Re-use of datasets needs to be given careful consideration by both the researcher and the research ethics committee, especially with regard to presumed consent and the potential risk of disclosure of personal data. This applies to the researcher collecting or producing the original data and the researchers re-using the data. Researchers who initially collect the data should be aware of the ESRC expectation that other researchers will re-use the data, given appropriate permissions. When researchers apply for funding they must include a statement on data sharing, and if data sharing is deemed impossible they must present a strong argument to justify this (see the ESRC Research Data Policy). The original researcher should therefore take into account, when obtaining consent, the long-term use including the potential for data linkage and preservation of data. Further advice on securing consent for data re-use, as well as exemplar consent forms, are available at the UK Data Service’s website. We strongly advise researchers to seek advice from the relevant ESRC data service provider (currently the UK Data Service) at the outset of the project to clarify how issues of confidentiality and sharing are to be addressed.

In cases where the researcher believes it may not be possible to sufficiently anonymise data to be available at the appropriate ESRC Data Service Provider (for example the UK Data Service) through a standard End User License, they should seek advice from the ESRC Data Service Provider, as alternative secure access methods could be arranged.

Data re-use falls into three categories:

  • Anonymised non-sensitive data or data where there is minimal risk of disclosure of the identity of individuals. These data have fewer legal restrictions.
  • Data protected by legislation, for example personal data (Data Protection Act 1998) and census data (Statistics and Registration Service Act 2007). Personal data provided for administrative purposes are protected under the Data Protection Act 1998. Here, the data producer and data custodian have a strong interest in how researchers will access the data, and may control access.
    This category of data may only be available via ‘safe settings’, see for example ESRC-funded Administrative Data Research Network and the Safe Haven Farr Institute collaboration, which have been set up to ensure the safe use of administrative data and patient and research data for medical research. 
  • Ethically sensitive data. This could include large datasets such as the National Child Development Study (NCDS), where the inclusion of information such as a birth date or postcode makes disclosure possible (perhaps via a link to other datasets). The governance arrangements and requirements for access to these datasets ensure compliance with the Data Protection Act 1998 and mitigate the risk of disclosure. Small-sample qualitative studies can also raise ethical sensitivities in relation to re-use, because of the potential identifiability of participants, and in cases where methods could generate identifiable data (eg photographs). This should be addressed in data management plans.