Regressions, Observations, and Figures, Oh My! An Exploration of Ukds Securelab Output Requests

Trusted Research Environments (TREs) routinely document and report which of their datasets are used by researchers, which are the most popular, how many times researchers request that their research results be released from secure computing environments, etc. (e.g., UK Data Service, 2025; Bennett Institute for Applied Data Science, 2022). They also dictate the types of allowable materials researchers can request be released. However, documenting the specific components of those output requests has not previously been done. As part of the SOCRATES project, a DARE UK-funded project to evaluate the potential utility of the Semi-Automated Checking of Research Outputs tools, the UK Data Service has surveyed a selection of UKDS SecureLab projects and their output requests.

The goal of this analysis was to identify the most prevalent types of output tables and figures, patterns of statistical program use, and the reasons why a given output request may be rejected. Our findings will be of interest to other social science-focused data archives and TREs, and provide a comparison for TREs in other disciplines (such as medical research) interested in the types of outputs their researchers are likely to produce with the data in their collections. This presentation will present the findings of this analysis and present recommendations for future TRE researcher training to address patterns of output request rejections.

UK Data Service. (2025). From strength to strength: Data, impact and future (Annual Report Nos. 2023–2024). UK Data Service. https://ukdataservice.ac.uk/app/uploads/ukdsannualreport2023-2024.pdf

Bennett Institute for Applied Data Science. (2022). Bennett Institute for Applied Data Science First Annual Report (Annual Report 2022). University of Oxford. https://www.bennett.ox.ac.uk/annual-reports/2022/