Robert Yin – Case Study Research
Chapter 1: Getting Started
Case study would be the preferred method when:
A 'how' or 'why' question is asked (explanatorily)
About a contemporary set of events
Over which a researcher has little or no control
Case study research is used in many situations, to contribute to our knowledge of individual, group,
organizational, social, political and related phenomena. The need for case study research arises out of the desire
to understand a complex social phenomena.
Various research methods should not be arrayed hierchically: this reinforces that case study research only is
a preliminary method and cannot be used to describe or test propositions (and that is not true).
METHOD Form of research question? Requires control of Focuses on
behavioural events? contemporary
events?
Experiment How and why? Yes Yes
Survey Who, what (how many/much), where, how No Yes
many, how much?
Archival Who, what (how many/much), where, how No Yes/no
Analysis many, how much?
History How and why? No No
Case Study How and why? No Yes
Form of research question
'What' questions --> exploratory (all methods) or about prevalence (invloed, surveys or the analysis of
archival records)
'How' and 'why' questions --> explanatory, likely to favor using a case study, experiment or history.
A research question should have both substance and form.
If the research question is a how or why question, two remaining conditions help to distinguish further among a
history, case study and experiment. History and case study rely on many ot the same techniques:
History is the preferred method when there is no such control or access (dealing with the 'dead' past).
Direct observations are not possible. The historian must rely on primary documents etc.
Case study is preferred when examining contemporary events, but when the relevant behaviours
cannot be manipulated. The difference with a history is the fact that direct observation and interviewing
persons who are involved is possible.
Experiments are done when an investigator can manipulate behaviour directly, precisely and
systematically (threating groups of people in different ways)
o Quasi-experimental / observational studies: when there is no control over the behavioural
event. Begins to overlap with histories.
--> You can use multiple methods in any given study.
Definition of the Case Study as a Research Method:
1. A case study is an empirical inquiry that
o Investigates a contemporary phenomenon in depth and within its real-world context
o Especially when the boundaries between phenomenon and context may not be clearly evident
This part helps you to continue distinguishing case study research from the other methods. For example: an
experiment separates a phenomenon from its context and a history is usually studying noncontemporary events.
The second part helps covering the features of a case study:
2. A case study inquiry:
o Copes with the technically distinctive situation in which there will be many more variables of
interest than data points, and as one result
o Relies on multiple sources of evidence, with data needing to converge in a triangulating fashion,
and as another result
o Benefits from the prior development of theoretical propositions to guide data collection and
analysis.
Realist versus relativist perspective:
,Robert Yin – Case Study Research
Realist perspective: there is a single reality, that is independent of any observer (in this book)
Relativist perspective: multiple realities have multiple meanings, findings are observer dependent.
Variations in Case Studies as a Research Method:
Case study research includes both single- and multiple-case studies.
Case study research can include (and be limited to) both qualitative as quantitative evidence
Case study research has its own place in doing evaluations. Four different applications:
o Explain presumed causal links that are to complex for surveys/experiments
o Describe an intervention and the real-world context in which it occurred
o Illustrate certain topics within an evaluation (in descriptive mode)
o Enlighten those situations in which the intervention being evaluated has no clear, single set of
outcomes
Concerns about case study research:
Need for greater rigor
Confusion with teaching cases
Generalizing from case studies must be done differently because their rarely based on a single case
Unmanageable level of effort (does not have to be)
Comparative advantage. Case studies may be valued as adjuncts to experiments rather than as
alternatives to them.
Chapter 2: Designing Case Studies
Research design = A logical plan from here (the initial set of questions) to there (conclusions). It helps to avoid the
situation in which the evidence does not address the initial research questions.
Research design is a plan that guides the investigator in the process of collecting, analyzing and interpreting
observations. It is a logical model of proof that allow the researcher to draw inferences concerning causal
relations among the variables under investigation.
In case study research, 5 components are especially important:
1. Case study questions
2. Propositions (if any)
3. Unit(s) of analysis
4. Linking the data to the propositions
5. The criteria for interpreting the findings
Case study questions - most likely 'how' and 'why' questions. Step 1: narrow your interest by using literature in 1
or 2 key topics. Step 2: examine a few key studies on your topic. Identify questions in those studies --> try to find
loose ends. Step 3: examine another set of studies on the same topic.
Case study propositions - each proposition directs attention to something that should be examined within the
scope of study. In explorative studies there are no propositions, instead you need to formulate your purpose.
Unit of analysis (the 'case') - two steps:
Defining the case: classic case studies usually focus on individuals. The case can also be small groups,
decisions, programs, organizational change and specific events.
, Robert Yin – Case Study Research
Bounding the case: spatial, temporal and other concrete boundaries. For example: units of analysis is a
small group. The persons to be included in this group (the immediate topic of the case study) must be
distinguished from those who are outside of it (context for the case study).
Define a specific, real-life 'case' to be a concrete manifestation of the abstraction.
The key definitions used, should not be idiosyncratic (afwijkend) compared to previous research.
