Propensity Score Analysis provides readers with a systematic review of the origins, history, and statistical foundations of PSA and illustrates how it can be used for solving evaluation problems. With a strong focus on practical applications, the authors explore various types of data and evaluation problems related to, strategies for employing, and the limitations of PSA. Unlike the existing textbooks on program evaluation,
Propensity Score Analysis delves into statistical concepts, formulas, and models underlying the application.
Key Features- Presents key information on model derivations
- Summarizes complex statistical arguments but omits their proofs
- Links each method found in this book to specific Stata programs and provides empirical examples
- Guides readers using two conceptual frameworks: the Neyman-Rubin counterfactual framework and the Heckman econometric model of causality
- Contains examples representing real challenges commonly found in social behavioral research
- Utilizes data simulation and Monte Carlo studies to illustrate key points
- Presents descriptions of new statistical approaches necessary for understanding the four evaluation methods incorporated throughout the text
Intended Audience
This text is appropriate for graduate and doctoral students taking Evaluation, Quantitative Methods, Survey Research, and Research Design courses across business, social work, public policy, psychology, sociology, and health/medicine disciplines.
"The approach the authors take in writing this book is very effective for novices and experiences users...This balance between the practical and applied approach is a useful model for researchers to understand the process and interpretation of these analyses...[it] goes a long way in making propensity score analysis techniques more accessible, understandable, and useful to psychologists." (Karl N. Kelley PsycCRITIQUES 2011-07-06)
"Guo and Fraser’s book Propensity Score Analysis: Statistical Methods and Applications is the first comprehensive book that discusses and compares different PS techniques from theoretical and practical points of view. One of the book’s strengths is its focus on the application of PS to real data.
[T]his textbook gives a good introduction to PS matching techniques and some alternative approaches for estimating causal treatment effects. With its many examples in Stata, it may be useful for graduate students and applied researchers who have no or limited experience with PS methods but are familiar with basic regression methods and mathematical/statistical notation." (Peter M. Steiner PSYCHOMETRIKA—VOL. 75, NO. 4, 775–777 2010-12-08)