3 Why Interviews?
I
In the first two chtpters we have noted a range of social science research methods
and seen that there can be a case for using a mix of methods in an enquiry. What
we have not yet done is consider what interviews might be good for. This chapter
does so: y u z v e r , we argue that when it comes to choosing research methods,
you ne& to take account of the expectations of both your sponsors and your
audiences.A further consideration is that before you interview, you need to assure
yourself that you have, or that you can develop, the qualities that are hallmarks of
a good interviewer.
The Power of Interviews
Some of the claims made for interviewing as a social research method are:
'The purpose of interviewing is to find out what is in and on a person's mind
. . . ,to access the perspective of the person being interviewed . . .,to find out
from them things that we cannot directly observe' (Patton, 1990: 278).
Qualitative interviewing is a way of uncovering and exploring the meanings
that underpin people's lives, routines, behaviours, feelings etc. (Rubin and
Rubin, 1995).
These interviews focus on the informants' understandings rather than checking
the accuracy of the interviewers' account, which is the case with survey
interviews and questionnaires; 'it allows both parties to explore the meaning of
the questions and the answers involved, which is not so central, and not so
often present, in other research procedures' (Brenner et al., 1985: 3).
So, such interviews allow answers to be clarified, which is not the case with
self-completion questionnaires.
Furthermore, interviews, especially qualitative interviews, allow for under-
standing and meanings to be explored in depth. This is particularly the case
with longitudinal research, where each informant is interviewed on several
occasions over a period of months or years.
Qualitative interviews examine the context of thought, feeling and action and
can be a way of exploring relationships between different aspects of a situation.
Interviewing is a powerful way of helping people to make explicit things that
have hitherto been implicit - to articulate their tacit perceptions, feelings and
understandings.
As we shall show, survey interviews can be cheaper, per response, than self-
completion questionnaires.
Why interviews? 33
Survey interviews can allow targeting of samples that it would be hard to
identify in other ways - intercepting people shopping at a particular store is an
example of that.
Collective interviews can be used to explore the dynamics of intact social
groupings, such as a family, work-group, or sports team.
All interviews have advantages in research involving those who are not fluent
readers, or who read reluctantly.
Oral history interviews allow the reconstruction of aspects of the recent past
that are omitted from documentary historical sources.
Interviews and questionnaires
Here we wish to sharpen an understanding of the strengths of interviews as a
research method by comparing them to self-administered questionnaires, their
main rival. Table 3.1 is a conventional summary of the main differences between
these questionnaires and qualitative interviews. It must be stressed that some
interviews, notably surveys, are little more than spoken questionnaires, and
that some open-ended questionnaires have the same aim as qualitative inter-
views. Equally, as telephone interviews are becoming more widely used in social
science research, the case for mailing out questionnaires (which are often
immediately filed in the waste basket) is weakening. And the development of
e-mail questionnaires and internet conferences bring exciting possibilities to
research design, although their 'reach' is restricted to those with regular internet
access (Selwyn and Robson, 1998). Consequently, Table 3.1 should be taken
as a prompt to reflection but the complexity of reality cannot be tabulated so
neatly.
Rather than seeing interviews and questionnaires in opposition, it might be
better to see them as complementary within a multi-part study. We are not saying
that interviews have to be followed by questionnaires. However, questionnaires
are a good way of checking the strength and incidence of the story that the
interviews seem to contain. In that sense, they can be a good check on an
interpretation of interview data, as well as a way of exploring how widely views,
feelings and understandings are shared. But questionnaires, like the one in Box
3.1, can also be used as a quick-and-dirty way of getting a sense of issues to be
explored in interviews. The permutations of different types of questionnaires and
interviews are as unlimited as your imagination and the research problem.
Not an all-or-nothing decision
We return to a theme of Chapters 1and 2 in saying that deciding whether to inter-
view is not an all-or-nothing decision. It is not the case that we use interviews or
other methods. Better to consider interviews and other methods. Box 3.2 contains
an example of the use of multiple methods, showing how a clinical psychologist
whom one of us interviewed had used interviews in the process of developing a
numerical rating scale to describe children's levels of pain and breathlessness.