Linking data to propositions - require that you combine or assemble your case study data as a direct reflection of
your initial study propositions.
Criteria for interpreting the findings - the statistical estimates serve as the criteria for interpreting the findings.
However: much case study analysis will not rely on the use of statistics --> alternative strategy is to identify and
address rival (concurrerende, tegenovergestelde) explanations for your findings.
Role of theory development in case study research
Before collection your data, you should construct theoretical propositions (deductive). This is different than in
some qualitative methods, such as ethnograpy and the grounded theory (inductive).
Theoretical propositions = key issues from the research literature or practical matters such as differing types of
instructional leadership styles or partnering arrangement in a study of organizations. These enable the complete
research design to provide strongly guidance in determining the data to collect and the strategies for analyzing
the data.
The complete research design embodies a 'theory' of what is being studied.
Types of theory:
Individual theories
Group theories
Organizational theories
Societal theories
Having some theory or theoretical propositions will help you to generalize the lessons learned from the case
study. 2 ways of using theory to generalize:
1. Statistical generalizing = In statistical generalizing, an inference is made about a population (or universe)
on the basis of empirical data collected from a sample of that universe. Commonly used when doing
surveys or analyzing archival data.
--> less relevant in case study research: your case(s) are not 'sampling units' and also will be too small to
serve as an adequately sized sample to represent any larger population.
2. Analytical generalizing = this is needed when doing case studies. Analytic generalization is not
generalization to some defined population that has been sampled, but to a theory of the phenomenon
being studied, a theory that may have much wider applicability than the particular case studied.
Criteria for judging the quality of research designs
Construct validity
Identifying correct operational measures for the concepts being studied
For example: investigation neighbourhood change:
1. Define 'neighbourhood change' in terms of specific concepts
2. Identify operational measures that match the concepts (by using literature)
Internal validity
Seeking to establish a causal relationship, whereby certain conditions are believed to lead to other conditions, as
distinguished from spurious relationships (Is the research done right?)
Mainly a concern for explanatory or causal studies (how and why led event x to event y), because the
investigator has to conclude correctly that there is a causal relationship between x and y, without some third factor
(z) that actually has caused y.
External validity
Defining the domain to which a study's findings can be generalized
Will the same thing happen in another setting?
Reliability
Demonstrating that the operations of study can be repeated, with the same results.
If a later researcher follows the same procedures as described and conducts the same case study over again, the
later investigator should arrive at the same findings and conclusions.
Case Study Tactics for Four Design Tests:
TESTS Case Study Tactic (how to increase the criteria) Phase of research in
Chapter 1: Getting Started
Case study would be the preferred method when:
A 'how' or 'why' question is asked (explanatorily)
About a contemporary set of events
Over which a researcher has little or no control
Case study research is used in many situations, to contribute to our knowledge of individual, group,
organizational, social, political and related phenomena. The need for case study research arises out of the desire
to understand a complex social phenomena.
Various research methods should not be arrayed hierchically: this reinforces that case study research only is
a preliminary method and cannot be used to describe or test propositions (and that is not true).
METHOD Form of research question? Requires control of Focuses on
behavioural events? contemporary
events?
Experiment How and why? Yes Yes
Survey Who, what (how many/much), where, how No Yes
many, how much?
Archival Who, what (how many/much), where, how No Yes/no
Analysis many, how much?
History How and why? No No
Case Study How and why? No Yes
Form of research question
'What' questions --> exploratory (all methods) or about prevalence (invloed, surveys or the analysis of
archival records)
'How' and 'why' questions --> explanatory, likely to favor using a case study, experiment or history.
A research question should have both substance and form.
If the research question is a how or why question, two remaining conditions help to distinguish further among a
history, case study and experiment. History and case study rely on many ot the same techniques:
History is the preferred method when there is no such control or access (dealing with the 'dead' past).
Direct observations are not possible. The historian must rely on primary documents etc.
Case study is preferred when examining contemporary events, but when the relevant behaviours
cannot be manipulated. The difference with a history is the fact that direct observation and interviewing
persons who are involved is possible.
Experiments are done when an investigator can manipulate behaviour directly, precisely and
systematically (threating groups of people in different ways)
o Quasi-experimental / observational studies: when there is no control over the behavioural
event. Begins to overlap with histories.
--> You can use multiple methods in any given study.
Definition of the Case Study as a Research Method:
1. A case study is an empirical inquiry that
o Investigates a contemporary phenomenon in depth and within its real-world context
o Especially when the boundaries between phenomenon and context may not be clearly evident
This part helps you to continue distinguishing case study research from the other methods. For example: an
experiment separates a phenomenon from its context and a history is usually studying noncontemporary events.
The second part helps covering the features of a case study:
2. A case study inquiry:
o Copes with the technically distinctive situation in which there will be many more variables of
interest than data points, and as one result
o Relies on multiple sources of evidence, with data needing to converge in a triangulating fashion,
and as another result
o Benefits from the prior development of theoretical propositions to guide data collection and
analysis.
