Kvale, Steinar (1996) Interviews.

An Introduction to Qualitative
Research Interviewing. London SAGE, Chapter 7: The
Interview Situation, pp. 124-135; Chapter 8: The Quality of the
Interview, pp. 144-159
 The Interview Situation
In the interview, knowledge is created inter the points of view of the
interviewer and the interviewee. The interviews with the subjects are
the most engaging stage of an interview inquiry. The personal contact
and the continually new insights into the subjects’ lived world make
interviewing an exciting and enriching experience. Different forms of
interview conversations were discussed in Chapter 2 and the mode of
understanding in the qualitative research interview described. In this
chapter I outline in more detail some guidelines and techniques for
carrying out interviews and give an illustration with an interview on
grading.

The Interview Conversation

The purpose of a qualitative research interview was described

earlier as obtaining qualitative descriptions of the life world of the
subject with respect to interpretation of their meaning. The interview
form treated here is a semistructured interview: It has a sequence of
themes to be covered, as well as suggested questions. Yet at the same
time there is an openness to changes of sequence and forms of
questions in order to follow up the answers given and the stories told
by the subjects. I will discuss the interview interaction in line with the
mode of understanding depicted earlier with respect to 12 aspects of
the interview: life world, meaning, qualitative, descriptive, specificity,
deliberate naiveté, focus, ambiguity, change, sensitivity, interpersonal
situation, and a positive experience (see Box 2,1 in Chapter 2).
The Interview Situation 125

An open phenomenological approach to learning from the inter-

viewee is well expressed in this introduction from Spradley (1979):

I want to understand the world from your point of view. I want to know
what you know in the way you know it. I want to understand the meaning
of your experience, to walk in your shoes, to feel things as you feel them,
to explain things as you explain them. Will you become my teacher and
help me understand? (p. 34)

The research interview is an interpersonal situation, a conversation

between two partners about a theme of mutual interest. It is a specific
form of human interaction in which knowledge evolves through a
dialogue. The interaction is neither as anonymous and neutral as when
a subject responds to a survey questionnaire, nor as personal and
emotional as a therapeutic interview. Patients seek therapists for help:
They are motivated to be as open as possible with the therapist, with
whom a trusting relationship is established over time. In a research
setting it is up to the interviewer to create in a short time a contact
that allows the interaction to get beyond merely a polite conversation
or exchange of ideas. The interviewer must establish an atmosphere
in which the subject feels safe enough to talk freely about his or her
experiences and feelings. This involves a delicate balance between
cognitive knowledge seeking and the ethical aspects of emotional
human interaction. Thus, at the same time that personal expressions
and emotions are encouraged, the interviewer must avoid allowing the
interview to turn into a therapeutic situation, which he or she may not
be able to handle.
The interviewer has an empathic access to the world of the inter-
viewee; the interviewee’s lived meanings may be immediately accessi-
ble 1n the situation, communicated not only by words, but by tone of
voice, expressions, and gestures in the natural flow of a conversation.
The research interviewer uses him- or herself as a research instrument,
drawing upon an implicit bodily and emotional mode of knowing that
allows a privileged access to the subject’s lived world.
A research interview follows an unwritten script, with different
roles specified for the two actors. The implicit rules of their interaction
become visible when they are broken, such as in thls interview
exchange with an unemployed man about traveling, in which the
interviewer is caught off guard when the subject reverses the roles:
126 I nte rVi ews

Subject: When you are on vacation there is some silly time factor, the
only thing you have time for is to go down and throw yourself
on the beach. Do you sunbathe?
Interviewer: What?
S: Do you sunbathe?
I: Well, no I do not.
S: You have a nice color.
I: I don’t spend one single summer day on that, but as a whole I look
tanned. Furthermore I get very easily tanned, I only need to
put one finger out of the window to catch the sun.
S: A lot of people would envy you that.
I: Well, where do we begin. What are you doing with your friends?
(Berg Sorensen, 1988, p. 124).

The conversation in a research interview is not the reciprocal

interaction of two equal partners. There is a definite asymmetry of
power: The interviewer defines the situation, introduces the topics of
the conversation, and through further questions steers the course of
the interview. This was the case in the rather open interview reported
by Giorgi (Chapter 2). Socrates’ interview, despite the conversational
partners being formally equal and the polite introduction, took the
form of harsh interrogation, relentlessly driving Agathon around in
his contradictory conceptions of love and beauty, until Agathon
throws in the towel and concedes that he knows nothing of what he
was talking about (Chapter 2).
Advance preparation is essential to the interaction and outcome of
an interview. A substantial part of the investigation should take place
before the tape recorder is turned on in the actual interview situation.
The key issues of the interview concern what, why, and how: mfinf—
acquiring a preknowledge of the subject matter to be investigated;
why—formulating a clear purpose for the interview; and for—being
familiar with different interview techniques and deciding which to
apply in the investigation. Also, before the first interviews in a study
are undertaken, thought should have been given to how the interviews
will be analyzed and how the findings will be verified and reported.
Research interviews vary on a series of dimensions. They differ in
degree of structure, from well-organized interviews that follow a
the Interview Situation 127

sequence of standard questlon formulations, to open interviews where

specific themes are in focus but without a predetermined sequence and
formulation of questions. Sometimes only a first, topic-introducing
question is asked and the remainder of the interview proceeds as a
follow-up and expansion on the interviewee’s answer to the first
questions, such as in the interview on learning reported by Giorgi. The
interviews also differ in their openness of purpose; the interviewer can
explain the purpose and pose direct questions from the start or can
adopt a roundabout approach, with indirect questions, and reveal the
purpose only when the interview is over.
The interviews can differ further in their emphasis on exploration
versus hypothesis testing, as mentioned in the d’iscussion of design.
Interviews also vary concerning description versus interpretation. The
interviewer might seek mainly to obtain nuanced descriptions of the
phenomena investigate d or can, during the interview, also attempt
to clarify and interpret the descriptions together with the subject.
Interviews also vary on an intellectual-emotional dimension, from a
rational logical discourse between interviewer and subject analyti-
cally clarifying conceptions of the phenomena investigated, to the
interviewer attempting to get spontaneous and emotional descriptions
of, and reactions about, a topic. Two extreme interviews on the
intellectual-emotional dimension were presented earlier—the discur-
sive argumentation of Socrates and the emotional therapeutic inter-
change reported by Rogers.

