Key informant interviews:

a practical guide for refugee and displacement researchers

Dr. Jeff Crisp

Refugee Studies Centre
University of Oxford

[email protected]
@jfcrisp

Interviews with key informants are an important component of any research project
dealing with refugee, displacement and humanitarian issues. Whether you are talking to
a politician, a government official, a UN or NGO employee or a civil society leader, they
provide an invaluable means of gaining access to factual and topical information; an
understanding of the historical context of your project; as well as ideas, insights and
opinions that can shed a new and different light on the evidence you have collected by
other means. In my experience, they can also be one of the most stimulating parts of
the research process, more enjoyable in many ways than administering a survey,
trawling through archival records or reading reams of secondary literature.

Unfortunately, however, too many refugee and displacement researchers adopt a rather
sloppy attitude towards key informant interviews, essentially making up their
methodology as they go along, rather than thinking through the purpose, format and use
of each interview in advance. On the one hand, this means that they do not get the full
benefit of the interviews that they manage to arrange and undertake. On the other hand,
when interviews are conducted in a deficient manner, and particularly when the
informant feels that they have had a negative experience, they will be much less likely
to agree to set aside any time for any future researchers that they meet in the course of
their work.

In that respect, the golden rule to remember is that your key informants are usually busy
people with many other items on their agenda. A bad interview will reflect badly on you
and the organization that you represent, will represent a lost opportunity in terms of the
research that you are undertaking, and will be a waste of time for a person who almost
invariably has more important and urgent things to do.

To avoid such negative outcomes, I would like to provide a set of 8 key guidelines,
based on my own experience as an interviewer and interviewee over the past 40 years.
While your approach will evidently have to be revised and adapted according to the

1
person you are meeting and the context in which you are meeting them, I believe that
these guidelines provide a solid basis for any key informant interviews that you
undertake in the course of your research. Please note that these guidelines do not apply
to interviews with refugees, displaced people and host population members.

1. Preparation

In my experience, too many researchers conduct interviews in an ad hoc manner,

without too much advance consideration of the purpose of the meeting, the issues that it
should cover or the potential difficulties that might arise in the course of the
conversation. To avoid such mistakes, preparation is essential.

Before you undertake an interview, therefore, make sure you know as much as possible
about the situation or issue that you want to discuss. As an interviewee, there is nothing
more frustrating than to be asked a question such as ‘what does UNHCR actually do?’,
‘how many displaced people are there in Kenya?’, or ‘has your country signed the
Kampala Convention?’ These are questions to which you should already know the
answer.

Similarly, before meeting any informant, try to find out as much as possible about the
organization that he or she is working for, the way that organization is structured, the
policies that it pursues and that person’s position within the institutional hierarchy. At the
same time, before undertaking any interview, try to develop a well-rounded
understanding of the operational context in which you are working.

Have there been any recent events or developments of which you should be aware? Is
the operational environment characterized by any particular issues, complexities or
controversies? What comments have been made about those controversies by other
stakeholders, including the national and international media, human rights organizations
and the United Nations?

2. Professionalism

If you want to be taken seriously as a researcher, then there are some minimum
standards of professionalism that you should always try to meet. Try to arrange
meetings in advance, rather than showing up at someone’s office in the hope that they
will agree to meet you. If at all possible, send them an email in advance, briefly
explaining who you are, what you are doing your research on and which issues you
would like to discuss.

2
In very practical terms, professionalism means being on time for your meeting - even if
your interlocutor is known to be a poor timekeeper. While you do not need to dress
smartly, unless, perhaps, you are meeting a senior politician, government official or
businessperson, try to look as clean and tidy as possible. Always present your
interlocutor with a business card when you first arrive for your meeting.

As a general rule, I always accept refreshments when they are offered to me, as to turn
them down might cause offense. More generally, make sure that you are aware of any
local customs with respect to what you wear, the way you sit and the title you use to
address the person you are interviewing. In order to maintain a high degree of
professionalism, I always try to avoid interviews that take place over drinks, dinner, at a
reception or in a moving vehicle.

3. Personal engagement

Effective interviewing is as much a social skill as it is an academic or technical one. And

in that respect, I always try to engage with my interlocutors at a personal level, before
moving on to the substance of each interview. There is no hard and fast advice that I
can provide on this matter, other than to underline the importance of finding a point of
connection with your interview, thereby breaking down the barrier that almost inevitably
exists between two people who are meeting each other for the first time and whose
backgrounds might well be very different.

To give some very simple examples, if you start a conversation with a sentence such as
“Well, it is a great pleasure for me to be in Amman again,” your interlocutor is likely to
be interested or pleased to know that you have visited their city before, and may well go
on to ask how it has changed since your last visit. If you mention that you are studying
sociology, law or international politics, you may find that your interlocutor has similar
academic interests. And a reference to mutual acquaintances often acts as a useful ice-
breaker. Needless to say, your efforts to find a point of connection in this way should be
done in as natural a manner as possible. If it doesn’t happen in a fairly spontaneous
way, then don’t try to push it!