34 Interviewing for socid scientists Why interviews? 35
Characteristics Qualitative interviews Self-administeredauestionnairesTABLE 3.1 A comparison of qualitative intewiews with self-administered questionnaires
Distributing questionnairesCharacteristics Qualitative interviews Self-administeredquestionnaires Time costs
(continued)
Arranging interviews - less of
a problem witb intercept (or
'clipboarding') and telephone
surveys
Establishing trust and rapport -
less of a problem witb surveys
Allowing time for diversions
during the interview -less of
a problem for surveys
Travelling, getting lost and
hanging around (unanswered
calls in phone interviews,
refusals in intercept interviews)
Limit number of interviews
to guard against interviewer
fatigue
Transcription of data
(frequently estimated as 7-10
hours transcription time per
hour of tape)
Data analysis (time needed
for analysis regularly under-
estimated, especially where
qualitative data are involved)
Provide information about As for questionnaires, but far Attitudes, motivation, accounts
better at exploring these things of behaviour, opinions, events
in depth, learning about the
informants' perspectives and
about what matters to them.
Best at
Can also be useful for
Exploring the stories and
perspectives of informants
(Understanding)
Checking how far the
researcher's hypothesis or
world view is shared by the
sample (Surveys)
Open-ended questions
allow researcher to ask for
informants' stories and
perspectives
Surveys: closed questions
can be asked, as in opinion
polls about voting intentions
Important when subjects
neither read nor write easily
and willingly
Usually swift, especially where
optical mark readers are used
Richness of response Can be a dialogue between
researcher and informant,
allowing nuances to be
captured (especially where
video recording is used) and for
questions ro be clarified and
adapted, and new ones
to be improvised. Long
interviews quite common
Questions cannot be modified
once printed, nor can the
nuances of the respondents'
voices easily be detected
in their writing. Long
questionnaires are rarely
acceptable -short (and
superficial?) is the norm
Usually swift. However, open-
ended questions are difficultto
analyse, whether they are used
in interviews or questionnaires
Mainly cost of printing,
distributing and retrieving
questionnaires. Looks cheap
per questionnaire distributed
but typical low response rates
can make it an expensive
method per returned response
Money costs Hire of interviewers (perhaps),
travel costs, tapes, batteries,
transcription of tapes
Phone bills and costs of
sending respondents an
advance letter telling them
about the research project and
the phone contact to come
Sensitive to Informants. Good for finding
out about the individual,
specific and particular
The research literature and the
range of responses amongst
groups. (Also true of survey
interviews)
Anonymity Can be awkward for some
people to say some things in
face-to-face settings - danger
of only bearing from the
confident members of the
population
Sensitive questions may be
more acceptable in anonymous
questionnaires Audience
While research designs should be sensitive to the matters we have discussed here
and in the previous chapter, they will also reflect the stances of any sponsors and
of the intended audience for the research findings.
For example, much research is done to order, commissionedwith deadlines and
budgets to meet, which effectively fixes what can and cannot be done. This is as
true for undergraduates who have to finance their own dissertation study to be
submitted before the end of the year, as it is for us, doing funded research for
commissioning agencies. Research sponsors will have their own preconceptions
about the best way to do a study, and they will often be positivist preconceptions.
Furthermore, some sponsors will want a report that highlights action points and
areas of concern but, as we recount in Chapter 5, will not permit any publication
that draws on that work.
For students, the audience is likely to be a member of faculty. Quite apart from
matters to do with deadlines and resources, it is important to remember that these
Ethics Interviewers know whom they
have interviewed, although
transcripts can be anonymized
Anonymous questionnaire
responses easily ensured
Sample size With the exception of phone
and intercept interviews, less
suitable for wide coverage -
better for detailed work with
fewer people
Can be very large, and since the
aim is often to generahze from
the sample to the population,
samples often need to be big
Time costs
7
Devising interview guide
(checking validity and
reliability, see next chapter),
piloting (check on usability) -
may be less of an issue with
qualitative research
Devising questionnaire
(checking validity and
reliability - see next chapter).
piloting (check on usability)

36 Interviewing for socid scientists
Box 3.1 An open-ended,self-administeredquestionnaire
I (Used by Peter in 1995, early in a study and as one enquiry method) I
Learner autonomy is a key goal of Higher Education. Can you see
better ways of fostering it? If so, please outline them here. [one-third of
page left blank for responses]
What do you think learners need to master if they are to become -and
be recognized as - autonomous learners by the end of their course?