Framing the Interview

The interview is a stage upon which knowledge is constructed

through the interaction of interviewer and interviewee roles. Some
directions are suggested here for setting the interview stage so the
interviewees will be encouraged to put words to their points of view
on their lives and worlds. The directions pertain to interviews with
middle-class persons in Northern Europe and North America. In other
cultures, different norms may hold for interactions with strangers
concerning initiative, directness, openness, and the like.
The interviewees should be provided with a context for the inter-
view by a briefing before and a debriefing afterward. The context is
izs I n te r V i e ws

introduced wlth a briefing in which the interviewer defines the situ-

ation for the subject; briefly tells about the purpose of the interview,
the use of a tape recorder, and so on; and asks if the subject has any
questions before starting the interview. Further explanations about
the interview investigation should preferably wait until the interview
is over.
The first minutes of an interview are decisive. The subjects will want
to have a grasp of the interviewer before they allow themselves to talk
freely, exposing their experiences and feelings to a stranger. A good
contact is established by attentive listening, with the interviewer
showing interest, understanding, and respect for what the subject says;
at the same time, the interviewer is at ease and clear about what he or
she wants to know.
The initial briefing should be followed up by a debTie[ing after the
interview. At the end of the interview there may be 5ome tension or
anxiety, because the subject has been open about often personal and
emotional experiences and may be wondering about the interview’s
purpose and how it will be used. There may perhaps also be feelings
of emptiness; the subject has given much information about his or her
life and may not have received anything in return. This being said, a
common experience after research interviews is that the subjects have
experienced the intervlew as gentlinely enriching, have enjoyed talk-
ing freely with an attentive listener, and have sometimes obtained new
insights into important themes of their life world.
The interaction can be rounded off by the interviewer mentioning
some of the main points learned from the interview. The subject may
then want to comment on this feedback. The interaction can thereafter
be concluded by the interviewer saying, for example, “I have no
further questions. Do you have anything more you want to bring up,
or ask about, before we finish the interview?” This gives the subject
an additional opportunity to deal with issues he or she has been
thinking or worrying about during the interview.
The debriefing is likely to continue after the tape recorder has been
turned off. After a first gasp of relief, the interviewee may bring up
topics he or she did not feel safe raising with the tape recorder on.
And the interviewer can now, insofar as the subject is interested, tell
more fully about the purpose and design of the interview study.
The Interview Situation 129

The lived interview situation, with the lnterviewee s voice and

facial and bodily expressions accompanying the statements, provides
a richer access to the subjects’ meanings than the transcribed texts will
later. It may be worthwhile for the interviewer to set aside 10 minutes
of quiet time after each interview to recall and reflect on what has
been learned from the particular interview, including the interper-
sonal interaction. These immediate impressions, based on the inter-
viewer’s empathic access to the meanings communicated, may—in the
form of notes or simply recorded onto the interview tape—provide a
valuable context for the later analysis of transcripts.

The Interview Guide

An interview guide indicates the topics and their sequence in the

interview. The guide can contain just some rough topics to be covered
or it can be a detailed sequence of carefully worded questions. For the
semistructured type of interview discussed here, the guide will contain
an outline of tO 1CS to be covered, with suggested questions. It will
depend on the particular design chosen whether the questions and
their sequence are strictly predetermined and binding on the inter-
viewers, or whether it is up to an interviewer’s judgment and tact how
closely to follow the guide and how strongly to pursue an individual
subject’s answers.
Each interview question can be evaluated with respect to both a
thematic and a dynamic dimension: thematically with regard to its
relevance for the research theme, and dynamically with regard to the
interpersonal relationship in the interview. A good interview question
should contribute thematically to knowledge production and dynami-
cally to promoting a good interview interaction.
Tfiumn/icnJJy the questions relate to the topic of the interview, to
the theoretical conceptions at the root of an investigation, and to the
subsequent analysis. The questions will be different when interviewing
for spontaneous descriptions of the lived world, or interviewing for a
conceptual analysis of the person’s concepts of a topic. Simply ex-
pressed, the more spontaneous the interview procedure, the more
likely one is to obtain spontaneous, lively, and unexpected answers
from the interviewees. And vice versa: The more structured the
130 I n te rV i e ws

interview situation is, the easier the later structuring of the interview
by analysis will be.
In line with the principle of “pushing forward” in an interview proj-
ect, the later stages should be taken into account when preparing the
interview questions. If the method of analysis will involve categorizing
the answers, then clarify continually during the interview the mean-
ings of the answers with respect to the categories to be used later. If
a narrative analysis is to be employed, then give the subjects ample free-
dom and time to unfold their own stories, and follow up with ques-
tions to clarify the main episodes and characters in their narratives.
Dynamically, the questions should promote a positive interaction;
keep the flow of the conversation going and motivate the subjects to
talk about their experiences and feelings. The questions should be easy
to understand, short, and devoid of academic language.
A good conceptual thematic research question need not be a good
dynamic interview question. When preparing an interview it may be
useful to develop two guides, one with the project’s main thematic
research questions and the other with the questions to be posed during
the interview, which takes both the thematic and the dynamic dimen-
sions into account.
Table 7.1 depicts the translation of thematic research questions in
the grading study into interview questions to provide thematic knowl-
edge and contribute dynamically to a natural conversational flow. The
abstract wording of the research questions would hardly lead to off-
the-cuff answers from high school pupils. The academic research
questions need to be translated into an easy-going, colloquial form to
generate spontaneous and rich descriptions. One research question
can be investigated through several interview questions, thus obtain-
ing rich and varied information by approaching a topic from several
angles. And one interview question might provide answers to several
research questions.
The roles of the “why,” “what,” and “how” questions are different
in research versus interview questions. It has been repeatedly empha-
sized that when designing an interview project, the “why” and “what”
questions should be asked and answered before the question of “how”
is posed. In the interview situation, the priority of the question types
change. In the interview itself, the main questions should be in a
descr lptive form: “What happened and how did it happen?” “How
the InteTview Sitnation 131

TABLE 7.1 Research Questions and Interview Questions

Research Questions Intert iew Questions

Do you find the subjects you learn

important?

