Nigel King.

The Qualitative Research Interview

In Qualitative methods in organizational research
A practical guide
Edited by Catherine Cassel and Gillian Symon
(London, Sage Publications, 1994. 253 pp.)
Types of research interview and their uses
The qualitative research interview
The goal of any qualitative research interview covers a range of approaches to research
interviewing. Its goal is to see the research topic from the perspective of the interviewee, and to
understand how and why he or she comes to have this particular perspective. To meet this goal,
qualitative research interviews will generally have the following characteristics: a low degree of
structure imposed by the interviewer; a preponderance of open questions; a focus of specific
situations and action sequences in the world of the interviewee rather than abstractions and
general opinions. (pp.14-15)
A key feature of the qualitative research interview method is the nature of the relationship
between interviewer and interviewee.
In a quantitative study using structured interviews, the interviewee is seen as a research subject.
In contrast the qualitative researcher believes that there can be no such thing as a relationshipfree interview. The interviewee is seen as a participant in the research, actively shaping the
course of the interview rather than passively responding to the interviewers questions.
The diametrical opposite of the qualitative research interview is the structured interview.
In this the interviewer uses a detailed schedule with questions asked in a specific order.
Every effort is made to control the way these questions are asked in order not to bias the
responses of different interviewees.
Questions are mostly closed
The emphasis is heavily on easily quantifiable information (p.15)
Another type of research interview lies somewhere between the qualitative research interview
and the structured interview. The authors refer to it as structured open-response interview.
It uses an interview schedule (like the structured interview)
Questions follow in a set order
However, many more of them are open-ended
There is flexibility in the order groups of questions are asked
These interviews tend to focus on factual information and general evaluative comments.
The type of interview to be used in a study depends on the nature of the research question to be
addressed.
Unfortunately, it is all too common to come across studies in which decisions about
methodology seem to have preceded any careful consideration of the research question,
leaving the researcher with data which do not properly address his or her question (p.16).

pp. 16-17
The qualitative research interview is most appropriate:
1. Where a study focuses on the meaning of particular phenomena to the participants.
2. Where individual perceptions of processes within a social unit are to be studied prospectively,
using a series of interviews.
3. Where individual historical accounts are required of how a particular phenomenon developed.
4. Where exploratory work is required before a quantitative study can be carried out.
5. Where a quantitative study has been carried out, and qualitative data are required to validate
particular measures or to clarify and illustrate the meaning of the findings.
The structured interview is most appropriate:
1. Where testing of a formal hypothesis (-ses) is desired.
2. Where data gathered can be readily (and meaningfully) quantified.
3. Where factual information is to be collected and the researcher knows in advance the type of
information the participants will be able to provide.
4. Where a postal survey would be likely to produce a very poor response rate.
5. Where the generalizability of previously obtained qualitative findings is to be tested.
The structured open-response interview is most appropriate:
1. Where a quick, descriptive account of a topic is required, without formal hypothesis-testing.
2. Where factual information is to be collected, but there is uncertainty about what and how
much information participants will be able to provide.
3. Where the nature and range of participants likely opinions about the research topic are not
well known in advance, and cannot easily be quantified.
Description of the method
Constructing and carrying out qualitative research interviews
Four steps in constructing and using qualitative research interviews (p.18):
1 defining the research question;
2 crating the interview guide;
3 recruiting participants;
4 carrying out the interviews.
Defining the research question. The research question should focus on how participants describe
and make sense of particular element(s) of their lives.
The primary concern not to quantify individual experience.
Researcher should not frame the research question in a way which reflects his/her own
presuppositions or biases.
Creating the interview guide (p.19)

In the qualitative research interview there is no formal schedule of questions to be asked wordfor-word in a set order.
Interview guide lists topics the interviewer should attempt to cover in the course of the interview.

Three sources for topics

The research literature

The interviewers personal

knowledge and experience of
the area

Informal preliminary work

(unstructured discussions with
people who have personal
experience of the research
area)

