The procedures of research have drastically changed in these “modern” times. The introduction of statistics and mathematics courses for students in the social sciences and the development of statistical software have caused some of the beauty of the subject to be forgotten. The intellectual underpinnings of statistics are mostly ignored. I cannot speak for other fields, but in economics the general attitude is I have a software that will calculate things and as long as the results come close to what is generally accepted as standard I should be ok. Too often we simply learn (and teach) the plug-and-chug method. Rarely does one see a great deal of analysis of variables before doing regression, for example. This bothers the author of this book.
Statistical Models and Causal Inference is an attempt to make researchers in the social sciences conscious about what they are doing. Freedman concentrates on a key element in statistical research: using diverse statistical methods prior to plug-and-chug game. The key to every good study should be first to understand the data very well. This would, one hopes, make the later results of regression more meaningful. Often this process is neglected and practitioners use regression techniques as a first encounter with the data in the study. This is very wrong.
The book is written in a clear Freedman style, including the usual humor, which makes this a fun and interesting read. The author concentrates on the use of regression methods and analysis of data. As the title suggests, the book is full with examples from real studies. In fact, the book is more a collection of case studies than a monograph. The reader should be familiar with basic statistics and regression methods, since the intention of the book is not to teach the methods, but rather to guide the reader on how they should be used, and to show where we usually make mistakes. There are numerous references in the book providing the means for further reading and exploration.
Statistical Models and Causal Inference is a tremendous book that should be read by every practitioner who ever does statistical data analysis and modeling. For students this would be an eye opener, especially after taking theoretical statistics classes and even after having some experience in data analysis and modeling.
Ita Cirovic Donev holds a Masters degree in statistics from Rice University. Her main research areas are in mathematical finance; more precisely, statistical methods for credit and market risk. Apart from the academic work she does statistical consulting work for financial institutions in the area of risk management.
Editors' introduction: inference and shoe leather; Part I. Statistical Modeling: Foundations and Limitations: 1. Some issues in the foundations of statistics: probability and model validation; 2. Statistical assumptions as empirical commitments; 3. Statistical models and shoe leather; Part II. Studies in Political Science, Public Policy, and Epidemiology: 4. Methods for Census 2000 and statistical adjustments; 5. On 'solutions' to the ecological inference problem; 6. Rejoinder to King; 7. Black ravens, white shoes, and case selection: inference with categorical variables; 8. What is the chance of an earthquake?; 9. Salt and blood pressure: conventional wisdom reconsidered; 10. The Swine Flu vaccine and Guillain-Barré Syndrome: relative risk and specific causation; 11. Survival analysis: an epidemiological hazard?; Part III. New Developments: Progress or Regress?: 12. On regression adjustments in experiments with several treatments; 13. Randomization does not justify logistic regression; 14. The grand leap; 15. On specifying graphical models for causation, and the identification problem; 16. Weighting regressions by propensity scores; 17. On the so-called 'Huber sandwich estimator' and 'robust standard errors'; 18. Endogeneity in probit response models; 19. Diagnostics cannot have much power against general alternatives; Part IV. Shoe Leather, Revisited: 20. On types of scientific inquiry: the role of quantitative reasoning.