Which form of learning motivation Do you find learning

dominates in high school? interesting in itself?

What is your main purpose

in going to high school?

Do the grades promote an external, Have you experienced a conflict

instrumental motivation at the between what you wanted to read
expense of an intrinsic interest (study) and what you had to read to
motivation for learning? obtain a good grade?

Does learning for grades socialize Have you been rewarded with money
to working for wages? for good grades?

Do you see any connection between

money and grade5?

did you feel then?” “What did you experience?” and the like. The aim
is to elicit spontaneous descriptions from the subjects rather than to
get their own, more or less speculative explanations of why something
rook place. “Why” questions about the subjects’ own reasons for their
actions may be important in their own right. Many “why” questions
in an interview may, however, lead to an intellectualized interview,
perhaps evoking memories of oral examinations. Figuring out the
reasons and explanations for why something happened is primarily
the task of the investigator.

Interview Questions

The research interview proceeds rather like a normal conversation

but has a specific purpose and structure: It is characterized by a
132 I n t e rV i e ws

systematic form of questioning. The interviewer’s questions should be

brief and simple. In the life world interviews described here, an
opening question may ask about a concrete situation. The different
dimensions introduced in the answer can then be pursued. The deci-
sive issue is the interviewer’s ability to sense the immediate meaning
of an answer and the horizon of possible meanings that it opens up.
This, again, requires a knowledge of, and interest in, both the theme
and the human interaction of the interview. Decisions about which of
the many dimensions to pursue that are introduced by a subject’s
answer will depend on the purpose and content of the interview, as
well as on the social interaction in the interview situation.
Box 7.1 depicts some main types of questions that may be useful in
the semistructured interview form treated here. A more extended
discussion of interview questions is given by Seidman (1991). In
addition to paying attention to the thematic and dynamic aspects of
the questions, the interviewer should also try to keep in rriind the later
analysis, verification, and reporting of the interviews. Interviewers
who know what they are asking about, and why they are asking, will
attempt to clarlfy the meanings relevant to the project during the
interview, obtaining a disambiguation of the statements made, and
thereby provide a more trustworthy point of departure for the later
analysis. Such a process of meaning clarification during the interview
may also communicate to the subjects that the interviewer actually is
listening to and interested in what they are saying. Ideally, the testing
of hypotheses and interpretations is finished by the end of the inter-
view, with the interviewer’s hypotheses having been verified or falsi-
fied during the interview.
If an interview is to be reported, perhaps quoted at length, then
attempt when feasible to make the social context explicit during the
interview, and when possible the emotional tone of the interaction,
so that what is said is understandable for the readers, who have not
witnessed the live interview situation. Much is to be learned from
journalists and novelists about conveying the setting and mood of a
conversation.
The focus here has been on the interviewer’s questions. Active
listening—the interviewer’s ability to listen actively to what the inter-
viewee says—can be more important than the specific mastery of
questioning tecliniqtles. Therapists’ educatlon emphasizes their skills
The Interview Situation 133

Box 7.1

Types of Interview Questions

A. Introducing Questions: “Can you tell me about ... ?”;

“Do you remember an occasion when ... ?”; “What hap-
pened in the episode you mentioned?”; and “Could you
describe in as much detail as possible a situation in which
learning occurred for you?” Such opening questions may
yield spontaneous, rich, descriptions where the subjects
themselves provide what they experience as the main di-
mensions of the phenomena investigated. The remainder
of the interview can then proceed as following up of
dimensions introduced in the story told in response to the
initial questlon.
B. follow-Vp Questions: The subjects’ answers may be
extended through a curious, persistent, and critical attitude
of the interviewer. This can be done through direct ques-
tioning of what has just been said. Also a mere nod, or
“mm,” or just a pause can indicate to the subject to go on
with the description. Repeating significant words of an
answer can lead to further elaborations. Interviewers can
train themselves to notice “red lights” in the answers—such
as unusual terms, strong intonations, and the like—which
may signal a whole complex of topics important to the sub-
ject. The key issue here is the interviewer’s ability to listen
to what is important to the subjects, and at the same time
to keep in mind the research questions of an investigation.
C. Probing Questions: “Could you say something more
about that?”; “Can you give a more detailed description of
what happened?”; “Do you have further examples of this?”
The interviewer here pursues the answers, probing their
content but without stating what dimensions are to be
taken into account.
(continued)
134 I rite r Views

