Transcript Are qualitative interviews appropriate for your research project

Qualitative interviews
Marika Lüders /[email protected]
Structure of lecture
1. What are qulitative interviews and when are qualitative interviews
appropriate?
2. The hows of interviewing
3. Analysis of qualitative interviews with and without software
4. Ethical considerations
2>
Department of Media and Communication
A research interview
”a purposeful conversatoin in which one person asks prepared questions
and another answers them in order to gain information on a particular
topic or a particular area to be researched” (Frey and Oishi 1995 How to
conduct interviews by Telephone and in person: 1)
Structured (closed style) by a standard list of questions
Unstructured (open style) - autonomy for the informant to answer freely
3>
Department of Media and Communication
Qualitative interviews vs.
surveys
Qualitative interviews
4>
Surveys
Few interviewees
Many respondens
Un/semistructured - open answers
Structured - predefined answers
Depth, detail
Overview
Complex phenomena
Causality
Qualitative analysis
Quantitative analysis
Purpose: develop theory, analytical
generalisations
Purpose: generalisations
Department of Media and Communication
When are qualitative interviews
appropriate?
”Methods are mere instruments designed to identify and analyze the
obdurate character of the empirical world, and as such their value exists
only in their suitability in enabling this task to be done.”
Herbert Blumer (1969) Symbolic interactionism: Perspectives and
Methods: 27.
Hence: when do qualitative interviews appropriate for analyzing the
obdurate character of the world?
5>
Department of Media and Communication
When are qualitative interviews
appropriate?
The purpose of qualitative interviews is to collect data that reflect
comprehensive life-world stories of the interviewees.
To obtain an in-depth, lively and nuanced understanding of the
phenomena being studied.
Statistics have their very essential purpose, but strips away contexts,
letting go of the richness and complexity of the reality.
Some things can be counted (political preference),
others cannot as easily be subject for surveys (worldview, values,
experiences).
6>
Department of Media and Communication
When are qualitative interviews
appropriate?
Qualitative interviews conducted to understand experiences and
individual practices: How do journalists experience living and reporting
from war areas such as Kongo, Afghanistan or Iraq?
For studying social and political processes, how and why things change.
For studying personal and intimate issues.
Main advantage: interviews are unique and can be tailored to fit the
experiences of the informants.
7>
Department of Media and Communication
Research project design
Topic
Research questions -> appropriate methods (book is kind of weird here)
Recruit informants
Prepare questions and how to ask them
(can be compared to operationalisations in quantitative methods: how can
you be sure your questions will actually provide you with answers to
your research questions. The validity of the research project).
Are qualitative interviews appropriate for your research project (as
explained intitially).
8>
Department of Media and Communication
Selecting a topic
How something happened, details of an event
Bring new light on a topic, uncover complicated relationshps and
evolving events
Enquire how present situations of conflict or processes of peace are result
of past decitions/incidents.
Clearly, in order to uncover complicated processes of societal
developments, qualitative interviews with the right people have the
potential to uncover new aspects and information.
9>
Department of Media and Communication
Selecting a topic
Shed light on invisible problem
Give voice to the voiceles
Rubin and Rubing: a topic is important if it addresses some unsolved
problem of considerable scope.
10 >
Department of Media and Communication
Topic - what do you want to
research?
In my case for my PhD-project (and forgive me for not choosing
something more appropriate for this course):
Mediated personal expressions among adolescent users.
Where did this topic come from?
Partly assigned - had to be concerned with personal media
Importantly from curiosity: the interplay between forms of mediated
expressions - how to understand the social and individual consequences
of the individual becoming a producer of media content and of social life
taking place in mediated spaces.
11 >
Department of Media and Communication
Selecting a topic
Access
To institutions, key persons
To material
To resources, time, money
12 >
Department of Media and Communication
Selecting a topic - summing up
Intererst
Own experience
Social consciousness, wanting to influence
Relevance and broader significance
Will access be a problem?
13 >
Department of Media and Communication
3 minute discussion: what topics would
you consider for a research project?
Discuss in groups of three and four what kind of topic you would be
curious to research (as a collaborative research project).
Make sure qualitiative interviews are the suitable method to choose.
After three minutes I will ask for suggestions.
14 >
Department of Media and Communication
From topic to research questions
Might require some initial interviews: do your potential research question
resonate with the interviewees?
What can you learn from previous research (research being cumulative)?
The research questions help to keep the project focused, but might be modfied
and rewritten throughout the research project.
In my case: article-based, five articles with different questions to be answered.
One example:
How do digital network technologies affect possibilities to shape representations
of selves, and how do users create convincing representations?
15 >
Department of Media and Communication
From research questions to
interviews
Obviously, I could not ask my adolescent informants ”how they represent
themselves in mediated spaces”.
Questions need to be lucid and understandable for your informants.
Several overlapping questions might be needed, as well as probing for
more in-depth answers.
Develop an interview-guide, specifically for each informant. The
interview guide should not restrict the flow of the interview!