Realist versus relativist perspective:
,Robert Yin – Case Study Research
Realist perspective: there is a single reality, that is independent of any observer (in this book)
Relativist perspective: multiple realities have multiple meanings, findings are observer dependent.
Variations in Case Studies as a Research Method:
Case study research includes both single- and multiple-case studies.
Case study research can include (and be limited to) both qualitative as quantitative evidence
Case study research has its own place in doing evaluations. Four different applications:
o Explain presumed causal links that are to complex for surveys/experiments
o Describe an intervention and the real-world context in which it occurred
o Illustrate certain topics within an evaluation (in descriptive mode)
o Enlighten those situations in which the intervention being evaluated has no clear, single set of
outcomes
Concerns about case study research:
Need for greater rigor
Confusion with teaching cases
Generalizing from case studies must be done differently because their rarely based on a single case
Unmanageable level of effort (does not have to be)
Comparative advantage. Case studies may be valued as adjuncts to experiments rather than as
alternatives to them.
Chapter 2: Designing Case Studies
Research design = A logical plan from here (the initial set of questions) to there (conclusions). It helps to avoid the
situation in which the evidence does not address the initial research questions.
Research design is a plan that guides the investigator in the process of collecting, analyzing and interpreting
observations. It is a logical model of proof that allow the researcher to draw inferences concerning causal
relations among the variables under investigation.
In case study research, 5 components are especially important:
1. Case study questions
2. Propositions (if any)
3. Unit(s) of analysis
4. Linking the data to the propositions
5. The criteria for interpreting the findings
Case study questions - most likely 'how' and 'why' questions. Step 1: narrow your interest by using literature in 1
or 2 key topics. Step 2: examine a few key studies on your topic. Identify questions in those studies --> try to find
loose ends. Step 3: examine another set of studies on the same topic.
Case study propositions - each proposition directs attention to something that should be examined within the
scope of study. In explorative studies there are no propositions, instead you need to formulate your purpose.
Unit of analysis (the 'case') - two steps:
Defining the case: classic case studies usually focus on individuals. The case can also be small groups,
decisions, programs, organizational change and specific events.
, Robert Yin – Case Study Research
Bounding the case: spatial, temporal and other concrete boundaries. For example: units of analysis is a
small group. The persons to be included in this group (the immediate topic of the case study) must be
distinguished from those who are outside of it (context for the case study).
Define a specific, real-life 'case' to be a concrete manifestation of the abstraction.
The key definitions used, should not be idiosyncratic (afwijkend) compared to previous research.
Linking data to propositions - require that you combine or assemble your case study data as a direct reflection of
your initial study propositions.
Criteria for interpreting the findings - the statistical estimates serve as the criteria for interpreting the findings.
However: much case study analysis will not rely on the use of statistics --> alternative strategy is to identify and
address rival (concurrerende, tegenovergestelde) explanations for your findings.
Role of theory development in case study research
Before collection your data, you should construct theoretical propositions (deductive). This is different than in
some qualitative methods, such as ethnograpy and the grounded theory (inductive).
Theoretical propositions = key issues from the research literature or practical matters such as differing types of
instructional leadership styles or partnering arrangement in a study of organizations. These enable the complete
research design to provide strongly guidance in determining the data to collect and the strategies for analyzing
the data.
The complete research design embodies a 'theory' of what is being studied.
Types of theory:
Individual theories
Group theories
Organizational theories
Societal theories
Having some theory or theoretical propositions will help you to generalize the lessons learned from the case
study. 2 ways of using theory to generalize:
1. Statistical generalizing = In statistical generalizing, an inference is made about a population (or universe)
on the basis of empirical data collected from a sample of that universe. Commonly used when doing
surveys or analyzing archival data.
--> less relevant in case study research: your case(s) are not 'sampling units' and also will be too small to
serve as an adequately sized sample to represent any larger population.
2. Analytical generalizing = this is needed when doing case studies. Analytic generalization is not
generalization to some defined population that has been sampled, but to a theory of the phenomenon
being studied, a theory that may have much wider applicability than the particular case studied.
Criteria for judging the quality of research designs
Construct validity
Identifying correct operational measures for the concepts being studied
For example: investigation neighbourhood change:
1. Define 'neighbourhood change' in terms of specific concepts
2. Identify operational measures that match the concepts (by using literature)
Internal validity
Seeking to establish a causal relationship, whereby certain conditions are believed to lead to other conditions, as
distinguished from spurious relationships (Is the research done right?)
Mainly a concern for explanatory or causal studies (how and why led event x to event y), because the
investigator has to conclude correctly that there is a causal relationship between x and y, without some third factor
(z) that actually has caused y.
External validity
Defining the domain to which a study's findings can be generalized
Will the same thing happen in another setting?
Reliability
Demonstrating that the operations of study can be repeated, with the same results.
If a later researcher follows the same procedures as described and conducts the same case study over again, the
later investigator should arrive at the same findings and conclusions.
Case Study Tactics for Four Design Tests:
TESTS Case Study Tactic (how to increase the criteria) Phase of research in
