- Introduction: Doing Philosophy of Social Science
- Micro, Macro, and Mechanisms
- Mechanisms, Causal Modeling, and the Limitations of Traditional Multiple Regression
- Process Tracing and Causal Mechanisms
- Descriptive-Causal Generalizations: “Empirical Laws” in the Social Sciences?
- Useful Complex Causality
- Partial Explanations in Social Science
- Mechanistic Social Probability: How Individual Choices and Varying Circumstances Produce Stable Social Patterns
- The Impact of Duhemian Principles on Social Science Testing and Progress
- Philosophy and the Practice of Bayesian Statistics in the Social Sciences
- Sciences of Historical Tokens and Theoretical Types: History and the Social Sciences
- RCTs, Evidence, and Predicting Policy Effectiveness
- Bringing Context and Variability Back into Causal Analysis
- The Potential Value of Computational Models in Social Science Research
- Models of Culture
- The Evolutionary Program in Social Philosophy
- Cultural Evolution: Integration and Skepticism
- Coordination and the Foundations of Social Intelligence
- Making Race Out of Nothing: Psychologically Constrained Social Roles
- A Feminist Empirical and Integrative Approach in Political Science: Breaking Down the Glass Wall?
- Social Constructions of Mental Illness
- Cooperation and Reciprocity: Empirical Evidence and Normative Implications
- Evaluating Social Policy
- Values and the Science of Well-Being: A Recipe for Mixing
Abstract and Keywords
This article reports the perspective on the philosophy of Bayesian statistics, based on the idiosyncratic readings of the philosophical literature and, more importantly, the experiences doing applied statistics in the social sciences and elsewhere. It is noted that Bayes need not be linked with subjectivity and inductive reasoning. Bayesian statistics is connected with a formal inductive approach. A problem with the inductive philosophy of Bayesian statistics is that it assumes that the true model is one of the possibilities being considered. The Bayesian data analysis fits well into the falsificationist approach. Because it is felt that the status quo perception of Bayesian philosophy is wrong, it is thought that it is more helpful to present the author's perspective forcefully, with the understanding that this is only part of the larger philosophical picture.
Andrew Gelman is a professor of statistics and political science and director of the Applied Statistics Center at Columbia University.
Cosma Rohilla Shalizi is an assistant professor of statistics at Carnegie Mellon University and an external professor at the Santa Fe Institute. He received his PhD in theoretical physics from the University of Wisconsin-Madison in 2001. His research focuses on time series prediction, network analysis, and inference in complex systems.
Access to the complete content on Oxford Handbooks Online requires a subscription or purchase. Public users are able to search the site and view the abstracts and keywords for each book and chapter without a subscription.
If you have purchased a print title that contains an access token, please see the token for information about how to register your code.