16 >
Department of Media and Communication
Design < analysis and theory
You need to anticipate the end of the project: what are your aims, how do
you want to contribute?
Implies reading articles and reports from similar research projects.
Anticipating the anlysis, and learning from early interviews: your focus
may change throughout the process - research is a hermeneutic process!
Reformulating research questions.
Means interviews must be analysed while the project is underway.
The aim of qualitative interviewing is not statistical generalisations but
analytical generalisations - developing theories.
17 >
Department of Media and Communication
Design < analysis and theory
The aim of qualitative interviewing is not statistical generalisations but
analytical generalisations - developing theories.
If your theory is right, similar projects should come up with similar
findings.
In my case:
hyperpersonal communication
Interrelations between mediated and face-to-face communication
with significant consequences for sociability in network societies.
18 >
Department of Media and Communication
Flexible design
Adjusting as you go along: necessary part of qualitative research
You may discover that your initial thesis is wrong, and consequently
need to alter your approach and questions in interviews.
If you already knew all the answers beforehand, why conduct interviews
at all?
19 >
Department of Media and Communication
An iterative process
A process towards a model of the phenomenon
1.Beginning: gathering themes and ideas
2.Middle: focusing, winnowing
3.End: analysing, forming theories
”The iterative design stops when the information you are putting together
supports a small number of intergrated themes and each additional
interview adds no more ideas or issues” (Rubin and Rubin: 47)
Theoretical saturation
20 >
Department of Media and Communication
Continous design
Allows exploration of new topics while keeping the research organised
and focused
Allows the researcher to be flexible yet organised.
Understandig you need to talk to other informants.
21 >
Department of Media and Communication
Recruiting interviewees
”To enhance credibility, you choose interviewees who are knowledgable,
whose combined views present a balanced perspective, and who can help
you test your emerging theory” (Rubin and Rubin 2005: 64).
First-hand knowledge
Whose views reflect different perspectives
Who can help you develop and modify your theory
22 >
Department of Media and Communication
Recruiting interviewees
Experienced: Relevant, first-hand experience: after all your empirical
knowledge of the phenomenon of study stems from those you interview.
Talk to the people concerned.
Knowledgable: preliminary research is often indicative.
But you will rarely find one individual with comprehensive knowledge.
Piecing togeher information from different interviews to depict the
overall picture.
23 >
Department of Media and Communication
Preparing an interview guide
Unique to each informant as each informant will have different
experiences and perspectives (yes, means you need to be very well
prepared).
Themes?
Specific questions? Helpful for the unexperienced researcher
Interview guides, if followed too mechanically, can easily interrupt the
conversational flow of the interview. Should be kept as a background
check-list.
Start with information about the research project, then easy to answer
questions, before taking up more sensitive issues.
24 >
Department of Media and Communication
Topical interviewing
Exploring what, when, how and why something happened.
Examples from R&R: what is wrong with welfare programs, why immigration policy is
ineffective, how health care can be improved.
Piecing together perspectives from different people to a coherent
narrative.
Researchers must have comprehensive factual knowledge on the matter
in question. Be prepared!
Researcher more actively guide the questioning.
25 >
Department of Media and Communication
Interviewees as conversational
partners
Unlike survey interviews where repondents are more passive and less
able to elaborate.
Conversational partner: the uniqueness of each person with whom the
researcher talks.
The aim is to discover his or her distinct knowledge.
26 >
Department of Media and Communication
Qualitative interviews as
extenstions of conversations
Similar to and different from conversations:
Similar as a conversation of questions and answers (though the
interviewer must learn to keep quiet)
Different: relies on the interviewer being able to proces what the
informants says. Hearing the meaning requires specific skills: think of
appropriate questions, stay focused (both the interviewer and informant),
trust, persuading people to be interviewed, specificity of questions.
27 >
Department of Media and Communication
Qualitative interviews as
extenstions of conversations
A conversation between strangers.
Some strangers are more accustomed to talking than others (interviewing
journalists as opposed to interviewing teenagers).
Clifford Geertz (1973): Thick descriptions - the depth, detail and richness
Requires main questions, probes and follow-ups. Intense listening.
Informants do not know when they have said enough. The responsibility
of the researcher to probe for more depth and detail.
28 >
Department of Media and Communication
Structuring the interview
Starting out with a selection of preplanned main questions, which in turn
may stimulate probes and follow-up questions.
-> how much time to spend on each topic, you need to cover all areas of
interest within a restricted time span (important people are busy!)
The aim: complete a narrative from various perspectives, resolve
contradictions.
29 >
Department of Media and Communication
Structuring the interview: main
questions
Breaking the subject into smaller parts
Prepared based on studying background information or preliminary
interviews. Though not to be followed rigidly.
The answers you obtain will reflect the quality of your questions: again
emphasising the importance of being prepared and to think through and
analyse your interviews throughout the research project.
30 >
Department of Media and Communication
Structuring the interviews:
probes
Encouraging the interviewee to expand on narratives, explain in more
depth: nodding, uh-huhing.
Steering the interview back on track: keeping the interview on track.
Getting the details right and trying to understand the narrative your
interviewee offers.
