An Overview of Case Study
Research Methods
What is a case study?
Why conduct a case study?
• In-depth analysis of topic using multiple sources
of evidence
• Richly descriptive, grounded in deep and varied
sources of information
• Good for showing how something happens/works
in a real life situation / Understanding a complex
issue
• Explore; Describe; Explain
Definition of Case Study
Yin defines the case study research method as “an
empirical inquiry that investigates a contemporary
phenomenon within its real-life context; when the
boundaries between phenomenon and context are
not clearly evident; and in which multiple sources
of evidence are used” (Yin, 1984, p. 23).
Issues
• Studying a small number of cases is
•
•
•
insufficient for establishing reliability or
generality of findings;
The intense involvement of the researcher
in the study of the case could bias the
findings;
Some critics suggest case study research
is useful only as an exploratory tool or for
establishing a hypothesis;
Some would claim it is unscientific.
Benefits
• Answer why? Questions;
• Gain an in depth understanding of a
particular phenomenon;
• Develop an initial hypothesis;
• Establish a basis for future research;
• Develop new theories;
• Extend existing theories;
• Test existing theory.
How many cases?
ideal;
Eisenhardt (1989), suggests from 4 to 10 is
can
Yin (1984) and Stake (1995) suggest that one
be acceptable;
The key is to justify your choice.
Stages
•
•
•
•
•
•
•
•
Determine the research questions;
Decide if case study is appropriate;
Decide how many cases are appropriate;
Select the cases and determine data gathering and
analysis techniques;
Prepare to collect the data;
Collect data in the field;
Evaluate and analyze the data;
Write up the research and present the findings.
Choosing Your Case Study
• Unit of Analysis
• Case selected because they are:
• highly effective; not effective; representative; of
special interest
3 Types of Case Studies
1. Intrinsic Case Study
2. Instrumental Case Study
3. Multiple or Collective Case Study
Intrinsic Case Study
• Researcher interested in understanding
a specific individual or situation
• Why?
• Goal = understand the case in all its
parts
• Exploratory Research
Instrumental Case
Study
• Researcher interested in understanding
something more than just a particular
case
• Studies the case only as a means to
some larger goal
• Goal = global
• Draws conclusions that apply beyond a
particular case
Multiple or Collective
Case Study
• Researcher studies multiple cases at
the same time
• Overall study
Methods of
Analysis
(Types of Evidence)
• Documentation
• Interviews
• Direct Observation
• Participant Observation
• ...and Others (Survey, Artifacts, etc.)
• Why multiple? ...Triangulation
How to Conduct Case Studies
•
•
•
•
•
•
•
•
•
Determine / define research question
good for “how” & “why” questions / limited number of
events and their interrelationships
Select cases & determine data gathering / analysis
techniques
not a “sampling technique” -- select to maximize what can
be learned
with multiple cases, each is a single case with conclusions
contributing to the whole study
Prepare to collect data
systematic organization is key!
Collect / evaluate data
Prepare the report (and graduate!)
•
•
•
•
•
•
Analyzing the Evidence
Before writing up your case study, you need to analyze the
data -- keep focused on your research question
Pattern matching: a set of results is predicted, then
compared to actual results
Explanation building: a particular explanation (e.g. theory)
is used to analyze the data
Time-series analysis: looks at trends over time, matching
with possible explanations
Logic models: a complex chain of events and looks at
relationship between independent variables (causes) and
dependent variables (events)
Cross-case synthesis: findings are analyzed across cases
and generalizations made
Data collection and analysis 1
• Decide how to collect the data, interviews,
surveys etc;
• If survey method is chosen - paper or
internet, postal or facilitated in person;
• If interviews, decide if to be group or
individual;
• Decide how to capture data, written notes,
audio tape, video;
• Decide transcription and language issues.
Data collection and analysis 2
• Decide how to code the data;
• If you have quantitative data, decide
which statistical tools to use;
• If qualitative decide if you need
software to analyse it;
• Decide how to analyse the data and
integrate theory;
• Decide how to present the findings and
analysis.
Conducting In-Depth Interviews
Lecture adapted from Prof. Shishir Mathur
Purpose of Interviews
• Generate ideas
• Develop hypothesis
• Gain insight into complex issues
• Seek expert opinion
• Get people’s opinion; learn how they
look at the world
• Collect information
Pre-Interview
• Prepare, prepare, prepare! Know your objectives
and information needed.
• Select interviewees -- rationale
• Prepare questions
• Clearly worded; neutral; value of open-ended
questions; possible probes/follow-up questions
• But, be flexible -- take advantage of opportunities
• Initial contact
Interview Process
• Face-to-face; phone
• Non-distracting setting (for face-to-face)
• Introduction/review purpose/informed consent
• Redundant note taking
• Carefully constructed question order (more later)
• Be flexible (skip questions, ask probes)
• Closing question
Post-Interview
• Reserve time to re-write/organize notes
• Time-consuming, but an essential piece
• Follow-up with interviewee
•
•
•
Types of Interviews
Informal, conversational: no predetermined questions
asked (based on rough topic list you want to cover);
adaptable to interviewee’s nature & priorities; “go with
the flow” BUT difficult to do effectively, tough to
analyze/compare with others, can get off-topic quickly,
can be time-consuming
Standardized, open-ended interview: same openended questions asked to all; faster interviews; easier
to analyze, less flexibility/opportunity to explore new
avenues (can build in prompts to allow this)
Closed, fixed-response interview: same questions
and same set of pre-determined responses (like a
•
•
•
•
•
•
Sequence of Questions
Get respondents involved in interview as soon as
possible (limited time, ask only “important”
questions)
Start off with “easy” questions (e.g. facts rather than
controversial issues)
Intersperse fact-based questions throughout
Make sure your “key” questions aren’t at very end (in
case you run out of time)
Transition between major topics
Importance of closing question
Review (Interview) Research
• Thematizing: why andStages
what of the research
•
•
•
•
•
•
Designing: plan the study (including interview guide)
Interviewing: conduct based on a guide (best format for
298 research)
Transcribing: prepare material for analysis, timeconsuming
Analyzing: looks for themes, key points related to
research questions, compare/contrast across interviews
Verifying: assess validity of findings (multiple sources of
information)
Reporting: write up report using information/analysis
Sources of Error/Bias
•
Interviewee/Respondent
Induced Bias
•
•
Memory, exaggeration,
hidden agenda,
misunderstand, lack of
expertise, incomplete
answer, courtesy
Interview Induced Bias
•
React to response, voice
inflections, desire to help,
biased questions, question
order, appearance