4.. Language

If you can undertake an interview in the first language of the person that you are
speaking to, that is fantastic! It will almost certainly prove to be more fruitful than one
undertaken in a person’s second language, and even more so than one undertaken by

3
means of an interpreter. Unfortunately, if you and your interlocutor do not have a
common language, and the interpreter you are using is ineffective, there is not a great
deal that can be done to improve the situation!

Remember that your job as a researcher is to communicate, and in that respect the
language that you use is paramount. Try to speak in a clear, coherent and measured
manner, especially when your interlocutor is speaking in a second language. Unless
you are talking to an academic, a bureaucrat or UN official respectively, rigorously avoid
the use of all academic, bureaucratic or UN jargon. You cannot expect a local
government official to discuss the finer points of Agamben or Foucault, just as you
cannot expect an academic to have a full knowledge of the UN’s Cluster System!

5. Timing

With respect to timing, my advice is to avoid meetings as much as possible at the very
beginning or very end of the working day, when your interlocutor is most likely to have
other things on her or his mind. Preventing someone from having a lunch-break is also
generally not a good idea.

I usually try to keep my interviews down to around 45 minutes, although I also usually
ask at the beginning of each meeting how much time the other person can spare. When
I discover that the person I am interviewing does not have anything very important or
interesting to say, I try to keep the interview going for at least 20 minutes, so as to avoid
the impression of wanting to leave in a hurry.

This approach can have some amusing consequences. I once interviewed a

government official in northern Iraq who could not answer any of my questions. I
managed to keep some kind of conversation going for 20 minutes, and concluded by
thanking him for his very informative comments. He looked at me with a quizzical look
on his face and said, “Well, I’m very surprised to hear you say that I have been so
informative, as I only started this job today and haven’t really learned what I am
supposed to be doing.”

Finally, don’t try to cram in too many interviews in a day: two in the morning and two in
the afternoon is usually quite enough if you want to allow time for preparation, to read
through your notes and to get from one place to another.

4
6. Questions

Questions are at the heart of any key informant interview, and your effectiveness as a
researcher will depend to a great extent on your ability to pose them in the right manner.
Some key points:

a) Arrive at each interview with a pre-prepared list of the issues that you would like
to cover with your interlocutor. Do not, however, write those issues down as
specific questions, and certainly DO NOT run through a list of scripted questions
as if you were completing a survey or questionnaire. The whole point of an
interview is to have a free-flowing conversation in which you can adapt and add
to your questions as you go along and respond to the remarks your interlocutor
has made.

b) Keep your questions short and to the point. Be careful about expressing strong
opinions or, even worse, getting into an argument with your interviewee. I often
find it useful to distance myself from opinions by using the following format: “As
you might have seen from their recent press release, Oxfam has suggested that
it is too soon to encourage the refugees here to repatriate. How do you respond
to their comment on this matter?”

c) I like to use questions with a numerical basis. For example, “What are the three
biggest challenges confronting you in this operation?” “If you were given
additional resources for IDP education, what would your top four priorities be in
terms of increased expenditure?” “Can you identify three main reasons why
coordination between humanitarian and development actors has proved to be so
difficult in this operation?”

d) If I suspect that an interlocutor might not want to provide me with honest

answers, I sometimes ask them a question to which I already know the answer,
so as to check on how truthful they are prepared to be. It’s a cheap trick but it
can work!

7. Answers

You evidently need to keep a record of your interviews, and there are two main ways of
doing this: electronic recording or taking manual notes. I much prefer the latter, mainly
because some people are intimidated by the sight of recording devices. There is also
more to go wrong technically. I once did an hour-long interview with a researcher that I

5
had to repeat for a second time because the person had pushed the wrong button on
their device!

If you are taking notes by hand, then keep taking notes by hand! There is nothing more
dispiriting than to give an interview to a person who seems to think that nothing you say
is worth writing down!

I find it useful to go through my interview notes the day that I have written them, to
highlight key points made in the course of the interview and to identify any issues that
need to be exploited further with future interlocutors. I personally do not produce typed
transcripts of my interviews and have never used any kind of coding method to analyze
their contents.

8. Ending the interview

I always conclude my interviews by thanking the interlocutor for their cooperation and by
asking them if they have any questions from me. I usually ask if there are any other
particular people that I should talk to, and whether they have an email address for them.
If the interview has gone well, I may ask the interlocutor if they have any appropriate
documents that they can share with me, especially if they have mentioned any specific
reports in the course of the interview.