[one-third of page left blank for responses]
The programme that I described overleaf has somewhat limited appeal
to undergraduates. Can you suggest any developments that are
consistent with current thinking about good practice in teaching and
learning that might enhance the programme's appeal to students?
[one-third of page left blank for responses]
academic staff will often have their own preferences about research orientations
and methods: the wise student is alert to them (Arksey, 1992). Faculty and
departments too often have working (and often tacit) notions of what counts as
'proper' research, and may take deviance from those norms as evidence of sloppy
research. The 'paradigm wars' may be muted nowadays but their aftermath
lingers in some places. We have repeatedly heard North American colleagues tell
us that quantitative, positivist research is seen as 'normal' research in social
science, especially in psychology, with qualitative and descriptive research being
judged to be inherently inferior. So, the choice of methods is best made with the
audience in mind.
Nor"are,~rofessionalresearchers exempt from such expectations. Many of us
have come across referees whose comments on research proposals seem based on
assumptions about the properness of certain research designs and who seem to
have taken slight account of the problem to be investigated. Journals have their
peccadi[oese. Even at the design stage (perhaps, especially at the design
stage),the 'streetwise' researcher will have an idea of possible publication outlets
for the research findings and be aware of the range of methods that is likely to be
acceptable to them.
In choosing methods, think of the outcome in the sense of the relationship
between methods and research findings, but also think of social outcomes, in
terms of the relationship between the methods and the reactions of the intended
audience.
Why interviews? 37
Box 3.2 The use of serial researchmethods in the constructionof rating
scales for child pain and breathlessness
(This is based on an interview with the clinician who developed the rating
scales)
Researchers wanted to create a scale that would allow them to develop a
way of estimatingthe levelsofpain feltby childrenwho, because of develop-
mental problems, were not easily able to communicate. Consequently, they
were not able to say how much post-surgical pain they felt. Although
physicians could estimate a pain level on the basis of the physiological
consequences of the surgery, pain thresholds differ from person to person.
Sincethe children could not report how much pain they felt, the decision
was taken to treat parents as key informants about children's manifestations
of pain. Earlier attempts to investigate parents' observations of children's
pain responses by having them write descriptions had been inhibiting for
some parents and had produced rather short accounts. Interviews with
parents were seen as a way of getting richer and more extensive data.
Twenty parents were interviewed and asked how they knew when their
child was in pain, and what signs of pain the child showed. They were
happy to talk, and talked so much that they brought in a lot of material that
was not relevant to the interviewer's concern.
The researchers thought it important to capture the nuances of the spoken
voice, nuances that might have been lost if the transcripts alone had been
analysed. Consequently, the interviews were analysed by two coders, who
listened to the tapes while reading the transcripts. When they had reached
a high level of agreement about the analysis of the interviews, the basis
was laid for piloting a paediatric pain manifestation scale. It is not our
purpose to describe the complex and rigorous procedures necessary to
develop a reliable scale, but we do want to draw attention to the way in
which interviewing was used to explore the territory and to establish the
content validity of the scale. (That means whether the scale included the
full range of pain symptoms.) Interviews were one part of a multi-method
research approach.
A similar study looked at child breathlessness. Formally speaking, the
objectivebreathlessnessmeasures of respirologydo not match well sufferers'
feelings of breathlessness. If the sense of breathlessness is attended by
feelings of panic, or if it is unexpected, then feelings can be stronger than
when the child had a greater sense of being in control of similar
physiological symptoms. Researchers wished to develop apictorial scaleof
breathlessness, which would indicate how breathless child patients felt.
