23 September 2014

ANDS Sensitive Data Guide

The Australian National Data Service (ANDS) has released its guide to Publishing and Sharing Sensitive Data [PDF], with the rubric that the guide
outlines best practice for the publication and sharing of sensitive research data in the Australian context. The Guide follows the sequence of steps that are necessary for publishing and sharing sensitive data, ... By following the sections below, and steps within, you will be able to make clear, lawful, and ethical decisions about sharing your data safely. It can be done in most cases!
Looking on the bright side the ANDS offers "key messages" -
  • The advantages of publishing your sensitive data will probably far outweigh any potential disadvantages when simple and appropriate steps are taken 
  • Publishing your data, or just a description of your data (i.e., the metadata), means that others can discover it and cite it 
  • You can publish a description of your data without making the data itself openly accessible 
  • You can place conditions around access to published data 
  • Sensitive data that has been confidentialised can be shared
ANDS characterises 'Sensitive data' as
data that can be used to identify an individual, species, object, process, or location that introduces a risk of discrimination, harm, or unwanted attention. Under law and the research ethics governance of most institutions, sensitive data cannot typically be shared in this form, with few exceptions. 
 The guide argues that sensitive data
has often been excluded from discussions about data publication and sharing. It was believed that sharing sensitive data is not ethical or that it is too difficult to do safely. This opinion has changed with greater understanding and use of methods to ‘de-sensitise’ (i.e., confidentialise) data; that is, modify the data to remove information so that participants or subjects are no longer identifiable, and the capacity to grant ‘conditional access’ to data. Requirements of publishers and funding bodies for researchers to publish and share their data have also seen sensitive data sharing increase.
Australia and other nations have significant and high-quality datasets that contain potentially-sensitive information. This includes epidemiological surveys of health, medical trial data, and ecological studies of endangered species. For reasons of finance, efficiency, conservation, and participant fatigue and disturbance, these can be reused rather than repeated. And in most cases, this can be done by removing or modifying aspects of the dataset that make its subjects identifiable (see Confidentialise your data) or attaching conditions to data access and reuse.
The advantages to sharing data go beyond meeting publisher or possible funder requirements. The benefits to the researcher and institution are clear. If data or a description of a dataset are published they are discoverable by others, and can thus be cited in favour of the original data collector or owner. This is the primary goal of data publication. There is also evidence that scholarly papers that are accompanied by published data are cited more often than those without. New collaborations and publications may eventuate if you choose to share your data with others for reuse. And finally, storing your data in a public repository affords secure and ongoing storage that may not be available in your current or changing research environment.