Comparative case studies
Jan Osička
Lecture outline
• Case selection process
• Comparative case study
• Exercises
Case selection
• Starts with definition of population
• Am I interested in particular case? Why? What is it that I want to study?
What is this case a case of?
• Theory building or theory testing?
• Restricting the population is crucial for inference we are about to make.
• Case boundaries: apples or fruit?
• Temporal boundaries: what makes different periods of time different?
Exercise I
What was the second war in Iraq a case of?
Case selection
Theory building
Phenomena
CasePopulation
Theory testing
Theory
CasePopulation
Theory building (Gerring 2007)
Diverse
Cases (two or more) illuminate the full range of
variation on X1, Y, or X1/Y.
Extreme
Cases (one or more) exemplify extreme or unusual
values on X1 or Y.
Deviant
Cases (one or more) deviate from some cross-case
relationship.
Most-
similar
Cases (two or more) are similar on specified
variables other than X1 and/or Y.
Most-
different
Cases (two or more) are different on specified
variables other than X1 and Y.
Exercise 2
Suggest at least one case for each of outlined theory building case
selection techniques.
Theory testing (Gerring 2007)
Influential
Cases (one or more) with influential configurations of the
independent variables.
Crucial
Cases (one or more) are most- or least likely to exhibit a given
outcome.
Pathway
Cases (one or more) where X1, and not X2, is likely to have
caused a positive outcome.
Typical
Cases (one or more) are typical examples of some cross-case
relationship.
Diverse
Cases (two or more) illuminate the full range of variation on X1,
Y, or X1/Y.
Most-similar
Cases (two or more) are similar on specified variables other
than X1 and/or Y.
Most-
different
Cases (two or more) are different on specified variables other
than X1 and Y.
Exercise 3
Suggest at least one case for each of outlined theory testing case
selection techniques.
Comparative CS
Aims
• Simultaneous work with theory
• Generating, testing, illustrating on two or more cases
• Macro-context analysis
• New fundamental questions arising from comparisons
• Within-population analysis
• Logic of understanding
Many variables, small N
• Social reality is complex
• Equifinality is inherent
• How to differentiate among the effects of particular variables?
→ Reducing variables (aggregation, hierarchy)
→ Adding cases
Comparability
• Cases must be comparable for a comparative CS to be valid.
• Comparability: majority of independent variables (X) show similar
values.
• Achieving comparability
• Geographical - simply more units
• Diachronical - same unit in more time periods
Exercise 1
Suggest comparable pair of cases.
Mill‘s methods
Elimination methods of Agreement x Difference
Agr. X1 X2 X3 X3 X4 Y
Case 1 + - + - + +
Case 2 + + - + - +
Dif. X1 X2 X3 X3 X4 Y
Case 1 + + + + + +
Case 2 - + + + + -
Exercise 2
Suggest examples of two cases based on (a) method of agreement, (b)
method of difference