Practical Propensity Score Methods Using R / Edition 1

by Walter L. Leite

Overview

Practical Propensity Score Methods Using R by Walter Leite is a practical book that uses a step-by-step analysis of realistic examples to help students understand the theory and code for implementing propensity score analysis with the R statistical language. With a comparison of both well-established and cutting-edge propensity score methods, the text highlights where solid guidelines exist to support best practices and where there is scarcity of research. Readers will find that this scaffolded approach to R and the book’s free online resources help them apply the text’s concepts to the analysis of their own data.

Product Details

ISBN-13:	9781452288888
Publisher:	SAGE Publications
Publication date:	12/01/2016
Edition description:	First Edition
Pages:	224
Product dimensions:	7.38(w) x 9.12(h) x (d)

Table of Contents

Preface
Acknowledgments
About the Author
Chapter 1. Overview of Propensity Score Analysis
Learning Objectives
1.1 Introduction
1.2 Rubin’s Causal Model
1.3 Campbell’s Framework
1.4 Propensity Scores
1.5 Description of Example
1.6 Steps of Propensity Score Analysis
1.7 Propensity Score Analysis With Complex Survey Data
1.8 Resources for Learning R
1.9 Conclusion
Study Questions
Chapter 2. Propensity Score Estimation
Learning Objectives
2.1 Introduction
2.2 Description of Example
2.3 Selection of Covariates
2.4 Dealing With Missing Data
2.5 Methods for Propensity Score Estimation
2.6 Evaluation of Common Support
2.7 Conclusion
Study Questions
Chapter 3. Propensity Score Weighting
Learning Objectives
3.1 Introduction
3.2 Description of Example
3.3 Calculation of Weights
3.4 Covariate Balance Check
3.5 Estimation of Treatment Effects With Propensity Score Weighting
3.6 Propensity Score Weighting With Multiple Imputed Data Sets
3.7 Doubly Robust Estimation of Treatment Effect With Propensity Score Weighting
3.8 Sensitivity Analysis
3.9 Conclusion
Study Questions
Chapter 4. Propensity Score Stratification
Learning Objectives
4.1 Introduction
4.2 Description of Example
4.3 Propensity Score Estimation
4.4 Propensity Score Stratification
4.5 Marginal Mean Weighting Through Stratification
4.6 Conclusion
Study Questions
Chapter 5. Propensity Score Matching
Learning Objectives
5.1 Introduction
5.2 Description of Example
5.3 Propensity Score Estimation
5.4 Propensity Score Matching Algorithms
5.5 Evaluation of Covariate Balance
5.6 Estimation of Treatment Effects
5.7 Sensitivity Analysis
5.8 Conclusion
Study Questions
Chapter 6. Propensity Score Methods for Multiple Treatments
Learning Objectives
6.1 Introduction
6.2 Description of Example
6.3 Estimation of Generalized Propensity Scores With Multinomial Logistic Regression
6.4 Estimation of Generalized Propensity Scores With Data Mining Methods
6.5 Propensity Score Weighting for Multiple Treatments
6.6 Estimation of Treatment Effect of Multiple Treatments
6.7 Conclusion
Study Questions
Chapter 7. Propensity Score Methods for Continuous Treatment Doses
Learning Objectives
7.1 Introduction
7.2 Description of Example
7.3 Generalized Propensity Scores
7.4 Inverse Probability Weighting
7.5 Conclusion
Study Questions
Chapter 8. Propensity Score Analysis With Structural Equation Models
Learning Objectives
8.1 Introduction
8.2 Description of Example
8.3 Latent Confounding Variables
8.4 Estimation of Propensity Scores
8.5 Propensity Score Methods
8.6 Treatment Effect Estimation With Multiple-Group Structural Equation Models
8.7 Treatment Effect Estimation With Multiple-Indicator and Multiple-Causes Models
8.8 Conclusion
Study Questions
Chapter 9. Weighting Methods for Time-Varying Treatments
Learning Objectives
9.1 Introduction
9.2 Description of Example
9.3 Inverse Probability of Treatment Weights
9.4 Stabilized Inverse Probability of Treatment Weights
9.5 Evaluation of Covariate Balance
9.6 Estimation of Treatment Effects
9.7 Conclusion
Study Questions
Chapter 10. Propensity Score Methods With Multilevel Data
Learning Objectives
10.1 Introduction
10.2 Description of Example
10.3 Estimation of Propensity Scores With Multilevel Data
10.4 Propensity Score Weighting
10.5 Treatment Effect Estimation
10.6 Conclusion
Study Questions
References
Index

Editorial Reviews

This book provides an overview of propensity score analysis. The author’s introduction situates propensity score analysis within Rubin’s Causal Model and Campbell’s Framework. This text will be good for the advanced user with previous knowledge of the R language, complex survey design, and missing data.

S. Jeanne Horst

This book offers a comprehensive, accessible, and timely treatment of propensity score analysis and its application for estimating treatment effects from observational data with varying levels of complexity. Both novice and advanced users of this methodology will appreciate the breadth and depth of the practical knowledge that Walter Leite offers, and the useful examples he provides.

Itzhak Yanovitzky

This book provides an excellent definition of propensity scores and the sequential steps required in its application.

Mansoor A. F. Kazi

In general, the book is well-crafted and focuses on practical implementation of propensity score methods featuring the free software R. Even though there is room for improvement that could be addressed in a second edition, we believe that it is a useful book for researchers and graduate students, and therefore, many readers will find it beneficial.

Haiyan Bai

With the use of examples consisting of real survey data, Practical Propensity Score Methods Using R provides a wide range of detailed information on how to reduce bias in research studies that seek to test treatment effects in situations where random assignment was not implemented.

Jason Popan

It is a well-crafted practical book on propensity score methods and features the free software R. I believe many students will like it.

Wei Pan

Clearly written and technically sound, this text should be a staple for researchers and methodologists alike. Not only is the text an excellent resource for understanding propensity score analysis, but the author has recognized the messiness of real data, and helps the reader understand and appropriately address issues such as missing data and complex samples. This is extremely refreshing.

Debbie Hahs-Vaughn