The case study approach;

how to design a good interview

guide and make arrangements
for interviews
Arild Holt-Jensen
Professor, University of Bergen
Norway
 What is a case study?
Misunderstandings related to case studies
(based on Flyvbjerg 2011)
Quantitative and Qualitative methods
Types of qualitative research
Forms of interviews
Semistructured interviews interview
guide
Formulate appropriate questions
Interviewing practice; preparation and
afterwork structure of report
 What is a case study?
An intensive analysis of an individual unit
stressing environmental factors in relation to
environment (Webster dictionary 2009)
Today we see case studies as a main approach
to social scientific knowledge, but many
scientists think case studies cannot provide
reliable information about general laws and
structures.
Case studies are, however, the main focus in
what Sayer (1984) calls intensive concrete
research and recommends as a major
approach in human geography
Types of research (Sayer 1984)
Place (case study area) links physical, social
relations and meaning (Sack 1997)
 Misunderstandings about case studies
1. General theoretical knowledge is more valuable
than concrete case knowledge
2. One cannot generalize on the basis of case
studies; so useless for scientific development
3. Case studies useful for generating hypotheses in
first stages of a total research process, but other
methods needed for theory building
4. The case study contains a bias towards
verification, a tendency to confirm researchers
preconceived ideas
5. Difficult to summarize and develop general
theories on the basis of specific case studies
Flyvbjerg(2011):the cited notions are wrong
Case study work needed for students to
develop from booklearning beginners to
virtuose experts! You learn more from
experienced reality than from books or
lectures!
Deviant cases are main sources of theory
development and prove that place
matters in a world of general globalisation.
Formal generalisation is overvalued as a
source of scientific development, the force
of the example and transferability are
underestimated
 Flyvbjerg (2011) continued
The case study is useful for both
generating and testing of hypotheses, but
is not limited to that.
You may learn more from deviant cases
than the typical or average case: the
region with (deviant) successful
development can ideas be transferred?
Researchers often learn from case studies
that their preconceived views were wrong
and lead to new ideas(AHJ:Rauland 1968)
 Conclusion on case studies
The case study approach often lead to
falsification of precoceived notions rather than
theory verification, But there is danger of
subjectivism (f.ex in selection of informants) .
There are more discoveries stemming from
intense observations than by statistics from
a large group.
Students can safely be let loose in case
study reality, which provides a useful
training ground with insights into real life
practices which academic teaching often
does not provide.
 We need both quantitative and
qualitative methods
There is
complementarity
between Case studies
and Statistical methods
like there is between
extensive
generalisation and
intensive concrete
research in Sayers
model.
 Quantitative and qualitative methods
Questionnaire (many units extensive)
Structured interview
Semistructured interview (interview guide)
Focus group (round table)
Field talks
Observations (on excursion)
Participant observation (few units-intensive)
Few units does not offer general conclusions,
but deeper understanding of social phenomena
 Main types of qualitative research
Oral : most used. To get knowledge by
speaking to people (interviews)
Textual : collect documentaries, fiction
and also landscapes used as text (photos,
paintings)
Observational: excursion, get aquainted,
can be participant observation if you have
time to live in the area.
ORAL METHODS

Biography Autobiography Individual

Biography
Oral history
Interviews
Unstructured
Semi-structured
Focus groups *
Structured
Surveys Surveys structured
Questionnaires-structured General/
Structural
 Textual sources and observational
methods
Textual sources and methods
Documentary sources: newspapers, archives, maps
Creative sources: novels, poems, films, arts and music
Landscape: as text: social differences, building types

