Building ethics into the research design

Limits of confidentiality: professional and elite interviews

Concern about limits to confidentiality in social science research are not necessarily restricted to participants who are potentially vulnerable to risk of harm.  Elite and professional interviews can pose particular ethics challenges because of the specialist professional roles of your potential participants.  For example:

  • There may be a power differential if you are interviewing people who are used to setting their own agendas. 
  • Specialist professionals can be difficult to access, and it can be difficult to find alternative interviewees if they decline to take part. 
  • Elite professional participants are often very busy, and fit research into small amounts of time in between other meetings.  They can sometimes be intimidating, or impatient, and may hurry you through the initial consent process.  Perhaps they feel they have already consented - even though you have not been through your consent process - and want to get on with it.  Perhaps, too, they have less time than you (or they) had anticipated.
  • In other cases, a professional interviewee may have been nominated by their manager to take part in your research, raising the question of whether their participation is truly voluntary, or whether it is presented as a requirement of their job. 

These factors can put pressure on participants, and on you as a researcher, and so they can undermine processes of valid consent

In particular, fully informed consent depends on agreeing the limits of confidentiality in the research.  Confidentiality and anonymity can sometimes be limited by the very nature of participants’ professional roles, and so the confidentiality agreement in professional and elite interviews is absolutely central:  ask yourself what you can realistically promise your participants. 

Questions to consider

Can you extend confidentiality to those who choose not to participate in research?  For example, if someone declines to take part in your study, will you tell their manager?  What consequences could that have for them, and what can you do about that?

Can you realistically offer confidentiality or anonymity when you are interviewing a small number of high profile and distinctive individuals?  They may be identifiable to others who work in their area (including other participants). 

If you cannot guarantee that participants will not be identifiable, it may be better to offer to name them, and interview them ‘on the record’.  This may be more straightforward in some cases than an unrealistic promise of anonymity, but bear in mind that the information you get will be markedly different.

What if you are interviewing ‘on the record, and the participant reveals something ‘off the record’?  Can you use that information?  How?

For example, imagine you were studying the management of cross-department working in central government.  You interviewed a senior civil servant in one department, who was very critical of management practices in another department.  She explained, off the record, that a senior policy adviser in the other department was an alcoholic; this had caused problems in his team, and she had heard that disciplinary proceedings were being instituted against him.  The following week, you interviewed this senior policy adviser. He gave a glowing account, and did not reveal any tensions or difficulties.  What could you do?  Participants were aware of who was being interviewed for the study.  How could you probe for more information?  He might guess what had been said about him, and by whom.   When you come to write up the research, could you report the difficulties that the first interviewee had described?  How could you do that, without those comments being attributable to the first interviewee?

While this example is extreme, it is not unusual for participants to be critical of the work of other participants, or even to ask you what another participant has said.  You need to consider how you handle those different perspectives whilst maintaining confidentiality.

These considerations do not only apply to elite or professional interviews, but to any research that is focused within a group or organisation.  Difficulties can arise because participants are known to each other, or because those in positions of authority may be able to guess who has said what, when reading or hearing findings from your study.  See our ethics dilemma on unintended consequences for an example of this tension, involving research with students.

Finally, participants may not agree with what you say - particularly if you are evaluating their organisation or working practices:

  • What if a participant asks to read your report, but strongly disagrees with how you have interpreted what they said, or with the conclusions you have drawn? 
  • What if they demand changes? They could require that you withdraw their data from the study. 

A related issue is that an organisation may commission you to evaluate their service, but then the commissioners of your study may be unhappy with any critical conclusions that you draw. 

In general, the key message is that  you should be careful what you offer, and make sure it is agreed at the outset:

  • For participants, it may be better to offer interviewees the chance to check their interview note but not to comment on study outputs or reports (although you can still provide copies of these).
  • For project commissioners, you may want to have a written agreement at the outset of the project, which limits their influence on the substance of what you report, and sets out any agreements about whether you can publish the research findings more widely.  The research office in your organisation should be able to advise on these contractual issues.

Summing up

There are no easy answers to the questions outlined above.  What is critical is that you recognise the particular tensions in elite and professional interviews, and plan how you will address those in your research.  In particular, you must ensure that your confidentiality agreement with participants explicitly addresses:

  • the limits of confidentiality and anonymity;
  • the use of the data provided - and the participants right to comment or amend material.

For elite and professional interviews, in particular, it may be a good idea to get written consent, so you can be sure that participants understand and accept these limitations.

Because of the complexity and sensitivity of professional and elite interviews, we would also recommend that you take advice from experienced colleagues - don’t try to make difficult decisions on your own.  You can also refer also to your relevant professional guidelines and to our suggested additional reading.