Kooperativer Bibliotheksverbund

UID:

Umfang: 1 online resource (xix, 661 pages) : , digital, PDF file(s).

ISBN: 1-108-38394-7 , 1-108-37752-1

Serie: Strategies for social inquiry

Inhalt: Fairfield and Charman provide a modern, rigorous and intuitive methodology for case-study research to help social scientists and analysts make better inferences from qualitative evidence. The book develops concrete guidelines for conducting inference to best explanation given incomplete information; no previous exposure to Bayesian analysis or specialized mathematical skills are needed. Topics covered include constructing rival hypotheses that are neither too simple nor overly complex, assessing the inferential weight of evidence, counteracting cognitive biases, selecting cases, and iterating between theory development, data collection, and analysis. Extensive worked examples apply Bayesian guidelines, showcasing both exemplars of intuitive Bayesian reasoning and departures from Bayesian principles in published case studies drawn from process-tracing, comparative, and multimethod research. Beyond improving inference and analytic transparency, an overarching goal of this book is to revalue qualitative research and place it on more equal footing with respect to quantitative and experimental traditions by illustrating that Bayesianism provides a universally applicable inferential fram

Anmerkung: Title from publisher's bibliographic system (viewed on 28 Jul 2022). , Cover -- Half-title -- Series information -- Title page -- Copyright information -- Contents -- List of Figures -- List of Tables -- Acknowledgments -- A Note on the Cover -- Part I Foundations -- 1 Introduction: Bayesian Reasoning for Qualitative Research -- 1.1 Placing Our Approach in Perspective -- 1.1.1 Process Tracing -- 1.1.2 Qualitative Methods -- 1.1.3 Multi-Method Research -- 1.2 A Guide for Readers -- 1.3 Scope of the Book -- 1.3.1 Foundations -- 1.3.2 Operationalizing Bayesian Reasoning in Qualitative Research -- 1.3.3 Bayesianism in Methodological Perspective -- 1.3.4 Bayesian Implications for Research Design -- 1.4 A Spotlight on Low-Tech Best Practices -- 2 Fundamentals of Bayesian Probability -- 2.1 Introduction -- 2.2 Deductive Logic -- 2.2.1 Boolean Algebra -- 2.2.2 Deductive Structure -- 2.3 Probability Theory -- 2.3.1 Disjunction Rule -- 2.3.2 Law of Total Probability (Marginalization) -- 2.3.3 Bayes' Rule (Conditionalization) -- 2.4 Applying the Laws of Probability -- 2.4.1 A Walk in the Park -- 2.4.2 Partisan Coupling -- 2.4.3 Game Show -- 2.4.4 A Rare Disease -- Appendix 2.A Deduction, Induction, and Abduction -- Appendix 2.B Deriving the Product and Sum Rules from the Rationality Desiderata -- 2.B.1 Conjunction -- 2.B.2 Logical Negation -- 2.B.3 Emergence of the Product and Sum Rules -- Appendix 2.C Deriving the Principle of Indifference from the Rationality Desiderata -- Part II Operationalizing Bayesian Reasoning in Qualitative Research -- 3 Heuristic Bayesian Reasoning -- 3.1 Introduction -- 3.2 Bayes' Rule for Updating Degrees of Belief in Hypotheses -- 3.3 The Hypothesis Space -- 3.3.1 Hypotheses as Logical Propositions -- 3.3.2 Specifying Concrete Alternatives -- 3.3.3 Mutual Exclusivity and Exhaustiveness -- 3.4 Prior Odds -- 3.5 Evidence -- 3.6 Likelihood Ratios -- 3.6.1 Inhabit the World of Each Hypothesis. , 3.6.2 Testimonial Evidence: Assess the Likelihood That Source S Stated X -- 3.6.3 Condition on Previously Incorporated Evidence -- 3.7 Drawing an Inference from Multiple Pieces of Evidence -- 3.8 Conclusion -- 3.9 Exercises: Constructing Mutually Exclusive Hypotheses from