Asking for examples.
31 >
Department of Media and Communication
Structuring the interview:
follow-up questions
To get richer, more in-depth answers
Asking for details and examples
Filling in narrative blanks, discovered when you study already
conducted interviews. Follow-up interviews or in the same interview.
Figuring out slant, interviewees will be biased, follow-up questions will
help you understand their perspective and point of view.
Following up contradictions,
32 >
Department of Media and Communication
Credibility through
transparency
The future reader must be able to see the process by which the data were
collected and analysed.
Keep notes and recordings, but ensure anonymity and confidentiality are
protected.
Encourages the researcher to stay close to the data in writing up a report.
Quotes are good (and interesting to read).
The research process must be visible in the analysis.
33 >
Department of Media and Communication
3 minute discussion: Analysing
qualitative interviews
Discuss in groups of three and four how you would go on to analyse the
interviews.
34 >
Department of Media and Communication
Analysing qualitative interviews
Preliminary analysis after each interview.
Fine-grained analysis when interviews have been completed:
Making sense of the narratives you have ended up with.
Interview data must be coded and structured in order to develop theory
and an overall explanation.
CODING
+
REASSEMBLING for comparing narratives
35 >
Department of Media and Communication
Coding
The process of grouping responses into categories that bring together
similar (or contradictory) ideas, concepts, or themes.
Coding unit: a word, a sentence, a paragraph
Using brackets, underline or otherwise mark coding unit.
Requires close-reading and re-reading the material.
Your analysis will only be as reliable and good as your coding.
Material with the same codes are ultimately put together and compared.
36 >
Department of Media and Communication
Final stages of analysis
Organise data that help you formulate themes, refine concepts and link
them together to create a description or explanation.
Material is interpreted in terms of literature and theory within the field.
Your research project should increase the knowledge level of the studied
phenomenon within the field.
37 >
Department of Media and Communication
Analysing with software
Computer-assisted qualitative data analysis (CAQDAS) supports the
resercher in the process of coding and retrieving data.
The actual analytical process of analysing the data can only be done by
the researcher.
Methods that accelerate the routine and mechanical tasks -> more time
and effort can be devoted to the depth of the analysis.
The coding process remains very much the same, only made more
efficient.
38 >
Department of Media and Communication
Analysing with software
TAMS Analyazer
A selection of
codes that I
applied to the
interviews.
A total of 190
codes.
Codes were
defined to secure a
coherent coding
process.
39 >
Department of Media and Communication
40 >
Department of Media and Communication
Analysing with software
Oversimplifying the process of analysing complex qualitative data?
”It is important for qualitative researchers to keep interviews in the
context in which it was gathered” (Roberts and Wilson 2002: ”ICT and
the research process”: paragraph 11)
Well, the context is never lost even if the researcher uses software, rather
it is very easily accesssible
41 >
Department of Media and Communication
Analysing using software
CAQDAS facilitates a multifaceted and creative analysis
Several codes can be applied to the same text elements
Software only supports the researcher in the process of comparing and
consequently discovering patterns, similarities and differences between
the interviewees.
The interpreting role of the researcher remains the most fundamental part
of the data analysis process.
Compare with Rubin & Rubin’s chapter on analysing data - the
similarities in analytical process are evident.
42 >
Department of Media and Communication
Analysing using software
For mac, open software:
TAMS analyzer http://www.macupdate.com/info.php/id/8358
For PC
MaxQDA: http://www.maxqda.com/
Nud*ist: http://tinyurl.com/2b4qnc
43 >
Department of Media and Communication
Ethical considerations
Informed consent: received from the subject after s/he has been carefully
informed about the research
Right to privacy: protecting the identity of the subject
Protection from harm: physical or emotional.
Always ask for permission to record interviews.
Behave ethically.
For the most famous example of how not to:
44 >
Department of Media and Communication
Ethical considerations
Laud Humphrey’s Tearoom Trade (1970)
”Humphrey studied homosexual encounters in public restrooms in parks (”tearooms”)
by acting as a lookout (”watchqueen”). This fact in itself may be seen as ethically
incorrect, but it is the following one that has raised many academic eyebrows. Unable
to interview the men in the ”tearooms”, Humphrey recorded their car license plate
numbers, which he used to trace the men to their residences. He then changed his
appearance and interviewed many of the men in their homes, without being
recognized.” (Fontana and Frey: ”Interviewing: The Art of Science”: 71)
Which is why we have NSD:
45 >
Department of Media and Communication
Reporting your project to NSD
Norwegian Social Science Data Services (NSD) is the Privacy
Ombudsman for Research for all Norwegian universities, colleges and
research institutions. Concerns students as well as researchers.
Research projects that imply the collection of personal and confidential
information must be reported to the Privacy Issues Unit at NSD. The unit
ensures that collection, safeguarding, storing and reusing personal data
comply with ethical and legal standards.
http://www.nsd.uib.no/personvern/melding/pvo_melding.cfm
Letter of consent must be signed by all informants.
46 >
Department of Media and Communication