Observational methods
Tourist lookaround: get a feeling of the place
Excursion with guide or with textual background
Participant observation: passive or active? Problem/concern:
influence/affect the study/
object-subject relations- to understand semiotics (metaphors,signs)
 Which methods should we use?
Oral, textual, observational
We must use the tools most appropriate
for the research questions of the case
Often we have to use a combination of
methods in case investigations, but with
main focus on the oral tools
Interviews gives direct access to
informants. In contrast to questionnaires
they do not intend to be representative;
but to understand individuals.
My MA in Rauland mountain farm district
Very simple structured
interviews (questionaire)in
all the 350 households.
Hypotheses: transition to
market economy with
reduction of old farm
practices lead to population
and industrial decline.
That summer: local festival
theater on farm. Local
tourism!! NEW INDUSTRY
Hypotheses falsified!
 Interviews: direct access to informants
In Rauland I just had one page of factual questions as
in population/agricultural census.
But used 2 summers on the survey,lived in families
and used bicycle. Learned much from observations,
social life, informal talks.
The questionaire approach gave statistical data like a
census, but due to my presence and observations it
gave indepth knowledge that was absent in the recent
census!
Combination of structured questionnaire and
unstructured talks was an asset, but in most case
studies time limit means semi-structured interviews
is the best compromise
 Semi-structured interviews
Aim: Not to be representative but to
understand how individual people
(informants) experience the problem you
focus on.
Interviewing: Face to face conversation
with a purpose
Critics from positivists: Interviewers bias
the respondents answers by leading
questions. This is a problem we have to
deal with! But: Objectivism; does it exsist?
 More about forms of interviews
Structured interviews: Predetermined
/standardised list of questions (How many
times have you crossed the border to
Russia the last 3 years?)
Semi-structured interviews: Degree of
predetermined questions, but more like a
talk based on an interview guide to secure
you cover questions related to your problem
formulation.
Unstructured interviews: Talks where
questions are defined by the informants
response (Often used: life stories or tell us
about what you did last week)
 Strength of interviewing
Filling the gaps of knowledge
Investigate complex behaviour /motivations
Collect diversity of meaning
Opinions, experiences
Variations by position in society, ethnicity, class,
gender, age / How meaning differ among people
Example: NEHOM cases: 20 in depth interviews in
each urban housing area=ca 5 as
informants/decision makers, 15 with
respondents/local inhabitants, which the local
initiatives we had as focus, were aimed at
 Interviewing techniques
Interview guide:Used in semi-structured interviews to
remind the researcher on issues to cover topics and
key concepts to cover but flexibility in progress so
that it works as a conversation between equals/ create
trust to avoid biased answers.
Formulate appropriate and good questions:
Language: use neutral, commonly understood concepts
problems in daily speech with scientific defined concepts
Avoid ambiguity/double meanings / be neutral and non-
offensive
Avoid leading questions!!!!! (Do you agree with the
community elders that inhabitants living in the border region
to Russia should be allowed crossing the border any time?)
Types of questions/ progress in interviews
Descriptive: Roles ice breaking
Contextual (storytelling): Identify events
/people /places
Opinions: Personal /individual meanings
Structural: Why this meanings into peoples
ideology
Contrast /comparison: Reflect on contrary
meanings
Controversial questions: confronting more
sensitive issues in end of interview
 Interview guide things to consider
Research questions are usually too broad to
serve as productive interview questions So
focus on the local context in which the locals
are the specialists.
If you ask a question they will answer it. But
this means that the evidence you are gathering
may not accurately reflect real opinions. The
respondent may not have understood the
question.
Therefore: Prepare follow-up questions
Adress topics from the side (compare with
similar cases)
 Interview guide- things to consider
Peoples espoused theories (the things they
believe they believe) differ from the theories
they use. So to avoid this trap: ask about
concrete examples (what they have done rather
than about general principles)
Interviews are social occasions. The
interviewed will want to put his best foot
forward. Dont take the first answer as final.
Therefore: Ask for elaboration
Ask for opposing ideas
Ask about Other Influencing Conditions
 Interview guide things to consider
Testimony statement (this is the truth)
alone is weak evidence: In general we need
multiple sources of evidence with
triangulation between different sources and
methods.
If your study is based entirely on interviews
you need to interview so many in each group
of respondents that you feel you do not need
more interviews to get new opinions or data.
It may be an idea to ask the same question
with different phrases or from different angles.
 Interviewing practice preparations
Decide how many you need to
interview, which groups of key
informants and respondents you need to
meet (considering the time at disposal).
Agree on formulations in the interview
guide
Take contact by phone /internet as early
as possible to arrange time of interview
meetings with key informants and
respondents
Decide division of work within group
 After the interviews
Write down your notes immediately: Write
all you remember; behaviour, non-textual
responses, what possibly can have biased the
answers you got
Transcribe from recording device as soon as
possible.
Interpretation of the data collected
(Hermeneutic approach)- reduce your data
Write the report divide the work between you
and decide how to present it.
 Structure of report (possible)
Problem formulation (aims of study)
Case study areay (based on statistics,
written and internet sources)
Theory that is relevant
Methods used and weaknesses in
fieldwork situation/ interviews done
Results of fieldwork (presentation of your
work
Conclusions (playback to problem
formulation - aims of study)
TWO LAST BULLPOINTS ca 50% of report
Fieldwork means
knowledge
victory!
GOOD LUCK !
 Some references
Flyvbjerg, B.(2011) Case study p. 301-316 in Denzin,N.K. &
Lincoln,Y.S(eds) The Sage Handbook of Qualitative
Research.4th ed. Sage; Thousand Oaks,CAL
Holt-Jensen, A. (1968) Fjellbygda Rauland. Ad Novas
Norwegian Geographical Studies. Universitetsforlaget, Oslo
Dunn, K.(2005) Interviewing.Ch.6 p. 79-105 in
Hay,I.(ed):Qualitative Research Methods in Human
Geography.2nd Ed. Oxford University Press; UK
Sack, R.D. (1997) Homo Geographicus. John Hopkins;
Baltimore
Yin, R.K.(2009) Case Study Research. 4th ed. Sage; London
Thousand Oaks, CAL. Particularly Ch.4: Collecting Case
Study Evidence