Contributing Causes -- 3.10 Exercises: Envisioning Discriminating Evidence -- Appendix 3.A Scenarios Within Worlds: Partitioning Likelihoods -- 4 Explicit Bayesian Analysis -- 4.1 Introduction -- 4.2 Bayes' Rule in Log-Odds Form -- 4.3 Quantifying Probabilities on a Logarithmic Scale -- 4.4 Consistency Checks -- 4.4.1 Parsing the Evidence Differently -- 4.4.2 Altering the Order of Evidence -- 4.4.3 Considering Other Pairs of Hypotheses -- 4.4.4 Additional Guidance -- 4.5 Application: Tax Reform in Chile -- 4.5.1 Research Question and Hypotheses -- 4.5.2 Weighing the Evidence -- 4.5.3 Aggregating Weights of Evidence -- 4.5.4 Assessing Whether a Joint Hypothesis Is Merited -- 4.5.5 Key Conceptual Points -- 4.6 How Much Evidence to Collect and When to Stop -- 4.7 Conclusion -- 4.8 Exercise: Silver Blaze Weights of Evidence -- 5 Bayesian Analysis with Multiple Cases -- 5.1 Introduction -- 5.2 Cross-Case Inference: Democratic Mobilizationin Southeast Asia -- 5.2.1 The Philippines -- 5.2.2 Vietnam -- 5.2.3 Aggregating Weights of Evidence Across Cases -- 5.2.4 Overview of Cross-Case Analysis -- 5.3 Cross-Case Inference: State-Building in Latin America -- 5.3.1 Peru and Chile -- 5.3.2 Conceptual and Practical Lessons for Bayesian Inference -- 5.4 Generalizing Scope Conditions -- 5.4.1 State-Building in Latin American and Beyond: Prussia -- 5.4.2 Evaluating the Results and Deciding How to Move Forward -- 5.4.2.1 Patchwork Hypotheses: Mixing and Matching Causal Logics -- 5.4.2.2 Revised Hypotheses: Altering the Causal Logic for Wider Applicability. , 5.5 Comparative Thinking in Bayesian Analysis -- 5.6 Conclusion -- 6 Hypotheses and Priors Revisited -- 6.1 Introduction -- 6.2 Hypothesis Space -- 6.2.1 What MEE Means -- 6.2.2 What MEE Does Not Mean -- 6.2.3 Why We Need MEE Hypotheses -- 6.2.4 Strategies for Constructing Mutually Exclusive Hypotheses -- 6.2.5 Contributing Causes vs. Mutually Exclusive Hypotheses -- 6.3 Occam Factors: A Conceptual Introduction -- 6.4 Occam Penalties in Bayesian Model Comparison -- 6.4.1 The Six of Spades Example -- 6.4.2 The Chilean Tax Reform Case -- 6.4.2.1 The Core Constituency Model -- 6.4.2.2 The Equity Appeal Hypothesis -- 6.5 Joint Hypotheses and Priors: Democratic Mobilization Revisited -- 6.5.1 Rival Hypotheses -- 6.5.2 Weights of Evidence -- 6.6 Conclusion -- 6.7 Exercise: Silver Blaze Priors -- 7 Scrutinizing Qualitative Research -- 7.1 Introduction -- 7.2 A Framework for Scrutiny -- 7.2.1 Assessing Whether Inference Follows Bayesian Logic -- 7.2.2 Scrutinizing Weight of Evidence -- 7.2.2.1 Nuclear Non-Use -- 7.2.2.2 Neoliberalism by Surprise -- 7.2.2.3 State-Building -- 7.2.3 Sensitivity Analysis -- 7.2.3.1 Varying the Priors -- 7.2.3.2 Varying the Weights of Evidence -- 7.2.3.3 Aggregate Results -- 7.3 Scrutinizing Narrative Form Analysis: The Fashoda Crisis -- 7.3.1 Hypotheses, Evidence, and Suggested Exercises -- 7.3.2 Weights of Evidence -- 7.3.3 Evaluation -- 7.4 Scrutinizing Narrative Form Analysis: Land Reform in Egypt -- 7.4.1 The Power Consolidation Hypothesis -- 7.4.2 Power Consolidation vs. Humanitarianism -- 7.4.3 Power Consolidation vs. US Pressure -- 7.4.4 Power Consolidation vs. Developmental Goals -- 7.4.5 Evaluation -- 7.5 Writing Bayesian Case Narratives -- 7.6 Conclusion -- 7.7 Exercises -- Part III Bayesianism in Methodological Perspective -- 8 Contrasting Logical Bayesianism and Frequentism -- 8.1 Introduction. , 8.2 Conceptualizations of Probability -- 8.3 Frequentist Inference -- 8.3.1 Null Hypothesis Significance Testing -- 8.3.2 Parameter Estimation -- 8.4 Critique of Frequentism -- 8.4.1 Probability as