Box 7.1 Continued

D. Specifying Questions: The interviewer may also follow

up with more operationalizing questions, for instance:
“What did you think then?”; “What did you actually do
when you felt a mounting anxiety?”; “How did your body
react?” In an interview with many general statements, the
interviewer can attempt to get more precise descriptions by
asking “Have you also experienced this yourself?”
E. Direct Questions: The interviewer here directly intro-
duces topics and dimensions, for example: “Have you ever
received money for good grades?”; “When you mention
competition, do you then think of a sportsmanlike or a
destructive competition?” Such direct questions may pref-
erably be postponed until the later parts of the interview,
after the subjects have given their own spontaneous de-
scriptions and thereby indicated what aspects of the phe-
nomena are central to them.
F. Indirect Questions. Here the interviewer may apply
projective questions such as “How do you believe other
pupils regard the competition for grades?” The answer may
refer directly to the attitudes of others; it may also be an
indlrect statement of the pupil’s own attitude, which he or
she does not state directly. Careful further questioning will
be necessary here to interpret the answer.
G. St Fuct 2f Ting Questions: The interviewer is responsible
for the course of the interview and should indicate when a
theme has been exhausted. The interviewer may directly
and politely break off long answers that are irrelevant to
the topic of the investigation, for example by saylng, I
would now like to introduce another topic: ... ”
H. Silence: Rather than making rhe interview a cross ex-
amination by continually firing off questions, the research
interviewer can take a lead from therapists in employing
silence to further the interview. By allowing pauses in the
The Interview Situation

Box 7.1 Continued

conversation the subjects have ample time to associate and

reflect and then break the silence themselves with signifi-
cant information.
I. Interpreting Questions. The degree of interpretation may
involve merely rephrasing an answer, for instance: “You
then mean that . . ?” or attempts at clarification: “Is it
correct that you feel that . . ?”; “Does the expres-
sion ... cover what you have just expressed?” There may
also be more direct interpretations of what the pupil has
said: “Is it correct that your main anxiety about the grades
concerns the reaction from your parents?” More speculative
questions can take the form of: “Do you see any connections
between the two situations of competing with the other
pupils for grades and the relation to your siblings at home?”

as listeners, furthering an empathic active listening to the many

nuances and layers of meanings of what their patients tell them. Freud
(1963) recommended that therapists listen to their patients with an
“evenly hovering attention” to attend to the meaning of their accounts
(Chapter 4, Psychoanalytical Knowledge Production).
The importance of listening also appears in phenomenological and
hermeneutical approaches to interviewing (Chapter 3, sections titled
Hermeneutical Interpretation; and Phenomenological Description).
There is the phenomenological ideal of listening without prejudice,
allowing the interviewees’ descriptions of their experiences unfold
without interruptions from interviewer questions and the presuppo-
sitions these involve. A hermeneutical approach involves an interpre-
tative listening to the multiple horizons of meaning involved in the
interviewees’ statements, with an attention to the possibilities of
continual reinterpretations within the hermeneutical circle of the
interview. Attention will also be paid to the influence of the presup-
positions of the subjects’ answers as well as the presuppositions of the
interviewer’s questions.
 The Quality o/ the Interview
In the first parts of this chapter I will address issues of quality in
interview research. Criteria for evaluating the quality of a research
interview are suggested and related to characteristics of the interview-
ees and the interviewers. An interview by Hamlet is then presented as
an illustration of problems that may arise when using fixed criteria for
appraising the quality of an interview. Thereafter, the moral quality
of an interview is discussed in relation to ethical research guidelines,
and finally, a common objectlon to the scien tific quality of an inter-
view is addressed—the question of leading questions.

Interview Quality

The interview is the raw material for the later process of meaning
analysis. The quality of the original interview is decisive fof the quality
of the later analysis, verification, and reporting of the interviews.
Of the six quality criteria for an interview depicted in Box 8.1, the
last three in particular refer to an ideal interview—requiring that the
meaning of what is said is interpreted, verified, and communicated by
the time the tape recorder is turned off. This demands craftsmanship
and expertise and presupposes that the interviewer knows what he or
she is interviewing about, as well as why and how. Although such
quality criteria might seem to be unreachable ideals, they can serve as
guidelines.
Qualit y o[ the fnferrigm 145

Box 8.1

Quality Criteria for an Interview

• The extent of spontaneous, rich, specific, and relevant

answers from the interviewee.
• The shorter the interviewer’s questions and the longer
the subjects' answers, the better.
• The degree to which the interviewer follows up and clar-
ifies the meanings of the relevant aspects of the answers.
• The ideal interview is to a large extent interpreted
throughout the interview.
• The interviewer attempts to verify his or her interpreta-
tions of the subject’s answers in the course of the interview.
• The interview is “self-communicating”—it is a story
contained in itself that hardly requires much extra de-
scriptions and explanations.

The three interviews in Chapter 2 live up to the quality criteria

suggested here in different ways. Thus the interview reported by
Giorgi has brief questions and long answers; it provides rich and
spontaneous relevant answers about learning in everyday life, and the
answers are followed up and clarified. In the session reported by
Rogers the counselor follows up and clarifies the meanings of the
client’s answers, and at the end the client herself spontaneously in-
terprets the meaning of the interaction. In the third example, Socrates
critically interprets the meanings and contradictions of Agathon’s
statements about love and beauty, and concludes by constructing a
logical chain of arguments, the validity of which Agathon ends up ac-
cepting. All three interviews are in one respect self-communicating—
they convey important knowledge as they stand, and they also open
to further interpretations.
146 I n te rV i e ws

These quality criteria are not without exceptions. At a workshop,

one participant told about an unsuccessful interview she had con-
ductsd with a young author at a “writing school.” The topic was the
author’s own writing process; the contact during the interview had,
however, been poor, and the author’s statements were (ragmented and
superficial. There were no coherent stories and descriptions, the
interviewer could not find any unity or deeper meaning in the answers.
The resulting interview appeared worthless to her. Then she ventured
the information that the author had told her that he was trying to be
a postmodern author. From this perspective, the fragmentation, the
incoherence, and the surface statements that abstain from deeper
interpretations of meaning need not be due to a poor interview
technique, but stem from the very topic—writing literature—with the
interviewee playing the role of a postmodern author.