Interview guide may be modified through use: adding probes or even whole topics which had
originally not been included, but have emerged spontaneously in the interviews; dropping or
reformulating those which are incomprehensible to participants or consistently fail to elicit
responses in any way relevant to research question(s).
Recruiting participants for the study (p.20)
In deciding how many participants to recruit, the amount of time and resources available is a
critical factor. Time needed to undertake a study based on qualitative research interviews
(transcribing, analyzing transcripts, developing interview guide, recruiting participants, carrying
out interviews, traveling to and from them, feedback findings) should not be underestimated.
Participants must be assured of confidentiality and should be told clearly for whom the
research is being carried out and what it hopes to achieve.
These points should be repeated at the start of the interview itself, and permission to taperecord the interview must be obtained.
The interviewee should be told what kind of feedback about the study he/she will receive and
approximately when this will happen.
Practical issues in carrying out qualitative interviews (p.21)
Flexibility is the most important factor here.
Starting the interview
It is best to open with a question which the interviewee can answer easily and without potential
embarrassment or distress. More difficult or sensitive questions should be held back until some
way into the interview, to give time for both interviewer and interviewee to relax and feel they
are getting to know each other.
Good for opening questions requests for factual or descriptive information.
Phrasing questions.

 It is advisable to avoid multiple questions where in one question you ask about more than
one thing, and phrase the questions as simply as possible.
Leading questions imposing your own perceptions on the interviewee should be avoided.
You need to beware of assuming that the answer to a question is so obvious that it need not be
asked.
You should not tell the interviewee what his/her answers mean So, what youre really
saying is.
It is sometimes useful to repeat an answer back to the interviewee to seek clarification.
Ending the interview (p.22)
Avoid ending the interview on a topic which is difficult, threatening or painful.
The concluding questions should steer the interview towards positive experiences.
It may be useful to finish by giving the interviewee the opportunity to make any comments
about the subject at hand which have not been covered in the rest of the interview.
Difficult interviews and some tips about how to deal with them (pp.22-24)
The uncommunicative interviewee
Some interviewees may be defensive about the topic being discussed; they may be trying to get
the interview over with as quickly as possible; they may think that brief answers are what you
want; they may just be habitually laconic.
The risk of monosyllabism can be reduced before the interview begins by being quite clear
about how much time you require and that the interviewee has the time available and by
stressing the anonymity of all answers.
If the interviewee is still unresponsive, the first thing to check is that you are phrasing
questions in as open a way as possible. E.g. question like How useful did you find the course?
may lead to a one or two word answer. But if you rephrase it to In what ways if any did you
find the course useful? would be more likely to elicit expansive replies.
If you are anyway facing problems, a useful strategy is silence. Instead of moving on to your
next question when the interviewee provides another short response, pause for a few seconds.
This will mean that you would like to hear more on the subject.
The over-communicative interviewee
An opposite problem the interviewee who repeatedly indulges in long-winded digressions from
the interview topic.
Although some degree of digression should be tolerated, if the interviewee is repeatedly staying
far from your questions without adding anything of significant interest, you need to attempt to
impose more direction on the course of the interview. This should be done as subtly as possible,
to avoid causing offence.
A good strategy is to interrupt the digression politely at a natural pause or break and refer back
to an earlier point made by the interviewee which was relevant to your research question; Thats
very interesting. Could we go back to what you were saying earlier about as Id like you to tell
me more about

The high-status interviewee

When interviewing people of high status, it is important to set your relationship with them at am
appropriate level.
If you are over-familiar, or appear to show off your knowledge in their domain, you may cause
offence.
Conversely, if you are overly nervous or submissive you are likely to be patronized.
You need to be respectful especially in regard to their areas of professional or expert
knowledge but at the same time confident of the worth of what you are doing and of your own
expertise.
Interviews on emotionally charged subjects
When interviewees are finding an area difficult to talk about because of their emotional reactions
to it, make sure that you give them the time they require to answer your questions.
If their distress is great, let them known that it is perfectly all right for them to leave the question
altogether, or to return to it later if they feel able to.
The would-be interviewer
Some interviewees persistently ask the interviewer about their own opinions, experiences and so
on. It can be a good sign of the rapport established, but the danger is that if you simply concur
and state your views, you may bias the interviewees subsequent responses in the same way as
can happen with leading questions by implying that certain answers are more acceptable than
others.
A useful strategy would be to say to interviewees that you will be happy to answer any of their
questions at the end, but for now you would like to concentrate on their views.
Analysing data from qualitative interviews
(p.25)
This discussion of data analysis makes the assumption that the researcher has available full
transcripts of interviews. There is no satisfactory alternative to recording and fully transcribing
qualitative research interviews.
Although tape-recording is the most efficient, a participant might refuse to allow it; then the best
the interviewer can do is make brief memory-jogging notes during the interview and reconstruct
as much as possible immediately afterwards.
Familiarisation with the data
It is essential that the researcher is thoroughly familiar with the data before commencing any kind
of analysis.
Not only will the transcripts have to be read more than once, but it is also strongly recommended
that the researcher listens to each interview as well in order to take into account nuances of
speech, tone of voice, hesitations and other such paralinguistic information.
Approaches to analysing data
Miller and Crabtree (Doing Qualitative Research. Newbury Park, CA: sage. 1992) provide a
useful framework for summarising the various approaches to data analysis. They propose four
mail approaches: quasi-statistical, templates, editing and immersion/crystallisation.