Relative Frequency -- 8.4.2 Alternative-Free Hypothesis Testing -- 8.4.3 Hypothetical Data -- 8.4.4 Pre-Data Questions and Long-Run Answers -- 8.4.5 Arbitrariness and Subjectivity -- 8.4.6 Meta-Analysis -- 8.5 Conclusion -- 8.6 Exercise: Opening a Can of Beans - What Is Random? -- Appendix 8.A Example: P-Values vs. Weights of Evidence -- Appendix 8.B Example: Frequentist vs. Bayesian Rocket Science -- Appendix 8.C Example: Frequentist Confidence Intervals vs. Bayesian Credible Intervals -- 9 A Unified Framework for Inference -- 9.1 Introduction -- 9.2 The Intellectual Terrain of Multi-Method Research -- 9.3 Synthesizing Quantitative vs. Qualitative and Cross-Case vs. Within-Case Analysis -- 9.4 Toward a Unified Approach to Qualitative Methods -- 9.4.1 Surveying the Terrain -- 9.4.2 Bayesianism and Process Tracing -- 9.4.3 Remapping the Qualitative Methods Landscape from a Bayesian Perspective -- 9.5 Bridging the Experimental vs. Observational Divide -- 9.5.1 Randomization Is Not a Silver Bullet -- 9.5.2 Where "No Learning from Observational Research" Goes Wrong -- 9.5.3 Learning from Both Observational and Experimental Research: John Snow on Cholera -- 9.5.3.1 Snow's Observational Evidence -- 9.5.3.2 Snow's Unrealized Natural Experiment -- 9.5.3.3 Drawing a Cumulative Inference -- 9.5.4 Control and Informativeness -- 9.6 A More Fundamental Dimension: Objective vs. Subjective Likelihoods -- 9.7 Conclusion -- Appendix 9.A Distinctions between Logical Bayesianism and Earlier Bayesian Process-Tracing Literature -- Appendix 9.B Irrelevance of Randomization to Post-Data Inference. , Appendix 9.C Objective Likelihoods from Categorical Data: A Party Preference Survey -- Part IV Bayesian Implications for Research Design -- 10 Iterative Research -- 10.1 Introduction -- 10.2 Perspectives on Iterative Research -- 10.3 Bayesian Logic of Iterative Research -- 10.3.1 Old vs. New Evidence and Prior vs. Posterior Probabilities -- 10.3.2 "New" vs. "Independent" Evidence -- 10.3.3 Curtailing Confirmation Bias and Ad Hoc Theorizing -- 10.4 Iteration in Practice -- 10.4.1 Assigning Prior Log-Odds after Devising a New Hypothesis -- 10.4.2 Weights of Evidence for the New Hypothesis -- 10.4.3 Posterior Log-Odds -- 10.4.4 Conceptual and Practical Lessons -- 10.5 Common Concerns -- 10.5.1 Biased Priors -- 10.5.2 Biased Likelihoods -- 10.5.3 Intellectual Integrity -- 10.6 Conclusion -- Appendix 10.A Extending the Hypothesis Space -- 11 Test Strength -- 11.1 Introduction -- 11.2 Van Evera's Tests -- 11.2.1 Bayesian Intuition -- 11.2.2 Subsequent Interpretations -- 11.3 Sensitivity and Specificity: Advances and Limitations -- 11.3.1 From Certainty and Uniqueness to Sensitivity and Specificity -- 11.3.2 Shortcomings of Sensitivity and Specificity Coordinates -- 11.4 Improved Coordinates for Generalizing Van Evera's Tests -- 11.4.1 Likelihood Ratios or Weights of Evidence -- 11.4.2 Relative Entropies -- 11.4.3 Expected Information Gain and Asymmetry -- 11.5 Comparison of Test Coordinates and Numerical Examples -- 11.6 Social Science vs. Medical Diagnosis: Decisions, Utility, and Test Strength -- 11.7 Conclusion -- Appendix 11.A Dinosaur Extinction -- 11.A.1 The Chicxulub Crater -- 11.A.2 Tanis Fossil Site -- 12 Case Selection -- 12.1 Introduction -- 12.2 Touring the Terrain -- 12.3 Optimal Bayesian Case Selection -- 12.3.1 The Information-Theoretic Perspective -- 12.3.2 Relative Entropy -- 12.3.3 Expected Information Gain. , 12.3.4 Practical Caveats and Principled Insights.

Weitere Ausg.: ISBN 1-108-42164-4

Sprache: Englisch

URL: lizenzpflichtig

URL: https://doi.org/10.1017/9781108377522