The Interview Subject

Some interview subjects appear to be better than others. Good

interviewees are cooperative and well motivated, they are eloquent
and knowledgeable. They are truthful and consistent, they give concise
and precise answers to the interviewer’s questions, they provide
coherent accounts and do not continually contradict themselves, they
stick to the interview topic and do not repeatedly wander off. Good
subjects can give long and lively descrlptions of their life situation,
they tell capturing stories well suited for reporting. The subjects of the
learning interview reported by Giorgi and of the interview on grades
were both good interview subjects according to these criteria.
As pleasant as such interview subjects may appear to the inter-
viewer, it is by no means a given that they provide the most valuable
knowledge about the research topics in question. The above idealized
subject appears rather similar to an upper-middle-cla5s intellectual
whose views are not necessarily representative of the general popula-
tion. Well-polished eloquence and coherency may in Some instances
gloss over more contradictory relations to the research themes.
The ideal interview subject does not exist—different persons are
suitable for different types of interviews, such as providing accurate
witness observations, versus giving sensitive accounts of personal
Qualit y o[ the Intervieui 147

experiences and emotional states, versus telling capturing stories. The

subjects of the three interviews in Chapter 2 were thus all good
subjects with respect to different purposes—Agathon providing logical
contradictions for Socrates to clarify; the therapeutic client living out,
and learning from, the emotional nature of the therapeutic relation-
ship; and the woman learning interior decorating giving rich, sponta-
neous descriptions or learning in everyday life.
Recognizing that some people may be harder to interview than
others3 it remains the task of the interviewer to motivate and facilitate
the subjects’ accounts and to obtain interviews rich in knowledge from
virtually every subject.

Interviewer Qualifications

The interviewer is him- or herself the research instrument. A good

interviewer is an expert in the topic of the interview as well as in
human interaction. The interviewer must continually make quick
choices about what to ask and how; which aspects of a subject’s answer
to follow up and which not; which answers to interpret—and which
not. Interviewers should be knowledgeable in the topics investigated,
master conversational skills, and be proficient in language with an ear
for their subjects’ linguistic style. The interviewer should have a sense
for good stories and be able to assist the subjects in the unfolding of
their narratives.
Learning to become an interviewer takes place through interview-
ing. Reading books may give some guidelines, but practice remains the
main road to mastering the craft of interviewing. This involves reading
interviews, listening to interview tapes, and watching more experi-
enced interviewers, but learning is primarily through one’s own
experience with interviewing. An interviewer’s self-confidence is ac-
quired through practice; conducting several pilot interviews before
the actual project interviews Wlll increase his or her ability to create
safe and stimulating interactions.
Role playing can be included in pilot interviews for the purpose of
training, with subjects playing such roles as the Tacit Oyster, the
Nonstop Talker, the Intellectualizing Academician, and the Powef
Player who tries to take control of the interview.
148 I n t e rV i e w s

Box 8.2 outlines some criterla for interviewer qualifications that

may lead to good interviews in the sense of producing rich knowledge
and doing justice to the ethical demand of creating a beneficial
situation for the subjects.
Recall once more the three interviews in Chapter 2. The interviewer
inquiring about interior decorating poses clear questions, is gentle and
open to what is said, follows up sensitively, and steers the interview

Box8.2

Qualification Criteria for the Interviewer

I. Knowledgeable. Has an extensive knowledge of the

interview theme, can conduct an informed conversation
about the topic; being familiar with its main aspects the
interviewer will know what issues are important to pursue,
without attempting to shine with his or her extensive
knowledge.
2. StYuC tffring.- Introduces a purpose for the interview,
oritlines the procedure in passing, and rounds off the
interview by, for example, briefly telling what was learned
in the course of the conversation and asking whether the
interviewee has any questions concerning the situation.
3. Clear.- Poses clear, simple, easy, and short questions;
speaks distinctly and understandably, does not use aca-
demic language or professional jargon. The exception is in
a stress interview: Then the questions can be complex and
ambiguous, with the subjects’ answers revealing their reac-
tions to stress.
4. Gentle.- Allows subjects to finish what they are saying,
lets them proceed at their own rate of thinking and speak-
ing. Is easy-going, tolerates pauses, indicates that it is ac-
ceptable to put forward unconventional and provocative
opinions and to treat emotional issues.
Qualit y o[ the Interview 149

in the direction of what she wants to know about the learning

experience. The therapeutic interviewer is gentle and safe, allows the

Box 8.2 Continued

5. Sensitive.- Listens actively to the content of what is

said, hears the many nuances of meaning in an answer, and
seeks to get the nuances of meaning described more fully.
The interviewer is empathic, listens to the emotional mes-
sage in what is said, not only hearing what is said but also
how it is said, and notices as well what is not said. The
interviewer feels when a topic is too emotional to pursue
in the interview.
6. Open: Hears which aspects of the interview topic are
important for the interviewee. Listens with an evenly hov-
ering attention, is open to new aspects that can be intro-
duced by the interviewee, and follows them up.
7. Steering: Knows what he or she wants to find out: is
familiar with the purpose of the interview, what it is
important to acquire knowledge about. The interviewer
controls the course of the interview and is not afraid of
interrupting digressions from the interviewee.
8. Critical: Does not take everything that is said at face
value, but questions critically to test the reliability and
validity of what the interviewees tell. This critical checking
can pertain to the observational evidence of the inter-
viewee’s statements as well as to their logical consistency.
9. Remembering.- Retains what a subject has said during
the interview, can recall earlier statements and ask to have
them elaborated, and can relate what has been said during
different parts of the interview to each other.
10. Interpreting: Manages throughout the interview to
clarify and extend the meanings of the interviewee’s state-
ments; provides interpretations of what is said, which may
then be disconfirmed or confirmed by the interviewee.
150 I nte rVi e ws