The advantage of a pictorial scale is that children, who may not commun-
icate fluently, could identify the picture from the scale that best described
their state.
co~~fmued

38 Interviewing for social scientists
Interviews were held with a group of parents of children prone to breath-
lessness, with a group of 12-16-year-olds and with a group of 8-12-year-
olds. This technique of using focus groups can have the advantage that the
social interaction can trigger off new ideas, but there is also the risk that
dominant individuals can overshadow quieter people (see Chapter 6 for
further discussion of this). It is cheaper to run three groups of eight people
than to do 24 individual interviews. Since the purposes were qualitative,
the limitations of focus group interviews were acceptable.
The discussions were not taped, since this was felt to be intrusive.
Instead, a note was kept of the adjectives that respondents used to describe
their breathlessness, or that of their children. In addition, the children were
asked to draw pictures to represent breathlessness and they then explained
what they were trying to convey. In retrospect, it was a mistake to have a
respirologist present, who intervened too much in the discussions. It would
also have been better to run focus groups containing only a single class of
breathless children or parents, since the onset, symptoms and prognosis of
asthma and cystic fibrosis are quite different.
However, the results were useful. The researchers had expected children
to describe breathlessness in terms of images of physical exercise, such as
running. In fact, childrenused imagesto do with the throat and its constriction
(especially asthma sufferers) and with the lungs and their congestion
(especially cystic fibrosis sufferers). On the basis of the children's pictures,
which were collected, and of the adjectives they used, the researchers and a
graphic artist constructed three provisional, pictorial breathlessness scales.
These scales were then piloted and refined by a mixture of further inter-
views and physiological measures. Statistical techniques were then used to
orgabize the findings on an interval scale (see Table 11.2for an explanation
of this concept).
e...# A,
Can you interview?
While you may be persuaded that interviewing is the method that is best fitted to
your research purpose, you need to check that you have the qualities necessary to
interview effectively, or to be sure that you can develop them.
Even with survey interviews, potential informants decide to participate partly
on the basis of a snap judgement about the interviewer. They wonder 'is this a
person whom I wish to spend time with - after all, the only benefit for me is likely
to be an interesting encounter? What's the cost? What do they want from me?'
The more personable the interviewer and the more intriguing their status, the
more likely it is that an interview will be granted.
For example, if we are approached in the street by an intercept interviewer, our
thoughts might be:
Why interviews? 39
Are they trying to sell me anything?
Am I busy? Is it raining or freezing, or blowing a pestilential gale?
Do I like the look of them?
How far am I prepared to say something about myself to a stranger, when I
have no idea where the data will end up?
Or, if we are trying to get an interview with someone who thinks they are
important (which means they think they are busy), getting the interview may
depend upon:
* Your status: a student doing a thesis has lower status than someone of similar
standing to the respondent. By and large senior people are more likelyto get an
interview, especially with elite informants, than arejunior researchers.
Theproject: if the project is intriguing, or if the potential respondent can see a
possible pay-off, then there is a greater likelihood of getting the interview.
Yourselj if you sound personable, trustworthy and interesting, whether your
approach has been in writing, in person, or by phone, agreement to doing an
interview is more likely.
In survey interviews, the interviewer usually works from a schedule that must
be closely followed and the self comes through only (but very importantly) in
appearance, body language and voice.
In qualitative interviews, the interviewer is more like a jazz musician in ajam
session. The key may have been set and there is an initial theme: thereafter it is
improvisation. Your ability to 'jam' is crucial to the success of these interviews,
since there will be times when it will be right to improvise in any of the following
ways:
Vary the question order to fit the flow of the interview.
Vary the phrasing of the questions to help the conversation to seem natural.
Probe, or ask follow-up and clarification questions, ad lib.
Let the interview seem to eo off track.-Give a lot of attention to building trust and rapport, which will often involve
putting something of the interviewer's self into the interview, perhaps by
raising similar or different experiences, by saying 'something that has always
puzzled me . ..Can you help me to understand that?', and so on.
