be able to find the data easily, but in practice this isn’t always the case.
To streamline this process, the easiest way for others to find associated data after reading your article is through a Data Availability Statement (DAS).
*data is any information you analyse to
produce the results of your research.
What is a DAS and why have one
A Data Availability Statement, sometimes called a Data Access Statement and often referred to as a DAS is a short statement within your paper, or papers metadata, that tells readers where your data can be found. It may be that you cannot share your data because it belongs to someone else or you don’t have the appropriate permissions to share it. A DAS is where you have the opportunity to explain why the data cannot be shared or, if it belongs to someone else, inform the reader of how to request access to it.
Including a DAS in your paper is good research practice.
In 2025 UKRN undertook a pilot project looking at exploring the prevalence of DAS in research articles and assess the quality of DAS and their usefulness. The project identified 8 elements that make up a good DAS and maximise usability:
- Describes where and how materials supporting the results reported in a research output can be accessed.
- Includes hyperlinks and persistent identifiers (e.g. Digital Object Identifier* or accession number where available) if materials are recorded or stored in a catalogue or repository.
- Explains why materials cannot be shared openly, for example to protect study participant privacy.
- If a barrier to immediate access exists, provides clear reasoning (e.g. sensitivity or volume) and with a clear point of contact (e.g. access committee with the likelihood of a timely response).
- Clearly separates raw data and any secondary data reuse that supports results and analyses.
- If appropriate, clearly states that no materials were produced in the research.
- Is positioned in a DAS Section of the paper.
*A Digital Object Identifier, or DOI,
is a type of persistent identifier:
a string of numbers, letters and
symbols used to permanently identify
an object and link to it on the web.
Writing a Data Availability Statement
If you aren’t sure what kind of data you have and therefore what to include in a DAS, this flowchart can help break the process down.
The university website holds a series of useful resources Data access statements – Research – University of St Andrews including some examples of DAS.
Many journals will have a section in their templates for a DAS. Alternatively, the DAS can be in a section at the end of the paper, usually after your acknowledgements.
If your research is not based on any new or pre-existing data sources, or if (as in many humanities research projects) the data are published works represented in the bibliography, then you will still need a statement to fulfill funder or institutional requirements. This may require more adaptation to your personal research, but can be as simple as:
“This study is not based on any new or pre-existing data.”
If you analysed information in order to produce your paper then you have used data. In some cases the information you analysed will have belonged to someone else, it may be from an archive or other data holder such as the NHS. In these cases, you still need to share information about the data you used. The DAS is the place to do this. Examples of DAS for secondary data include:
“This study brought together existing research data obtained upon request and subject to licence conditions from a number of different sources. A full list of the sources is available in the reference list.”
“This study is a re-analysis of existing data which are available at [insert DOI or URL]”
“Data used for this study were obtained from [insert details with URLs if possible] and are subject to access restrictions”
Reasonable request is insufficient
In the short term it may feel easier to use the DAS “data available on reasonable request”. However this is poor practice. Studies have shown that papers saying data is available on reasonable request are not actually shared when requested. This is highly frustrating when trying to advance research in those fields.
One of the reasons researchers use a “reasonable request” DAS is the need to finish organising and tidying the data so it can be easily shared when requested. In reality, the next project quickly comes along and this falls through the cracks, making it harder to share the data later. It adds to the researcher’s cognitive load. Uploading data to a repository and sharing the DOI in a DAS can help prevent this. If the data belongs to someone else, a DAS explaining how to contact the owner enables others to easily do this, again reducing the cognitive load for the researcher. It also provides the original researcher with a reminder of how they had access to the data, in case they want to revisit it.
Instead of saying “data is available on reasonable request” it is best to either share the data and include details of where to find it in the DAS or explain the conditions under which the data can be shared, why this is required and how to make the request or explain why the data cannot be shared at all.
DAS as a requirement
Like most universities, the University of St Andrews has a research data management policy (https://www.st-andrews.ac.uk/policy/research-open-research/research-data-management-policy.pdf). At St Andrews, this and other policies require researchers to include a DAS in all research publications, from articles to PhD theses. More than this, major funders such as UKRI and Wellcome require outputs funded by them to include a DAS. Finally, many journals require a DAS to be included as part of their publications.
If this post has been helpful for you in thinking about data availability statements, look out for our other posts in the ‘D’s of Data’ series.