CREATE & MANAGE DATA
CONSENT & ETHICS
CONSENT / GAINING CONSENT
Discussing and obtaining consent for participation in research, use of information gathered for analyses, publications and outputs, and sharing data beyond the research can be a one-off occurrence or an ongoing process. Both approaches have advantages and disadvantages.
One-off consent is simple, practical, avoids repeated requests to participants, and meets the formal requirements of most Research Ethics Committees.
Consent is gained early in the research process and covers all aspects of participation and data use. For research where no confidential or sensitive information is gathered (e.g. surveys), or where data collection is a one-off event, this is usually sufficient and most practical. Criticism of this method is that it places too much emphasis on 'ticking boxes'. It can also be difficult to gain consent in one go in exploratory research or where not all data uses, research outputs and even methods are known in advance.
Process consent is considered throughout the research process and assures active informed consent from participants. This is generally recommended, such as in ESRC's Research Ethics Framework and is especially important in a research design which involves more than one point of contact with a respondent.
Consent for participation in research, for data use and for data sharing can be considered at different stages of the research, giving participants a clearer view of what participating in the research involves and what the data to be shared consist of.
Consent for various uses of data can also be sought after their research contribution is complete. However if a participant cannot be traced, the status of publishing or sharing the collected data may be uncertain. Participants should not be overwhelmed with too many issues of consent early on in the project. Process consent can be repetitive and irritating when participants have already expressed their willingness to participate.
Special considerations for consent occur in more complex research situations. Examples are research with patients, children, people with learning difficulties, asking about criminal matters and interviewing in the workplace.
Withdrawal from participation in research is always at the discretion of the participant. There should be no penalty for withdrawing and the participant is not required to provide any reason. However, the status of existing data, if any, is more ambiguous.
For large-scale longitudinal surveys no retroactive withdrawal is standard policy. This is because data have typically been highly anonymised. However, for qualitative studies the situation is more complex. In particular, for qualitative longitudinal work, withdrawal of existing data can be very damaging to a study with a small number of participants. However, no researcher wants to be in a position of retaining data if a participant wants complete and total withdrawal.
There is no single solution. Much depends on the state of processing or anonymity of the data. The Archive recommends that projects consider some of the following strategies for dealing with participants wishing to withdraw:
- If a participant requests retroactive withdrawal (of all their contribution), you should request a meeting to explain to the participant the costs to the project. If appropriate, the participant should be asked to reconsider their decision.
- Although the option of being able to withdraw as a project proceeds should be explained in information and consent forms, the option of retroactive withdrawal need not be made explicit as an option. It can be dealt with on a case-by-case basis, only if it arises.
- Projects should be clear in their own planning what they will do in the event of a participant continuing to insist on full withdrawal. In most cases, the ethical duty to the participant and the risk to a project for loss of goodwill are more than sufficient to offset any possible gain by retaining the data, and the project will accede to the participant's request in the final instance.
- If all fails and they do wish to withdraw retroactively, participants could be asked to consider whether some of the data could be used, for example, that which could be completely anonymised.