Emma White: It may be legal, but is it right? Ethics and secondary data use
20 December 2016

Ethical research seeks to maximise benefit and minimise harm.

The Economic and Social Research Council revised its ethical framework in 2015, and, for the first time, specifically referenced the importance of ethics review for research using secondary data.

At the time, this created some concern about additional burden throughout the academic sector, but guidelines explaining that ‘light-touch’ review was appropriate in some circumstances eg when using aggregate data or data that are not identifiable, allayed those worries for the most part.

For reviewers on ethical panels, experience is needed to appropriately assess risk and suggest mitigation. For example, there is a risk of (unintentional) re-identification of individuals in de-identified data when two or more datasets are linked together following the Administrative Data Research Network (ADRN) model, since additional information may generate more unique sets of characteristics at more granular levels of geography. Allowing access in carefully managed secure settings with the independent application of disclosure control to outputs is one way to address this risk.

The academic sector has a tradition of transparent ethical review and is generally well placed to accept and adapt to updated ethical review requirements. Nonetheless, ethical review in many cases is still considered to be a barrier to be overcome, rather than a cornerstone of rigorous and robust research.

The necessity of ethical review for research projects supported by the ADRN has led to the establishment of the National Statistician’s Data Ethical Advisory Committee, which is open to organisations that would not necessarily have internal ethical review processes of their own.

In general however, non-academic researchers do not necessarily apply ethical review to their use and re-use of secondary data, identifiable, linked or otherwise. Strictly speaking, they are not obliged to do so, once the requirements of the Data Protection Act (DPA), 1988, and other relevant legislation are met.

The fact that something is legal does not mean that it is ethical. (The converse may not be true either.) There is ongoing confusion and concern about the re-use of personal data, yet people still share information extensively with organisations and one another eg on social media – despite not always fully appreciating what can and cannot subsequently be done with it. At the same time, data that are publicly available can still be personal data as defined by the DPA, 1988, and must be treated accordingly.

The Digital Economy Bill, 2016-17, which is currently progressing through parliament, and the General Data Protection Regulation, to be implemented in 2018, will soon need to be taken into account with respect to research using secondary data, but neither is designed to answer ethical questions.

Informed consent is one part of the answer to the “is it ethical?” question, but it is not always feasible, especially for secondary data research purposes, where the researcher may not know the identities of the individuals in the study in the first place. The Information Commissioner’s Office takes the view that consent is one but not the only gateway to legal and ethical data sharing for research purposes, and it is very willing to engage in discussion to facilitate research while ensuring that the rights of data subjects are protected.

When secondary data research starts with ethics, we have the chance to transform our mind set to really get to grips with the public perception difficulties around re-use of data. We can even frame our early thoughts from the first person perspective and ask ourselves how we would feel if our data were to be used in the ways being suggested.

Ultimately, no matter what the data source, ethical research, that is, research that is designed to do no harm, to be beneficial, and to be fair and transparent, leads to increased trust and loyalty in organisations and researchers, and over time enhances reputation and builds impact.

Dr Emma White is the Associate Director of the Administrative Data Research Centre for England, based in the University of Southampton. She is also Head of Administrative Data for NatCen Social Research. In October 2016, Emma took up the role of Deputy Chair of the Market Research Society’s Census and Geodemographics Group. She has a PhD in Mathematics, and is a specialist ethics reviewer for the Faculty of Social, Human and Mathematical Sciences in the University of Southampton.

Contact details

e.e.white@soton.ac.uk

emma.white@natcen.ac.uk

@edefaoite

Any views or opinions presented are solely those of the author and do not necessarily represent those of the MRS Census and Geodemographic Group unless otherwise specifically stated.