Designing Social Inquiry: Scientific Inference in Qualitative Research

King, Keohane, and Verba argue that all good social science—whether qualitative or quantitative—derives from the same underlying logic of inference. The differences between the two traditions are stylistic, not fundamental . Both aim at valid descriptive and causal inference from incomplete data. The authors exploit the metaphor of the researcher-as-statistician to show that qualitative researchers can and should adopt the same standards of rigor: specifying observable implications, avoiding selection bias, and maximizing leverage . This unified framework generated decades of debate about whether case-study research is genuinely scientific.

The book's most influential recommendation is that researchers should not select cases on the dependent variable. If a study examines the causes of revolution but only looks at cases where revolutions occurred, there is no variation in the outcome, making causal inference impossible . The authors recommend selecting cases based on the explanatory variable instead. This advice sparked substantial debate. Critics like Barbara Geddes showed that selecting only high-growth cases in development studies produced spurious findings , while defenders argued the precept is sound but requires refinement for process-tracing research.

The authors characterized process-tracing as simply another means of increasing the number of observable implications of a theory—a way to generate more data points . Critics like George and Bennett argued this fundamentally misunderstands process-tracing. Within-case evidence is not equivalent to cross-case observations; a single unexpected piece of process-tracing evidence can force a revision of a case's interpretation, whereas additional cases may not alter statistical estimates . The debate revealed a deeper disagreement about whether qualitative research operates on the same inferential logic as statistical analysis.

Designing Social Inquiry triggered what became known as the 'KKV debate,' one of the most consequential methodological controversies in political science. As a Chinese retrospective notes, KKV is 'a milestone' that 'triggered a major debate between quantitativists and qualitativists and thus stimulated the development and innovations of political methodology' . It forced qualitative researchers to articulate their methods more rigorously and prompted the development of alternatives like process-tracing and set-theoretic approaches. The book also became a standard reference for graduate methods training, though its quantitative orientation made it difficult for some qualitative scholars .

Critics argue the book's foundation in classical statistics and the epistemology of Hempel and Popper is an inadequate basis for evaluating case-study research . The 'researcher-as-statistician' metaphor imposes a statistical worldview that cannot make sense of important aspects of case studies, particularly the value attached to findings from a single case . Some scholars contend that qualitative research has different strengths—interpretive understanding, attention to context, and causal mechanism identification—that do not reduce to the same inferential logic. The debate reflects a deeper divide between positivist and interpretivist approaches to social science.

Designing Social Inquiry represents a methodological turn in Keohane's career, moving from substantive IR theory (After Hegemony, Power and Interdependence) to questions of research design. Notably, the book cites none of Keohane's own qualitative studies as examples of recommended methods . Critics observed this irony: Keohane's own research was largely qualitative and employed methods closely resembling structured comparison and process-tracing, yet the book's examples are predominantly quantitative or hypothetical . The co-authorship with King (a quantitative methodologist) and Verba (a survey researcher) shaped the book's statistical orientation.