• Choose a category
• All
• Classic-Fiction

By Root 731 0

in conjunction with or to test rational choice theories.17 Social constructivists, cognitive theorists, and historical institutionalists may welcome the comparative advantages of case studies for addressing qualitative variables, individual actors, decision-making processes, historical and social contexts, and path dependencies. Meanwhile, structuralists may worry that case studies are more amenable to these social and institutional theories than to materialist theories. We maintain, however, that case studies (as well as statistical and formal methods) are useful for theory development across all these schools of thought and that they can incorporate both material and ideational variables. Postmodernists will be skeptical of our aspiration to cumulative theoretical knowledge, but even they may find our version of case study methods useful in studying discourses, identities, and interactions systematically.

Fourth, there is growing interest across the social and physical sciences in modeling and assessing complex causal relations, such as path dependence, tipping points, multiple interactions effects, selection effects, disproportionate feedback loops, equifinality (many alternative causal paths to the same outcome), and multifinality (many outcomes consistent with a particular value of one variable). Case study methods, particularly when used in the development of typological theories, are good at exploring many of these aspects of complex causality.

Fifth, we found it necessary to address an imbalance in our field, and perhaps in others, between the mix of methods that we and our colleagues use in our own research and that which we teach to our students. Although almost half the articles published in the top political science journals in recent years used case studies, only about two-thirds of the thirty top-ranked graduate programs in political science offer a dedicated graduate course in qualitative or case study methods, and only two of these departments require such a course.18 In contrast, all of the top thirty departments offer courses in statistics, and almost all of these departments require some training in statistics, often several courses. We believe that graduate students should be trained to produce cutting-edge research in their method of choice (which requires more courses for statistical methods than for qualitative methods) and to be critically aware consumers of research using the other two methods.19 In this regard, this book is designed as a text for teaching students cutting-edge qualitative methods.

Finally, the publication of Designing Social Inquiry: Scientific Inference in Qualitative Research (DSI) by Gary King, Robert O. Keohane, and Sidney Verba has greatly influenced our field and usefully forced us to clarify our thinking on case study methods.20 We find much to agree with in this important work.21 At the same time, we find it necessary to qualify DSI’s central argument that there is one “logic of inference.” If this logic of inference refers in a broad sense to the epistemological logic of deriving testable implications from alternative theories, testing these implications against quantitative or case study data, and modifying theories or our confidence in them in accordance with the results, then perhaps on a very general level there is one logic that is the modern successor of the still-evolving positivist tradition, although many disagreements remain about particular aspects of this logic.22

If, however, the logic of inference refers to specific methodological injunctions on such issues as the value of single-case studies, the procedures for choosing which cases to study, the role of process-tracing, and the relative importance of causal effects (the expected change in the dependent variable given a unit change in an independent variable) and causal mechanisms as bases for inference and explanation, as DSI appears to argue, then we disagree with the overall argument as well as some of the methodological advice DSI provides to case study researchers on these issues. DSI risks conflating