- Throw the interview guide away, if necessary (see Box 3.3 for an example).
The following things about us relate to our effectiveness as interviewers.
Understanding the topic
The investigatorwho is well versed in the literature now has a set of expectations that
the [forthcominginterview] data can deny. ..It is, however, true that preconceptions
can be the enemyof qualitativeresearch ...but the benefitsof the 'preconceptions' that
spring from the literature review are, perhaps,much greater than their costs .. . a good
literaturereview createsmuch more distance than it collapses.(McCracken, 1988:31)

40 Interviewing for social scientists Why interviews? 41
Box 3.3 Throwing away an interview guide in qualitative research
(This is based on an interview with the researcher whose experience is
described here)
A researcher was investigating the attitudes of people with terminal AIDS
to experimental drug treatments. Should treatments that might offer relief
and remission be permitted, despite the dangers of serious side effects and
of hastening death?
Qualitative interviews were used. The interviews were distressing to the
researcher who 'sobbed for a year': we return to this theme in Chapter 9.
One involved a man whose face was swollen and misshapen beyond
recognition by sarcoma. He could no longer see. By his bed was a picture
of how he had been, which he asked the interviewer to look at. He wanted
to talk about his illness and what it meant, saying virtually nothing about
experimental drug treatments. Respecting him and his needs, the researcher
did not press the interview guide but instead took on the more therapeutic
ro)e of listening and talking with him. To do otherwise would be ethically
uneasy in any circumstances, and quite wrong with a dying man. He
thanked ger for the conversation. Despite having got nothing from the
interview that was usable within the research, and despite being very upset
by the experience, the researcher now looks back on it as a lesson in the
C..,.-
imp&nce of being more sensitive to the informants than to any research
agenda: she sees it as a lesson in practical ethics.
The better we know or understand an area, whether through sensitive reading or
from our own experience, the better we can connect with the interviewee.
Sometimes, it is necessary to become an 'insider' in order to make this connection
(see Box 1.2, above; Chapters 7 and 8 extend this point). Connecting is not, of
course, the same as overwhelming. It means that we have a knowledge that allows
us to interpret, understand and respond.
Understanding people (including ourselves)
If what people have to say about their world is generallyboring to you, then you will
never be a great interviewer.Unless you are fascinatedby the rich variation in human
experience, qualitativeinterviewing will become drudgery. (Patton, 1990:279)
Being interested in others is certainly important in sustaining enthusiasm for the
job of interviewing. Being self-aware helps us to do the job well. The more we
have recognized the different sides of ourselves and glimpsed the different people
we can be, the more we can seethem in others. This helps to make our questioning
and prompting more sensitive. Self-understanding is also valuable in alerting us to
ways in which who we are can skew the interview. For example, if the interviewer
knows that he or she has a tendency to try to stamp their presence on a situation,
that self-awareness can help them to hold back and try to listen to the informant,
hearing the informant's view of the world without overlaying the interview with
their own presence.
Body language
The person taking me to get the train to get to my interview said that in her research
department, she'd not be allowed to do an interview in jeans (and I mentally added 'let
alone in blue suede shoes,openneck shirt and earring').However, the dress seemsto me
to be okay for the circumstances (a university in the longvacation).(Peter's field notes,
July 1997)
The Guurdian newspaper carried a report on 12 March 1998 claiming that 7 per
cent of the impact of any piece of communication is verbal, 37 per cent comes
from the tone of voice and 56 per cent derives from body language. Are we aware
of what messages are sent by our dress, appearance and body language? We have
more to say about dress in Chapter 7.
The researcher's empathy, sensitivity,humour and sincerity are important tools for the
research. ..How the researcher asks questions depends on how he or she feels about
the topic or the interviewee.And what the researcher hears from the answermay depend
on his or her mood and prior experience.(Rubin and Rubin, 1995: 12)
Can you build trust that brings rapport? In other words, how socially skilled are
you? And being socially skilled is not the same as being voluble. For example, a
quiet and sensitive person may make a far better qualitative interviewer than an
egregious party-goer who overwhelms others. If social skill is a minimum require-
ment for qualitative interviewing, it is not sufficient for interviewing within a
research stancethat places great importance on building close relationships between
the partners in interview research (see also Box 1.3and Chapter 7). And, as Rubin
and Rubin suggest, it is helpful to use self-awareness to monitor and to try to
regulate our moods. The fieldnotes in Box 3.4 show something of the emotional,
or affective, experience of interviewing.