client to state her emotional critique of himself, is sensitive to what

the client says, and reflects it back to her with a mild degree of inter-
pretation. Socrates structures his interview by starting with Agathon’s
speech and stating the purpose of his questioning about the nature of
love; he then steers his opponent through relentless questioning,
remembers well Agathon’s earlier answers, interprets contradictions
in and among the answers, and critically questions their logic on the
basis of his own thorough conceptual knowledge of love and beauty.
The brief interview passage on talkativity and grades (Chapter 1,
Conversation as Research) can also be mentioned. By bringing up an
answer from earlier in the interview and asking in an open way, the
interviewer obtains two interesting pupil hypotheses about a con-
nection between how much a pupil talks and his grades as well as
between agreement with a teacher’s opinions and grades. The inter-
viewer does not take the statements at face value, but follows up first
by questioning the postulated connection in an open, rather naive way,
and then in a second question openly disputes the connection, with
the pupil who still holds to his own offering to provide examples of
his postulate.
In spite of such criteria as those given in Box 8.2, there are no abso-
lute standards for interviewer qualifications. In interviews in which
the topic really matters, the above technical rules and criteria may lose
relevance in face of the existential importance of the interview topic.
With extensive practice in different interview forms and with different
subjects, an experienced interviewer might go beyond technical rec-
ommendations and criteria, and—sometimes—deliberately disregard
or break the rules.
One example of breaking rules in the interest of good interviewing
took place at an interview workshop. The participants were divided
into groups in which one of the group members interviewed another.
The instruction to the interviewers was to explore the meaning of
authority for the subjects. All but one group returned to the plenary
session with lively positive reports. The negative experience came
from a group where the interviewer had been demanding, hostile, and
aloof and had continually interrupted the interviewee's answers,
breaking most of the above criteria for good interviewing, with the
result that his group returned dislntegrated and angry. The inter-
viewer’s explanation of his bad behavior was simple—instead of
Qualii y o[ the Inteniieu 151

asking about the meaning of authority, he had played the role of an

authoritarian interviewer, thereby obtaining a rich spectrum of spon-
taneous interviewee reactions to the phenomenon of authority.

Hamlet’s Interview

There are no definite criteria for evaluating the quality of an

interview. An example from literature may show how the appraisal of
an interview technique depends on the purpose and the content of the
interview.

Hamlet: Do you see yonder cloud that’s almost in shape of a camel?

Polonius: By th’ mass, and ’ris like a camel indeed.
Hamlet: Methinks it is like a weasel.
Polonius: It is back’d like a weasel.
Hamlet: Or like a whale?
Polonius: Very like a whale.
Hamlet: ... (Aside) They fool me to the top of my bent.
!Hamlet, act III, scene 2)

A first comment on the quality of this interview concerns its length.

Hamlet’s interview is brief. The seven lines are, however, dense and
rich enough to be the subject of more lengthy comments. In contrast,
current research interviews are often too long and filled with idle
chatter. If one knows what to ask for, why one is asking, and how to
ask, one can conduct short interviews rich in meaning.
The quality of Hamlet’s interview technique depends on how the
interview is interpreted. This short passage gives rise to several inter-
pretations. At first glance the interview is an example of an unreliable
technique—by using three leading questions Hamlet leads Polonius to
glve three entirely different answers. The interview thus does not yield
any reproducible, reliable knowledge about the shape of the cloud in
question.
At a second glance, the topic of the interview might change: The
figure in question is no longer the cloud, but the personalit o[
Polonius, his trustworthiness. The interview then provides reliable,
isi I nt e rV i e w s

thrice-checked knowledge about Polonius as an unreliable person—to

all three questions his answers are led by Hamlet’s questions. With the
change in the purpose and the topic of the interview the leading
questions do not produce unreliable knowledge, but become a sophis-
ticated, indirect, interview technique.
Hamlet’s interview then approximates the threefold ideal of being
interpreted, validated, and communicated by the end of the interview.
By repeating the question in different versions and each time getting
the “same” indirect answer about Polonius’s personality, the inter-
view is “self-interpreted” before Hamlet closes off with his aside inter-
pretation: “They fool me to the top of my bent.” As to the second
requirement—verification—few interview researchers today repeat
so consistently as Hamlet a question in different versions to test
the reliability of the answers. Regarding the third requirement—
communication—the short interview has been carried out so well that
it speaks for itself. I would think that, when watching the play, the
audience would generally experience a Gestalt switch from cloud
shape to personal credibility as the interview topic even before Hamlet
gives his aside conclusion.
So far, I have discussed Hamlet’s interview isolated from its con-
text, its position in the broader drama. At a third glance the interview
appears as a display of the power relations at a royal court. The prince
demonstrates his power to make a courtier say anything he wants. Or,
the courtier demonstrates his mode of managing the power relations
at the court. In an earlier scene in the play, Polonius himself gave a
lesson in what in current textbooks of method is called an indirect,
funnel-shaped, interview technique. Polonius requests a messenger
going to Paris to inquire into the behavior of his son studying music
in the city. The messenger is instructed to start with a broad approach:
“Enquire me first what Danskers are in Paris” and then gradually to
advance the subject, to end up with suggesting such vices as drinking,
quarreling, and visiting brothels, where “Your bait of falsehood take
this carp of truth,” concluding the lesson, “By indirections find
directions out” (Hamlet, act II, scene 1). When Polonius is that well
versed in indirect questioning techniques, is he actually caught by
Hamlet’s questioning technique? Or does he see through the scheme
and play up to Hamlet as a courtier?
Qualit y o[ the Intervieui lJ3

A central theme of the play, which was written at the transition

from the medieval to the modern age, is a questioning of reality; not
just a suspicion of the motives of others, but also a preoccupation with
the frail nature of reality. Hamlet’s interview may in that case be seen
as an illustration of a pervasive doubt about the appearance o[ the
WOTld, including the shape of a cloud and the personalities of fellow
players.
From an ethical perspective, the evaluation of Hamlet’s interview
again depends on the interpretation of its purpose and content. In the
first reading, the leading questions merely lead to unreliable knowl-
edge of the shape of the cloud. In the second reading, the interview
entails the deliberate deception of Polonius; there is no question of
informed consent, and the consequences may be a matter of life and
death for the protagonists. An ethics of principles is here overruled by
a utilitarian interest 1n survival.
In conclusion, the quality of the knowledge obtained by Hamlet’s
interview, as well as the ethical evaluation of the interview, depends
on the interpretation of the purpose and the topic of the interview.
Some further issues of ethics and of leading questions will now be
addressed more specifically. Neither are mere technical problems:
They raise basic questions as to the nature of the human interaction
in the interview and the reality the interview is about.

