it3010 lecture 8 case study research
DESCRIPTION
In this lecture you will learn about case study as a research strategy and how to conduct one.TRANSCRIPT
Figure 3.1 in: B. J. Oates, Researching Information Systems and Computing. London: Sage Publications, 2006.
The research process
Case study: Definition
• A case study focuses on one instance (case) of the ‘thing’ that is to be investigated.– In-depth,– Using multiple data generation methods,– The goal: A rich and detailed description of the thing and its
relationships and processes.
• As opposed to surveys and experiments where the goal is to eliminate complex relationships and focus on few parameters.
Phenomenon (case) of study
Context
Phenomenon
Important to identify the edge
Case study studies the phenomenon, its context, and the relationship between them.
Case study: Characteristics
• Depth rather than breadth
• Natural settings
• Holistic study
• Multiple sources and data generation methods
Types of case study
• Exploratory– Used initially in order to define hypotheses about a new research topic.– When there is little literature available about the research topic.
• Descriptive– Describes without speculating about why things happened.
• Explanatory– Tries to explain why things happened.
Types of case study
Context
Phenomenon
Context
Phenomenon
EmbeddedUnit of
Analysis 1
EmbeddedUnit of
Analysis 2
Context
Phenomenon
Context
Phenomenon
Context
Phenomenon
Context
Phenomenon
Context
Phenomenon
Context
Phenomenon
Context
Phenomenon
Context
Phenomenon
U1 U2
U1 U2
U1 U2
U1 U2
Embedded(multiple units of analysis)
Holistic(single unitof analysis)
Single-case Designs Multiple-case Designs
Types of case study
• Typical instance: So typical that it can actually be generalized.
• Extreme instance: If it works here, it will work everywhere!
• Test-bed for theory: School example of an instantiation of a theory.
• Convenience: The simplest. Because you were lazy.• Unique opportunity: An offer you could not refuse.
Generalization
• Normally not required from a case study– But not impossible.
• What type of findings can be generalized?– Concepts, e.g. "boundary object", "informate", "coordination
mechanism".– Theories, i.e. a theory that describes case A can describe also another
case B. Needs a large number of cases for proper validity.• Build a theory, e.g. a grounded theory.
• Test a theory,
• Evaluate alternative theories.
– Implications, not as strong as theories but can guide decisions in similar other cases.
– Rich insight, even if the researcher does not attempt to generalize.
Advantages and disadvantages
• Advantages:– Good for complex problems with lots of parameters (social).– Good where there is low degree of control (over e.g. parameters).– Helps us understand attitudes, cultures, meanings and other non-
objective aspects of a situation.
• Disadvantages:– It is not considered a hard-core method by the positivist camp.– It is not an easy method.– Can take time.– Can be confusing (e.g. difficult to create good research questions).– Can be difficult to get access to a good case.