Conclusion
The thinking behind the choice - or rejection - of any interview approaches to a
research problem has now been set out. In suggesting that it frequently makes
sense to use interviews as one of several enquiry methods, we have previewed one
aspect of research design, namely triangulation. In the following chapter, we
consider more systematically factors that will affect the ways in which you design
your interview research project.

42 Interviewingfor sociaIscientists
I
Box 3.4 Feelings about an interview
(This is an extract from a researcher's fieldnotes)
I'm a little phased by having been away from home for two days, not sure
that this interview will really happen and dogged by low-level anxiety
about some personal difficulties.But I've done hundreds of interviews and
think I'm skilled at relating to and being easy with people. Spend some
time on breathing, relaxation and affirmations ...
"Like all colleagues in this study, he [the informant] is a prolific and
engagingtalk$r, with nice, open and engaging body language and a speech
style that leaves little room for me to talk. To begin with, I'm happy to put
aside my three page interview guide and let him shape the conversation. It
is ~lqvantinteresting and, above all, it tells me what he thinks is
+,'
important.. . . I become increasingly worried about what we're not
covenng . .. I'm enjoying listening to my colleague and, as with other
interviews in this series, I feel an emerging sense of who he is, as well as
warmingto this person. But I can feel tension -I'm doing it wrong, I'm not
getting through the questions on the guide. I start looking for points to
interrupthim. I manage to slip in a rather complicated question and, when
he simplifiesit, I settle for his interpretation at the cost of losing data that
would have been interesting for comparative purposes. Eventually, well
aware of the time, 1rather derail my informant, cover some of the closed
questions in the protocol, scan it to see what important areas we've not
covered (very few, thank God!) and turn him towards the remaining big
area.
Relief. It more-or-lessworked.Confidencehigh and self-esteemgroomed.
4 Designing an Interview-Based Study
This chapter is based on the assumption that you will be doing a project in which
interview approaches offer a good prospect of getting answers to your research
questions. Most of the points we discuss have to be confronted regardless of
whether an enquiry is interview-based or led by other methods. It also sets the
scene for Chapters 5 and 6, which also focus on design issues, although the dis-
cussion becomes increasingly concerned with design issues that are particularly
important for interview research.
We begin this chapter with some thoughts on how you might identifya topic to
investigate. Then, you have to choose the enquirymethods,which we assumewill
include interviews.Here, the guidingprinciple is that the methodsshouldbe fitfor
the purpose - they should have the power to provide data that you can use to
answer your particular research questions: different questions, different research
designs. While you are engaged in the twin processes of choosing the topic and
the methods, you should also be doing a search of the literature. We suggestthat
this is the right time for you to begin drafting your report and we make some
suggestions for drafting the literature review section of it. Lastly, we address
some of the more detailed design considerations that you will have to consider in
an interview-based study -sampling issues, the trustworthinessof the design,and
the claims that might be made about its generalizability.
Staking out the territory
What is the topic that you will investigate? This decision causes beginners
problems and is not unknown to experts either. Thekind of rationalthinkingabout
research that pervades much of the writing about social research points to the
following strategy for identifying a research topic:
Your reading on a topic might make you think that there's an aspect, an
area, or an application of an idea that hasn't been pursued. You might wonder
whether findings would apply to different groups, or to different situations.
You might suppose that the conclusions were a result of the research methods
used, and that different methods (substituting interviewsfor questionnaires,for
example)or differentuses of a method (substitutingunstructured for structured
interviews) would lead to different conclusions. Or you might reasonablywant
to do something that is commonplace in natural science, which is to replicate
(copy as closely as possible) a study in order to check out the strength of the
conclusions.