Quasi-statistical Seeks to turn the textual data into quantitative data which can be manipulated
statistically.
Illustration content analysis selects a suitable unit of measurement single words, phrases or
themes and then categorises each unit found. Statistical analyses can be then carried out.
Content analysis is firmly within the quantitative, logical-positivist tradition, concerned with
hypothesis-testing, generalisability and the separation of the researcher from the data for the sake
of objectivity.
As such, it should not be used to answer research questions which are essentially qualitative.
However, elements of content analysis can be usefully applied in the early stages of some
qualitative analysis. (p.26)
Template here text is analysed through the use of an analysis guide, or codebook, consisting
of a number or categories or themes relevant to the research question(s).
Difference with content analysis codebook is revised through exposure to the textual data.
the pattern of themes emerging is interpreted qualitatively,
rather than statistically.
Codebook can be built upon existing knowledge (a priori) or developed from initial analysis of
the interview data (a posteriori).
Editing the interpreter enters the text much like an editor searching for meaningful segments,
cutting, pasting and rearranging until the reduced summary reveals the interpretive truth in the
text (Miller and Crabtree, 1992; also more about editing techniques see: Strauss, A.L. (1987).
Qualitative analysis for social scientists. Cambridge: Cambridge University Press.)
A key feature of most editing techniques is their cyclical nature: interpretations are repeatedly
compared with the original textual data. In grounded theory this process is called constant
comparison to achieve a point of theoretical saturation where additional analysis no longer
contributes to discovering anything new about a category (Strauss, 1987: 21).
Immersion/crystallisation researchers immerse themselves in the research subject (via
qualitative interviews, in the form of spontaneous conversations, observation, introspection and
reading of non-academic literature) over a prolonged period of time, and produce an account of
their findings through analytical reflection and intuitive crystallisation of meaning. (p. 27)
An example of data analysis can be found on p.27
Issues of reliability and validity (p.30)
Reliability
In quantitative research the measures used should produced the same results when applied to
the same subjects by different researchers.
Qualitative research, in seeking to describe and understand how people make sense of their
world, does not require researchers to strive for objectivity and to distance themselves from
research participants.
The issue of possible researcher bias cannot be ignored though. The findings cannot be a
product of researchers prejudices and prior expectations.

This can be guarded in two ways:

- researchers should recognise their presuppositions and in the analysis of the data
make an effort to set these aside.
- at the stage of coding for themes or categories, inter-rater comparisons can be used
(co-researchers coding can be compared to the one of the researcher)
On p.31 there is an example of a coding programme that helps checking the level of reliability.
Validity
In quantitative research a valid instrument is one which actually measures what it claims to
measure.
In qualitative research a study is valid if it truly examines the topic which it claims to have
examined.
The difference is that in quantitative research notions of validity centre on methods, and in
qualitative on the validity of interpretations.
The involvement of other people is crucial to considerations of validity in interpreting data from
qualitative research interviews.
The use of computers in qualitative analysis
On p. 32 some computer programmes used for transcript and text analysis, their advantages and
disadvantages are described.
Advantages and disadvantages of the method (p.33)
Advantages:
- one of the most flexible methods available;
- it can address quite focused questions about aspects of organisational life;
- it can be used to examine much broader issues;
- suits ideally to examining topics in which different levels of meaning need to be explored;
- it is a method which most research participants accept readily (partly due to the familiarity with
the interviews in general);
- an important fact is that most people like talking about their work but rarely have the
opportunity to do so with interested outsiders.
Disadvantages (p.34):
- developing an interview guide, carrying out interviews, and analysing transcripts are all highly
time-consuming;
- interviews are time-consuming for interviewees, thus it might be difficult to recruit them;
- difficult interviewees;
- data overload. There are three directions out of this situation: (1) there are the original aims of
the study and the researcher should regularly ask him/her-self whether the new information
adds value to the study; (2) there is literature describing other studies using qualitative research
interviews which can be consulted; (3) personal networking if the researcher is alone using
qualitative methods in an institution domnated by quantitative techniques.