Secure data archives – coming soon!
Often, when we acquire sensitive/confidential sources of data for analysis, the data have to be destroyed at the end of the analysis. This is in keeping with data sharing agreements/contracts and potentially with data protection legislation.
However, there are downsides to this: the ability to reproduce results that somebody produced some time ago is key to scientific validation, and requires the original data source used. That can’t happen if we’ve destroyed the data.
The group are thinking about solutions for securely archiving data. MORE SOON!
Statistical Disclosure Control Handbook –
MARCH 2019 BETA RELEASE
This Handbook will provide analysts and staff responsible for statistical disclosure control checks with guidance about how to assess statistical results produced from confidential sources of data, before releasing from their secure data environments.
Intended to act as a handy reference, the Handbook will provide guidance about managing the statistical disclosure control process; and will contain specific guidance about assessing a variety of statistics for statistical disclosure.
Archiving files in a Secure Data Environment
After a project has completed, what should be done with all the files used? This includes data files, syntax, report and article drafts etc. This presentation and accompany notes produced by Yannis Kotrotsios (Cancer Research UK, formerly UK Data Archive) provides some useful tips.
Imports and Outputs
Key to the operation of a secure data environment is having procedures in place to manage requests of imports of files, data etc., by users, as well as removing files. This presentation by the UK Data Service provides some information about how they manage these processes for their Secure Lab.
Developing a career as a secure data access professional? Check out our Competency Framework. Designed for staff working in, and managing, secure data access facilities to develop their careers.
Installing and using R in a secure data environment
Many researchers now use R for their statistical analyses. As well as being free to use, the software offers a lot of flexibility when it comes to undertaking analysis, and many graduate courses now teach data analysis with R as a matter of course.
R generally relies on using the internet, to go and find packages to use.
But secure data environments don’t generally permit access to the internet. However, there is a way to install R packages and work with them locally inside the secure data environment.
Here are some instructions that should help (developed by Carlotta Greci, The Health Foundation, and Richard Welpton, Cancer Research UK).
If you would like to use our work please do so but please also acknowledge us!
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.