Improving the Transparency of Data Access Conditions in the SSH Domain: Recommendations based on a small-scale analysis of the conditions applied to restricted access datasets

Abstract

The benefits of making data available for reuse are recognized by many. While some datasets can be openly available, others contain sensitive or personal data that needs to be protected. To allow the sharing of the latter datasets, many trustworthy digital repositories provide options to publish data restricted access. However, detailed standardized information about the access conditions for these restricted access datasets are often lacking from the metadata. Researchers interested in reusing these datasets can thus not judge whether they are eligible to reuse data and under what conditions.  

To get a better understanding of how we can increase the transparency of access conditions, this paper aimed to investigate the access conditions and procedures that are commonly applied by depositors within the (Dutch) Social Sciences and Humanities (SSH) research community. The results of a survey that was conducted (n=45) indicated that various conditions are applied and while some datasets can have few restrictions, for others reuse is highly restricted. Most respondents limit reuse to research purposes and prohibit commercial use. Some datasets are available for students or teaching, but often with additional requirements. A large majority of respondents required a motivation letter to evaluate before allowing reuse. Notably, respondents often chose ‘it depends’ when asked whether a specific condition was applied, showing a lot of nuances in the conditions and the evaluation of access requests. An important result from our survey was that clear procedures and decision-making guidelines seem to be lacking for many respondents. Requests are often evaluated ad-hoc and through email. Decisions are said to require an evaluation of the quality of the application, yet the evaluation criteria seem to be rarely specified and explicitly communicated at the time of data deposit.

Based on the results of the small-scale survey, we conclude with a set of six recommendations directed at researchers and infrastructure providers outlining how information about access conditions and procedures can be made more transparent in the future. This work should be seen as a starting point to improve the Accessibility and Reusability of restricted access datasets in the SSH domain.