Hypothesis Testing and Model Selection in the Social Sciences

Examining the major approaches to hypothesis testing and model selection, this book blends statistical theory with recommendations for practice, illustrated with real-world social science examples. It systematically compares classical (frequentist) and Bayesian approaches, showing how they are applied, exploring ways to reconcile the differences between them, and evaluating key controversies and criticisms. The book also addresses the role of hypothesis testing in the evaluation of theories, the relationship between hypothesis tests and confidence intervals, and the role of prior knowledge in Bayesian estimation and Bayesian hypothesis testing. Two easily calculated alternatives to standard hypothesis tests are discussed in depth: the Akaike information criterion (AIC) and Bayesian information criterion (BIC). The companion website ([ital]www.guilford.com/weakliem-materials[/ital]) supplies data and syntax files for the book's examples.

David L. Weakliem, PhD, is Professor of Sociology at the University of Connecticut. He has been a fellow at the Center for Advanced Study in the Behavioral Sciences at Stanford University and at the Australian National University. Dr. Weakliem is Editor-in-Chief of Comparative Sociology and a past Deputy Editor of the American Sociological Review.

1. Hypothesis Testing and Model Selection
1.1. Introduction
1.2. Standard Procedure of Hypothesis Testing
1.3. Model Selection
1.4. Purpose and Plan of the Book
2. Hypothesis Testing: Criticisms and Alternatives
2.1. Hypothesis Testing and Its Discontents
2.2. Uses of Hypothesis Tests
2.3. Criticisms of Conventional Hypothesis Testing
2.4. Implications of the Criticisms
2.5. Alternatives to Conventional Tests
2.6. Examples
2.7. Summary and Conclusions
Recommended Reading
3. The Classical Approach
3.1. Random Sampling and Classical Tests
3.2. Two Approaches to Hypothesis Tests
3.3. Confidence Intervals
3.4. Choosing a Significance Level
3.5. Comparison to Conventional Practice
3.6. Implications of Choosing an α-level
3.7. Other Kinds of Errors
3.8. Example of Choosing an α-level
3.9. Evaluation of Criticisms
3.10. Conclusions
Recommended Reading
4. Bayesian Hypothesis Tests
4.1. Bayes's Theorem
4.2. Bayesian Estimation
4.3. Bayes Factors
4.4. Bayesian Confidence Intervals and Bayes Factors
4.5. Approaches to Bayesian Hypothesis Testing
4.6. The Unit Information Prior
4.7. Limits on Bayes Factors
4.8. Bayes Factors for Multiple Parameters
4.9. Conclusions
Recommended Reading
5. The Akaike Information Criterion
5.1. Information
5.2. Prediction and Model Selection
5.3. The AIC
5.4. Consistency and Efficiency
5.5. Cross-Validation and the AIC
5.6. A Classical Perspective on the AIC
5.7. A Bayesian Perspective on the AIC
5.8. A General Class of Model Selection Criteria
5.9. Summary and Conclusions
Recommended Reading
6. Three-Way Decisions
6.1. Substantive and Statistical Hypotheses
6.2. Bayes Factors for Directional Hypotheses
6.3. Bayes Factors for Three-Way Decisions
6.4. Summary and Conclusions
Recommended Reading
7. Model Selection
7.1. Introduction
7.2. Bayesian Model Selection
7.3. The Value of Model Selection
7.4. The Risks of Model Selection
7.5. Examples of Model Selection
7.6. Conclusions
Recommended Reading
8. Hypothesis Tests
8.1. Hypothesis Tests and the Strength of Evidence
8.2. When Should Hypotheses Be Tested?
8.3. The Role of Hypothesis Tests
8.4. Overfitting
8.5. Hypothesis Tests and the Development of Theory
8.6. Conclusions
Recommended Reading
References