"Researchers often use observational data to estimate treatment effects when randomized controlled trials or experimental designs are not feasible for social, behavioral, and health studies. Unfortunately, using observational data may threaten the internal validity of a study by introducing selection bias. To tackle this problem, Rosenbaum and Rubin (1983) introduced propensity score methods (PSM), which balances the distributions of observed covariates between treatment conditions (i.e., treatment vs. control), to reduce selection bias. Over the past three decades, PSM has become increasingly popular for making causal inferences based on observational studies. This volume provides a concise, introductory text on propensity score methods that is easy to comprehend by those who have limited background in statistics, and is practical enough for researchers to quickly generalize and apply the methods. Although there are other books on PSM, most either focus on advanced topics in PSM or on theories of PSM. This volume covers basic concepts, assumptions, procedures, available software packages, and step-by-step examples for implementing PSM using real world data, with exercises at the end of each chapter. Software code and datasets are available on an accompanying website"--