The Ethics of Interviewing

The moral qualities of an interview are here first addressed with

regard to the ethical guidelines of informed consent, confidentiality,
and consequences. Thereafter, comparisons with therapeutic inter-
views will serve to emphasize some of the ethical issues involved in
research interviews.

Informed Consent. Through briefing and debriefing, the subjects

should be informed about the purpose and the procedure of the
intervlew. When it comes to later use of the interview it may be
preferable to have a written agreement, signed by both interviewer
and subject, thereby obtaining the informed consent of the interviewee
154 I n te rV i e ws

to participate in the study and allow future use of the interviews. This
may include information about confidentiality and who will have
access to the interview; the researcher’s right to publish the whole
interview or parts of it; and the interviewee’s possible right to see the
transcription and the interpretations. In most cases such issues may
not matter much to the subjects interviewed, but if the investigation
should treat or instigate issues of conflict, particularly within institu-
tional settings, a written agreement may serve as a protection for both
the interviewees and the researcher.

Con/idenfinfify. The qualitative research interview involves differ-

ent ethical issues than those of a standardized questionnaire or a
therapeutic conversation. Confidentiality in these cases is assured by
the computed averages in survey responses and by the closed doors of
the therapist’s office. It is more problematic in a research interview.
Confidentiality issues involved in transcribing and reporting inter-
views will be addressed in later chapters (Chapter 9, Transcribing
Interviews; Chapter 14, Ethics of Reporting).

Consequences. The consequences for the interviewees concern the

situation itself as well as later effects of participating in the interviews.
In the grading study the general benefits of the investigation appeared
unproblematic, no harms were foreseen for the pupils interviewed,
and more knowledge about the effects of grades was considered to be
in their own interests. There were, however, some problematic con-
sequences for the subjects. The Danish pupils appeared rather embar-
rassed when describing their own relation to certain forms of grading
behavior, such as competition about, and wheedling for, good grades
(see also the interview on grades, Chapter 7, An Interview About
Grades). At such points their accounts often became general and
vague; for research purposes it would have been desirable to probe
more lntenslvely and critically in order to obtain reliable knowledge
about such grade-affected behaviors. This was not done out of concern
for the pupils’ well-belng; when consenting to be interviewed, they
had not been told that they might be questioned about topics that
could be penible to themselves or might involve changes in their
self-concept.
Qualit y o[ the Interview ISS

Research and TheTapeutic Interviews. Some consequences of a

research interview can be highlighted by a comparison with thera-
peutic interviews. Although the research interviewer can learn much
from therapeutic interviews, it is important to distinguish between the
two types. The main goal in therapy is change in the patient; in
research it is the acquisition of knowledge. A research interviewer’S
ability to listen attentively may, however, in some cases lead to quasi-
therapeutic relationships, for which most research interviewers have
neither the training nor the time. A research interview can come to
approximate a therapeutic interview, depending on the extent, the
topic, and the subjects of the interview. A quasi-therapeutic relation-
ship may be promoted through long and repeated interviews with the
same subject, where a close personal rapport may develop. If the
interview topics involve strongly personal and emotional issues, they
may in some cases br1ng forth deeper personal problems requiring
therapeutic assistance. Emotionally unstable subjects, more or less
consciously seeking the advice of a professional, may attempt to turn
a research interview into personal therapy.
Any possibility that an interview situation might come close to a
therapeutic relationship should be taken into account when designing
the study. This can be done by seeing to it that the interviews do not
promote a therapeutic relation. If sensitive issues and subjects are
involved in an interview study, arrangements might be made with a
therapist to serve as a “backup” for dealing with personal problems
that might be brought up by the interviews.
In some cases it may be possible to give interview subjects a fair
return for their services. One such study concerned the transition of
mental patients from living in a state hospital to living under normal
circumstances in the city of Aarhus. For their combined master’s
thesis, three psychology students conducted intensive personal inter-
views with the patients during the transition period, and in return
organized a consultation group to facilitate the patients’ transition to
normal living situations. In some cases, such as in therapeutic research,
it may thus be possible to exchange therapy for information, to offer
benefits that may alleviate troubles the interviewees may have had.
A suspicious attitude toward the subjects’ statements, as in Hamlet’s
interview, has also been common in some forms of psychological
156 I n t e r V i ews

