David Collier – “The Comparative Method: Two decades of Change” in Dankwart Rustow and Kenneth Paul Erickson, Comparative Political Dynamics: Global research perspectives, New York: HarperCollins Publishers, 1991, pp. 7-31.
The aims of comparison:
* It sharpens our powers of description
* It can be a stimulus to concept formation.
* It provides criteria for testing hypotheses
* It contributes to the inductive discovery of new hypotheses and to theory building.
The comparative method refers to the partially distinctive methodological issues that arise in the systematic analysis of a small number of cases, or a “small N”.
Why study a small N?
* Because the phenomena under study occur relatively infrequently
* Because, even if these phenomena are more common, it is believed that they are better understood through the close analysis of relatively few observations.
* The problem of having more variables to analyse than cases to observe: the quandary of “many variables, small N” (Arend Lijphart).
Synopsis of Lijpart:
Four methods of analysis:
1. The Case Study method
2. The Comparative method
3. The Experimental method
4. The Comparative method
Case Study method
Merit: permits intensive examination of cases even with limited resources.
Inherent Problem: Contributes less to building theory than studies with more cases.
Types of Case Studies: 1. Atheoretical 2. Interpretive 3. Hypothesis-generating 4. Theory-confirming 5. Theory-infirming 6. Deviant case studies|
Defined as: Systematic analysis of small number of cases (small N analysis).Merit: “Given inevitable scarcity of time, energy, and financial resources, the intensive analysis of a few cases may be more promising than the superficial statistical analysis of many cases” (Lijphart, 1971).Inherent problem: Weak capacity to sort out rival explanations, specifically, the problem of “many variables, few cases”.Potential Solutions: 1. Increase number of cases 2. Focus on comparable cases 3. Reduce number of variablesa. Combine variablesb. Employ more parsimonious theory.|
Merit: Eliminates rival explanations through experimental control.Inherent Problem: Experimental control is impossible for many or most topics of relevance to field of comparative politics|
Merit: Assesses rival explanation through statistical control.Inherent Problem: Difficult to collect adequate information in a sufficient number of cases, due to limited time and resources. |
Lijphart’s proposals: How to solve the limitations of ‘small N’ (many variables, few cases).
1. The use of triangulation among the different approaches (among the comparative method defined as the small N approach and the statistical method) 2. The focus on ‘comparable cases’:
a. Cases that are matched on many variables that are not central to the study, controlling thus these variables b. Cases that differ in terms of the key variables that are the focus of analysis, assessing more adequately their influence. 3. The reduction of variables through their combination or through the elaboration of a theory that focuses on a small number of explanatory factors.
Innovations relevant to the comparative method
1. Justification of Small N.
Giovanni Sartori introduces the idea of ‘concept stretching’. He suggests that if researchers apply concepts to a broader range of cases, it could lead to conceptual stretching, as the meaning associated with the original concept fail to fit the reality of new cases. In this way, the concepts become highly abstract and they do not reflect the conditions specific to the different (numerous) cases under analysis.
Clifford Geertz advocates the idea of ‘thick description’, a label according to which it is very important to unravel the underlying meaning of political phenomena and to understand that this meaning is rooted in particular contexts. He justifies the need for research with relatively few cases in order to emphasise the importance of the explicative variables that vary according to the context.
Other advocates of the small N are authors, such as Skocpol and Somers who introduced the concept of ‘contrast of contexts’ (the use of comparison in order to contextualise the research findings) or Charles Ragin, who draw attention to the concept of ‘conjunctural causation’ (causal patterns that vary according to the context) (p.14).
2. Solutions to the Problem of many Variables, Small N
c. Increasing the Number of Cases
d. Comparable Cases versus contrasting cases (Lijphart versus Przeworski and Teune)
e. Reducing the number of variables in conjunction with using stronger theory.
The central goal in the field of comparative method must be to sustain the communication between the comparative quantifiers and the experts in qualitative comparison. In this sense, the country specialists and experts in qualitative small N comparison will push the comparative quantifiers toward more carefully contextualised analysis; and the comparative quantifiers will push the country specialists and experts in qualitative comparison toward more systematic measurement and hypothesis testing (p.25).