interview research and may be based in a therapeutic frame of refer-

ence. In therapy it is ethical to be skeptical of what patients say: They
are at a loss about the meaning and purpose of their life and go to a
therapist for help in learning what they really mean and want. In
contrast to the therapeutic interview, where a suspicious attitude
toward the patients’ statements may be part of an implicit therapeu-
tic contract, using concealed techniques and interpreting meanings
implying a distrust of the subjects’ motives in a research interview
would raise ethical problems.
When a researcher makes interpretations going beyond the self-
understanding of the interviewees, a series of issues are raised: Should
subjects be confronted with interpretations of themselves, which they
may not have asked for? During the interview? In the course of the
analysis of the interviews? When reporting the interviews? And what
should be done about disagreements between the subject’s and the
researcher’s interpretations of a theme? In therapy the answer to such
questions is relatively simple: Patients have sought therapy and they
are paying the therapist to help them change, involving often painful
changes in self-understanding and that may take place in a dialogue
that continues over many years. In psychoanalysis, the working
through of the patient’s resistance to the therapist’s interpretations is
an essential part of the therapeutic process. In research, however, it is
the interviewer who has sought out the interviewees; they have not
asked for interpretations leading to fundamental changes in the way
they understand themselves and their world. To emphasize the point:
In therapy ir may be unethical if the therapeutic conversations the
patient has asked for, and often paid highly for, do not lead to new
insights or emotional changes. But in a research interview, which the
interviewee has not asked for, it may be unethical to instigate new self-
interpretations or emotional changes.
An inherent contradiction between pursuing scientific knowledge
and ethically respecting the integrity of the interviewee is illustrated
in the following example (see Fog, 1992). As a therapist conducting
research interviews, Fog addresses the dilemma of the researcher
wantlng the lnterview to be as deep and probing as possible, with the
risk of trespassing on the person, but on the other hand to be as
respectful of the interviewed person as possible and thereby risk
getting empirical material that only scratches the surface. She men-
Qualit y o[ the Interview 157

tions a woman who repeatedly and energetically tells the interviewer

how happy she is in her marriage. The woman also gives many verbal
and nonverbal signals denying the happiness and reports situations
where she is angry about the marriage. The information obtained in
the interview is thus ambiguous and puts the interviewer in a difficult
situation between scientific and ethical considerations. Should she
leave the woman’s version uncommented, or should she follow her
own hunch that the woman is denying the realities of the marriage
and probe further and point out to the woman the many inconsisten-
cies and contradictions in what she tells about her marriage? A
consequence of the latter cotild be a radical challenge of the woman’s
understanding of herself and her marriage. This would have been part
of an implicit contract in a therapeutic interview, but it is definitely
beyond the contract of a normal research interview and was not
attempted in this case.

Leading Questions

The most frequently asked question about interview studies today

probably concerns the effects of leading questions. The issue is some-
times raised in the form of a question such as: “Can the interview
results not be due to leading questions?” The very form of the question
involves a liar’s paradox—an answer of “Yes, this is a serious danger”
may be due to the suggestive formulation of the question leading to
this answer. And a “No, this is not the case” may demonstrate that
leading questions are not that powerful.
It is a well-documented finding that even a slight rewording of a
question in a questionnaire or in the interrogation of eye witnesses
may influence the answer. When the results of public opinion polls
are published, the proponents of a political party receiving low
support are usually quick to find biases in the wording of the poll’s
questions. In a psychological experiment on witness reliability, differ-
ent subjects were shown the same film of two cars colliding and were
then asked about the cars’ speed. The average speed estimate in reply
to the question “About how fast were the cars going when they
smashed into each other?” was 41 mph. Other subjects—seeing the
same film, but with smashed replaced by contacted in the question
 I n te r V i # ws

above—gave an average speed estimate of 32 mph (Loftus & Palmer,

1974). Politicians are well experienced in warding off leading ques-
tions from reporters; but if leading questions are posed to subjects
who are easily suggestible, such as small children, research findings
may be invalidated, a key issue in the current focus on child abuse.
Although the wording of a question can inadvertently shape the
content of an answer, it is often overlooked that leading questions are
also necessary parts of many questioning procedures; their use de-
pends on the topic and purpose of the investigation. Leading questions
may be deliberately posed by interrogators to obtain information they
suspect is being withheld. The burden of denial is then put on the
subject, as with the question, “When did you stop beating your wife?”
Police officers and lawyers also systematically apply leading questions
to test the consistency and reliability of a person’s statements. In the
Rorschach personality test, leading questions are employed by the
psychologist to “test the limits” for specific forms of perceiving the
ambiguous ink blots. In Piaget’s interviews with children about their
understanding of physical concepts, questions leading in wrong direc-
tions were used to test the strength of the child’s concept of, for
example, weight. In Socrates’ dialogue on love, he repeatedly em-
ployed such leading questions as “Surely you would say ... would you
not?” with the intention of exposing the contradictions of Agathon’s
understanding of love and beauty.
The qualitative research interview is particularly well suited for
employing leading questions to check repeatedly the reliability of the
interviewees’ answers, as well as to verify the interviewer’s interpre-
tations. Thus, contrary to popular opinion, leading questions do not
always reduce the reliability of interviews, but may enhance it; rather
than being used too much, deliberately leading questions are today
probably applied too little in qualitative research interviews.
It should be noted that not only may the questions preceding an
answer be leading, but the interviewer’s own verbal and bodily re-
sponses following an answer can act as positive or negative reinforcers
for the answer given and thereby influence the subject’s answers to
further questions. The technical issue of using leading questions in an
interview has been rather overemphasized, but the leading effects of
pfoject-based research questions have received less attention. Recall
the different kinds of answers obtained by a Rogerian, Freudian, and
 159

Skinnerian approach in the imaginary interview on teasing (Chapter

5, Thematizing). A project’s orienting research questions determine
what kind of answers may be obtained. The task is, again, not to avoid
leading research questions, but to recognize the primacy of the ques-
tion and attempt to make the orienting questions explicit, thereby
providing the reader with the possibility of evaluating their influence
on the research findings and of assessing the validity of the findings.
The fact that the issue of leading questions has received so much
attention in interview research may be due to a naive empiricism.
There may be a belief in a neutral observational access to an objective
social reality independent of the investigator, implying that an inter-
viewer collects verbal responses like a botanist collects plants in nature
or a miner unearths precious buried metals. In an alternative view,
which follows from a postmodern perspective on knowledge construc-
tion, the interview is a conversation in which the data arise in an
interpersonal relationship, coauthored and coproduced by inter-
viewer and interviewee. The decisive issue is then not whether to lead
or not to lead, but where the interview questions should lead, and
whether they will lead in important directions, producing new, trust-
worthy, and interesting knowledge.